The hospitalist movement continues to grow. Are there unintended consequences?
A few months ago, a family member was in the hospital and called me, quite upset, saying his PCP hadn't been by to see him since he was admitted. For a moment I didn't even really understand what he was talking about. Then I, somewhat sheepishly replied, "We don't really do that anymore".
The reason we don’t do that anymore is that now we have hospitalists. The number of hospitalists has been growing dramatically over time, as outpatient offices have gotten busier and more complex, leaving PCPs with little time to make rounds on the local wards.
But this state of affairs leads to an interesting question. Would your primary care provider, someone who knows the ins and outs of your medical history, or a hospitalist, someone who knows the ins and outs of a complex healthcare behemoth, treat you better when you are hospitalized?
That's the question that this article, appearing in JAMA Internal Medicine attempts to answer.
The researchers used Medicare data to identify 650,651 older adults with a hospitalization in the US in the year 2013. Based on billing records, they determined if the primary physician caring for the patient was a hospitalist, the patient's own PCP, or a "other generalist". That third category were basically non-hospitalists who also didn't have a prior relationship with the patient - potentially cross-covering outpatient docs.
After accounting for severity of illness, hospital factors, patient comorbidities and diagnosis-related group, some really interesting results emerged.
I'll start with the big finding. 30-day mortality was 10.8% among those cared for by a hospitalist, compared to 8.6% among those cared for by their own PCP - a small, but significant difference. If this difference were true and causal, that means that there is 1 excess death for every 50 patients treated by a hospitalist - and that could add up quickly.
But is it causal? I spoke with Dr. Jennifer Stevens, the study’s lead author to get to the bottom of it.
“We would expect if this were all unmeasured confounding, then the readmission findings would be different”.
What she is pointing out is that, if hospitalists just take care of sicker patients than PCPs, wouldn’t the readmission rate be higher for the hospitalists? But to be fair, she also cautioned me that these results haven’t been replicated and should only be hypothesis-generating.
But I think it is likely that there is some value in knowing your patient well. There is some value in knowing the hospital well too. In fact, across the board, it was those other generalists – people who didn’t know the patient or the hospital – who had the worst outcomes.
Dr. Stevens believes that some patients might need that special relationship that comes from a PCP.
“There might be some populations of patients who might particularly benefit from that level of familiarity. Maybe they are the very old. Or maybe they are the patients with complex clinical conditions.”
Hospitalists are here to stay, and I, for one, am very thankful for the hard work they do. But perhaps finding small ways to integrate PCP care into a hospital stay might change the dynamic for the better. Your move, Skype.