Over the past several years, the traditional American system of care for hospitalized adult patients has undergone a revolutionary change. The traditional system, in which primary care physicians (generally internists or family physicians) managed their own patients in the hospital, is being replaced in many areas of the country by a new system in which primary care physicians relinquish the care of hospitalized patients to a new group of inpatient specialists called hospitalists.1 The most important forces driving the hospitalist movement in adult medicine are the demand for higher quality hospital care and more economical use of resources. Although the use of specialists in inpatient medicine has existed in Europe and Canada for many years and also in some academic centers in this country, the widespread adoption of the hospitalist model in the United States has important implications for patients and physicians. This article explores the rationale for this change in the organization of inpatient care, the experience to date with hospitalists in the care of adult patients, and the potential implications of this change for the care of hospitalized children.
THE HOSPITALIST MOVEMENT IN ADULT MEDICINE
Wachter has defined hospitalists as physicians who spend at least 25% of their time serving as the physician of record for hospitalized patients who have been referred by primary care physicians and who are referred back to their primary care physician at the time of discharge.2 The rationale for hospitalists is a natural extension of the burgeoning literature that supports the idea that “practice makes perfect” in the care of patients with illnesses such as acute myocardial infarction, stroke, and AIDS.3–5This minimum time commitment for hospitalists (here defined as 25%, although most community hospitalists serve in this role for 100% of their time) permits the development of expertise related to clinical content and efficiency and also ensures a high level of accountability for the inpatient setting. The time commitment also invests the hospitalist in activities designed to improve the system itself. This is an important issue because hospitalists who are familiar with the hospital system arguably can work more effectively with hospital administration, nursing services, and other departments than can primary care physicians who use the services only occasionally.
As important as medical expertise is the issue of availability. Over the past 20 years, the number of hospitalized patients has decreased substantially and their acuity of illness has increased markedly. In addition, physicians now undergo intensive scrutiny on length of stay and hospital costs. Many primary care physicians now want hospital-based physicians to care for their patients because they can not meet the demands of both hospital and outpatient practice. For others, it is physically impossible to care for patients in the hospital because their practice is too far away.
Hundreds of American hospitals and medical groups have adopted the hospitalist model for the care of adult inpatients. The majority of the hospitals in Northern California now use hospitalists for inpatient care and large managed care organizations such as Kaiser Permanente, Humana, and United Health Care all have expressed their support for the model.1,6–8 A new society, The National Association of Inpatient Physicians, now has more than 2000 members and recently became affiliated with the American College of Physicians. A recent policy conference, “The Emerging Role of Hospitalists in American Health Care: A National Conference,” sponsored by the US Agency for Health Care Policy and Research, was held in December 1997 and drew more than 500 registrants.
Yet a number of concerns have been raised about the hospitalist model. Does the transfer of care from the primary care physician to the hospitalist and back again to the primary care physician lead to problems with continuity of care? Are patients satisfied with the care provided by hospitalists? Are primary care physicians pleased with hospitalist care and communication? These are extremely important issues. Although there is not much information concerning the hospitalist model and patient and primary care physician satisfaction, two centers have reported high levels of patient and physician satisfaction in hospitalist programs that have found ways to ensure excellent communication between hospitalists and primary care physicians.9,,10 Furthermore, data from the same centers report average decreases in length of stay and hospital costs of 10% to 20%.9,,10 Even with the paucity of data, it is clear that one key to the hospitalist model is excellent communication between hospitalists and referring physicians. In adult medicine, many innovations are being developed including computer links, faxing of admitting and discharge information, and follow-up clinics where hospitalists see patients one or two times after discharge. However, even with excellent communication and follow-up, both hospitalists and referring physicians must be sensitive to the needs of patients and families. If patients understand that their physicians are working together, they will be more inclined to trust and work with them. Despite legitimate concerns about the hospitalist model, few have challenged the underlying premise of the hospitalist movement—that the use of dedicated inpatient physicians is likely to decrease the cost of inpatient care while maintaining and perhaps improving its quality.
