Share this article on:

Psychosocial Training in U.S. Internal Medicine and Family Practice Residency Programs

Gaufberg, Elizabeth H. MD, MPH; Joseph, Robert C. MD, MS; Pels, Richard J. MD; Wyshak, Grace PhD; Wieman, Dow PhD; Nadelson, Carol C. MD

Educating Physicians: Research Reports

Purpose Competence in the psychosocial aspects of medical care is necessary for primary care physicians to function effectively. This study investigated the psychosocial training internal medicine and family practice residents receive in U.S. programs.

Methods In 1996, program directors of all U.S. internal medicine (IM) and family practice (FP) residency programs were surveyed regarding the format, content, and quantity of psychosocial training provided in their programs, their opinions on topics related to psychosocial training, and demographics of their programs.

Results The response rate was 61%. Ninety-nine percent of FP and 62% of IM program directors reported requiring at least one psychosocial training experience. Family practice programs required an average of 352 hours (SD ± 175; range 27-2,664) of psychosocial training compared with 118 hours (SD ± 272; range 0-1,050) for IM programs. Most IM and FP program directors expected residents to achieve at least basic competency in virtually all psychosocial topic areas; however, FP programs provided a greater range of psychosocial experiences. FP program directors most often identified psychologists and IM program directors most often identified internists as providing the most psychosocial training in their programs. Both IM and FP program directors considered lack of curricular time to be the main obstacle to development of psychosocial training.

Conclusion Residents' competence in psychosocial areas is important to both IM and FP program directors. However, content and time devoted to psychosocial training vary considerably both within and between program types.

Dr. Gaufberg is instructor in medicine (she was instructor in psychiatry at the time the paper was written), Dr. Joseph is assistant professor of psychiatry, Dr. Pels is instructor in medicine, Dr. Wieman is lecturer on psychiatry, Dr. Nadelson is clinical professor of psychiatry, all at the Harvard Medical School, Boston, Massachusetts. Dr. Wyshak is associate professor, Department of Psychiatry, Harvard Medical School and Biostatistics and Population and International Health, Harvard School of Public Health, Boston.

Correspondence and requests for reprints should be addressed to Dr. Gaufberg, The Cambridge Hospital, 1493 Cambridge Street, Cambridge, MA 02139; e-mail: 〈〉.

The authors acknowledge David Bor, MD, Chief of the Department of Medicine at The Cambridge Hospital, and (posthumously) Robert Dorwart, MD, former chief and chair of the Department of Psychiatry at The Cambridge Hospital, for their intellectual and material support of this work.

This research was supported by the Harvard Medical School Department of Psychiatry, Livingston Fellowship Fund.

Graduates of primary care residency programs must be proficient not only in biomedical medicine, but also in providing humane and psychologically sophisticated care. Illness and behaviors associated with illness are known to be influenced by psychological issues. Furthermore, patients with overt psychiatric illnesses are common in the primary care setting. These illnesses are debilitating and costly, and primary care physicians may underdiagnose and undertreat psychiatric problems.1

Recognizing the need for greater psychosocial training in primary care disciplines, the Accreditation Council for Graduate Medical Education has developed requirements for behavioral health curricula in internal medicine (IM) and family practice (FP) training programs.2 Educators and professional organizations have responded by describing model rotations and articulating learning competencies for training in the behavioral sciences.3–8 However, little is known about the quantity and quality of psychosocial training offered nationally in IM and FP residency programs. Three older reports9–11 may not accurately reflect current training, and a fourth report that focused on didactic training did not describe required rotations.12

Our report describes and quantifies psychosocial training in IM and FP residency programs in the United States.

Back to Top | Article Outline


Program directors of all 416 U.S. internal medicine and 455 U.S. family practice residency programs listed in the American Medical Association (AMA) database13 for the academic year 1995 were surveyed in September 1996 about the psychosocial training (such as block rotations and structured longitudinal experiences) that they required and formally integrated into their residency programs. On the survey instrument, psychosocial training was defined broadly as formal training in psychiatric, behavioral, or addictive problems, interviewing, and relevant social and ethical issues.

