Precision Medicine’s Impact on the Doctor-Patient Relationship

By Paul S. Appelbaum and Gil Eyal

Healthcare is now entering a much-anticipated era of “precision medicine” (PM), an effort characterized “[b]y taking into account individual differences in lifestyle, environment, and biology” to “accelerate research and improve health”.

The extent to which these goals are achievable and generalizable has been extensively debated, but often lost in the discussion is the likely impact of PM on the doctor-patient relationship, still the primary means of delivering medical care. As we and our colleagues noted in our recent commentary in Genetics in Medicine, multiple changes are looming in both the doctor’s and the patient’s roles. Here we highlight just two of them: the proliferation of uncertainty and its downloading to the patient; and the diminution in the control and authority of the physician.

Insofar as PM involves the ongoing collection of information on biomarkers (including but not limited to genetic data), environmental exposures (such as pollutants in the air), and behaviors (from sleep to exercise), it will become possible over time to identify a growing number of risk factors. Some—such as elevated blood lipids—may call for specific prophylactic interventions, but the implications and actionability of other markers will be much less certain.

For many risk genes—think of the Alzheimer gene variant ApoE4—no proven interventions exist. And unless a city-dweller is inclined to give up her job and move to more bucolic environs, exposure to air pollution will continue to be an irremediable fact of her life.

Indeed, as polygenic risk scores are developed for complex disorders, data on health risks will multiply for all of us, but their implications for action will become even murkier.

We will move from a dichotomous status of “healthy” or “sick,” with all of the supports for the latter role that have evolved over the years, to the liminal status of “at risk,” in many cases able to do nothing but wait to see if the risk materializes or encouraged to take costly precautionary measures that may prove unnecessary.

Physicians’ roles will also be affected by some of these same trends. Their patients will turn to them for answers regarding the implications of the copious data collected about their health status. However, most physicians today report discomfort in interpreting genetic findings, much less grasping the effects of interactions among genetic, environmental, and behavioral risks.

Physicians and patients alike—and increasingly patients alone—will turn to new services, offering the ability to integrate large amounts and diverse types of data, that will arise to interpret the findings and identify risk-mitigation strategies. Although some commentators have suggested this will lead to a “patient-centered” system, with patients in greater control of decisions about their care, it seems more likely to us that both patients and physicians will surrender control to new entities with access to and mastery of their data.

Although these perturbations in doctor and patient roles seem inevitable, their negative consequences could be mitigated by thoughtful steps taken now.

Careful consideration of the types of data to be collected, based on a risk/benefit analysis with patient benefit as the dependent variable, is a path that could avoid overwhelming people with inherently uncertain information about their health risks. The development of interpretive approaches that can be utilized by physicians, who themselves will need to be educated to deal with the “datafication” of medicine, is a way of maintaining their value as mediators and advocates for their patients.

The time to start taking these and other steps is now.

 

Paul S. Appelbaum, MD is Dollard Professor of Psychiatry, Medicine and Law at Columbia University. Gil Eyal, PhD is Professor of Sociology at Columbia University.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.