By Christopher Guadagnino, Ph.D.

Rusty Holman, M.D.

Published February 2008

The hospitalist model of inpatient care – well into its second decade of existence – has spread so rapidly that questions about its viability have long since given way to examination of its important clinical and financial implications, as well as its impact on the medical profession.

Hospitalists are distinct from most other physicians in that they do not have an office-based practice but instead practice full-time within an institutional setting, serving as the physician of record after admitting patients to the hospital from the emergency room or from a primary care physician (PCP) or subspecialist, then returning patients to their PCPs at the time of hospital discharge. Capable of managing most medical problems encountered in the hospital setting, some hospitalists also perform general medical procedures such as central venous catheterization, arterial puncture, lumbar puncture and endotracheal intubation. As their clinical role evolves, hospitalists are increasingly teaming with subspecialists such as intensivists to care for frail, medically complicated patients.

The hospitalist model has gained widespread acceptance among U.S. hospitals, driven by the underlying principle that frequent follow-up care by a small number of inpatient-based physicians – nimbly weaving patients through the hospital maze of tests and services – can lead to more prompt and efficient therapy, potentially decreasing the number of adverse outcomes and the length of stay.

But research remains surprisingly inconclusive as to whether hospitalists have improved the efficiency and quality of patient care, or lowered costs; and concerns persist among some physicians that hospitalists may erode continuity of care, the physician-patient relationship, the inpatient skill set of PCPs, and – ironically – the already scarce supply of office-based general internists who are attracted to the hospitalist profession.

Economic and quality impacts of hospitalists received national attention once again in December, when the New England Journal of Medicine published a wide-scale literature review of studies focusing on whether the use of hospitalists saves hospitals time and money, and improves quality. The review followed nearly 77,000 patients admitted to 45 hospitals between September 2002 and June 2005, and concluded that despite a 12 percent – or a half-day – reduction off the average four-day hospital stay, the use of a hospitalist offered only modest savings compared with care by a general internist, and no significant savings over care provided by a family doctor, while inpatient death rates and 14-day readmission rates were no different for patients of hospitalists and PCPs. Researchers speculated that hospitalists may simply do the same amount of work, but in less time, or they may order more tests than regular doctors since the hospitalists are not intimately familiar with patients’ histories.

While the promise of significant quality enhancement and cost savings of the hospitalist model has yet to be fulfilled conclusively, hospitalists have nevertheless become an integral component of the health care delivery system, and attention has shifted to how that reality should be managed with efficacy – with particular emphasis on the importance of effective hospitalist-PCP communication to mitigate the inherent discontinuity during patient handoffs between PCPs and hospitalists.

Advocates note that hospitalists do not erode quality of care, while bringing at least modest efficiency gains, while one of their greatest benefits is allowing PCPs to increase productivity in their office-based practices by freeing them from the burdens of hospital rounding and on-call duties.

The Society of Hospital Medicine (SHM) believes the hospitalist profession is capable of demonstrating greater cost and quality benefits once it has the guidance of a uniform set of consensus guidelines to reduce variation, optimize the model’s efficiencies, and ensure effective communication during patient handoffs. SHM, in collaboration with other organizations, expects to publish a set of such guidelines in a few months, according to Rusty Holman, M.D., president of SHM’s board of directors.

Hospital medicine "is the most rapidly growing profession in medicine" and hospitalists are now present in 70 percent of U.S. hospitals, where they care for two-thirds of hospitalized patients, on average, according to Hospitalists and the Changing Hospital Environment, a report released last June by the American Medical Association Organized Medical Staff Section (AMA-OMSS). Growing from fewer than 1,000 hospitalists in the mid 1990s to nearly 13,000 in 2004, the profession projects as many as 30,000 hospitalists to be in practice by 2010, according to 2005 survey data from the American Hospital Association. About 75 percent of hospitalists are general internists, and another 15 percent are internal medicine subspecialists or pediatricians.