IMPLICATIONS OF THE HOSPITALIST MOVEMENT FOR PEDIATRICS
In the United States, most hospitalized children with general pediatric problems are cared for by their own physician or by a physician member of a medical group designated to care for all the inpatients of the group on a rotating basis. This was precisely the situation in adult medicine until recently. Not surprisingly, the same forces driving the hospitalist movement in adult medicine— demand for higher quality hospital care and more economical use of resources—are driving the pediatric hospitalist movement. A discussion of the specific shape of these forces in pediatrics; the implications of the movement in terms of education, manpower, and finances; and a case study of one pediatric hospitalist program established recently follows.
SPECIFIC FORCES PROMOTING PEDIATRIC HOSPITALISTS
The forces promoting the hospitalist model in pediatrics are, if anything, more persuasive than those in adult medicine. The average primary care pediatrician will have even less experience caring for hospitalized patients than the average internist, the difference in illness between outpatient and hospitalized patients will be greater, and the demand for availability in the outpatient setting will be more acute. Also, many pediatricians as well as other primary care physicians feel the pressure of increasing demands of their office practice, whereas others practice in a geographic location far from the hospital. However, what makes it easier to establish hospitalist programs in adult medicine compared with pediatrics is the larger number of hospitalized patients and referring physicians.
A PEDIATRIC HOSPITALIST SERVICE
Although a number of hospitalist programs have been established in adult medicine, few programs have been established formally in pediatric medicine. One such program, called the Generalist Inpatient Service (GIS), was established in September 1996, at Children's Hospital Medical Center (CHMC), a large academic teaching hospital in Cincinnati, Ohio. Some of the impetus for the development of this service came from requests of community physicians who wanted to identify a smaller number of hospital-based physicians who would care for their inpatients, for geographic or other reasons. More than 300 primary care physicians within a 100-mile radius of Cincinnati refer to the GIS, which has >2000 admissions per year.
At CHMC, the academic pediatric generalists primarily supervise the GIS, although several pediatric subspecialists with interest, experience, and expertise in general pediatrics also serve as attending physicians on the service. Neonatologists care for infants in the Regional Newborn Intensive Care Unit and critical care specialists care for children in the Pediatric Intensive Care Unit. Similarly, GIS patients admitted to subspecialty services are under the care of the subspecialty attending physicians.
When GIS patients are transferred from the intensive care unit to a general pediatric unit, the GIS physician assumes responsibility. The GIS has generally replaced the previous traditional ward-attending system, in which many academic pediatric generalists were attending physicians in the hospital for 1 month per year. With the advent of the GIS, faculty now are asked to decide whether they would like to spend most of their clinical time taking care of inpatients or outpatients. Three faculty members have chosen to spend 3 or more months per year on the GIS, which accounts for ∼40% of the total faculty service time. Most of the other faculty members serve for 6 weeks per year as GIS physicians.
When the GIS began, the service was small and was managed by one GIS physician and one senior resident. Now, with the large number of patients cared for by the GIS, this service has become one of the three general pediatric units in the hospital. The GIS is divided into two teams, A and B. Each team consists of one GIS physician, one senior resident, two first-year residents, and two third-year medical students. The teams admit every other day, consistent with an every fourth night on call system for the first year residents, and function independently except for cross-coverage by residents on nights and weekends. The fewer members on the teams allow for more bedside participation and teaching. The average census is 10 to 15 patients for each team, which is a number comparable with the usual adult hospitalist census.
One of the keys to an effective hospitalist system is communication with primary care physicians. In the case of the GIS, primary care physicians only have to make one telephone call through the hospital Priority Link system to speak to the hospitalist on call and arrange the admission. When patients are admitted through the emergency department, the GIS physician is notified. Within 24 hours of admission, the first-year resident responsible for the patient calls the referring physician and discusses management plans. If there are difficult or complicated problems, the GIS physician also will call. The referring physician also is called as necessary during the hospitalization and at discharge. A brief letter is dictated by the senior resident, reviewed by the GIS physician, and faxed to the referring physician as soon as possible after discharge, and a signed copy is sent by mail. The resident's discharge summary follows. Assuming these responsibilities helps residents learn these important communication skills, which are necessary in medical practice.
