[This Transcript is Unedited]
Department of Health and Human Services
National committee on Vital and Health Statistics (NCVHS)
Next Generation Vital Statistics:
A Hearing on Current Status, Issue, and Future Possibilities
September 11, 2017
Hubert H. Humphrey Building
200 Independence Ave., SW
Washington, D.C.
TABLE OF CONTENTS
- Welcome – Bruce Cohen
- Overview: Importance and Need for Vital Records: Local, State, Federal, International – Speakers: Tara Das, Valerie Gaston
- Panel 1: Current Status for Vital Records: State and Local Data Producers
- Panel 2: Current Status of Vital Records: Federal Perspective
- Panel 3: Current Status for Vital Records: Other Users of Vital Statistics Data
P R O C E E D I N G S (9:05 a.m.)
Agenda Item: Welcome and Introductions
DR. COHEN: Good morning and welcome. Thank you all for coming. Welcome to the National Committee on Vital and Health Statistics Hearing on the Next Generation Vital Statistics, Current Status, Issues and Future Possibilities, sponsored by the Population Subcommittee of NCVHS, which I co-chair with Bob Phillips, who is sitting over there. Before we begin, we need to take care of some statutory duties related to all of the committee members and our status as a federal advisory committee.
So my name is Bruce Cohen, and I am retired from the Massachusetts Department of Public Health. I am a member of the full committee, co-chair of the population health subcommittee, and I have no conflicts.
DR. ROSS: Good morning. I am Dave Ross. I am a member of the Full Committee and I am the CEO and President for the Task Force for Global Health. We are affiliated with Emory University. I have no conflicts.
DR. STEAD: Good morning. I am Bill Stead from Vanderbilt University Medical Center. I am chair of the Full Committee. No conflicts.
DR. PHILLIPS: Bob Phillips with the American Board of Family Medicine. Member of the Full Committee and the Population Health Subcommittee. No conflicts.
DR. MAYS: Vickie Mays, University of California Los Angeles. I am a member of the Full Committee, Privacy and Pop, and I have no conflicts.
MS. GOSS: Good morning. I am Alix Goss. I am with Imprado, a consultancy of Dynavet Solutions. I am a member of the Full Committee, co-chair of the Standards Subcommittee. No conflicts.
MR. COUSSOULE: My name is Nick Coussoule. I work for BlueCross BlueShield of Tennessee. I am a member of the Full Committee, co-chair of the Standards Subcommittee and member of the Privacy, Security Subcommittee, and I have no conflicts.
DR. COHEN: Did you just join the Population Health?
MR. COUSSOULE: I said Privacy and Security, sorry, I know you are recruiting new people every day.
DR. RIPPEN: Helga Rippen, Alertgy, and I am a member of the Full Committee and a member of the subcommittees Privacy and also Population Health. No conflicts.
DR. COHEN: Are there any committee members on the phone?
DR. CORNELIUS: Yes, good morning. Lee Cornelius, University of Georgia. Member of the Full Committee and the Population Health Subcommittee. No conflicts.
MS. STRICKLAND: Debra Strickland. I work for CONDUENT. I am a member of the Full Committee and members of the Standards Subcommittee. I have no conflicts.
DR. COHEN: Thank you. Thank you for joining us on the phone. I hope things go well with you down there.
Let us have the Committee staff introduce themselves as well.
DR. BRETT: I am Kate Brett. I work at the National Center for Health Statistics. I am the lead staff for the Subcommittee on Population Health.
MS. HINES: Good morning. I am Rebecca Hines. I am also with the National Center for Health Statistics and I am the Executive Secretary and the Designated Federal Official for this Committee. I thank you all for coming.
MS. DORSEY: Rashida Dorsey. I am the Executive Staff Director for the National Committee on Vital and Health Statistics. I am with the Office of the Assistant Secretary for Planning and Evaluation.
DR. THORPE: Good morning. Roland Thorpe, on the Committee for Vital Health Statistics. No conflicts.
DR. COHEN: For the next two days, we are going to have over 40 people testifying and presenting. Our agenda is really tight, so we are going to try to hold folks rigorously to the amount of time that we told them. We have a draconian sign master here. Kate will give you the three-minute warning, the yellow one-minute warning and the red, you had better stop now, or we will use the ejector seat button.
Although this is a hearing, we hope there will be plenty of opportunity to discuss the issues. The way hearings work are presentations, and then the committee has a chance to ask questions and engage in conversation. And then we will have opportunities for others who aren’t on the committee designated throughout the day to continue the discussions.
We also want to try to get feedback in other ways as we move forward. So we are going to be using something called Poll Everywhere website, as well as classic index cards. Do you want to describe the index cards and the Poll Everywhere?
DR. BRETT: Sure. So we really want everyone’s feedback in whatever way makes it easy and comfortable for you. The Poll Everywhere software allows you to either use your computer or tablet and go to a website, log in and type comments. Or you can text on your phone, and we will have all the instructions as it comes up when we are ready for this. But you text 22333/NCVHS, and that opens you up to that website. Then you can text in however much you want, as multiple times over the course of the whole two days.
And what we will be doing on our end is being able to see the comments that are coming in. They will be recorded for us. It is just easy for us to then take that information for use when we put together the summary report after the effect.
Also, if you like index cards better, we have index cards because some of our co-chairs like index cards. We will just type them in, so that we get it in the right place. Now for questions?
DR. COHEN: On the index cards, if you have a comment, just label it C. And if there is a question that you would like us to discuss or address, just put a Q. I think that is the best way to do it.
DR. BRETT: We will not be using the Poll Everywhere to pull in questions. So if someone is online, or there is an overflow room for this hearing because we are concerned that at some point we might run out of room, so we do have a room on the fourth floor. If people are participating in that way, we will take questions using the WebEx chat box and pass it to the committee to ask those questions.
DR. COHEN: Thanks, Kate. So before we get to our first set of speakers, I want to start off with a couple of quotes. The first quote, most people take vital statistics for granted, assuming that statistics they need should be freely available as part of today’s culture. Death rates are among the typical vital statistics that most people assume we have always had available and, without much effort, will continue to have available.
The real story is quite different. National statistics of deaths and births were achieved only within the present generation after two centuries of intermittent struggle and building. This was written 67 years ago in 1950. Some progress, but not a lot. The issues still remain.
The second quote, US vital statistics systems rely on the original information reported and the consistency of that information reported by a myriad of physicians, parents, funeral directors, channeled through state and local information systems of widely varying levels of sophistication and automation. Unfortunately, the data providers do not work for vital statistics. The challenge for state and local vital statistics staff is to maintain and improve data quality and timeliness.
The data are then processed by NCHS, a federal statistical agency that has experienced relatively flat funding for many years. The challenges facing vitals and the continuing importance of the resulting data make it an important topic for periodic re-examination, assessing both the current and emerging uses of data, and considering the methodology and organizational features of compiling vital statistics data. This is from a National Research Council report on vital statistics issued in 2009.
So as we can see, vitals has always been on our mind in a variety of ways. Vitals touch us all, whether it is through civil registration or public health information and surveillance, or securing identity which certainly became a much more prominent focus of vitals 16 years ago today in this country.
I am looking forward to the next two days where we have discussions that will help us all better understand, sustain and improve the vital statistics system in this system. So thank you for all participating, and I look forward to a wonderful engagement over the next two days.
And with that, I will introduce our first panel. To set the stage, we have two speakers. Tara Das, who is the registrar and director of vital statistics for the state of Texas, and Valerie Gaston, who is the director of health statistics for the chief of the Division of Vital Statistics for Stats Canada. Thank you. Tara?
Agenda Item: Overview: Importance and Need for Vital Records: Local, State, Federal, International
DR. DAS: Good morning, everyone. I am Dr. Tara Das, state registrar and director of Vital Statistics for the Texas Department of State Health Services. Previously, I was the director for the Office of the Registrar in the Bureau of Vital Statistics for the New York City Department of Health and Mental Hygiene.
So vital statistics is a field that I am very passionate about. I am very happy to have the opportunity to talk with all of you about it.
Let’s start off things by talking about births and deaths. These are milestone vital events. There is not one of us here who has not been personally affected by a birth or a death taking place.
Births and deaths are significant moments. They are often marked by families and loved ones coming together. They are also marked by legal documents, birth and death certificates, which serve a number of critical government and societal functions by having marked that a vital event has occurred at a specific time, date and place to an individual as stated on the birth or death certificate.
Critical government and societal functions for which a birth or a death certificate is needed and often required include obtaining legal identity documents like passports, driver’s licenses and social security cards. Enrolling in schools, registering to vote, receiving benefits, insurance and social security, and settling estates of family affairs.
As such, birth and death certificates need to ensure confidentiality and integrity of personal information, prevent identity theft and minimize fraudulent use. Excellence in customer service, security and data quality is a must for our operations to be successful.
So how are these birth and death certificates created? When a birth or a death occurs, it must be reported by a legally-authorized data provider. For births, there is usually the medical practitioner or midwife who attended the birth. And for deaths, the process is slightly more complicated as it entails two data providers for completion, usually the medical practitioner who attended the death and the funeral home working with the family.
The data provider reports that birth or death life event to the responsible government entity, usually the state health department, in that vital records jurisdiction, so that the events can be reviewed and registered as a vital record, and issued as a birth or death certificate. In the United States, there are 57 vital records jurisdictions. They are responsible for receiving birth and death reports, and reviewing and registering them as vital records. So that is all 50 states, five US territories, New York City and Washington, DC.
Each of these jurisdictions is overseen by a state registrar who is the official custodian of birth and death records for that jurisdiction. So if you need to report a birth or death, you need to report it to that vital records jurisdiction where the event took place. If you need to order a birth or death certificate, you need to order it from that vital records jurisdiction where the event took place. If you need a correction to a birth or death certificate, you need to request it from that vital records jurisdiction where the event took place. Do you see the pattern?
The legislation rules and policies that determine who are the authorized data providers who can report births and deaths, and by what deadlines, who can request or order birth or death certificates, and who can request corrections to those certificates are all governed locally by that state, territory or jurisdiction. The system of vital records in this country is truly a federated system.
For me, there is no clear or more timely way to demonstrate the federated nature of our vital records system than through hurricanes. There has been a lot in the news lately about hurricanes unfortunately. You have read, and perhaps been affected by, the devastation brought by Hurricane Harvey which hit my state, Texas.
For me, it has been reminiscent of events during Hurricane Sandy, which at New York City and other areas in October 2012 when I was working at New York City Vital Statistics. At the time, Hurricane Sandy was the second deadliest hurricane to hit mainland US after Katrina with about 41 deaths in New York City alone, in addition to mass evacuations and disruptions.
New York City is the only city that is an independent vital records jurisdiction, meaning that all of its births and deaths are reported, registered and issued through New York City alone. I mentioned that reporting deadlines for data providers vary according to the laws and rules of that vital records jurisdiction, and that also across jurisdictions, death certificates have to be completed in two parts, the medical portion and the funeral home portion.
Well, in New York City, the medical portion needs to be reported within 24 hours of death, and the death certificate completed and registered in 24 hours of death. New York City also has a 24/7 death registration and certificate issuance unit that has to be up at all times, even during times of disaster. So for those of you familiar with continuity of operations planning, death registration in New York City has a 0 RTO recovery time objective.
During Hurricane Sandy, I and many others had to evacuate our homes. We were later notified that we had to evacuate our vital statistics offices, but maintain continuous uptime and continue our operations. So needless to say, it was a hectic and stressful time. We did manage to relocate our offices, and we also maintained communications with hospitals and medical examiners. For its medical examiners who would be reporting those hurricane-related deaths as such deaths are not natural, so they cannot be reported by hospitals.
We also contributed to disaster mortality surveillance because deaths have to be medically reported in New York City within 24 hours and registered within 72 hours through our electronic registration system. So the death registration process was integral to emergency response.
Death certificates across the US and the world contain medical information on manner and cause of death, place of death and injury, and nature of injuries, as well as demographic information on age and race. So it can be extremely useful for mortality surveillance and disaster response if deaths are reported timely.
My personal and work experiences during Hurricane Harvey have been different, despite the seriousness of both hurricanes and the human impact. Unlike New York City, Texas is a dual-registration state. There are approximately 400 local registrars in our counties over whom I have supervisory power as state registrar, but who retained their independence.
For the deaths reported by data providers, for their local registration districts, it is the local registrars who review and release those death reports in our statewide electronic system prior to the state registering them as death certificates and official vital records. In Texas, reports of death are initiated by funeral homes, whereas in New York City, it is medical facilities. And deaths in Texas need to be registered within 10 days of death.
Texas is also the second largest state in the US. Also just as an aside, for me, it is very exciting to be the state registrar in not only the second largest state, but the second most populated state in this country. Because our vital records system is federated, the unique strengths and challenges of each jurisdiction influence how we can achieve our goals. Because it is such a large state, Texas not only has local registrars and a dual registration system, but Texas residents have not been equally affected by Hurricane Harvey, as was the case during Hurricane Sandy in New York City.
The state government buildings where I work are located in Austin, which is in Central Texas, and the hurricane hit east. We in Central and West Texas have not had to evacuate our homes or close our offices, but our neighbors and families in East Texas have. In fact, many of our local registrars in East Texas had to close their offices, which affected the death registration process.
As part of disaster and public health surveillance, I had to provide education around the death registration process, how it produces mortality statistics, and how it is affected by local geographies, laws and rules. This need for education around vital records, and how the reporting and registration process impacts public health and disaster response, is not a surprise, nor is it a rare occurrence.
I have been talking thus far about vital records, whereas the title for this hearing refers to next generation vital statistics. The terms vital records and vital statistics are often used interchangeably, but there are distinctions.
When data providers report births and deaths, they complete standardized data collection forms that are prescribed by individual vital records jurisdictions. Those standardized forms contain legal, demographic and medical information. So the data collection form for the birth certificate not only contains fields for names, dates of birth and birthplaces for the child and the parents, but also medical fields for risk factors for pregnancy, infections present and treated during this pregnancy, and characteristics of labor and delivery procedures amidst other medical information.
Likewise, the data collection form for the death certificate that data providers report not only contains legal and demographic information like names, race, names of surviving spouse, dates of birth and death, date, method and place of disposition or burial. But also medical information on manner and cause of death, pregnancy status of female descendants, whether tobacco use contributed to death, and injury information for deaths that are not natural.
The medical data that is collected as part of birth and death reporting are not all issued as part of birth or death certificates and stored as vital records, but they do constitute vital statistics. Vital statistics facilitates public health and policy planning into areas like disaster mortality surveillance, improving prenatal care, identifying relationships between pregnancy factors and newborn health, identifying leading causes of death by age and race, and reducing deaths due to specific causes.
With the legal, demographic and medical information that is collected, vital events reporting by our data providers in medical facilities and funeral homes serve dual purposes, legal and administrative, public health and research. Data collected in the course of vital events reporting are transformed into vital records for issuance of birth and death certificates, for legal and administrative purposes, and that data collected is also transformed into vital statistics for dissemination of data and statistical files and reports for public health and research purposes.
With these dual purposes of vital events reporting, legal and administrative, public health and research, both sides, customers of vital records and customers of vital statistics do not always realize that their work stems from the same source, that initial vital events reports from the data provider who oversaw the birth or death event. And that the work is thereby affected by local geographies, laws and realities, as well as by the timeliness, quality and accuracy of the information that data providers report.
Consequently, for the vital records and vital statistics system to improve, there is a strong need for leadership and advocacy that can educate and speak to the mission-critical nature of birth and death certificates, vital records and vital statistics. And that can educate and speak to the need for excellence in customer service, security and data quality given the dual purpose of vital events reporting, the legal and administrative public health and research.
This need for advocacy and education does not only apply to stakeholders and customers of birth and death records and statistics, but also to internal staff who work in vital records jurisdictions on a day-to-day basis. I want to take some time to talk about what our staff do in the trenches because their work is not often recognized.
The heavy lifting of vital records jurisdictions is predominantly done in health departments, within bureaus or offices titled vital records, vital statistics, health statistics or some variant thereof. The majority of staff in these offices are usually assigned to process orders and requests for birth and death certificates, and process corrections and amendments to those vital records.
These staff work with the general public, and they can get overwhelmed by the number of customer applications and phone calls they receive regarding why couldn’t a customer get a particular correction approved? Why couldn’t they receive the birth or death certificate that they ordered? Or why is it so important for them to get a particular service?
Stressful working conditions can ensue in these offices with low salaries, high learning requirements for examining customer identity, documents and applications to reduce identity theft and fraud, high customer demands, low recognition, and high public visibility of mistakes that they make in the course of their work. This happens particularly in vital records jurisdictions that seem high volumes of customers.
I do think it is in important for our vital statistics, research and public health colleagues to understand that these public and family-facing demands can occupy much of our time, not just in vital records and vital statistics offices, but in the medical and funeral home facilities for our data providers. You may think, for instance, that physicians would all be equally aware of the public health need for their timely and quality birth and death reporting. But when data providers are reporting births and deaths, and our staff are processing those records, they are usually more so considering the legal and administrative purposes of that individual certificate in dealing directly with the family who is going through some of the most stressful moments in their lives.
And in those moments, our data providers and staff are less so considering the public health importance of the aggregate vital statistics to which that individual family certificate is contributing. Given this context, it can be difficult for our staff and data providers to take a step back and see the larger government and societal purpose to the work that they do.
I think the responsibility lies with the state registrar and other leadership to demonstrate to them how important and valued their work truly is, and ensure that they get the necessary support. That through their day-to-day actions of working with data providers, reporting births and deaths, reviewing and registering birth and death records, issuing and correcting birth and death certificates, and producing birth and death reports, our staff and data providers are fulfilling mission-critical functions that are essential to the running of government and society. Concerns around identity theft, fraud prevention, security, data quality and public health are in their hands, the hands of our staff and our data providers, to address.
So in this hearing as we focus on our guiding question, how do we transform today’s vulnerable vital records data collection network into a network of state systems that produce accurate and timely information supporting a breadth of local, state and federal data needs. Let’s not forget the support that is needed for the hard-working folks who work day in and day out to provide you with the birth and death certificates, records, and statistics that you need, and the challenges that they face in working directly with the families who experience those momentous life and death, birth and death vital events.
Now, let’s take this discussion to the next level up. From the experiences within the vital records jurisdiction, the individual one, to the networking community of vital records jurisdiction in the US and the world. Recall that the legislation rules and policies that specify the authorized data providers, reporting deadlines and data collection forms for birth and death reporting are governed locally in the US by each vital records jurisdiction.
Moreover, levels of government funding and bureaucratic procedures, as well as leadership and political will, vary across the vital records jurisdictions. That influences the state of modernization in each jurisdiction. The extent of electronic registration, certificate imaging, digital archiving, electronic identity verification, health information exchange and data analytics.
The system of vital records and thereby vital statistics in the US is truly a federated system. It is rooted in local realities, but essential for national security and public health. Recalling our discussions of Hurricane Sandy and Harvey, that differences in reporting deadlines and registration procedures can impact how death certificates can be used in disaster surveillance across vital records jurisdictions. Despite the differences, in both New York City and Texas, there exists a robust electronic registration system, which is not the case across all jurisdictions.
Electronic birth and death reporting and registration are legislatively mandated in both New York City and Texas. And that facilitates more timely and quality research, surveillance, and outreach with data providers on the timeliness and data quality of the information they report.
Electronic registration also facilitates real-time data analysis on birth and death certificates as they are being data entered, which was done during Hurricane Sandy. In paper-based birth and death registration systems, all such activities would have to wait until the completed paper certificate arrives in the vital records office for registration.
So with the state of modernization and extent of electronic systems in such variance across the US and the world, when the vital records and statistics community comes together, the discussions around improvement and innovation often devolve to the lowest common denominator. That is how to get all states and territories to the same basic level of electronic system use. While perhaps beneficial to some who haven’t reached that basic level, it is a disservice to others who have already exceeded it.
We need more national and international conversations and dialogue that look to raise the ceiling on customer service, security, and data quality in our operations. So how can we cultivate a community where ideas and innovation for all are discussed to truly advance our field of vital records and vital statistics, such that leaders remain engaged, laggards not discouraged, and variation in governing laws and rules, not a hindrance, but perhaps an opportunity to capitalize upon in our federate system.
Education and engagement in the dual purposes of vital events reporting, legal and administrative, public health and research, and how best to achieve these purposes with regard to customer service, security and data quality are essential at multiple levels with our data providers who report births and deaths, our vital records staff, our public health colleagues, data partners and other vital records jurisdictions. How can such education and engagement be facilitated at these multiple levels?
I have not yet spoken about the rule of federal government in day-to-day vital statistics operations. Each of the 57 vital records jurisdictions has a contract with social security administration to regularly report births for assignment of social security numbers to newborns and regularly report deaths for the termination of federal benefits.
We also each have a contract with the CDC National Center for Health Statistic, NCHS, to regularly report births and deaths with required sets of core data items to constitute national vital statistics files for national-level research surveillance and policymaking. There are advantages to our vital records and statistics system being federated. To be able to quickly respond to local disasters, emergencies and policy needs without having to wait for national-level data to be made available.
But with national-level vital statistics only as good as the slowest state performers with regard to timeliness, data quality and modernization. These conversations also tend to remain focused on the lowest common denominator amongst all jurisdictions, and how to get everyone to the same floor, as opposed to raising the ceiling on how we can truly innovate customer service, security and data quality.
Now, such conversations about getting everyone to the same floor for improving the timeliness and quality of national vital statistics are definitely not without merit or need. There is a great deal of work that needs to be done in this area.
But perhaps the federal government can also play a role in helping our community identify ideas for improvement and innovation that would address our day-to-day local realities, challenges and goals in our respective vital records jurisdictions, so that we can all efficiently and effectively fulfill the legal and administrative public health and research purposes to the work we do.
Our vital records and vital statistics community needs a clear path forward, one that would benefit all jurisdictions, so that we can all grow, improve and innovate without leaving anyone behind, a clear path forward that will leverage the strengths and mitigate the weaknesses of our federated system and quickly responding and adjusting to national level, as well as state and local level needs, demands and concerns for vital events reporting, vital records and vital statistics. Thank you.
DR. COHEN: Thank you for that wonderful overview and for the challenge you have laid out for us. Two quick bookkeeping items, it just vanished. There was a slide that gave a summary of the vital statistics system in the US. Hard copies of those are available for committee members. If others wish, let us know, and we can provide that to you.
I see a new committee member. Denise, could you check in and let us know, and Linda, as well?
MS. LOVE: Denise Love. National Association of Health Data Organizations. Committee member, subcommittees Standards and Population Health. No conflicts.
MS. KLOSS: Linda Kloss. Member of the Full Committee, Co-Chair of the Privacy, Confidentiality, and Security Subcommittee. Member of the Standards Subcommittee. No conflicts.
DR. COHEN: Thank you. Valerie?
MS. GASTON: Good morning. I would like to start by thanking you for inviting me to this hearing. I am really excited to be here to talk to you about vital statistics in Canada, but also for hearing what you are working on and how it is going to affect us. I feel like a little brother or sister of the US, so I hope that I can bring something to the table as opposed to just taking things away.
During my talk today, I have been asked to focus my comments on what we have done that maybe has worked well, and hopefully you could borrow. So I am going to touch a little bit about vital statistics in Canada. I would like to thank Tara because a lot of it, you have said, so a lot of it is the same.
I want to talk about the vehicle for data collection, a dataset that we have on the Canadian coroner and medical examiners data. And finally, I want to talk a little bit about data access and challenges in civil registrations that affect vital statistics.
I want to start by just giving you a brief overview of vital statistics in Canada. I think that Tara spoke really well about the duality. So the first point I have is civil regulations. And in Canada, the civil registration of birth, death, still births and marriages occurs in our ten provinces and three territories. And this is governed by each province and each territory’s vital statistics act.
And in the last 10 years, I would say that most of the vital statistics agencies have moved out of the health domain departments in favor of service departments. This has added a challenge for us because where the link to health was really clear and part of their mandate, right now, their mandate is more of a service standard. It is issuing these birth registrations and death registrations within a specific period of time.
I represent the vital statistics part of it at Statistics Canada. So we are mandated with reporting on births, deaths and stillbirths at a national level. And we have been releasing these statistics since 1921. And the data are collected under the Statistics Act. This act governs the collection, the use, the retention, as well as the disclosure or sharing of these data.
So I want to point out one key difference that will affect some of the things that we do that may work for us is that Canada is like a California. Our population is about 35.8 million people. We have some provinces and territories that have under 500,000 people, so they are spread out, but there is not a lot of them. So the cornerstone of our system of vital statistics is really the cooperation and collaboration between the 13 provincial vital statistics registrars and the federal government.
So we have the Vital Statistic Council for Canada. We rely heavily on the VSCC, I will use acronyms, which is an inter-jurisdictional advisory group, which is composed of the heads of the vital statistics agencies from each province and territory, as well as the health statistics division at Statistics Canada.
It was created in 1945 and meets annually in person and monthly via teleconference. Of course, with 13, it is a lot easier to bring 13 people together to meet annually, as well as via teleconference. So we get full participation at almost every call, which is a blessing.
I mentioned before about many of them moving out of health departments and into service departments. One of the key struggles for us, and the reminders, is making sure that we stress the importance to continue to collect the data on the health variables. Now, we collect a much smaller set of health-related variables than you do in the US, which might be a bit easier to keep that pressure on. But it is also a concern of ours.
I wanted to mention, as well, that the vital statistics council at the annual meeting, we have greatly benefited from the director of vital statistics from NCHS attending, as well as the president from NAPHSIS. And we learn a lot from those presentations. I feel like in the last few years, I have just been following Delton around and asking more questions.
We also try to attend NAPHSIS as much as we can because that is also a great learning experience for us. This year actually was wonderful. They asked us to give a short presentation. And again, I felt a little bit like what do we have to say that is new.
I think I was actually pleased that many people did find that there were different issues a little bit, and that we treat them a little bit differently in Canada. Hopefully, some of that will come up today.
Okay, so our main vehicle for data collection is the National Routing System. And so this is a set of technical and messaging standards, and I imagine you have the same with Steve. I think the main difference for us is that it is point-to-point. So we have the data provider that has a bilateral agreement with the receiving organization. So it is a little bit complicated in that it makes for a lot of agreements. Again, this is 13, not 57 I think you said in your total, yes, 57 jurisdictions. So that is a little bit easier.
But it also allows for greater control from the data provider to control what elements will be disclosed to the data recipient, and also ensures that the data recipient only gets the elements that they are authorized to receive. So this is near real-time. We have messages right now for birth, death and stillbirth notifications. We have birth certificate validations. And we are working on future messages for change of name and change of sex.
So the NRS partners right now are the 13 vitals registrars in the provinces and territories. We have three federal partners, so Statistics Canada, the Canada Revenue Agency and Service Canada. And we are also looking at including a couple of other federal departments, Immigration, Refugees and Citizenship Canada, basically passports, and Indigenous and Northern Affairs Canada. Also I put Nunavut, which is the last territory that has not yet — sorry, I have three minutes. I am going to run through this.
I think one of the advantages of the NRS is we have been able to leverage the funding from the different departments. Obviously the vital statistics registrars depend on our funding to exist. So they are getting funding from each partner for each message. And also, when we got the system in, that was a way to sort of share the cost for building the network.
So the NRS is governed through a subset of federal, provincial and territorial partners. We currently have committees that are working on gender identity and how to include the gender X. The change of name, change of sex notifications, and also the inclusion of aboriginal characters. So what we are trying to do through this sort of small group is to find a common approach, so that when different provinces, different jurisdictions come in and are ready to deal with an issue because the larger jurisdictions usually see them quicker, which would be your situation, Tara, that there is already something, a structure for them in place that they can follow. And it gives them a little bit of something to rely on to speak to their colleagues in government and say, well, we should do it this way.
The Canadian Coroner and Medical Examiner Database has really been a complementary source of data. It provides us with a standardized subset of death data that are investigated by coroners. It provides additional detail on specific elements that are not part of the vital registry’s data elements. So activity, place of death, any circumstances around death, that sort of thing. We have a minimum dataset that contains basic information that is available on all deaths.
The continuum of access, I wanted to talk about this a little bit because data access is a huge issue. On the left, you have this sort of general statistical product. But on the right, the access to micro data, that is a big thing for us. We have clear policies on how these data can be shared. It is very important to share access to vital data.
