Saturday, November 23, 2013

Outliers, Hotspotting, and the Social Determinants of Health

We all have, or will have, our personal health problems, and the health problems that confront those close to us in our family and among our friends. Some are relatively minor like colds, or are temporary like injuries from which we will heal. Some are big but acute and will eventually get all better, like emergency surgery for appendicitis, and others are big and may kill us or leave us debilitated and suffering from chronic disease. Some of us have more resources to help us deal with these problems and others fewer. Those resources obviously include things like how much money we have and how good our health insurance is, but also a variety of other things that have a great impact in our ability to cope with illness, survive when survival is possible, and make the most of our lives even when afflicted with chronic disease. 

These other things are often grouped under the heading of “social determinants of health”. They include factors clearly related to money, such as having safe, stable and warm housing, having enough to eat, and otherwise having our basic needs met. They also include support systems – having a family and friends that is supportive and helpful, or alternatively not having one or having family and friends whose influence is destructive. It includes having a community that is safe and livable and nurtures and protects us and insulates us from some potential harm. This concept, “social capital”, is most well-described in Robert Putnam’s “Bowling Alone”[1] and its health consequences in Eric Klinenberg’s “Heat Wave”[2], discussed in my post “Capability: understanding why people may not adopt healthful behaviors”, September 14, 2010.

How this affects communities is a focus of the work of Dr. Jeffrey Brenner, a family physician who practiced in one of the nation’s poorest, sickest, and most dangerous cities, Camden, NJ, and is a founder of the “Camden Coalition”. I have written about him and his work before (“Camden and you: the cost of health care to communities”, February 18, 2012); his work drew national attention in the New Yorker article by Dr. Atul Gawande in January, 2011, “The Hot Spotters”. Brenner and his colleagues have taken on that name to describe the work that they do, and have collaborated with the Association of American Medical Colleges (AAMC) to focus on “hotspotting” ( and produce a downloadable guide to help health professionals become “hot spotters” in their own communities in ten not-easy steps. The focus of this work is on identifying outliers, people who stand out by their exceptionally high use of health care services, and develop systems for intervening by identifying the causes of their high use and addressing them to the extent possible, activities for which traditional medical providers are often ill-suited and health care systems are ill-designed.

The essential starting point in this process, emphasized by Brenner in two talks that he gave at the recent Annual Meeting of the AAMC in Philadelphia (his home town) in early November, 2013, is identifying “outliers”.  The concept of recognizing outliers was the topic of a major best seller by Malcolm Gladwell a few years ago (called “Outliers”[3]), and Brenner notes that they are the “gems” that help us figure out where the flaws, and the costs, in our system are. As described in Gawande’s article, Brenner was stimulated by looking at work done by the NYC Police Department to identify which communities, which street corners, and which individuals were centers of crime; rather than developing a police presence (and, hopefully, pro-active community intervention) for the “average” community, they were able to concentrate their work on “hot spots”. Moving out of a crime-prevention and policing model, Brenner and his colleagues were able to link to hospital admissions data that was tied to people and performed a “utilization EKG” of their community, looking at who had the highest rates of admissions, ER visits, 911 calls and sought to determine what the reasons were.

Unsurprisingly, the individuals identified most often had the combination of multiple chronic diseases, poverty, and a lack of social supports – pictures of the impact of poor social determinants of health. Sometimes there were individual, specific issues – like the person who called 911 multiple times a day and was found to both live alone and have early Alzheimer’s so that he couldn’t remember that he already had called. Often, there were predictable community and poverty related issues, related to inadequate housing , food, transportation, and poor understanding of the instructions given them by the health care providers that they had seen.

One example of such an effort is “medicine reconciliation”, in which (usually) pharmacists review the medications that a patient entering the hospital, clinic or ER is supposed to be on (per their records) and what they say they are taking. It sounds like a good idea, and it has received a great deal of emphasis in the last several years, but it is one that Brenner calls a “fantasy” because it doesn’t involve going into people’s homes and (with them) searching through their medicine cabinets and drawers to find the piles of medications they have, and often have no idea of how to take, which ones are expired, which ones have been replaced by others, which ones are duplicated (maybe brand vs. generic names or from samples). He showed a slide of a kitchen table piled high with medicines found in one house, and says that his group has collected $50,000 in medicines found in people’s houses that their current providers did not know they were taking or wanted them to take.

Brenner notes that continuous ongoing stress weakens the body and the immune system, enhancing production of cortisol (a stress hormone) that has effects like taking long-term steroids, increasing the probability of developing “metabolic syndrome” and a variety of other physical conditions. He also cites the work of Vincent Felitti[4] and his colleagues that have identified Adverse Childhood Events (ACEs), such as abuse, neglect, etc., being associated with the presence of being a high-utilizer sick person in middle age (and, if they reach it, old age). This is, he indicates, exactly what they have found doing life histories of these “outliers”. It suggests that while interventions at the time of being identified as a high utilizer can be helpful for the individual patient, for the cost to the health system, and even to the community; but it also reinforces what we should already know – that the interventions need to occur much earlier and be community-wide, ensuring safe housing and streets, effective education, and adequate nurturance for our children and their families.

We need, Brenner says, half as many doctors, twice as many nurses, and three times as many health coaches, the intensively trained community-based workers who do go out and visit and work with people at home. I do not know if those numbers are true, but it is clear that we need to have comprehensive interventions, both to meet the needs of those who are sickest now and to prevent them from developing in the future. We are not doing it now; Brenner says “Like any market system, if you pay too much for something you’ll get too much of it, and if you pay too little you’ll get too little.”

We need to have a system that pays the right amount for what it is that we need.

[1] Putnam, Robert D. Bowling Alone: The Collapse and Revival of American Community.Simon & Schuster, New York, NY. 2000.
[2] Klinenberg, Eric. Heat Wave: A Social Autopsy of a Disaster in Chicago. University of Chicago Press, Chicago. 2002.
[3] Gladwell, Malcolm. Outliers: the story of success. Little Brown. New York. 2008.
[4] Felitti, V et al., “Relationship of Childhood Abuse and Household Dysfunction to Many of the Leading Causes of Death in Adults The Adverse Childhood Experiences (ACE) Study”, Am J Prev Med 1998;14(4) (and many subsequent publications).

No comments:

Total Pageviews