A product of the managed care industry, the hospitalist movement has largely been driven by economic imperatives, according to the SHM. Medicare’s prospective payment system created incentives for hospitals to increase patient turn over, control costs and increase efficiency, while medically uncomplicated patients are increasingly shifted to the ambulatory setting, leaving hospitalized patients who are sicker and require a much more hands-on and intensive model of care – which PCPs are less comfortable providing. Restrictions on work hours of residents have further intensified the need for inpatient care professionals. Reimbursement pressures on PCPs have raised the opportunity cost of time taken away from office productivity to do hospital rounding for what might be a small number of inpatients.

Hospitalists are touted as being able to address these imperatives, as well as bring other clinical improvements to hospital care, as outlined by the American Medical Association’s Council on Medical Service:

· Improved quality of care and clinical outcomes in the inpatient setting due to the increased expertise and experience of hospitalists, particularly with respect to severely ill patients.

· Improved efficiency in the inpatient setting because the hospitalist is available throughout the entire day to see hospitalized patients and to assess potential admissions from the emergency room; and in the outpatient setting because the practice of the office-based physician is not interrupted by inpatient rounds and mid-day emergencies with hospitalized patients, and time is not wasted traveling to and from the hospital.

· Enhanced care as an "inpatient safety net" for indigent patients who were previously admitted as "no doctor" patients, and for whom hospitals are finding it increasingly difficult to find physicians willing to provide this coverage.

· Enhanced accountability and investment in the hospital quality improvement process due to the hospitalist being located in the hospital for a considerable portion of each day.

· Enhanced educational and training opportunities by teaming residents and medical students with experienced hospitalists.

The Joint Commission puts its confidence in hospitalists’ ability to improve patient safety. In its July 2006 Perspectives on Patient Safety, the agency states that hospitalists’ ongoing presence allows them to respond to patients in real-time and prescribe treatment, order additional tests or arrange consultations with specialists much more quickly than having to locate a patient’s regular physician. The agency also notes that hospitalists are familiar with the use of hospital protocols, guidelines and communication tools that affect the safety of care delivery, and have the opportunity to develop close working relationships with staff from all departments, potentially enhancing cooperation and collaboration. The SHM states that hospitalists, many of whom participate on quality improvement teams and safety committees, have unique insights into how best to deploy and improve health information systems within their hospitals because they are familiar with systemic quality and efficiency improvement efforts.

The hospitalist profession could "dramatically change what it means to be an American PCP," the AMA-OMSS report adds, as inpatient and outpatient care duties become increasingly more distinct and hospital medicine continues to attract PCPs as a career option, with academic medical centers developing residency and fellowship programs in hospital medicine.

There were six hospitalist employment models in 2005-2006, the report notes: hospital or hospital corporation (34 percent), academic institution (20 percent), multi-state hospitalist-only group or management company (19 percent), multi-specialty/primary care medical group (14 percent), local hospitalist-only group (12 percent), and emergency or critical care physician medical group (2 percent).

Hospitalists’ median total compensation package in 2005 was $168,000, while hospitalist incomes vary widely but are typically similar or slightly higher than a doctor with the same training working in a traditional (inpatient and outpatient) practice in the same market, writes SHM Co-President John R. Nelson, M.D., on the career counseling page of the American College of Physicians’ (ACP) website. Because hospitalist practice is a relatively new phenomenon, Nelson adds, there may be considerable negotiating room for a new hospitalist practice when discussing compensation.

Evidence that hospitalists improve quality of care remains elusive, communication between hospital-based and primary care physicians is notoriously deficient, and critics continue to question whether hospitalists threaten continuity of care and the doctor-patient relationship. The hospitalist profession is mindful of these challenges and is actively working on meeting them.

