Topics for Research about Personal Health Records (PHRs)

This short review is intended to promote discussion about PHRs among the members of the National Health Information Infrastructure (NHII) Workgroup of the National Committee on Vital and Health Statistics (NCVHS).   This preliminary document outlines broad categories of issues and lists potential questions of interest as a first step to developing a research agenda on personal health records.  It does not imply who should be responsible for the research agenda.  The next steps for the Workgroup are to confirm the scope and audience(s) and suggest the level of granularity that is desired in the next draft. 

Contributors to this review include: R Agarwal, University of Maryland; G Christopherson, Centers for Medicare and Medicaid Services (CMS); MJ Deering, National Cancer Institute; RT Kambic, CMS; H Lehmann, Johns Hopkins University; E Ortiz, Veterans Administration.

The first research step is a review of the field and needs assessment. We ask, “Who wants a PHR and what will they do with it?” These simple questions become analyses of the functional specifications and the business case of a PHR. 

Additionally, terminology clarification is needed. There are currently multiple definitions of a PHR.  (A separate project intended to create a “taxonomy” of PHRs is underway.) In the short history of the PHR, discussion about it has moved from a focus on personal health data to a system of applications that might include communication, searching, decision support, and other sophisticated techniques. To clarify these differences, we suggest that a PHR refer only to the application that manages personal data and, following HL7 use of “Electronic Health Record System” (EHR-S), the term Personal Health Record System (PHR-S) be used to include the universe of potential functions that might be associated with a PHR.

PHR BUSINESS CASE FOR PATIENT, PROVIDER, PAYER, PURCHASER, AND INTERMEDIARIES

Because a PHR/PHR-S involves any and potentially all information related to an individual, the participants in a PHR/PHR-S will include not only the patient, but, in some circumstances, also the provider, the payer, and any intermediaries among these entities.  In those instances where the PHR/PHR-S is totally patient-controlled, the patient will determine which other users may access the PHR/PHR-S. 

The next step is to consider what the business case is for each user or entity. What is the business case for its involvement in the PHR/PHR-S?

PATIENTS

• What do patients do with their own health information? How do people use their health information?  We would find an impressive variety of healthy, ill, chronic, and acute situations where individuals use their health information.
  • What do patients want from PHR/PHR-Ss and what are they willing to pay for?
• Are we able to determine differences and similarities in the use of health information across these varied groups? What is the level and spread of health information desired across the disparate groups? What information do lay persons want? How much do they need? What are the temporal components of this information need, i.e. when do they need it? How often?
• What are key, or representative, use cases for patients and how their workflow around a particular health issue involves information? Examine the information the patient uses, ask why it is important and how the workflow might be improved.  For example: the patient (or caregiver) needs to monitor and maintain specific health status measures for a chronic disease; the patient needs to keep track of a series of appointments and payments for therapy; the patient gathers information from the Web and other sources about childhood asthma; the patient keeps track of payments from private insurers and Medicare; the patient keeps her entire health record on her home computer.

PROVIDER

• Under what circumstances are doctors, pharmacists, therapists, nurses, and others who provide care interested in sharing their patient records with the patient?
  • What is the provider experience of sharing such data?
  • How much information do they currently give and how much does the patient take away?  
  • How much concern does the provider have over proper use and misuse of the data?
   
• Under what circumstances do providers need or value patient-initiated information?
  • What are the issues when providers use patient-entered data?
  • What is the provider experience in receiving such data? 
  • How are concerns about the integrity of the patient-initiated data addressed?
  • What does progressive disclosure of PHI by the patient to clinician actually mean clinically, and what impact may be expected on provider clinical decision making?
• What are potential untoward effects of progressive disclosure? How does medical information have to be translated and repurposed for the patient?
  • What kind of consumer health vocabularies do we need?
• What are key or representative use cases for providers and how their workflow involves providing information to the patient or receiving information from the patient? For example: ensuring medication compliance; following directions to lose weight, stop smoking, exercise, etc., consulting with the patient by phone, letter, email, video, etc., providing referrals to another provider; receiving health status reports from patients with heart disease or diabetes etc. .

