Influence of educational programs on attitudes of medical students towards psychiatry: Effects of psychiatric experience, gender, and personality dimensions

Abstract

Background: Attitudes of medical students form the basis for medical actions. Because of the specific characteristics of psychiatric patients, positive attitudes of medical students towards psychiatry should be a higher goal in medical education.

Aim: We hypothesize that medical students in different educational programs develop different attitudes towards psychiatry.

Methods: In a cross-sectional study, students enrolled in different educational programs completed the ‘attitudes towards psychiatry’ questionnaire (ATP-30). Data concerning experiences in psychiatry, personality traits and socio-demographic variables including gender were also analyzed.

Results: The response rate of students in the PBL-curriculum (n = 61) was >90%, in the traditional curriculum (n = 280) >75%. Attitudes towards psychiatry of male students in the Problem-Based Learning program were equal to the female students’ attitudes in both programs. Female students’ attitudes in the traditional curriculum reached comparably good results while male students’ displayed the worst attitudes.

The personality factors ‘openness to experience’ and ‘agreeableness’ correlated significantly with positive attitudes towards psychiatry. PBL-students showed significantly more ‘openness to experience’.

Conclusion: Educational programs might play a role for the development of attitudes towards psychiatry, especially in male students. Factors influencing enrollment into special educational programs should also have been taken into account. An independent study with a larger number of participants will be required to support these findings.

Introduction

Medical students’ attitudes towards psychiatry and psychotherapy as well as towards the mentally ill form the basis of their later actions as medical doctors (Eaton & Goldstein 1977) and have been studied since the 1960s (Moos & Yalom 1966). Previous research has shown that students’ attitudes are directly related to the patient's management at the level of primary care (Eaton & Goldstein 1977). Furthermore, students’ attitudes influence their choice of medical specialty. The recruitment of psychiatric trainees also depends on their positive view of this specialty (Pan et al. 1990; Weissman et al. 1994). Attitudes towards psychiatry are supposedly formed during the undergraduate training. To measure attitudes instruments are necessary which disclose a certain quality standard and can be used easily. These requirements were accomplished in 1982 with a test for ‘Attitudes Towards Psychiatry’ (ATP-30) developed by Burra et al. (1982). This validated self-assessment questionnaire allows comparisons between universities, different educational systems and countries.

Table

Practice points

Previous psychiatric experiences are associated with more positive attitudes towards psychiatry – this should be considered in curriculum development. Studying in a problem-based program might affect students’ attitudes towards psychiatry positively. Gender aspects have to be taken into account to improve attitudes towards psychiatry while planning a curriculum – male students in traditional curricula showed the least positive attitudes towards psychiatry.

In the past there have been three main focuses of medical education research in psychiatry:

1. Students’ attitudes towards psychiatry and patients with mental illnesses: with the help of the ATP-30 it has been demonstrated that attitudes towards psychiatry are in general moderately positive (McParland et al. 2003) and improve after clinical exposure to patients with psychiatric problems (Burra et al. 1982; Sloan et al. 1996; Strebel et al. 2000). Attitudes of female students were more positive than those of male students (Baptista et al. 1993; Kuhnigk et al. 2007).

2. Number of medical school graduates who wish to specialize in psychiatry: A decrease of interest in psychiatry has been reported internationally in medical students since the 1960s from approximately 10% (Albee 1959) to currently approximately 2–5% (Ghadirian & Engelsmann 1982; Malhi et al. 2003; Baboolal & Hutchinson 2007; Kuhnigk et al. 2007). The percentage of students considering psychiatry as a future career increases after undergraduate training in psychiatry (Bulbena et al. 2005). Furthermore, it has been shown that attitudes are related to speciality; yet, it still remains unclear whether professional socialization reinforces existing attitudes or whether existing attitudes result in speciality preference (Batenburg et al. 1999).

