Published on August 17, 2018
Take a minute to look outside your window. Are you surrounded by parks, health food stores, and neatly maintained sidewalks … or fast-food restaurants, liquor stores, and bowling alleys?
Physical environments play a significant, but undervalued, role in determining our overall health status, quality of life, and longevity. In fact, a person’s ZIP code is a stronger predictor of his or her overall health than other factors, including race and genetics.
Are there specific neighborhood characteristics that correlate with poor health outcomes and high healthcare spending? To answer this question, the Carrot Health data science team analyzed a wide range of geographic and proximity data variables (e.g. distances to certain types of businesses and “walkability” scores), and developed a mathematical model that predicted which areas of our country are in the top 5% of total standardized healthcare spending. (This measure is normalized for the different amounts that insurers pay for the same service in different areas). When we validated our model, we found that it was able to explain 47% of the actual observed healthcare spending in a given neighborhood.
The following table summarizes some of the key predictive factors that we discovered. Neighborhoods with lots of fast food restaurants and long distances from fitness centers, for example, were much more likely to have high healthcare spending. Places with many fitness centers and plentiful retail/shopping options, on the other hand, were very unlikely to have high spending.
On a relative basis, school quality (as measured by standardized test scores) turned out to be five times more predictive than any other geographic variable. In other words, it takes a ton of fitness centers to overcome the burden of poor education outcomes in a given neighborhood.
As a group, residents of neighborhoods with high healthcare spending face enormous economic, social, and behavioral challenges. They tend to have much lower incomes than average, and face far higher risks of food insecurity, violence, and tobacco usage, as the following table illustrates:
If we want to tackle the problem of high healthcare spending, we need to address these underlying social determinants of health — along with the neighborhood environments that reinforce and foster them.