Personal Ecology in Accountable Care Outcomes

Two days ago, the Institute of Medicine (IOM) published the first of two reports on the role of social and behavioral factors as determinants of health, and advocated for the inclusion of social and behavioral items in the electronic medical record (EMR). The report is available for free, and you can get it here:

http://www.iom.edu/Reports/2014/Capturing-Social-and-Behavioral-Domains-...

Actually, the IOM is not the first to say this, as the Information Systems group at the National Institutes of Health long ago created a list of what those social and behavioral items should be.

https://www.gem-beta.org/public/EHRInitiative.aspx?cat=4 

At Sustainable Health Systems, we call these factors the personal ecology - the individuals social surroundings, their family or childhood experiences, lifestyle choices such as diet, tobacco, stress-coping strategies and amount of regular exercise, transportation resources, financial situation, and so forth. When I was in college and in the first year of medical school, this was taught to me as the biopsychosocial model, but the further I passed from those years of training the less that phrase was used. Medical treatment today extensively exploits the biomolecular model, while surgery exploits materials and high-technology engineering, and fewer and fewer physicians use the word biopsychosocial.

Actually, even the biopsychosocial model is probably obsolete today, because in the 30 years since I was a med student, epidemiological studies have made it increasingly apparent that economic factors are also a strong determinant of health outcomes. Rather than use biopsychosocioeconomic, at Sustainable Health Systems we just call it ecology. We couldn't agree more with the IOM, and we've been drawing attention to patient ecology as a missing element by calling it the "dark matter of health care", because, like dark matter, we know it's out there and that it's ~85% of the universe, but we can't detect it.

http://www.countyhealthrankings.org/roadmaps/what-works-for-health 

Why can't we detect patient ecology as factor in health outcomes? Because it's not in the EMR nor in the payer's claims database. If the boundary conditions are not in the control variables of a experimental data set, then one cannot statistically demonstrate their effect on the outcome of the experiment. The Pennsylvania Chronic Care Initiative failed to work on 10 of 11 outcome variables (http://jama.jamanetwork.com/article.aspx?articleid=1832540), and I suspect that's because the study design focused on practice protocol and made little effort to control for patient ecology as a boundary condition for that study.

Despite the fact that public health researchers have thoroughly documented that non-clinical factors determine ~80% of health outcomes, health care administrators have forged ahead with business models that seek to make physicians, physician networks and hospitals accountable for those health outcomes. Physicians have been to much like lemmings, agreeing to being held accountable despite the fact that they were trained that the vast majority of things that determine health outcomes aren't medical factors! I know they should know better, because we all sat in the same classes that discussed these ecological factors! 

When I was a student, I kept wondering why physicians only try to fix things after they're broken, and why so little effort was spent on preventing harm from coming in the first place. The answer is complex, involving an evolution of bioethical understanding of the physician's role vis-a-vis a patient's personal life choices (even though much is not a choice of volition, e.g., adverse childhood experiences), as well as a distorted business model that rewards disease-care and is especially lucrative when it involves high-tech drugs or devices. I'll take this new report from the IOM as a sign that the issues I've spent my medical career working on have been important after all.

- GEM