The Subcommittee on Populations of the National Committee on Vital and Health Statistics held a meeting on September 15, 1998, in the Hubert H. Humphrey Building, Washington, D.C. The meeting was open to the public. Present:
The Subcommittee on Populations held a breakout meeting on September 15, 1998 to review several Subcommittee projects.
Ms. Rosenbaum described the current challenge to fashion a new legal framework for managed care, and outlined the scope of the project.
Ms. Markus described the purpose, methodology, and major findings of a series of recent interviews on data collection with state officials. The objectives of the study were to describe: (1) how state Medicaid agencies decide which data they want collected, (2) types of data that managed care organizations (MCOs) submit to the state as a result of these choices, and (3) states' capacity to analyze the data to inform the purchasing process.
It was observed that states focus less on minority populations and multiculturalism than on the financial and operational side of their business ventures.
Ms. Rosenbaum introduced the process by which specification language was to be drafted in the areas of data collection and data reporting for public health surveillance and community- wide disease tracking activities. The specifications will address data related to health care access and service utilization, measurement of health care outcomes or inputs, and health care quality performance. Subcommittee members discussed the composition of the drafting work group.
The members compiled a "to do list" for the Subcommittee, including:
Members discussed the composition and timing of the Medicaid managed care report to be compiled.
Review of the draft was deferred.
Drs. Newacheck and Starfield presented a draft letter on the State Children's Health Insurance Program. Members agreed to circulate a revised letter to members of the full Committee and to include it as an action item in the Subcommittee report.
In-depth discussion was tabled until the next meeting of the Subcommittee.
Plans were made to discuss, on September 17, 1998, integration of the Subcommittee's various mandates, staffing, meeting schedules, and balancing the work load. (An off-mike decision on the meeting date in October does not appear in the transcript.)
Dr. Iezzoni convened the breakout meeting of the Subcommittee on Populations at 1:15 p.m., September 15, 1998, and participants introduced themselves.
Sara Rosenbaum, J.D., and Anne Markus, Center for Health Policy Research, George Washington University, led the discussion.
Ms. Rosenbaum outlined the background and content of the Medicaid managed care project conducted by the Center for Health Policy Research. The movement toward managed care began 25 years ago with the establishment of ERISA and the application of the Sherman Anti-trust Act to doctors. These events deregulated most of health care, in that health systems were able to grow outside the established regulatory framework. The current struggle is to fashion a new legal framework for managed care; the framework that now governs the system consists of the contracts between buyers and sellers of managed health care.
Ms. Rosenbaum noted that the Center's initial managed care research focused on large purchasing agreements. Many findings can be generalized beyond Medicaid to other sectors. The Centers for Disease Control and Prevention (CDC), Substance Abuse and Mental Health Services Administration, Health Resources and Services Administration, and a number of private entities have funded this work, whose initial results illustrate the structure, content, and operation of the contracts for all states. Based on this body of knowledge, the researchers developed a series of sample purchasing specificationsÄexamples of "well-drafted legal language for a realistic range of responses to a particular issue"Äin 18 separate areas of interest.
Ms. Rosenbaum described the specification development process including: multiple reviews of the drafts by diverse subject matter experts. It is hoped that good policy decisions will be enhanced by the availability of good legal language for a variety of possible policies on an issue.
She then described certain cross-cutting issues for which language is being drafted. One issue is memoranda of agreement, or memoranda of understanding. These are agreements between managed care companies and other stakeholder agencies, such as public health units, that are not the purchasers of services. Another issue is data availability; the Center has prepared a background analysis of the data collection requirements in the law, an analysis has been conducted of data collection in selected states, and the Center plans to solicit comment from Committee members.
Ms. Markus described the purpose, methodology, and major findings of a series of recent interviews on data collection with state officials. The objectives of the study were to describe: (1) how state Medicaid agencies decide which data they want collected, (2) types of data that managed care organizations (MCOs) submit to the state as a result of these choices, and (3) states' capacity to analyze the data to inform the purchasing process.
The focus of the study, Ms. Markus explained, was on MCOs that provide a comprehensive range of services. With a total of 11 states, the study covered 44 percent of people enrolled in full-risk Medicaid managed care; in terms of contract language used and the extent to which the provisions related to data in the contract, the states were representative of the nation.
