Both students and residents were asked about six specific conditions—hypertension, cardiovascular disease, insomnia, headache disorders, low-back pain, and arthritis—and their perceptions regarding the value-added if behavioral/mind–body methods were included as part of the overall treatment for these conditions. The majority of both students (465, or 70%) and residents (380, or 69%) stated that there would be large value added (four or five on a five-point scale) if behavioral/mind–body methods were included as part of the treatment for insomnia, whereas 395 students (60%) and 333 residents (62%) stated that the use of such methods would add large value to the treatment of headache disorders (See Table 1). The number indicating that there would be large value added was considerably lower for the remaining conditions:
- ▪ low-back pain (240 residents [45%], 282 students [43%]);
- ▪ hypertension (210 residents [39%], 287 students [44%]);
- ▪ cardiovascular disease (149 residents [28%], 198 students [30%]);
- ▪ arthritis (128 residents [24%], 167 students [26%]).
Perceptions about training
A minority of both students (171, or 26%) and residents (125, or 23%) indicated that their formal training was helpful (score of four or five on a five-point scale) in learning how to address psychosocial factors in treatment, whereas 284 students (43%) and 267 residents (49%) rated the quality of their training in these areas as not helpful (score of one or two on the five-point scale) (see Table 1). When specifically asked about the quality of mentoring they had received, a majority of medical students (421, or 64%) and a smaller majority of residents (294, or 54%) indicated that mentors did a good job with respect to training them in diagnosing psychosocial factors, whereas a significantly smaller percentage of students (249, or 38%) and residents (162, or 30%) stated that mentors did a good job with respect to training them in treating psychosocial factors.
To assess students’ and residents’ future intentions with respect to incorporating behavioral/mind–body methods, we asked them to comment on their level of interest in receiving further training in these areas, and the extent to which they felt committed to adopting behavioral/mind–body approaches in their future clinical practice. A minority of both students (217 or 33%) and residents (150 or 27%) expressed high or very high interest (three or four on a four-point scale) in obtaining additional training (see Table 1). The remaining reported either moderate (two on the four-point scale) (294 or 45% students; 233 or 42% residents) or low (one on the four-point scale) (150 or 23% students; 166 or 30% residents) interest in receiving further behavioral/mind–body training. There was also considerable variability regarding how committed students and residents were to adopting behavioral/mind–body approaches clinically. Approximately one third (207 or 31%) of students and two fifths (233 or 42%) of residents indicated that they were either “not very committed” or “not at all committed” to the clinical adoption of behavioral/mind–body methods.
Perceptions of barriers
Respondents were also asked about factors that might be serving as obstacles to the further integration of psychosocial factors and behavioral/mind–body methods in medicine. Among respondents, the most frequently cited barrier was lack of training with 444 residents (81%) and 554 students (84%) indicating that this was contributing greatly to the failure of medicine to address psychosocial factors and behavioral/mind–body methods. The second most frequently cited item was lack of time, with 429 residents (84%) and 482 students (74%) stating that this was serving as a significant barrier. Additional barriers were
- ▪ physician discomfort with the feelings of uncertainty that can arise when addressing psychosocial factors (356 residents [65%], 437 students [67%]);
- ▪ appeal of the “quick fix” (355 residents [65%], 455 students [69%]);
- ▪ inadequate reimbursement for these methods (339 residents [62%], 415 students [63%]);
- ▪ physicians’ lack of expertise in behavioral/mind–body approaches (317 residents [60%], 382 students [58%]);
- ▪ lack of evidence for behavioral/mind–body methods (208 residents [40%], 265 students [42%]);
- ▪ reluctance of physicians to examine the role of psychosocial factors in their own health (211 residents [38%], 212 students [32%]);
- ▪ physicians’ need to maintain a sense of control (180 residents [33%], 197 students [30%]);
- ▪ lack of patient willingness to address psychosocial factors (187 residents [34%], 178 students [27%]); and
- ▪ lack of support from peers (198 residents [38%], 132 students [21%]).
We also examined whether there were any significant differences in what male and female residents considered to be the most significant barriers to the integration of psychosocial factors and behavioral/mind–body methods. Male residents were more likely (P < .05) to state that the following were barriers: absence of evidence of these therapies’ effectiveness; the belief that when drug therapies are used alone they are just as effective; lack of acceptance by professional peers; and, feeling that owing to the nature of their practice, it would not be appropriate to address psychosocial factors. Female residents, on the other hand, were more likely to say that insufficient clinical time to address psychosocial factors issues was serving as a significant barrier.
