A Comparison Study of Communication Skills Between General Surgery and General Practice Residents on First-time Patient Visits.

BACKGROUND
There is little published research about differences in doctor-patient communication of different specialties. Accordingly, we compared doctor-patient communication skills in two different specialties, general surgery (GS) and general practice (GP).


METHODS
Twenty residents from the Bahrain Defence Force Hospital (10 men and 10 women; mean age 28 years; 10 GS and 10 GP) participated in 200 patient first visit consultations. The consultations were video-recorded and analysed by four trained observers using the MAAS Global scale.


RESULTS
1) Internal consistency reliability of the MAAS Global (> 0.91) and Ep(2) = 0.84 for raters was high, 2) GP residents spent more time (12 minutes) than GS residents (7 minutes), in the visits, 3) There were several differences on the MAAS Global items between GP and GS residents (GS > GP, p < 0.05 on history taking, diagnosis and medical aspects; GP > GS, p < 0.05 on information giving), and 4) The present participants performed well compared to normative samples as well as to criterion-referenced cut-off scores. The general level of communication skills in both specialties, however, was 'unsatisfactory' and 'doubtful', as it is for normative samples.


CONCLUSION
Excellent doctor-patient communication is essential but does not appear to receive the amount of attention that it deserves in practice settings. There are some differences between specialties as well as unsatisfactory communication skills for both specialties, since residents from both programs spent less time than recommended on each consultation. Our findings emphasize the need to improve the communication skills of physicians and surgeons in general.


Introduction
Some believe that general practitioners use a predominant patient-centred style and general surgeons tend to use a doctor-centred approach. 14 This may suggest communication differences between doctors who deal with surgical procedures compared to doctors who do not. This, in part, may reflect the changing role of the doctor-patient relationship in the past two or so decades which now involves greater patient control, reduced physician dominance, and more mutual participation. 15 Accordingly, the main purpose of the present study was to compare the communication patterns of doctor-patient consultations in two medical specialties, general surgery and general practice. To systematically study these differences, we conducted a comparative quantitative analysis of the similarities and differences in doctor-patient communication between residents in general surgery and general practice.

Residents
A total of twenty residents, ten from general practice (GP) and ten from general surgery (GS) from year 1, 2, and 3 of their residency program, were chosen at random and invited to participate in the study. The age of residents ranged from 25-33 years old with a mean age of 28 years. In general surgery there were 6 (60%) male doctors and 4 (40%) female doctors, whereas in general practice, there were 8

Design and Procedures
This comparative study used a quantitative analysis of checklist scores to compare doctor-patient consultations in general surgery and general practice. The study was conducted in the outpatient departments of the two specialties, general surgery and general practice, in the Bahrain Defence Force Hospital (BDFH). Patients come mainly from the military population and their relatives, and some from the civilian population.
All residents in both specialties received a written or verbal invitation, explaining the aim of the research, and they agreed to the video-recording of twelve first visit patient consultations. There was no dropout of doctors from the study. Each resident was evaluated with twelve first visit patients. The first two patients in the evaluation were not used in the study as they were practice sessions to allow the doctors to become comfortable with the videotaping.
Twelve first visit patients were selected for residents to do an entire patient consultation in 15-20 minutes. A selection of ten patients is considered to be sufficient to allow comparisons of mean scores at the group level. 16 Accordingly, 10 patient consultations for each of 20 residents (i.e., 200 in total) were analyzed. The consultations took place at the Bahrain Defence Force Hospital (BDFH). Residents were video-recorded, and the first two patients for each resident (i.e., 20 x 2 = 40) were not included in the study.

Assessment of communication skills
The MAAS Global instrument for rating doctor communication skills was used. This instrument consists of a checklist and a 30-page scoring manual, listing criteria for each item. 17 The MAAS Global instrument consists of three main aspects: 1) the communication skills for each separate phase such as introduction, follow up, consultation, and diagnosis, 2) general communication skills such as exploration, emotions, and empathy, and 3) medical aspects such as history taking, physical examination, and management. In the checklist seventeen caseindependent items are used and rated on a 7-point scale: 0 = not present, 1 = poor, 2 = unsatisfactory, 3 = doubtful, 4 = satisfactory, 5 = good, 6 = excellent (see Appendix A).
The focus of the first thirteen items was on the communication skills, while the last 4 items related to the medical content. Since we limited our study to first-time patient visits, we excluded item number two which relates to follow up consultations. Several studies have provided evidence for the validity of the MAAS Global for assessing communication skills. 17 In addition, the MAAS-Global has high internal consistency reliability (alpha > 0.90) and reproducibility (r > 0.80). 16 Each rater reviewed 50 video-recorded consultations and gave a score for each consultation. The raters received standardized training, carried out by one trainer. The communication was evaluated by 4 trained observers who rated the videotapes using the MAAS Global checklist. The inter-observer reliability coefficient for each group of ratings was high (> 0.90).

