HM@15 - Are You Living Up to High Expectations of Efficiency?


In 2002, a summary article in the Journal of the American Medical Association helped put the relatively small but rapidly growing HM profession on the map. Reviewing the available data, Robert Wachter, MD, MHM, and Lee Goldman, MD, MPH, of the University of California at San Francisco (UCSF) concluded that implementing a hospitalist program yielded an average savings of 13.4% in hospital costs and a 16.6% reduction in the length of stay (LOS).1

A decade later, the idea of efficiency has become so intertwined with hospitalists that SHM has included the concept in its definition of a profession that now comprises more than 30,000 doctors, nurses, and other care providers. HM practitioners work to enhance hospital and healthcare performance, in part, through “efficient use of hospital and healthcare resources,” according to SHM.

The growth of any profession can create exceptions and outliers, and observers point out that HM programs have become as varied as the hospitals in which they reside, complicating any attempt at broad generalizations. As a core part of the job description, though, efficiency and its implied benefit on costs have been widely promoted as arguments for expanding HM’s reach.

So are hospitalists meeting the lofty expectations?

A Look at the Evidence

A large retrospective study that examined outcomes of care for nearly 77,000 patients in 45 hospitals found that those cared for by hospitalists had a “modestly shorter” stay (by 0.4 days) in the hospital than those cared for by either general internists or family physicians.2 Hospitalists saved about $270 per hospitalization compared with general internists but only about $125 per stay compared with family physicians, the latter of which was not deemed statistically significant.

Reversing Hospital Medicine’s Gains? The Effects of Care Fragmentation

Just as well-coordinated hospitalist programs can help reduce inpatient length of stay, other studies have found that the gains can be easily reversed by models that lead to more fragmentation of care.

A recent study by Kenneth Epstein, MD, MBA, FHM, FACP, found that such fragmentation, defined as the percentage of care provided by hospitalists other than the one who sees a specific patient the most, can significantly prolong LOS. The study linked a 10% increase in fragmentation to an increase of 0.39 days and 0.30 days in the LOS among pneumonia patients and heart failure patients, respectively.12

Arranging physician schedules to increase continuity, he says, can help avoid excessive fragmentation, as can ensuring efficient communication so that providers coming onto service understand the previous provider’s plan. Although Dr. Epstein’s study looked specifically at a hospitalist model, he stresses that other models could have as many, if not more, discontinuities of care.

“Hospitalist groups are more structured and aligned to try to resolve these issues—trying to reduce fragmentation of care and length of stay—whereas nonhospitalist groups may have as much fragmentation but may not be as motivated to try to resolve these issues,” he says.

Hospitalists are, in fact, leading efforts to address fragmentation-of-care concerns. At the 335-bed Johns Hopkins Bayview Medical Center in Baltimore, the hospitalist-instituted CICLE (Creating Incentives and Continuity Leading to Efficiency) model led to a 13% decline in unique providers and a 16% decrease in LOS, to 3.60 days from 4.31.13

Even more significantly, the average total hospital charge per admitted patient dropped roughly 20%, to about $8,000 from nearly $10,000. The study’s authors documented the most significant savings in medications, laboratory charges, and radiology charges


A more recent review of 33 studies found general agreement that hospitalist care led to reduced costs and length of stay but revealed less uniformity in the impacts on quality and patient outcomes.3

A more dramatic—albeit smaller—affirmation of HM as an efficient force has come from a study of patients admitted to 200-bed Olive View-UCLA Medical Center in Sylmar, Calif. The study, led by assistant medical director Scott Lundberg, MD, concluded that the arrival of an academic hospitalist program led to a one-year increase of $2.3 million in reimbursements from Medi-Cal, California’s Medicaid program.4

“Most other places that have demonstrated the cost-effectiveness of hospitalists generally point to reducing length of stay, which therefore reduces the costs,” Dr. Lundberg says. Under Medicare’s diagnosis-based reimbursement (DRG) system, hospitals could get paid the same amount whether the patient stays one day or five.

Medi-Cal, however, uses a straight-up per diem reimbursement system. “So reducing someone’s length of stay is not necessarily desirable if Medi-Cal would have paid you for all of those days,” Dr. Lundberg says. The state’s Medicare program also can deny coverage for days deemed medically unnecessary after a review of patient charts.

Hospitalists, he says, helped boost revenue in two ways. First, the program helped the hospital avoid denied coverage days by ensuring that patients stayed only as long as necessary. Average LOS, in fact, dropped to 1.92 days from 2.48 days, decreasing the Medi-Cal denial rate to 31.8% (from 43.8%) and bumping up the average reimbursement per inpatient day to $955 from $787.