3. Effects of medical curriculum on attitudes towards psychiatry: A positive change in attitudes towards psychiatry in medical students could be shown most often after their exposure to psychiatry and completion of psychiatric courses (Burra et al. 1982; Chung & Prasher 1995; Sloan et al. 1996). However, this effect declined significantly over time during undergraduate training (Baxter et al. 2001; Altindag et al. 2006). Positive changes took place among students whose training consisted primarily of clinical aspects of psychiatry with a considerable number of contact hours with patients. Similar results were found by McParland et al. (2003) in two consecutive classes (semesters) of fourth-year medical students before and after establishing a new Problem-Based Learning (PBL) curriculum and before and after an 8-week course in psychiatry, respectively. Both classes showed a significant positive change of attitudes towards psychiatry and an increase of students nominating psychiatry as a field of specialty after having completed the 8-week course. In an additional study, McParland et al. (2004) could not find a correlation between studying in a PBL curriculum and attitudes towards psychiatry and drew the conclusion that the content of courses was more important than the educational program itself to improve students’ attitudes towards psychiatry. However, attention was drawn to the fact that changes in curricular programs towards PBL were performed within ongoing training and that students had already studied for 3 years in a traditional curriculum. Singh et al. (1998) could also not find differences in attitudes towards psychiatry with regard to different educational programs or gender.

The aim of our study was to examine the attitudes of medical students towards psychiatry in different years of undergraduate training in different educational programs (PBL vs. traditional subject-orientated program). The specificity of influencing variables such as previous experience with psychiatry, gender aspects, and personal interest in psychiatry as a subject was comparatively analyzed. Furthermore, by using the five-factor model of personality (NEO-FFI), we analyzed whether any of the ‘main five personality factors’ differed between student groups and if they correlated with positive attitudes towards psychiatry (Borkenau 1992; Borkenau & Ostendorf 1993). In addition, a potential connection between attitudes towards psychiatry and intended career as psychiatrist was evaluated.

Methods

The 30-item questionnaire ‘Attitudes Towards Psychiatry-30’ (ATP-30) was developed by Burra et al. (1982) to study attitudes towards various aspects of psychiatry on the basis of a 5-point Likert scale. This scale measures attitudes towards mental illnesses, psychiatric patients, psychiatric institutions and psychiatrists, teaching, knowledge, and career choice. The higher the score, the more favorable the attitude towards psychiatry (minimum score 30 = very negative attitude; maximum score 150 = very positive attitude; score of 90 = computed neutral attitude value). The ATP-30 has shown good validity and reliability (Burra et al. 1982). The German version of the ATP-30 (Strebel et al. 2000) is based on an expert translation by three psychiatrists involved in clinical and scientific work as well as a translation by a professional interpreter (medical doctor, English as first language). In addition to the ATP-30 socio-demographic characteristics of the students (age, gender, etc.), their experience with psychiatry (contact with mentally ill relatives or patients during their studies or personal experience) was also documented. In addition, the probability of working in psychiatry in the future was recorded by the use of a visual analog scale from 0 to 8. Furthermore, the five factor model of personality (McCrae & Costa 1987) was applied in the German version (Borkenau & Ostendorf 1993) to study students’ dimensions of personality.

Sample and design

In Germany, the undergraduate medical courses comprise 12 semesters with each semester covering 14 weeks. At the Medical Faculty of the University of Hamburg 61 students were enrolled in a mere PBL program. A total of 84 interdisciplinary PBL cases based on somatic diseases were used within 3 years by groups of seven students each. Forty students of semester two (cohort No. 2: n = 40) and 21 students of semester four (cohort No. 1: n = 21) were included in this study. The number of students per semester differs because of ‘teething troubles’ with cohort No. 1 and a consecutive high drop out from the PBL program. At the Medical Faculty of the University of Duisburg-Essen 280 students enrolled in a traditional curriculum were included in this study, 200 students from semester two (n = 200) and 80 students from semester five (n = 80). The number of students per semester differs because admission for pre-clinical education (semesters one to four) takes place only in the winter term, whereas admission for clinical education (semesters five to twelve) takes place twice a year (Kuhnigk et al. 2007). Within the first four semesters of the traditional curriculum, science and basic medical knowledge were taught subject centered. Within the PBL curriculum (semester one to six) science, basic medical knowledge as well as clinical skills (physical examination) were taught with an interdisciplinary understanding and on the basis of patient case reports. Psychosocial learning objectives were an inherent part of this curriculum and relevant to exams. After six semesters students from the PBL curriculum at Hamburg University were integrated into the traditional program where courses in psychiatry took place.

The cover page of the questionnaire included all information about the exact purpose of this study. Questionnaires were filled out during mandatory courses of each semester and were distributed at the beginning of each course and immediately completed by the students within 15–20 minutes. In Hamburg O. Kuhnigk was available for questions during this time and collected the questionnaires after completion. Participation in the study was voluntary and anonymous.

