Comparing three experiential learning methods and their effect on medical students’ attitudes to learning communication skills

Abstract

Background: Despite numerous studies exploring medical students’ attitudes to communication skills learning (CSL), there are apparently no studies comparing different experiential learning methods and their influence on students’ attitudes.

Aims: We compared medical students’ attitudes to learning communication skills before and after a communication course in the data as a whole, by gender and when divided into three groups using different methods.

Method: Second-year medical students (n = 129) were randomly assigned to three groups. In group A (n = 42) the theatre in education method, in group B (n = 44) simulated patients and in group C (n = 43) role-play were used. The data were gathered before and after the course using Communication Skills Attitude Scale.

Results: Students’ positive attitudes to learning communication skills (PAS; positive attitude scale) increased significantly and their negative attitudes (NAS; negative attitude scale) decreased significantly between the beginning and end of the course. Female students had more positive attitudes than the male students. There were no significant differences in the three groups in the mean scores for PAS or NAS measured before or after the course.

Conclusions: The use of experiential methods and integrating communication skills training with visits to health centres may help medical students to appreciate the importance of CSL.

Introduction

A number of successful outcomes have been linked to provider–patient communication: patients’ satisfaction with care, compliance with recommended treatment and understanding and recall of information (Ong et al. 1995; Brown et al. 2003). Therefore, young doctors are expected to communicate effectively with patients, the patients’ significant others and colleagues. There are three challenges in communication skills training: medical students need to understand the theory of communication (knowledge); they need to have a positive attitude towards communication skills (attitude); and they need to be trained to command a repertoire of specific communication skills in their daily practice (skills) (Kaufman et al. 2000). In this study we evaluate Finnish medical students’ attitudes to learning communication skills in three groups using different experiential methods: theatre in education (TIE), simulated patients (SPs) and role-play (RP).

Attitudes are commonly viewed as summary evaluations of objects (e.g. oneself, other people, issues, etc.) along a continuum ranging from positive to negative (Petty et al. 1997). Medical students’ attitudes to learning communication skills have been explored during the past 10 years using both qualitative and quantitative approaches. The results are presented in Table 1.

Table 1.  Earlier studies of medical students’ attitudes towards CSL.