Robert M. Wachter, M.D., and co-author Lee Goldman, M.D., MPH – advocates of the hospitalist model who are credited with having coined the very term "hospitalist" in 1996 – noted in a Jan. 2002 article in the Journal of the American Medical Association that most studies of hospitalists had found no change in quality measures relative to inpatient care by a PCP. After analyzing 19 published cost and quality studies, Wachter and Goldman concluded that hospitalist programs did reduce hospital costs by an average of 13.4 percent and decreased average length of stay by 16.6 percent, while findings related to improved outcomes were inconsistent.

The Dec. 2007 NEJM literature review of hospitalist impacts reached essentially the same conclusion regarding neutral impact on quality and only modest impact on cost, noting that inpatient death rates and 14-day readmission rates were no different for patients of hospitalists and PCPs, while hospitalists achieved small average cost savings per hospitalization: $268 lower compared with a patient cared for by a general internist, and $125 lower compared with care by a family physician.

"The benefits didn’t come at the expense of increased mortality or readmissions," says Holman. No uniform system of care is yet in place that is designed to reduce variation in hospitalist practice across the country, and yield measurably improved results, Holman explains.

While research findings about hospitalists’ impact on quality of care appear to be inconclusive, adding another physician to a patient’s hospital care inevitably introduces potential confusion and disrupts continuity of care, believes Jim King, M.D., president of the American Academy of Family Physicians (AAFP). Most of the time, the PCP is out of the inpatient care loop completely when a hospitalist is involved, and is usually not consulted when inpatient decisions are made, he says. "Medication can be changed by a hospitalist without my knowledge, and work-ups can be ordered that have failed with my patient in the past," notes King. Handoff communication is also spotty. "Some hospitalists have contacted me after discharging my patients, but more often, they don’t," says King. Patients also have more trust in physicians they have known for a long time, compared with those they have just met, he adds: "My hospitalized patients are usually very positive about me being there as their advocate, instead of a complete stranger."

Deficits in communication and information transfer between hospital-based and primary care physicians are "substantial and ubiquitous," while delays and omissions are consistently large, and traditional methods of completing and delivering discharge summaries are "suboptimal for communicating timely, accurate, and medically important data to the physicians who will be responsible for follow-up care," according to a review of medical literature published in JAMA last February. Before the advent of hospitalists, the same physician often provided inpatient and outpatient care and there was little need for information transfer via discharge summaries, the JAMA study authors note. In the new hospitalist model of care, however, the discharge summary becomes a vital tool for communication and information transfer, but the current Joint Commission performance standard that discharge summaries be completed within 30 days of hospitalization falls far short of the needs of PCPs and patients, they add.

To ensure consistent and reliable information exchange, the JAMA researchers recommend that hospitalists send a summary document to PCPs on the day of discharge that includes, at minimum, the diagnoses, discharge medications, results of procedures, follow-up needs, and pending test results. A more complete discharge summary should contain more detailed information, the authors write, including:

· Primary and secondary diagnoses.

· Pertinent medical history and physical findings.

· Dates of hospitalization, treatment provided, brief hospital course.

· Results of procedures and abnormal laboratory test results.

· Recommendations of any subspecialty consultants.

· Information given to the patient and family.

· The patient’s condition or functional status at discharge.

· Reconciled discharge medication regimen, with reasons for any changes and indications for newly prescribed medications.

· Details of follow-up arrangements made.

· Specific follow-up needs, including appointments or procedures to be scheduled, and tests pending at discharge.

· Name and contact information of the responsible hospital physician.

Communication other than discharge summaries also must be improved, the researchers maintain, as many PCPs are not routinely notified about patient admissions or complications during the hospital course; while some PCPs may fail to provide sufficient information to hospitalists at admission, fail to visit or call hospitalized patients, or fail to participate in discharge planning. For patients with chronic illnesses and frequent hospitalization, those deficits are multiplied, and completeness of information handoffs is particularly important.