The survey instrument consisted of three sections: (1) seven questions (multiple-choice, rating-scale, and write-in) about the format, content, and quantity of psychosocial training; (2) six questions (multiple-choice and rating-scale) eliciting program directors' opinions about various topics related to psychosocial training; and (3) seven write-in questions about demographic characteristics of the program director, the program, and the base institution. The survey instrument was pre-tested and evaluated by six current or former program directors of IM or FP residency programs, as well as by psychosocial educators.

We performed simple univariate statistical analyses using a standard statistical software package. Institutional review board approval was granted from the researchers' home institution.

Back to Top | Article Outline


The overall survey response rate was 61% (64% of all FP program directors responded and 58% of all IM program directors responded). Chi-square analyses and t-tests indicated no significant difference between the respondents and non-respondents with respect to elements found in the AMA database, such as geographic region, program size, and program type. Totals of 99% of FP and 62% of IM program directors reported requiring at least one psychosocial training experience in their residency programs (p < .001).

Back to Top | Article Outline

Information about Psychosocial Training

In FP programs with at least one required psychosocial training experience, 48% of the program directors reported psychologists provided the most psychosocial training, 22% reported family physicians provided the most training, 16% reported social workers provided the most training, and 8% reported psychiatrists provided the most training. In IM programs with at least one required psychosocial training experience, 41% of the program directors identified internists as providing the most psychosocial training, 36% identified psychiatrists, 14% identified psychologists, and 4% identified social workers.

Back to Top | Article Outline

Expected Levels of Competency for Psychosocial Topics

Table 1 lists the levels of competency that these program directors expected of their graduates with respect to various psychosocial topics. Both FP and IM program directors rated the doctor—patient relationship (84.6% and 80.7%, respectively) and interviewing techniques (76.8% and 59.7%, respectively) as areas in which they expected high levels of knowledge and clinical expertise. While most of the program directors expected at least “basic competency” in most psychosocial topics, 73% of IM program directors versus 32% of FP program directors did not expect basic competency for psychotherapy and counseling.

Table 1

Table 1

Back to Top | Article Outline

Content and Quantity of Psychosocial Training

The program directors' responses about the content and quantity of required psychosocial training for a typical resident over the three years in their programs were categorized into two basic formats: longitudinal format experiences and block-rotation experiences. These results are reported in Table 2.

Table 2

Table 2

The total amounts of time devoted to required psychosocial training during residency (sum of longitudinal and block experiences) were 352 hours (SD ± 175, range 27-2,664) for FP and 118 hours (SD ± 272, range 0-1,050) for IM programs (p < .001). The mean total numbers of hours devoted to the longitudinal experiences (excluding precepting) were 163 (SD ± 198, range 0-2,184) for FP programs and 34 (SD ± 88, 0-450) for IM programs (p < .001). A greater percentage of FP programs provided each category of longitudinal experience and devoted more curricular time to psychosocial training than did IM programs. In all, 81% of FP and 39% of IM programs that responded had at least one psychosocial block rotation (p < .001). The total means of block rotation hours were 189 (SD ± 160, range 0-1,120) for FP and 84 (SD ± 147, 0-960) for IM programs (p < .001).

Back to Top | Article Outline

Relative Emphasis on Psychosocial Versus Traditional Medical Topics

To gauge relative emphasis on psychosocial versus traditional medical topics, we compared the numbers of resident lectures in the individual programs in the preceding year devoted to depression and hypertension, because the prevalences of these conditions are approximately equivalent and both cause substantial disability and mortality. A total of 51% of the IM program directors reported having two or more resident lectures on depression versus 76% of FP program directors, and 80% of IM program directors reported having two or more resident lectures on hypertension versus 65% of FP program directors. Overall, IM programs had more lectures on hypertension than on depression (p < .0022), and FP programs had more lectures on depression than did IM programs (p < .0001).