Study	Theme	Participants	Method	Results
Rees & Garrud (2001)	Qualitative study to explore medical students’ views and experiences of CSL	First-year medical students (N = 5)	Focus groups	Three attitude-related themes: (1) positive attitudes towards CSL; (2) negative attitudes towards CSL; and (3) relationship between students’ attitudes and their education, age and communicative abilities
Rees et al. (2002b)	Qualitative study to explore medical students’ attitudes towards learning communication skills	First- to fifth-year medical students (N = 32)	Focus groups	– Two attitude-related themes were found: (1) positive and (2) negative attitudes towards CSL – Negative attitudes related to: communication skills as a social science; communication skills courses being taught in the preclinical part of the curriculum; and socialization into the same negative attitudes of older students and qualified doctors
Rees et al. (2002a)	Developing the CSAS to measure medical students’ attitudes towards CSL	First time: First- and second-year medical students (n = 490) Second time: first- and second-year medical students (n = 39)	– Factor analysis – The test–retest reliability	– The CSAS consists of 26 items – The scale has two subscales, each with 13 items – In subscale I, PAS, the items relate to positive attitudes to CSL – In subscale II, NAS, the items relate to negative attitudes to CSL
Rees & Sheard (2002)	Exploring the relationship between medical students’ attitudes and their demographic and education-related characteristics at the University of Nottingham and Leicester	First- and second-year medical students (N = 490)	CSAS	– Female students had more positive attitudes to CSL than male – Students’ whose parents were doctors had more negative attitudes than students whose parents were not doctors – Students who thought their communication skills needed improving had more positive attitudes – Leicester students had higher scores on the PAS and lower scores on the NAS than Nottingham students
Rees & Sheard (2003)	Evaluating medical students’ attitudes towards CSL before and after a communication skills course	First-year medical students’ (N = 216)	– CSAS – A personal details questionnaire	– Students rated their communication skills lower at the end of the course – Students’ PAS scores decreased at the end of the course – Gender, language and ethnicity exhibited two-way interactions with attitudes over time – Female students developed significantly less negative attitudes than males by the end of the course
Cleland et al. (2005)	Assess medical students’ attitudes towards CSL in the University of Aberdeen and validate the use of the CSAS in a different population	First- to third-year medical students (N = 467)	CSAS	– First-year students had higher PAS scores than second and third-year students – First-year students’ NAS scores were lower than scores for second-year students but not for third-year students – The PAS and NAS scores for second- and third-year students were not significantly different – Female students had more positive attitudes than males – The CSAS was an appropriate tool to measure attitudes of medical students in different population
Wright et al. (2006)	Exploring medical students’ attitudes towards CSL, knowledge of appropriate provider–patient communication and confidence of communicating with patients in USA	First- and fourth-year medical students (N = 118)	– CSAS – Perceived importance of medical communication – Perceived confidence about communicating with patients – Appropriate provider communication knowledge	– First- and fourth-year students did not differ in terms of attitudes or knowledge, but fourth-year students were significantly more confident about communicating with patients – Students’ PAS scores were significantly related to perceived importance of communication skills and confidence of communicating with patients – Female students had more positive attitudes than males
Shankar et al. (2006)	Exploring positive and negative attitudes among medical students at the Manipal College of Medical Sciences, Pokhara, Nepal	Third- and fourth-year medical students (N = 123)	CSAS	– Median PAS score was 51. Mean NAS score was 31.18 – No difference in PAS scores of female and male students – Communication skills training in this particular medical school should be modified and strengthened
Anvik et al. (2007)	Exploring medical students’ attitudes towards CSL in Norway and comparing the findings with earlier reports	Norwegian medical students (N = 1833)	– Translated version of the CSAS – A PCA	– CSAS measured three factors which were different from the formerly described two-factor model with PAS and NAS – The CSAS may be used in order to distinguish between affective and cognitive components of attitudes
Anvik et al. (2008)	Exploring cognitive and affective attitudes towards CSL among medical students in Norway	Norwegian medical students (N = 1833)	Translated version of the CSAS	– Norwegian medical students had positive attitudes to learning and using communication skills – Cognitive attitudes stayed high during medical school but affective attitudes declined – Female students had more positive attitudes than males
Harlak et al. (2008a)	Validating a Turkish translation of the CSAS	First- to fifth-year medical students (N = 197)	– Translated version of the CSAS – Factor analysis	Confirmed that the translated version of the CSAS is a valid and internally consistent instrument to measure medical students’ attitudes to CSL
Harlak et al. (2008b)	Exploring and comparing Turkish medical students’ attitudes to CSL and their empathy tendency before and after a communication course	Turkish first-year medical students (N = 59)	– Translated version of the CSAS – The Empathy Tendency Scale – Questions about age and gender	– Mean scores in the positive attitude group decreased significantly by the end of the course compared with the start – No significant change in the negative attitude group – Female students had more positive attitudes than males
Ahn et al. (2009)	Develop a Korean version of the CSAS, and explore pre-medical and medical students’ attitudes towards CSL	First survey: pre-medical and medical students (n = 325) Second survey: medical students and doctors-in-training (n = 257)	First survey: – Translated version of the CSAS – A PCA Second survey: – Open questions	Korean pre-medical and medical students’ attitudes consist of five factors: (1) facilitation of interpersonal skills, (2) doubts about the importance of CSL in medicine, (3) motivation, (4) negative attitudes to assessment, and (5) overconfidence. – Some modifications to the CSAS to improve its fit for measuring Korean students’ attitudes
Molinuevo & Torrubia (2011)	Validating a Catalan version of the CSAS	First- and second-year medical students and first-year nursing students (N = 569)	Translated version of the CSAS – A PCA	– Confirmed that the translated version of the CSAS is a valid and internally consistent instrument to measure medical and nursing students’ attitudes towards CSL – Female students had more positive attitudes than males
Bombeke et al. (2011)	Comparing medical students’ patient-centred attitudes and attitudes to communication skills in two cohorts before and after clinical clerkship	First group (n = 48) did not receive training and second group (n = 37) had received 5 years of training	Translated version of the CSAS and four other validated questionnaires	– Communication-trained students declined in patient-centred attitudes and attitudes to CSL compared to non-trained students – Female students had more positive attitudes than males

