This week, JAMA Internal Medicine published interesting original research by Dr. David Grande et al entitled “Public Preferences About Secondary Use of Electronic Health Information” (subscription required). The researchers used conjoint analysis to determine the patients’ preferences in regards to sharing personal health information in several different scenarios. The scenarios elicited preferences pertaining to sharing of data with/without genetic information (about their risk for cancer) with one of three entities (a university hospital, commercial entities, or public health), and for one of three purposes (research, quality improvement, or commercial marketing).
Interestingly, among the 3,300+ adults that did the survey, the purpose that garnered the most support was research; quality improvement was second. The authors were somewhat surprised that respondents supported research over quality improvement because QI is more likely to generate immediate benefits for individuals than research. Respondents also indicated that while they care about the user, they care about the use of the data more — but don’t particularly care about the sensitivity of the data shared (at least given the scenario presented).
The authors drew two additional conclusions:
- Racial and ethnic minorities were no less supportive than white participants of sharing health information.
- Individuals with distrust of the health system, no usual source of care, and those who had experienced cost barriers to care were less supportive of secondary use of their health information.
The authors suggest that this research should help inform debates about the secondary use of health information given that, in some cases, identifiable data used with permission might be more valuable and informative than deidentified data. They also conclude that:
Our findings also suggest that organizations and institutions within the healthcare delivery system are the most trusted stewards of health information and should play an instrumental role in efforts to extend the uses of population-level health information.
The American College of Physicians published a policy paper in 2011 entitled, “Health Information Technology & Privacy.” The following are some of the key statements from that position paper relevant to these issues:
- Privacy policies should accommodate patient preference/choice as long as those preferences/choices do not negatively impact clinical care, public health, or safety.
- ACP believes that under a revised privacy rule, permitted activities not requiring consent should include well-defined socially valuable activities involving public health reporting, population health management, quality measurement, education, and certain types of clinical research. Further, ACP supports the following principles on the use of Protected Health Information (PHI) and Individually Identifiable Health Information (IIHI):
- The sale of any IIHI without the patient’s permission should be expressly prohibited.
- Whenever possible and appropriate, de-identified, anonymized, or pseudonomized data should be used. The method used to remove identifiers should be publically disclosed.
- IIHI should only be supplied in cases where such information is necessary for proper performance of a specific function. For example, if the goal is to count incidence of a disease or count the number of patients receiving an intervention, there is no need to include IIHI. Determination of the need for identifiable information should be made by appropriate publicly accountable decision-making bodies (e.g., Department of Health and Human Services, regional or local Institutional Review Boards [IRBs]).
- Regarding research, a revised privacy rule should maximize appropriate uses of information to achieve scientific advances without compromising ethical obligations to protect individual welfare and privacy.
The findings of this study lend some credibility to the anecdotal observations that people are surprisingly willing to share personal information on social networks — including personal medical information — presumably because they perceive value from doing so. As health systems aggregate larger and larger data sets, national registries grow (see NEJM article on The Randomized Registry Trial), broader discussions about the secondary use of personal health information, and education of the public could accelerate the opportunities for novel research studies and quality improvement efforts.