Statistical analysis

All correlations were measured as Pearson Product Moment Correlations. The significance of differences between arithmetic means was investigated by analysis of variance. The a priori hypothesis that PBL students had a more positive attitude towards psychiatry than students from a traditional curriculum was tested with reference to the significance level of 0.05 as recommended (Bortz & Döhring 1995; Bortz 1999). The significance level for a posteriori hypotheses was corrected for multiple comparisons (Bonferroni, 11 statistical tests) and was set to p ≤ 0.0045.

Results

Fifty-seven students of the PBL program in Hamburg (93% return quota) and 241 students of the traditional program in Duisburg-Essen (86% return quota) returned the ATP-30 questionnaire. Sixty-five percent of the students in the PBL program and 64% in the traditional program were females. In Hamburg the average age of students was 22.7 years and 23.2 years in Duisburg-Essen.

Because there was no significant difference between the ATP-30 sum of scores from different semesters at both universities, all further calculations comprise pooled data from all students of one university (Table 1). Attitudes towards psychiatry and psychotherapy were found to be generally positive in both educational programs, since results in ATP-30 sum of scores in both groups were found to be above the calculatory mean of 90. However, ATP-30 cumulative scores of students within the PBL program were significantly higher (p < 0.045) than those of students within the traditional program. In order to evaluate the influence of sample size on these results the effect size was calculated. The effect size for this difference was ε = 0.30 and can be characterized as small to moderate (Bortz & Dohring 1995). Forty-nine percent of the students in the PBL program and 39% of the students in the traditional program said that they had previous experience with psychiatry or psychotherapy (practical training, night watch, personal experience, mentally ill relatives). A total of 5.7% of the students in the PBL program and 5% of the students in the traditional program nominated psychiatry as their preferred future field of speciality (Table 2).

Table 1.  Return rate of questionnaires and attitudes of medical students (sum of scores) towards psychiatry (ATP-30), (ATP-30 sum of scores: 30 = very negative attitude, 150 = very positive attitude)

	PBL program Medical Faculty of Hamburg University			Traditional program Medical Faculty of Duisburg-Essen University
	2nd semester	4th semester	Total	2nd semester	5th semester	Total
Number of distributed questionnaires	40	21	61	200	80	280
Return rate	38 (95%)	19 (90.5%)	57 (93.4%)	181 (91%)	60 (75%)	241 (86.1%)
ATP-30 Sum of scores S ± SD	111.7 (±15.5)	110.5 (±7.2)	111.3 (±13.3)	106.9 (±13.8)	108.8 (±11.9)	107.3 (±13.4)

Table 2.  First choice for students' future field of specialty

		PBL program (Medical Faculty of Hamburg University)			Traditional program (Medical Faculty of Duisburg-Essen University)
Rank	Field of specialty	Desired field of specialty (n = 53)	Rank	Field of specialty	Desired field of specialty (n = 219)
1	Pediatrics	26.4% (n = 14)	1	Surgery	20.1% (n = 44)
2	Internal medicine	15.1% (n = 8)	2	Pediatrics	17.4% (n = 38)
3	Surgery	13.2% (n = 7)	3	Anesthetics	8.7% (n = 19)
4	Gynecology	9.4% (n = 5)	4	Gynecology	7.8% (n = 17)
4	Neurology	9.4% (n = 5)	5	Internal medicine	5.9% (n = 13)
5	Psychiatry	5.7% (n = 3)	6	Neurology	5.5% (n = 12)
5	Orthopedics	5.7% (n = 3)	7	Psychiatry	5% (n = 11)
5	General medicine	5.7% (n = 3)	7	General medicine	5% (n = 11)
		*Others: (<5%; n = 5)	7	Orthopedics	5% (n = 11)
					Others: (<5%; n = 43)

Note: *Others (desired field of specialty <5%).

Female and male students in the PBL program had similar positive ATP-30 scores (n.s.). Female students in the traditional program had similar scores to PBL students, but higher scores compared to their male colleagues (p < 0.001) (Table 3). Similar results were noted concerning the effect of gathered experience in psychiatry. Students with experience in psychiatry studying in a traditional program showed significantly higher ATP-30 sum scores. This difference does not appear in students studying in a PBL program. Within the traditional program men differ greatly from women (ε = 0.60) whereas the difference within the PBL program is marginal and not significant (ε < 0.10). Differences between students with and without experience in psychiatry studying in a traditional program were of medium effect size (ε = 0.38) while differences concerning PBL students were small (ε = 0.14).

