A new study out this month in Medical Care by Yongkang Zhang and colleagues finds that people in struggling neighborhoods have considerably higher spending on potentially preventable health care. While overall health care spending was similar between the most deprived areas and average communities, the finding about spending on potentially preventable health care tells an important story of primary care.
Interpreting health care spending is difficult. In an economically struggling community, we expect to see higher spending. It would reflect the poorer health that such communities tend to experience. But when we see spending is comparable, we should wonder whether patients are having trouble getting needed care.
This study aimed to provide a clearer picture by controlling for measures of health status. But that muddied the waters for me. One would expect that neighborhood deprivation would lead to differences in health care spending because of differences in health. Ruling health out of the equation might, in fact, explain the null finding about overall spending. The authors did find higher spending among the most well-off neighborhoods, even after controlling for health status, suggesting some overuse.
Analyses of potentially preventable care offer a clearer picture
What I found more valuable about this study was its exploration of Medicare spending on potentially preventable health care services. If patients have a reliable source of primary care, they are unlikely to need these high-cost services. We know that quite a bit of health care is considered of low or no value. This includes preventable health care services, and many organizations are hard at work to rid the system of this unneeded care.
The research team examined three types of preventable health care: emergency department visits, hospitalizations, and 30-day hospital readmissions. First, they applied an algorithm to identify “non-emergent” and “emergency, but primary care treatable” emergency visits. Then they used the Agency for Healthcare Research and Quality’s Prevention Quality Indicators to identify preventable hospitalizations (e.g., for diabetes and hypertension). Finally, they used the 30-day unplanned readmission measure from the Centers for Medicare and Medicaid Services.
The study found that spending on potentially preventable health care was $53 higher per person in the most disadvantaged communities, compared to the average. Higher preventable emergency department visits and higher preventable hospitalizations drove this, not 30-day readmissions. It is also worth noting that the most well-off communities had lower rates of preventable health care use compared to the average. This was also driven entirely by fewer preventable emergency and hospital visits.
A hidden story of primary care
The authors add plenty to the literature by showing that measures of neighborhood deprivation predict health care spending. This is a growing field of study with a lot of momentum thanks to the development and public availability of the Area Deprivation Index.
The most valuable finding, in my opinion, is what the study reveals about primary care. Rather than rely on other research to show that primary care reduces preventable health care use, the team tests it. The team describes the results as a sensitivity analysis, and, thus, buries the tables in a set of online supplemental files. But the conclusion is critical and headline worthy.
The authors controlled for measures of primary care physician supply and utilization. When they did, the relationship between neighborhood deprivation and preventable health care use was reduced. It also became non-significant (see the figure below from the online files). In as much as a study can show the effects of primary care on unnecessary, expensive care, this one does it.
Note: Quintile 5 reflects the least well-off (most deprived) communities, and Quintile 1 reflects the most well-off (least deprived). The red line is the comparison group, reflecting communities with an average level of deprivation.
A conclusion worth amplifying
That primary care is the cornerstone of the health care system is not a new idea. I’ve argued before for the value of primary care, as have many of my colleagues. They have written about building the primary care workforce, improving the workplace climate in primary care, and tailoring primary care to unique communities. Yet it is easy to forget how critical primary care is for improving the health care system.
The publication last week of a tribute to Barbara Starfield, who defined and built the field of primary care research, reminds us that the answers to many of our health system conundrums (here in the U.S. and abroad) lie with improving primary care first.
Zhang and colleagues are right to call for research into whether neighborhood deprivation and health care are causally related. And they argue that these measures could help identify high-risk patients or, at least, patients in high-risk communities.
The story of primary care that the authors uncovered is the one that needs the most amplification. Yes, we should all work to improve neighborhood social conditions for many reasons. But even in the most well-off neighborhoods, we can do much to reduce health care spending by boosting primary care.
Gregory Stevens