|Ted Schettler, MD, MPH
CHE Science Director, and Science Director of the Science and Environmental Health Network; Coordinator of CHE’s Science Working Group
What is health? How do we measure it? What determines it? A new report from the Robert Wood Johnson Foundation and the University of Wisconsin Population Health Institute takes a turn at answering these questions. “County Health Rankings: Mobilizing action toward community health” combines weighted measures of health outcomes and health determinants to define and rank the health of individual counties throughout the US.
Premature mortality and morbidity, estimated by a combination of how healthy people feel and the percent of low birth weight babies, were given equal weight as measures of health outcomes. Various measures of health behaviors, clinical care, social and economic factors, and the physical environment were weighted as measures of health determinants and expressed as composite scores. The website also provides comparative rankings of health outcomes and measures of health determinants at the county level. The authors describe their methods of data collection and analysis in some detail, including a justification of their final choices for weighting individual variables in the composite score. This remarkable effort is worth exploring.
At the outset, the authors had to choose among many candidate health outcomes and determinants. How did they do? Are these the right measures of health or just available statistics? What other health determinants could reveal important insights? Several limits are notable. Among them: This is a cross-sectional analysis and cannot determine causal relationships. Genetics, gender, race, and ethnicity are left out. The relative contribution of each group of health determinants always adds up to 100%, regardless of context. (Health behaviors were assigned a weight of 40%, health care 10%, socioeconomic factors 40%, and the physical environment 10% for the final composite score. The justification for these relative weights and alternative opinions are available on the website.) Thus, potential interactions among health determinants are not considered.
For example, the report uses air pollution as a measure of environmental quality, yet it doesn’t acknowledge that people who are socioeconomically disadvantaged are more susceptible to the health effects of air pollution than people who are better off. It’s increasingly clear that air pollution causes more asthma and asthma attacks in children lower on the socioeconomic ladder, independent of other environmental exposures. How should we think about this? Is the problem air pollution, socioeconomic stressors, or both? How does our answer influence what we propose to do? If we ignore interactions, we not only underestimate the impacts of combined eco-social variables in vulnerable groups but also set ourselves up to fail to identify interventions that can have multiple, cross-cutting benefits.
Despite its inevitable simplifications and assumptions, this detailed report deserves attention and discussion, especially among those of us embracing an ecological model of health. It raises many interesting questions. Is this the right mix of individual and county-wide variables? Are there other measures of health at the county level worth identifying? Are there other measures of environmental quality and integrity that should be added?
Maybe the report’s biggest contribution will be to set the stage for soliciting ideas about what to do with the information. Should high-ranking counties be complacent? Should they compete with themselves to improve? Counties struggling with poor health outcomes and multiple adverse health determinants will need something more than disconnected, poorly-coordinated activities. They must understand that risk factors don’t exist in isolation but rather in a complex, interactive web of causation. In those counties, the entire web needs fundamental transformation, achieved through creative, strategic interventions. This is no small task. I think of Donella Meadows’ Places to Intervene in a System in which she says: “There are no cheap tickets to systems change. The higher the leverage point, the more the system resists changing it.” Perhaps this is where collaborations like CHE come in…strength in numbers, ideas, and mutual support.