Back to Top | Article Outline

Factors Associated with Psychosocial Training

The program directors were asked to rate, on a scale of 1 to 5 (1 = very poor, 5 = excellent), the quality of the psychosocial training they had received during their own residencies. The mean ratings were 3.6 for FP program directors and 2.8 for IM program directors (p < .0001).

Fifty-six percent of FP and 43% of IM program directors identified short visit length as the top factor that limited the members of their specialties in diagnosing and treating mental disorders in the outpatient setting (p < .0046). Lack of physician's knowledge and skills was perceived at the top factor by 32% of the IM program directors, but by only 10% of the FP program directors (p < 0.001).

Forty-five percent of the FP and 50% of the IM program directors identified lack of curricular time as the greatest barrier to change or development of psychosocial training (p < 0.25). Lack of effective teachers was perceived as the greatest barrier by 24% of the IM program directors versus 9% of the FP program directors (p < .0001).

Back to Top | Article Outline

Perceived Need to Expand Psychosocial Training

Totals of 28% of the FP program directors and 58% of the IM program directors responded that they would like to expand the time devoted to required psychosocial training in their programs. Those IM program directors who expressed a wish to expand psychosocial training had a significantly lower mean for total psychosocial training hours (76 hours) than did those IM program directors who did not wish to expand psychosocial training (153 hours; p =.016). For FP program directors, there was no significant difference between those wishing to expand and those not wishing to expand the training hours.

Back to Top | Article Outline


We found that a significantly higher percentage of FP programs than IM programs required psychosocial training. Virtually all FP program directors who responded said that their programs offered some psychosocial training (as specifically required in FP residency accreditation guidelines), whereas over a third of the IM programs offered none. FP program directors had a more positive impression of the quality of their own psychosocial training than did IM program directors, which may indicate that believing oneself to be well trained leads to confidence in the relevance of the subject matter and affects future teaching practices. In addition to the differences we found between program types, the considerable range of curricular hours for programs of the same specialty is also noteworthy. Recall bias, which is inherent in self-report surveys, may account for some of this variation, but lack of clear accreditation requirements regarding curricular time and content of psychosocial training may also be an important contributing factor.

Historically, the discipline of family practice has been grounded in a biopsychosocial framework, whereas internal medicine has been more biomedically and subspecialty oriented. The results of our survey support the continued validity of this construct, which likely contributes to shaping psychosocial training in residency programs. In addition to the greater amount of time spent on psychosocial training in FP programs, the differences we found between program types are most evident in three areas: expected level of psychosocial competency, the disciplines of psychosocial teachers, and the relative emphasis placed on psychosocial versus traditional medical topics.

Family practitioners, for example, have broadly defined roles, and this characteristic probably is reflected in the higher percentage of FP program directors expecting at least basic competency in areas traditionally considered to be at the core of the field of psychiatry: psychiatric mental status exam, psychiatric differential diagnosis, and psychopharmacology. The traditional subspecialty orientation of many IM programs may lead to more referrals to psychiatrists and other mental health professionals and less perceived need for competence by practitioners in these areas. The fact that both the FP and the IM program directors rated the doctor—patient relationship and interviewing skills so highly suggests that these competencies provide the foundation for all other clinical activities, a prerequisite for applying one's medical knowledge in an effective manner.

Further, because the IM programs used more MDs (primarily internists and psychiatrists) to train residents in psychosocial topics, there may be a tighter integration of psychopharmacology as well as more immediate role models for MD trainees. The same system may also constrain the available pool of teachers due to limited faculty or the cost of their time.

Finally, a similar analysis emerges for the relative amounts of emphasis placed on traditional medical topics (represented by hypertension) and psychosocial topics (represented by depression). Fewer lectures on depression in IM programs may suggest a disparity between the prevalence and morbidity of psychiatric disorders in medical settings and the emphasis they are accorded in IM curricula. However, many IM programs wish to expand psychosocial training, perhaps suggesting a shift away from the traditional emphasis on biomedicine.