These studies show that medical students’ positive attitudes (PAS; positive attitude scale) have declined when measured on the Communication Skills Attitude Scale (CSAS) before and after training in communication skills (Rees & Sheard 2003; Harlak et al. 2008b). The CSAS was developed by Rees et al. (2002a). Female medical students seem to have more positive attitudes than males to learning communication skills (Rees & Sheard 2003; Cleland et al. 2005; Wright et al. 2006; Harlak et al. 2008b). It has also been argued that medical students’ positive attitudes may be associated with more problem-based and self-directed methods of learning in comparison with didactic methods (Rees & Sheard 2002). In addition, it has been suggested that medical educators should consider the role of affective learning when teaching courses emphasizing communication skills and develop teaching and learning methods that present communication skills training in a more positive light or emphasize their importance, because this may help medical students to appreciate the importance of these skills (Wright et al. 2006). However, there do not appear to be any studies comparing different experiential learning methods and their influence on medical students’ attitudes to learning communication skills.

In a review of communication skills training (Aspegren 1999), experiential methods were reported to be more effective in teaching communication skills than instructional methods. Nowadays experiential methods such as SPs and RP are widely used in communication skills training (Lane & Rollnick 2007; Cleland et al. 2009; May et al. 2009; Hargie et al. 2010). SPs portray live interactive simulations of specific communication challenges and medical problems, offering the students a safe learning environment with no risk of harm to real patients (Kurtz et al. 2005). However, SPs are expensive, and some students consider them artificial (Kurtz et al. 2005). In RP, fellow trainees play the role of the patient and the doctor (Lane & Rollnick 2007). RP offers an effective and cost-effective way to practise communication skills (Kurtz et al. 2005). Some students find RP with fellow students unrealistic, whereas others find it more relaxing than simulations (Nestel & Tierney 2007; Koponen et al. 2011).

Alongside SPs and RP, another experiential learning method, TIE, has recently been applied in medical students’ communication skills training (Koponen et al. 2010). TIE connects theatre techniques with education, utilizes elements of traditional theatre, educational drama and simulation, and contains structured patterns of activities around the selected topic (Jackson 1993). TIE differs from SPs and RP, because the idea is to watch a short play of doctor–patient encounter and reflect the play through drama conventions, and the whole group of students is actively involved throughout the TIE workshop (Koponen et al. 2010).

Our earlier study (Koponen et al. 2011) compared Finnish second-year medical students’ perceptions of the suitability of TIE, SPs and RP with peers in learning interpersonal communication competence. This study includes a detailed description of each method used (Koponen et al. 2011). The results showed that the respondents were very positively disposed to the use of all three methods; however, this study did not measure the effectiveness of the methods. Even though several researchers have investigated medical students’ attitudes to communication skills learning (CSL) with CSAS (Rees et al. 2002a,b; Rees & Sheard 2002, 2003; Cleland et al. 2005; Wright et al. 2006; Harlak et al. 2008a,b), these studies have not compared students’ attitudes in three groups using different experiential learning methods. In addition, Finnish medical students’ attitudes to learning communication skills have not been measured. A recent study of Finnish university students’ (N = 1323) study-related social anxiety (Almonkari 2007) showed that practising communication skills was one of the contexts where students experienced most social anxiety. It is therefore of interest to examine the attitudes of Finnish medical students to learning communication skills.

The evaluation focuses on educational programmes to determine their merit or worth and assessment on learners to determine how well the learner is doing (Cook 2010). In this study, we aimed to evaluate and compare second-year medical students’ attitudes to learning communication skills before and after communication skills training in the data as a whole, divided by gender and divided into three groups where different experiential learning methods were used.

Table

Practice points

All three learning methods, TIE, SPs and RPs, had a positive impact on medical students attitudes to learning communication skills. Female students had more positive attitudes than male students. Longitudinal research on medical students’ attitudes to CSL is required.

Method

The pilot course in communication

In Finland, communication skills training for medical students started at the beginning of the 1990s (Pyörälä 2006). However, in a middle-sized university in eastern Finland, this area of the curriculum has not been widely developed, and experiential learning methods have not been used in teaching communication skills to medical students. Due to a curriculum reform, in 2006 we developed a pilot course in communication for second-year medical students in co-operation with a speech communication lecturer and clinical lecturers. Three experiential methods, namely TIE, SP interview with amateur actors (SP) and RP with peers, were adapted in a pilot course. The pilot course had three objectives: (1) to help students learn to analyse doctor–patient communication, (2) to practise interpersonal communication skills, and (3) to help students to understand [and appreciate] the meaning of communication and interpersonal communication skills in the doctor–patient relationship. The third objective refers to cognitive learning, but implicitly it refers to affective learning (McCroskey 1994), since we also aimed to help the students to appreciate the doctor's interpersonal communication competence and to foster positive attitudes to learning communication skills.