Hospitalists also helped alleviate the work-hour limits for residents imposed by the Accreditation Council for Graduate Medical Education (ACGME), which had effectively capped the number of inpatients the center could admit. Because Olive View-UCLA receives per diem payments from Medi-Cal, making room to accept more patients into the hospital has meant increased revenues. Among the other benefits, the program has improved patient satisfaction and relieved some of the pressure on teaching teams.

With $310,000 for salary outlay in the hospitalist program’s first year, the study found a net cost benefit of $2 million. “One of the real challenges in getting this hospitalist thing going was getting our administrators to shell out the money for the salaries,” Dr. Lundberg says. The study demonstrated that a hospitalist program not only pays for itself, but also can substantially ramp up revenue. “I’m guessing that others, especially at public hospitals, face the same challenges,” he says. “I’m hoping they can point to this analysis and say, ‘Look, here’s what L.A. County did. They were able to show a net increase in revenue from this hospitalist service.’ ”

On the opposite side of the country, hospitalists are pointing to a success story in pediatric care. At the 120-bed Children’s Hospital at Montefiore at Albert Einstein College of Medicine in the Bronx, N.Y., a recent study concluded that establishing a pediatric HM program led to a significant reduction in LOS for patients with asthma or bronchiolitis.5 Nora Esteban-Cruciani, MD, MS, assistant director of pediatric hospital medicine and lead author of the report, which was presented at HM11, says it’s the first study to demonstrate such an effect for asthma in an inner-city academic setting.

Compared to a traditional resident-attending team, care administered by a resident-physician’s assistant-hospitalist team reduced LOS for bronchiolitis by 15.5% and asthma by 11.8%. With the 378 hospital-bed days saved annually, Children’s Hospital at Montefiore achieved an estimated savings of about $944,000 before taking salaries into account. “We anticipate seeing similar benefits in other groups of patients, and the total savings will far exceed the hospitalist salaries,” Dr. Esteban-Cruciani says.

After the pediatric HM program launched, her study also documented a 17% to 25% decrease in rehospitalizations among asthmatic children at four, six, and 12 months after their initial hospital discharge. As a result of the demonstrated value, Dr. Esteban-Cruciani says, the children’s hospital is expanding its HM program and hiring another 4.5 full-time equivalents.

So how did hospitalists achieve the positive results?

Listen to Dr. Meltzer discuss the need for experienced hospitalists.

“Knowing the most up-to-date and evidence-based treatment plans, understanding how to use the hospital systems in the most efficient manner, being on the ward for eight to 12 hours per day to respond to issues that arise, as well as 24-hour availability by phone for the residents,” she says. “The day-to-day continuity, as well as the ability to consistently improve systems of care, are distinctive advantages to hospital medicine.”

The case for HM as a model of efficiency comes with a major caveat, however. David Meltzer, MD, PhD, FHM, chief of the section of hospital medicine and an economist and public-policy expert at the University of Chicago, points out that healthcare costs don’t end with a patient’s hospital discharge. Could savings achieved during inpatient care be offset by greater costs afterward?

A new study in the Annals of Internal Medicine by researchers at the University of Texas Medical Branch in Galveston has sharpened that question with the suggestion that, at least in some cases, hospitalist-procured savings might not last.6 When compared to care delivered by primary-care physicians (PCPs), the researchers found that hospitalist care yielded an average inpatient savings of $282 per Medicare beneficiary. But that reduction was wiped out by an extra $332 average cost in the month after discharge, due to higher readmissions, more emergency department visits, and more patients sent to nursing facilities instead of to their own homes. An accompanying editorial raises the uncomfortable question: “Are hospitalists discharging their patients more quickly but less appropriately, such that some of their patients bounce back?”7

A program that is structured in such a way as to hire or retain experienced hospitalists is likely to have a higher cost savings than one that doesn’t.

—David Meltzer, MD, PhD, FHM, chief, section of hospital medicine, economist, University of Chicago

The study itself has its own share of caveats: Data were collected only until 2006, before reducing 30-day readmissions became a widespread focal point. The editorial also highlights the possibility that hospitalists might care for patients whose weaker relationships with outpatient providers could be the true driver of increased readmissions. In a statement, SHM President Joe Li, MD, SFHM, adds that constructive talks about healthcare costs must include the notion of quality, something the organization has worked to improve with interventions like Project BOOST.

