Over the past few years, the care of medical inpatients increasingly has been turned over to an emerging group of professionals called hospitalists. Typically, hospitalists are internists who devote the majority of their clinical effort to caring for inpatients.1,2,3,4 Outpatient responsibilities, if they exist, are minimal. The defining characteristic of hospitalists is the “hand-off” cycle: a primary care provider admits the patient to the designated hospitalist, who provides inpatient care and then sends the patient back to the primary care provider upon discharge.
Many early arguments for hospitalists centered on the positive impact that this model of care would have on resource utilization and patient outcomes. Indeed, early data suggest that care by hospitalists is associated with reductions in length of stay, lower readmission rates, and improved resource utilization,5,6,7,8 and there seems to be little negative impact on patients' satisfaction.9 Among the issues that have not been fully addressed is the role that hospitalists play in medical education. Potential issues have been discussed, such as a diminished sense of autonomy among residents,10,11 perhaps counterbalanced by increased satisfaction and better supervision of patients.4,9 For other issues, such as the presence of hospitalists in academic medical centers and their teaching responsibilities, few data have been presented. Historically, the cornerstone of both undergraduate and graduate medical education has been inpatient-based. Though ambulatory care training has been emphasized in recent years, the inpatient wards remain the major site of clinical teaching. If beds and/or wards are being turned over to hospitalists, it is important to determine the impact this may have on educational programs.
The purpose of this study was to address such educational issues. Separate questionnaires were mailed to the chairs and program directors of all internal medicine training programs in the United States to learn (1) how many programs have hospitalists on staff, and to gain information about related census issues (e.g., number hired, plans for future hires); (2) the role of hospitalists in teaching activities, and (3) attitudes regarding the roles hospitalists play in general and their role in teaching, specifically.
The questionnaire was sent to all chairs and program directors of accredited internal medicine training programs who were identified in the spring of 1999 using the 1998–1999 AMA Graduate Medical Education Directory. This process resulted in a roster of 106 chairs, 382 program directors, and 22 individuals who filled both roles. Three separate questionnaires were developed. Content was defined by the study team, taking ideas from current literature as well as discussions that had taken place locally in the course of developing a hospitalist service in 1998. Draft instruments were revised numerous times to improve clarity and breadth, after piloting them with faculty.
The questionnaire for chairs was brief (eight questions) and focused on asking whether hospitalists were employed at the sites and if so, defining how long they had been there and their training backgrounds and responsibilities. A more extensive questionnaire was developed for program directors. In addition to general program descriptions, directors were asked whether hospitalists were employed and provided 12 attitude statements about hospitalists to be answered on a five-point Likert scale from “strongly disagree” to “strongly agree.” For programs that had hospitalists on staff, a set of questions focused on teaching responsibilities, participation in other educational activities, and 13 more attitude statements about hospitalists' roles and their impact upon teaching. The questionnaire for the few individuals who were both chairs and program directors was a collection of the unique items from the other two versions. The first mailings were sent in April 1999. A second mailing with a new copy of the instrument was sent in June 1999. Because the response rate for program directors was low, for the third mailing, items asking about activities at each training site were omitted to reduce the respondents' burden, thus shortening the questionnaire from four to two pages. The third mailing was sent in August 1999. The final response rates were 78.3% (n = 83) for chairs and 57.6% (n = 220) for program directors. The eight responses from the 22 chairs-program directors were added to both data files, for analytic sample sizes of 91 and 228.
Analyses of the responses focus on description. We used standard univariate statistics (frequencies and percentages) to characterize the sample. To test for differences between programs that did and did not respond, between responses to the long and short survey forms, and between programs that did and did not employ hospitalists, we used chi-square, t-tests, and the Wilcoxon two-sample test.
Respondents and Non-respondents. Program characteristics available from the 1998–1999 AMA Graduate Medical Education Directory allowed limited comparison of non-respondents with respondents. Overall, the program sizes were the same for respondents (mean = 52.9, SD = 33.2) and non-respondents (mean = 55.8, SD = 36.9, p =.40). The respondents and the non-respondents did not come from different regions of the country (p =.086).