As shown in Table 3, for both students and residents, the following factors emerged as significant predictors of attitude (P < .05): (1) being female, (2) use of behavioral/mind–body methods to manage their own health, and (3) perception that medical school training in these areas had been helpful. Students’ and residents’ health-related behaviors (i.e., smoking, alcohol consumption, and exercise) were not correlated with either attitude toward or practice of behavioral/mind–body methods.
Together, these factors accounted for 36% of the variability in student and 40% in resident attitudes regarding the need to address psychosocial factors.
To better understand the nature of the relationships among the above predictors and respondents’ attitudes toward the role of psychosocial factors, we examined cross-tabulations. Averaging across the 550 female students and residents, 402 respondents (73%) indicated that the addition of psychosocial factors and behavioral/mind–body methods would lead to marked (moderate to large) improvements in treatment and diagnosis compared with 363 (55%) of the 657 male respondents. A significantly higher percentage of female students and residents (179, 64%) of the 280 who reported using behavioral/mind–body methods in practice also indicated that they find engaging with patients about psychosocial factors very satisfying compared with 117 (50%) of the 334 male students and residents who answered the question (see Table 1). Female students and residents were also nearly twice as likely as male students and residents—40% versus 22%—to express interest in receiving further training in behavioral/mind–body methods (see Table 1).
Among 339 residents who indicated that they used one or more behavioral/mind–body methods to manage their own health, 252 (75%) stated that such methods would lead to moderate or large improvements in treatment outcomes, whereas among those 209 residents who did not use behavioral/mind–body methods for their own personal health, 120 (57%) felt that such methods would result in improved outcomes. A similar pattern emerged for medical students. Among those 458 students who used behavioral/mind–body methods to manage their own health, 372 (81%) stated that such methods would lead to moderate or large improvements in treatment outcomes, whereas among 201 students who did not personally use behavioral/mind–body methods, 130 (65%) felt that such methods would result in improved medical outcomes.
Among those 277 residents who rated their formal training as not helpful, 166 (60%) stated that the inclusion of behavioral/mind–body methods would lead to either moderate or large improvements in treatment outcomes, whereas for those 125 residents who felt their training in addressing psychosocial factors was helpful, 109 (87%) felt that behavioral/mind–body methods would lead to improved treatment outcomes. A similar pattern emerged for medical students. Of those 284 students who indicated that their training was ineffective, 185 (65%) stated that the addition of psychosocial factors and behavioral/mind–body methods would significantly improve treatment outcomes, whereas among the 169 students who said their training was beneficial, 150 (89%) said that such methods would lead to improvements in treatment.
We examined whether year graduated from medical school was related to residents’ perceptions of the quality of training they had received with respect to addressing the impact of psychosocial factors in health. Somewhat surprisingly, there was a negative, albeit small, correlation (r = −0.10; P = .02) between these variables, suggesting that residents graduating more recently experienced their medical school and residency training in these areas to have been less, rather than more, helpful.
Predictors of practice
Similar to those factors that predicted attitude, the following variables were associated with frequency of use of, and/or referral to, both psychological counseling and relaxation techniques among residents (n = 550): (1) perception that formal training in these areas, during both medical school and residency, had been helpful, (2) use of behavioral/mind–body methods to manage their own health, and (3) indication that their religious/spiritual beliefs were important to them in their work. As was the case with attitude toward addressing psychosocial issues, gender (being female) was significantly correlated (P = .03) with more frequent use of psychological counseling. However, the effects of this variable became statistically nonsignificant when perceived quality of training was entered into the regression equation (see discussion section for a potential explanation of this finding).
Among the 277 residents who indicated that their training in addressing psychosocial issues was poor, 110 (40%) stated that they never use or refer patients for psychological counseling, whereas only 24 of these residents (8%) indicated that they often use or refer to counseling. In contrast, among 125 residents who indicated their training in these areas was helpful, 14 (11%) stated that they never use or refer patients for counseling, whereas 57 (46%) indicated that they often do. Similarly, the 125 residents who rated their educational training in psychosocial factors more favorably were considerably more likely (15% compared with 4%) to say that they often use or refer patients to relaxation therapies.