Because the use of the hospitalist system at CHMC is voluntary, many primary care physicians continue to care for their own hospitalized patients. This arrangement suits the needs of both groups of primary care physicians—those who refer to the hospitalists and those who care for their own patients in the hospital. In our experience, the physicians most likely to refer to the GIS are family practice physicians, whereas primary care pediatricians are more likely to care for their hospitalized patients, especially if their office is near the hospital. We have not required that physicians always or never use the GIS. For example, physicians who care primarily for their own inpatients can request that hospitalists assume care for their patients when they are away on vacation or at a meeting. This has worked particularly well for physicians in solo practice and has allowed skeptical primary care physicians to gain comfort with the use of hospitalists before deciding to use them extensively. Although physician satisfaction with our pediatric hospitalist system has not been studied formally, referring physicians seem to be pleased because of easy access to the hospitalist physicians and improved communication by timely telephone calls and letters. Through this communication, both referring physicians and hospitalists have come to know each other better, and the increased emphasis on communication has helped the entire system run more effectively.
To our knowledge, patient satisfaction with a pediatric hospitalist system has not been studied, but in our experience, patients and their families have accepted the hospitalists as the physicians responsible for their hospital care and understand that follow-up will be provided by their own physician. When patients are admitted to the hospitalist service, the hospitalist physicians explain that they are working with the patient's primary care physician. This emphasizes the ongoing close relationship between these two groups of physicians, which helps gain the patient's trust.
As the demand for hospitalists grows, academic programs must reconsider the best ways to create effective training models. The Department of Medicine of the University of California San Francisco recently has received a grant from the Josiah Macy Jr. Foundation to establish a residency track for hospitalists within the internal medicine residency program. This training will augment traditional inpatient training with more emphasis on communication skills; medical consultation, pre- and postoperative care, and palliative care skills; quality improvement, utilization management, and practice guideline development; information management skills; and a more comprehensive understanding of the continuum of care, including the role of home care, the hospice, and the skilled nursing facility. The hospitalist track also will retain significant outpatient training, because hospitalists must be keenly aware of what can and cannot be done outside the hospital.
In designing training pathways for pediatric hospitalists, we believe that traditional pediatric training programs also provide an excellent core experience for future hospitalists. As far as we know, no hospitalist residency tracks are yet planned in pediatrics; however, whether or not these tracks are developed, residents should have sufficient time in intensive care and consultative medicine during residency training to be highly competent in these roles. The increasing emphasis on outpatient activity in both internal medicine and pediatric residency programs will only enhance the ability of the hospitalist to integrate inpatient services with the combination of services managed by the primary care physician.11Pediatricians who desire an academic career should consider a fellowship in general academic pediatrics, with significant time devoted to inpatient and consultative care, as well as to research training. Other opportunities for research training include the Robert Wood Johnson Clinical Scholars Program or an appropriate advanced degree. As with their adult medicine colleagues, pediatric hospitalists also need to hone their leadership, communication, and information management skills through residency, fellowship, or continuing education programs.
Although community-based hospitalists generally will spend 100% of their time caring for inpatients, hospitalists in academia may spend from 25% to 50% of their time in this clinical role and the remainder in teaching, research, or other clinical work. In academic medicine, it is difficult for a physician to spend a considerable period as an inpatient attending physician and manage a primary care practice with the necessary continuity to attract patients. Individual faculty members generally will need to decide whether to work primarily in an inpatient or outpatient setting, so that they can gain more expertise in clinical care and teaching and address important research questions in their respective areas. For hospitalists, their role will vary depending on the size and type of hospital in which they work.
