I’ve been working with a team of graduate students in the Health Informatics Program at Indianapolis University Purdue University at Indianapolis (IUPUI) studying the relationship between food/housing insecurity and health status. We are completing a paper for submission to a peer-reviewed journal. In this and the next post, I’ll talk about some of the interesting findings.
It is intuitive that if you are always or usually worried about having enough money for food or a place to live, that your health status is likely to not be as good as if you rarely or never worry about food or housing. Conversely, it is also intuitive that if you have poor or fair health (as opposed to good, very good or excellent health), that your income potential is less and you are more likely to always or usually worry about food or housing.
But from a policy perspective, or from the view of a healthcare or social services provider focused on sustainability during times of uncertainty, its not enough to say “That makes sense. It seems logical.” You need to quantify the relationships or at least have metrics that are indicators of more complex relationships in order to establish a base line, to measure interventions and to predict/quantify/evaluate outcomes.
This is particularly important when the Congress and the Trump Administration have created new opportunities, or at least reduced financial uncertainty for healthcare innovation. And this at a time when social determinants of health are receiving increased attention resulting in innovations that promote community-wide coordination and collaboration of care between physical health, behavioral health and social service providers.
Our team of graduate students from IUPUI have been thinking about this. How can we measure the association between food/housing insecurity and general health? It turns out the data is available to at least get a decent start answering this important question.
WHAT DATA DID WE ANALYZE? We used data from the 11 states and the District of Columbia that completed the social context module of the 2015 Behavioral Risk Factor Surveillance System (BRFSS), an annual survey of health and social behaviors administered by the CDC and each of the states and territories.
HOW DID WE ANALYZE THE DATA? In total and by state/district, we estimated the prevalence of housing and food insecurity among persons who reported their general health was poor or fair. Logistic regression models were used to assess the effects of food and housing insecurity on general health and, conversely, of general health status on food and housing insecurity.
KEY RESULTS: We calculated a 48.12% probability that people who are always or usually worried about having enough money for food and housing have poor or fair health, compared to 27.28% of people who only sometimes worry about having enough money for food or housing. (Rather than looking at the probabilities, the graph on the right shows similar results when simply looking at the percentages of people who report poor or fair health by level of food/housing insecurity.)
DISCUSSION: Innovations that move people from “always or usually” worried about food and housing to just “sometimes” worried, could have a significant impact on general health status. Conversely, the data suggest that moving someone from “poor or fair” health to “good” health could yield a similar measurable improvement in food/housing insecurity. In fact, while this analysis is not sufficient to predict the specific change in dependent variables, it does suggest that a shift from always or usually worried to just sometimes worried could result in a decrease in the number reporting poor or fair health on the order of 15%! That would be huge.
CONCLUSION: In this time when the federal government and states are looking for innovations that will improve health outcomes and lower cost, this kind of analysis, using the BRFSS in concert with screening and health risk assessment tools, could be of tremendous value. A more complete understanding of how the BRFSS could be used to measure the effectiveness of new innovations, and the effect on total cost, is needed. For our next study, we want to also look at what such a change in health status would do to the total cost of health care.
In my next post, I’ll talk about the similarities and differences between states with respect to the relationship between food/housing insecurity and health status. It’s pretty interesting stuff.
If you want to see our working draft of our paper, send me a note.