The pilot course emphasized history-taking skills at the beginning of the medical consultation, especially building a relationship with the patient, exploring the patient's problems (questioning, listening and non-verbal communication) and providing structure to the consultation. These skills were selected on the basis of the enhanced Calgary–Cambridge Observation Guide (Kurtz et al. 2005). The structure of the pilot course is presented in Table 2. First, a lesson in doctor–patient communication was given, and the students analysed doctor–patient interaction shown on DVD. Then, the students practised communication skills through the TIE, SPs or RP method. The patient scenarios were based on real patient cases in primary health care and were the same in each learning method. In the TIE workshop, we used theatre performance and drama techniques in order to activate the students to think about doctor–patient interaction and take part in the fictional context created in the doctor's role. The patient encounter was further developed with the student group, and several elements of drama education were used in the learning process. SPs and RPs involved practising patient interview with an amateur actor (SP) or a peer (RP) followed by self-evaluation and constructive feedback discussions with peers and facilitator. Video-recording was not used in any of these learning encounters. Next, all students spent 2 days in health centres observing and analysing real doctor–patient encounters. Finally, the students discussed their learning experiences in small groups. The study period was 3 months. Grading was pass/fail.

Table 2.  Structure of the pilot course in speech communication.

Preparation	Training		Evaluation
All students – 1 h introduction – 2 h/seven small groups – Short lesson in doctor–patient communication – Students analysed doctor–patient interaction shown on DVD – Students formulated their learning goals	Group A (TIE) – 2 × 2 h – TIE workshop	All – Students spent 2 days in health centres – Students observed real doctor–patient consultations and wrote reports on their findings (self-study exercise)	All – 2 h/seven small groups – Students discussed the self-study exercises – Students’ learning experiences were discussed
	Group B (SP) – 2 × 2 h – SP interviews with amateur actors
	Group C (RP) – 2 × 2 h – RP with peers

Participants’ demographics

All second-year medical students (N = 136) in our university were invited to take part in the study. Of these, 129 (95%) completed the questionnaire twice: before the pilot course (T1; time 1), and after the pilot course (T2; time 2). Respondents’ age ranged from 19 to 34 years (M = 22.15; Mdn = 22), 45 (35%) were male and 84 (65 %) female students.

The students were randomly assigned to three groups.1 In group A we used the TIE method, in group B SPs and in group C RP. In group A (n = 42), 13 (31 %) males and 29 (69 %) females took part in the study, and their ages ranged from 20 to 29 years (M = 21.69; Mdn = 21). In group B (n = 44), 14 (32 %) males and 30 (68 %) females took part, and their ages ranged from 20 to 34 years (M = 22.66; Mdn = 22). In group C (n = 43), 18 (42 %) males and 25 (58 %) females took part, and their ages ranged from 19 to 31 years (M = 22.07; Mdn = 22). Randomization to three groups was successful (χ² = 1.389; df = 2; p = 0.499).

Instruments

We used the CSAS, developed by Rees et al. (2002a). The CSAS consists of 26 items, all of which have response options along a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The scale has two subscales. In subscale I, the PAS, 13 items relate to positive attitudes to CSL. In subscale II, the negative attitude scale (NAS), 13 items relate to negative attitudes to CSL. The CSAS has been found to show satisfactory internal consistency and test–retest reliability (Rees et al. 2002a; Cleland et al. 2005).

The CSAS was translated from English into Finnish independently by a speech communication lecturer and an English lecturer. For the back translation, a similar group applied the same procedure. Together they agreed on a Finnish version of the CSAS. After back translation the speech communication lecturer compared the Finnish version and back-translated items and finalized the Finnish version of the scale. The items were in the same order as in the original CSAC and the same five-point Likert response options were used.

In addition, the questionnaire included respondents’ age, gender, year of training, name (optional), letter of the group (A, B or C) and consent to participate in the study.