Compared with adult medicine, in which some have argued that intensivists are well-suited to be hospitalists for general adult inpatients,12 the problems in general inpatient pediatrics are sufficiently distinct from those of intensive care patients (neonatal and others) that neonatologists and critical care specialists are unlikely to assume the hospitalist role, especially in large academic centers. In these centers, hospitalists will assume the care of children with general pediatric problems and teach residents and medical students. They also will be available to provide general pediatric consultation to other services and physicians. Neonatologists will continue to care for infants in the neonatal intensive care unit, and critical care physicians will care for children in the intensive care unit. However, in smaller, nonteaching community hospitals, hospitalists will not only care for the children on a general pediatric unit, but they also may assume care for infants in levels I and II neonatal units and for children in the intensive care unit. They also will provide consultation to other services, particularly the surgical services and the emergency department, and help transport ill children to and from their hospital. In teaching hospitals, residents are available 24 hours a day, which allows them to respond to the needs of both patients and families around the clock. Hospitalists in nonteaching hospitals can serve in the same role when they provide in-hospital coverage 24 hours a day 7 days a week.
Although the hospitalist role may attract both pediatric generalists and subspecialists, we believe that this role requires a generalist perspective. Effective hospitalists view the patient as a whole and do not focus just on one organ system or disease process. They are concerned with the psychosocial aspects of illness and appreciate the need to communicate well with patients, families, primary care physicians, subspecialists, case managers, and others. For academic general pediatricians, the role is especially appropriate because many are already hospital-based. The hospitalist role provides an excellent model for teaching inpatient medicine to residents and medical students. The progressive change in the intensity and complexity of illness among pediatric inpatients demands specific expertise and availability that hospitalists can provide. They also can teach communication skills, evidence-based medicine, and cost-effective management.
In adult medicine, most hospitalists are general internists, although a substantial number are subspecialists (most commonly pulmonologists or critical care specialists).13 Whether generalists or subspecialists do a better job in the care of hospitalized children is an issue that needs to be studied. Although hospitalists should be comfortable with the management of common inpatient problems and quickly become adept in the management of common inpatient specialty disorders, there is no question that subspecialists are required in certain cases because of their expertise and experience. Even when the management of complex problems requires the use of multiple consultants, hospitalists have much to offer in integration and overall management, especially if there are difficult psychosocial issues.
Even with hospitalists, pediatrics is in no danger of ceasing to be a primary care specialty, because many children require the expertise of primary care pediatricians for the management of problems that do not require hospitalization, such as many acute and chronic illnesses, as well as psychosocial and behavior problems. These physicians also are needed to care for children with more complex problems in the outpatient setting, especially when these children are discharged earlier from the hospital with ongoing treatment plans such as the use of intravenous antibiotics and tube feedings.
It is difficult to estimate how many pediatric hospitalists will be required, but it will be far less than the number of adult hospitalists (for which estimates have ranged from 3000 to 30 000: one hospitalist for eight generalist physicians).14 The use of pediatric hospitalists thus may be feasible only in those areas with large populations and numbers of physicians. Our prediction is that the current supply of general pediatricians will be adequate to support the hospitalist movement in pediatrics.15,,16
As in adult medicine, the organizational arrangements for pediatric hospitalists will vary. Some hospitalists will be salaried by a hospital or health system, others will be part of a hospitalist group that may cover multiple hospitals, and still others may be self-used. Academic clinician–educators will spend up to 6 months per year on the inpatient service, and promotion will be based on outstanding clinical and teaching skills. Clinical researchers will do clinical trials, outcome studies, and health services research to improve the care of the individual patient as well as the entire inpatient system of care while spending fewer months in clinical inpatient practice. In both community and academic settings, it is likely that hospitalists will oversee quality assurance and utilization review functions and participate in the development of practice guidelines. Not only are these endeavors important for children's hospitals, but also for general hospitals with pediatric inpatient units that manage only a fraction of the patients in the hospital. For these units, knowledgeable advocates are needed more than ever to ensure that these inpatient services have the necessary resources to deliver high-quality care.
The financial issues surrounding the hospitalist model are complex, and only a small amount of data is available to begin to understand them.9,,10,17 Despite the attractiveness of the hospitalist concept on many levels, most hospitalist systems will not prove to be financially self-sufficient.17 Although this may seem counterintuitive, the experience with adult hospitalist systems is that the revenues generated by hospitalists through fee-for-service billings and/or capitation dollars generally equal ∼60% to 80% of the hospitalists' salaries (the only exceptions are those hospitalists who work in markets in which there is little managed care and who care for insured patients only). The reasons for this shortfall are multiple: hospitalists often staff the hospital 24 hours per day, 7 days a week; they are available throughout the hospital day to see patients multiple times if necessary (even if reimbursed for only one visit); and many hospitalists care for the uninsured or unassigned patients admitted from the emergency department as one of their core functions.
