Sunday, June 05, 2005

Hospitalists serve no purpose

Here's a summary of the findings of a study presented during the last Society of Hospital Medicine meeting. David Meltzer MD, a well-known academic in the field of inpatient medicine studied the impact of hospitalists on outcomes, length of stay and costs in six large academic centers. He looked at some 31,000 hospital admissions over two years.

There have been many smaller studies that have demonstrated that having physicians who "specialize" in inpatient medicine improves all of the above. Meltzer's study however showed no significant improvements in any of these parameters.

Great. It was for my expertise as a hospitalist that I received my academic appointment at UCLA. I can imagine my department head at the next budget meeting. 'Now remind me, why did we hire Ford?'

In fact, I'm giving a grand rounds lecture on the benefits of the hospitalist paradigm in a few weeks. I guess I can make it a very brief one. I'm reminded of Gilda Radner's line while playing character Emily Litella: "Never mind!"

OK, maybe the title of this post is a bit melodramatic. I haven't thoroughly evaluated Meltzer's study as it hasn't yet been published. But judging from some of the methodology used, I'd say his results were somewhat predictable.

The whole idea of developing a hospitalist program is to improve the efficiency of managing inpatients. By having a physician who specializes in seeing inpatients and who is generally stationed in the hospital for most or all of the day, it is hoped that they will provide better care than primarily office based physicians.

The theory is that when doctors specialize in inpatients, they build up special skills in that unique area. They also become far more efficient and adept at navigating through the idiosyncrasies of their particular base hospitals simply because they spend so much time there.

In addition, most successful hospitalist programs have other features that distinguish themselves from traditional models. Generally they make a significant capital investment in ancillary services such as case managers (usually special trained RN's) who assist patient and family in their progress through their hospitalizations. There is also investment in other infrastructure that is important post-discharge such as home nursing, outpatient IV infusion services, outpatient physical therapy, etc.

These other features go a long way towards shortening hospital stays and optimizing outcomes.

There are many controversies that have arisen from what is increasingly being seen as the new norm when it comes to inpatient care (far more than I'm able to address in this short post). However, one of the most important motivations driving this practice is that hospital costs, lengths of stay and patient outcomes should all improve. It was axiomatic that specialists in inpatient care and the infrastructure designed around them had to be more efficient than an office doc making special trips to several hospitals to see only one or two patients and then rushing back to a full clinic.

So why didn't Meltzer's study confirm this? He himself speculates on the most likely reasons. What it come down to is that the "study" group, the patients followed by hospitalists, wasn't managed sufficiently differently from the control group. This was likely for the following reasons:
  • This study took place in academic institutions. Instruction from hospitalist to house staff may be expected to make the residents and interns function more like hospitalists on later rotations even though their future attendings might be non-hospitalists.

  • Since more hospitalists were placed on inpatient rotations during the study period, there was less demand for non-hospitalist attendings. Programs might then be expected to have placed their "better" non-hospitalist attendings to staff these inpatient rotations.
Both of these factors could help explain why the hospitalists didn't significantly outperform the non-hospitalists.

Another factor wasn't specifically addressed in the presentation. I seriously doubt that the study made provisions for the hospitalist group to have better ancillary services as described above than the control group. The presence of such services is known to be an integral feature of and greatly contributes to the success of any hospitalist program. Again, this would tend to make the two groups more similar than different.

For these reasons, I still have faith in this paradigm. I've seen it make dramatic differences in several large commercial settings when adopted (with very good metrics in each case) and there are multiple formal studies in community hospitals that bear this out.

Bottom line: I'm not yet looking for another career change.


Anonymous Anonymous said...

My only experience with the adult hospitalist paradigm was 10 years ago when I was working IV therapy in a community hospital. I only saw her at codes. I watched more than one private attending step aside when she walked in the room - not because she demanded it, but because they wanted the best for their patients.

I watched her help a cardiologist decide when it was time to call a code. She did it gently and very professionally. He was reluctant to quit, because he'd been speaking to the patient only moments before he coded and the patient was fairly young. She suspected a ventricular aneurysm, and explained why.

I watched her end of the conversation about DNR status for a "house case" patient whose only living relatives were 300 miles away.

I know she did a lot more than the little I saw but she was certainly a very valuable member of the team. It would be interesting to repeat the study in community hospitals large enough to have hospitalists.

June 07, 2005 9:01 AM  
Anonymous Anonymous said...