Ms. Markus summarized the first set of findings. Seven states characterized their data decision-making process as formal. For half of those, data collection is just part of the general purchasing process. Decision makers included the Medicaid director, managed care program director in consultation with Medicaid director, or other program directors in the agency; most states relied on teams to make decisions. Larger groups of decision makers were involved in informal decision-making processes. Most states involve MCOs in data decision making; states with competitive procurements do not.
Ms. Markus also found that states that involved MCOs in data discussions characterized the MCO responses in a range: receptive, collaborative, basically good relationships to a mixed experience, to an initially negative experience. Impediments included: limited power to impose conditions and problems in negotiating encounter data. Post-contract discussions with MCOs were cited as a key factor for successful implementation of a managed care program. Ms. Markus noted that states that involved MCOs in discussions tended to involve other participants in the process. Other stakeholders included state legislators, consumers, providers, fiscal contractors, and other departments or agenciesÄbut no other health agencies.
Ms. Rosenbaum noted policy implications for the fact that, under the law, an MCO in a Medicaid program is considered an agent of the state or of HCFA, and not a provider. While most states may see MCOs as providers, the courts see it otherwise; a hybrid relationship must be developed.
Subcommittee members discussed a focus on minority populations and multiculturalism. The survey found little concern collecting information on minority populations and multiculturalism. Most states were more concerned with the financial and operational side of their enormous business ventures. For MCOs, ethnic and race information is not identified as a business need.
Ms. Markus itemized states' reasons for collecting data: compliance with federal and state requirements (or contract specifications) and measurement of quality. Half the states mentioned performance measurement, and one mentioned national public health goals. Ms. Rosenbaum noted that states are not named in this study. Selection criteria included federal requirements, HEDIS, and the national standard format, HCFA 1500 and UB92; other criteria were added financial burden and consistency of standardization.
Ms. Markus turned the discussion to encounter, access, and quality data. Contracts in all states in the study required submission of encounter data. Ms. Rosenbaum noted that some states can link Medicaid data to enrollment files, to provider files, to fee-for service encounter data, but some have no patient identifier to permit that. Most states require MCOs to submit parallel enrollment and disenrollment data. Few if any personal characteristics are collected. Half the states collect complete encounter data; two said they lacked pharmacy data. Many states that currently lack pharmacy data will be putting that provision in their contracts in the future. The contracts list the data elements. Most states receive the information electronically from the plan to the state.
Participants discussed the need for additional detail concerning the respondent states' encounter data format and enrollment data elements. Also discussed was the need for information in the future on the role of other standards, such as electronic standards for health care transactions. Ms. Markus noted that half the states use encounter data to measure quality, a third use it to set and adjust rates, and the remainder measure plan compliance with performance and compare utilization rates within the plan. It was not determined how the data influence policy decisions, according to Ms. Markus, and Ms. Rosenbaum added that the data, at this point in time, are not sufficiently accurate or complete for them to be used in a legal sense. Data are being used to connect certain use patterns with certain broad populations. Drs. Newacheck and Takeuchi suggested the value in obtaining specific examples of where data have been useful to states in any area, not just capitation. Another participant suggested examining data vis-a-vis use patterns.
Dr. Iezzoni noted that some states have no data analysts, and Ms. Markus observed that most states did not cite the need for technical assistance. Concerns about the data included quality, timeliness, and accuracy of the data; incremental implementation of the data collection process; insufficient resources; high turnover rate of programmers; and consensus of the definition of an encounter. Participants discussed the advisability of encouraging states to improve their capacity to make policy decisions on what data to collect.
With regard to capacity to understand the quality of data, Ms. Markus stated that a couple of states reported using a system to validate the data. Ms. Rosenbaum pointed out that sanctions for data submission violations are part of the contracts, but no sanctions have yet been invoked.
Participants discussed the minimum availability of integrated data. In response to a question from Ms. Coltin, Ms. Rosenbaum noted that variation in enforcement of data submission and other reporting requirements would be in the practice; the contract would be standard. But it is recognized that adhering to some requirements might push smaller companies out of business. Nevertheless, variable enforcement has ramifications public managed care. Ms. Markus noted that a couple of states mentioned QSMIC as a future requirement.
Ms. Markus then discussed access data. Federal requirements of MCOs include medical audits, collection of management data, provision of data to the state in compliance with a Medicare Plus Choice or other specified data set. Medical audits refer to evaluating the operation of the plan. In response to a question on HEDIS, Ms. Rosenbaum replied that almost every state is asking for a consumer satisfaction measure, and parts of HEDIS may be selected.
