The President’s New Freedom Commission on Mental Health (1) identified several goals for reforming mental health services and improving public views of persons with serious mental illness. This report also addresses the need for primary care providers to further their own knowledge regarding the treatment of mental illness, emphasizes the importance of conducting research on the current system of mental health service delivery, and recommends that both persons in recovery and their families be actively involved in each stage of treatment.
A recent study (2) has documented many impediments to the partnership between family members of persons with serious mental illness and mental health service providers in developing and implementing treatment plans. The barriers faced by service providers and families tend to have a negative influence on family members’ involvement in the mental health care of the person in recovery (3) and, on occasion, the relationship between the person in recovery and his or her family members. In their survey of members of the National Alliance for the Mentally Ill, Marshall and Solomon (4) recently found that only 36 percent of persons in recovery were encouraged by their mental health service provider to engage a family member in their treatment. Thus, it appears that service providers are not promoting an active role for family members in treatment, despite a documented relationship between structured family involvement and recovery for individuals with psychiatric disabilities (5).
The present study examines several additional issues relevant to the recommendations of the New Freedom Commission on Mental Health. With medical school educators conceptualizing the etiology and treatment of mental illness as primarily biological, while services and support to persons with mental illness are now predominantly community-based, we attempted to determine how familiar an unselected population of medical students participating in a six-week psychiatric rotation was with current thinking on causes of mental illness and the "best practices" in psychiatric treatment. Given the documented absence of collaborative treatment efforts, we also saw a clear need to assess the attitudes of future physicians towards the role of the family in the treatment of relatives with psychiatric disabilities.
Specifically, we hypothesized that following a rotation in psychiatry, medical students would report stronger beliefs in biological (compared to social) causes for mental illness, the effectiveness of medication and psychotherapy in the treatment of mental disorders, the importance of including families of persons in recovery in treatment decisions (given the students’ exposure in the rotation to representatives of community-based organizations promoting an integrative treatment approach), and greater interest in considering psychiatry as a specialty than they did prior to the rotation.
Participants in this study were 955 third-year medical students at the Indiana University School of Medicine. Participation was completely voluntary and anonymous and was approved by the Indiana University-Purdue University Indianapolis-Clarian Institutional Review Board. A survey was administered to five cohorts of medical students, before and after their six-week psychiatric rotation, from the summer of 1997 to the summer of 2001. Although 955 students participated in at least one part of the survey, the current analyses only include 672 students whose identification numbers allowed us to match them on the pre- and postsurveys.
Four hundred eleven (61%) of the participants were male and 261 (39%) were female; 84% identified themselves as Caucasian, 10% as Asian, 3% as African American, and 3% of other ethnic backgrounds. Sixty nine percent of the respondents were between 18 and 25 years of age while 31% were 26 or older. χ2 analyses revealed that students who only completed one administration of the survey (either pre- or postrotation) were more likely to be male (χ2 = 4.7; p = 0.03). Specifically, while 61% of all respondents completed both waves of the survey, the group completing only the pretest or posttest was 73% and 71% male, respectively. No other significant differences between students who completed both surveys and those who completed one were found with regard to age or race.
We created a survey to understand the medical students’ attitudes towards the causes of mental illness, the role of the family in serious mental illness, and the field of psychiatry and treatment for mental illness. In addition, questions were asked about the medical students’ familiarity with the Alliance for the Mentally Ill (AMI) and the Key Consumer Organization, a self-help organization for individuals with serious mental illness. The survey was administered during an orientation session at the start of the psychiatry rotation and again following the final examination. To match their pre- and postsurveys, students selected a unique identifying code (that could not be directly linked to them) that was written on both the pre- and postsurvey. The survey was voluntary, and failure to complete both waves probably reflected fatigue following the final exam and inaccurate recall of the arbitrary code entered on the pretest. The survey took approximately five minutes to complete, and we obtained written consent from all participants.
The medical school junior clerkship in psychiatry we studied was six weeks in length. Students’ specific experiences during the clerkship varied due to the large number of medical students at this school (280 per class) and the number of different hospitals (six) and treatment facilities that are integral parts of the school's clinical efforts. Within the experience, junior students were assigned a major rotation on which they spent the majority of their time, usually an inpatient service, and a minor rotation that consisted of one half-day per week in an outpatient clinic. There was a very active attempt to have both adult and child psychiatry exposure as part of this experience, although this was possible for only two-thirds of the students. There were also six didactic areas covered during afternoon seminars, including a session with families from the National Alliance for the Mentally Ill (NAMI), discussions of personality disorders, child psychiatry, geriatric psychiatry, psychopharmacology, and a review of the mental status examination. In addition, students were assigned to several half-days of chemical dependency treatment, which included both lectures and exposure to self-help groups.