To minimize "information voltage drop" during patient handoffs, effective hospitalist programs have such information transfer protocols, says Holman. Most hospitalist programs have the explicit expectation that the hospitalist informs a patient’s PCP within 24 hours of admission and discharge, and have some standards for the content of the communication, believes John Bulger, D.O., chair of the Task Force on Hospitalist Medicine for the American College of Osteopathic Internists, and head of the hospitalist program at Geisinger Medical Center in Danville, Pa.

Ethical concerns are also raised by patient handoffs between PCPs and hospitalists, inasmuch as a lack of pre-existing relationship between a hospitalist and a patient represents a communication barrier that inhibit patients from expressing their values and medical care preferences at a time when they are sickest, write Jeremy Snyder and Brian C. Zanoni, M.D., in the Feb. 2006 Virtual Mentor, the AMA’s ethics journal. PCPs have greater familiarity with the values of their patients and families, an intimacy that allows the physician to have greater knowledge of the patient’s attitudes toward risk and willingness to engage in important health treatment decisions, particularly major surgery or end-of-life issues – and such knowledge takes on added significance during a hospitalization, they note.

"The mission of hospitalists is to facilitate informed choices of patients and families regarding their medical care and not merely to execute the [hospitalist’s] own medical judgement effectively and efficiently. Hospitalists must weigh families’ personal values with objective data regarding prognosis, risk, and benefit," which can be time-consuming, often entailing family meetings for which physicians are not reimbursed, Snyder and Zanoni write.

"My experience has been the opposite – that having a different mind and set of eyes to apply to a patient’s care can be a catalyst to conversation that would not have occurred in the outpatient setting," says Holman. A good hospitalist will establish a relationship of trust and will put a patient at ease by introducing themselves in association with the patient’s PCP, e.g., "I’m Dr. Holman, and I work in close association with your primary care doctor. I’ll be in close communication with him," says Holman. The hospitalist should ask the patient what are their goals for care, which Holman says is sometimes not even asked in the outpatient setting, unless a patient has been with the same PCP over many years.

To reduce discontinuity of care, a PCP can still be involved in the care of their patients under the hospitalist model through visits or phone calls with patients and through better communication with hospitalists – the amount of which varies widely, ranging from "call me when they’re done," to the periodic exchange of progress notes, according to Sanford Melzer, M.D., MBA, senior vice president of strategic planning and business development at Seattle Children’s Hospital, and a member of the American Academy of Pediatrics’ committee on hospital care. The PCP should clearly communicate his or her expectations with the hospitalist at the time of admission. "For a successful partnership, the PCP should tell the hospitalist how often they want to hear from them, and the hospitalist must follow through," says Melzer. At a minimum, he adds, "A community physician should know what’s going on at discharge, or during a major change in the patient’s status – a life-threatening diagnosis, a positive lab test result, movement to the ICU, preparation for an operation."

When a PCP’s patient is under the care of a hospitalist, the PCP’s role is primarily to be a consultant for the hospitalist, and much of their communication is to prevent the PCP from being uninformed when they get a call from the patient’s family, according to Bulger, who says that consults with a PCP during a patient’s hospitalization happen occasionally, rather than routinely. PCPs who use hospitalists generally don’t do hospital rounds, although some do courtesy and social visits to their patients, albeit rarely, he adds.

A consensus statement clarifying the roles and responsibilities of PCPs and hospitalists can address many of these issues, says Holman, and one is currently in draft form, with a final version expected to be published in a few months. Developed by the SHM, working with the Joint Commission, the National Quality Foundation (NQF), the American College of Physicians, and other groups, the consensus statement will include measurable standards for transitions of care, including what information needs to be available to a hospitalist at the time of admission; what information needs to be communicated back to the PCP at the time of patient discharge – and how soon; what level of patient education is required at the time of discharge – including guidance on who is responsible for following up on post-discharge instructions and pending lab results, and in what manner; and guidance on how to address preventable readmissions – including timely surveillance of patient progress after discharge, compliance with medication, and access to outpatient services.