Ms. Markus completed her presentation of the state study with a discussion of public health database. Few states require MCOs to report direct information to public health agencies. Few states contractually require reporting to a public health authority. When states trust the public health agency, it was more possible for them to integrate data.
The discussion turned to the activity on drafting contract language. Ms. Rosenbaum explained to the Subcommittee members the document they were to review in order to consider appropriate draft specification language for data collection and data reporting for public health surveillance and community-wide disease tracking activities. The specifications address data related to health care access and service utilization, measurement of health care outcomes or inputs, and health care quality performance. Ms. Rosenbaum explained that members could make alterations to the basic draft language in the following ways: address the issue in clearer language, enumerate more than one way of conducting a procedure, address the issue in more elaborate language, add contemporaneous notes or commentary, and add definitions.
Once the "assignment" was set out, members discussed how to implement the review. Ms. Rosenbaum described the generally iterative, small-group nature of the process. She suggested that two or three Subcommittee members or designated people from the Department to speak for the Subcommittee should work on the document. Subcommittee members discussed the composition of the working group. Reviewers would be cognizant that they are drafting legal documents, not practice guidelines, that require minimum legal statements that confer the right of a buyer to get certain data in a useful format. Three or four iterative review cycles would be required before the document would be ready for review by the full Subcommittee.
The members compiled a "to do list" for the Subcommittee, including:
The members discussed the composition and timing of the Medicaid managed care report to be compiled. Dr. Iezzoni stated that the George Washington University report would be incorporated as an appendix (or appendixes), and that the report will be 20 single-spaced pages with a 5-page executive summary. Ms. Coltin suggested the gap analysis model for the report.
Mary Moien led the discussion. She presented the history of the report, including her production of a first draft in August that focused, per guidance from the March Washington meeting and the Boston meeting, on who needed data and what data was being used for, but not on specific data items. The document presented to the Subcommittee for the present meeting includes several draft recommendations.
Dr. Iezzoni suggested that a review of the draft be deferred until after off-line discussions are held concerning staffing, timing, and preparation of the final report, decisions to be made within the next few days. Those members who reviewed the draft were asked to submit notes to Ms. Moien and the other members.
Drs. Newacheck and Starfield presented a draft letter that suggests that a system be implemented to collect information on the State Children's Health Insurance Program that would be useful at both the federal and state level, for accountability, program monitoring, and evaluation purposes. The specific approach suggested is the State and Local Area Integrated Telephone Survey (SLAITS) developed by the National Center for Health Statistics. A plan was developed to circulate the letter to members of the full Committee and to include it as an action item in the Subcommittee report to be given the following day. Members suggested modifications to the letter, including the highlighting of a specific recommendation and the enumeration of implementation issues still to be addressed.
Ms. Coltin led the discussion. Ms. Greenberg stated that a work group needs a charge, which could be identical to a piece in the overall charge. Ms. Coltin stated that she had identified two goals and objectives in an analysis of the recommendations made by the President's Advisory Commission. One task would be to take each of the recommendations and identify the underlying data issues, and then to assess whether current data systems are up to the challenge and to determine what action would be necessary to realize the objectives. Another task would be to use NCQA's informational systems road map model in terms of the private sector to evaluate whether its provisions are adequate, and to modify them where necessary. Ms. Coltin clarified that the road map model addresses standardization of data collection.
Members discussed the advisability of the next major thrust undertaken by the Subcommittee being "just" quality, or quality related to the population focus, rather than managed care. The discussion was tabled until the next meeting of the group on September 17, at which time a work plan was to be fashioned. Dr. Newacheck cautioned against taking on an agenda too broad to do well. Ms. Coltin suggested the attractiveness of a transition from the work on Medicaid managed care to quality issues related to vulnerable populations, augmented by linkages with groups working on related issues.
Dr. Iezzoni suggested that on September 17 the Subcommittee discuss integration of its various mandates, staffing, and meeting schedules. Ms. Coltin added the subject of balancing the work load. One member suggested that the operation of the Work Group on Quality may be an evolutionary process.
Dr. Iezzoni adjourned the meeting at 5:15 p.m.
I hereby certify that, to the best of my knowledge, the foregoing summary of minutes is accurate and complete.
/s Lisa I. Iezzoni - January 21, 1999
Lisa I. Iezzoni, Chair - Date