So one of the key things we have is a Research Data Center. It is a network with the universities. It is a sort of a Stat Can mini office in the universities. So researchers can go and access the data. The data outputs are vetted. So the data stay in the RDC. The outputs are vetted, so that any confidentiality rules are applied. And then there is a whole process around it on research proposals and how they are accepted and that sort of thing.
I will just finish with our challenges. So we have some challenges that are probably the same here. That is around the collection of gender or birth, sex at birth. We have the parentage issues. For us at the vital stats, it is important to know when you have four parents on a birth registration which one is the birth mom. So those are the types of things, timeliness, relevance are other ones. I will finish on that. I am sorry for going over my time. Thank you so much. I will be around if you have any questions.
DR. COHEN: Thank you so much, Valerie. I apologize for the short amount of time that you had, but I am sure we will have many discussions. The way this hearing will work is we will have questions now from the committee. Are we going to put up the URL for people who want to make comments using our electronic feedback system? Where are those index cards if people want? They will be available at the front desk if folks want index cards to write comments on.
After the committee has a chance to comment, we are going to roll right into the first half of our next panel, which will focus on people who work at the local level generating vital statistics. So questions from the full committee? Do you want to just raise your tents the way we normally do? This is the high-tech way. I will start at the end with Denise.
MS. LOVE: I am interested in Canada. We have had some interaction with the hospital discharge data people and the provinces. So in general, really quick, is there much interchange or linkage or cross-sharing of data from the provinces and Statistics Canada?
MS. GASTON: We receive all the datasets. We can only share datasets when we have received the permission of the data providers to do so. It is actually a fairly complicated process.
I think what we can add is linking these different datasets together. We have expertise in linkage. Then having those linked datasets available in the RDCs I think is a way that is manageable. Of course, it doesn’t provide access for everybody, but it is a nice sort of in-between. I hope that answers your question.
DR. MAYS: Thank you both for great presentations. They were very stimulating. I have a ton of questions, but I will keep it down to just a few. What would you want to see in terms of what happens during a disaster that could make your job better in terms of the quality of the records?
I did work in Katrina, so I know a lot about kind of the disaster for people trying to get records. So do you have any suggestions about anything that we should do differently during disasters? And then my second is what agencies do you interact with in terms of fraud prevention? Again, if you had your wish, what would you want?
DR. DAS: I wish more people in Texas asked me what were my wishes and what do I want. So thank you for that. I think with regard to both, disasters and fraud prevention, the impetus, the support from within our state and local government to do something, doesn’t happen until it is too late.
I think for those of us who are constantly thinking about these issues on a day-to-day basis when nothing has blown up, it is difficult to get the attention from our leadership at whatever levels they happen to be, to do something about it.
DR. MAYS: What do you need?
DR. DAS: In terms of very broad what we need is I think, at the federal level, to keep that focus for the states and jurisdictions on a day-to-day routine basis, to be able to – funding priorities, mandates, to have structures in place, so that we are dealing with disasters and fraud prevention, not after something had happened. I think that is the biggest thing.
And so I think the more that our leaders within state and local government here that mandate and those priorities set at the federal level, that helps us on a day-to-day basis to get what we need during routine circumstances, so that we are prepared when disasters or some sort of security breach happens. We are already prepared for that.
Because again sort of when you asked the question about what agencies do we work with on fraud prevention, that again sort of is dependent on the state registrars and the vital statistics offices leadership and initiative to engage proactively with other agencies. I just think that fraud prevention, as well as disasters, are some things that the focus is not maintained on them during routine circumstances.
MS. KLOSS: Thank you both very much for great presentations. I would like to follow Vickie’s lead. I have a quick question for both of you. Valerie, I would like to give you an opportunity to summarize the slide you had to skip over on access issues. Is it your advisory group that is responsible for creating the policies for access? What is your mechanism for getting that job done?
MS. GASTON: It is not actually. It falls under Statistics Canada’s policies. It is a double-edged sword because we have really strong legislative support for maintaining the confidentiality and the privacy. That also means that there are many, well, requirements to provide access.
So we have to first think of what the different uses will be. We ask permission from the registrars for those. So that we discussed sort of as a group with the vital statistics council, but of course, it depends one each jurisdiction’s legislation.
So we try to, through those meetings, have an opportunity to perhaps express how important it is for them to try to change their legislation, if that is an option. It doesn’t always work, so sometimes we have to take one jurisdiction out of a dataset if we just can’t meet their requirements.
MS. KLOSS: Tara, you mentioned data quality several times. I was just curious as to whether there is any kind of standardization process for data quality? Or is that very much local and jurisdictional and research-based? Or are there some general guidelines on what quality levels are to be expected or what mechanisms for assessing data quality are used?
DR. DAS: So, I think one is that having very specific standards and expectations for what good data quality looks like, can definitely be refined. I think that could be something that is one of the action items to come out of here. I think for all of those, I mentioned customer service, security and data quality, is having sort of specific standards and best practices for them would be really welcomed.
With regard to data quality, one thing that we do look at across jurisdictions, and with NCHS’ help, is a percent of data items left unknown or left blank. It is very basic, missing an unknown value. Cause of death data quality is something that is spoken about like everywhere and at all levels, state, local, national, international.
And so far, the gold standard has been does it match what is in the medical chart. But are there other standards that we could be using to look at data quality? They haven’t really been identified yet.
DR. COHEN: That is across jurisdictions or within jurisdictions? Just the general quality question.
DR. DAS: I can’t speak for every jurisdiction, what they are doing within. But the percent of missings and unknowns, that is standard. And then, it is really does this reflect what is in the medical chart. But other jurisdictions may have come up with something more, so I don’t know.
DR. ROSS: Thank you again both of you. Tara, following on Linda’s question, you spoke very eloquently about the challenges of security, maintaining secure data, also getting high quality data with this tension on customer service. And in this federated system that we have in the US, that is a fact of life, you spoke directly to this need to cultivate this community across the vitals community that really leads to excellence rather than the lowest common denominator.
Do you have any suggestions or wishes on the federal involvement that would advance that? What would that take? What does that look like to you from a state perspective?
DR. DAS: I think something that would be really helpful and kind of speaks to Bruce’s question on the standards of data quality within each vital records jurisdiction, which is something that I can speak to because I am not familiar with all that goes on in all 57 jurisdictions. But it would be really helpful if we did have some sort of repository that at least can sort of collect what are the levels of efforts or innovations happening at the state and local levels.
I think that that, just establish what exists, what have innovators within states and jurisdictions have done for someone to be able to provide that work, so that we can sort of capture where we are at, at all the varying levels, and then establish standards. What is the ceiling for excellence in customer service, securing data quality, given these varying efforts and standards that have just sort of sprung up in our federated system. I think that would be a wonderful thing if the federal level can help us with just shepherding that through.
DR. ROSS: So, like creating and maintaining a really vibrant community of practice that produces those best practices and agreements?
DR. DAS: Right. Sorry, I was going to say, and also looking at, you know, maintaining what is going, as much as we can learn, in the private sector with regards to customer service and security, just seeing what are advances going on in other industries and other sectors that we could benefit from, as well.
DR. ROSS: Valerie, you pointed to the similarities to the Canadian and US systems. But some of the things that you are doing in Canada, among those that you think you are doing well that you would recommend we do? Is there one thing that you would thing. Like, hey, US, you are missing the boat. Do this. We have tried. It works.
MS. GASTON: I am not sure I understand all the intricacies of your system and where the differences are. I think what has been really helpful for us, has been in the implementation of the NRS has been getting different federal departments involved. It means that, well, first of all, we can focus on our mandate, and then the other federal departments can focus on theirs and deal directly with the vital registrars.
It also ensures kind of that the NRS will live on because there is always people with great ideas to do something bigger and larger and knowing how difficult it is to just do something small, these ideas are sometimes a little bit frightening for us. I don’t know if that is me being close-minded. But I don’t want to have something that would harm our sort of system now because it is working.
I think that has leveraged the funding, as well. It has taken some funding pressure off of our department and also provides the registrars with extra sources of funding. So I feel that is maybe something that if it is not in use, now could be. I think that is the main one I can think of.
DR. COHEN: Just a reminder, I have tried this text up here, and it works really well. It is easy to use. If you want to do that to provide feedback to the committee, that would be really helpful.
This session of discussion is just for committee members. We will open up a broader discussion later this morning for anyone in the audience and on the phone who wishes to discuss and ask questions.
DR. RIPPEN: I have a question for each one of you, a little bit different. I guess, Tara, with regards to kind of not being recognized. If there was something that you could really brag about, that people really should know, what would that be? And then if you could identify the most urgent issue, but then also one that you see looming in the long-term, what would that be? I am cheating. I am asking three.
DR. DAS: That is really nice of you to ask what is the one thing I would brag about, just to get a little bit more recognition to our hardworking staff. I think it would be that despite all the challenges, stresses and frustrations that they feel in dealing with a tsunami of emotions that come their way from our families and customers who are, like I said, going through some of the most stressful moments of their lives.
Our staff are very committed to ensuring they are providing excellent customer service. They face barriers in doing so, but they want to do an excellent job. Folks don’t necessarily come into vital records and vital statistics thinking that is going to be the career path that they will take.
But I will say that for almost everyone who does take this path, the heart that you see, it is really wonderful. I have started that process of education within the state of Texas about that, taking a step back and seeing the larger purpose to the work that we do. It resonates. It makes people feel even that much better about the work that they do.
I would say it is the heart that folks have and definitely the sense of mission that they have. They know that the work that they are doing, even if it is not being recognized, they know how important it is to get those birth or death certificates to the families, and also to ensure that our research and policy partners and colleagues get the statistics that they need. That is what I would brag about. I love working with people in vital records at all levels. It is a lot of fun.
Urgent issues, I think data privacy is a big one, especially now, and again, I am speaking from the perspective of having electronic registration systems in both New York City and now in Texas. We have the ability. Other folks, they know how valuable our birth and death data is, and they want to link to it. They want to add it to lifetime health registries or what have you.
I think sometimes the compelling public health purpose of having such data sometimes overrides people’s thoughts about data privacy. These are people’s personal and private information. I think maintain data privacy, not just for the paper certificates, but for that electronic data, is a big thing. Again coming back to standards and the community of practice, having that, what we should be linking to, what we shouldn’t be hasn’t really been established. I would say that is an urgent issue.
And then thinking now, just again giving what is in the news lately, well, data privacy has been in the news lately, but also disaster response. Perhaps that in itself will hopefully compel leadership to ensure that we can meet disasters because we do it in vital records and statistics in two ways. Not just in providing mortality surveillance where it is possible given the death registration procedures in each jurisdiction.
But also, folks, families having to evacuate their homes or their homes being destroyed, they now don’t have their birth certificates or driver’s licenses or passports. So that is another area where our community is extremely important when it comes to disaster response. It was not just in providing data to epidemiologists and other policymakers, but also in helping families, perhaps waiving, we are looking into that right now, waiving fees for birth certificates, so that folks can start rebuilding their lives.
Since it seems that we have been getting one disaster after another, I would say that is really an urgent issue to make sure that we are prepared like I said, during routine times, so that we can answer the call for non-routine times.
DR. RIPPEN: Do you provide access to data for researchers in a controlled environment? Or do you actually physically give them a dataset?
MS. GASTON: We give them a dataset. We give them access to the dataset. They can go work in that physical location. But then what they leave with are their outputs that have been vetted for confidentiality. So micro data files, in rare instances, are actually shared with other federal departments. We are getting away from sharing them with individual researchers.
DR. COHEN: Thank you both. That was tremendous. We are going to move on now to the first half of our next panel and then take a break. The four speakers we will have in the first half of our next panel, which focuses on state and local data collection and data producers are Shawna Webster, who is the executive director of NAPHSIS, the National Association for Public Health Statistics and Information Systems. Rich McCoy, who is the president elect of the board for NAPHSIS and also the director of the Vermont Center for Health Statistics. Angela Shaw, birth registrar for MedStar Washington Hospital Center, and Susan Zannis, who is the vital stat supervisor in the Southern Nevada Health District. Shawna, would you like to begin, please?
Agenda Item: PANEL 1: Current Status for Vital Records: State and Local Data Producers
MS. WEBSTER: Thank you for providing NAPHSIS the opportunity to provide testimony at this important event. On behalf of the NAPHSIS board of directors and all of its members we are extremely grateful that the committee is investigating ways to improve and sustain the National Vital Statistics System as a whole.
My name is Shawna Webster. I am the executive director of NAPHSIS. We are the non-profit association of vital records offices in all 57 jurisdictions. The organization’s mission is to provide national leadership for both vital records and related information systems in order to establish and protect individual identity and improve population health.
I am here today to provide a high-level more national context from the point of view of the association. But I am joined by my colleagues and president elect of NAPHSIS, Richard McCoy, registrar out of the state of Vermont, who is here to speak from the jurisdiction’s perspective and provide some real-world examples to illustrate our points.
NAPHSIS has a 501C3 non-profit membership association plays an essential role in representing and advocating for vital records agencies. We not only promote the profession and the importance of the work our members do, but we also act as a convener, collaborator and a clearinghouse of standards, best practices and resources.
The NAPHSIS board of directors and robust committee structure guide the association’s work and inform our advocacy and education efforts. NAPHSIS works with its partners to educate policymakers about the importance of vital records, modernizing the system’s infrastructure, and to secure continued federal investment in these priorities.
One way NAPHSIS advocates for the importance of vital records data is by promoting the use of its various electronic systems like even STEVE. The services these systems provide local, state and federal government agencies help reduce identity theft and fraud. They also promote public health by providing timely and accurate delivery of jurisdiction data.
NAPHSIS developed these systems with and for its members, allowing the flexibility required for a federated system in which each state’s laws and statutes govern who can access the dataset, when and how. We have included more information about these systems in our written testimony. But for now, I will turn it over to my colleague to address your questions about legal challenges that are agencies-based.
MR. MCCOY: So, I am going to talk a little bit about two primary challenges, which are legal and financial. Then we will move onto a few other examples. First on the legal side, there are statuses in every state and jurisdiction that govern how the vital records shall be used, such as the birth and death certificates, and how the information shall be shared. Who may access that information and under what conditions.
The variances in those statutes make it particularly challenging for the jurisdictions to meet the ongoing request for their birth and death data. So for example, a state or a jurisdiction may have what is called open record statutes, which allow virtually any person to access those records and information with no restrictions, no questions asked. States or jurisdictions may also be closed records, and there are stringent restrictions in regards to who may access, utilize or even see those records and use that information. Then there are variations in between.
What is considered public information in one state may actually be restricted in another state. In Vermont, we have been open records. We will be closing very soon. But information such as cause of death has been open and available, whereas that is not available in our neighboring states.
So this level of variance creates several challenges and confusion for the potential users of birth and death data. It is difficult for the vital records staff who must protect that information. What is important for us to share with you is that there is a national model law that has revised every 10 to 15 years. This is a joint effort of NAPHSIS and NCHS. The states and jurisdictions are encouraged to adopt the model law since it does reflect best practices and standards for vital records administration.
However, it is a difficult process for vital records offices to navigate their state government to request or advocate for such legislation. This topic is often seen as a lower priority when compared to other public health legislation. So even though vital records and statistics impact public health, and are part of public safety, security and privacy, we have difficulty getting attention to pass a model law.
Now, adoption of a model law by health states and jurisdictions would bring increased opportunities for efficiencies and collaboration, while also still appropriately protecting personal information. This has been developed in concert with our membership and NCHS, but the model law has not been fully endorsed by HHS. This lack of support does further disable our agencies from affecting change with this important industry standard.
MS. WEBSTER: The NAPHSIS system allows the jurisdictions flexibility to participate or not as their state statutes allow, and facilitates legal use of the data by both government and non-government agencies alike. Still, the differences in state statutes, or the interpretation of those statutes, continues to block some of our jurisdictions from participating fully. This is a significant problem for the data users who want access to all 57 jurisdictions’ data, consistency in the laws and the implementation and interpretation of those laws would cause significant improvements to the completeness of the datasets and greatly facilitate access to the data.
Although vital records are the primary source for calculating health statistics, the jurisdictions are also responsible for registering vital events, providing certified copies and paternity acknowledgements, maintaining adoption records, preserving and securing vital records and preventing fraud and identity theft, as Tara just pointed out earlier. Rich is going to talk a little bit more about the financial challenges.
MR. MCCOY: So, I put these into four categories. I would be happy to talk more detail afterwards. I will just give you sort of a topline summary. So there are four categories, inadequate staffing, lack of support from IT services, lack of sustainable funding modernization, and inadequate or sometimes unpredictable funding streams.
So the past 20 years have been particularly challenging for vital records offices. They have been diligently implanting electronic registration systems, moving away from typewriters, 3×5 cards and the like, and going to electronic. But rarely is the money available to do these projects.
So these projects don’t require just software, but also staffing to implement and maintain those systems. The maintaining is often overlooked, as well. So first on staffing, there is a massive change happening in our world which is most of our staff have been at their jobs for 20, 30, even 40 years. I am aware of some who have been in the industry for 50 years.
A large number of retirements are starting to occur with approximately 20 to 30 percent turnover each year now. It creates opportunities for us to bring in newly trained, experienced staff with new skills and knowledge. But as you heard earlier, this isn’t really seen as a career as it used to be.
Also, some of our current staff who have been around a long time don’t have the public health informatics, project management or experience with newer analytical software packages. So it is critical for vital records to have the support it needs to hire new younger professionals coming out of school systems that can provide these sets of skills.
We also need new strategies to attract the staff to see this as a profession, as a career that has an impact. On that IT support topic, most vital records offices do not have dedicated IT support. They usually share it with the health department or other agencies. Vital records struggle to get attention for that level of support.
Additionally, many states’ IT departments have been focusing their resources on meeting meaningful use requirements. So bringing state laboratories, immunization registries, cancer registries and the like up to those standards of vital records has been overlooked.
Third, the sustainable funding topic, as states have modernized their state health care systems, including public health surveillance systems with a variety of federal grants, very little has trickled down to vital records. The enormous price tag of implementing electronic birth or death system can vary between 3 million and 5 million per state, per system. That does not vary based on the size of the state.
Development, as I mentioned, is part of the price tag. There is also the maintenance, which can go on for a life cycle of 10 to 20 years. That requires human and financial capital to support, as well as up rates as statutes change or as federal requirements change.
And then finally, the problematic funding streams, funding that goes to vital records offices from federal agencies has been level or has only seen minor increases over the last 15 years. Also, a portion of state general fund allocated to these offices almost never increases. In many states and jurisdictions, the fees collected for copies of certificates are a major part of their operational budget. But those fees rarely cover the full administrative cost.
In addition, we now have ongoing budget deficits in many states and jurisdictions, staff rescissions or vacancy freezes creating obstacles for our improvement of the vital records systems.
MS. WEBSTER: So all of these changes impact our membership’s ability to participate in data exchange and interoperability projects. As resources dwindle or maybe never actually increase, it becomes important for NAPHSIS and its members to participate in cross-organizational partnerships and workgroups like the Joint Public Health Informatics Task Force or JPHITF, which was just defunded a couple of months ago. These collaborations break down silos within the industry and are essential to all of our constituents who just can’t be at every meeting all of the time.
CDC is currently working to find additional funding for JPHITF, but support from the NCVHS community would really help solidify and extend the impact of this important group and our community’s ability to focus on interoperability, which as you know, greatly affects timeliness and access to the data.
MR. MCCOY: So finally, I am going to talk about data users. Vital records, and as you heard, vital statistics, which are two different areas, are essential to nearly all parts of the federal government, as well as state level government, advocacy groups, medical researchers, genealogists, non-profit and for-profit organizations and, of course, the public.
But without question, the largest users of the data is the federal government. And as the largest users, there does need to be adequate financial support to ensure that these data will continue to be available in a timely manner and provide accurate and complete information.
So I am going to talk briefly about the death certificates, since that is one of the most sough-after pieces of information. It is critically important to know when someone has died, so that federal and state benefits can be concluded, but also to ensure their identity cannot be used to commit fraud, and to also of course provide timely benefits to next of kin.
This creates a critical or ongoing need for updates of death records to the Social Security Administration, the Office of Personnel Management, the Internal Revenue Service, the Department of Veteran’s Affairs, Department of Defense, the Centers for Disease Control and the Centers for Medicare and Medicaid just to name a few. I could give a list of at least a dozen more.
Unfortunately, there has been some issues in the past, such as in the Social Security Act, which allows SSA to share our jurisdictions data with several other enormous benefits-paying agencies like CMS for free. They should not be allowed to happen without adequate remuneration to the states.
In addition to the need for timely and accurate death certificate information, there is also the need for birth certificate data. The Centers for Disease Control are the primary user for that information, and uses it for purposes of tracking the health of our newborns, mothers, but also some of the data is used for real-time surveillance, such as monitoring for indications of ZIKA-related birth defects or monitoring flu-related deaths.
Outside of the federal government, there are hundreds, if not thousands, of organizations that rely on both the birth and death certificate data. Their access and use of these data are often detrimentally impacted by our limited resources, technology and of course our statutes. In some states, our agencies can charge for the time to respond to a request, but many states, they cannot. As those requests increase, there is no corresponding increase in staffing.
The most common scenario I am going to give you is the following. In the vital records office that cannot charge for its time, but its request from a sister agency, such as the state’s Department of Motor Vehicles. DMV is an excellent example, where federal requirements on driver bureaus to validate birth certificate documents is creating pressure on the vital records office, but without any funding support to the vital records staff that is necessary to meet that demand.
There are other examples I could give you, but for reasons of time, I will skip over this. I just want to say as an example, state-level, our vital records office is required to provide data to the immunization registry, cancer registry, birth defects registry, the advanced directives registry, organ donor registry, prescription monitoring system, PRAMs, I could go on. None of those programs provide any funding to the vital records office.
MS. WEBSTER: So in conclusion, I just wanted to say that NAPHSIS and its membership has built the infrastructure to deliver the data and the right pricing structure, to be the solutions to many of the problems we are going to be discussing over the next two days. To support the National Vital Statistics System, federal and state agencies should use these systems NAPHSIS members have developed. We ask that the committee strongly support a mandate for federal agencies to use these systems and pay the appropriate fees for that use.
And we ask that the committee advocate for more consistent legal statues that will allow jurisdictions to participate fully. With this leverage, the National Vital Statistics System will be able to quickly evolve into the robust, timely, integrated, sustainable, multi-purpose information system that we all want and need. Thank you for your time and your consideration.
DR. COHEN: Thank you for that and somewhat, in my opinion, a chilling testimony. It is an important perspective for us to have. That was a general overview around state and local issues. We are going to get a little more specific now with respect to births, hearing from Angela and Susan. Angela?
MS. SHAW: Good morning. I am the birth and death registrar for MedStar Washington Hospital Center here in DC. We will start with the birth. We have the clinician staff document all data related information to the birth in our official electronic medical record.
I have the privilege to go up and meet with the parents and speak with them in reference to completing the mother’s worksheet for the child’s birth certificate, which is in English and Spanish. Health information management abstracts the data from the medical record and the mother’s worksheet, and we enter it into the electronic registration birth system.
With our deaths, the clinical staff document all data related information to the death in the medical record. Our clinician completes the patient death fact sheet to certify the death information. Health information management abstract the death information from the medical record in the patient death fact sheet into the electronic death registration system.
HIM then sends the death information death certificate to the clinician for electronic signature of the certificate. After the certificate is signed by the clinician, HIM, we then send the certificate to the funeral home through the EDRS system. In addition, we have some concerns about EDRS and EBRS.
We would like to have additional diagnosis and procedures added to the EBRS dropdown menu like anemia, GVS, genital HSV active and tubal litigation. We would also like to add spellcheck to both features in the module. We would also like the staff to be able to add clinicians to the modules.
We would like the staff to be able to rest passwords for the clinicians and to be able to modify the cause of death recommended by the Office of Chief Medical Examiner’s Office. That is it for us.
MS. ZANNIS: Good morning. My name is Susan Zannis. I am with the Southern Nevada Health District out of Las Vegas, Nevada. So coming into this position, what I recognized the most was that there is a huge lack of understanding about why we do what we do. It is easy to come in and make sure that all the boxes are checked. But it is more difficult to understand why those boxes are checked and how they serve the public, and how they help us to ensure better health within our district.
So in order to help my staff understand what we do, we came up with a vision and mission statement. It goes a long with a lot of what Tara said and along with what has been said already to provide quality customer service through accurate and secure documentation of life events. The other part I wanted to read was to contribute positively to the efforts for a healthy Southern Nevada by ensuring we collect all essential data to be used as a foundation for improving public health.
I think it is important for us who are doing the job to understand why we are doing the job. That gives us the passion behind why we do what we do. To just come in and sign your name is pretty easy, right?
Clark County serves more than 77 percent of Nevada, so we register the majority of the births in Clark County. And so for me going into this, the best practice is education and training. It is education and training for all of our birth clerks, all of our death registrars. For all of the hospital clerks and registrars, there is a huge amount of turnover in Clark County with registrars, birth clerks. And so, the information again about why we do what we do gets lost. You just come in, and we push the button again.
What do we do when we are registering births? We examine the birth certificates. Number one, is everything complete? Does it make sense? Was this baby born at 22 pounds? Or was this baby born at 12 pounds? I mean, you look at those differences because they make a difference when that comes out to NAPHSIS and they look at things. There is a huge margin of error, so caring about what we do matters.
And then the documents that go with it, are they the correct documents? Is there a minor father who has the correct paperwork signed? Are they of age? Do they have the parental consent? The declaration of paternity, do we have the correct signatures? Have they been read their legal rights before they sign these documents? Do we have the correct dates? Do they match up?
And now, we are working in Nevada on a declaration of parentage, which is going to be used for surrogacy issues. My concern there is are we capturing the birth mother’s information? Did the birth mother smoke? Or are we getting the other mother’s information because that corrupts here data. We need to make sure that we are getting the correct information when that happens. And then in Nevada, we have a 10-day registration period. So we try to do a report and then let the hospitals know where they are in that reporting stage.
This just kinda goes into the different fields that we look at, education, race, ethnicity, marital status. Is the father there? Were they married? Is he allowed to be on the birth certificate with the correct documentation? It goes into the labor and delivery. Then the different anomalies that can happen, and we use those for research and studies.
Are all the signatures there? Do we have all the proper signatures to make sure that this record is ready and complete and can be signed? This is just an example of some of the documents that we have to look at. We look at legal documents all day long. There is a lack of attention when it comes to vital records. It is easy to think that we are just paper-pushers. This is not an easy job. Understanding all of these documents, and are they relative to the record, there is a learning curve to it. You cannot just walk into this department and go for it.
So we act as the collector for all of the records in Clark County. This just touches again, we provide the oral and written instruction for the declaration of paternities. We do that in our office. So parents come in, that is not easy. They come in. They are fighting. They are not agreeing. How do you deal with that when they come into your office, and they are not real sure they want to sign these documents?
One of the main things that we need to do is maintain confidentiality, protect the integrity of our records, so that we can ensure that they are not being fraudently used. Okay, I will hurry. So how do we do that? We do background checks on the people we hire. We do annually signed confidentiality agreements. We have key card access to all of our areas’ cameras. We have a secured vault for our security papers, so that not everybody has access to it. And we have paper tracking systems.
And for our applicants, we have the NRS statutes we follow. It is somebody with direct and tangible interest. So what do we do to find that? Everybody that comes in to request a record must have proof. If you are the grandmother, you had better have your child’s certificate to show that you are her mother.
And data sharing, so we send everybody to the state, but then we also get stuff back, so that we can do our own QI in reporting, so that we have stuff at a county level. These are just some uses we have used, birth certificates, perinatal hepatitis B, teen pregnancy, tuberculosis exposure and notification, birth defects, media, public information, sexual health, HIV and syphilis. That is it.
DR. COHEN: Thank you all so much. This was a wonderful panel. We are going to take a 10-minute break. We are funning a little behind. Hopefully, we will be able to make it up throughout the day. We will come back, and we will talk about deaths and IT-related issues. We will open it up for committee questions and then for general discussion, and then have our lunch break. Please be back in 10 minutes.
(Brief recess)
DR. COHEN: Continuing on the local data collection theme, we have four presenters, Rob Moore, the past president of the National Funeral Directors Association, John Fudenberg, Clark County coroner and former president of the International Association of Coroners and Medical Examiners. Kari Guida from the Minnesota Department of Health, where she is a senior health informatician, and Linette Scott hopefully will be calling in from California, where she is the chief medical information officer for the California State Department of Health Care Services. Rob, take it away.