Although this balance sheet would appear to make a hospitalist system seem unattractive from a financial standpoint, when hospitalists reduce lengths of stay or utilization of unnecessary tests, the savings generally far exceed their salaries. Therefore, hospitalist systems need to be subsidized by whichever entity is benefiting from the value of the hospitalists.17 For example, in a capitated system in which risk has been delegated to the medical group, the group benefits when hospital costs can be cut while preserving quality and patient satisfaction. Under these circumstances, it is the medical group that often will choose to support the hospitalist system. In other circumstances in which the hospital is being paid a fixed rate per diagnosis-related group, the hospital will be the beneficiary of cost savings and will need to help support the hospitalist program. More research needs to be conducted to understand fully the financial implications of the hospitalist model. At CHMC we are studying patient and financial outcomes of the model.
Although the growth of the hospitalist model will be influenced strongly by the characteristics of local medical practice, we believe that the arguments promoting this model are sufficiently compelling that its widespread adoption, already ensured in adult medicine, will come to be realized in pediatrics. If the situation for adults is any guide, there will be wide variations in the types of hospitalist models used, and local circumstances will make the model inappropriate or even impossible in some areas of the country. In others, the model will exist side-by-side with other models of care, at least in the near future. From the experience of one pediatric hospitalist model, our impression is that general pediatricians can assume the hospitalist role, do it very well, and enjoy a satisfying career. Ultimately, the models that are promoted should be those that can be demonstrated to yield the best clinical outcomes and the highest patient satisfaction at the lowest cost. Whether these benefits can be realized through pediatric hospitalist systems can be determined only through research studies, which we hope will be performed in the near future.
- Received March 10, 1998.
- Accepted October 9, 1998.
Reprint requests to (P.S.B.) Department of Pediatrics, Children's Hospital Medical Center, 3333 Burnet Ave, Cincinnati, OH 45229-3039.
- GIS =
- Generalist Inpatient Service •
- CHMC =
- Children's Hospital Medical Center
- Wachter RM
- Stroke Unit Trialists' Collaboration
- ↵Likosky WH, Craig D, Hartka L. Implementation and outcomes of a hospitalist program: Kaiser Permanente Northern California. Ann Intern Med. In press
- ↵Shepherd G. Humana “hospitalists” usher in new era in medicine. Tampa Bay Business Journal. Oct 13, 1997:1
- ↵Newcomer LN. Hospitalists: a managed care view. In: Wachter RM, ed. Proceedings of the National Conference on the Emerging Role of Hospitalists in American Health Care. San Francisco, CA: UCSF Publications; 1997
- ↵Freese RB. The Park Nicollet experience in establishing a hospitalist system. Ann Intern Med. In press
- ↵Graduate Medical Education Directory 1997–1998. Chicago, IL: American Medical Association; 1997
- Gipe B
- ↵Lindenauer PK, Pantilat SZ, Katz PP, Wachter RM. Hospitalists and the practice of inpatient medicine: results of a national survey. Ann Intern Med. In press
- ↵O'Neil EH, Ruhlin C. Health workforce needs of hospitalist based health systems. In: Proceedings of the National Conference on the Emerging Role of Hospitalists in American Health Care. San Francisco, CA: UCSF Publications; 1997
- ↵Physician Characteristics and Distribution in the US 1997–1998. Chicago, IL: American Medical Association; 1997
- Oliver TK,
- Tunnessen WW Jr.,
- Butzin D,
- Guerin R,
- Stockman JA III.
- ↵Wachter RM, Whitcomb WF, Nelson JR. Financial implications of the hospitalist movement. Healthcare Financial Management. In press
- Copyright © 1999 American Academy of Pediatrics