There are very serious inequities in our healthcare system, and one of the most disturbing is that there are many pieces of evidence that minorities receive worse care than white people and have commensurately worse clinical outcomes. That is an important part of the narrative we are taught in medical school to motivate what amounts to sensitivity training. We are essentially taught that clinicians have latent levels of racism, that those who are inattentive to cultural differences or not sensitive to the different backgrounds of their patients are delivering substantially worse care. There are important issues, including difficulty in communication with patients from different linguistic and cultural backgrounds and sometimes a lack of trust in Western medicine and the predominantly white culture it represents. We are exposed to anecdotal horror stories. However, are physicians really that ethnocentric, racist and insensitive as a group?
The more current research suggests not. Motivated to look into these serious racial inequalities, Darrell Gaskin, professor of health economics & African American studies at UMD (now at Hopkins), and his colleagues did a very extensive analysis which covered 44% of the US population. 44%!
They did find that there was substantially worse care for minority patients across several different conditions and treatments. However, when they looked closely, they saw that the differences were all attributable to the fact that the minority patients were much more likely to be receiving care at a worse hospital with less well trained doctors.
“There is empirical evidence demonstrating that compared to white patients, minority patients tend to use specialists with poorer clinical outcomes and primary care physicians with less clinical training and with less access to specialists and hospital-based specialty services.”
Was there anything left to indicate that minority patients were receiving worse care at some hospitals?
“For African American and Hispanic patients, we found significantly higher rates of adverse events in less than 5 percent of hospitals, with the exception of decubitus ulcer for African Americans. More than 13 percent of hospitals had higher decubitus ulcer rates for their African American patients than for their white patients.”
If statistical significance is test at the p=0.05 threshold, that means that about 5% of the time you can expect to find a difference; it is a attributable to random error. That suggests that even in hospitals where there seems to be some difference, it could be due to random chance.
However, the important part of the story is that at a given hospital, patients seem to be all treated about the same. You can find minority patients who receive poor quality care, but you can find white patients as well who receive crappy care, in about the same proportion. Healthcare providers seem to be decent at providing roughly the same quality of care, on average, to all their patients.
How do we fix this problem if there are poorly performing hospitals that are treating minorities in high proportions?
“Several studies have shown that minority patients are concentrated in a relatively fewer hospitals and that these hospitals tend to be lower performing.”
That should be good news. That means that a lot of the effect is from relatively few hospitals and that attention can be focused there if we want to reduce iniquities in healthcare delivery. It’s not a bad thing to teach cultural and racial sensitivity to medical students, in fact it’s a great thing. However, it is not clear to me that is where the change needs to happen. We should be figuring out ways to incentivize specialists and physicians with experience and advanced training to work in underserved communities and what are currently poorly functioning hospitals. It will be interesting to see if expanded healthcare coverage will improve some of these serious issues.