Among the 146 residents who reported that religious/spiritual beliefs were very important to them in their work, 36 (25%) indicated they often use or refer to counseling, and 16 (11%) to relaxation. In contrast, the 185 residents who stated their spiritual beliefs were unimportant were significantly less likely to use such methods (29 [16%] for counseling, and 7 [4%] for relaxation).
This study sought to examine medical students’ and residents’ attitudes toward the role of psychosocial factors and behavioral/mind–body methods in medicine. Several themes emerged.
In this sample, a substantial number of both students and residents appear to recognize the need to address psychosocial factors and the potential value of incorporating behavioral/mind–body methods in the treatment of common medical conditions. For example, 40% to 60% of respondents indicated that there would be large value-added if approaches such as psychological counseling and other behavioral/mind–body therapies (e.g., relaxation, meditation, guided imagery) were included as part of the treatment of insomnia, headache, low-back pain, and hypertension.
However, despite this fairly widespread recognition of the benefits when psychosocial factors are addressed, many students and residents expressed skepticism regarding the benefit of integrating psychosocial factors into medical diagnoses and treatment. For example, approximately one in three students and two in five residents stated that the inclusion of psychosocial factors and behavioral/mind–body methods would result in either small or no improvement in medical diagnoses and treatment. Similarly, when asked about their level of commitment to adopting behavioral/mind–body approaches, approximately one-third of students and two in five residents indicated that they were either not very, or not at all, committed to such adoption.
Our multivariate analyses identified several factors that were associated with both medical student and resident attitudes and practice patterns. First, women were considerably more likely to recognize the need to address psychosocial factors. Compared with male respondents, female students and residents were not only more likely to say that they enjoyed engaging patients in discussions about psychosocial factors but also appeared more interested in receiving further training in these areas (see Table 1). Second, both students and residents who reported using behavioral/mind–body methods to manage their own health were significantly more likely to believe that such methods would be useful in the treatment of patients.
Similar to our previously reported findings with physicians,9,12,13 students’ and residents’ perceptions of the quality of their formal training in behavioral/mind–body methods and the role of psychosocial factors in health also emerged as a significant predictor of attitude. Those who reported that the training and mentoring they received in these areas were subpar were significantly less open to the need to address psychosocial factors. Not surprisingly, residents’ perception of the quality of training was also predictive of their actual use of psychosocial factors and behavioral/mind–body methods in clinical practice. For example, those who felt more positive about the psychosocial training they had received were almost four times as likely to report frequent use of, or referral of patients to, psychological counseling as were those residents who felt their training was less effective.
Along with being more likely to indicate that addressing psychosocial factors is necessary in medical diagnosis and treatment, female residents also tended to be more knowledgeable about behavioral/mind–body therapies and, to a lesser extent, more likely to report using or referring patients to these approaches. However, in our multivariate analyses, the effects of gender on practice became nonsignificant in the presence of the variable quality of training, suggesting that female residents’ greater tendency to use psychosocial factors and behavioral/mind–body approaches may be explained in part by their more favorable impressions of the training they had received in these areas.
In an effort to identify additional barriers, we asked both students and residents to indicate the extent to which they felt that certain factors were serving as obstacles to medicine’s incorporating psychosocial factors and behavioral/mind–body approaches. Lack of training and insufficient time emerged as the most frequently cited obstacles. Additional factors included the appeal of the “quick fix,” physicians’ personal discomfort with addressing psychosocial factors, inadequate reimbursement for these methods, and to a lesser extent, lack of evidence for behavioral/mind–body methods, reluctance of physicians to examine the role of psychosocial factors in their own health, physicians’ need to maintain a sense of control, patient unwillingness to address these factors, and lack of support from peers.
The findings that females are more likely to recognize the need to address psychosocial factors, are more committed to adopting behavioral/mind–body approaches in clinical practice, and express greater interest in receiving additional training in these areas are intriguing. Although speculative, the fact that there was a higher response rate among female students may be reflective of woman’s greater interest in the subject matter of the survey. In the follow-up, in-depth interviews we are currently conducting (as part of the same study reported here), physicians and physicians-in-training have hypothesized about these findings. Although only speculative at this point, it is possible that women may be socialized and/or biologically more predisposed to value relationships, emotional nurturance, interpersonal communication, and the importance of subjective experience, and may therefore be more likely to believe in the need to address the interior (psychosocial) dimension of patients’ lives. In addition, it is possible that women are more likely to pursue specialties (e.g., family practice) that afford them the opportunity and require them to engage more frequently in relationship-based interactions with their patients.