Data analysis

Data were analysed using SPSS version 17.0. Both PAS and NAS scores were calculated by summing response values for the 13 items of each subscale. Exploratory data analysis revealed that all continuous variables including PAS and NAS scores at T1 and T2 were non-normally distributed. Therefore, non-parametric statistical tests were used.

The internal structure of the translated CSAS was examined by a principal component analysis (PCA), as has been done in earlier studies (Harlak et al. 2008a; Molinuevo & Torrubia 2011). The Kaiser–Meyer–Olkin measure of sampling adequate was 0.88 and Bartlett's test of sphericity was significant (p < 0.000), confirming the appropriateness of PCA. The PCA with Varimax rotation revealed six factors with eigenvalues over 1 and explained 61.93 % of total variance. Next, two factors were explicitly extracted in order to see if the original two-factor structure was maintained. Two factors accounted for 40.93 % of the variance. Eigenvalues were 8.966 for factor 1 and 1.676 for factor 2. The factor loadings of the individual items supported the distinction of the positive and negative attitude subscales reported by Rees et al. (2002a), thus supporting the construct validity of the translated scale. In order to assess the reliability of the translated CSAS, the Cronbach's α's were calculated for PAS and NAS measured before and after the pilot course. PAS-before was α = 0.882 and NAS-before was α = 0.794. PAS-after was α = 0.895 and NAS-after was α = 0.828. NAS-before had the lowest α value. If item 20 had been deleted, Cronbach's α would have been 0.811; however, with the inclusion of item 20 it was 0.794. Since the difference was not statistically significant, we decided to retain item 20 in NAS-before in order to have as comparable results as possible with earlier studies.

Non-parametric statistical tests (Wilcoxon test, Mann–Whitney U-test, Kruskall–Wallis test) were used to establish whether students’ scores on the PAS and NAS differed significantly in the data as a whole, between genders and in three groups at T1 and T2.

Results

Participants’ positive attitudes at T1 and T2

Scores on the PAS at T1 and T2 in the data as a whole, divided by gender and by each group are presented in Table 3. In the data as a whole, medical students’ scores on PAS increased significantly after training. In addition, male and female students’ scores on PAS increased significantly after training. Mann–Whitney U-test showed that female and male students’ scores on PAS differed significantly at T1 (U = 1176, z = −2.968, p = 0.003) and also at T2 (U = 1308.5, z = −2.794, p = 0.005). This shows that female students had more positive attitudes than males at the beginning and at the end of the pilot course.

Table 3.  Relationships between PAS scores at T1 and T2.

	Median PAS score at T1 (inter-quartile range)	Median PAS score at T2 (inter-quartile range)	Wilcoxon test statistic, p-value
Whole data	45 (40.5–49)	47 (42–51)	z = −3.932, p = 0.000***
Male (n = 42)	42 (37.75–47)	43 (40–48.5)	z = −2.877, p = 0.004**
Female (n = 83)	46 (42–50)	48 (43–52)	z = −2.795, p = 0.005**
Group A (n = 42)	45 (42–49.5)	49 (42.75–54)	z = −2.768, p = 0.006**
Group B (n = 44)	45 (42–48)	46 (43–50)	z = −2.221, p = 0.026*
Group C (n = 43)	43 (37.5–49)	45 (39–49)	z = −1.952, p = 0.051

Notes: ***Significant at the p < 0.001 level; ** significant at the p < 0.01 level; and *significant at the p < 0.05 level.

In groups A and B, scores on PAS increased significantly at the end of the course. However, in group C there were no significant differences with students’ PAS scores at T1 and T2. Kruskall–Wallis test showed that the means of students’ scores on PAS did not differ significantly between the three groups at T1 (χ² = 1.570; df = 2; p = 0.456) or at T2 (χ² = 4.333; df = 2; p = 0.115). In addition, Kruskall–Wallis test showed that there were no significant differences in male students’ PAS scores in these three groups measured before the course (χ² = 0.830; df = 2; p = 0.660) or after the course (χ² = 1.171; df = 2; p = 0.557). The results were the same for female students’ PAS scores: there were no significant differences in female students’ PAS scores in these three groups measured before (χ² = 1.057; df = 2; p = 0.589) or after the course (χ² = 4.113; df = 2; p = 0.128).

