Attitudes towards General Practice: a comparative cross-sectional survey of 1^st and 5^th year medical students

Abstract

Objective: Positive attitudes towards General Practice can be understood as a prerequisite for becoming a General Practitioner (GP) and for collaboration with GPs later on. This study aimed to assess attitudes of medical students at the beginning and the end of medical school.

Methods: A total of 160 1^st year students at Hannover Medical School were surveyed. Their attitudes were compared to those of 287 5^th year students. Descriptive, bi- and multivariate analyses were performed to investigate influences of year of study and gender.

Results: Year of study and gender both were associated with the attitudes towards General Practice. The interest in General Practice and patient-orientation (communication, care of older patients with chronic diseases) was higher in 1^st year students compared to more advanced students. Female students valued such requirements more than male students, the differences in attitudes between the years of study being more pronounced in male students.

Conclusion: Despite some limitations caused by the cross-sectional design, the attitudes towards General Practice competencies changed to their disadvantage during medical school. This suggests a formative influence of the strategies used in medical education. Educational strategies, however, could be used to bring about a change of attitudes in the other direction.

Introduction

The shortage of general and rural doctors is often reported. In order to identify possible causes, the training and the registration requirements which could be deterring potential candidates are regularily scrutinised. As a consequence, the training in many areas has been restructured via the creation of vocational training collaborations (Weiterbildungsverbünde) with fixed rotations from the beginning, which are supported by structured, content-related seminars [1], [2], [3].

In this country, the causes of primary care talent shortage have less often been studied in medical school itself. From abroad, it is known that role models for potential career paths are important during this period [4], [5]. Also early and, if possible, prolonged contact with general practice [4], [6] can positively influence the choice towards primary care training. However, differences in higher education and health systems limit the transferability of these findings from abroad. A more recent survey regarding the career plans of a large sample of medical students in Baden-Württemberg (n=1,299) describes the importance of concrete activity- and personality-related factors [7], such as an interest in patient orientation, which was associated with the career choice of family doctor.

This publication focuses more on the basic attitudes of medical students towards general practice. Attitudes (so-called affective learning goals) together with knowledge and practical skills (including communication) represent one of the three educational objectives in the study of medicine [8]. Attitudes are already "brought" from school and previous experience but these change during study through the continuing maturation of these young adults, but can also be consciously shaped through the use of intended learning goals by the faculty. A positive attitude towards general practice, is taken as a prerequisite for the choice of a general practice career goal in this study [9], [10] and is a desirable educational goal given the need for subsequent inter-professional collaboration. For these reasons, the attitudes of medical students at the MHH in their 1^st and 5^th academic year were sought for comparison.

Methods

Study design, participants and procedure

For the study, a survey of attitudes to general practice was undertaken in the form of a written questionnaire. For this publication, students in the 1st academic year at MHH were surveyed before their first lectures during the “Introduction Week” in October 2011. These results are compared to previously published results of students surveyed during their 5th academic year in the period 2008-2010 [11]. We therefore present a comparative cross-sectional study, as the same individuals were not followed over time.

In the 1^st academic year, a total of n=160 participants were recruited for the current survey (participation rate: 58.4%). Recruitment took place in the auditorium by directly addressing the students at the beginning of a lecture during the introduction week briefing on aspects of general practice and the Skills Lab at MHH (a training center for medical skills) by the author CK. Following a brief explanation of the survey students were requested to first complete the questionnaires in silence (about 10 minutes). Questionnaires were then collected and the lecture continued.

The survey of n=287 participants in the 5th academic year is described in detail elsewhere [11]: The respondents were from a total of 5 trimesters from 2 consecutive academic years (participation rate: 67.8%). Questionnaires were always distributed at the beginning of the block practical in general practice as part of the introduction held in the auditorium as well as a few weeks after the end of the block placement for pre-/post-comparison. Only the baseline data from the 5^th year survey (ie, before the primary care training module and the accompanying seminar series) were used for comparison with the survey from the 1^st academic year.

The study design for the 5^th academic year was submitted and approved by the chairman of the ethics committee of the MHH prior to the survey (No. 611). The current survey of 1^st academic year students was practically identical and the only slightly modified questionnaire was also presented to the Dean of Studies (field of evaluation and capacity) and was approved as a survey into specific aspects of the programme of medical teaching under § 2, paragraph 1c of the evaluation procedure.

Design of the questionnaire

The questionnaire and its development is discussed in detail in [11]. The questionnaire included a number of socio-demographic characteristics such as age, sex, origin (eg country, city), and details of the Abitur final grade, previous qualifications and planned field of study (see Table 1 (Tab. 1)). The question regarding experience in general practice (eg electives) was omitted from this current survey as it was designed for first year students.

Table 1. Socio-demographic characteristics of the study participants.

The questionnaire also included another 39 real "attitude items". After an extensive literature review some questionnaires previously only used abroad were identified [11], examined, compared and a selection of relevant items made. These were translated into German and simultaneously culturally adapted. Of these translations, reciprocal translations into English were also prepared in order to identify possible distortions of meaning.