Table 3.  Attitudes of medical students towards psychiatry (ATP 30) according to gender and psychiatric experience

University	Statistical values	Comparison groups		Significance
		Female students	Male students
PBL program (Medical Faculty of Hamburg University)	n/percent ATP-30 sum of scores SD	37/65%	20/35%	n.s.
		111.7**	110.5**
		13.7	12.8
		With*	Without*
	n/percent ATP-30 sum of scores SD	Previous psychiatric experience		n.s.
		28/49%	29/51%
		110.4**	112.2**
		16.9	8.7
Traditional program (Medical Faculty of Essen-Duisburg University)	n/percent ATP-30 sum of scores SD	Female students	Male students	p < 0.001
		152/63%	89/37%
		110,3**	102.2**
		12.5	13.2
		With*	Without*
	n/percent ATP-30 sum of scores SD	Previous psychiatric experience		p < 0.002
		96/39%	149/61%
		110.5**	105.4**
		14.0	12.6

Note: Responses expressed as percentage (*Do you have any previous psychiatric experience (e.g. clinicaltraineeship, night watch, personal experience, mentally ill relatives, etc.?), ATP-30 sum of scores (**30 = very negative attitude; 150 = very positive attitude; S = sum values; SD = standard deviation).

Results of the personality questionnaire NEO-FFI showed a significant positive correlation of high scores on the scale of openness to new experiences and compatibility with high ATP sum scores in students from both educational programs (Table 4).

Table 4.  Correlation of attitudes towards psychiatry (ATP-30 sum of scores) with the personality field, openness and agreeableness (NEO-FFI means) from students of the PBL curriculum in Hamburg and of the traditional curriculum in Duisburg-Essen

	PBL curriculum Medical Faculty of Hamburg University (n = 57)	Traditional curriculum Medical Faculty of Duisburg-Essen University (n = 240)
ATP-30	Attitudes towards psychiatry	Attitudes towards psychiatry
NEO-FFI
Openness to experience	r = 0.518*, p ≤ 0.0001	r = 0.245*, p ≤ 0.0001
Agreeableness	r = 0.575*, p ≤ 0.0001	r = 0.268*, p ≥ 0.0001

Note: r = Pearson's Product Moment Correlation, p = level of significance, NEO-FFI-Scale: 0 = strongly disagree, 4 = strongly agree, ATP-30: five-point Likert Scale with 1 = strongly agree and 5 = strongly disagree/ATP-30 sum of scores: smallest negative value = 30, highest positive value = 150; *direction of correlation: the higher openness to experience/agreeableness the more positive attitudes towards psychiatry.

PBL students displayed significantly higher marks concerning openness to experience (M = 2.82, SD = 0.47) than their colleagues in the traditional program (M = 2.56, SD = 0.51; Table 5). The effect size for ‘openness to experience’ necessary to identify significance was ε = 0.53. Following Bortz & Döhring (1995) this is the equivalent of a medium effect (Bortz, 1999). Within the categories ‘agreeableness’, ‘neuroticism’, ‘extraversion’, and ‘conscientiousness’ no notable differences were detected.

Table 5.  Students' personality dimensions based on the NEO-FFI (Score: 0 = strongly disagree, 4 = strongly agree, M = means, SD = standard deviation)

	PBL curriculum (Medical Faculty of Hamburg University)	Traditional curriculum (Medical Faculty of Duisburg-Essen University)
	n = 57	n = 245
NEO-FFI-scale	M/SD	M/SD	p
Neuroticism	1.66/0.60	1.66/0.66	n.s.
Extraversion	2.48/0.47	2.50/0.51	n.s.
Openness to experience	2.82/0.47	2.56/0.51	<0.0007
Agreeableness	2.74/0.49	2.56/0.48	n.s.
Conscientiousness	2.65/0.56	2.69/0.57	n.s.

Students with a high school diploma mainly focusing on humanities showed the highest ATP-30 scores in both groups (data not shown). Students with a high school diploma mainly focusing on sciences showed lower ATP scores irrespective of the educational program at university.