MR. MOORE: Thank you. Thank you for this opportunity to present some testimony on the critical role that funeral directors play in the vital statistics system. As a funeral director myself who has witnessed the initial changeover from paper death certificates to our current electronic system, I know how important it is to stay current and up to date with technology. I want to thank the NCVHS for hosting this hearing, so that we can maybe move everything forward again. I appreciate that.
It is not secrete that funeral directors are primarily a part of the death certificate component of the vital statistics system. We function as the conduit between the public and the jurisdictions information gathering. We gather and input the necessary data into the EDRS system to complete the parts of the death certificate that are required, other than typically the medical and registrar sections.
So the death certificate process is similar in most jurisdictions, but as you have heard from some of the other testifiers this morning, it can vary from jurisdiction to jurisdiction in the type of information that is asked on the death certificate, and also how we do file. So with that in mind, what I would like to do, I think it would be of value to the committee today is just to give you a typical case flow from the funeral home perspective. I am from New Jersey, so it is going to be a little bit New Jersey-centered on what we do there.
Everything obviously happens when the death takes place. In New Jersey, the EDRS case is initiated most often by the health care staff facility. That would be at a hospital, hospice or nursing center if the nurse is an attended death. It can also be initiated electronically by the medical examiner or coroner if it is an unattended death, such as an accident or homicide. And occasionally by the funeral director itself, and I think you heard from Tara that in some states, the funeral director does it mostly. In New Jersey, it is not that way. It is kind of rare that we actually start the case electronically. I could give you some examples, but I want to be time sensitive.
So there are separate sections of the EDRS case that need to be completed. Have you heard about the demographic material or statistical information that I call it that is completed by the funeral director. You have the medical data, which in New Jersey, there is a pronouncement part for the pronouncing physician. And then there is a certification of the cause by another physician, which could be the same person also. And then, of course, you have the registrar section, too.
So what do funeral directors do typically? We conduct an arrangement conference with the next of kin or responsible party. While we are helping them plan a meaningful ceremony to acknowledge the life of their loved one that has passed away, we gather that data that we need to complete the death certificate at the same time. And you have seen the different things, birthdate, place of birth, social security, all the little bits of information that are asked for.
We typically then go onto talk to the family and explain the need for certified copies of the death certificate. I think this crowd understands the need for those. We will come to a decision on a number of copies that they might need, and then we will conclude our arrangement conference at that point.
So once we are done, then we would sit down and log into the EDRS system. In New Jersey, you participate in that with a user name and password, which has to be renewed every 90 days. We log into electronically at our computer, and then we put in all that information that is asked for. Many times, at that point, the case is in the system, but it has not been medically certified. You might have the pronouncers push and done. We will go in. We will put our portion in of the background information.
Then we kind of play a little bit of a waiting game when we wait for the doctor to come back and put in the cause of death. In New Jersey, they have 24 hours to do that. I know it is different in different states. Once the doctor does do that, he or she gets in there and puts in that cause of death. We get an email in New Jersey, that will say that the pronouncement portion has been filled in. At that point, we would log back in with our user name and password again, and finish the case, so to speak. We would assign it to the jurisdiction that would be filing it.
You have a couple of options in New Jersey. I would say about 99 percent of the time, we will file the case with the local municipality jurisdiction that our funeral home is located in. At that point, one of the most important parts for the funeral home side is once the case has been accepted and finalized by the registrar’s office, we are allowed at that point to print out in our office what we call the disposition permit. That is the document that needs to accompany the remains to the cemetery or to the crematory if the person is being flown on a common carrier to another location around the country, that piece of paper accompanies the remains until its final place of disposition.
At that point, we can order the copies that the family has requested. They come in a couple of different ways, short forms, long forms, cause of death, no cause of death, all the different things. I will be way under three minutes, thank you.
Then what happens is with the EDRS system in New Jersey, every funeral home has to have a bank account that is linked to the system. It is almost like EZ Pass for paying tolls. They take the funds based upon the number of copies that we are purchasing for the family out of the system, and then that would be something that we would include on the family statement as a case advance or cash distribution type of item.
So once we do that, we are allowed to get our actual paper certificates locally. We do that at our firm as a courtesy to the families because they are always very anxious to get their certified copies. You can get them mailed from the state, too, but we get them locally because it is a speedier process.
And then, pretty much that is it. We are done with the system. We don’t really use the data beyond getting that. We would perhaps go back into a case if there was a correction that might need to be done. Unfortunately, corrections are still paper in New Jersey as our fetal death certificates. But once we accomplish that and get the disposition permit and the certified copies, our role pretty much ends within the vital statistics system. So that is pretty much the short and sweet version. I will be happy to answer questions as we get closer.
MR. FUDENBERG: Good morning. My name is John Fudenberg. I am the coroner in Clark County, which is also in the Las Vegas area. I am also here representing the International Association of Coroners and Medical Examiners. I will take his leftover two minutes because I talk a lot. Thank goodness I didn’t have to disclose my conflicts because I have all kinds of conflicts and issues and everything going on. I might need to see somebody here later.
First of all, I am honored to be presenting in front of this group. More importantly, I am very grateful that you are included coroners and medical examiners in this process, as I think they are oftentimes overlooked when we are discussing things at this level, so thank you for that.
I think my responsibility is to give you a very general high-level overview of the medical legal profession. And when I saw medical legal profession, some of you don’t know I am referring to coroner and medical examiner systems. With that being said, I will tell you a little bit about who we are. We are predominantly coroners and medical examiners throughout the United States. There are nearly 2400 systems in the United States. Around 2000 of those systems are coroner systems. The remainder are medical examiner systems.
I think it is important to know the difference between coroners and medical examiner systems. Typically, it means who is in charge of the office. Typically, coroner systems are non-physicians. Nearly 82 percent of them are elected in the country. The rest are appointed. Oftentimes, they don’t have to have qualifications to be a coroner. More and more, states are adopting standards and training and education for coroners as prerequisites for running for the office. I can say I am not elected. Otherwise, I wouldn’t work there. I wouldn’t deal with that madness.
I think in addition to certifying death, which is why we are here, I think again it is important for you to know what we do. We investigate as a general rule all unnatural deaths. So we investigate all homicides, suicides, accidents, which are predominantly made up of motor vehicle accidents and, of course, drug overdose, which is the big issue nowadays, and which is I imagine why a lot of us are sitting in this room today.
We are also the organizations that identified the decedents in most jurisdictions. We are the organizations that notify the next of kin in these cases. So we are the people that are knocking on the door at 2:00 am notifying a mother that her 17-year old daughter died in a motor vehicle accident, which I think is very telling to what we do. I think most of us would consider ourselves being in the business of life rather than death because we are really there to serve the families, as Tara mentioned, during one of the most difficult times of our lives.
So when we talk about stress and vital statistics employees, we work hand-in-hand with our Office of Vital Statistics. And I know that they deal with a lot of stress. But I think we can multiply that by about 100 when we are out there notifying a mother that a young daughter died in the middle of the night. So I think that is important for you all to know.
Let me say something, a little bit about our data. I am not going to get into the where we fall in that process. I think that is very clear. Robert mentioned that. We sign the death certificates. We certify the deaths in most cases that fall under the jurisdiction of the coroner and medical examiner, so that is our role.
Most of us have case management systems in our office. Many of us do not. Most states have EDRS, which I believe everybody in here knows what EDRS is. And I can tell you that the majority of our case management systems do not integrate with the EDRS, which is problem number one.
Furthermore, the data that we collect, I think it is safe to say, and I think most people in this room would agree, that our data is like gold. We have hundreds of agencies requesting our data, more and more every day with the opioid epidemic rising in popularity. I think again it is safe to say we have a couple of dozen federal agencies that rely on our data. I can tell you our office has nearly 135 reports that we send out on a weekly and monthly basis.
We are very fortunate. We are somewhat well-funded compared to most coroner and medical examiner offices. But most are not, and most do not have any reporting capability. They may keep their data on spreadsheets at best. I think if there is a message that I could convey to all of you is that we in the medical legal profession need help. We need help in our systems. We have a whole lot of other things that we need help with.
But as it relates to this group, I think the number one thing is that we need help with our systems. We need to develop some standards in how our data is collected, how it is reported. And I think that is a very easy lift for us because there are very few vendors that deal with coroner and medical examiner case management systems.
If we could get to those vendors, have them develop some standards, if we could encourage offices through the health departments or whomever governs their agencies to report that data consistently and share that data consistently, I think that we would be in a much better place. I don’t know how many minutes I have left. I am trying to talk as fast as I can.
The last thing that I will say is I heard a couple of people, I was going to say whining about funding, but they weren’t whining, I am going to whine about funding. I think that the medical legal profession is probably the most underfunded forensic discipline by the federal government, and the state and local governments. We are usually not funded well. That explains why we are where we are in our data systems.
So if I could ask anything, I would ask all of you to look at that and look at the grant money that is going to the medical legal profession compared to the health departments, compared to the other forensic science disciplines. I think you will find that it is miniscule at best. That is where we need help. I think that is where we can be good partners, better partners to all these federal agencies in getting them the data that they use and that is so valuable. And again, I have a lot more to say, but I will end now. I am available for any questions. I could go many different places, but I will save it for some of the questions later. Thank you.
DR. COHEN: Thank you very much. We are going to sort of switch from death-related issues to IT.
D) State/Jurisdictional Health Information Technology and Information Technology
MS. GUIDA: It is not very often that IT is uplifting compared to the previous presentation. Thank you for that order. Before I begin, I just want to recognize some of my colleagues at the Minnesota Department of Health who have supported me in my work and in the projects that I am going to talk about, Marty LaVenture, Jennifer Fritz, Sally Almond, Molly Crawford, the Minnesota e-Birth Records Project Team, as well as my Office of Health Information Technology.
As my title implies, I am going to use a public health informatics lens for my comments today. And also, they will be based in part on the work that Minnesota did on a project called the Minnesota e-Birth Records Project, which assessed the readiness for e-Birth records. That was funded by NCHS. It was also submitted, if you want the full report, as part of my testimony.
So this is how in Minnesota, and I imagine it is a similar process throughout the US, our birth records information is collected, used and shared. Briefly, it goes from our prenatal care provider to the birthing hospital, and then it ends up at the Minnesota Department of Health. From there, it goes many places.
It is used for our birth record certificate. It is sent here to NCHS. We also see it being exchanged with other state and territories. It goes to external programs, such as local public health in the school districts. And it is used with an MDH, the Minnesota Department of Health, for such programs as immunizations, PRAMs, infectious disease.
All of this information in this flow of information is done within the tightly held privacy laws that we have in Minnesota. The Minnesota Health Records Act is more restrictive than HIPAA, as well as what is in statute. Yes, more restrictive than HIPAA.
This shows not only the flow of our birth records information, but it really shows how foundational birth records information, as well as death records information, is to our public health work. We hear again and again within the Minnesota Department of Health that the Office of Vital Records is that first domino that needs to fall, so we can have a truly interoperable health department within and externally with all of our partners. As you look forward, I really encourage you to consider how information needs to be used by our internal and external partners, so we can really meet our community’s needs looking towards the future.
Our e-Birth Records Project actually just focused on the two middle blue circles, the flow of information between the birthing center and the Minnesota Department of Health. And what we found, and some of you have alluded to, is that just collecting that information is very complex. So this is at a high level, the 17 steps that happen within a hospital to register a birth. I am not going to walk through all 17 steps because I was given eight minutes.
But I will point out a few things. This process collects both medical and civil information. It collects information with the 58 or so concepts or questions for the US birth certificate, as well as some Minnesota-specific data. All states have some very specific state questions. So like we had a project going on about Hep B, so we have some Hep B questions.
The information flows from a variety of sources, paper, electronic health record and the mom. We also find that a lot of the information is entered into the electronic health record, and then it is re-entered into our web-based reporting system, MRNC. So retyped information is put in there.
And often, the prenatal care record, which is kind of the paper record that floats. It is the orange circle all the way around. It doesn’t actually get entered into the ERH until after discharge. And it is usually a scanned document. So it is not very easy querable.
I would like to point out when we share this with our partners, a lot of them point out that although this is really important, none of this is actual care. I mean, this is work that the hospital is doing, as well as doing care. And as was pointed out to me when we were interviewing hospitals to document this process, the people who are entering the information into MRNC are also in charge of making sure that no one walks out with babies and checking bags. This is just one of their many activities they are doing. And in Minnesota, this happens bout 70,000 time a year. I imagine this process happens probably several hundred or thousand times a year across the country.
So as you guys look towards NextGen vital statistics, we have to address this complexity, as well as many other complexities. I would argue that one of the best ways to do it is to look at implementing e-health standards. This isn’t an e-health standards subcommittee. I won’t go into a lot of detail about the standards and the technicalities. But I really want to talk about what are some high-level things we can do when we look at e-health standards and policies for interoperability.
What we need to see is hospitals, jurisdictional and NAPHSIS policies must be aligned to support the implementation use and maintenance of all e-health standards for both internal and external interoperability. I would also argue and recognize that this cannot be done without funding for public health departments, as well. It is not just the individual Office of Health Vital Statistics.
We also need to see resources and technical assistance for our Office of Vital Records hospitals and prenatal care providers. I believe prenatal care providers are often left out of these discussions. And they have that first source of information about the mom and the health of the pregnancy up to that point, so really thinking about how we can get that information incorporated.
And then finally looking at continuing to build the informatics capacity at offices of vital records. I believe it is being incorporated. We need to continue to show that it can be part of the work, not another piece of work.
Also, we need to start thinking about 2030. We have to start talking about 2050. We need to really consider the who/what/where/when and why for our collection, use and sharing of the information birth and death in the future. We need to figure out what do we wish we were collecting in 2030 to make good decisions. We need to start figuring out what is it in 2050 we want to be looking at and understanding because we have to start collecting that in 2030.
We also must continue to figure out how we can support public health in the country as a whole to deal with the direct and indirect implications of climate change, air pollution, extreme heat, the floods and droughts, as well as our ecosystem threats. And finally, and most importantly, I urge you, as you look at this work, to really think about how we can use advanced health equity. Vital statistics is a tool to advanced health equity, but the lack of excess to this information creates disparities in our communities that we serve. So we need to be able to provide timely actionable information to communities to advance health equity.
I should have a bullet actually before that, that we need to actually ask those communities what they want. I think we need to rethink about and redefine what the word community means, and start reaching out to different types of communities.
And then also, I think we need to look at the disparities that are happening between our offices of vital records. When we hear about the inconsistencies and funding, as well as policies, we are going to see disparities that are going to filter down from our offices to our Departments of Health and to the communities that we serve.
So I look forward to the work and findings of this committee. I will be working in 2030, hopefully 2050, but it is possible. I hope to see that continued work on e-Health standards, planning for the future and the advancement of health equity. Thank you for your time.
DR. COHEN: Thank you very much. That is wonderful. Linette, are you on the phone?
DR. SCOTT: I am. Can you hear me okay? Thank you for this opportunity to respond to the question to provide testimony today. My name is Linette Scott. As Brice had mentioned, I am the chief medical information officer at the Department of Health Care Services, which in California runs our Medicaid program.
In a previous position, I also worked very closely with vital records. I served as the state registrar with our Department of Public Health, and also worked very closely with NAPHSIS on various committees and as a board member. I bring a past experience with vital records and a current experience in working with vital records as part of the Medicaid program.
I have supplied written testimony that has additional detail and, in particular, links to a variety of the things that I will touch on. But I will provide highlights during the short seven additional minutes now.
So as I mentioned, the Department of Health Care Services runs our Medicaid program, which in California we call MediCal. We have approximately 13 million beneficiaries and a very complex and diverse population, as well as a diverse health care environment. We primarily are a managed care organization with about 11 million of our members in managed care. But we still have about 2 million members in fee for service environments. We support a variety of models of managed care within California.
In the various business processes and activities that we do in MediCal, we are dependent on vital records data in a number of different ways. To provide just some examples of those, in terms of enrollment and eligibility, both for our members and our providers, birth records come into play in terms of identifying, as was mentioned previously, civil purposes in terms of identifying individuals and identifying the types of benefits that they are eligible for.
Also in terms of our providers, making sure that we disenroll, as well as enroll. Providers, there is significant concern with having individuals receiving services or providing services if they are already deceased. That is an area that gets a significant amount of focus from a fraud and abuse perspective, as was mentioned earlier with other presenters.
Another type of arena that we are very dependent on in terms of using vital records data relates to looking at quality outcomes. I think this was mentioned earlier with other presenters. When we look at how well the health care service is performing, and what kind of services our members are receiving, quality of care is receiving a significant focus. And there are a number of kinds of quality measures that we can only perform when we are using vital records data.
So I have included in my testimony some links to the core set of measures that were identified for the Centers for Medicare and Medicaid services. These core measures look at things like live births weighing less than 2500 grams, prenatal and postpartum care, looking at timeliness of prenatal care and postpartum care occurring.
And these measures, because of the interaction between mom’s data and baby’s data, and understanding when the baby’s date of birth is in conjunction with the mom’s claims data, birth records can become very, very valuable in terms of looking at these. We are very focused on that. So in California, we have about 500 to 550,000 births each year. About half of those are paid for by MediCal. So we are very focused on quality of care. So those are just some examples in terms of how vital records data comes into play in MediCal and are part of the core set of measures that are supported and requested by our primary funder center for Medicare and Medicaid services.
One of the things I wanted to highlight though is the importance of national standards as we do this work, both in the vital records, as well as more broadly. In Medicaid, there are a number of ways that we are focused on the use of national standards. And I have highlighted those in my testimony in terms of the Medicaid Information Technology Architecture initiative, HIPAA, which was mentioned earlier, again HIPAA, while people think of it as for privacy, it really focuses on the use of standardized code sets to streamline administrative processes. And our Transformed Medicaid Statistical Information System, which is how we, as Medicaid programs, report to CMS.
Similarly, the state and jurisdictional vital records report to the National Center for Health Statistics, however on the Medicaid side, it is required by law because we have received federal funds for Medicaid programs we provide this data back. But part of that is to establish those national standards, so that you can have comparability in terms of how you talk about performance between states, and nationally. I think that is where the similarity really lies.
So national standards are critically important, otherwise, we can’t have state-to-state comparisons or national comparisons, in terms of understanding the health of our communities and our populations. We also then can’t assess how the quality of performance is in terms of our services that we are providing both in terms of communities and in the health care industry.
As the Health Information Technology Coordinator in California, as part of our Medicaid program, I am responsible for our EHR electronic health record incentive program in the Medicaid program, as well as working with many of our partners throughout the health information technology arena. In that capacity, I do want to highlight as was mentioned earlier that the work done with NAPHSIS and the National Center for Health Statistics related to establishing standards for birth and death data to be captured in electronic health record.
The focus on how we integrate our systems is both vital records, public health, and other administrative data with the data that is being collected in electronic health records gives us an opportunity to look at decreasing the burden of reporting for providers. So we want to be able to focus on that, as well. There has been initial work done. Many of my other colleagues on this panel have been a part of that and have been leaders in that. But there is still a lot of work to do.
So even though we have started that initial process of looking to the electronic health records to help support, in particular, I think the clinical data and health data, there is still very much a legal and civil responsibility around the information on birth and death records for which the rigor is not part of the electronic health record. So we need to look at how to fully integrate those two functions and how to again look at how we streamline the burden in terms of reporting.
As Carrie showed in her slides with the many lines going between lots of different places, reporting burden is a significant issue that we need to address because the higher the burden, the lower the data quality just because of the number of touch points. That is something that we will want to look at as we move forward and think about the next vital statistics system.
In terms of the kind of final thoughts here, the Department of Public Health is again where our state registrar lies in California. We have been working very closely with them, as was mentioned earlier. The importance of this data in terms of having complete data is critical.
And so we have been working very closely with our Department of Public Health to create a pathway for receiving the real-time data for both birth and death data. We had requested and received approval from Centers for Medicare and Medicaid Services to help support that with funding. And so, we also are contributing to that. As was mentioned, vital records often does not receive state general funds, but is supported by fees from certificates and funding provided by data users. And so that was part of our reason for requesting that funding through our federal partner to help support that because of the need for real-time data.
In terms of the protection of data, as was mentioned, we are a HIPAA-covered entity as the Medicaid program. But California also has specific laws related to vital records. So as we receive vital record data, we make sure that any of our staff within the department that are using that data sign acknowledgements to be responsive to the specific requirements of California state law.
We also have been working closely with our Health and Human Service Agency and Department of Public Health in relation to how we release data publicly and have developed a set of data de-identification guidelines that we are commonly using to help create a common language around that, when we move into the public arena as opposed to the confidential arena. We have been working on how we share data publicly through our agency’s open data portal, as well.
So in conclusion, vital records infrastructure and funding have significant variability and complexity. However, the work done to date through partnership between states and jurisdictions and the National Center for Health Statistics had resulted in datasets that form the basis of a multitude of activities and have data quality and consistency that is a goal for many administrative data systems. Continued support for this foundational data and the state registration processes is essential as we continue to work to transform the health care system and improve health outcomes for our constituents. Thank you for the opportunity to testify. I look forward to continuing conversation as we move into questions and answers.
DR. COHEN: Thank you for dialing in and providing those comments. Next, we will take a round of questions from committee members. Then we will open it up for general discussion before lunch.
DR. ROSS: There seems to be this overriding theme of a need to pay for play, both within the states and at the national level. So first, on the federal level, are you aware has there ever been a study that enumerated all of the federal uses of these data?
MS. WEBSTER: Not to my knowledge. But we have compiled quite a list, as Rich had said. I mean, we could, off the top of our heads, name probably 20 or 30 right now. And we also have a list of the organizations that are receiving the full DMF, the death master file, from SSA, as well.
DR. ROSS: Am I right that one of the things you were saying is those who benefit from the process of local and state collection of these data at the federal level, those who benefit, aren’t really paying their part of the way?
MS. WEBSTER: That is correct. I mean, I think it really has fallen on NCHS and to, maybe a lesser extent, SSA, who we have worked very cooperatively with over the past 10, 20 years or so, to help sustain this enormous system as it evolves into modernization and electronic systems. But not just them, not just NCHS and SSA, but also on the backs of moms and dads who need their kids’ birth certificates to get access to health and other social services. It is not right. It is just not right. It is a system that is developed out of necessity, but it needs to change.
DR. ROSS: Also, part of that theme, some of you mentioned, if I picked it up right, that you think some kind of fee-based reimbursement by the data users, it would be useful. Are there any examples of where that is done and how that works?
MS. WEBSTER: Sure, there are a lot both at the state level, as Rich can speak to. All of our jurisdictions have sort of one-off contracts with either other state public health programs or other non-public health, but other programs in the state that use vital records data like the DMVs, for instance. But at the federal level, the SSA is a great example of a well-formed and well-established partnership that pays for use of the data. It is just the reuse of the data that we have a problem with. I think that model can be reused with any federal agency.
MR. MCCOY: I should add, so on the SSA case, NAPHSIS has worked with its membership to come up with a fair pricing model based on what the administrative costs were at the jurisdictional level. So we provided that information to SSA to show here is how the jurisdictional costs break out, and why we feel this is a fair reimbursement.
DR. COHEN: Another quick comment, with SSA, there is remuneration at birth where the states get, I don’t know what, it is a little over $3.00 now or was $3.00 per birth to enumerate and get Social Security numbers assigned to newborns. The cost studies show that, the way I did it when I was a kid, having to go to Social Security offices, would cost between $15.00 and $20.00, so it is a huge potential savings by using vital statistics that way. Linda?
MS. KLOSS: Carrie, I wanted to follow up on your graphic, the flow process. Has your work taken you to the next step to look at working with any organizations to optimize that flow? What has happened as a result of spotlighting the complexity?
MS. GUIDA: So that was actually just like a first step of the work we did. In the report that I submitted that you will see, we looked at what that workflow would look like if we implemented the e-Birth records we were testing at the time. The picture then went from 17 high-level steps to about 8 high-level steps. You really saw much more, both the mom and the EHR at the center of that instead of on the peripheral view of it.
We really identified and gave some good feedback on what some of the next steps need to be for that development. We went to the Connectathon two years. We did go way beyond that. We have seen some of that work implemented just into the workflow within our local public health department.
MS. KLOSS: Did you have an opportunity to cost out what could be saved if we went from 17 steps to 8 steps in this process that gets repeated millions of times a year?
MS. GUIDA: We did not. But in Minnesota, we have an advisory committee, an e-Health advisory committee, which has a lot of CEOs of our health systems and our payers. When they saw that figure, not the money figure, the actual figure, they were very excited at just seeing an actual return on investment in using the EHR. They could see it. They didn’t enumerate it, but they could see it.
DR. MAYS: Thank you for a set of great presentations. As I said, they raise so many questions. My first question I want to start is was raised by your presentation, Susan. When we think about the birth record, we often are thinking about how to collect the data from the mom and kind of the accuracy of the mom’s data.
But what I am starting to see is a lot of problems in terms of the father. Issues of paternity, signing a birth certificate, then late in the heat of the moment, they want to be a wonderful partner, and then later it really wasn’t their kid. Then they are in a state where they can’t remove their name, but yet have to say support, so a lot of really complex issues that, as I am starting to learn about men’s participation in this process.
So I am trying to understand what would you recommend in order to make sure that for a man, when he signs this or he participates, that there is enough health literacy, that he understands the legal issues, is well informed of his rights, how would we do that? I mean, you don’t have like tons of time. So do you have some thoughts about how we can do this better?
MS. ZANNIS: I can tell you that in Clark County, the education to fathers was not being done for a long time. They were handed the document in the hospital, or it was right out there where they could print it out. We didn’t know who was signing it. That has been tightened down over the last year. Now, the document is a confidential document that is held at the hospital or at our office.
And so, the parents have to come in, and part of that is going over their legal rights. That is a federal mandate actually. When you do a paternity, you are supposed to read them. They are supposed to have it in writing, as well as orally or audio.
DR. MAYS: So every state is to do that?
MS. ZANNIS: Yes. But is that happening, probably not because who is monitoring that? I know in Nevada, nobody was monitoring that until it came to light that we were missing that stuff. I think that there needs to be some emphasis on that from the federal level that this is a mandate. It is required.
We need to have somebody monitoring that these hospitals are doing it because it is real easy to walk in a rush, and you have 30 babies born that day. You are going to hand them the paperwork and come back later, and it is signed. Well, who signed it? I think that we need to have somebody monitoring. I think that we need to have somebody monitoring.
DR. MAYS: Well, monitoring, and also maybe research on what are the most effective practices to make sure that the participant understands and is consenting in a way in which it is a understandable consent.
My second question is for those of you who are funeral directors, coroners, et cetera. I want to talk a little bit about the narratives that you often have to participate in and the ways in which funeral directors, for example, get information. Can you talk about what you would need in order to do that better?
Like narratives, I am one who studies them. In these narratives, there is a standard section, but it is not standard training in terms of the quality of what we get in there. I don’t know what would make that better. I guess the same in terms of funeral directors.
I think we have all, at least at this age, for some of us, have enough experience with death that you watch, for example, how people ask information. They may skip asking about race because it doesn’t seem like it is polite to do that. So I am trying to get a sense of how can the standard of what you collect become consistent, higher quality and not cost us a ton of money, but in some way, give you what you need to do your job better.
DR. FUNDENBERG: I am not exactly sure. I think that we are working. There are many different groups right now that are working on standards as they relate to the medical legal profession. I think that a lot of that can be accomplished through those standards groups. The SWIGMDI, the Scientific Working Group on Medical Legal Death Investigation was a standard-setting group. Don’t write that down, Dave, because that group has gone away.
I say that because it is transitioned to now what is referred to the NIST OSAC Medical Legal Subcommittee. NIST is the National Institute for Standards and Technology, I think. OSAC is the Organization of Scientific Area Committee. They have 21 or 22 subcommittees that each are represented by forensic science discipline, one of which is the medical legal subcommittee. I think some standards can be developed within those groups.
The other thing I think that I wanted to bring to Dave’s attention in response to his question is I believe you asked who uses the data. And we have a lot of overlap in our data clearly. There is a group that was convened by the White House, the National Science and Technology Council of the White House, I think. You know the layout of the federal government a lot better than I do.
They generated a report called Strengthening the Medical Legal Death Investigation System and Improving Data Systems. I would recommend you review that report. They outline a lot of the detailed data that different federal agencies use and why they use them. It really stresses the importance of those datasets. I think that is a very valuable resource, and it is brand new, so it is fresh.