There were few significant differences between the responses of the 130 program directors who responded to the long form of the program directors' questionnaire and the 90 who responded to the short one. For example, there was no difference in numbers of inpatient training sites (p =.63) or numbers of categorical residents at the PGY1 level (p =.35). There was no difference in the percentages who had hospitalists (p =.80), were planning to hire hospitalists (p =.59), or had rejected the idea of having hospitalists (p =.47). Those responding to the short form had more favorable attitudes with respect to one of the 13 attitude items.
Chairs. Overall, 50 of the chairs (55.6%) reported that hospitalists were employed at one or more of their training sites. The numbers of hospitalists per institution ranged from one to 15, with a median of four. (The total number of hospitalists employed by the 44 programs that reported having them was 206.5.) Twenty-nine (64.4%) planned to hire more hospitalists. The tenure of hospitalist programs was a median of two years, with a range of 0.5 to 7.5 years. Nearly three fourths of the hospitalists (71.9%) had completed residencies in internal medicine, 4.4% had completed general internal medicine fellowships, and 11.4% had completed subspecialty fellowships.
The reported duties of the hospitalists were quite variable. The numbers of months of inpatient responsibilities ranged from one to 12, with a median of eight. The percentages of the responding department chairs reporting other responsibilities for hospitalists were: 55.3% reporting hospitalists with outpatient practices (with a median of 10% full-time equivalent); 77.8%, medicine consultation; 46.8%, clinical pathways/disease management development or implementation; 31.9%, quality assurance; 27.7%, medical directorships; and 17.4%, insurance company or managed care liaisons. Of note, 53.2% required academic productivity for promotion.
Of the programs that did not have hospitalists, 37.1% planned to hire them in the future and 16.1% had considered but rejected the idea.
Program Directors. Overall, 50.5% of the responding programs employed hospitalists. As shown in Table 1, many program directors' attitudes about hospitalists were positive. For example, the majority agreed that hospitalists are more familiar with practical aspects of inpatient care, that they are good for the hospital financially, and that patients of hospitals are satisfied with their inpatient care. Most disagreed that they needed more training beyond that gained in an internal medicine residency. On the other hand, most also thought that use of hospitalists disrupted the continuity of patient care, and only one third agreed that hospitalists provide better inpatient care than other general iternists.
The last three columns of Table 1 shows the means of the Likertscale responses of the program directors with and without hospitalists. Differences were significant, and in the anticipated direction, for seven of the 12 attitude statements.
In addition, respondents of the 109 programs with hospitalists were asked whether the hospitalists participated in a number of different activities related to education. Nearly all participated in the teaching of medical students (80.2%) and residents (84.5%). Other educational activities in which they participated included attending physicians' rounds (74.7%); residents' reports (58.6%); management conferences (53.5%); curriculum development (55.6%), and journal club (48.5%). Specific topics taught by the hospitalists included cost-effective care (57.1%); resource utilization (57.1%); health economics (42.9%); clinical pathways/disease management (38.8%); and insurance principles (26.5%). In nearly all programs (78.1%), students and housestaff evaluated the hospitalists.
Table 2 lists additional attitudes of the program directors who employed hospitalists, especially their perceptions of hospitalists' role in and impact on teaching activities. Over 70% agreed that that hospitalists are viewed as good educators and are respected. The majority thought that hospitalists have led to improved housestaff supervision and are more accessible to housestaff than other teaching faculty. They were less certain that hospitalists had an impact on housestaff's considerations of lengths of stay and costs of tests and procedures, or that the housestaff had learned to order fewer tests and procedures.
The results of the surveys reported above show that hospitalists have a presence in both undergraduate and graduate medical education: at least half of the responding training programs employed hospitalists, who in most cases played roles in teaching students and/or residents. Attitudes expressed by the total sample of program directors were generally positive, naturally more so for those representing programs with hospitalists. In particular, program directors from programs with hospitalists were especially complementary about the hospitalists' familiarity with practical aspects of care, their positive financial impact on the hospitals, their positive impact on patients' satisfaction, and the improvements in residency training. On the other hand, most programs had only a few hospitalists, they had had them for only one or two years, the numbers of months in inpatient responsibility ranged from one to 12, and their involvement in a variety of specific teaching activities was varied. Given this variation, it might not be feasible to characterize “the” teaching role of hospitalists.