With respect to the probability of choosing a career in psychiatry it was found for students of both educational programs that the more positive students evaluate the ATP-30, the more likely they will decide to work in psychiatry later. There is a high positive correlation between the two variables ‘ATP-30’ and ‘career in psychiatry’. A similar result was found for the correlation between the variables ‘ATP-30’ and ‘interest in psychiatry’. The probability to work in psychiatry in the future was generally very low for both student groups. However, students from the traditional program rate the probability to work in psychiatry lower than their colleagues from the PBL program (Table 6). Both groups showed positive results having personal interest in psychiatry as a subject. The students from the PBL program rated the item more positive than their colleagues from the traditional program (Table 6).

Table 6.  Coefficient of correlation of the attitudes towards psychiatry (ATP-30 sum of scores) with the items, working in psychiatry and interest in psychiatry and students' probability of working in psychiatry and their personal interest in psychiatry as a subject

		Study participants
Topic	Statistical analysis	PBL program (Medical Faculty of Hamburg University) (n = 57)	Traditional program (Medical Faculty of Duisburg-Essen University) (n = 245)
Choice of profession	Probability of working in psychiatry in the future? Mean/SD	4.8*/1.9	5.3*/2.3
	Correlation: ATP-S/Probability of working in psychiatry	| r | = 0.50 (p < 0.0001)**	| r | = 0.59 (p < 0.0001)**
Interest in specialty	Personal interest in psychiatry as a subject? Mean/SD	2.5*/1.3	3.3*/2.1
	Correlation: Personal interest in psychiatry as a subject/Probability of working in psychiatry	| r | = 0.50 (p < 0.0001)**	| r | = 0.59 (p < 0.0001)**

Note: (r = Coefficient of correlation based on Pearson, p = Level of significance); Probability of working in psychiatry (‘How probable is it that you will work in the field of psychiatry in the future?’) and Personal interest in psychiatry as a subject ('Please rate your personal interest in psychiatry as a subject’) (*Visual analogue scale: 0 = very probable/very high; 8 = very improbable/very poor; SD = standard deviation). **Direction of correlation: the better the attitude towards psychiatry (ATP), respectively, the higher the interest in the subject, the higher the possibility to work in psychiatry.

Discussion

One of the key questions of this study was whether a change of attitudes towards psychiatry can be achieved at medical school by a specific educational program. In summary, all medical students participating in the PBL curriculum at the University of Hamburg and in the traditional curriculum at the University of Essen-Duisburg had only moderate positive attitudes towards psychiatry independent of the educational system. This result generally confirms former research (Burra et al. 1982; Alexander & Eagles 1990; Sloan et al. 1996; Strebel et al. 2000). However, students studying in the PBL curriculum showed a significantly more positive attitude towards psychiatry than students in the traditional curriculum despite the low test power. Despite the complexity of the variable the medium effect size was 0.30. Although these differences can be mainly attributed to the educational program, it has to be taken into account that they could also have been caused by differences within the samples. Approximately 50% of the PBL students had gathered experiences in psychiatry, within the traditional curriculum only 39% of the students. The positive attitude towards psychiatry (independent of experiences) and ‘more frequently gathered experiences in psychiatry’ indicate a ‘special selection’ of students in the PBL curriculum. Furthermore, students from both curricula differed in the personality trait ‘openness to experiences’. The fact that students with high openness to new experiences are also more open towards mentally ill patients and towards psychiatry seems plausible, a result which could be demonstrated in previous research (Kuhnigk et al. 2007). With respect to further studies investigating attitudes towards psychiatry matching for these variables would be recommended.

In the traditional educational program, female students showed significantly better attitudes towards psychiatry than their male colleagues, a phenomenon also observed by others (Alexander & Eagles 1986, 1990; Fabrega et al. 1994; Kuhnigk et al. 2007). Female students almost regularly assess psychosocial disciplines more positively than their male colleagues (Alexander & Eagles 1990; Fabrega et al. 1994). Relational-emotional aspects tend to be of greater importance to female doctors, which is probably a reflection of the specific role that women still play within our society (Goerg et al. 1999). Female students of the traditional curriculum showed ATP-scores comparable to all students of the PBL curriculum irrespective of gender. Interestingly, female and male students of the PBL curriculum showed no difference. A comparable result could not be found in the literature which raises the question whether the educational PBL system might be the cause for this difference. Male students of the PBL curriculum showed unusually high ATP-scores and therefore more positive attitudes towards psychiatry than so far described in the literature. This could be due to either the pre-selection of a group with specific interest in being a potential PBL student or the influence of the educational program with its psychosocial learning objectives. Within the traditional curriculum these learning objectives played a minor role.