Dr. Margaret Warner, who is in the room here, she co-chaired that committee. She really has a good understanding of both the vital statistics world and the medical legal world. I think she put together a very good report. That would be a resource. Vickie, I am not sure if I answered your question.
MR. MOORE: We are kind of at the mercy of the people that are supplying us the data. There will be cases where it might be a spouse who is coming in to make arrangements for their partner or their husband or their wife. In those scenarios, you are pretty confident that the information that you are getting is relatively good.
But there are many times where the person who is passed away could be indigent or could not have that person in their life who can come forward with that information. So there are many times where we are at a loss to even complete what the death certificate is asking for, much less know that the validity of it is good.
DR. MAYS: I guess my question is what is the training that, in your profession, you have for ascertaining this data? To what extent are there any quality controls in making sure it is really asked the way that it is supposed to be? Are there differences by states what have you.
MR. MOORE: Yes, that is the short answer. Services largely regulated at the state level, there is license requirements that are varied from state to state. I know at our professional level of the National Funeral Directors Association, we have an arranger training course that we recommend a lot of our members take that has to do with approaching each family in a meaningful way.
But getting the data for the death certificate, I would say the help from that comes from the states. I know there are some. In New Jersey, we have had some pamphlets or guidelines, if you will, that come out. My wife was sitting behind me, and she sent me a picture of a brochure that we have from the CDC. I am not sure how old that is, 2002, that was put out to help funeral directors gather data in a way, as well. I would say it is limited to get something consistent across all jurisdictions.
MS. WEBSTER: So, our vital records offices also do, state-by-state is different by how much resources they have, how many staff people they have, how many staff field rep trainers that they have on their staff. But they do a lot of outreach to the data providers, physicians, the medical examiners, coroners, the birth clerks. It falls to their staff to really do that quality control check and build the relationships between the VROs and all of those data provider groups. It is a really hard job.
DR. MAYS: Is it one you feel you are funded to do?
MS. WEBSTER: No, not adequately.
MR. MCCOY: I will also, though, give a shout out to NCHS. Delton and NCHS, in the last two years, have been providing funding to a couple of states, Vermont is an example, to do comparisons of the death certificate to the originating information, so the medical record and other information sources. Those results of those pilots are just starting to become available because again it has only been the last two years. But NCHS has been very supportive of starting to look at these quality issues, so that we can design training programs to go back and address the issues.
MR. COUSSOULE: This may sound like a long-winded question. It may be a little more of a statement, I think, in some ways. As I hear the challenges inherent in this very complicated system, I always tend to think. There are perpetual resource challenges in any place you are at. It doesn’t matter whether you are in the federal government, the state government or private companies, et cetera. That is a normal challenge.
I always have in my head to kind of start with the end in mind. The problem is, as I hear this, there are about 100 different kinds of ends in different intermediate steps. So one of the challenges I see with this is the end needs kind of flow back into the data capture question. How do I get the data I need? If I am trying to do something at whatever level it is, what data do I actually need to make that valuable?
For each endpoint, there is some limited subset of data. When you add all of those up, it is a vast amount of data, which then the capture points become very complicated and expensive. And so, I am trying to understand a little bit about the incentives that drive the behavior at the capture level to then transfer into the end uses.
And so, I think in my head, it is easy to think back and say, well, the Social Security Administration needs this information, so I will pay $3.00 or whatever the amount happens to be. The difficulty is translating all those needs into how do we actually fund the capture at the source level. I am trying to understand how you all even think about that in some ways to figure out how to push the data capture piece, if the incentives are very difficult to figure out. Does that question make sense? I am not sure who I am directing it to. I am more trying to think through the system incentive as a whole, trying to drive the behavior, when it is not clear what those incentives are in total.
MR. MCCOY: I will unfortunately answer your question with a question. Should 57 jurisdictions each be paying $3 to $5 million to develop and implement their own EBRS and EDRS, meaning should Vermont spend $5 million to pay its IT staff to build in-house a system, and then New York State next door purchases off the shelf for $3 million, and then Massachusetts does something else. Couldn’t that money be better spent accomplishing the quality improvements that we are looking for?
But it is because we have 57 jurisdictions with different statutes and laws and requirements that this patchwork is developed. But is there something we could do at a higher level, national level of standards to encourage perhaps more technology sharing, more reuse of technology, versus 57 different pieces of technology?
MR. COUSSOULE: Just from a personal perspective, having run IT organizations for about 30 years, I get that challenge. The interesting part then is, I follow it with a detailed question, which again is open to whoever would like to answer. Part of the question gets into the federalism versus not discussion. So how do we clarify the needs at some macro level with allowing the solutions to maybe be different a little bit.
I think what you are saying is how do I create leverage? And where do we believe in this system we can create leverage, such that we don’t have to spend $10 at every place. Instead, I spend it once or twice, something like that. Where do we think the leverage points might be in the system that more federal oversight or direction or guidance could be helpful, but not stifle the creativity in the local system, which is a point you brought up earlier.
DR. DAS: I like that our system is federated. I do. I think there are strengths in each jurisdiction.
MR. MCCOY: Currently, there is a process by which we have the US standard birth certificate, death certificate, and so it is revised every, we will say, 15 or 20 years. That process involves NCHS bringing together experts nationally, so physicians and hospitals, researchers and other groups, as well as the jurisdictions. That process is expected to involve coming up with what is the core deliverable that we need nationally, then allowing for room for jurisdictions to add their own individual items.
So for example, some states have added questions to death certificates, was the death related to diabetes? A totally separate question, a couple of states have added that on their own. Some states add to the birth certificate certain birth defects. That process has worked pretty well. The challenge has been getting the jurisdictions to be able to implement the new versions of those certificates. Whereas we nationally may come to agreement on what that core is, getting the jurisdictions the resources to implement it has been a challenge.
And so I am going to get some of the numbers wrong. I think in terms of the revised birth and death certificate standards, it took about literally 15 years to get all jurisdictions to implement. That is a long time to get everybody on the same core.
DR. ROSS: I think the piece that Nick might have been focused on, or at least what I got from it, was have you done an assessment of all of the federal users, what they need and what reasonable costs would be for the individual federal users that probably could be aggregated into one federal bucket across the different agencies that need the data? Has that work been done or is it being done?
MR. FUDENBERG: Certainly, through our process of rolling out the EVVE system nationally, we have a pretty definitive list of potential users for death certificate information, so federal agencies that have a need for death information.
DR. ROSS: Think of it this way, what if they didn’t get it? What would they want to pay to get it?
MR. FUDENBERG: Are you saying that we should start charging you for all of our reports that we send you every day? Is that what you mean?
DR. ROSS: What I am hearing is there is this sort of pay for play. A theme is saying a lot of people at the federal level, a lot of agencies, benefit, but they don’t necessarily pay for the benefit that accrues. They are assuming it on the backs of states, local entities, funeral directors and others that are all the partners that actually gather the data. You are the data-gathering enterprise, but other people benefit. That is what I am hearing. Am I wrong that there is a mismatch here between those who benefit and those that bear the cost of gathering it?
MS. WEBSTER: Yes, I would say the overall answer is yes. We know what the market is. We have also gotten ten years of feedback about the pricing structure and made adjustments and made deep discounts and tried to work out deals here and there. The main problem we have had with getting federal agencies on our system is that it takes years to change the direction of the boat.
Unless you have one person, unless you have a federal mandate, some kind of, as our President Lou always says, the hammer, we don’t have a hammer. There is no leverage for us to make anybody do anything, not really. The language is not specific enough. It doesn’t mandate use of any particular system, let alone ours, which is the right system to be using. So there is that issue.
But unless you have somebody in the federal agency who is willing to go to bat and really understands the importance and the value of this particular dataset, it takes years to ever get anything done, if you get anything done at all.
DR. COHEN: We are spinning a little here. We can continue this conversation maybe in our discussion at the end of tomorrow’s session to wrap up. I want to make sure we get everyone whose tents are up to have a chance to comment.
DR. THORPE: Can you briefly tell me what your system is, and will it actually fix the problem? What I have heard this morning from you all is that everything is jurisdictional, and there is a lot of heterogeneity. Either we come top down, there has to be some difference and some latitude and some flexibility. I am not sure I understand how the costs. I am with Nick. I am not sure I understand how the costs of these things got costed out with so much heterogeneity across the jurisdictions.
MS. WEBSTER: The cost estimates have been developed by our membership. We do that in committee. We do lots of research because we are statisticians and that is what we do. We come up with those cost estimates. We renegotiate these contracts every couple of years with SSA, so it is fairly up to date information. That is one.
The systems that I am talking about are our STEVE system. It is the State Territorial Electronic Verification Vital Events. It is the system that we developed sort of in coordination with NCHS that helps jurisdictions exchange data with each other. It is the way that Florida tells Colorado that someone that was born in Colorado has just died. That way, nobody can use a birth certificate and get a credit card and commit fraud.
That system is also set up to deliver datasets to other public health programs or government agencies that have the legal right to the data. That is one. That is exchanging data with each other.
Then we also have EVVE, the Electronic Verification of Vital Events. That system is a point-to-point system that allows someone in a passport office, say that is looking at a birth certificate, to get on EVVE, put on a little bit of information, and within 30 seconds, verify that is a legal, legitimate document. In some cases, they can also certify. So if you don’t walk in with a valid birth certificate, you can pay the state’s fee for certification and actually get right away a certification of the birth document. So that person doesn’t have to go away from the office and come back a month later, and run the same process again.
We have just rolled out our Fact of Death System, as well. That provides fact of death information to both government, anybody can use it for government agencies, but many states can also provide fact of death information to non-government agencies like pension companies, insurance companies, measurable use of that dataset. In that way, that is what I am saying when I say we have the system set up to deliver the data. It is just a question of having folks actually use them.
DR. STEAD: Just to push us out of the box a bit, would it be possible, or would it be useful, to decouple the legal reporting from the health reporting? Would that be possible? Or would that decrease the complexity sufficiently to let you get one piece of it standardized?
DR. DAS: I think considering that for the birth certificate can be discussed because the distinctions between the confidential medical report and the certificate that is issued to families and others for legal and administrative purposes are pretty distinct. It is more difficult on the death certificate because the information is more integrated. That also varies by jurisdictions as to which of the medical data fields are part of the legal death certificate. But for many of the jurisdictions, it is combined. When you are issuing a death certificate to a family, they are getting cause of death, manner of death, injury information because those are also things that people like insurance companies need. That is my answer on it. I am worth exploring anything that is out of the box. That is fine with me. But I do think it is a much more straightforward process to do on births than on deaths.
MR. MCCOY: I will add, I do think it is very possible because NCHS and NAPHSIS has been working together for the past eight years to establish technical standards for extracting the medical information and directing it to the health departments’ data systems. So essentially doing what you are saying, decoupling the legal registration piece from the medical piece. So we have spent years now coming up with a very strong set of standards, which places like Carrie’s state and California have been testing. I think the model exists. We just need to pilot it some more.
MS. GUIDA: I would just say one of the things when we were doing that project, and as we look towards health equity, we have to make sure that we are not losing some of those social determinants of health information that is not always in the EHR that may influence health more than all of the birth outcomes, such as the mom zip code and education. As long as we get that together —
DR. DAS: I think one of the reasons why our public health colleagues love the fact that a birth certificate is used for legal and administrative purposes, it is because of that, we get all of the information for the public health and medical information, in addition to those legal and administrative fields. So in a way, it is our public health colleagues. They like that. It helps them get the information that they need on all births.
DR. STEAD: I get that. I just wonder whether we are making the baby too big to deliver. As a total generalist in this space, reading all the material coming up here, it looked to me that maybe what happened in 2003 was a great overreach for the amount of information to collect. Maybe that is why it has taken so many years for it to take effect.
I just put that because if you think about, like the way you described all the pieces that are assembled through the death certificate process, there is no reason all of that needs to be tightly coupled in today’s technical world. You might actually could get to something that could be much more nimble. I just put it out there for you to think about.
DR. COHEN: I would like to get these last two questions from the committee and then open it up for more general discussion before our lunch break.
DR. RIPPEN: Just based on the perspective of a lot of times, people want to touch and make the person who does the first touch responsible for everything. It happened with quality reporting in the emergency department. You had all these things that were totally relevant for stabilizing a patient, but you had to ask them all these questions that were totally irrelevant to care.
So again, as we start thinking about, well, how do we actually allow for the efficient use of limited resources, but then again, not have the potential of linking to needed information from a public health perspective and for other purposes. So as we move forward, again, there are a lot of lessons learned and a lot of different health-related IT data approaches. It is just an important thing to consider, so yes.
MS. GOSS: I am going to hold part of my question until we get into this afternoon because it was back in the box dealing with standards and of course some interest there, so I will make this a general comment. Last month, the standard subcommittee had a hearing on predictability in adoption of HIPAA standards and operating roles. I am hearing some very common themes.
I am thinking that from a collaboration perspective across the subcommittees, we need to chat because I am seeing a lot of stuff woven together. It is all about the data, about the data model. It is all about the data flowing and the systems that it needs to. It has lots of people who want to grab the data, a lot of legal and technology issues for us to discuss.
Just in the short couple of hours that we have been here today, I know we tried to isolate the roadmap into its own discreet activity. But I am thinking between this conversation and terminologies and vocabulary conversation, there is a larger visionary piece of work for us to do as a committee.
DR. COHEN: Thank you. That is really important for us committee members to hear about integrating this work around vitals with a variety of the other streams of inquiries that we are planning. Are there any questions or comments from members on the phone? Okay, so I am going to open it up for any comments or questions from folks in the audience. There is a mic right behind me if anyone has questions or comments, or if any of the panelists have any questions for any of the other panelists?
MR. FUDENBERG: Valerie, you mentioned you have a coroner medical examiner vital statistics system in Canada. Is that basically used as the case management system for the coroner medical examiner’s offices? And you at vital statistics have access to that data?
MS. GASTON: No. So what we have done, we basically have a mapping. We map to their own systems. Again, it is much simplified. There are only 13 systems. It is a lot easier to map and make sure that their elements go to the minimum dataset that we established with them.
But that being said, even that is not perfect. So there are some times when we look at the data, we realize that what should have been mapped somewhere wasn’t. But that is how we have done it.
MR. FUDENBERG: Sure, and the reason I ask is back to Nick’s question is that, talking about leverage, there is only two or three vendors in the medical legal case management system world that covers probably 70 percent of the United States population. There is no reason we couldn’t develop some standards that they could at least map the data. I am not a technical guy, but they could map the data to wherever it needs to be mapped. I know that there has been some discussions, but that is one way that we could extract a lot of the data from the coroner medical examiner’s offices.
DR. COHEN: Lou, could you please introduce yourself?
DR. SAADI: I am Lou Saadi. I am state registrar for Kansas and president for NAPHSIS, board of director for NAPHSIS. I wanted to follow up on your question about what is our incentive to participate. I mean, first of all, we have to collect this information. It goes beyond civil registration. We have to collect who has been born, who has died, who has gotten married, who had various vital events. Our incentive is to our own citizens.
But then we have to, because it is a local initiative, local requirement, we belong to the larger country. So we have to participate nationally. Vital statistics is probably one of the few government functions that I can imagine that actually can run like a business. We have probably heard that. How can government function like a business? We have various components that require us to do that, and we have various obstacles.
But the incentive then is to participate on a national level and to be good stewards of the information. It is the opportunity to collect such rich public health information on our population birth and death, the beginning and the end. So our issues become when we try to make the information available, more timely, more completely, and of a high quality, such that others can use it.
I mean, as one of my board members stated very profoundly is we really don’t need all this rapid reporting and these sophisticated IT systems to serve our customers. But there is a responsibility to share this information. So how do we get it? The crux of it becomes how do we recoup the costs from a business perspective of these IT systems and the facts that it costs us to meet payroll, it costs us to meet some of the requirements that we have.
So what is our incentive to provide the information? We have so many obstacles to overcome in the mindset, oh, it is in the computer. You can just hand it over to us. I mean, how many times have we heard that. It is just there, hand it over. But in the end, we are going to have to recoup our costs, so that we can innovate, so we can make things better.
I have got like eight pages of notes of things that could be of interest to this group. You have all hit on many of them. But the incentive, what is it? And it is because we are good partners that we are members of this country. We have to participate. It is our responsibility. But then, how do we do it in a way that we can survive and innovate and become standardized across the whole nation?
DR. COHEN: Thank you, Lou. We will be hearing from Dr. Saadi tomorrow about some of the ideas and suggestions about where we can go from here.
MS. HINES: There are two questions that came in from the polling, if no one in the room has any questions. If every state is paying $3 to $5 million for electronic death reporting system, has there been discussion of developing shared services? That is one that would be relevant. And are state and jurisdictions impacted by the consolidation of or centralization of IT services?
MR. MCCOY: I will take the centralization question. The answer is yes. It hasn’t occurred in all jurisdictions, but several such as myself, Kansas and others have been impacted. So what it means is we compete for those limited IT resources. We have to go through a much more stringent scoring and approval processes. So it does mean we have longer development and implementation times. And also, it means the people that we need to know to understand our business process are further away from us. They are not co-located with us. They report to a whole different group or organization, and they don’t understand why we do what we do or how we do it. So they don’t always jump on, say, the birth system being down for three days, even though the emergency plan for the state says 24 hours. So that is a real challenge for us. I know some other states.
DR. SARDY: I want to echo what Rich has said. Many of us are being significantly impacted. The other thing is it is just not a birth registration system or a death registration system. It is an entire backend processing enterprise. So many of us in the late 80s invested in imaging systems, so that we can capture these documents. That was one of the first innovations that my state did to be able to automate vital records processing.
But we have queue management systems. We have the locator activities, a series of things that have to be in these systems, not just the fact that we are filling the records. We have to build indexes. There is just very sophisticated systems that are associated with these. That is why they cost so much, the security provisions, et cetera.
The other thing that we can’t participate in, and I wish we did have more of our IT colleagues at the table to describe some of this, is many systems are being developed, and they are in some of the open source technology, our state IT directors won’t let us implement those. I saw an article in publication city and county government, only 20 percent of open source products are adapted at the local government level.
It is worse for us. There are so many requirements. I mean, I manage the infectious disease data collection system. We started with, just on a lark, to be able to accept an open source system from Utah. I had to pay to have it retooled into a more standard IT product. They wouldn’t allow because it didn’t meet the security requirements.
So we have a number of obstacles in that department now. Can we begin to look at ways to share these resources, now that we have cloud computing. Some states now are accepting cloud computing as a way of doing business. There is a lot of opportunity in that area. But we are hampered in many ways from just our IT requirements.
MR. MCCOY: We are talking births and deaths, but we all are responsible for marriages, divorces and, in some cases, fetal deaths, stillbirths. There are other vital events that we have systems to support.
I echo again what Lou said about open source. Our IT for our state prohibits any open source software application, which is quite challenging because a lot of the statisticians I hire come out of school trained using a program called R. R is open source. I then have to retrain them to use SPSS or SASS because we are not allowed to use R.
DR. COHEN: We didn’t address that second question, which is potential economies of scale about pooling resources maybe from different federal agencies or whatever. Why don’t we hold that for the discussion for tomorrow afternoon, after we have heard from everybody for a day and a half, I would like to maybe reconsider that is a possible way forward.
I want to really thank this incredible panel. You got us off to a fantastic start this morning, laying the groundwork from a broad overview, from an international perspective, and boots on the ground in our communities, in our states, in our counties. You did a fantastic job.
Two quick announcements. One, there is a cafeteria upstairs. We can be back in an hour. I really need to thank Kate who is the chief cook and bottle washer here for our subcommittee, and has really pooled everything together. She has done a phenomenal job, so thank you very much.
We will reconvene at 1:15.
(Luncheon recess at 12:15 p.m.)
A F T E R N O O N S E S S I O N (1:15 p.m.)
DR. COHEN: Let’s all take our seats and re-gather for this afternoon’s session. This morning we focused on primary data collection from the states and jurisdictions and other data collectors. This afternoon we are going to focus on the federal perspective of data users and purveyors and a variety of other users — commercial users and researchers.
Before we begin our panel this afternoon, this morning the issue of several natural disasters came up in our discussions. I just wanted to take a moment for all of us to acknowledge and remember 16 years ago on September 11th there was a man-made tragedy in this country that bears remembering, so let’s just take a moment of silence to reflect personally.
(Pause)
Thank you.
Agenda Item: PANEL 2: Current Status of Vital Records: Federal Perspective
DR. COHEN: This afternoon we have got quite a cast of federal folks. I will try to read them all. Delton Atkinson, Director of the Division of Vital Statistics at NCHS, will start. In no particular order, federal users, Cathy Carter, who is the Director of the Enterprise Systems Solutions Group at CMS; Karen Matsuoka, who is Director of the Division of Quality Outcome for Child and Adult Health Programs at CMS; Bill Riley, NIH Associate Director of Behavioral and Social Research; Ursula Bauer, Director of the National Center for Chronic Disease Prevention, CDC; Mike Kogan, Director of the Office of Epidemiology and Research, MCH Bureau in HRSA; Stephen Goss, Chief Actuary, SSA; Corryne Carter, Program Analyst from Passport Services at State Department; Carolyn Clancy, Deputy Under-Secretary for Health at the VA; and Aaron Firoved, Director of the National Biosurveillance Integration Center, Department of Homeland Security.
Did I miss anybody? Okay. My job is done. Delton, please.
MR. ATKINSON: Good afternoon. As mentioned, I am the Director of the Division of Vital Statistics at the National Center for Health Statistics. I think my comments today will be from a couple of different perspectives — as a division director and, at one time in my life, I was a state center director in North Carolina, and another time in my life I was with NAPHSIS as its Executive Director, and at another time in my life I was with the private sector working on vital statistics issues, so I have gotten to see vitals in a number of different perspectives.
What I want to do today is talk about the National Vital Statistics System and the National Death Index. Both are different. The National Death Index is not a part of the National Vital Statistics System. I want to talk about the need for these records by governance and some of that you heard this morning, and the financing, the sharing and use of the records, and then the issue of timeliness and quality of the records.
Now, as a part of the National Vital Statistics System we collect data from 57 registration jurisdictions, and you can see the states. We collect a complete census — births, deaths and fetal deaths. Notice you do not see marriages and divorces because those are datasets that, because of funding issues, we ended the collection of that data. We do get marriage and divorce counts from states.
Annually, you can see the numbers here — about four million birth events, a little over 2.6 million death events, and about 26,000 fetal deaths that are 20 weeks of gestation or longer.
Why do we collect these records? There are really four different reasons. One is that it helps us to describe the health status of the nation — what are the births and deaths outcomes that our citizens are experiencing, and is that improving, getting worse, and so forth.
Second, it has been used for the surveillance of new and emerging diseases. Because of some of the successes we have had in improving the timeliness, we have transitioned the National Vital Statistics System to be more of a surveillance system than what it has been in the past. An actual example here is the 122 cities mortality surveillance system. That system has gone away. We are now using the mortality data as a part of the National Vital Statistics System to feed that program in terms of the pneumonia and influenza data that it needs.
Third is program management and policymaking, and then the final one, obviously, is health research that needs to go on here.
We have heard a little bit this morning about the financing of Vital Statistics, and I have here statistics versus civil registration. Tara and some others talked about the civil registration and how that is done, which is financed through the sale of birth and death certificates or the sale of services. If you think about Vital Statistics we really have transitioned, whether we wanted to or not — we have transitioned it to be a fee-for-service system. It’s the sale of something in order to make it work.
We at NCHS provide funding to states for the National Vital Statistics System through our Vital Statistics Cooperative Program contract. These funds that the states get — it’s about $21.1 million that they get — they can use for whatever purposes they want. We don’t tell them that they have to use it for this or that. Some get to use it; some do not get to use it.
If you look at the components of that $21.1 million, it’s births $9.5 million, deaths $8.5 million, fetal deaths $1.1 million, infant/death linkages $2 million. If you take this whole thing and divide by 57 jurisdictions, you’re only talking about $400,000 a year. That’s peanuts relative to the cost of having a top flight Vital Statistics System in this country.
In terms of the governance, again, it was mentioned a bit this morning that vital records are governed by state laws, not federal. That is very important. One of the things that we seem to have difficulty with sometimes when we are working with researchers is because the first thing they will say is you have the records, and why can’t you give them to us and why can’t we use them in the way that we want to use them. That is not how the system is set up. So it’s the laws and statutes and regulations at the state level that govern the use of the records, and we have to try to follow those laws and regulations to the extent that we can.
How the records are used is outlined in our VSCP contract, and the participation in the contract is optional. States really do not have to participate, but we get 100 percent participation. I think that’s because they get something out of this process as well as we do.
We have tried to work with states in moving them towards electronic vital registration systems. Almost all of the jurisdictions — that would exclude the territories — have some kind of electronic birth system in place. Roughly, about 46 jurisdictions have an electronic death registration system and some of those others are working towards trying to move in that direction. Even though they have an electronic registration system, not all of those states are statewide, nor do they have complete coverage, and I will talk about that a little bit later in the process.
If you look at the flow of records from the states to us, the state are providing us births, deaths and fetal death records that come into our transactional database. There, the records are being processed as they are being received, and at the end of the year we spin off out of that a mortality database, a fetal death database and a birth database and, those are used for analysis.
Let me give you an example. If you take deaths, we have about 2.5 million individual records for deaths. We have close to 7.5 million transactions that happened in order to create those 2.5 million records. Why? We have updates that are coming in, changes and modifications. I have a staff down at RTP that spends 100 percent of their time working with states trying to improve the quality of the records. As they find errors, they work with the states to get corrections and we get corrections in. So, there’s about three times as many transactions as we have individual records that have to be gone through on a regular basis in order to get there.
In terms of the data and what we do with it, we have a number of annual vital statistics publications and topic-based data briefs on birth deaths and fetal deaths. We recently initiated a vital statistics rapid release program. Because we are getting the data in faster than we had in the past, we are able to implement a surveillance program. We create public use files for births and deaths. We have files on CDC WONDER. We have confidential research files that one can access with approvals, and we have the data in the research data centers.
Let me go to the National Death Index. The National Death Index is comprised of all deaths occurring within the U.S. from 1979 forward. It is also comprised of all military deaths occurring overseas — something that a lot of folks didn’t know. The jurisdictions and DOD own the records. In many ways, we are an intermediary. We are there to try to facilitate research happening.
Our actions are governed by data use agreements. When a researcher comes to us and matches the files and then says, well, I want to re-use it for this project or that project, and so forth, we cannot do those kinds of things, again, because we have to abide by the state rules. We have to abide by whatever DOD rules they provide to us.
If you think about it, the National Death Index is not something that’s mandated in our statutes for NCHS. It really is a collaborative activity. And to be able to get the 57 jurisdictions and to be able to get DOD to all agree that this is something valuable and useful and this is the way that we want to be able to do it I think is tremendous. Currently, we have over 90 million records in the database, and we link all kinds of files with that.
You remember I had a diagram that shows the flow of records. What I have here is just the mortality. You see the morality records coming from states into the transaction database. With the tremendous support that NIH has provided to us we were able to build the DOD transaction — well, no, let me take that back. The DOD transaction database we were able to build with funds from DOD. We get monthly files from DOD that go into a transaction database, and that gets added to the NDI database, so you see how the mortality data is created here.
In terms of the use of the National Death Index, the researcher must submit an IRB-approved project to NDI. That goes through an approval process. We have an NDI Advisory Committee that is comprised of state professionals and federal, state and private researchers who actually review and approve each of the NDI applications. Those individuals are very stringent about what it is they are looking for.
Each of those approved applications then comes to me for final approval, and then my staff goes through the process of matching the researcher’s records with the NDI database, using one or more of the following criteria that you see up here. Then, we provide the results back to the researcher — it’s either the fact and/or cause of death — and then assess a fee for the matching services.
That fee is one of the things that create a lot of consternation. NDI is 100 percent receipts-operated. There are no appropriated dollars in NDI. That means that the receipts must cover the payments for the states for the use of the records, my staff’s salary and fringes, the operating expenses including travel, IT costs and NAPHSIS cooperative agreement, the NDI Advisory Committee. Again, the records are owned by the states, so they get to set the fee structure. They also get to set the access and use rules for the use of NDI.
Again, what we try to do is have the NDI cover — and it was talked about a little bit earlier about how the states have different laws and statutes and so forth. They don’t have 57 similar laws; they are different. So we try to set this up such that it will operate given those 57 different sets of rules. If you have ever tried to put something together that has 57 different sets of rules to try to come up with something that can operate, this is a tremendous challenge, but we have been able to do that.