At the very least, the new research highlights the importance of context when considering HM impacts on cost and quality. Separate studies, meanwhile, suggest that the jury is still out on whether other hospitalist-led models can consistently improve outcomes and costs. At academic centers, for instance, work-hour limits for medical residents have provided a strong impetus for joint-care arrangements, such as comanagement systems. A 2004 study found that an orthopedics-hospitalist comanagement structure led to a modest reduction in complications after elective hip and knee surgery. But the report documented no difference in costs or actual length of stay.8

Listen to Dr. Meltzer discuss the need for experienced hospitalists.

click for large version

More recently, a study of nearly 7,600 patients at UCSF Medical Center found that an HM-neurosurgery comanagement model had no significant impact on the center’s patient mortality, readmissions, LOS, or patient satisfaction. The comanagement system, however, yielded an average savings of $1,439 per hospitalization and boosted physicians’ perceptions of quality and safety.9

Andrew Auerbach, MD, MPH, SFHM, associate professor of medicine at UCSF Medical Center, says the savings, while not dramatic, nevertheless can add up when applied to the thousands of patients seen by the service every year. “That’s compelling because I think one of the things that you’re arguing when you’re doing these services is what the return on investment is going to be,” he says. “Traditionally, these have been implemented without any specific financial return on investment being applied, but the large expectation that clinical improvement is going to happen.”

His study at UCSF found just the opposite: no clinical improvement but a net cost benefit. “We were a little disappointed in some ways, but in other ways not surprised because there are very few data out in the community that suggest comanagement improves any outcomes,” Dr. Auerbach says. Among complicated neurosurgery patients, the strongest determinants of outcome might be beyond the scope of hospitalist-aided medical care.

With hospitals nervously eyeing their bottom lines, however, any financial improvement that does not adversely affect quality can still be seen as a positive development, and Dr. Auerbach says his study was the first to demonstrate that benefit. At UCSF Medical Center, at least, comanagement has proven compelling enough to spur plans for extending the service to orthopedic surgery patients.

Regardless of the care model, other studies suggest that specific interventions at key moments can yield substantial savings. A small, randomized controlled study led by hospitalists at Johns Hopkins University in Baltimore, for example, supports the idea that “simply showing providers the cost of some diagnostic tests at the time of order entry can affect behavior.”10 Although the study didn’t focus exclusively on hospitalists, experts say they’re in the best position to take the lead in curbing unnecessary costs.

“Hospitalists, I think, have a better understanding of the impact of resource utilization on the total cost of care and can be more prudent in the use of technologies,” says Kenneth Epstein, MD, MBA, FHM, FACP, chief medical officer for Traverse City, Mich.-based Hospitalist Consultants Inc. One reason is that hospitalists aren’t beholden to any specific technology, whether endoscopies or cardiac catheterization.

I’m hoping they can point to this analysis and say, “Look, here’s what L.A. County did. They were able to show a net increase in revenue from this hospitalist service.”

—Scott Lundberg, MD, assistant medical director, Olive View-UCLA Medical Center, Sylmar, Calif.

Mark Graban, author of the book “Lean Hospitals: Improving Quality, Patient Safety, and Employee Satisfaction,” says hospitalists can play another critical role in controlling costs by mapping out and simplifying the discharge processes. He recalls how hospitalists helped coordinate the effort by one of his hospital clients to prevent discharge delays that would have unnecessarily kept patients in the hospital for an additional night or two.

“That length-of-stay reduction, especially in a fixed-reimbursement setting, can have a huge financial impact,” Graban says. “And, inarguably, it’s the right thing to do for the patient, because it’s patients that are medically ready to be discharged. It gets them home and it reduces their increased risk of picking up infections or being involved in hospital errors.”

Focusing on patient safety could translate into big cost savings under the new Medicare system that penalizes providers for certain hospital-acquired conditions, such as skin ulcers and urinary tract infections, Dr. Epstein says. “There’s an emphasis by hospitalists in understanding the system and being willing to put energy into things like documenting ‘present on admission,’ which then has a huge impact on the hospital,” he says. Close monitoring of patients and developing standardization of care can likewise minimize the risk of conditions, such as catheter-associated infections, from cropping up in the hospital.

Dr. Meltzer says his own research suggests that experienced hospitalists are most effective at controlling costs. “So a program that is structured in such a way as to hire or retain experienced hospitalists is likely to have a higher cost savings than one that doesn’t,” he says.

Listen to Scott Lundberg talk about the effectiveness of hospitalists in recouping reimbursement.

In a broader sense, the maturation of the HM model and more widespread adoption of effective methods by practitioners might be boosting the overall impact of hospitalist care. A study that examined nearly 2 million Medicare admissions over six years found that the effects of the hospitalist care model on LOS became progressively more pronounced over time, from an average reduction of only 0.02 inpatient days in 2001-2002 to a decrease of 0.35 days by 2005-2006.11

Interestingly, the study’s authors suggest that effects attributable to hospitalists were most pronounced among older, complicated, nonsurgical patients cared for at nonprofit community hospitals.