Compared to the survey given to the 5^th academic year students, only the item regarding perceived self-efficacy was omitted as this was not appropriate for first-year students. All items were reported on a five-point Likert scale (1="strongly disagree" to 5="strongly agree"). As some questions were considered difficult to understand for first year students the current survey also added a "do not know" option.

This publication presents a selection of particularly relevant attitude items in tablular format. Choices were made for reasons of space and content overlap. A factor analysis was carried out in order to identify and group similar items [11]. This analysis identified 13 factors - as shown in Table 2 (Tab. 2) that reflected the different aspects of the questionnaire. These form the basis for the selection of items presented in Tables 3 (Tab. 3) and 4 (Tab. 4), with as many aspects of the questionnaire presented. The focus was kept specifically on the possible attitudes of the medical students in terms of concrete primary care working situations. As an example, this was not applicable for the item relating to electronic patient records [factor = 11], which is therefore not shown here.

Table 2. Factors and number of the corresponding items selected for Table 3 and 4.

Table 3. Attitudes regarding the discipline of general practice in the 1^st and 5^th years of study, presented as median ± the standard deviation from the Likert scale values (from 1 [do not agree] to 5 [agree]). In brackets is the corresponding factor is indicated (see Table 2).

Table 4. Multivariate analysis of the association of gender and year of study (independent variables) with attitudes. Bold = Items from Figure 1.

Statististical analyses

Socio-demographic characteristics and all other data were analysed using descriptive statistics (mean, standard deviation, percentage).

Attitude items were analysed and reported using descriptive statistics (represented by the mean of the Likert scale scores). The results of the 1^st and 5^th academic year were compared by means of non-parametric Mann-Whitney U tests. For this statistical comparison of individual items, a Bonferroni correction was applied because of the multitude of tests conducted. For n=39 items a p value less than 0.05/39=0.00128 was interpreted as statistically significant.

In a multivariate analysis, the gender of the participants was considered alongside the academic year (1^st or 5^th) as an independent variable and the association with the attitudes (dependent variable) is shown. This was done using an ordinal logistic regression reporting odds ratios (OR) and 95% confidence intervals (CI). The influence of the simultaneous, independent influence of gender and year was also illustrated (see Figure 1 (Fig. 1)). The different types of influence (A to D) arose as a summary of the findings of the bi- and multivariate analyses.

Figure 1. Differences of attitudes depending on year of study and gender (presented as the mean of 1 [do not agree] to 5 [agree]). Examples from Tab. 3 and 4.

Results

Characterisation of the study participants

By comparison to the 5^th academic year, students surveyed in their first year of study were consistently younger (average 21 versus 26 years) and had slightly better school grades (mean average grade of 1.6 versus 1.8). The students were slightly more likely to come from small towns rather than big cities and were more often undecided regarding the choice of their future specialisation (about half in the 1^st year versus a third in the 5^th year of study) - 11% (1^st year) and 13% (5^th year) wanted to pursue general practice, including those with professional intentions of internal medicine without specialisation. A detailed overview of the socio-demographic data of the study participants is given in Table 1 (Tab. 1).

Differences in attitude between the 1^st and 5^th years of study

Significant differences (defined as p<0.001) were found in 8 (20.5%) of the 39 items requested. A selection of the items, identified according to the criteria described in the Methods section, are shown in Table 3 (Tab. 3).

Already in the 1^st year, general practice was not counted amongst the three most fascinating subjects (mean Likert score 2.50). By the 5^th year this was still the case, but significantly less so (2.15). Nevertheless, experience in general practice were considered highly valuable for the medical profession, although less so in the 5^th academic year than in the 1^st (4.54 and 4.05). Items relating to psycho-social and conversational skills were viewed with greater openmindedness in the 1^st academic year, while such aspects were considered less interesting by the 5^th year.

Similar opinions were also reflected in items not presented here. For example, the question of whether the treatment of acute symptoms is more interesting than the care of chronically ill patients (1^st year: 3.09; is more applicable in the 5^th year: 3.39, p=0.009). Also, the care of geriatric patients interested 5^th year students rather less than those of the 1^st academic year (1^st year: 3.26; 5^th year: 3.70, p=0.001).

Coordination and filter functions of primary care physicians were rated less important by the younger students. The interest in research was less pronounced in the 5^th than in the 1^st year academic year.

Differences in attitude between men and women

The attitude towards general practice of men and women in the 1st academic year was comparable (Examples A, B and D in Figure 1 (Fig. 1)). In the 5^th year however statistically significant differences in attitudes between the sexes were frequently found (see results published elsewhere [11]). The cause of these differences between the cohorts was mainly a result of the different attitudes of men, whereas women's attitudes in the 1^st and 5th academic years were more often similar. This aspect is evident in the results of the multivariant analysis shown in Table 4 (Tab. 4), in which factors such as "gender" and "academic year" were taken into account simultaneously. ORs with CI, not including the "1", suggest independently significant associations between gender and / or academic year with the measured attitudes.