Given that background, what then are the challenges and opportunities for the National Vital Statistics System and the National Death Index? You will have to come back tomorrow, because my presentation tomorrow will deal with these challenges and opportunities, but it was important that you understand this background in order that you can understand some of the challenges and opportunities that I will outline tomorrow. Some of those, Bruce and David, will be some very specific recommendations that I would like for the committee to think about.
DR. COHEN: I’m disappointed; I wanted to hear them now. Thanks, Delton, that’s great. Karen?
DR. MATSUOKA: Thank you. On behalf of the Centers for Medicare and Medicaid Services I want to thank you for having us here today. Whether it’s covering the birth of a child or providing for hospice care at the end of life, over 100 million Americans depend on CMS for their healthcare through Medicare, Medicaid, the Children’s Health Insurance Program or the Health Insurance Marketplace. For that reason, vital records data are critical to the mission of CMS, and that is why we’re here today.
My name is Karen Matsuoka; In addition to Division Director for Quality in Health Outcomes, I am also the Chief Quality Officer for Medicaid and CHIP, and I am joined by my colleague, Cathy Carter, from the Office of Information Technology.
It would be absolutely impossible to describe the myriad ways in which CMS uses vital records data in just the few minutes that we have with you today. What we are hoping to do instead is to submit a more comprehensive written statement following today’s testimony, but for today, I thought I would briefly describe how we use vital records data in the Medicaid and CHIP programs, and then I will turn it over to Cathy who can provide an overview of how vital records data are used in the Medicare program.
Medicaid and CHIP provide health coverage to over 70 million Americans including roughly 20 million low-income adults, 35 million children, and 15 million elderly adults and individuals with disabilities. Not commonly known, nearly two out of every three adult women enrolled in Medicaid are in their reproductive years, and Medicaid currently finances about one-half of all births in the United States. In some states, that percentage is actually much higher. As such, improving the health of mothers and newborns and reducing complications of childbirth is an incredibly high priority for CMS.
CMS promotes quality measurement and improvement in this focus area in a variety of ways including the inclusion of perinatal quality measures in the Medicaid and CHIP child and adult core set measures, quality improvement grants and learning collaboratives for states, and demonstration programs. Vital records data, especially when they are linked with Medicaid and CHIP administrative data, are an important source of information for measuring maternal and infant health outcomes, for understanding material demographics and risk factors, and for informing and monitoring the effectiveness of our quality improvement activities.
For example, we have included the Joint Commission measure on early elective deliveries in the Medicaid and CHIP core sets because it is a key indicator of perinatal care quality. Approximately 10 percent to 15 percent of all births in the United States are performed early without a medical reason. These births have increased risk of maternal and infant complications and result in additional unnecessary costs for patients, providers and payers. While Medicaid claims data can tell us whether a delivery happened, gestational age information is also needed in order to determine whether the delivery was early or full term. Vital records information provides that key variable.
The same principle applies to birth outcomes measures like rates of low birth weight. While Medicaid administrative data is helpful in tracking the utilization of services like prenatal visits and postpartum care, it is when this data is linked with vital records data that we can obtain information about the clinical outcomes of those services, which is ultimately what is most important for the beneficiaries that we serve.
In turn, states have been able to use these data to drive quality improvement on these measures. Leveraging an adult Medicaid quality grant — for example, the state of Louisiana was able to link its Medicaid and vital records data and then use this data infrastructure for both quality measurement and improvement purposes. First, they provided hospital-based training on the importance of full-term births; second, they reinforced this training with policies that stopped payment for early elective deliveries that were not medically indicated; and, third, with this linked data infrastructure the state was able to establish a registry of women at high risk of pre-term birth, enabling the state to conduct more proactive care management. As a result of these efforts, rates of non-medically indicated, early elective deliveries declined from 18 percent to 5.5 percent in just three years.
This is just one of many ways that we use vital records data in Medicaid and CHIP. In the wake of the opioid crisis, for example, we rely just as much on vital records data for monitoring overdose deaths. These data are, in that sense, absolutely vital to the mission of our work in Medicaid and CHIP.
With that, I will turn it over to Cathy for an overview of Medicare.
MS. CARTER: Thanks for having me today. The Medicare program is huge. I am listing just a few statistics here. The Enrollment Database and the Common Medicare Environment are a system and a database that my group in CMS maintains. It is the authoritative source for all Medicare demographic data. We collect the data from a variety of places — from our friends at SSA — that is actually the majority of the data; that is actually where it starts from — from the Railroad Retirement Board and from health plans and other places.
What we do once we collect that data is we distribute it to those same entities — we have an exchange program with them — as well as the downstream systems such as those — the ones that are probably the most well-known are for claims processing.
Examples of data that is collected and distributed. There has been a lot of talk about date of death. That is very important for Medicare purposes. We collect that data from SSA primarily although we also get it from claims, because in some cases there will be a discharge that is based on — the reason is death, and in those cases, we forward that information back to SSA. In many cases, they already know about it but in some cases they don’t.
We also are very interested in lawful presence data and incarceration data because we are not allowed to pay Medicare claims for people in managed care plans if the person is incarcerated or is not lawfully present. We also deal with all of the states for buy-in data because they buy in and pay the premiums for Medicare to have Medicare pay first.
One very critical element for us, and I guess we consider it a vital statistic, is the address. We get our addresses from SSA. That is our source. If anybody changes their address we send them to SSA to make that change. And then we get coordination of benefits data from insurers and payers to make sure that we are only paying first if there is no other insurance available.
Here is a little bit of a description of why we need this data. No payments obviously can be made after date of death, although there are some exceptions because the Medicare program is full of exceptions. We can’t pay if a person is incarcerated or if they are not lawfully present or if there are other payers that should pay prior to Medicare.
For purposes of the address, there are an awful lot of mailings that we do to Medicare beneficiaries, and we are in the process of convincing beneficiaries to get their information online more and more, but there are still a lot of mailings that go out. I have listed a few of those — the Handbook that probably all of you have seen. You haven’t maybe gotten it yourself but your parents or your aunts and uncles and relatives are getting it.
The initial enrollment packages when someone is first eligible for Medicare. The Qualifying Health Coverage notice was something that was required under the Marketplace. And in terms of state interaction and premium adjustments.
The last slide here shows some statistics. We have 59 million Medicare beneficiaries in that database that I was talking about. A couple of interesting statistics — 10 million that are low-income, six million that are considered high income and therefore they have to pay a higher premium than the average person, 45 million of those people enrolled in Part C and D plans, and the states pay premiums for about 10 million beneficiaries, and we are paying over one billion claims a year. In other words, our vital statistics is a big business for Medicare, and we are interested in improvements to those areas and in our work with our data exchange partners. Thank you.
DR. COHEN: Thanks, Cathy. Bill?
DR. RILEY: Thank you all, and thanks for this opportunity to present to the committee. I appreciate the presentations this morning. They gave me certainly a greater appreciation for the complexity that we face as we move forward. I am going to try to give an NIH perspective on the vital need that we have for vital statistics in this country and how it’s important for our researchers.
Obviously, one of the first things that we use this data for is to get a better sense of early warning signs for the types of things we should focus more research on. The national data, among other things, is one of the ways in which we do that to determine what are going to be our priorities, what sort of things should we put more money into as we move forward.
The opioid crisis is a good example of that, although, as you will notice, we were a little slow to the trigger on that one. Before we actually had smartphones in our pockets we sort of knew that there was an opioid crisis going on but we didn’t really pay that much attention to it for various reasons.
It certainly is also important for us for population health research data, looking at some of the drivers of that. This is the Angus Deaton study that many of you are familiar with — Angus Deaton is one of our Nobel Laureates in the field — showing that some of the opioid crisis that we now see was from decades before, and some of the economic and social drivers of that from decades earlier, especially among working class whites, has led to some of the increases or laid the groundwork for some of that moving forward.
Obviously, one of the places where the NIH is particular interested in death data is among epidemiologic research linked to deaths, so we support a number of both surveillance studies and longitudinal studies, clinical registries across the board across many sort of disease entities, and these epidemiologic studies are obviously critical not only for documenting this against prevalence of diseases but also assessing predictors and moderators and their relationships between morbidity and mortality.
And the linking has been one of the difficult aspects of that. Most of the other data we can collect. But the one thing that our epidemiologic researchers don’t have that much control of is what is the death data of these participants, so that’s an important component of it as well.
These are just a few examples of the NIH-supported epidemiologic studies that are currently ongoing that use death data. These are not all, but this gives you some sense of what we’re dealing with, all the way from the NCI SEER database for monitoring cancer surveillance, a number of the NHLBI studies from Framingham on to some of the newer ones, the special populations and underserved populations, and some of our emerging databases that are upcoming including the Environmental Child Health Outcomes Study, the Adolescent Brain and Cognitive Development Study and, the largest one of all, the Precision Medicine Initiative, now called the All of Us Study. So, if you start thinking about one million-plus people and eventually getting the death data on one million-plus people, this becomes a difficult process moving forward.
Just one example of the importance of getting quality death data — this is cancer survivor estimates, and I can make this pretty simple. If your health provider has given you or a family member some data on the five-year survival given that you have been diagnosed with cancer X, whatever that happens to be, that data comes from two places. It comes from our ability to, one, be able to early diagnose and identify someone with breast cancer, prostate cancer or whatever and monitoring them over time, and then getting accurate cause-specific death data to determine whether they died from that cancer or they died from some other cause other than cancer. So, it makes this work critical. Without death registry data and cause of death data, we actually do have to develop that from actuarial data estimated instead of being able to use real data to be able to pull that together.
The other point I want to focus on — I think most of this committee is aware of some of the linkage issues that we have with epidemiologic studies in relationship to the National Death Index. The other place where it’s particularly important is in clinical research and clinical trials research. In this space, we support right now — in FY2016 the NIH supported over 1,000 new trials. That’s pretty common for us. We’re funding probably 5,000 or 6,000 clinical trials currently at this point in time. Death is a primary outcome in a significant proportion of those studies, not all, but even in the ones where it is not, it’s an important indicator for us of serious adverse events and other things that we have to monitor in those clinical trials as we move forward.
This is the part that I think always amazes people, when I tell them this is how we determine in a clinical trial whether our primary outcome occurred or didn’t occur. People always go, you’ve got to be kidding me. This is the way you do it? No. But this is the way we do it. If we’re lucky, we have them at a hospital system that we are currently monitoring in that clinical trial. They’re in the hospital system where the clinical trial is occurring and we can move forward with that and get that data that way. That’s the fortunate way that happens.
The more unfortunate way it happens is that I am doing a follow-up call to a prospective study participant and I find out from their spouse or loved one who answers the phone that they are no longer alive. And they always sort of wonder why, even though I know their genomic data and I know lots of laboratory data, I know their clinical record, I know all that other data, why I wasn’t aware that they were dead. But that’s a pretty common phenomenon in those clinical trials.
Obviously, this is one of the reasons why we think it is particularly important to have data from the National Death Index to be able to do this work. Even with the variations that we have in quality and timeliness, it is really our only real method to verify death in most of these studies. Either we’re relying on the medical records or relying on the self-report of survivors to be able to get that data.
Again, we can’t do this state by state. When I first began doing this work we could certainly rely on a single state because we were mostly doing a clinical trial in a single state or doing relatively circumscribed work. We are now doing multi-site trials across the country. We’re doing national epidemiologic studies. It isn’t possible to do this on a state-by-state basis; we have to have some national system to be able to do that.
Just quickly, there’s a reason why we rely on the National Death Index and not so much the Death Master File anymore. Back in 2011, researchers before that had access to the full file with unlimited searches. Then the determination was made that the SSA really didn’t have the flexibility to be able to give this data to researchers that they had been previously doing, which then put a lot more load on Delton’s group because now the National Death Index is the primary way in which most researchers obtain that data moving forward.
One other quick point I wanted to make. Most people think about research as the typical research that the NIH funds, but, increasingly, it’s the healthcare systems themselves that are doing this research and monitoring the outcomes and being able to become rapid and responsive in their ability to be able to assess the outcomes from the care they provide. One of the ways they do that is by determining death, and again, their ability to be able to access all the deaths that occur not only in their hospital system but also outside of it requires data from the National Death Index and other vital statistics like that.
Just to wrap up, I focused on death; that’s the predominant data point that we get from Vital Health Statistics in most of the NIH research that we support. Obviously, there is research that goes on in teenage pregnancy and other types of things that rely on birth data, and even on marriages, births and adoption and some of the family aggregation and family constellation research that goes on at the NIH. I’ll stop there, thank you.
DR. COHEN: Thank you. Ursula?
DR. BAUER: Good afternoon, everyone. I am glad to be here with you. My name is Ursula Bauer and I am the Director of the National Center for Chronic Disease Prevention and Health Promotion at the Centers for Disease Control and Prevention.
CDC is the nation’s public health agency working 24/7 to protect Americans from health and safety threats. CDC, we like to say, increases the health security of our nation. I am the Chronic Disease Director so I will mostly draw examples from my center but I will endeavor to speak for the agency.
I will point out that, prior to joining CDC in 2010, I had 15 years’ experience in state health departments In Louisiana, Florida and New York, so I certainly appreciate the state perspective as well. While the federal government is a major user of vital statistics, state health departments are certainly an equal major user of the statistics as well.
At CDC, the main purpose of the Vital Records System — I will also say, for most of us who use the vital statistics side of the system, we don’t actually think often about the vital records reason that this system was set up in the first place, so the morning panel was, I’m sure, enlightening to a lot of us on the vital statistics side.
But we use those data to track trends in health events over time, basic information, of course, like the leading causes of death, but also to answer the kinds of questions that you have heard among my fellow panelists — so, are heart disease death rates going up or going down, has the progress that we have made in reducing lung cancer deaths stalled or accelerated, what’s happening with opioids, what’s happening with motor vehicle crash deaths and so on. The Vital Statistics System really allows us to monitor those kinds of trends and is the foundational surveillance system for CDC’s work.
To answer these and other questions and to take effective action to prevent and reduce premature mortality, which is the work of CDC, we do need the underlying cause of death. That gives us the most information regarding the death and how we might take preventive action. And we also look, of course, at the immediate and the contributing cause. For example, for influenza, the death record helps establish when the flu season has reached the epidemic threshold, while additional data included on the death record can inform strategies during the flu season to best protect the health of Americans — information like age, geographic location and so on.
The contributing cause of death field is helpful as well for enriching our understanding of low prevalence conditions — something like epilepsy, for example, which may not be commonly reported on the death record as an immediate or underlying cause, but these kinds of mentions as a contributing cause help fill out the picture of the burden, the geographic variation, racial and ethnic distribution and other features of a condition that we might not have another surveillance system to study well.
Vital records are the go-to system for much of our work and help advance both our public health function of disseminating information as well as adding to our knowledge.
Translating the death record into publicly accessible systems like mortality atlases that you will find on our website which are very useful to the public and to other decision-makers, linking the death record to other data systems to help complete information about a condition — for example, how diabetes or cancer survival has changed over time.
The death record, as we heard earlier, is a source of last resort, if you will, for our national program of cancer registries so that we can identify as completely as possible all our cancer cases. And we use the death records for some of our work to understand the impact of behaviors. Our smoking-attributable mortality and economic costs calculator or our alcohol-related disease impact calculator quantify the cost of those behaviors to society and, of course, are completely dependent on vital records.
Birth records are likewise the primary source of data that we have as a nation to track, monitor and report on trends on birth. We are interested in the number of births, gestational age, weight, sex, plurality, and maternal place of residence, age, risk factors such as hypertension, diabetes and tobacco and so on. All of this information is critical to our understanding of trends in disease and disease burden and how we can intervene on the public health side.
As with the death records, linking birth records to other data systems provides new information. For example, we can link our cancer registry data to birth records to understand birth outcomes in survivors of childhood cancer. The birth record, as we heard this morning, constitutes the sampling frame for our pregnancy risk assessment monitoring system. That surveillance system couldn’t exist without the birth record and helps us understand the new mom’s experience in the prenatal and postpartum periods. It’s critical for our understanding of behaviors and experiences associated with healthy birth outcomes as well as adverse pregnancy outcomes including most recently potential exposure to Zika, availability of contraceptives and so on.
The birth record also provides information including as a source of controls for case controlled studies. For example, in our acter bacterial core surveillance system we can characterize the cohort from which that infection might emerge and also calculate the appropriate denominator.
I have tried to show you how critical the Vital Statistics System is to CDC and how essential it is to our work for monitoring the nation’s health and for guiding CDC’s efforts to improve the nation’s health. We do have quality concerns, and, as we heard earlier, we will be spending most of our time tomorrow talking about those.
I will call out one issue which is accuracy around the race variable. This does lead — although it affects all of our racial and ethnic groups in the U.S., it does, in a particular way, affect American Indians and Alaska Natives who are severely under-reported in that race category making the system really substantially less useful for our American Indians and Alaska Natives.
We heard a lot about how we obtain our data. With that, thank you.
DR. COHEN: Thank you. As everyone is going through, I don’t know whether we are going to have time to address this question, but I would be very interested at some point if you could all let the committee know what the source of your data is — whether you have agreements with NCHS, whether you use NDI or whether you work with jurisdictions or NAPHSIS to get your vitals data. It would be very helpful for us to know where the data streams come from both for births and deaths.
I don’t think we will have a chance to specifically address that in our later discussions, but if you could let the committee know how you get the data, that would be really helpful. Thanks.
DR. KOGAN: Good afternoon, everyone. I am Michael Kogan from the Health Resources and Services Administration, the Maternal and Child Health Bureau. This is kind of a homecoming. I started off as a child at CDC NCHS, so it’s nice to see a number of familiar faces. Like Ursula, even before that, I worked in a state health department, the department in Massachusetts.
The Maternal and Child Health Bureau focuses mainly on programs for mothers and infants and families in the U.S., and vital statistics in many ways form the undergirding for all of these programs. I am going to talk about three major programs and the role that vital statistics play in all of them, and at the end I’m going to talk a little bit about research.
First, our major program is the Maternal and Child Health Block Grant. This is the Title V program, and it is the oldest continuous public health program in the U.S. It started in 1935 as part of the Social Security Act. As you can see, it supports the improved health of America’s children and families in all 50 states, D.C. and in the eight territories. Like Medicaid, we serve about one-half of all pregnant women, about one-third of all infants and children, and about four million children with special healthcare needs.
This became a block grant in 1981. At the most recent funding level, the block grant is funded at about $650 million, and for every $4.00 that the federal government puts in the state is obligated to put in $3.00. You can see that. since it became a block grant in 1981, we have helped the states reduce infant mortality by 51 percent and child mortality by nearly 53 percent.
Probably everybody here is familiar with the concept of a block grant. Its purpose is to give states more flexibility in how they address the needs, because states obviously have very different needs. What we want to do is assure accountability through performance measurement and technical assistance. As part of our legislation back in 1935, it was stipulated that we had to collect information on certain outcomes and certain conditions. All of them come from vital statistics. Again, vital stats have formed the bedrock for our data collection, and we also use vital stats to monitor progress in state priorities at the state level.
We did a major revision of how we evaluate the block grant just two years ago, and we think it is unique in the federal government. We now have three levels on how we measure performance starting with what the states collect for evidence-based strategy measures. We wanted to take advantage of at least 50 years of maternal and child health research. We found that certain interventions can influence performance measures, which we know can then have an effect on national outcome measures.
Here you see the role of vital stats in this system. For our national outcome measures we have about 26 in total, and nine of them come from vital statistics. In our national performance measures, we have 15, and three come from vital statistics.
The next program I want to talk about is Healthy Start. That is a $100 million program. It began in 1990 with the idea that there were certain pockets in our country that had a significantly higher infant mortality rate, whether it was in certain rural areas or certain urban areas, and it had to be an infant mortality rate at least one and one-half times the national average. This program supports at-risk mothers and infants from entry into prenatal care through two years after birth. We now have about 100 grantees. In 2015, the infant mortality rate among Healthy Start participants over 159 counties was 16 percent lower than the overall infant mortality rate in the same counties.
Vital stats play a role in how we determine who is going to be a community for Healthy Start. We also use it for our national evaluation and for OMB guidelines and HHS.
Here I would like to spend a minute on something we started in 2012. This is an infant mortality collaborative improvement and innovation network, or the CoIIN. It is done usually electronically. You get a group of people together and the idea is you exchange information rapidly, you test out ideas rapidly, and if it is not working you go to other ideas. This was done in the southern states to start. This would have been impossible without vital stats. We used provisional data. We gave money to all the states through our state supplemental data initiative, and for purposes of quality improvement we have gotten results in lowering infant mortality and a number of the factors that have been associated with infant mortality in those states.
Next we have the Maternal, Infant and Early Childhood Home Visiting Program. It is a voluntary program for pregnant women and young children. It’s an evidence-based program. We know there are decades of research showing that home visiting can be effective. Vital stats play a role. For 17 states, we used vital records data to measure impact on selected outcomes and covariates, and we also combined it with other data like Medicaid and Census.
For research, we have a grant program like NIH. We call it our 40-grant program where we use secondary data analysis for researchers, and they can analyze vital statistics data. We fund a number of grants every year in that area.
And then we also do research at the Maternal and Child Health Bureau. Here is a paper that came out last month in Health Affairs on Widening Disparities between Infant Mortality and Life Expectancy between Appalachia and the rest of the U.S. over the last 25 years.
And as a final comment I will go to life expectancy. We were able to break it down, infant mortality, at the county level by race from 1990 through 2013. But there’s a really interesting question talking about the rapidity of vital statistics. When we were interviewed by the media often the question came up what about opioid deaths. You only have it through 2013. Can’t you have it later? And we actually had to talk about the vital statistics process to the media.
Thank you very much.
DR. COHEN: We have an opportunity now with our first set of panelists for the committee to ask questions. We will have a chance later at the end of the day for everyone else. While the committee asks questions, if anybody else wants to use the Internet or their phones to generate questions, go at it.
We will have 10 minutes for questions from the committee.
MS. HINES: While we are waiting for that to come up, just a quick question for Delton that came in through the WebEx. Are there any vital registration areas capturing the abortion data, and how are those data used?
MR. ATKINSON: There are still a few states that are collecting abortion data. I don’t have the names of those at this particular point. Kansas is one of those. There are a few that collect it. We don’t at the national level.
DR. BAUER: We do at CDC. We publish an annual abortion report. Usually it comes out late in November each year.
MS. HINES: Where do you get those data?
DR. BAUER: States report them. Most states actually do report that but it’s not the 50.
DR. MAYS: Thank you very much. Each panel is more and more enlightening, so things are coming together.
My first question is to you, Delton. In terms of the quality of the data, for example, we all know if you’re born an American Indian you end up white at death. I guess the question is to what extent is race and ethnicity still a factor. What are the next movements towards — as the states collect the gender identity data, what is the next step in that? Is it a set of data that’s going to be regularly available for people to kind of be able to monitor that group as well?
MR. ATKINSON: That’s a difficult question, the whole race and ethnicity issue. There have been some efforts to link some of our data with Census data to see what was reported in the Census files in terms of the race and ethnicity. That is probably a better indicator than sometimes looking at what was the race at birth versus the race at death. Obviously, the race at death is being reported by somebody else, or it has just been indicated by the funeral home.
I think linkage is probably going to be the best way to go.
DR. MAYS: And that would have to only be done by your group, or you would ask the states to do it? Who would do the linkages?
MR. ATKINSON: We have had a couple of pilot efforts to do linkages on that. There are probably some states who have tried to do that also. I am not certain.
DR. MAYS: My other question is for Bill in terms of NIH. I know what the needs are in terms of the research you’re already doing, but the real question for me is whether or not you believe in this mortality issue enough that you will consider, for example, what kind of RFAs, et cetera, could be put out to help us in terms of thinking about the quality of this data, who do we need to target, are there best practices — some of the things that have come up previously which are research questions. I’m wondering if you can foresee that this was something that you might be able to take back and discuss.
DR. RILEY: I wouldn’t commit my boss to that but I would —
(Laughter)
It has certainly been enlightening today to understand some of the — as you said, there are research questions involved here, and how that data is collected and ways we could improve upon that, as well as just improving the access to those data and the ways we are able to do that more readily than we do right now. It’s clearly a problem.
I think, right now, what researchers spend more time doing — and Delton can probably attest to this — they spend more time complaining about their limited access to the data than they do about how do we research this to find better ways to get more quality and timely data.
DR. BAUER: If I could just speak to the American Indian and Alaska Native data linkage, CDC did pay to link I believe it was 1996 through 2009 data and run through the Indian Health Service Encounter data so that we could capture American Indian and Alaska Native deaths. We also routinely run our National Program of Cancer Registry Cases through that Indian Health Service linkage as well.
We do not have the resources, though, to update that since 2009, so we have a current gap.
DR. MAYS: What was the difference? When you did it, what was the —
DR. BAUER: For mortality, there was misclassification. We picked up 40 percent more cases. It is especially problematic for American Indians and Alaska Natives.
DR. PHILLIPS: I just have a quick question about paying to support the vital statistics infrastructure. I heard CMS say that clearly there is value returned to the states for the linkage of data. NIH, is there payment to support vital statistics and records that come back from the states?
DR. RILEY: Most of it comes through that fee-for-service agreement and the payment for each of those matching processes as those go on. There are a few data use agreements on some of our researchers I think as well that are broader than that, but there really is sort of a limited amount of support I think for the Vital Statistics System and for the National Death Index, being able to access.
MR. ATKINSON: I will say that NIH has supported us on a sort of special project basis where we have tried to take those dollars and improve the National Death Index. For example, we have now established a preliminary National Death Index. That dataset is available in January after the close of a calendar year. That dataset has at least 90 percent of all deaths occurring in the country. A researcher who uses that particular dataset, we provide them a free run of the dataset, of their files and of the final dataset when it comes in generally October-November.
DR. COHEN: I’m sorry, I am going to have to quickly go to the next questions because we have some conflicts in scheduling. If you have a quick question and a quick response — Denise?
MS. LOVE: I may have to punt this to later because mine is about linkage and using Census.
DR. COHEN: Let’s get to that later. Also, the issue has come up around the source of the data and whether funds go to NCHS or how do the jurisdictions get payment in kind for the data that is originally theirs. I want to postpone that discussion for tomorrow or even in the future because that seems to be an emerging point. Census will be here tomorrow.
Carolyn, shall we continue with you?
DR. CLANCY: Good afternoon, Chairman Cohen and members of the committee, and a special shout-out to Dr. Vickie Mays who had reached out to us with this opportunity — in case any of you wonder what I am doing here. I have been in many, many meetings in this very conference room over the years after 23 years at AHRQ until I went to the VA four years ago.
I am going to take a much sharper focus on the issue of vital statistics. We work with the National Death Index, and the specific focus I am going to hone in on is suicide. To cut to the chase, 20 veterans die of suicide every day, and we need to find the 14. I will get to that in just a moment. This is National Suicide Prevention Week so it seemed like a very opportune time to be here. I am going to tell you a little bit about what we do about this.
Suicide, as all of you know, is the tenth leading cause of death for Americans overall, and I think it is higher on the Top 10 list of years of productive life lost, which of course I know from my colleagues at CDC. We saw for the American public a 24 percent increase in age-adjusted death rates from suicide. It’s a little difficult to know how much of that is diminished reluctance to report a death as a suicide when, in fact, that is what has occurred.
Our research suggests that 20 veterans die of suicide every day. Six of them are getting care from us. Not all veterans are enrolled in the Veterans Health Administration — eligibility rules and so forth — but we feel responsible and that it’s part of our imperative, and it is one of a short list of priorities for our new Secretary Shulkin. We feel it is our responsibility to reach all 20. We can’t do that without your help.
We have a very large integrated suicide prevention program throughout the VA. Just to give you a feeling for this, I had the privilege of testifying about this before our Senate committee several months ago with a suicide prevention coordinator from Cantus, who turned out to be a good friend of Kathleen Sebelius. When asked how her day was, she said she started her day answering phone calls from our crisis line, people that she was to follow up with, and it was like juggling except she didn’t know until she got on the phone whether she was juggling rubber balls or glass balls. I think you could have heard a pin drop in the room when she described that to the senators.
We have extensive training programs because we really believe that every single person who interfaces with the veterans we serve — suicide prevention is everyone’s business. We have used our data to develop some cutting edge predictive models and actually started rolling that out last fall, which is a very interesting new world kind of way of dealing with clinical medicine — I believe, based on data that you can’t see or feel, that you may be at increased risk and we want to do everything possible to help you.