I do not think a study of the utility of hospitalists done in a university setting (or any setting with residency training programs) would be a valid one. House officers generally act as hospitalists in that there is always someone in the hospital available to do re-evaluations and make decisions on in-patients and that is where the majority of the efficiencies are achieved. The study would have to be done in a (relatively) large community hospital such as mine which has >300 beds and has two general hospitalists, a pediatric hospitalist, and several intensivists. My experience (I am in private single-specialty practice) has been that these physicians do add to the quality of care and better outcomes for the patients (especially the intensivists) as well as shorter lengths of stay. I would certainly like to see a well designed study done.

June 07, 2005 12:00 PM  
Anonymous Anonymous said...

I work with and around a number of hospitalists, and like other physicians, some are better than others. Sometimes it seems that maybe they are so busy trying to keep up with all the patients coming in that they do not have time to step back and see the big picture.
In some hospitals, it seems that the presence of hospitalists allows the ED physicians to now freely admit patients, then let the hospitalist figure out whether they need to be in the hospital or not.
Sometimes hospitalists (in my view) overuse testing, overuse consultants and can't seem to bring what should have been a brief stay to a close. Others underuse tests and consultants, and discharge patients with a lot of loose ends and unanswered questions (that impact on preventative medical decisions).
So there is nothing unique about hospitalists; perhaps the benefit that comes from them is that they have to a large extent replaced some older physicians with antiquated management styles -- if that's the case one would expect the lustre of the hospitalist to fade with time.

June 07, 2005 6:57 PM  
Anonymous Anonymous said...

In my experience, the most valuable asset of the hospitalist is the presence of a physician who will see/admit an ER patient. It's like pulling teeth to get a private practitoner to get in there and do his job. After numerous acrimonious complaints about whose responsible for admitting this or that ER patient, one hospital literally put all the miserable QA questions on the back burner waiting in anticipation for their hospitalist program to get up and running. It was like the calvary finally rescued them.
Big problem with admissions, though. Surgeons complain the hospitalist will bring in a surgical abdomen so they go in to make rounds and find a consult that needed to go to the OR four hours ago.
One hospitalist admitted an infected total hip, consulted ortho only to have the orthopod basically tell him to get lost. The orthopod wrote a single note in the chart indicating he was not going to take care of the patient then signed off. Patient sat there for a week, none of the orthopods willing to take him on. The hospitalist finally gave him a one-way bus ticket to the university and discharged him.
The bottom line:
1. They're essential. Some hospitals would implode without them. I don't know why, but these days, you just can't get these private practitioners to come in and admit people. Constant battle.
2. Relations between specialists and hospitalists, particularly in the ER admissions scenario are most extraordinary. Hospitalist are in danger of getting stuck with a difficult case when they go the extra mile to admit a patient.
Is the orthopod in the above case liable for an EMTALA violation? Abandonment? Anything? I don't know.

June 04, 2008 4:34 PM  
Anonymous Anonymous said...

Most times in medicine, when we try to fix something we break something else. One example of hospitalist programs creating costs is due to their inherit tendency to be owned and run by people other than the hospitalists themselves.

This leads to an inability of the hospitalist to say "NO" The result...

While hospitalists MAY decrease length of stay and costs. The program itself leads to a greater number of unnecessary admissions.

Patients that were babied over the weekend by the traditional internists, and now "Just admit to the hospitalists".

It's just easier to rid oneself of ALL responsibility by "just admitting to the hospitalist" than to do the right thing, the less expensive albeit more work intensive thing for the Internist on call at night or on the weekend, OR ON CALL FOR ANY OTHER TIME.

Hospitalist programs increase unnecessary admits.

June 20, 2008 7:19 AM  
Anonymous Anonymous said...

My experiences with Hospitalists have been horrific. The large university hospital in NC uses Hospitalists. That hospital also has an accompanying clinic.

In the situations I have observed, the Hospitalists have refused to communicate with the clinic physicians - including to the extent of denying to acknowledge what the clinic physician thought should be done for the patient.

I see turf wars. I see the hospitalists starting from scratch with complicated (elderly, cancer patients) who have had recent clinic contact.

If Hospitalists would put aside their egos and utilize a procedure to inform the clinic physicians when their patients are admitted then a lot of time with poor assumptions, wasteful tests would be avoided.

And, moreover, the Hospitalist may indeed be made aware of a condition of the complex patient that would change the course of the hospital plan.

I have observed Hospitalists looking for an acute problem and then losing interest when no such acute problem exists. Effectively, that puts the Hospitalist in the category of a triage practitioner.

October 29, 2008 1:38 PM  

Post a Comment

<< Home