Patient Care

Surgery’s Waterboys


The query came from the audience: “But isn’t comanagement really about us becoming the surgeon’s waterboy?” Encouraged by the chortling crowd, he furthered, “I mean, think about it: How much more demeaning can it get than to be the admit-ologist and discharge-ologist for the surgeon? They make all the coin and we just follow after them picking up their jock straps.”

Slack-jawed, I mustered what was, under the circumstances, a rather confident “Umm … ?”

This comment, from a talk I gave a couple of years ago at an SHM annual meeting about comanagement, took me a bit by surprise. Not because of the sentiment; that I get. It’s easy to feel that the comanagement we do suffices only to sate the surgeon at the hospitalist’s expense. Rather, I was taken aback because of its indication of the distance with which we’ve missed the comanagement bull’s-eye.

New Comanagement Data

A recent article regarding the comanagement of neurosurgical patients drudged this oratorical memory from its peaceful cerebral resting place between the 1982 Milwaukee Brewers’ starting outfield (Ogilvie, Thomas, Moore), my wife’s least favorite Beatle (Ringo), and the number of macaroni noodles my grade-school friend Mike could stuff into his nostril and cough up through his mouth (nine with aspiration, five without). In the paper, Auerbach et al report a retrospective, before-and-after study of 7,596 patients admitted to the neurosurgery service at the University of California at San Francisco Medical Center.1 The authors compared administrative, financial, and survey data for 4,203 patients before a hospitalist-neurosurgery comanagement arrangement to 3,393 patients after the program implementation—by far the largest trial of hospitalist comanagement to date.

The real comanagement story—indeed, the story of the whole of hospital medicine—is our need to fundamentally improve outcomes through systems improvements.

They found:

  • Shockingly, surgeons (“hospitalists make it easier for me to do my job”) and nurses (“I can easily and promptly reach a physician”) liked having us around.
  • Curiously, patients were rather indifferent (measured via patient satisfaction indicators) to our presence.
  • The cost of care decreased by about $1,500 per patient after the intercalation of hospitalists—this despite the fact that the length of stay was unchanged before and after model implementation.
  • Unfortunately, such traditional markers of quality as mortality and readmission rate remained stubbornly unchanged.
  • Encouragingly, nontraditional-but-likely-important indicators of quality (e.g. nursing and physician perception of improvements in care provision) were achieved.


This study adds significantly to our understanding of the comanagement model. The finding of costs savings is as expected (nearly all studies of hospitalist programs have shown cost savings) as it is unexpected (prior studies of comanagement models reported no cost savings).2 Likewise, the lack of improvement of hard quality endpoints (mortality and readmission rates) is consistent with most studies of hospitalist programs, including a previous report of comanagement of orthopedic patients that showed improvements only in minor complications, such as rates of electrolyte abnormalities, while improvement in the softer quality endpoints—nursing and surgeon satisfaction and perceptions of quality—is consistent with most reports and conventional wisdom.2

Within hours of publication, the blogs were throbbing with discussion of what this meant for the field of hospital medicine. Did this prove comanagement to be the godsend many believe (perceptions of improved quality), the complete farce that many believe (no evidence of mortality benefit), or was this just further confirmation that hospitalists are really nothing more than cost reduction-ists?

My opinion? This is just the comanagement MacGuffin.

MacGuffin Explained

Fans of film will know that the MacGuffin is a Hitchcockian plot device that uses a meaningless but often mysterious and intriguing element to drive the plot. So while everyone, it seems, is concerned with the MacGuffin, the MacGuffin exists only to help the story unfold. Think of the “government secrets” driving the plot in Hitchcock’s North by Northwest, or “unobtainium” in the movie Avatar. In both cases, the MacGuffin preoccupied the cast (they had to have it, or defend it), but in the end, the MacGuffin was insignificant except to move the plot forward.

In much the same way, the debate about whether the shared-care model of surgical patients is a good thing is comanagment’s MacGuffin; it definitely drives the plot but ultimately it misses the point. The real comanagement story—indeed, the story of the whole of hospital medicine—is our need to fundamentally improve outcomes through systems improvements. The true benefit of comanagement is not in one doctor (hospitalist) taking over the medical care of another doctor (surgeon). That will only slightly improve outcomes of the medical issues at which the hospitalist is more expert (e.g. minor electrolyte disorders). Meanwhile, this model continues to allow the same harms that the underlying unsafe hospital system imparts. The comanagement model itself won’t fix this. Rather, the model simply acts as a mechanism for us to accomplish our desired goals of system redesign.

Put another way, I am better at internal-medicine care than a neurosurgeon is. As such, I have no doubt that if I manage the medical issues of neurosurgical patients, I will do it better. However, this system of hospitalist provision of internal-medicine care can ultimately only lead to the type of marginal, not meaningful, improvements these comanagement studies have shown.

The real potential for the comanagement model comes when I take off my internal-medicine hat (diabetes care, electrolyte management, etc.) and put on my HM hat (ability to execute systematic quality and process improvements that result in safer systems that effect ALL patients, ALL the time, and is not dependent on the individual provider to do the right thing).

In doing this, the MacGuffin—the comanagement model that cohorts a lot of patients in the hands of a relatively few hospitalists—affords us the opportunity to truly advance the patient-safety plot by building better systems, the type of systems that ensure that every patient systematically gets appropriate VTE prophylaxis, avoids medication errors, has unnecessary urinary and central venous catheters removed, avoids pressure ulcers, doesn’t fall or get delirious, and has expert transitions of care. I have no doubt that if we achieved these kinds of interventions, rather than just managing patients’ medical issues, we’d see the kind of profound changes the comanagement model can offer.

MacGuffin or not, comanagement is likely here to stay. The challenge, then, is to find a way in which these care arrangements can go beyond scut to systematically and comprehensively improve the flawed systems of care that envelop our surgical patients.

Doing this will vastly improve patient outcomes, add significant value to the care we provide, and clearly signal to the surgeon that it’s time to bring us the water bottle. TH

Dr. Glasheen is physician editor of The Hospitalist. He is associate professor of medicine at the University of Colorado Denver, where he serves as director of the Hospital Medicine Program and the Hospitalist Training Program, and as associate program director of the Internal Medicine Residency Program.


  1. Auerbach AD, Wachter RM, Cheng HQ, et al. Comanagement of surgical patients between neurosurgeons and hospitalists. Arch Intern Med. 2010;170(22):2004-2010.
  2. Huddleston JM, Long KH, Naessens JM, et al. Medical and surgical comanagement after elective hip and knee arthroplasty: a randomized, controlled trial. Ann Intern Med. 2004;141(1):28-38.

Next Article:

   Comments ()