Our study was subject to several limitations. We defined psychosocial training broadly to include training in psychiatric, behavioral, or addictive problems, interviewing, and relevant social and ethical issues, which reflects common parlance about psychosocial training in graduate medical education. However, not restricting our focus to a single topic (e.g., interviewing only or psychiatric diagnosis only) may have obscured differences in training among the various topic categories. Further, this study looked only at psychosocial training that is formally integrated into the training program, such as block rotations or structured longitudinal experiences, and did not assess informal psychosocial teaching that might happen on ward rounds or during impromptu case discussions. The relative emphasis placed on psychosocial issues can also be affected by factors such as the culture of an institution and individual residents' levels of interests in psychosocial topics. Other limitations of this study include a 39% non-response rate and lack of a formal validity check of the survey instrument.

Our results indicate that both FP and IM program directors recognize that adequate psychosocial training is clearly essential to the effective functioning of future primary care physicians. However, competency is essential in many areas within medicine, and the incorporation of psychosocial training experiences involves setting priorities relative to other elements of the curriculum, as well as meeting accreditation requirements. The program directors' response that “lack of curricular time” is the major barrier to change or development of psychosocial training underscores the difficult job of establishing priorities for various important curricular elements. In the current health care climate, however, improving psychosocial training alone may not solve the problem. Program directors more often identify “short visit length,” rather than “lack of physician knowledge and skills,” as the top factor limiting diagnosis and treatment of mental disorders.

The results of this study provide useful baseline information for program directors and curriculum developers to bridge the gap between the current psychosocial knowledge and skills of primary care physicians and the competencies required to meet their patients' needs.

Back to Top | Article Outline


1. Eisenberg L. Treating depression and anxiety in primary care: closing the gap between knowledge and practice. N Engl J Med. 1992;326:1080–4.
2. American Medical Association. Graduate Medical Education Directory 1998-1999. Chicago, IL: AMA, 1998.
3. American Academy of Family Physicians. Recommended Core Educational Guidelines for Family Practice Residents, Human Behavior and Mental Health. Reprint #270, July 1994.
4. Cole SA, Sullivan M, Kathol R, Warshaw C. A model curriculum for mental disorders and behavioral problems in primary care. Gen Hosp Psychiatry. 1995;17:13–8.
5. Sox HC (Chairman). Graduate Education in Internal Medicine: A Resource Guide to Curriculum Development. Report of the Federated Council for Internal Medicine Task Force on the Internal Medicine Residency Curriculum. Philadelphia, PA: American College of Physicians, 1997.
6. Kern DE, Grayson M, Barker L, et al. Residency training in interviewing skills and the psychosocial domain of medical practice. J Gen Intern Med. 1989;4:421–31.
7. Smith RC, Lyles JS, Mettler JM, et al. The effectiveness of intensive training for residents in interviewing. A randomized, controlled study. Ann Intern Med. 1998;128:118–26.
8. Williamson PR, Smith RC, Kern DE. The medical interview and psychosocial aspects of medicine: block curricula for residents. J Gen Intern Med. 1992;7:237–42.
9. Goldberg RJ, Novack D, Fulton J, Wartman S. A survey of psychiatry and behavioral science curricula in primary care residency training. J Psychiatric Educ. 1985;9:3–11.
10. Merkel WT, Margolis RB, Smith RC. Teaching humanistic and psychosocial aspects of care: current practices and attitudes. J Gen Intern Med. 1990;5:34–41.
11. Strain JJ, Pincus HA, Gise LH, Houpt J. Mental health education in three primary care specialties. J Med Educ. 1986;61:958–66.
12. Sullivan M, Cole S, Gordon G, Hahn S, Kathol RG. Psychiatric training in medical residencies. Gen Hosp Psychiatry. 1996;18:95–101.
13. American Medical Association Medical Education Research and Information Database. Chicago, IL: American Medical Association, 1995.
© 2001 Association of American Medical Colleges