Debate continues on the value of the hospitalist model

Two academic leaders in hospital medicine, Dr. Mark Williams and Dr. Robert Centor (our own DB) debated the model in the current issue of Archives of Internal Medicine. Links to the point-counter point exchange can be found in DB’s post on the topic.

In short, Dr. Williams maintains that the model improves outcomes and efficiency, that the evidence is in and the debate is over. DB says “not so fast”, citing mixed evidence and marked variation in the model. As much as I’m excited about the potential for the hospitalist model and as much as I love my hospitalist career (that’s my conflict of interest disclosure) I have to go with DB on this one.

For those who can access the articles in the original, many aspects of hospital medicine were covered. I’ll restrict my comments to Dr. Williams’s claim that the model has been proven to improve efficiency and outcomes. From where I sit the evidence is all over the map. It’s mixed at best. I’ve blogged about it many times, most recently here. The most talked about study last year was this one published in NEJM. The conclusion of that largest ever study on hospitalist outcomes and efficiency was underwhelming. There was no improvement in patient outcomes. Hospitalist care was associated with decreased charges per case in comparison with internists but not family practitioners.

Although a number of smaller studies showed superior efficiency with the hospitalist model the next largest study, and one with superior design, showed no efficiency or outcome benefits. It was a prospective multicenter study presented at SHM 2005 which you can access in this issue of The Hospitalist. The results:

Twelve thousand and onepatients were cared for by hospitalists and 19,890 by non-hospitalists. There were no statistically significant differences in age, race, gender, Charlson Index, or distribution of primary diagnosis be­tween the 2 groups. There were no statistically significant differences in in-hospital mortality, 30-day readmission and emergency room use, 30-day self-reported health status, or patient satisfaction. Mortality data up to 1 year after admission are pending. Average length of stay was 0.05 days shorter for hospitalist patients but this difference was not statistically significant. Costs were also similar between the groups.

We’re still waiting, by the way, for the one year mortality data. What’s important about that study? It’s the fact that, following some early hype in the blogosphere (here, here, here and here) it got buried. It wasn’t mentioned in Dr. Williams’s article. And because it was never published in a Medline indexed journal it was not included in the systematic review cited by Williams and others who promote the model.

This debate will never be settled. There will be few, if any, new studies. With traditional practitioners fleeing hospitals in droves, soon there will be no comparison groups against which to study the model. It’s a moot point. The model is here to stay. We don’t need these metrics to establish our value.