Although PBL curricula with a focus on psychosocial topics still remain an exception in Germany, it can be assumed in general that the rising percentage of women in medical training and the high interest of female students in the subject of psychiatry will generate a positive influence on the number of students choosing psychiatry as their profession (Chung & Prasher 1995). A reason why male students in a traditional curriculum show a less positive attitude towards psychiatry could be seen in a widely held but wrong belief by students that this specialty is less scientific and offers fewer clinically effective treatments than other subjects (Malhi et al. 2003). Since male students more often have a high school diploma focussing on sciences, they might be led by this wrong belief in rating their attitudes towards psychiatry more negatively than female students. Furthermore, other variables such as national or university specific factors, like career opportunities, prestige, income, and the condition of further medical specialization have to be taken into account. They play a key role in later career choice (Firth-Cozens et al. 1999). Since experiences gathered during medical training greatly influence which field of specialty is chosen (Feifel et al. 1999), universities need to design their curricula very carefully.

We were able to demonstrate for the first time that even in early semesters of a traditional medical curriculum, personal experience in the area of psychiatry plays a key role with respect to attitudes towards psychiatry and seems to influence them positively. Students with previous experiences in psychiatry studying in a traditional curriculum revealed significantly more positive attitudes than students without these experiences. This difference could not be demonstrated within the cohorts of students studying in a PBL curriculum. An a posteriori comparison of students without these experiences indicates that students in a PBL curriculum show most significantly more positive attitudes towards psychiatry than students from a traditional curriculum. In summary, literature findings concerning attitudes of medical students in traditional curricula do not apply to students in PBL curricula. The question of additional influencing factors should therefore be investigated irrespective of curricular programs.

Further variables correlating with attitudes towards psychiatry were found in the area of personality and qualification. ‘Openness to experience’ correlates positively with attitudes towards psychiatry and students with more scientifically based high school diplomas tend to have less positive attitudes towards psychiatry. Whether these students need special interventions concerning the subject of psychiatry during their medical studies should be further considered. The aim of such interventions would be to develop a certain degree of open-mindedness for the integrative medical care of mentally ill patients rather than becoming a psychiatrist. Although students from both curricula showed relatively high interest in psychiatry as a specialty, the probability of later working in this field was low. This dissonance between positive attitudes towards psychiatry and the choice of psychiatry as a potential career was also observed in other studies (Ndetei et al. 2008). It has to be noted that both groups of students were in their pre-clinical medical training. The slightly higher interest in psychiatry of students in PBL curricula as well as the slightly higher probability to work in this field requires additional studies. Also the small sample size of PBL students within this study needs to be taken into account. Irrespectively, psychiatry as a future field of specialty was at the bottom of the list. However, consolidated knowledge about psychiatric disease patterns and the need of positive attitudes towards psychiatry and mentally ill patients is a basic requirement for all medical doctors irrespective of career aspirations, since a 40% comorbidity rate for psychiatric problems is seen by general practitioners, half of which remain unrecognized (Kruse et al. 1999).

Some restrictions must be taken into account in the interpretation of the results of this study. It is a ‘paper and pencil’ and a cross-section investigation within the university and at two different universities. On the basis of international comparability and test statistical possibilities we favored a questionnaire based on the ATP-30, even though this test instrument appears to be outdated in some parts. Items like ‘Psychiatrist seem to talk about nothing but sex’ (item 7, ATP-30) appear not to be up-to-date anymore. On the basis of our high return quota for an empirical questionnaire (75–95%) the results are representative. A possible critique of the design of our study could be that the control group (traditional curriculum) was based at the University of Duisburg-Essen. Since Bargel & Ramm (1994) did not find any relevant differences in traditional medical curricula at German universities, because of compulsory state exams this is no problem. Up to now there are very few studies available showing connections between attitudes of medical students towards psychiatry and educational programs at medical schools. All were conducted as student surveys at the beginning and end of psychiatric teaching units (Singh et al. 1998; McParland et al. 2003; Reuster et al., 2003). In contrast to these surveys our study was conducted without any context to psychiatric teaching units. Hence, attitudes of students from both curricula seem to be based on self-directed learning, lectures, and seminars in other fields of specialty or additional environmental influences. Whether positive attitudes towards psychiatry of students in the PBL curriculum existed before the start of medical training or whether they developed in the course of medical training cannot be answered conclusively. All PBL students at the Faculty of Medicine in Hamburg had actively decided to join the PBL track. As PBL students showed higher scores in the personality area of ‘openness to experiences’ measured with the NEO-FFI (positive correlation with ATP-sum of scores) these personality traits seem to have a supportive effect on the results.