First, it is difficult to determine the extent to which the attitude and practice patterns we identified can be generalized to the larger population of medical students and residents in the United States. For example, it is possible that our lower-than-anticipated response rate (40%) may have reflected some general lack of interest in or enthusiasm for the topic among the sample of students and residents we initially contacted. Although we cannot be certain, given this response rate, it is possible that students and residents in the general population may actually be less likely to address psychosocial factors than the group who responded to our survey, and that our findings therefore represent an overestimate of the degree of interest in and openness to these factors.
Second, the cross-sectional nature of the study precludes our ability to draw definitive conclusions regarding the direction of causation between study variables. For example, as noted, a central finding was that negative perceptions regarding the quality of training in addressing psychosocial issues was predictive of residents and students being less open to the need to address such factors in clinical care. However, it may be the case that, rather than quality of training influencing attitude and practice patterns, believers (in addressing psychosocial factors) may simply rate their training in these areas as more effective as a result of already holding positive attitudes toward this area. Although certainly possible, it also seems equally, if not more, plausible that those who are more interested in and open to the role of psychosocial factors in health would, in turn, have higher, rather than lower, expectations regarding the quality of their medical training in these areas. This latter interpretation would argue for our initial supposition: namely, that education is a likely driver of both attitude and behavior relative to addressing psychosocial factors in medicine.
Our finding that students as well as residents tend to rate their training in psychosocial factors (both in terms of diagnosis and treatment) as inadequate lends strong support to the recent Institute of Medicine report,1 which, as noted, emphasized the need to improve the quality of medical school and residency training to address the complex interplay of biological, psychological, and social factors and their influence on human physiology and health.14,15
The importance of the improvement of medical education in these areas is further supported by our finding that quality of training predicted both attitude and practice such that those students and residents who indicated that the psychosocial training and mentoring they had received was helpful were significantly more likely both to acknowledge the need to address psychosocial factors and to use behavioral/mind–body methods in clinical practice. The need to improve training and education in addressing psychosocial factors was further highlighted by our finding that for both students and residents, lack of physician training was the most frequently cited obstacle to the appropriate integration of psychosocial factors and behavioral/mind–body methods in medicine. Finally, the fact that residents who graduated from medical school more recently tended not to rate their training in psychosocial factors and behavioral/mind–body methods as more effective than those who had graduated before them raises questions about the effectiveness of recent efforts to improve training in these areas. Given these findings, we believe it may be very useful for the Association of American Medical Colleges to convene a Medical School Objectives Projects focused on training in behavioral/mind–body methods.
As noted earlier, a significant predictor of both attitude toward the role of psychosocial factors and the likelihood that residents would actually use behavioral/mind–body approaches in clinical practice was the use of such methods to care for their own health. This finding suggests that as part of educating students and residents about the potential value of addressing psychosocial factors and incorporating evidence-based, behavioral/mind–body methods, it may be important to make the training in these areas experiential as well as didactic, providing physicians-in-training opportunities to experiment with such methods in the “laboratory” of their own personal lives as a precursor to introducing patients to them. This idea is supported by studies suggesting that interventions designed to change physician behavior are most effective when they are not merely didactic in nature but also include both active participation and direct experience.16 Several U.S. medical schools have begun incorporating behavioral/mind–body practices into the curriculum. For example, since 2002, Georgetown University has been offering an elective to first- and second-year students that introduces them experientially to various behavioral/mind–body methods (e.g., relaxation, imagery, meditation) to increase self-awareness and self-care. Their initial findings suggest that this 22-contact-hour course is effective in increasing students’ empathy, self awareness, ability to cope with stress, and regard for addressing the mental–emotional well-being of patients as part of clinical care.17,18
Finally, the finding that lack of time as well as inadequate reimbursement were cited by the vast majority of students and residents as significant barriers to the incorporation of psychosocial factors and behavioral/mind–body methods suggests that our current health care delivery system may, in many respects, be antithetical to the biopsychosocial model. As we noted in our previous report summarizing physicians’ attitudes toward the role of psychosocial factors,13 medicine as it is currently practiced, particularly within a managed care context, may, at best, be suboptimal precisely because the very way in which care is structured limits the extent to which practitioners are actually able to adequately address the psychosocial dimension of patients’ lives, either diagnostically, or in terms of the actual treatment strategies they employ.