The Verdict

Despite the variable design and scope of individual programs, experts say, HM’s overall net positive on the efficiency of inpatient care is fairly well documented. Future considerations of hospitalists’ true effects on costs, however, will demand an accounting of healthcare across an entire system, where the HM impact is decidedly less certain. “The right comparison in some sense is, What are the total costs of care for a patient cared for in a system that uses hospitalists versus the totals costs of similar patients cared for in a system that doesn’t use hospitalists?” Dr. Meltzer says.

David Mitchell, MD, PhD, a hospitalist at Sibley Memorial Hospital in Washington, D.C., and a member of SHM’s Performance Standards Committee, is among those with an additional concern: Providers may not be taking full advantage of their position to control costs.

“The reason is primarily that the reimbursement structure is not set up to incentivize us to cut costs,” he says. Dr. Mitchell, who has worked in 12 hospitals in six states, argues that hospitalists still are too detached from the true price of ordered tests. “That’s what I fear in hospital medicine, that we just become robots: chest pain means CT scan without thinking,” he says. “This just doesn’t make sense.” Dr. Mitchell also contends that the focus of some HM programs on seeing as many patients as possible to maximize reimbursements is leading to less efficiency. At HM11 in May, he met another hospitalist who said he regularly saw 40 to 45 patients every day. “I know there’s absolutely no way you can see that many patients and do an efficient job,” Dr. Mitchell says.

If one of the clearest areas of success for hospitalists has been in reducing length of stay within a hospital, experts acknowledge that it may no longer be enough. “In the new payment model, success is going to be defined differently, and it will be in terms of reducing the total cost of care,” Dr. Meltzer says.

Over the next decade, hospitalists will need to respond to new set of incentives. “And I think one of the really interesting questions will be how hospitalists can best do that, and the extent to which it causes them to rethink the ways in which they organize their practice,” he says.

Bryn Nelson is a freelance medical writer based in Seattle.


  1. Wachter RM, Goldman L. The hospitalist movement 5 years later. JAMA. 2002;287(4):487-494.
  2. Lindenauer PK, Rothberg MB, Pekow PS, Kenwood C, Benjamin EM, Auerbach AD. Outcomes of care by hospitalists, general internists, and family physicians. N Engl J Med. 2007;357:2589-2600.
  3. Peterson MC. A systematic review of outcomes and quality measures in adult patients cared for by hospitalists vs nonhospitalists. Mayo Clin Proc. 2009;84(3): 248-254.
  4. Lundberg S, Balingit P, Wali S, Cope D. Cost-effectiveness of a hospitalist service in a public teaching hospital. Acad Med. 2010;85(8):1312-1315.
  5. Esteban-Cruciani N, Montejo J, Azzarone G, Douglas L, et al. Impact of a pediatric hospital medicine program on resource utilization for children with respiratory disorders. J Hosp Med. 2011;6(4)Supp 2:S27.
  6. Kuo Y-F, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med. 2011;155(3): 152-159.
  7. Chen LM, Saint S. Moments in time. Ann Intern Med. 2011;155(3):194-195.
  8. 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.
  9. Auerbach AD, Wachter RM, Cheng HQ, et al. Comanagement of surgical patients between neurosurgeons and hospitalists. Arch Intern Med. 2010;170(22): 2004-2010.
  10. Feldman L, Thiemann D, Brotman D. Financial impact of presenting lab cost data to providers at the time of order entry: a randomized controlled clinical trial. J Hosp Med. 2011;6(4)Supp 2:S93.
  11. Kuo Y-F, Goodwin JS. Effect of hospitalists on length of stay in the Medicare population: variation according to hospital and patient characteristics. J Am Geriatr Soc. 2010;58:1649-1657.
  12. Epstein K, Juarez E, Epstein A, Loya K, Singer A. The impact of fragmentation of hospitalist care on length of stay. J Hosp Med. 2010;5(6):335-8.
  13. Chandra S, Howell E, Wright S. CICLE: Creating incentives and continuity leading to efficiency. J Hosp Med. 2011;6(4)Supp 2:S17

Recommended Reading

ONLINE EXCLUSIVE: The “Weak Link” in Patient Handoffs
The Hospitalist
A Critical First Step
The Hospitalist
SQUINT Is Looking Out For You
The Hospitalist
Are You Delivering on the Promise of Higher Quality?
The Hospitalist
It Takes a Village
The Hospitalist
Discharge improvement
The Hospitalist
Marriage of Necessity
The Hospitalist
POLICY CORNER: SHM Pledges Support to Patient-Safety Initiative
The Hospitalist
HM=Improved Patient Care
The Hospitalist
Under the Microscope: Medication Reconciliation
The Hospitalist
   Comments ()