Participants’ negative attitudes at T1 and T2

Scores on the subscale NAS at T1 and T2 in the data as a whole, by gender and by each group are presented in Table 4. In the data, as a whole, medical students’ scores on NAS decreased significantly after training. Male students’ scores on NAS did not change significantly after training. However, female students’ scores on NAS decreased significantly after training. Mann–Whitney U-test showed that female and male students’ scores on NAS differed significantly at T1 (U = 1375, z = −2.111, p = 0.035) and also at T2 (U = 1302.5, z = −2.824, p = 0.005). This shows that female students had less negative attitudes than males at the beginning and at the end of the course.

Table 4.  Relationships between NAS scores at T1 and T2.

	Median NAS score at T1 (inter-quartile range)	Median NAS score at T2 (inter-quartile range)	Wilcoxon test statistic, p-value
Whole data	31 (27–35)	28 (25–33)	z = −3.677, p = 0.000***
Male (n = 43)	32 (27–38)	31 (26.5–36)	z = −1.490, p = 0.136
Female (n = 83)	30 (27–33)	27 (24–32)	z = −3.495, p = 0.000***
Group A (n = 42)	29.5 (26–35)	26 (22–33)	z = −3.757, p < 0.0001***
Group B (n = 44)	29.5 (27–33.25)	28 (25–33)	z = −1.218, p = 0.223
Group C (n = 43)	32 (27.75–36.25)	30 (27–35)	z = −1.538, p = 0.124

Note: ***Significant at the p < 0.001 level.

In group A, scores on NAS decreased significantly after training. In groups B and C, there were no significant differences in students’ NAS scores at T1 and T2. Kruskall–Wallis test showed that the means of students’ scores on NAS did not differ significantly between the three groups at T1 (χ² = 1.906; df = 2; p = 0.386) or at T2 (χ² = 4.097; df = 2; p = 0.129). In addition, Kruskall–Wallis test showed that there were no significant differences in male students’ NAS scores in these three groups measured before the course (χ² = 1.615; df = 2; p = 0.446) or after the course (χ² = 0.836; df = 2; p = 0.658). The same results were found for female students’ NAS scores: there were no significant differences in female students’ NAS scores in these three groups measured before (χ² = 2.973; df = 2; p = 0.226) or after the course (χ² = 3.725; df = 2; p = 0.155).

Discussion

This study evaluated Finnish second-year medical students’ attitudes to learning communication skills before and after a pilot course in communication. Overall, the study demonstrates that medical students’ (N = 129) positive attitudes to learning communication skills (PAS) increased significantly, and their negative attitudes (NAS) decreased significantly between the beginning and end of the course. In other words, students’ attitudes did indeed become more positive during the pilot course in communication. The positive change in students’ attitudes could be seen as affective learning (Mottet & Beebe 2006). It has been argued that experiential methods are more likely than didactic methods to lead to action or change in behaviour or attitude, and to result in experimentation with alternatives and in the development of skills and strategies (Kurtz et al. 2005). The results of this study show that a communication course based on experiential learning methods may have a positive effect on students’ attitudes in a reasonable amount of time, since the course lasted 3 months.

The increase in positive (PAS) and decrease in negative (NAS) attitudes found in this sample of students differs from that described by other researchers (Rees & Sheard 2003; Harlak et al. 2008b; Bombeke et al. 2011). However, it is difficult to directly compare these results since, for example, the course content, duration and methods used vary in different studies. Nevertheless, none of the studies exploring medical students’ attitudes before and after communication training with CSAS (Harlak et al. 2008b; Rees & Sheard 2003) describe integrating visits to health centres with communication skills training or using mainly experiential learning methods during training, as was done in this study.

Rees et al. (2002a) suggested that the PAS measures students’ positive attitudes towards learning communication skills and their beliefs that communication skills are important within medical practice. Therefore, the positive change in students’ PAS scores after training in this study indicates that students realized the importance of communication skills in medical practice. One of the reasons behind the positive result of this course may be that the communications skills training was integrated into a clinical context. After practising communication skills with experiential learning methods, the students spent 2 days in health centres observing and analysing real doctor–patient encounters. Therefore, the communication skills may become more meaningful to the students. It has been shown that if the communication training environment is too idealistic compared to the reality of medical practice, it may lead to a decline in medical students’ patient-centred attitudes and their attitudes to learning communication skills during clinical clerkships (Bombeke et al. 2011). Therefore, it seems relevant to maintain a connection between communication skills training and clinical reality.