Another big direction that we have started under Secretary Shulkin on the advice of many inside and outside is actually reaching out to service members who have been discharged with other than honorable discharges. I won’t get into the arcane details of that, but many of those former service members are believed to suffer from PTSD or other behavioral disorders. We believe there are about 500,000 of those people and we are working to provide them emergent mental health care for up to a period of 90 days.
We are thrilled that we were able to work with our colleagues from the National Death Index to come up with our most recent data from two years ago, and very happy that we believe the next release will be November. That is not fast enough. We need to learn a lot more if we’re going to figure out how to reach those people who don’t interface our system. How do we know that they have died of suicide?
And, obviously, there’s a considerable overlap with the number of opioid overdoses. It is my understanding that that is a judgment call made by coroners. It is not remotely clear how much reliability there is on that particular call. My hope is that enhancements to the nation’s system for vital statistics will address these problems by helping us strengthen research, education and training.
A couple other opportunities for improvement. One is it’s very clear that death certificate data are entered by funeral directors. Since this problem has touched my own immediate family, I can tell you they don’t always enter suicide if they believe it is protecting the family. I have seen the same scenario with a very close friend who was on active duty when he took his life. It is very important to recognize that we need to know more about this so that we can begin to appreciate and address the impact of military services on families. Obviously, when a veteran dies of suicide, it shatters the entire family structure.
Again, finding the 14 is a very big deal to us. We, working through the National Alliance to Prevent Suicide, have lots and lots of partners to work with, but it would really go a long way in helping us work with them, medical hot spotting, if you will, so that they could identify in their communities where their outreach efforts would be most effective.
I will just close on one point. My husband, who reads four or five newspapers a day — he is more into politics than other things — was completely shocked a year or two ago when I told him that the greatest prevalence of suicide among veterans was in those over 55. He couldn’t believe it because what he read in the newspapers was all focused on younger people. Imagine the power if we had the ability to share that information with communities. We are connected with associations of police chiefs and all kinds of organizations that very much want to help veterans, but we need better data.
Thank you very, very much for your attention. This has been like a reunion in the happiest possible way for me, hearing about data challenges.
DR. COHEN: Thank you for your insights. Steve?
MR. S. GOSS: I am Steve Goss, actuary at Social Security Administration. What do we do? First of all, I’m going to have to talk to Bill later about the difference between actuarial data and real data.
(Laughter)
At Social Security Administration we have to try to project for the purpose of Congress to understand what the future costs and revenues are for Social Security and its financial — I shouldn’t say viability, but the extent to which they are going to have to make changes in either the revenues coming in or the benefits going out.
There are actuaries at CMS — and thanks so much to Cathy for having described the nature of a lot of the death data that we compile together. What I want to talk to you more about really is the way that we make use of all of these various sources of data. There’s essentially nobody on this side of the panel here that doesn’t produce material that we use. We use everything that’s available from CDC, certainly from Homeland Security and everywhere else.
Our primary purpose is to try to project the cost of Social Security and the revenues coming in, and the good news is that that’s relatively easy because basically Social Security covers the whole county. Over 95 percent of everybody in the country is going to be paying in and/or expected to receive benefits at some point. So, total national level data is what we really want. We generally get our data not so much at a micro level from the vital statistics world, but more on a macro level.
Our primary purpose really is to project the overall population of the United States and the projection of the population by characteristics. I’ll get into that more in a minute about how critical that is. We project the population and it is used for our projections of the costs of Social Security, and it’s used also by the actuaries over in CMS for the cost of Medicare and also for SSI and Medicaid, and for the President’s budget itself, and the longer-term projections, our population projections are used for that purpose. So, having high quality, accurate, sound data available for all the vital statistics upon which to understand past trends and, therefore, our best estimate of future trends is actually critical.
Demographics it has been said is destiny, and it really is in this case because, in our case, as pretty much is true with Medicare also, the cost per beneficiary of benefits and the amount of revenue per worker tend to rise at roughly the same rate. Medicare may be a little faster on the benefit side; Social Security is pretty much at the same rate. So, really, it’s the age distribution of the population that is so critical
This graphic is something we refer to in demographic circles as the aged dependency ratio. Because Social Security, Medicare and a lot of programs are financed on what we call a pay-as-you-go basis, pretty much the workers of the day paying in money is what’s covering the cost for the people who are seeking the benefits. You can see this aged dependency ratio; it is simply the ratio of the number of people in our population 65 and older to the number of people of working age, 20 to 64. That was going pretty steadily up until around 2008 or so when the baby boomers started to retire, and you can see the middle line which is our intermediate projection. Over the next 20 years that is going to rise like crazy.
You might say this is all because of the baby boomers. The baby boomers are retiring so they’re going to create a pig in the python kind of hump. Not at all. You can see it stays up. The reason it stays up is that what’s really going on with the aged dependency ratio is not just that people are living longer; it’s principally the drop in the birth rates way back in the 1960s. You see these lower two lines — If, in fact, birth rates had stayed at three children per one family on average, or what it was during 1946-1965, the baby boom period, at 3.3, you can see what the aged dependency would look like. It would be very gently rising, reflecting basically the increasing life expectancy that we have in our population. But this gigantic level shift that we have in this ratio is really because of the drop in the birth rate.
The easiest way to always think about that is, back in the day, there used to be three kids born to every couple and when they retired there were three kids in the workforce supporting them, one way or the other. In the future, it’s going to two. Two supporting two is going to be more expensive for them or less money provided to the elderly than it was when three supported two.
That’s just to give you a sense of the relationship between death rates and birth rates in the applications for the cost of these programs. You will see this looks very familiar — cost relative to the aged dependency ratio. The top solid lines are the cost of our Old Age and Survivors’ Insurance Program as a percentage of our tax base, taxable payroll. This is pretty much the same, as you would imagine, as the relationship of the number of elderly versus the number of working age.
The Disability Insurance Program does not fluctuate as much. If we had a lot more minutes we could talk a lot about where the demographic fluctuations affect disability.
The vital events that matter for our projects and for the cost of Social Security in the future, the births and deaths — and I was sorry to hear earlier that we are not doing as much about the marital status — they are entirely critical to us. Other ones are immigration and emigration — speaking of DHS — which are incredibly important to us, and also disability, another vital event which I’m sure we are not talking about here today, but we monitor that very carefully.
For births, of course, we go to NCHS. We get the births by age of mother. We look at these relative to the number of people deemed to be in the population of birth ages and develop birth rates. From those, we have to try to project what these birth rates are going to be in the future. The total fertility rate — which is effectively the average number of births by a woman if she survives the entire child-bearing period — you can see how that has fluctuated in the past. In the longer-term past it averaged at about three, but since 1965, birth control undoubtedly being something that became available, the birth rate dropped down to about two. A big question, the recession, birth rates have dropped considerably. Are they going to stay down, as many think, or are they going to come back up?
Deaths, mortality, the other really important part. We go obviously to NCHS for our death rates and, highly critically, death rates by cause of death, which is extremely important for us to try to make our predictions.
One of the problems with death rates that we have, though, is that we divide them by the Census population. If there is age misstatement or underreporting that is different between the death reports and the population, then we have problems. Partly for that reason, we go to our collaborative work with CMS about looking at death rates at age 65 and older where we are able to get from the same source the number of people exposed to death and the number of people living and potentially dying, and the number of deaths, and we think we have really higher quality. If we had more time, we could talk a lot more about that.
Projection of mortality. On this little graph, you can see that historically we had a tremendous age gradient in the rate of decline in mortality, much faster at younger ages than mentioned in the prior submitted infant mortality. In the future, we’re projecting much less of a gradient. Some other countries like the U.K. are projecting no age gradient whatever. We are projecting a faster rate of improvement — you can see on the blue bar in the bottom panel — at age 85 and older than we have had in recent years, since 1992.
Marriage and divorce data — Sharon and Mark will be back tomorrow to talk to you about the challenges and will share with you then how much we really wish we could have more information on that. I’m sorry to hear that that hasn’t been funded lately. Our benefits do depend on marital status for social security. Certainly, for Medicare they depend on it; even SSI and Medicaid.
Just an indication of how critical the marriage rates are — You can see our marriage rates have dropped like crazy, and yet the age-adjusted divorce rates have not really changed very much historically. So we’re keeping an eye on this, and monitoring this is really important.
One I think I did not have a slide on here is disability. We have data internally on disability and I’m happy to share that with others. It is critical to the projections of our cost.
Here is a list of sources of information that we have available on the Web. I am looking very much forward to hearing the rest of the discussion and any questions.
DR. COHEN: Thanks, Steve. Wonderful demographic summary of SSA uses of the data.
Earlier on there were several mentions of the Death Master File. We might get back to that discussion tomorrow, how it’s used, and its limitations. I know SSA is huge and there are lots of different programs, but that seemed to be a potential point of discussion as well.
Corryne.
MS. CARTER: Good afternoon everyone. My name is Corryne Carter, and I am here today representing the U.S. Department of State, Bureau of Consular Affairs, Office of Passport Services. It is a pleasure to have this opportunity to discuss with you the Bureau of Consular Affairs, what we do and how vital records play such an important role in that.
The Bureau of Consular Affairs is the public face of the Department of State to millions of people around the world. The Bureau is responsible for the welfare and protection of U.S. citizens abroad, for the issuance of passports and other documentation to U.S. citizens and nationals, and for the protection of U.S. border security and the facilitation of legitimate travel to the United States.
The Bureau’s Office of Passport Services issues U.S. passports to U.S. citizens and nationals. We protect the integrity of this important document as proof of U.S. citizenship at home and around the world while providing the highest level of customer service, professionalism and integrity. Our office consists of 26 regional passport agencies and centers open to the public and located in metropolitan cities around the U.S.; the Charleston Passport Center, a special issuance agency responsible for processing special requests for armed service personnel and other government personnel traveling for official U.S. government assignments; two passport production centers; the National Passport Information Center which is our customer service call center; a network of 9,100 public offices across the U.S. such as public libraries and post offices designated as passport application acceptance facilities; and a headquarters office in Washington, D.C.
In fiscal year 2016, the Office of Passport Services issued 18,676,547 U.S. passports. Just to note, we haven’t gotten our current fiscal year numbers in, but it is expected that we will either meet or exceed the 20 million mark.
Our services consist of renewal requests of customers replacing expiring passports and requests of customers who must apply for the first time. Our process begins with customers filling out a passport application form, gathering supporting documents, a photo and fees, and submitting a completed application to us for review and processing. Customers generally submit identification of citizenship evidence in the form of birth, death and marriage certificates and divorce and adoption records as supporting documents with their passport application form.
So, how do vital records play a part? The Office of Passport Services uses vital records such birth, death, marriage, divorce, adoption and amendment records to determine if a person is who they say they are, a U.S. national or U.S. citizen and eligible for a U.S. passport. Vital records are critically important and reliable to us when making these determinations because they document the details of the vital event near the time it occurred.
Recall my earlier statement that the Office of Passport Services issued over 18 million passports in Fiscal Year 2016. That represents over 18 million potential cases where we and our customers rely on vital records. Currently, our customers submit a request for a certified copy of their vital record to the appropriate state or local office. The customer then submits the certified record to us where it is reviewed and returned to the customer. We retain a photocopy of the vital record for auditing and future requests from the customer.
The customer information and documents we collect are sensitive and private information, and they are handled as such and are protected by the Privacy Act of 1974 and Department policy.
Lastly, but certainly not least, we rely on our longstanding relationships with Vital Registrars and their staff to resolve discrepancies and to verify and confirm records issues by their jurisdictions. We also rely on our partnerships with organizations such as the National Association for Public Health Statistics and Information Systems to confirm the registration of a vital event in a jurisdiction.
Thank you.
DR. COHEN: Thank you very much. Aaron?
DR. FIROVED: I am from DHS, but I am actually not from Immigration and Naturalization Services. We’re going to talk about a little different flair. I fall into those “and other users” kind of thing. I am the Director of the National Biosurveillance Integration Center. We are responsible to an interagency mission to 14 different departments and agencies. We have liaisons in several including VA, USDA, and relationships with many more. I’m going to give you just two brief slides on who we are so you can understand why this is important to us.
We integrate and analyze data from a variety of sources and we use that to monitor and report out to federal, state, local and territorial, tribal government officials and recently some private sector coordinating councils in our critical infrastructures. What we are trying to do is to try to help them understand what are the diseases that are emerging around the globe or acts of potential bioterrorism that could affect us here in the homeland. We are also trying to help improve the state of biosurveillance through acquiring new data sources and new data surveillance tools.
We get our information from open sources. That could be in multiple languages; it could be media reports, it could be state and local public health sites, World Health Organization. We could be purchasing things like travel data from the airlines, academic and scientific publications as well as our interagency liaisons, and then we’re distributing that. We are also fielding requests for information. We are certainly seeing some pickup from Harvey and Hurricane Irma as people are trying to understand what is the disease likelihood coming out.
I am not going to go through some of the technologies, all of them one by one, but just to say that we have a lot of information services that we need, and we need them in real time. They are underlying a lot of what we do, but they also have to be done responsibly so they need access controls. These need to be flexible and timely to deal with multiple populations at multiple levels at the same time.
One that we are doing in coordination with our Veterans Affairs colleagues is trying to understand how electronic healthcare records within the system for the care of veterans, when it’s protected, HIPAA-compliant, de-identified and aggregated, can then be used for biosurveillance, not only for the broader community but then to go back into the care of those veteran populations. That is the kind of information that has to have multi-level access control. I don’t want PII; I don’t want HIPAA-controlled information; that is not what we’re trying to do.
The same thing happens with a partnership we have with the University of North Carolina looking at pre-hospital data, EMS data, 911 call data, motor vehicle crash data and other information like that.
This is, frankly, where the state of biosurveillance is and, thankfully, vital health statistics as well. We are providing alerts in weeks to months if not a year out, limited or no access to data. It’s usually public dat-oriented and not a lot of private sector-oriented data. And we are talking about what happened. Like many of you, we are closely following Irma this week. I have family that was in the Tampa area in the bull’s eye. They were allowed to prepare because they could see this coming a week out. They could take the necessary steps. That’s where we want to be with disease surveillance. That is where we need to move toward.
This is some of the data that we need. We need demographic data; we need death certificates; we need congenital malformation type information. I am just going to cherry-pick from some of these to show you that through the use of these we can start doing analysis of trends, signals of anomalies, understand the location and translocation. Diseases follow different pathways through different communities, and so understanding the communities and what’s going on in those is incredibly important.
This has already been mentioned several times today — Zika in the United States. In 2015 when we first started reporting it in Brazil, it steadily spread through South and Central America through the Caribbean to the United States. This was an interesting one. Most of the people who came down with Zika were asymptomatic. You didn’t know it. You couldn’t screen for it. But it had an incredible outstanding effect on pregnant health and neonatal healthcare. Understanding which population is at risk is incredibly important to us, and you need to do that, again, in real time if you’re going to have an intervention.
This is another one from 2014 – Enterovirus, EVD68. Just understand the case definition of this. It is in a patient 21 years old or younger with the acute onset of focal limb weakness, no earlier occurring than August 1, 2014, and an MRI scan suggesting a spinal cord lesion primarily in the grey matter. You had to meet all four of those criteria to be classified with this. So, understanding accurate death data, timely, understanding which communities — it’s really hard to pull this picture together sometimes. Again, this was primarily for the children population.
Avian influenza we heard about today. At any given time there are probably a dozen different hotspots for avian influenza that we are looking at, any one of which can jump into people. Trying to understand, of those that start jumping, are they primarily going after an elderly or infant population. That is often the case. But every once in a while we see a shift where it’s the children and young adults that are the primary ones at risk. So, understanding which of the ones that are suffering ill health effects and then, frankly, the deaths in which of these populations helps us make those interventions and helps allow us to prioritize vaccinations and which groups come first if there’s a limited supply.
And opioids — we have all talked about this. This is what we have been doing with the University of North Carolina. States have mandatory reporting requirements. They need to understand who is overdosing, what were they overdosing on, was it heroin, fentanyl, prescription opioids. They are trying to gather this data and they’re looking at all information sources they can.
Finally, vital health statistics are incredibly important to what we do in biosurveillance. They need to be timely and they need to be in real time. They need to be accurate. And it’s going to be the way that we transform this reactive reporting to projecting where these diseases are going to move. Thank you.
DR. COHEN: Thank you, Aaron. Questions or comments from the committee? Again, if you audience or folks at home have any questions or comments, please use the online website.
DR. MAYS: Thank you very much for your presentations. I have a couple of questions and I would like to start with Carolyn.
One of the issues that I heard when I was on the Board of the Health of Select Populations about veterans is that it is not the veterans who are in care but the difficulty you have is with those veterans who are out of care. Has there been any discussion with NCHS about what datasets they have that you might be able to insert a veteran question and potentially funding from either you or DOD to help with that?
DR. CLANCY: Great question. I neglected to mention that my brilliant colleague — and anything I said that sounds remotely intelligible is all due to Ira Katz, who is sitting over here and will be able to stay around longer than I can.
We, at the moment, are entering into an arrangement with AHRQ for the medical expenditure panel survey partly because we need a deep dive on costs. But it would be enormously helpful — back on the topic of suicide and the veterans who do not get care in our system — if there were some information about whether they had health insurance or not.
So, if we got more timely information about deaths and had information about health insurance, we might have some clue about what kind of care they were getting and so forth. Again, trying to figure out how do we reach this group, some of whom may be getting terrific care and others not so much, particularly those folks with other than honorable discharges who, I would have to guess, are likely to be in and out of care and so forth.
Yes, we have had that kind of conversation and are looking for any information we can get, for sure. Many veterans who are enrolled in our healthcare system use us for some of their care but not all of it, particularly in the Medicare age group.
DR. COHEN: Any other questions or comments? Anybody on the phone with questions or comments?
DR. MAYS: I actually wanted to ask a question about the Social Security Administration. If you could have your big ask, what would it be? That’s what I don’t understand, is exactly what you need in order to do what you do better in terms of mortality or birth data. What is your ask?
MR. S. GOSS: Perfect, real-time, absolutely on-target data that is complete.
(Laughter)
But there’s a ranking. The birth data and mortality data that we get, especially combined with what we get at older ages from Medicare, is pretty good. What we are really lacking at this point is the marital status. Marital status is pretty critical to us and there just hasn’t been much of that lately.
I guess on the mortality, we have this issue which we think is a lot less now than it used to be — again, Sam Press, University at Pennsylvania did a lot of work early in censuses looking at age misstatement. There’s a tendency in people who are under 21 to say they’re a little bit older. There is a tendency also when people get to be very old to say, hey, I’m going to say I’m 85 so you’ll say I look good for 85, and people tend to maybe overstate their ages in censuses.
Fortunately, in our records — which is why we think we have relatively better quality data for death rates at higher ages — if we have got you nailed down as to what your age was at a much younger age and then we’re following you through time, we really know what your actual age is. But if a census comes along every 10 years and you’re asked how old are you, you might say almost anything.
Similarly, on the death reports. It depends on who is actually filling out the death certificate and what is the age indicated. How accurate is that? We all know there are things like sort of a confluence of year of birth tending to end in a zero or end in a five in a lot of reports. Some of these kinds of things create difficulties in the data.
So, really, more accurate and more complete data would be better, but the marital status is the place where we are really kind of suffering at the moment.
One comment also is on race and ethnicity, which was mentioned a little bit earlier. I remember back — I forget how many decades ago it was — when Census said it was going to start saying don’t put down the race/ethnicity that best pertains to you, but put a check by anything that you think you might be closely related to you, and at that point there seemed to be a breakdown in the quality and our ability to really get a good sense of race and ethnicity.
For us, the area where race and ethnicity is probably most important is actually probably in birth rates with respect to ethnicity versus the rest of the population. Black and white — it used to be many years ago that there was a big differential in birth rates. Not so much anymore. Ethnicity is a really important characteristic to us. We do not have great information on race and ethnicity breakdown in the datasets. I’m not sure that we ever are going to have that in the future.
DR. COHEN: Although there will probably be changes in the Census so we get to revisit ethnicity all over again.
MR. S. GOSS: Speaking of changes, also, over time reporting cause of death, there was some mention earlier about maybe sometimes funeral homes are not eager to report certain causes of death. Just our understanding of causes of death has changed, so we like to look over long periods of time, especially the material we get from NIH and NCHS. Are there certain causes of death which now are reported much more because we understand them? You can’t report what you don’t know about. So that’s one of our concerns in understanding the data and the trends.
DR. COHEN: Go ahead, Gib.
DR. PARRISH: Given the point you made about age misstatement particularly among younger persons and then Vickie’s questions related to race and ethnicity discrepancies between birth certificates and death certificates, has there even been any thought given to, in fact, attempting linkages between birth certificate and death certificate at the time of death, at least for those persons for whom the birth certificates would be electronic at this point?
We do infant death-birth linkages now. But it seems to me we’re getting to a point where you could, in fact, try to link a much more complete birth-death for most deaths.
MR. S. GOSS: One point on that is, as was mentioned I think on our panel, Social Security has, in fact, been getting people, at birth, right there generally in the hospital, to issue the Social Security record at that point. At some point in the future, we really are going to have these data in really good order.
DR. COHEN: Is there anyone from NCHS or any of the states who are aware of general programs? Dr. Saadi.
DR. SAADI: Within our processing we do merge birth and death data. It is a way for us to prevent the issuance of a record that doesn’t belong to someone. It’s really important particularly in military deaths. We get those kind of haphazardly actually. But that is a recommended standard for our security processes, that all jurisdictions link birth and death records so that we don’t issue on someone inappropriately.
DR. COHEN: Do you know how many jurisdictions are doing ongoing linkages of all their birth and death data?
DR. SAADI: I don’t have any numbers for sure, but I can’t imagine that we don’t have just about everybody doing it that have got electronic systems, that have databases. Everybody has a database.
DR. COHEN: So there are jurisdictional systems that do have these data.
DR. SAADI: Yes. And they are linked. It’s a recommended feature our vendors have as well as practices that we do.
That’s why when Shawna mentioned the EVE system where we exchange data between each other, we link those records. When somebody has died, say, for me in Missouri, it’s really important to tag that record and link it with that unofficially provided information so that somebody doesn’t come in and ask for that person’s birth certificate and then go off and get $60,000 worth of credit.
I can speak definitely for myself, but I think all the jurisdictions — I really can’t speak to the territories, but I think most everybody. D.C., I think you guys do as well. It’s a recommended security practice, and that has been on the books for quite a while for us to look toward.
DR. COHEN: Great. Charlie.
DR. ROTHWELL: I am Charlie Rothwell. I work at NCHS. There was something called the Intelligence Reform and Terrorism Prevention Act that was passed, and part of it had to do with vital statistics, which was never really appropriately acted on. One of the ideas was to take the National Death Index where we have all the records and give them back to the states, and give them back to the state of birth. That would allow them to do just exactly what was being said by Lou in order to be able to, from a security perspective, make sure that a birth certificate in fact was not inappropriately issued because the death had already happened.
That never took place. But if it did take place, then in fact what we would have is a database at the state level that could do just the type of analysis you’re talking about. It was intended from a security perspective, but one of the benefits of that type of activity would be to look at differences in the reporting on the birth and death certificates and improve it.
When we look at infant mortality rates in the United States we look by race. We don’t look at the death certificate; we look at the linked birth and death file. That gives you the appropriate outcomes for infant deaths. So this is a real issue and it’s one that could be solved.
DR. COHEN: Thanks, Charlie. Nick.
MR. COUSSOULE: I have one question. I hear a lot of folks today talking about the timeliness needs of the data. One of the challenges of dealing with complicated data at any scale is the curation and validation of that process which tends to take a long time as well.
How do you all think about the distinction between perhaps getting it faster with maybe less validation and verification versus waiting until it is better, for lack of a better term? You obviously want to do both of those things at the same time, but if there are trade-offs, how do you view that aspect of it?
MR. ATKINSON: One of the things that happens at this particular point is we encourage states to send us the records immediately after they are registered, and then after any updates to that record send us another copy of it. That’s why I mentioned that we go through — in deaths, for example, which is a big one, we go through about seven and one-half transactions to get to the 2.6. It’s because of these updates that are coming in. As the record gets better, the state is to send us additional information, send us another copy of that particular record and we can process that record and update our files.
DR. RILEY: I will say from the researcher perspective, this normally happens at the end of the study. The study is over with. We wait until that is done. We batch the requests from all of the death records that we need to get and pull them all out at one time, which produces that lag that Mike was talking about. It takes years sometimes before all of that data is in as opposed to being able to do this on a more case-by-case basis or in smaller batches over the course of time.
DR. COHEN: Again, I think fact of death is different than cause of death or underlying cause of death versus multiple causes of death. So, depending upon what your specific objectives are in the studies there might be trade-offs —
MR. ATKINSON: And specific causes. As the medical examiner talked about this morning, there are certain causes for which you are waiting for the test results from the medical examiner, and once they get that, that information gets back to the records office and then that office sends that information on to us.
DR. COHEN: So natural-caused deaths get filed a lot more quickly than external-caused deaths.
MR. GOSS: I think we are about to get the real answer, but in Social Security I know we get preliminary and final both birth and death data, and we make use of both, so we don’t want to stop any of this.
I think, generally, the preliminary data turn out to be pretty close to what happens in the final statistics. Not as much detail.
MR. ROTHWELL: The two are tied completely together. This is an opportunity — Right now, at 10:00 o’clock this morning, we released a report on drug overdose deaths for the United States. It was through February of this year. It had all 50 states plus 18 states with specific drug-related events. These are counts and percentages; they are not rates. The data is pretty solid.
Why I bring this up is that we should be even faster than that, number one. Number two, we only had 18 states that had information enough to give us the specific drug overdose deaths.
There is an opportunity here now that when Delton’s organization is dealing directly with the states and going back and forth with them, we can, through our electronic systems, not go back and forth with the states but through systems to deal directly with the field director, the certifying physician, the medical examiner to get the information that we need. We are right on the cusp of this.
So, the speed-up of the systems at the state end and local end and at our end gives us the opportunity to improve data quality. That is going to be the payoff on timeliness I think, is finally getting vital statistics that you really feel good about from the cause of death perspective. That’s my opinion.
MS. KLOSS: I think my question is relevant to this last exchange, and this is a question for any of those who are using the data. To what extent are you concerned about variability of quality across jurisdictions? We have talked about the data as if it is one thing, but it’s the composite of 57 things that are probably quite variable in their own right.
I just wondered, for Delton, what capability — and maybe this is for tomorrow’s discussion, but how do you assess that, and how are you controlling or finding that to be an issue or not an issue?
MR. ATKINSON: It is an issue and it’s one of the things I am going to talk about tomorrow. There is a great deal of variability between states. We have some states who do a fantastic job based on the quality measures that we look at. Our systems will produce all kinds of indicators about the records that are coming in.
We have some jurisdictions where almost every record has a problem with it.
MS. KLOSS: Downstream, then, as you incorporate this data — because this is a great classic case of data use. Collect once, use many is what we have heard here, but with those caveats, and are those well known to everyone using it.
MR. ATKINSON: The other thing is that my staff down at RTP is focused 100 percent on working with the states with problems to try to resolve each and every problem on a record.
Now, do I have enough resources to solve all of those? No. Do the states have enough resources to solve all of those? No. But, as we move towards our electronic systems it is my contention that we should not receive a record that does not pass the NCHS edits, because minimally those edits ought to be built into every electronic system out there. I can tell you that is not the case.
MS. KLOSS: Just one comment that struck me and this is not my area of expertise. This morning we talked about finding two cents here and three dollars there and four dollars here, and then this afternoon we heard about these absolutely life-enhancing uses, and there is such a huge disconnect between making the case for the importance of this and so on. That’s probably tomorrow’s opportunity discussion.
DR. COHEN: Yes. And there will be lots of discussions, although NCHS and NAPHSIS and the jurisdictions work really hard at resolving all these issues, so, ultimately, the end data users I think get very consistent and clean files. But the devil is in the details.
For instance, all the cause-of-death coding is centralized; everybody does that the same way — but, to be discussed more tomorrow. Helga?
DR. RIPPEN: Again, thank you so much. It was very informative.
Just for a show of hands, how many actually only use data relating to vitals from NCHS? Okay. I guess the question is how many are getting vital statistics directly from the states? Okay.