Although it has been some 30 years since Engel19 first published his seminal paper on the importance of expanding the model of medicine to encompass not just biological but psychological and social factors as well, our findings suggest that medical training and practice have a considerable way to go before the biopsychosocial model moves from the realm of the theoretical to the actual. On the basis of our findings and those of other researchers (e.g., Waldstein et al3), we believe it is incumbent on medical educators and health care policy makers to ensure that the training and practice of medicine continue to reflect the ever-growing body of scientific evidence that points to the significant influence of psychosocial factors on human health and well-being.
1 Cuff PA, Vanselow N. Committee on Behavioral and Social Sciences in Medical School Curricula. Improving Medical Education: Enhancing the Behavioral and Social Science Content of Medical School Curricula. Washington, DC: Institute of Medicine, National Academies Press; 2004.
3 Waldstein SR, Neumann SA, Drossman DA, Novack DH. Teaching psychosomatic (biopsychosocial) medicine in United States medical schools: survey findings. Psychosom Med. 2001;63:335–343.
4 Kiecolt-Glaser JK, McGuire L, Robles TF, Glaser R. Psychoneuroimmunology: psychological influences on immune function and health. J Consult Clin Psychol. 2002;70:537–547.
5 Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation. 1999;99:2192–2217.
6 Salovey P, Detweiler JB, Steward WT, Rothman AJ. Emotional states and physical health. Am Psychol. 2000;55:110–121.
7 Astin JA, Shapiro SL, Eisenberg DM, Forys KL. Mind-body medicine: state of the science, implications for practice. J Am Board Fam Pract. 2003;16:131–147.
8 Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA. 1999;282:1458–1465.
9 Sierpina V, Levine R, Astin J, Tan A. Use of mind-body therapies in psychiatry and family medicine faculty and residents: attitudes, barriers, and gender differences. Explore J Sci Heal. 3:129–135.
10 Bero LA, Grilli R, Grimshaw JM, Harvey E, Oxman AD, Thomson MA. Closing the gap between research and practice: an overview of systematic reviews of interventions to promote the implementation of research findings. The Cochrane Effective Practice and Organization of Care Review Group. BMJ. 1998;317:465–468.
11 Wolsko PM, Eisenberg DM, Davis RB, Phillips RS. Use of mind-body medical therapies. J Gen Intern Med. 2004;19:43–50.
12 Astin JA, Goddard T, Forys K. Barriers to the integration of mind-body medicine: perceptions of physicians, residents and medical students. Explore J Sci Heal. 2005;1:278–283.
13 Astin JA, Soeken K, Sierpina VS, Clarridge BR. Barriers to the integration of psychosocial factors in medicine: results of a national survey of physicians. J Am Board Fam Med. 2006;19:557–565.
14 Baum A, Posluszny DM. Health psychology: mapping biobehavioral contributions to health and illness. Annu Rev Psychol. 1999;50:137–163.
15 Kiecolt-Glaser JK, McGuire L, Robles TF, Glaser R. Emotions, morbidity, and mortality: new perspectives from psychoneuroimmunology. Annu Rev Psychol. 2002;53:83–107.
16 Davis DA, Thomson MA, Oxman AD, Haynes RB. Changing physician performance. A systematic review of the effect of continuing medical education strategies. JAMA. 1995;274:700–705.
17 Haramati A, Paul G, Vincenzi F. Use of mind-body skills to decrease anxiety and increase self-awareness, self-care and empathy in medical students. Paper presented at: AAMC Annual Meeting; Seattle, Wash; 2006.
18 Tractenberg RE, Chaterji R, Haramati A. Assessing and analyzing change in attitudes in the classroom. Assess Eval High Educ. 2007;32:107–120.
19 Engel GL. The need for a new medical model: a challenge for biomedicine. Science. 1977;196:129–136.
*A small number of respondents opted not to report gender. For this reason, the values and percentages shown in Table 1 (which compares and contrasts attitudes by gender) do not always match the attitudes and practice patterns reported for the sample as a whole.
†In reporting these frequencies throughout this section, percentages are based on the number of respondents who answered each particular questionnaire item.© 2008 Association of American Medical Colleges