The first part of this survey required the respondent to rate his or her degree of agreement with a series of statements regarding the importance of seven possible causes of serious mental illness (individual items will be discussed in the subsequent data reduction section). The second part consisted of eight items reflecting the health professional’s identification of responsibilities associated with interacting with families of people with psychiatric disabilities. Responses generated for items on both of these sections were assessed on a Likert scale, with scores ranging from 1 (very important) to 4 (not at all important). The third part of this survey consisted of seven statements about psychiatry and mental health as a treatment field, with students’ agreement assessed using a Likert scale by scores ranging from 1 (strongly agree) to 5 (strongly disagree).
A principal components analysis, with varimax rotation, was carried out to identify subscales for the analysis of medical students’ understanding of the causes of mental illness. Reducing the data of the first section in this manner produced three factors with eigenvalues greater than one accounting for 30%, 18%, and 15% of the total variance, respectively. As a group the seven items in the scale demonstrated weak internal consistency (Cronbach alphas ranging from 0.56—0.58), and so for each of the three factors a subscale was created composed of the items that loaded highly on the particular factor. We labeled the first subscale Social Causes, as it included three items identifying external/environmental influences as the cause of mental illness (having parents who are inconsistent in the way they treat their children, having too much social pressure, and excessive use of drugs or alcohol). The maximum and minimum combined score of these three items is twelve and three, respectively; a score of three gives maximum importance to the social causes of mental illness. We labeled the second subscale Biological Causes, as it consisted of two items addressing the role of biological systems as a cause of mental illness (imbalance of chemicals in the body or brain, and inheriting a gene that causes a mental disturbance). The maximum and minimum combined score of these two items is eight and two, respectively; a score of two gives maximum importance to biological causes of mental illness. We labeled the third subscale as Chance Causes of mental illness, as it included two items reflecting one’s belief that "fate or luck" and "the will of God" are causes of mental illness. The maximum and minimum combined score of these two items is eight and two, respectively; a score of two gives maximum importance to chance causes of mental illness.
Responses to the eight items in the second section, attitudes toward the interaction of health professionals with families of people with serious mental illness, showed good internal consistency as a single scale (Cronbach alpha = 0.78). The maximum summed score of this scale is thirty-two (indicating the least favorable attitudes toward the interaction of the health professional with family members in the treatment of persons with mental illness); a score of twenty indicates the point where attitudes toward the interaction of the health professional with family members are theoretically neutral.
Finally, the items in the third section of the survey, attitudes towards psychiatry, did not have strong internal consistency (Cronbach alphas approached zero); therefore we examined each item individually.
+
Pre- and Postrotation Results
Results of paired sample t-tests (see t1) revealed that students possess stronger beliefs in both biological and social causes of mental illness following the psychiatric rotation. Significant pre- and postdifferences in two items assessing students’ understanding of the literature on treatment for mental illness revealed that their recognition of the proven effectiveness of psychotherapy and psychotropic medications increased after the rotation. After the rotation students were significantly more likely to agree that the cost of psychiatric treatment is worthwhile. Following the rotation students disagreed significantly more with the statement that persons with mental illness do not get better, regardless of treatment. Interestingly, however, students agreed more after the rotation that treating persons who abuse alcohol or drugs is a waste of time. Significant pre- and postchanges were found in students’ attitudes toward the settings in which mental health treatment is provided. After the rotation students become more aware of the benefits of multifaceted treatment efforts that extend beyond the hospital. Finally, the students indicated that they were more familiar with the AMI and the KEY Consumer Organization after the rotation.
There were no significant pre- and postdifferences in students’ attitudes towards chance causes of mental illness or in their attitudes towards the role of the family in the treatment and recovery of relatives with mental illness. The psychiatric rotation did not have a significant impact on the students’ willingness to consider a career in psychiatry.
Results indicated that participation in a psychiatry rotation strengthened medical students’ beliefs in biological and social causes of mental illness. These results are all the more striking given that students strongly endorsed these beliefs even before completing the rotation. Thus, even with the strong emphasis on the biological underpinnings of illness experienced by students in medical school, they recognize environmental factors that could potentially play a role in the onset of mental illness. Students’ responses show that they attribute the etiology of mental illness to both biological and environmental/psychological factors, as previous research has found (6).