The hospitalist consensus guidelines will be vetted by the NQF for public reporting and pay-for-performance initiatives, says Holman, who believes that the guidelines may help empower the hospitalist model’s potential to reduce discontinuity between PCPs and hospital-based physicians through better coordination of inpatient and outpatient care, says Holman.

In that role, hospitalists can even augment the patient-centered medical home model, believes Melzer. As most patients are not hospitalized that often, the PCP is still the anchor of care, while hospitalists can be effectively integrated into the medical home model by timely and careful handling of information exchange during patient handoffs, aided by the guidance of national standards, he says.

A cluster of concerns remains regarding the impacts of the hospitalist model on the medical profession. One is that the hospitalist movement may cause erosion of PCPs’ hospital-based skills and judgment in the course of a disease. While some groups of family physicians choose not to round in hospitals and welcome fewer on-call responsibilities, King personally believes that PCPs should continue to round. "You’ve got to get rusty if you choose not to round in the hospital for some time," he says. The AAFP cautions that family physicians should strongly consider the mid- and long-range implications for their practices before they relinquish hospital privileges. Such implications may include difficulty being credentialed and/or reimbursed by managed care companies for services and procedures in the ambulatory setting if one does not have hospital privileges for those same services and procedures; as well as the very real possibility of being unable to successfully reapply for hospital privileges at future points of career transition, without the necessity of seeking substantial additional education and retraining.

Even though the use of hospitalists remains largely voluntary, a PCP who wants to continue doing hospital medicine may find it unavailable, says Melzer. In some group practices, for example, a PCP may not be able to find on-call coverage for rounding by practice partners who may prefer to use hospitalists, he notes. That is also the case at closed-staff hospitals such as Geisinger Medical Center, which only uses hospitalists to admit patients, notes Bulger.

Hospitals are starting to limit hospital privileges of physicians who fall below a certain patient encounter volume – a trend which may widen the division between inpatient and outpatient physicians, says Richard Neubauer, M.D., a member of the ACP’s Board of Regents. If they become completely dissociated from hospitals, office-based physicians may lose much of their ability to interact with colleagues and become completely isolated, notes Neubauer, who spent 27 years in private practice as a general internist in Anchorage, Alaska, and now represents a group of hospitalists as Chief of Medicine at Alaska Native Medical Center.

That potential loss is counterbalanced by the opportunity for PCPs to make their office-based practice schedule more predictable and productive. A large cadre of Geisinger’s general internists did not like the interruption to their day of having to round at the hospital, and the hospitalist program – in place since 2000 – has given them peace of mind that someone is taking care of their patients, notes Bulger.

PCPs’ attitudes toward hospitalists appears to be generally positive, as published PCP survey studies show high levels of physician satisfaction several years after implementation of a hospitalist program. A 2002 survey of its membership by the American Academy of Pediatrics found that pediatricians’ satisfaction with the level of care provided by hospitalists was high, and there was general agreement that use of hospitalists increases the manageability and productivity of office practice. Of the respondents who used pediatric hospitalists, over half said they transferred care of hospitalized patients because attending inpatients takes too much time away from office practice, while one-third said they did so because they prefer concentrating on ambulatory pediatrics.

The relatively low professional fee component for nonprocedural inpatient care has made it easier for PCPs to forgo management of their hospitalized patients, write Wachter and Goldman, who cite one analysis showing that the average PCP would gain about $40,000 a year simply by replacing hospital care – and its wasted commute time – with increased productivity in the outpatient setting.

Finally, the hospitalist movement has physician workforce implications. "The biggest problem with the hospitalist movement," says Neubauer, "is that few to none who are coming out of internal medicine training are interested in strictly outpatient medicine, with its huge amount of clerical management problems and pathetic reimbursement." Retirement is depleting the supply of community-based general internists, which "is further diluted by the allure of the hospitalist profession, with its shift work and flexible lifestyle. We’re seeing the potential demise of outpatient general internal medicine and we haven’t figured out how to balance the physician workforce needs," says Neubauer.