For those who say they would pay $50 more per patient if the quality is better, here’s the problem: Show me the data that say hospitalist care is higher-quality.
I suspect most of you have reviewed the study or at least heard about it. Bob Wachter, MD, MHM, blogged about the study. An article about the study appeared in American Medical Association News. Even National Public Radio ran a piece about the study on their show “Morning Edition.”
I am, of course, referring to the study by Kuo and Goodwin, which was published in the Annals of Internal Medicine in early August.1
In this study, the authors looked at a sample of patients (5%) with primary-care physicians (PCPs) enrolled in Medicare who were cared for by their PCP or a hospitalist during a period from 2001 to 2006. The authors stated their underlying hypotheses as:
- Hospitalist care would be associated with costs shifting from the hospital to the post-hospital setting;
- Hospitalist care would be associated with a decrease in discharges directly to home; and
- Discontinuities of care associated with hospitalist care would lead to a greater rate of visits to the emergency room and readmissions to the hospital, resulting in increased Medicare costs.
Did the authors say hospitalist care cost more? They can’t possibly be correct, can they? Don’t all the hospitalist studies show that hospitalists provide the same quality of care as primary-care doctors, except the costs are lower and the hospital length of stay (LOS) is shorter when hospitalists care for patients?
The point here is that these investigators look at the care not only during a patient’s hospital stay, but also for 30 days after discharge. This is something that had not been done previously—at least not on this scale.
Focus on Facts
And what did the authors find? Patients cared for by hospitalists, as compared to their PCPs, had a shorter LOS and lower in-hospital costs, but these patients also were less likely to be discharged directly to home, less likely to see their PCPs post-discharge, and had more hospital readmissions, ED visits, and nursing home visits after discharge.
Since its release two months ago, I have heard a lot of discussion about the study. Here are a few of the comments I’ve heard:
- “This was an observational study. You can’t possibly remove all confounders in an observational study.”
- “The authors looked at a time period early in the hospitalist movement. If they did the study today, the results would be different.”
- “The additional costs hospitalists incurred were only $50 per patient. Wouldn’t you pay $50 more if the care was better?”
- “This is why hospitals hired hospitalists. They save money for the hospitals. What did they expect to find?”
I agree that observational studies have limitations (even the authors acknowledged this), but this doesn’t mean results from observational studies are invalid. Some of us don’t want to hear this, but this actually was a pretty well-done study with a robust statistical analysis. We should recognize the study has limitations and think about the results.
Kuo and Goodwin looked at data during a period of time early in the hospitalist movement; the results could be different if the study were to be repeated today. But we don’t know what the data would be today. I suppose the data could be better, worse, or about the same. The fact of the matter is that HM leaders—and most of the rest of us—knew that transitions of care, under the hospitalist model, were a potential weakness. How many times have you heard Win Whitcomb, MD, MHM, and John Nelson, MD, MHM, talk about the potential “voltage drop” with handoffs?