This rotation in psychiatry also appears to promote students’ interest in a more collaborative role in managing treatment using input and support from consumers, family members, and other service providers and may go some way toward reducing the kinds of professional barriers that prompted the President’s New Freedom Commission to issue its report. These students were early in their medical training and supervision, however, so we should not automatically assume that these positive attitudes are permanent. Nevertheless, such positive attitudes towards blending natural and formal support by encouraging family involvement in mental health treatment should not go unrecognized. In addition, after the rotation students were more likely to indicate that they were familiar with the AMI and the KEY Consumer Organization. The combination of favorable attitudes towards the role of the family and increased awareness of the AMI and consumer organizations suggests that these future medical practitioners are more likely to collaborate with important stakeholders, such as families, than previous generations of physicians have been.
After the rotation medical students’ recognition of the empirical research establishing the effectiveness of both psychotherapy and psychotropic medications was strengthened. This probably reflects students’ exposure to different treatment modalities for psychiatric disorders, both during their hands-on experience and their concurrent enrollment in the psychiatric courses and didactics that accompanied the rotation. With such awareness of the effectiveness of these well-researched treatment modalities, physicians will be better able to recommend multiple treatment options for persons with mental illness.
Following the rotation students agreed less with the notion that people with mental disorders do not get better and that the treatment of mental illness costs more than it is worth. The psychiatric rotation was therefore successful in developing more favorable attitudes towards the prognosis of persons with mental illness. Such an increase in positive attitudes is especially important given that the rotation was only six weeks long. Moreover, despite the fact that the students’ major rotation was in a hospital inpatient setting, they recognized that the best way to treat mental disorders is not necessarily in a hospital. Coupled with the finding that students are very open-minded and welcoming when it comes to working with families of persons with mental illness, we would expect these students, as future medical practitioners, to place more emphasis on community-based treatment than traditional hospital services. Participation in the psychiatry rotation did not have any effect on students’ consideration of psychiatry as a specialization, which was surprising given that such an effect had been found in previous studies (7, 8).
We did not expect to find that following the rotation students were more likely than before to see treating persons with alcohol or substance abuse as a waste of time. However, most of these students were in inpatient settings where the incidence of comorbid substance disorders was 80 percent or more, and students also spent time on-call, where they encountered persons in crisis with substance abuse complications who had been through multiple treatments and/or refused further treatment. We speculate that the overwhelming comorbidity of their patients and the paucity of good short-term treatment results are what discouraged these students about substance abuse treatment.
A limitation of this study is the absence of a long-term follow-up to see if changed attitudes towards mental illness caused by the psychiatric rotation remained after graduation from medical school. Although Sivakumar et al. (9) found that medical students’ more favorable attitudes towards psychiatry following an eight-week clerkship were not maintained two years later, this finding was obtained in England almost twenty years ago. Therefore, continued effort should be devoted to studying the long-term effects of psychiatric rotations on health professionals’ attitudes towards the causes and treatment of mental illness.
Finally, although significant differences were plentiful in this study, the modest effect sizes for most of them make evident the need for a replication study. In addition, no evaluation regarding students’ perceptions of how the rotation impacted their attitudes toward mental illness was obtained, nor was any information gathered on the students’ previous experience with mental illness, either academically or personally. Information of this nature might have provided a basis for predicting possible changes in the attitudes of particular subsets of medical students.
In light of the recommendations of the New Freedom Commission on Mental Health, the limitations of this study suggest the need for a more in-depth exploration of medical students’ attitudes, particularly those entering primary care, toward the treatment of mental illness. Effort should also be devoted to identifying the impact of psychiatry rotations on students’ attitudes toward the role of stakeholders, including families, in the treatment of mental illness. Nevertheless, this study does extend our current understanding of attitudes towards mental illness to a large number of recent medical students in the Midwestern United States. During the years this study was conducted, 50 to 60 percent of the graduates of this medical school went on to become primary care doctors. Their generally favorable attitudes indicate that, at least immediately before and after the psychiatric rotation, medical students acknowledge their responsibility to become informed of the best practices associated with working with persons who have mental illness.
This study was supported by the Indiana Consortium for Mental Health Services Research. We thank Bernice Pescosolido, Terry White, Harold Kooreman, Eric Wright and Jaya George for their contributions to this project, and anonymous reviewers for suggestions regarding the manuscript.