In summary, male students in a traditional curriculum and without pre-existing psychiatric experiences as well as students with less openness to new experiences had a less positive attitude towards psychiatry. Irrespective of the educational program and despite verbalized interest in psychiatry, all students at the beginning of their medical training rate the probability of later work in the field of psychiatry as rather low. As it was shown repeatedly that attitudes towards psychiatry improved after teaching units in this subject the variables responsible for this effect need to be identified and applied to medical courses (Guttmann et al. 1996; Sloan et al. 1996). Visible success of psychiatric therapy and active participation in therapy seem to be important factors (McParland et al. 2003). Our data also show that gender aspects have to be taken into account to improve the attitudes towards psychiatry in male students. As improvements in students’ attitudes towards psychiatry in general decline again months after completion of psychiatry training (Burra et al. 1982; Wilkonson et al. 1983; Creed & Goldberg 1987), training in psychiatry should appear throughout the whole curriculum. Psychosocial principles and psychiatric aspects were included in the interdisciplinary PBL cases from semester one in the PBL curriculum in Hamburg. In addition, practical psychiatric experiences seem to be highly beneficial since it has been demonstrated that direct contact with mentally ill patients is of greater effect in medical laymen with respect to change of attitudes than didactic concepts (Wolf et al. 1996). According to our data this is also valid for medical students in a traditional curriculum. Students should therefore be continuously confronted with knowledge-based psychiatry as well as hands-on experience in psychiatry throughout medical training. Negative attitudes towards psychiatry based on incorrect assumptions about psychiatry could be counterbalanced this way (Feifel et al. 1999). In addition, the promotion of students with a special interest in the field of psychiatry should be enhanced.

Acknowledgments

The authors thank the ‘Bundesministerium für Bildung und Forschung’, the ‘Behörde für Wissenschaft und Gesundheit’ and the University of Hamburg for funding of the PBL curriculum ‘Modellstudiengang Medizin’. Special thanks go to Dr Strebel, J. Schilauske, and Prof. Jüptner who provided comparative data of students from the University of Duisburg-Essen. We would especially like to thank the medical students of the Universities of Hamburg and Essen for their participation in the study.

O. Kuhnigk and M. Bullinger designed the study. C. Haufs collected and computerized the data under the supervision of A.M. Böthern and O. Kuhnigk. All authors were involved in the data analysis and in writing the manuscript.

The study was funded by the ‘Bundesministerium für Bildung und Forschung’, Germany, the ‘Behörde für Wissenschaft und Gesundheit’, Hamburg, and the Medical Faculty of the University of Hamburg.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Additional information

Notes on contributors

O. Kuhnigk

OLAF KUHNIGK, MD, MME, is Head of the Office of the Vice-Dean of Education and Resident in Psychiatry at the Medical Faculty of Hamburg University, Germany.

M. Hofmann

MARZELLUS HOFMANN, MME, is Head of the Office of the Vice-Dean of Education of the PBL Reform Curriculum at the Private University Witten/Herdecke, Germany.

A.M. Böthern

AENNE M. BÖTHERN, PhD, is Psychologist in the Office of the Vice-Dean of Education at the Medical Faculty of Hamburg University, Germany.

C. Haufs

CHRISTIN HAUFS, MD, is Resident in Internal Medicine, St. Elisabeth Hospital, Damme, Germany.

M. Bullinger

MONIKA BULLINGER, PhD, is Professor of Medical Psychology and Vice Director of the Institute for Medical Psychology at the Medical Faculty of Hamburg University.

S. Harendza

SIGRID HARENDZA, MD, MME, is Assistant Professor for Internal Medicine at the Medical Faculty of Hamburg University, Germany.