This association is shown as an example for some items and is illustrated in Figure 1 (Fig. 1). With a "gap" (B, D) or with more distant but parallel curves (C), a gender effect is apparant in addition to a time effect represented byincreasing/decreasing curves. The gender effect involved many of the substantial conversational and psycho-social aspects, including the care of chronically ill patients.Men in the 5^th academic year found this aspect less important than men in their 1^st academic year, an effect that was significantly more pronounced compared to women.

Discussion

Summary of the main results

This survey of medical students presents a comprehensive picture of a variety of attitudes of first year students to general practice and the relevant associated competencies. Compared to another cohort of students in their 5^th year of study, first year students were, in principle, more open and showed more interest in the associated skills. The difference between the years was often caused by the subgroup of men, in whose fifth years greater deviation was evident compared to the women.

Comparison with the literature

Fundamentally, this discussion is based on the assumption that a positive attitude towards general practice is a crucial prerequisite for the intended career of general practicioner [9], [10]. Attitudes are thus understood to represent mouldable educational objectives of medical studies [5], [8], [9]. The following discussion therefore focuses on those determinants of career choice, which could be influenced by medical education.

Existing values present upon entry into the study of medicine can correlate with the subsequent career choice of GP. This is reflected by the expectation of a high income, research interest (negatively correlated, respectively) and a favourable assessment of the importance of primary care physicians (positively correlated) [4]. In part, candidate selection during admission interviews focusing either on social skills or an interest in basic scientific research could influence the choice of students that are more or less inclined to general practice.

Fundamentally, the extent of general practice teaching (particularly in the third and fourth years) correlates with the decision to pursue the subject [4], [6]. Generally, real practical experiences in primary care are valuable, in order to favourably affect attitudes towards the subject and at the same time encourage those to become a general practicioner [11], [12], [13], [14], [15]. This supports current efforts to extend the prescribed time devoted to the general medical training module to two weeks [16]. It is interesting that the attitudes towards GPs reported in two studies from the United Kingdom improved between the 1^st to 5^th academic year - contrary to our data [9], [15]. This could be interpreted as an indication of the traditionally more strongly rooted role of general practice in a health system organised around primary care. However, for a more differentiated judgement, it would be necessary to compare curricula with respect to the extent of primary care-oriented teaching (see above). In addition, the attitudes of the faculty and the teachers must be considered, given the unfavourable effect generated by negative comments made about the specialisation [4], [17].

Assumptions regarding the characteristics of a subject, for example, the fear of being intellectually understimulated [4], [18] or a poor reputation [4], also play a role in career considerations. This suggests that students during training should be well informed regarding the profession of general practice and misconceptions should be corrected and positive role models provided [19], [20]. However, our data suggests that feeling undervalued or even reduced earnings was not a significant argument against the choice of a general practice, it was however more generally true for men than women. At 5 sites surveyed in Germany, students’ “career ambitions“ (with respect to research) was less important amongst those who wanted to be GPs. However, "patient orientation" (interest in long term relationships with patients and preventative medicine) played a greater role [7] (also for example [21]). Such attitudes examined by us in greater detail seem to develop to the disadvantage of general practice towards the end of the course of study.

Open-mindedness, particularly towards psychosocial aspects of medicine [21], unfortunately was less pronounced as students approached their final practical study year. Longer periods of training in general practice as described above serve to regularly nurture and promote distinctive characteristics and interests of medical students in favour of a career in primary care. This was also the case in our first pre-/post-questionnaire (before and after the 3-week general practice clerkship) in the 5^th academic year [11]. The training module changed things with respect to greater openness and had a pronounced effect particularily among male students. This is especially promising with regard to the variability of attitudes through training interventions.

Strengths and weaknesses

Overall, data from nearly 450 students have entered into the analysis described here. Although not a multicentre survey (as currently in Spain [22]), it was conducted at very different stages of training: at the beginning and in the penultimate year of study. This allowed analysis of possibly changing attitudes over almost the entire theoretical education. The differences between years could be due to a priori differences in attitude in the surveyed cohorts. However, excluding the age difference which was of course linked to progression through study, charcteristics such as gender, school grade and origin showed only small differences between groups. Ultimately, the present cross-sectional study using a historical comparison group can only provide an indication, which needs to be reviewed longitudinally.

On the positive side is the use of items already proven and partially validated abroad, which have been refined in multiple review processes [11]. The survey, based on a large number of items showed a coherent picture (face validity). However, a simplification and formal validation of our German questionnaire is still pending.

Conclusions

Our empirical study provides strong evidence of different attitudes of male and female medical students at the beginning and end of their training. This conclusion is limited by the cross-sectional design of the survey and should be examined longitudinally. During the course of study, positive views of general practice should be delivered in order not only to promote the career choice of the subject but also the subsequent inter-professional cooperation. This applies to efforts such as continuous training in psychosocial skills [23] and having the highest proportion possible of the general practice in the course e.g. in the form of longitudinal scale programmes [15], [24] such as the primary care Witten-Herdecke adoption programme [25]. It will be necessary to evaluate the extent to which such measures could also help to modify attitudes in Germany and to generate more recruits for general practice.

Acknowledgement

We thank all medical students that participated in the survey.

Competing interests

The authors declare that they have no competing interests.