PAYER

• Why are insurance companies and CMS interested in sharing their patient records with the patient?
• What is the payer experience of sharing such data? Payers have experience with sending records of payments to the patient.
• Why might payers be interested in having patient-initiated data as part of their records?

PURCHASER

• Why are employers who purchase health insurance interested in their employees having access to their personal health information?
• What kind of information do purchasers wish their employees to have? What is the employer expectation of employees’ use of information?
• Why might purchasers be interested in having patient-initiated data as part of their records?

INTERMEDIARIES

• Intermediaries such as clearinghouses under HIPAA, Third Party Benefit Administrators, and other businesses that use or pass personal health information may become involved in distributing personal health information to patients. What are the business cases for these agents?
• Under Health Savings Accounts and stand-alone PHR/PHR-S systems, banks and other third parties have access to and manage the flow of personal health information.  What are the business cases for these agents to adopt or accept PHR/PHR-Ss?

AGGREGATE QUESTIONS

• The preeminent question is one of individual record privacy. This might be broken into ownership and control of the individual health data.
  • We need law and regulations which will then be tested in the courts.
  • Who owns what in the EHR and PHR/PHR-S space? How is ownership related to the control of access to data within a system?
• What is the scope of logging each and every time that the PHR/PHR-S is touched or accessed by any of the above stakeholders? In order to be legally protected will a provider have to maintain a time stamped exact copy of each off-load of health information to the patient?
• Who pays for the PHR/PHR-S?

DESIGN ISSUES WHICH FOLLOW FROM THE FUNCTIONAL NEEDS OF THE ABOVE STAKEHOLDERS

WHAT IS THE BASIC STRUCTURE OF THE PHR/PHR-S AND HOW IS IT DEFINED?

• Is it digital or paper; where is it stored; how are ownership, control, and updating implemented; how do we define the differences between siloed PHR/PHR-Ss for specific conditions (chronic illness, immunization records), and more general PHR/PHR-Ss? Is tethered and non-tethered a useful distinction?

WHAT ARE THE USER INTERFACE ISSUES AND OTHER USABILITY ISSUES FOR EACH OF THE POTENTIAL USERS?  ARE SPECIFIC TECHNOLOGY PLATFORMS (E.G. pc, Web, PDA, cell phone etc.) BETTER SUITED TO DIFFERENT USERS?

• How do we best present data for patient use?

HOW CAN WE STRUCTURE THE PHR/PHR-S TO TAKE ADVANTAGE OF THE SEMANTIC WEB?

• How will automated service and highly functional agents change the way we view and use our own health information? Once user needs are specified, the semantic web should be useful in gathering user specific information, filtering and feeding it to those who ask for it.

SOCIETAL AND EVALUATION ISSUES

WHAT ARE THE METRICS BY WHICH WE WILL MEASURE THE PROCESS, OUTCOME, AND IMPACT OF THE PHR/PHR-S AND ITS USE?

• Can evaluation metrics be linked to the functional specifications?

WHAT ARE THE MAJOR INCENTIVES TO THE PHR/PHR-S? DISINCENTIVES?

HOW WILL THE PHR/PHR-S AFFECT PROVIDERS?

• What do providers think of the PHR/PHR-S? What level of access will they provide, etc.?

HOW MIGHT THE PHR/PHR-S BE USED TO POSITIVELY IMPACT HEALTHCARE QUALITY? COSTS?

HOW MIGHT WE MEASURE THE LONG TERM IMPACT OF A PHR/PHR-S ON MORBIDITY AND MORTALITY?

HOW MIGHT WE MEASURE THE QUALITY OF PHR/PHR-S DATA, THE VALIDITY AND RELIABILITY OF DATA STORED IN PHR/PHR-Ss?

• What is the concordance of patient-entered and provider-entered data (e.g., problem list)? Should we be interested in this data at all?
• Does aggregate PHR/PHR-S data have regional or national relevance?

HOW MIGHT A PHR/PHR-S BE STRUCTURED AND OR USED TO IMPROVE PERSONAL HEALTH BEHAVIORS?

• The use of preventive services?
• Improved lifestyle choices?
• How can PHR/PHR-Ss help patients create a personal health plan?
• What is the impact of PHR/PHR-S on health status?