Rees et al. (2002a) suggested that the NAS measures students’ negative attitudes towards the learning and teaching of communication skills. The results showed a significant decline in students’ NAS scores after training, thereby indicating that students’ attitudes towards teaching and learning communication skills became more positive. In group A, where the TIE method was used, students’ PAS scores increased significantly and their NAS scores decreased significantly. However, there were no significant differences in the mean scores for PAS and NAS measured in the three groups before or after the course. Therefore, we cannot claim that one method was more effective than others. It would nevertheless be interesting to explore the results of our earlier study, which compared medical students’ perceptions of the suitability of TIE, SPs and RP in learning interpersonal communication competence (Koponen et al. 2011). The respondents were the same medical students as in this study. The results showed that respondents had very positive views on the use of all three methods, and that these methods had five similar elements. The similarity and the students’ positive experience of training communication skills through all three methods may explain why their NAS scores declined and why there were no differences in attitudes in the three groups after training.

However, our earlier study showed that there were some differences in respondents’ perceptions of the method's helpful and unhelpful elements (Koponen et al. 2011). Respondents who had experienced communication skills training with the TIE method found being active (the method stimulated them to think and led to new thoughts) and reflective discussions together with peers in the modified forum theatre stage (respondents appreciated being able to intervene in the play by stopping the action and discussing with the group when several viewpoints and proposals came up) as helpful elements. These elements were not found in groups B or C. The techniques of TIE could be adopted to SPs and RPs in order to further develop these learning methods in CSL.

The results show that female students had more positive attitudes than males at the beginning and at the end of the course. The results are consistent with the results of several earlier studies showing that female medical students have more positive attitudes to learning communication skills than males (Rees & Sheard 2003; Cleland et al. 2005; Wright et al. 2006; Harlak et al. 2008b; Bombeke et al. 2011). Rees and Sheard (2003) note that Aspegren's (1999) review of teaching and learning methods for communication skills training showed that male medical students were slower at learning communication skills than females, and this may explain why in their study, female medical students had a decrease in NAS by the end of a communication skills course. At the beginning of communication skills training, more attention should be paid to male medical students’ attitudes to learning communication skills.

The Finnish version of CSAS seemed to be a suitable instrument to measure Finnish medical students’ attitudes to learning communication skills. Internal consistency analysis showed that the reliability of the two subscales was adequate, although the α’s calculated at T1 and T2 for the NAS (NAS-before α = 0.794; NAS-after α = 0.828) were somewhat lower than for PAS (PAS-before α = 0.882, PAS-after α = 0.895). However, an α of at least 0.70 suggests reasonable reliability (Metsämuuronen 2003), and therefore the α's calculated in this study were acceptable.

This study has limitations, which must be taken into consideration when evaluating and interpreting the results. In order to study the direct impact of a pilot course on the respondents’ attitudes, a randomized controlled trial should have been designed, which was impossible in our university in 2006. The pre-test, post-test design in this study means that differences found in respondents’ attitudes in T1 and T2 may be due to factors other than the pilot course in communication or the different experiential learning methods used. In addition, the overall sample size (N = 129) was satisfactory, but the number of respondents in the three groups were rather small. The findings on medical students’ attitudes in this study may be representative of this year-group, but not necessarily of all medical students in our medical school or in Finland in general.

According to McCroskey (1994), students who complete communication classes or programmes should have heightened positive affect regarding communication and the courses and programmes they have taken (appreciating their content, the worth of behaviour learned and the instructors who taught them). It has been suggested that CSL should be integrated with clinical experience, because only in the clinical context do medical students truly realize the benefits of CSL (Rees et al. 2002b). However, this study showed that the use of experiential learning methods and integrating communication skills training with health centre visits may actually help medical students to appreciate communication skills training at an early stage of their studies. In the future, we should repeat the measurement of students’ attitudes in order to see whether the change in their attitudes was sustained. Moreover, an assessment of medical students’ communication skills and their knowledge of doctors’ interpersonal communication competence could be conducted in order to have a comprehensive evaluation of effectiveness of the methods.

Declaration of interest: The authors report no declarations of interest.

Notes

1. Ethical approval to conduct the study was obtained from the legal adviser of our university, even though it was not required according to the regulations at our university.