The only reason I’m bringing this up — and maybe I used the wrong acronym — is, if we’re getting data from different sources and making copies of that data, would that be fair?
DR. BAUER: When we get data we have a data use agreement with whomever we’re getting the data from, mostly from NCHS, that specifies what the uses of those data are.
DR. RIPPEN: Okay. But are people getting data from other — are you only getting it from NCHS?
DR. FIROVED: As the alternative user here, frankly, it’s not fast enough, so we are in the midst of something a lot of times and it’s directly from reporting by states, whether it’s the communities themselves reporting out.
You were talking about fast versus completeness. We make that tradeoff daily. We try to look at reputable sources, but you need the best picture you can get at the time, so we actually look for a lot of alternative sources to get an idea of what’s going on so we can start making decisions now.
DR. RIPPEN: Again, it goes back to efforts and quality. If one group is improving the quality of the data through different methodologies — for example, Native American Indians — and you have this added value but it’s only used within that, then everyone may have a different picture because not everyone is using the same exact data source. You’re getting it from similar sources but you’re augmenting it. There’s a nuance here, right?
So, again, I’m just wanting to understand who is getting data from whom and the implications, because if you are improving data on one end, what that might mean in the long term for other organizations who might be using it a little different.
And then stepping back, just from a structural perspective, because everything has a cost. Right? We’re talking to the states about a federated — how can we speed it up, how can we leverage it? It seems like the agencies may have a federated potential, too, with regard to if the effort is perhaps focused — and maybe that’s for tomorrow. But again, just trying to understand if we can focus on ultimately what would the need be — like what is vitals. Is it the health information or is it just all the legal stuff and identity? And who can facilitate that so that it can be used? And every agency has its own purpose and wants to augment it.
It’s just something to think about because I just want to understand.
DR. BAUER: An example might be we have a keen interest in maternal mortality, so we would partner with a state and fund a maternal mortality death review where we would compile information from a whole host of sources. That doesn’t ever get back to the official death certificate to sort of enrich what is on the death certificate, but we would have that additional understanding of those deaths.
DR. RIPPEN: But just imagine, especially when we’re talking about working environments and collaborative groups for accelerating innovation, if you were able to share that and others would be able to understand that adding this particular piece of data can have such a major impact on another agency. Then, when you start thinking about, well, where are the resources and how do you want to leverage them to slowly kind of build up, you are able to accelerate it.
Especially with some of the work that you did with the Native Americans, it would be great to include that because other agencies might be very interested in that. But that’s actually really great to hear.
DR. COHEN: To be continued. I have one question and then we will end with Roland. I’m sorry —
MS. WEBSTER: I just wanted to make a clarifying point, which is that we have two groups of people, those using the NDI which does come from NCHS and the states get paid through that contract, and then you have the more specific programs using a data use agreement contract, and most of those specify that the data cannot be re-used for another purpose. That is an aspect I wanted to bring to light.
And then you have Passport who uses one of our systems, so the states are getting paid through Passport as well. But again, there is a very strict non-reuse aspect to the contract.
DR. COHEN: What system does Passport use?
MS. WEBSTER: EVE.
DR. COHEN: So you use EVE to verify —
MS. CARTER: Yes, in limited circumstances. Again, I think you can have very narrow use agreements, but as long as you can start building that in and you create a governing process that allows them the ability to discuss.
DR. THORPE: I just want to make a comment. I, too, like some of my colleagues around the table, am hearing some dis-concordance in messages, and I just look forward to the opportunities that we will be able to have some frank conversation or try to get some recommendations on the table to move forward tomorrow.
Aaron, you said, frankly, this takes a long time. Can you clarify?
DR. FIROVED: Whether it’s opioid or Zika, there is the information that people need to make a decision now about whether to start a public messaging campaign. Months later, we will get information usually through CDC and MWRs and whatnot that will actually come out with the solid data. But we are often put in the position of I need to know right now, and sometimes it’s the case I need to know right now within two hours, and I assemble the best picture I can. It’s a picture that tries to contextualize everything, so, deaths, illness, what is the frequency right now, so we will cobble together the best picture we can give the policymakers. But these are people who are trying to act to mitigate, so it is not fast or accurate; it’s both. Further down, when it’s something like an influenza outbreak that’s going to have a lot of phases, you have more time.
DR. COHEN: I just wanted to end with one clarification that I need. There were several comments around the funeral director being responsible for cause of death information on the death certificate. I thought that responsibility was for the medical certifier, for a physician or someone else rather than the funeral director.
PARTICIPANT: Thank you for bringing that up because I was going to bring up the same thing. I spoke to Robert about that.
I don’t believe any funeral director will ever write the cause of death on a death certificate, or the manner in which — somebody mentioned suicide. It’s a coroner. They could be both, but they’re not doing it as a funeral director; they’re doing it as a coroner. So it’s the coroner in medical legal cases — we talked a little bit about that. It’s the coroner or the medical examiner who is certifying the cause of death and the manner of death, not the funeral director.
DR. COHEN: Or the attending physician. There is never a funeral director in his role as funeral director who would be determining cause of death. Thank you for that clarification.
We are on break, 10 minutes.
(Break)
Agenda Item: PANEL 3: Current Status for Vital Records: Other Users of Vital Statistics Data
DR. COHEN: Before I introduce our final panel, Nick, someone came up to me and wanted to address your question as well about the tradeoff between timeliness and quality. NDI — I don’t know whether it was clear — issues a provisional file to turn around rapidly and then you get a freebie for the full file when that’s available. So that’s an example of a way, by issuing provisional and final data, to address both of those issues.
Our final panel is the “other users” panel, Other Users of Vital Statistics. Thank you, all other users. It starts off with Rebecca Coyle, Executive Director of the American Immunization Registry; Michael Warren, President of the Association of Maternal and Child Health programs and also Deputy Commissioner for Population Health in Tennessee; Mohsen Naghavi, Professor of Global Health at the University of Washington; Rebecca Russell, Director of the Perinatal Data Center from March of Dimes; Susan McDonald, President of Pension Benefits Information; Edward Hall, Associate General Counsel for Compliance for Verus Financial; and Steve Emmert from LexisNexis.
Thank you all, and thank you for your perseverance. It has been a long but I think incredibly productive day. Rebecca.
MS. COYLE: Thank you. I am happy to be here this afternoon to share our experience with vital statistics. I am going to start off with just a very brief overview because I am not sure what your familiarity is with Immunization Information Systems, often known as Immunization Registries.
To give you just a brief overlook, it’s a confidential population-based system. They help with a variety of public health efforts including identifying pockets of need and exchanging data with multiple providers primarily in the medical arena. They also help with schools and childcare stability and make sure that they have the immunizations needed to attend, and there’s a variety of other administrative efforts that are utilized by using the IIS.
From here I want to go into what we do and how we fit into this space. I am with the American Immunization Registry Association, and we are the member organization that works with all of the state and jurisdictional departments that implement and utilize an immunization information system. We have a long history of working with our jurisdictions — the states and local health departments — that operate a system. We were essentially designed to help stop the reinvention of the wheel, if you will. I can’t say we have been 100 percent successful but we really provide that community of practice for the states so that when something is done in one place we can hopefully replicate that in another.
In terms of the quantity of systems that are out there, there are about 63 different systems out there. That includes all of the U.S. states, the territories, and there are also some large cities and counties that also operate immunization registries. I will say that there are 64 immunization awardees that receive funding from CDC to operate their systems.
I should be clear, the 63 and 64 are pretty close. There are a few variations and it comes down to like two or three different variations in those, but we will often use the term 64 awardees to describe the folks that we work with.
So, what do we use the data for? The primary uses include populating the IIS with demographic birth information. We consider this the gold standard of information; we want to get this information in as soon as possible. If you think back to that first slide when I was talking about it being a population-based system, this is what makes our system population-based, that we are able to capture the birth information.
It also helps with managing that denominator — removing individuals who have passed away who are no longer in that area. We want to be able to take those folks out. And, also, the adoption data.
Why are they important? Populating the database with the birth demographic information is key for ensuring that the systems are, in fact, population-based. There is also additional information that’s populated in some jurisdictions, not all, which include the birth dose of Hepatitis B. Ideally, every baby is getting a birth dose of Hepatitis B in the hospital before they go home, within hours of birth.
It also happens with identity management and name changes. For the epidemiology side of things, in case management for any infant who is born to a mother that is Hepatitis B surface antigen-positive, we want to make sure that we’re collecting that information in the system because it is going to help with case reporting and it’s going to help with tracking that individual over time. They have a very rigid set of requirements and vaccines that they have to meet within a period of time, and immunization information systems help when some of these kids get lost in the system or maybe fall off.
The other piece that is sometimes collected but not always is HBIG administration. Again, for those babies born to surface antigen-positive moms, we want to make sure that we are collecting that dose of HBIG as well.
I also want to just point out that in terms of identifying the population or pockets of need, I am going to share an experience from Minnesota back in 2009. Please note, this all predates the measles outbreak that just occurred in Minnesota earlier this year.
In 2009 there were reports of under-vaccination in Somali children, so the Department of Health was able to work with Vital Statistics and they shared information there. They were able to match the birth information and identify whether a baby was born to a parent of Somali decent. They would like to see if the mother was born in Somalia or the father was born in Somalia, and from there, we were able to identify that child as being in that population.
As it turns out, the data actually reinforced the suspicions and rumors that were occurring, so it’s a really good example of where there was a pocket of need that was suspected but that also the data was able to be utilized to back that up. And they are hopefully able to start preparing the material in time.
So, what are our experiences with Vital Records? As I mentioned, we have a long history with Vital Records. IIS and Vital Records are one of the very early data exchange partners. Long before we were doing electronic health data exchange on a daily basis, this was the first type of data exchange.
Just to give you an example of how much data is going in and out of there, 49 jurisdictions reported receiving information from Vital Records. This is for last year. Fourteen jurisdictions reported not to have received information. There are seven states that were not reporting that information and seven territories that were not. Last year, New Hampshire did not have an IIS. The yellow indicates the states that are not exchanging data with IIS.
In terms of timeliness, I want to pull this out because it’s one of our functional standards to make sure that all immunization data is put into the system within 14 days of administration. This can present somewhat of a challenge for us. We are getting really close to closing in on that. You will see that a majority of the data is going in within less than 14 days; however, there is still a number of records that go into the IIS outside of that. This is the percent of records in an IIS that were established by Vital Records, so it’s pretty substantial.
How does the data transmission actually work? Unfortunately, we have somewhat of a clunky system in that each jurisdiction has to negotiate their own agreements, and that is generally within the jurisdiction. So, between vital records there are a number of different policies that might be in place. It may be an MOU, it may be an interagency agreement, it may be just a data use agreement. And there are still some that don’t have any of those at all. We know that will probably create some confusion for folks. There are system barriers there in terms of updating that to include additional data elements.
I am just going to give you a sneak peek at what the challenges are, but my colleague, Mary Beth Kurilo, will actually tackle this tomorrow. It probably just confirms most of what has already been said to date, but I do want to just say that, from the IIS community perspective, we value the relationship we have with Vital Statistics. We conducted an informal survey with our members and I just want to say the positive comments that we received back about that data exchange were just overwhelmingly in support of Vital Statistics. It is, again, a very valued relationship. Thank you.
DR. COHEN: Rebecca, let me just clarify, those agreements are with all the individual states — the state vitals, the state health department or the state registries of vital records. Great, thanks.
DR. WARREN: Good afternoon. I’m Michael Warren; I’m a pediatrician and Deputy Commissioner for Population Health with the Tennessee Department of Health and am here this afternoon representing the Association of Maternal and Child Health Programs, or AMCHP. I will be giving you some experiences from some of our member states as well as some specific examples from Tennessee. AMCHP is the national professional organization for state public health leaders and others who are working in maternal and child health programs across the country.
One of the most important ways that our members use vital records is to inform public health programs and policies. As you can see in this screenshot, in my own state of Tennessee, in 1927, the causes of death were very different than they are in 2017. They included infectious diseases primarily like typhoid fever and tuberculosis. Fortunately, we don’t have lots of folks dying from those things today. Those causes look very different. Other chronic diseases and injury lead the way there. As those leading causes of death change, so do our public health programs and policies that are aimed at addressing those.
Maternal and child health professionals across the country are very interested also in the other side of this data, in the birth certificate data, particularly birth outcome data such as gestational age, birth weight and complications associated with delivery. As an example — and you heard about this earlier — many states have recently engaged in efforts to reduce early elective deliveries. Those are induced deliveries or C-sections occurring before 39 weeks without medical indication. Vital statistics data was used to track states’ progress both within states and across states in reducing these unnecessary procedures that contribute to poor birth outcomes.
Vital records can be extremely useful in identifying emerging trends. For example, death certificate data analysis can alert public health officials to increasing suicide among adolescents and young adults or increasing overdose deaths from opioids. Another example would be using birth certificate data to monitor the incidence of birth defects. This has become particularly important to states and territories over the past 18 months with the emergence of Zika virus and its associated birth defect, although I think it points out an important challenge in that microcephaly, which was the predominant birth defect, is not actually one that is captured on the current standard birth certificate.
These vital records are powerful in their own right but they can be made even more useful by linking them with other datasets. For example, linking with other datasets that provide detailed information about factors such as poverty rates, employment and education can help inform us about the impact of social determinants of health on birth outcomes and deaths.
In my own state of Tennessee, we recently conducted a novel analysis funded by CDC and March of Dimes using vital records data linked with Medicaid claims data and educational outcome data. As a state hit hard by the opioid epidemic, we have a large number of babies born with neonatal abstinence syndrome. These are babies that are dependent on opioid substances that their mothers used during pregnancy. The linked analysis allowed us to show that infants with NAS are more likely to be referred for evaluation of developmental delay and to have a diagnosis of developmental delay as they enter school. This analysis would not have been possible without this unique linkage across these multiple datasets.
But, even linkage across distinct vital records can be helpful — in a classic maternal and child health example, the use of linked birth certificates with infant death certificates. Infant mortality has long been a core public health measure and these linked datasets can help identify important prenatal and perinatal risk factors that increase the likelihood of infant deaths.
State entities also use vital records datasets for various administrative functions, for example, verifying the identity of individuals, helping to determine the eligibility for certain benefits programs, and this often involves collaboration across state and federal agencies beyond public health and the use of various data-sharing agreements.
As you have heard, there is substantial variability in the current vital records experience across the country, and many of our member states wanted us to share this with you as a concern today. While many states have fully implemented electronic death registration systems, others are still relying on paper death certificates that are then manually entered into electronic databases.
There is also substantial variation in data output. Most states, of course, have moved beyond publication of paper vital statistics reports, but the online availability of vital statistics varies. A few states have robust data portals that allow the public to dynamically query vital records data, while many others have more static websites that feature prepared data reports, and then if individuals need other analyses beyond those, they have to request those through their public health department.
As you can imagine, the use of paper certificates and manual data entry impacts the timeliness of data being available for analysis and also increases the likelihood for human error in data entry. Data entry often starts outside of health departments, and the quality of the data is dependent upon the skill and experience of those who are doing the data entry. Some of these positions have very high turnover requiring frequent retraining and transfer of procedural knowledge. Even when those entering data are skilled, they may be limited by the quality or completeness of information that is available to them, such as from medical records or facility worksheets used to complete birth certificates.
It is not typical to have real-time vital records data available for public consumption. States may have delays from six to 18 months following the end of a time period before vital records are finalized and made public. Even when data is made available in a more timely fashion, comfort with use of provisional data varies by jurisdiction. This may be due in part to the broader public’s understanding of the nature of provisional data and the propensity for such data to change. Some public health departments are comfortable releasing provisional data to the public, although more are likely to use such data internally to drive programs and policies.
You have heard about this earlier — the integrity of vital records is assured through a number of mechanisms. Both physical and electronic security measures assure the access to vital records is limited to those with appropriate authorization. And, as you have heard, many partners outside of state government — academic researchers being one example — are very interested in accessing vital records data for analysis. Data use agreements in review by institutional review boards help to assure that data are stored appropriately and released according to agreed-upon parameters. Such agreements also specify the procedures that are to occur in the event of a data breach.
As we look forward to the future there are a number of opportunities for strengthening the Vital Records System. First, there is the need to assure that all states are using modern systems that allow electronic processing from start to finish. States will likely need additional resources to complete this work and to maintain the highest level of functionality on an ongoing basis.
Second, any systems development work needs to be done with a focus on interoperability. This means assuring that various systems can talk to one another. At a basic level, this allows state and federal vital records systems to communicate in a standard way. However, there are other opportunities for enhanced data acquisition. For example, interfaces between hospital electronic health records and birth registration systems could allow the automatic upload of data, minimizing the resources required for data entry and improving the timeliness and quality of data entry.
These efforts will go a long way to improving data quality and timeliness. Better data and data made available in a more real-time fashion will further enhance the availability of public health to identify emerging issues, explore root causes of morbidity and implement public health programs and policies aimed at improving the health of populations.
Finally, we must continue to think about training the various external end users. This not only includes hospital birth clerks and funeral home directors but also physicians who certify death records and who could benefit from additional training in the proper completion of death certificates.
We hope this input has been helpful and we appreciate the opportunity to be here and look forward to your questions.
DR. COHEN: Thank you. Are you going to be around tomorrow to discuss some of these issues? Good. Mohsen?
DR. NAGHAVI: Thank you. I will discuss about effect of the causes of this data in the U.S. on the global burden of diseases. But before I talk about this I will talk about what is the global burden of disease.
Global burden of disease is a scientific effort for the quantify comparative magnitude of health loss in the different countries by the different causes and different injuries.
We estimate we have every year for 315 diseases in the 519 geographical locations, including all countries – we included many countries, including the U.S., China, India, Brazil, as a subnational. We estimate every year for the subnational location for these countries. Our estimation is from 1990 every year, until 2016, 2017. We have to estimate every year for all locations and for these diseases. The burden of disease has very important implication on policymaking in the different countries, and now they are using these results.
All of our results are publicly available because we believe that data is a public good, and the result is a public good, and all of our results are publicly available and on the internet. Also, we publish every year in the Lancet, one issue of Lancet for the result of GBD and the result of GBD 2016, we publish next week in the Lancet in one issue.
This is an example, for example, what is the pattern of the causes of deaths or years lost by country. This is different causes for females and for males. This is from 2016. The most important thing is cardiovascular, ischemic heart disease, in some countries war, some countries TB. Also, we have this result by the county level in the U.S. from 1980 until 2014, 2015, or 2016. We will publish this September. It shows the pattern of mortality by each county, and the percent change from 1990 until 2014 here, for some examples. This is am example for ischemic heart disease by county.
But the problem is that in the causes of death data that we are using from your data, there are some things that are not underlying causes of death because, based on the definition, we have to include underlying causes of death. You have lots of causes that are sepsis, senility, headache, infertility, heart failure, low back pain and cardiac arrest, are not underlying causes of death. This is a big challenge that we have to fix because we have to compare country in the different years and the country-by-country comparability is the big problem.
This is some example for Japan by sub-national, Mexico by sub-national and U.S. by sub-national. You can see that some data in some states in Brazil, quality of data is higher than some states in the U.S.
This is an example — here is Oman, here is California, here is Mexico City, Australia and Japan and Russia. U.S. is not the best but is one of the best data that we have. For example, this is completeness of the data of U.S. Completeness of the data of U.S. is very good. But when we look to the quality based on the percent of “garbage” or non-underlying causes of deaths, you can see that the U.S. is not the best. We have some other countries that are better than the U.S. based on the higher quality of the data, but U.S. is one of the top countries that we have.
From this morning, I heard many times about we want to publish number of deaths due to overdoses or opium overdoses. How you can do that? Forty percent of the overdose deaths in the U.S. are assigned to unspecific, unintentional poison. That is on the death certificate. Any publication in this area is year by year, state by state. Sex by sex is different. Forty percent average lower than reality. The quality of the data is very important. You are very timely, you are with very good coverage; everything is good, but quality of the data — who will work on the quality of the data? Thank you.
DR. COHEN: Thank you. Very interesting perspective. Rebecca?
MS. RUSSELL: It is my pleasure to be here today. I think all the presenters so far have really provided compelling testimonies to the importance of vital statistics data, and I am really honored to have this opportunity to add my voice on the critical need for these data for researchers and, specifically, at the March of Dimes.
I would like to start by recognizing the vital records and vital statistics community for the work that you do. I am constantly impressed at the complexity of the work, your passion for vital statistics, and, as a user and beneficiary of your diligence, hard work and commitment, we want to say thank you.
The March of Dimes is a unique collaboration of scientists, clinicians, parents, maternal and child health experts, community organizations, members of the business community and other volunteers affiliated with offices and staff representing every state, the District of Columbia and Puerto Rico. Our mission is to improve the health of babies by preventing birth defects, premature birth and infant mortality.
As the Director of the Perinatal Data Center at the March of Dimes, I am responsible for providing the organization with the data, evidence and expertise our staff need to advocate for better health for moms and babies, to evaluate and improve our public health programs, to identify and target areas for populations at high risk in order to improve equity in birth outcomes, and to communicate our mission story.
To do all of this we need strong, reliable and timely public health data sources. The March of Dimes Perinatal Data Center relies heavily on the natality and linked birth infant death data as the primary source to drive our work, which is focused in four main areas.
The first is epidemiologic research. We have a history of impactful research and publications using vital statistics data that has advanced knowledge in the field of perinatal and material and infant health. In my own work, I have analyzed and published data on topics including multiple births, low birth weight, infant mortality and preterm birth. Monitoring, tracking and understanding trends in these outcomes is critically important to the work that we do.
After decades of increases in the preterm birth rate using vital statistics data, researchers, including some of us at March of Dimes, identified that these increases were happening primarily in the late preterm period, or just a few weeks early. These findings led to nationwide efforts to better understand these risks and also led to transformative initiatives to reduce non-medically indicated early deliveries and also educate providers and pregnant women of the risks of an early delivery.
Vital statistics data helped make this work possible, and they give us the tools to continue to monitor preterm birth rates as well as other outcomes including low birth weight, fetal deaths and infant deaths that may also be impacted by efforts to reduce preterm birth.
The second area where we use vital statistics data is in the production of nationally-recognized data products that are used by the organization and our partners to raise awareness, build partnerships and effect change in maternal and child health. One example is the March of Dimes Premature Birth Report Cards published annually in order to spur action by stakeholders on evidence-based interventions and advocacy priorities to reduce preterm birth and raise awareness of the seriousness of prematurity. The report cards are rich with natality data including state and county-specific preterm birth rates and indicators of racial and ethnic disparities.
In 2016, using 2015 natality data, five states had achieved the March of Dimes 2020 goal of 8.1 percent, receiving an A on their report card. Last year, these reports generated national and local news media coverage nationwide and more than 1600 media hits and nearly 60 million media impressions.
These report cards have proven to be powerful tools in activating our partners. In collaboration with the Association of State and Territorial Health Officials, state health officials were challenged to sign a pledge to reduce the preterm birth rate in their state by 8 percent, and by the following year, the health officials in all 50 states, the District of Columbia and Puerto Rico had signed the pledge to adopt this goal for their state or territorial health department. As of last year, 28 states and jurisdictions had achieved this goal, as reported through the vital statistics data system. These data have helped us to motivate, engage and hold states accountable and also to recognize them when these goals have been achieved.
The third area where we rely on vital statistics data is in our work to provide the public easy and free access to these data through our PeriStats website. With more than 40,000 users annually, we provide access to aggregated data from multiple federal and national agencies and organizations. Through a process involving NAPHSIS and NCHS, we obtain approval and are granted access to natality-linked birth infant death and fetal death data which include geographic identifiers for states, counties and large cities.
Using these files, we provide aggregated data by county which illustrate the wide geographic disparities in perinatal outcomes and risk factors for these outcomes. This granularity allows March of Dimes and other health professionals, researchers and students to better understand and describe disparities in poor birth outcomes and inform targeted program planning and implementation.
The last area where we utilize vital statistics data is in technical assistance for staff and volunteers who are on the front lines of our mission. Through the creation of easily digestible data-informed products like these fact sheets, we take that data and provide it in ways that empower them to use it to educate, advocate, raise public awareness and engage in conversations with key stakeholders working towards reducing prematurity, infant mortality and birth defects.
We have used data for decades to accomplish this work and more. Over the span of my nearly 20 years using these data I have seen many achievements and improvements in our vital statistics system including improved timeliness, the full implementation of the 2003 Birth Certificate Revision, and improved quality and accuracy of the data collected. Federal investments in state and national vital statistics systems have allowed for these improvements and have directly impacted our ability to use these data in meaningful ways to assess progress, examine trends and reset or refocus our efforts in timely ways. Having the confidence in these data and assurance that we are using the best available data to understand the issues is critical to our success.
As we look to the next generation of vital statistics, continued investment in these areas and others would enable us to do an even better job of providing these data to the public and empowering people to use the data for action to improve the health of every mother and baby, move us towards our goals of increased equity and give every baby the opportunity to live their best life.
Specifically, we envision a future that includes continued federal support for states to update their systems, a strong network of data users and increasing our ability to use data from vital statistics to address social determinants of health and improve health equity through engagement of those communities at highest risk of poor outcomes and empowering the use of these data to identify and address needs.
I will leave you today with this. Vital statistics data, in the very simplest of terms, help us to tell a story, like the story of my own twin sons born 10 weeks early, weighing 1800 grams apiece, delivered by a C-section, admitted to a NICU and requiring assisted ventilation immediately following delivery. All of these items are collected on the birth certificate.
On an individual level, these data represent real babies and real mothers and real families. In the collective, they tell the story of the healthy moms and babies in this country, the experiences of our youngest and most vulnerable populations. As such, these data are a powerful and essential resource that help us to understand the causes of prematurity and other poor birth outcomes, find solutions and target interventions appropriately.
Thank you for this opportunity to address the committee. I look forward to continuing to support the Vital Statistics System which provides such an incredibly important resource for us to improve the health of every mom and every baby.
DR. COHEN: Thank you very much. Thank you, panelists.
We are rolling on to the last group of panelists, commercial users. We thought it was really important to hear from lots of different perspectives, so I am really excited to be able to have you three provide us with your insights around your uses of vital statistics. We’ll start off with you, Susan.
MS. MCDONALD: Thank you for the opportunity. I really appreciate it. For the commercial users it’s a whole different ballgame than probably anything you have heard so far.
Who is PBI? I’m the President of Pension Benefit Information and we are a company that started over 35 years ago to answer a very simple question. How do you know when someone dies and you should stop making payments? This was back when ERISA was being formed, and all of a sudden there were all these fiduciary responsibilities for pension plans, and in a perfect world you are going to know when someone dies, but it’s unfortunately not the case.
We set out to introduce a solution to that market, and this was back in the day when the computer was the size of an office. We had to get tapes from SSA quarterly, and it was a real challenge.
But we went and delivered the technology to provide the solutions, and our client base quickly jumped from just pensions to insurance companies, financial institutions, reverse mortgages. There are all these industry segments that need to know about deaths, and some of them are mandated by state and regulatory requirements and yet they don’t always get access.
We are a professional corporation. We are dedicated to really delivering services that help our clients meet their fiduciary and regulatory requirements. We consider data security our highest priority. We are SOC-2 certified. When you work with the Hartfords and the Aetnas and the AIGs of the world, it all rolls downhill from the requirements of data security, and I will attest to LexisNexis and their requirements for secure data. It is a challenge.
One thing I would like to say is the implementation of EDR did amazing things for vital records. Back before the change in 2011 we were getting 70 percent of the deaths within two weeks. It was pretty amazing. Our company took a leadership role, and when the financial crisis hit we went to our clients and said why do this monthly or quarterly or once a year or semi-annually. We put them on continuous monitoring. We said, you’re not going to add that many new people; give us a new file once a month. But it allowed them to proactively stop an overpayment versus the risk of finding out 13 months later and good luck trying to collect it.
So, who needs this information? Pension plans. We deal with private plans, government plans, anywhere from federal, state, local, municipal. Third-party administrators. A lot of the large Fortune 500 outsourced benefit administrators, they are doing that job on behalf of their clients. Insurance companies. For them to remain viable, especially if you think about what we’re going through right now, they have to be able to stop annuity payments and not make them to dead people. Same with disability, long-term care.
Ever since 2010 with the John Hancock settlement, every life insurer is supposed to be using the Death Master File to proactively identify the deaths if the beneficiary has not come forward. This is about reuniting people with benefits that perhaps they weren’t even aware of. That has added a lot of challenge because, at the very same time that the insurance companies were mandated to do this, the DMF became crippled, which is a huge problem.