Ethics
The study was approved by the Ethics Committee of Bahrain Defence Force Hospital. Each doctor consented to take part and be video-recorded in the study. Consent for recording was also obtained from patients.

Data Analysis
Descriptive statistics were computed on the data, between specialty differences were explored with one-way multivariate analysis of variance (MANOVA) of the scale items (independent variable = surgeon vs general; dependent variables = 16 items) and was followed-up with post-hoc one-way ANOVAs. The internal consistency reliability was computed with Cronbach's α and overall generalizability analyses (Ep 2 ) were conducted to determine the generalizability of various facets, including interrater reliability. A p value of < 0.05 was considered the critical value for significance.

Results
Internal consistency reliability analyses for the total MAAS Global scale (k=16) produced α= 0.92 and the subscale α range from 0.65 to 0.87. A fully-crossed single-facet (4 raters x 10 consultations) generalizability analysis resulted in an Ep 2 = 0.84.
Accordingly, high reliability was achieved for both internal consistency and rater reliability. Although the time allocated for each first-time patient visit was 15-20 minutes, GS residents spent an average of 7 minutes with each patient and GP residents spent 12 minutes.
The between general surgery and general practice analyses were conducted with MANOVA (dependent variables = 16 items; independent variables = specialty) and are summarized in Table 1. There were overall differences between the two groups (Wilk's lambda = 0.747; F = 3.87, p < 0.001) and subsequent post-hoc ANOVAs showed that there were several differences on single items and on one subscale. A close inspection of Table 1 Table 2 contains the subscale scores and their descriptive statistics. When the items are summed into subscales or the total scales, there are no differences in the total MAAS Global scores between the two specialties. The lack of differences in the total scores between the two specialties is due to the cancelling effects of items 9, 13 and 15 in

Discussion
The results are 1) Internal consistency reliability of the MAAS Global and Ep 2 for raters was high, 2) The patient encounters were brief, but GP residents spent more time than GS residents, 3) There were several differences on the MAAS Global items in performance between GP and GS residents, and 4) The present participants performed well compared to normative samples as well as to criterionreferenced cut-off scores.
The α reliability of the MAAS Global was high (> 0.90) in the present study as has been found in previous research. 17 Additionally, we conducted the Ep 2 analyses and found high consistency across raters indicating that our data has high reliability  Reinders et al. 16 found that 2 raters across 5 consultations are adequate to achieve Ep 2 > 0.70.
We exceeded this minimum with 4 raters for each of 10 consultations.
The present study is the first one to investigate the differences and similarities in doctor-patient communication skills in these two specialties, general surgery and general practice. Our findings indicate that doctor-patient communication differs between the specialties in history taking, information giving, and diagnosis as well as medical aspects. Moreover, the actual time spent with patients differed between specialties (GP > GS). As well, GP residents did better than GS residents in information giving, but the reverse was true for history taking, diagnosis and medical aspects as medical content. These differences probably reflect the differences in practice specialties, with GS more focused on the surgical aspects of the consultation whereas the GP residents may focus more on information sharing or giving. These differences are small (effect sizes ≈ 0.30), however, and there are no differences between the specialties on 13 of the 16 items or on the total scale score of the MAAS Global. Accordingly, GS and GP residents appear to be quite similar in their communication skills at least for first time patient encounters, although GP residents tend to spend more time with patients than GS residents.
We found that the general level of communication skills in both specialties received unsatisfactory or doubtful ratings on the MAAS Global. This finding is in concordance with several other studies of GP

Conclusion
Excellent doctor-patient communication is essential in healthcare and does not appear to receive the amount of attention that it deserves in practice settings. We found that there are some differences between specialties as well as unsatisfactory communication skills for both specialties. Both GP and GS residents also spent less time than recommended on each consultation. Our findings emphasize the need to improve the communication skills of physicians and surgeons in general. OTHER FEEDBACK