This study has some limitations. The response rate for program directors was relatively low, although we found no evidence of bias. Second, we are able to create a composite picture of what hospitalists do, but we did not collect parallel data regarding non-hospitalist attending physicians. Thus, we are missing a piece of the total picture. Third, we did not ask for detailed data regarding the teaching activities of the hospitalists, e.g., what does participation in curriculum development mean? Nevertheless, to our knowledge this is the first study to detail the presence of hospitalists and provide an overview of their teaching activities in teaching institutions.
Overall, even though their numbers were small, in at least half of the U.S. internal medicine training programs that responded, hospitalists were present and played roles in teaching. Given the amount of time they spend in inpatient services, they have widespread exposure to learners on all levels. This visibility makes hospitalists as a group an ideal target for faculty development focused on teaching methods and feedback skills. Although the respondents generally viewed hospitalists as excellent teachers who had led to improved training for residents, hospitalists as teaching faculty should be evaluated compared with faculty involved in teaching on traditional services.
Much of the justification for hospitalists draws on arguments that they should be able to save money by reducing test ordering and lengths of stay. If they are really succeeding in these areas, as current data suggest they might be,5,6,7,8 it is logical to assume they could affect residents' behaviors via modeling and/or direct teaching of optimal management strategies. The fact that so few program directors believe hospitalists will have an effect on residents' future behaviors in the areas of ordering and effective management is somewhat surprising and deserves more study. Examining residents' behaviors and attitudes in terms of the relative amounts of exposure they have to services led by hospitalists would yield useful insights. An additional contribution would be an understanding of how, as a group, hospitalists' teaching activities and outcomes differ from those of their peers, and whether there is observable variation in the reasons hospitalists' services were developed (e.g., cost efficiency, excellence in inpatient teaching, as a “safety valve” for overburdened teaching services). Future studies will be most helpful if they are aimed at defining the unique contributions attributable to hospitalists.
1. Wachter RM. An introduction to the hospitalist model. Ann Intern Med. 1999; 130(4:part 2):338–42.
2. Wachter RM, Goldman L. The emerging role of hospitalists in the American health care system. N Engl J Med. 1996;335:514–7.
3. Weissler JC. The hospitalist movement: caution lights flashing at the crossroads. AJM. 1999;107:409–13.
4. Whitcomb WF, Nelson JR. The role of hospitalists in medical education. Am J Med. 1999;107:305–9.
5. Craig E, Hartka L, Likosky WH, Caplan WM, Litsky P, Smithey J. Implementation of a hospitalist system in a large health maintenance organization: the Kaiser Permante experience. Ann Intern Med. 1999;130(4:part 2):355–9.
6. Freese RB. The Park Nicollet experience in establishing a hospitalist system. Ann Intern Med. 1999;130(4:part 2):350–4.
7. Diamond HS, Goldberg E, Janosky JE. The effect of full time faculty hospitalists on the efficiency of care at a community teaching hospital. Ann Intern Med. 1998;129:197–203.
8. Brown MD, Halpert A, McKean S, Sussman A, Dzau VJ. Assessing the value of hospitalists to academic health centers: Brigham and Women's Hospital and Harvard Medical School. Am J Med. 1999;106:134–7.
9. Wachter RM, Katz P, Showstack J, Bindman AB, Goldman L. Reorganization of the academic medical service: impact on cost, quality, patient satisfaction and education. JAMA. 1998;279:1560–5.
10. Goldman L. The impact of hospitalists on medical education and the academic health center. Ann Intern Med. 1999;130(4:part 2):364–7.
11. Schroeder SA, Schapiro R. The hospitalist: new boon for internal medicine or retreat from primary care? Ann Intern Med. 1999;130(4:part 2):382–7.
Research in Medical Education: Proceedings of the Thirty-ninth Annual Conference. October 30 - November 1, 2000. Chair: Beth Dawson. Editor: M. Brownell Anderson. Foreword by Beth Dawson, PhD.