When it comes to reverse mortgage companies, HUD requires them to notify them within 60 days. Well, who is going to call up and say, hey, mom and dad are dead; come take the house? It’s just not going to happen.
And lottery winners. I don’t know why you would choose the monthly option, but if you did is your family going to call and say, hey, stop the payments?
We have done so many interesting projects over the years. The Masters Golf Tournament — we can’t do that anymore because of the Section 203. But those people never died. They had a waiting list forever to get Masters golf tickets —
(General discussion)
Another example is we helped the NBA. They wanted to provide benefits for players that played pre-1964. They never got any benefits. Using the Death Master File — they didn’t have socials, but we were able to come up with names and identify them and locate them and find beneficiaries. A lot of really interesting areas that we work in.
As I said, in 2011 everything changed. The process became extremely challenging. As one of the other presenters said, you can’t report what you don’t know. Well, you don’t know what you don’t know when it comes to making payments to someone. Most of the clients that we talk with assumed that they were getting all the deaths they needed, and now that it has been almost six years they’re finding out that they have been making payments for over six years to these individuals.
Let’s face it; we do have a pension funding crisis. The PBGC multi-employer plan is going under, they estimate, in six years. The pension de-risking that is going on — my company alone, and we are not a huge company, but we have reunited 700,000 people with their benefits so that they could either be put into an annuity or cashed out with a bulk lump sum. So, what is going to happen to the PBGC in another couple of years? They are not going to have the premiums coming in, and it’s going to be a disaster. And those plans are not able to identify the deaths and prevent the overpayments. It’s kind of crazy.
The DMF took a bad rap and it was assumed that that was what was causing the identity theft problems. I’m sure there were cases of that, but I think if you compare the unintended consequences to American citizens over the identity concerns, you get to see a little bit of a difference.
We have got a funding problem, and the elimination of the state vital records has exacerbated it. It is really a huge issue. Government pension plans — every one of you from state, guess what? They are not getting access to the full DMF. They are making payments to dead people, and we are identifying them. How are we doing it? We are using obituaries. Now, that is not the answer. I think we all know that. They don’t include SSNs. John Smith born on any given day. So it’s a huge challenge.
I would just like to say that I believe the Bipartisan Budget Act of 2013 put the protections in place, and with somebody needing to be certified to get access to the Death Master File, the controls were put in place; yet, all of those efforts that we are going through to get access to the limited access DMF does nothing to restore it. Thank you.
DR. COHEN: Thank you. Clearly, there needs to be a replacement for the DMF. For those unfamiliar with the issue, in 2011 there was a federal ruling that limited the Death Master File from Social Security Administration to include only names on fact of death that they got from their own resources and eliminated the use of data from states whose state law prohibited the use of their data in that way. So, the Death Master File after 2011 did not include information on death from states whose state law prohibited the Social Security Administration from re-releasing their information. Clearly, it is a hole in access for commercial use of death data that needs to be addressed.
Any other clarifications?
PARTICIPANT: What was the rationale for that change?
DR. COHEN: A state prohibits re-release of death data and the Social Security Administration was re-releasing death data from those states. If people wanted access to those data, they needed to go to the states to get that information.
MS. MCDONALD: I think there is one comment that the reason that SSA started putting the records into the public DMF was because of EDR and the fact that those records were verified through that process. They got rejected if they did not match.
As I said, we no longer have to research these records. Now, whether that was the right answer or not, I guess history will bear its way.
DR. COHEN: I think maybe part of our discussion tomorrow can be, moving forward, is there a way to fill the gap to provide fact-of-death information rapidly to a variety of users through a variety of different developments that are happening among jurisdictions and at the federal level, because clearly there is a need for access to fact-of-death and cause-of-death information that we have heard from a variety of users. So that will be an interesting point of discussion.
MS. WEBSTER: Just a follow-up. The contracts, as I said earlier, with the state forbid the reuse of the data. This is one of the things I talked about earlier today, was that it really hampers and handicaps the state to not have their data paid for in a way that’s going to support the entire system, as a whole.
SSA actually took the state data out of the Death Master File. They found the error and they took it out, and they worked with us on the aftermath. But, unfortunately, part of the Act still permits them to share the data for free, although I think there are some charges being exchanged with benefits-paying agencies. But it’s still government agencies; it is not our corporate partners.
That is one of the reasons we developed EVE, which we will probably get into more tomorrow. EVE is not a file. It’s sort of a different kind of system, so there are still some complaints about it and some economies of scale that are gained. The fact-of-death product of EVE was really meant to replace the Death Master File.
DR. COHEN: I apologize for taking the Chair’s prerogative but I wanted to clarify the issue around DMF. Edward?
MR. HALL: Good afternoon. Thank you for allowing us the opportunity to speak today.
I am here to speak about another beneficial use of the Death Master File. My name is Ed Hall. I am the Associate General Counsel of Verus Financial. We provide services to state agencies, and one of the services we provide is unclaimed property auditing services. I wasn’t 100 percent sure of how aware everyone in the room would be of unclaimed property’s concept, so I’m going to go through a little bit of that before getting into the impact of public health data on that particular process.
Just to give a little bit of background, every state has statutes and regulations which require the annual reporting of unclaimed property to the state. Generally, that’s defined as property that has been abandoned, and that is usually defined as an owner not indicating interest in a piece of property. An example would be if I leave a deposit with a company and I disappear five years later, that company should be reporting that deposit to the state. When it goes to the state, the state holds it basically indefinitely in custody for me, so they are actually performing outreach and they are attempting to locate the owners of this property as opposed to remaining with a company that would have a benefit to keep the property themselves.
This is extremely beneficial in the fact that, in 2015, state governments returned over $3 billion to rightful owners of property. Generally, these unclaimed property laws are there to protect the public to ensure that they receive money owed to them as opposed, as I said, to allowing it to reside with companies that may have it in their possession.
In order to determine whether or not there are companies out there that may be holding on to unclaimed property that may be due to be reported to the state, states will perform audits of those companies. In general, with those audits, over the past I think a lot of times there was a general ledger type auditing approach made in those audits, but as systems modernized and as companies are using a lot of data in the administration of their products, there now is also large use of data, and data analysis is being integrated into the performance of these unclaimed property audits. This has had a tremendous effect.
Starting in 2008, state treasurers and comptrollers began multistate audits of the life insurance industry. These audits — and this is kind of where the public health data aspect comes in — involved the comparison of insurer records against the NTIS Social Security Death Master File. These audits have been ongoing since that time and have involved the review and comparison of over 500 million records against the Death Master File and represents at this point in time, over 85 percent of the insurance industry. As Susan mentioned when she was speaking, there are a lot of jurisdictions that are now requiring that insurance companies perform these comparisons themselves.
Generally, the type of information that is being used here is really just the fact that a certain individual is deceased, their identifying information and the date of the death. That is being used to really determine whether or not there’s an insurance policy where the person who is deceased per the DMF is the insured, and that that insurance policy was still in force as of the date of death per the Death Master File. The result overall of these audits has been tremendous. To date, these audits have resulted in the reporting of nearly three million properties to states with a value of over $3 billion.
As part of the process, either during an audit or during annual reporting with unclaimed property, there is also a due diligence component. Before any piece of property is sent to a state, again, either on an annual basis or as part of an audit, companies are required to perform a mailing, and they may actually also do additional outreach, but they are required to do outreach to ensure that — you know, you may receive a letter in the mail and it will say we have a piece of property that we believe belongs to you. We are planning to report it to the state. Please indicate whether or not you would like that property to remain with us or you would like to receive that property.
So that process is ongoing during the course of these audits, and the estimation is that there is an additional $3 billion that has been returned directly to beneficiaries as a result of these audit processes, as a result of just the outreach and due diligence process that was performed.
Again, as I think I mentioned a little earlier, state treasurers and comptrollers are holding onto these properties in custody for the rightful owner indefinitely. Again, on an annual basis — there are state websites that you can go to look to see if you have unclaimed property available. Also, a lot of times they will advertise in newspapers, publish a list, they’ll go out to fairs and allow people the opportunity and try to reunite them with their property because that is overall their goal.
In general, the comparison of this information against the DMF, from our perspective, is a very necessary tool. What’s happening here, these are not particular pieces of property that would have eventually made it to the states or made it to the rightful owners but for these comparisons to the Death Master File. If an insurer has an insurance policy and no one informs the insurance company once the person that is insured is deceased they are going to stop paying the premium. So, if it’s a term policy it’s going to go away immediately. If it’s a whole life policy or policy with cash value in it, in that policy there are non-forfeiture provisions in place that will actually drain down the value of that policy over the years and then it will eventually extinguish in most cases. So, this is a lot of the property that ended up being returned to states and rightful owners that frankly would have just gone nowhere otherwise.
The other point I wanted to make about the usefulness of this is that the average value of the policies that we’re finding hovers around the $1,000 mark. If you have a life insurance policy and your benefit is $1,000, that money when it goes to your intended beneficiary ends up being extremely important and extremely useful.
There is applicability of this type of information outside even of that particular industry where it’s very pertinent. But you can imagine if unclaimed property is based on an owner not indicating interest in a piece of property, if you’re doing an audit, if a company is telling you, well, we received contact from this person a year ago and then you can demonstrate to them that they are on the Death Master File, they were deceased as of four years ago, obviously, at that point in time the company may be using something that was an owner-generated activity to actually keep that property still in motion at their company. So, again, there are a lot of uses for this information.
As Susan said, we have taken data security extremely seriously from the very beginning. The requirements, as you can imagine, we’re auditing industries and working on behalf of the states as a third-party auditor and we want to make sure they feel comfortable that their information is with us. We have gone through SOC audits, we’re ISO-27001 certified.
When the new regulations from the NTIS came along, it really just added another layer to that audit process. So we certainly understand the value in keeping that information safe and protected, but at the same time, now that the infrastructure is there to ensure that is happening, it’s obviously extremely important, based on the results of these audits, that that information is kept available.
That is all I have.
DR. COHEN: Thank you. Last but not least, Steve.
MR. EMMERT: Thank you. I am going to give more of a higher-level discussion as opposed to the sectors. For those of you who don’t know, LexisNexis is the sister company of Elsevier. We publish Lancet, and a few hundred other scientific and medical journals around the globe in addition to operating LexisNexis and another company in this space called VitalChek, but the comments today will focus on the use of vital records and the distribution of vital records.
For us in a commercial context, we will split the market into four sectors — insurance, business services, government and healthcare.
In a big picture sense, you look at the 50 largest U.S. banks and they are all customers of the Risk Solutions group. Property and casualty insurers, 100 percent are customers. Eighty percent of the Fortune 500 companies, all of the 10 largest pharmaceutical retailers, eight out of 10 of the largest life science companies, and 80 percent of all federal government agencies are customers of ours who have access to these services. In addition, we service state government agencies in all 50 states and 70 percent of local governments around the country. So we have a large customer base.
Notwithstanding the large base, there are limits on the right of access to the data that we are talking about, the right to use the data. If you don’t have a permissible purpose you do not have access to the data. Both of our industry representatives to my left happen to be among the customers that have access to our system.
I am going to flip over to VitalChek for just a moment. VitalChek is an interesting company that, in the marketplace, is in the position of facilitating access by consumers to vital records. It’s a different role than LexisNexis. You don’t have to be a subscriber. You can call up after a single record.
This service is actually made necessary by some of the more restrictive and sometimes, frankly, somewhat out-of-date state laws that are in place. I am aware of at least one jurisdiction where you must appear personally at the only office of the custodian of records in order to obtain a copy of a record. So, if you need your birth certificate and you were born in this state and happen to be living across the country, you either have to fly across the country to personally appear or, for a relatively modest fee, you can have VitalChek get the paperwork in order and make that personal appearance to pick up the copy of the record and then FedEx it to you directly.
My son wishes you could buy marriage licenses that way. He drove 450 miles over July 4th weekend to get a marriage license. Apparently, you couldn’t do that online and it wasn’t enough for him to just sign the papers and give them to his fiancé. VitalChek does provide some of those services.
There has been some discussion of why you need these records. As an individual, you will need them for things such as passports, sometimes for marriage licenses and enrollment in various benefit programs. Sometimes you have to have proof of the age of a beneficiary. For instance, you work in an office and you have three children of varying ages, and in order to enroll them on health insurance you actually have to produce certified copies of birth certificates to prove their eligibility both as children but also based on age. Again, if you have moved across the country that is not always the most convenient exercise, so VitalChek can help in that space as well.
To go back to more of the private sector for a minute and talk a little bit about those records, one of the things I want to talk about a little bit is some of the needs around updating some of the laws here. For a company like Lexis, one of the things we would try to do is obtain an entire database so that we can provide both combined data — birth record data — with information on addresses, on places of employment, on property and other assets and movement and the like. Those are used widely in the legal services industry.
They are also used by insurance companies, life insurance companies, property and casualty companies and the like for the administration of programs and administration of benefits and administration of policies. Financial services companies, of course, are using it as a way of authenticating identities to try to help prevent identity theft or fraud. Government agencies similarly use them for those purposes. Tax fraud prevention is one of those and benefit fraud prevention is another. So they are used for a host of reasons.
But, if you look at the underlying structure behind the vital records laws in the states, a lot of those laws were passed about 25 years ago. Many of those laws were written 50 years ago, 75 years ago, before the technology existed to move data as quickly as we can over the Internet.
There needs to be a balanced reach between facilitating access and use of vital records and protection of vital records. We don’t want access to vital records to create a huge security risk for individuals. That is something we need to be very careful about and we need to guard against.
On the other hand, some of the inconveniences that were built into the system originally such as we only actually had one office that had all the hard copies of the records and, therefore, you had to go to that office. That is an artificial constraint that doesn’t really exist today once you have entered the data into an electronic system of records, and you can make that system available electronically.
So, going forward, and one of the things I have been working on — we have worked on it with regard to the Death Master File, for instance — Going forward, states need to evaluate carefully how they can facilitate access to this data for responsible, publicly beneficial uses in a manner that brings it up to the 21st Century.
I will mention that we have been doing a lot of work with NAPHSIS in this regard and they have been a great partner in this regard.
One last caveat is none of the comments I have made are meant to criticize either state custodians or, for instance, the Social Security Administration or NIST. They have all tried very hard within the parameters of the laws they have to be responsible partners. But it’s an area where some updating on the laws would facilitate some actions that would have a dramatic public benefit if we were able to get there. Thank you.
DR. COHEN: Thank you very much. I will open this session up for committee questions and comments.
I can start. I wanted to ask Rebecca and Michael with respect to using birth data, do you find your users focus more on gaining access from federal data sources or the jurisdictions, or both?
DR. WARREN: Maternal and child health programs are typically situated within state health departments, so most of them access that data through their own state health departments and through relationships. There are still siloes even within departments and across programs and typically they will do that. They will rely on the federal sources sometimes to verify performance across states. Folks are often interested in how do we compare, how do we rank, so then they will tap into the federal sources. But typically, for our members, it is from their own states.
MS. RUSSELL: In the Perinatal Data Center we get the national files from the National Center for Health Statistics, but we do have staff in every state. Often, there are limitations around the data that we have and so we will encourage our state staff to work closely with their state vital records offices and their state departments of health to get data that maybe we don’t have access to. So I think it works both ways.
DR. ROSS: This has been very interesting. I am interested about the notion of a legal refresh, if you will, or updating. Is there a proposal that has been put forward? What does that involve? Is that a state-by-state issue? Is that a federal thing?
MR. EMMERT: It is actually a very complex scenario. Vital records themselves originate typically at the state level. In a few jurisdictions, they are actually even more local than even the state level. They may be aggregated at the state level and then go out to the counties.
Generally speaking, for the right of access to the data, you want to go back to the states themselves because the data is actually theirs. One of the advantages, if you will, is that in theory you should have more complete and more accurate data. The down side of that is that we have 57, 58 jurisdictions that we have to deal with, so you have everything from collection challenges and formatting and record layout challenges as you bring the data together.
One of the things that is sort of a hallmark of the service that LexisNexis offers is the ability to do, for instance, a national search. We bring the data from multiple jurisdictions and then you run a single search against all of those jurisdictions. It brings a great deal of value. No one state can offer that service.
The federal government, a lot of the data they have is secondary. When the resolutions team was putting together databases of personal data, very little of that actually, comes from the federal government because very little of it originates at the federal government. The Death Master File was one of the — we call it an exception only because, having aggregated the data under the federal FOIA, they were forced to disclose it more broadly publicly, so they inadvertently became a disseminator of state data, data that they aggregated to perform the offices of the Social Security Administration, but having done that, were forced to then publicly disclose.
Not being one to complain — that thing about not looking a gift horse in the mouth — 25, 26 years ago, I had the opportunity to acquire access to that data on behalf of the LexisNexis company, and that was when we first started using that data file. But it had its limits then and it certainly has even more limits today. So the states still are the preferred source, but, to actually free up access you really have to go state by state, legislature by legislature and persuade them as to the wisdom of broader utilitarian uses and broader access while imposing reasonable safety measures, and that is a difficult process.
DR. ROSS: Understandable. But are there any proposed model laws and, also, in effect, advocacy packages that persons within the state, state government or others, could use to propose that discussion to the legislatures, or is this just a hopeless cause?
MR. EMMERT: I don’t think it’s a hopeless cause but I will say that, to my knowledge right now, I don’t know that there is a model law, but that doesn’t mean that one couldn’t be prepared. It’s just that I am not aware of there being one today.
DR. COHEN: There is actually, a model law. There is a model vital statistics law that has been generated by a conversation amongst the jurisdictions in NCHS.
DR. ROTHWELL: In fact, the oldest model law in the United States is the model law for Vital Statistics. It was last tried to be updated during – just what I was talking about, the intelligence reform time, in order to improve security. So there is a process in place.
The model law, from an NCHS perspective, we staff it usually when it comes about, our interests are mostly from a physical perspective, but the model law covers all of vital registrations. So the states really have the responsibility to come up with what they feel a model law is. When we send out the model law it is just that, the states can pick and choose from that model law what they want.
So, if in fact we were to create a model law that had something like you were talking about, and that the states would agree with that — a lot of ifs there — then the individual states would still be making up their minds on what they would modify. And there are some states who have not changed, I think, their model law I think for several revisions.
So, what you were saying is correct in a way. Yes, there is a model law. States can participate in creating that model law but they are not then required to use it. But it has been probably the most important thing from NCHS’s perspective, to get data standards in place, so we feel very strongly about it.
But again, this is a confederated activity.
DR. COHEN: Is there access to the current draft of the model law?
DR. ROTHWELL: We gave it to NAPHSIS and NAPHSIS has it.
MS. WEBSTER: It is on our website right now. I did also want to mention that federal-level support for a model law would go a long way to help our states actually get it moving through their own legislatures.
DR. COHEN: That is the core question, how do we get this moving.
MS. WEBSTER: Exactly. Right. And one of the things that really hobbles us in our products like EVE for the fact of death is that some of the states can actually share data with non-government entities. They can share information with government entities all day long; that is not our problem. But for our corporate partners, who we want to be able to do business with and have them support the systems and have us support them, there are lots of states who are interpreting their statutes or their statutes to specifically say that they cannot share data with non-government entities. That is hurting us right now.
DR. MAYS: Thank you for the presentations. I guess this question is specifically because of the issues you were raising. What would be the business case in terms of talking to states to do this? Most of the states, it’s not that they’re not doing it because they don’t believe in it. They are not doing it because they can’t get the resources or it’s not something politically that they think they can get through.
So there has to be a very good business case and I don’t know if there is one that can cut across these very different states and territories.
MR. EMMERT: That is an interesting question, but, to be honest, there are some fairly easy business cases. Let’s take it to the state. In the state of Indiana, three years ago, we started partnering with them to simply try to help them authenticate and verify the identities of state tax return filers. The first year we blocked $88 million in fraud. All we are doing is authenticating the identities of the filers who submitted for a tax refund.
Now the good news is, once you do that the first year, the second year a lot of those fraudsters go away so you only block maybe $20 million, and the third year maybe it’s only $10 million. But the cost of doing that is relatively low.
One of the things you have to do right upfront is to make sure that deceased people don’t file tax returns. It is shocking how many dead people do that. On a financial services basis, a lot of dead people also get car loans, mortgages and credit cards. Very busy when they’re deceased.
Having accurate data about the date of death is a big deal for that very reason. Even birth data is helpful because it helps you to establish age. Somebody comes in and says I’m Susan Smith and here is my social security number and they don’t know anything else, and then you find out that Susan Smith is actually only three but she’s getting a new Visa card. It’s a problem.
Having reliable birth data, reliable death data — I’m not talking about posting it on the Web for the whole universe to have, but, with specific use cases in mind, you can use it to prevent fraud in government programs, you can use it to prevent fraud in financial services programs.
The student loan program — just wait. Give it another year or two years and we will suddenly hear shockingly that there are billions of dollars of fraudulently issued federally-guaranteed student loans because the fraudsters started applying for those four or five years ago and no payments are due yet. So, kids don’t know that they already have $40,000 or $50,000 in student loans that they didn’t apply for. Someday soon, they are going to start getting collection notices and that’s when you are going to hear about it. So give it another year or two years and you’re going to start hearing about that problem.
The use case in the fraud prevention space alone is enormous, and it benefits not just the agencies but think of it as benefiting the citizens, because, yeah, Visa may have issued the card to the wrong person but it’s somebody’s parents or somebody’s little children who are having to deal with the headaches around proving that their three-year old didn’t actually get a Visa card this summer. It’s a pretty easy case.
MS. KLOSS: Dr. Naghavi, do you have any insight into how other countries are doing a better job avoiding the garbage codes?
DR. NAGHAVI: In 2015 the best countries were New Zealand and Finland, but Australia is usually better than the U.S. New Zealand, Finland, Sweden, Norway are better. Some states in Mexico are better than the U.S. Some states in Brazil are better than the U.S. Some states in U.S. like California are very good but some states are not.
This is a big problem that I think we have to fix. This is a very important problem. How is it possible that, for example, 2,000 per year die due to low back pain, or 100 deaths due to infertility, or lots of deaths due to sepsis? Sepsis is not an underlying cause. Some part of the problem is software that now we are using.
From two or three years ago, Jordan — unbelievable — is better than the U.S. This is the situation that we have.
DR. COHEN: Not all the committee members were in our discussions where we decided to really focus this hearing on structural and broader issues. We realize that quality and timeliness and content are also extremely important, and there is no bright line between one and the other, but we hope to possibly explore more the issues about content of certificates and quality as we move forward.
MS. KLOSS: But I would argue that it is the governance and organizational structure that promotes and enables the quality.
DR. COHEN: You are absolutely right.
DR. NAGHAVI: Another problem is when we look to the source of the death certificate, the worst data comes from hospitals, not from the coroner. This is the problem. In the system in the hospitals, there is the problem that the medical doctor hands the certificate to the nurse, the nurse passes it to some other person and some other person fills it out. This is a problem in the system.
DR. COHEN: Any other comments from committee members? Any committee members on the phone?
I will open it up to comments and discussion from audience members.
DR. SAADI: This is Lou again. Just a clarification about EVE fact of death. We do have our model law and many of us wish to have that new capacity because, as Mr. Emmert spoke very eloquently and kindly about, we are dealing with some very antiquated language from the 1950s.
But we have difficulty getting those out of our public health agencies. Something called same-sex marriage has prevented me from being able to do that the last three years, for example. And a lot of other states have that difficulty because the political issues overwhelm us in being able to get our legislation changed. It is just a fact of life for us. Dr. Warren, I am sure you probably have heard some of this as well from Jay.
Regarding the model law, we did craft — our EVE committee put together some special language for jurisdictions to use that specifically says you can utilize vital records data to share with non-government entities. So, our members have that information at their disposal. It is just how do you get out of your agency, out to the legislature to promote e-commerce, to promote sharing.
Frankly, we think our corporate partners and the private sector can help us shore up this infrastructure and be able to support many of the public health needs. Now, that doesn’t mean public health and other research entities should be able to support the uses of the data, but we do think there are some infrastructure needs that can be met by some of the e-commerce that we think can happen with this information.
DR. COHEN: We received one question for Susan and Edward online, and the question is are your organizations willing to use EVE fact-of-death for your searches instead of the DMF?
MS. MCDONALD: I have requested access and have been turned down as a service provider.
DR. COHEN: So you requested access to EVE fact-of-death?
MS. MCDONALD: Because you have to say the vertical market, so my approach was okay, then, these are all my insurance claims, and that wasn’t good enough at least from the last I heard.
Keep in mind there are a lot of records being processed and no one understood this when NTIS had those initial hearings. We process over 160 million records a month. It’s huge.
DR. COHEN: Edward?
MR. HALL: That is certainly something we are currently looking into. I am not the specific person to ask that in the company, but I do know that’s something we are in the process of working towards right now.
Obviously, any available information that we can have that would assist in our audit process is something that we are looking into.
MS. WEBSTER: I can speak to that a little bit. This goes back to our jurisdictions needing to verify the uses of the data, so we have this very complicated and restrictive credentialing process for our users, although we have worked with Susan and PBI and Verus and many others. We also have a strict part of our contract with our vendor for EVE — which is LexisNexis VitalChek — that makes them the only party that can sell to re-sellers, and that is very restricted as well.
So the issue is us being able to determine the end use several layers down and verifying that that is a legal and administrative use of the data and it is not being re-shared. Again, it’s just very, very complicated.
DR. COHEN: Is it something you think you can work out? It would be easy in Massachusetts; death records are open.
DR. SAADI: Okay. They are not in Kansas; closed record. I have to have information about the policyholder, that you actually have a policy with the company that you are requesting it from. Again, this is from the 1950s, that we have to verify exactly that you are getting this information on this individual because you actually do have a policy with that individual. This is something we have to reconcile because there are so many third-party administrators.
Our jurisdictions have to be comfortable with this, and I think there are ways to do it. We have to get to the critical mass of all jurisdictions participating. We have a chicken and egg situation right now. For others to be able to use the system they want a full product, and then we have to convince our jurisdictions that there are ways for us to get around individuals having access to the system and companies having access to the system and that they can show they have a valid need and they are not just fishing for the information.
I hope that makes sense.
DR. COHEN: Thanks. Are there any other comments or questions? Who is on the phone? Speak up, please.
DR. SCOTT: Hi Bruce. This is Linette Scott. I just wanted to chime in. I was a participant on the model law committee when we put that last version together, and one of the things I don’t think I have heard today but was a key component as we were thinking that through is, again, what is essentially the cost reimbursement for doing these certifications or verifications electronically.
So, just echoing back to a comment I made earlier, the vital record is often funded based on the sale of certified paper copies and a minimal amount on data files and data in and of itself.
One thing that might help the conversation, especially as we hear this conversation, is thinking about how to include the cost reimbursement piece to this question, because most of our Vital Records budget — even in California, the overall budget for all of the registration, data protection, production and all of that is less than $30 million. It is not all that much in the scheme of things, but when you’re sitting in the vital records program it’s a lot, and it is very hard to get those additional resources.
One of the things to offer is, again, make sure that as this conversation happens about receiving data and the need to receive data, if you include how the jurisdictions will be reimbursed in order to have the infrastructure it may help the conversation, especially from a kind of political willingness perspective. That would be one comment.
My second comment is in terms of again thinking about these uses. It might make sense to take a look at HIPAA. Under HIPAA, obviously, it protects health information, but there are specific uses around healthcare operations that are identified that allow for the data-sharing among various entities that are involved in the processing of claims and delivery of healthcare. Maybe that would be something to take a look at in the context of vital records around the ecosystem of folks that need to interact with this data, so, in terms of thinking about allowable uses.
Much of the law has been written based on that paper-based environment of somebody coming to the office and requesting a piece of paper, so, when we did the last model law revision we really tried to think about the electronic environment and how to incorporate electronic capabilities into the model law. But even so, this has now been a few years back. Thank you for the chance to comment.
DR. COHEN: Great comments. I want to add one thing. You are fortunate you’re in a state where the vital statistics enterprise gets to keep the revenues generated from the sale of certificates. Many states require that that money goes to the general fund and the vitals enterprise never sees any of that, or at least certainly not full value.
Are there any other comments on the phone?
(No response)
Thank you all for a wonderful day. I hope you can re-join us tomorrow when we focus really on solutions. We are adjourned.
(Whereupon the meeting was adjourned at 5:12 p.m.)