Research and journalism

There's an article in last week's issue of Annals of Internal Medicine that compares retail health clinics to primary care physicians, urgent care centers, and hospital emergency departments. It arrives at the conclusion that these retail care clinics provide care statistically equivalent to that given by PCPs and urgent care centers and superior to that given by emergency departments in cases of otitis media (middle ear infection), pharyngitis (sore throat), and urinary tract infection (UTI). This article was reported on today by MSNBC; you can see that story here. There are, to my mind, enormous problems with both of these pieces, and I want to address some of them here.

I'll start with the MSNBC article. The egregious error lies in the author's assertion that "To track the quality of care, the researchers studied outcomes for three routine illnesses..." This is incorrect. The study does not (and does not claim to) measure outcomes. Rather, it uses quality metrics based on standardized instruments and professional guidelines. Outcome has absolutely nothing to do with it. The reason that this is such a serious lapse is that "outcome" is, essentially, the bottom line for any patient encounter. When all is said and done, how did the patient do? By carelessly using this term, the author effectively creates an illusion that patients "do better" in one setting than another. In this case, yes, one word makes THAT much of a difference.

For the Annals article (written, ironically enough, by an UPSOM professor and his colleagues), I also take issue with a number of things. Most of these issues stem from a single gripe: the emergency department is not a good choice for a comparison group in the study. EDs, unlike retail clinics, PCPs, and urgent care centers, must by law accept all comers. They tend to have higher average acuity (EMERGENCY, people!). They tend to be starting from scratch rather than from an established diagnosis. They don't generally do follow-up visits or chronic care; they have you follow up with your PCP. Many times, in non-emergent situations, after ascertaining the lack of immediate danger the emergency physicians will get in touch with the primary care docs and allow them to manage the course of treatment. And most of all, they are not designed or intended to deal with minor issues. Translation: THE ED IS NOT THE PLACE FOR DIAGNOSING AND TREATING OTITIS, PHARYNGITIS, OR UTIs (or vaccinations, Pap smears, colonoscopies, mammography, or really any preventive health interventions -- all metrics used in this paper). Also, the article does not make it clear what kind(s) of EDs provided data for this research. Were they large academic centers? Children's hospitals? Small community hospitals? It makes a difference. Another point: the study compares the costs of care at the different health care centers, and the ED is by far the most expensive. Why? Because if you have insurance, you're basically paying for all the people who come into the emergency department without it. I wonder how much the numbers would change if they factored in the patients who received full ED services gratis. Anyway, the long and the short of it is that the emergency department was a poor choice for a comparison group.

Of course, the point of the article is that the quality of health care (as defined by the metrics, and in these three particular illnesses -- the latter is a rather severe limitation) is no worse at retail clinics than in other settings and that the retail clinic costs less. Great. Maybe they can start taking all of the patients that don't belong in the ED. But the numbers (and the conclusions) must be taken with a good-sized grain of salt, especially in the comparisons between health care providers. Anyway, read the article (by Mehrotra et al) and tell me what you think.

Peace and God bless!