1 INTRODUCTION

Effective communication skill of physicians is an important component of high-quality healthcare delivery and safe patient care.¹ In a medical context, communication skills have two dimensions—social and clinical. From a social perspective, effective communication skills of physicians facilitate building a relationship of trust and respect,^2-6 exchange of information and emotion about disease and illness, and empower patients in decision-making.⁷ On the other hand, from the clinical perspective, communication skills improve health outcomes,^{8, 9} adherence to medical recommendations,^{10, 11} and patient satisfaction.^12-18 A physician has to communicate effectively not only with patients but with colleagues, nurses, administrators, relatives of patients and news media.¹ Because, any healthcare delivery is a team effort, effective communication with colleagues also contributes to a reduction of professional stress and burnout among healthcare professionals,^19-21 making fewer errors in clinical decision-making,^{22, 23} and health literacy of patients.²⁴ So, a physician should have interpersonal communication skills,^{25, 26} and appropriate nonverbal communication skills—silence, facial expression, touch, and closer physical proximity.^{27, 28} Improved nonverbal communication skills enhance active listening and help to develop empathy and intuition among physicians.²⁹ A study found that medical students' caring behavior is strongly associated with their attitudes toward communication skills.^{30, 31} Another study in South Korea found that medical students' patient-centered attitude and empathic concern are correlated with their attitude toward learning communication skills.³² This attitude is influenced by the academic curriculum of medical schools. In a 12-year-long research, Gude et al.³³ found that an emphasis on communication skills in the academic curriculum would increase positive attitudes toward learning communication skills among medical students compared with students from the traditional curriculum.

The Communication Skills Attitude Scale (CSAS) is a widely used tool to measure the attitude of medical students toward communication skills. The scale was developed by Rees et al.³⁴ It has widely been used and validated in different populations, settings and countries such as Nepal,³⁵ the UK³⁶ India,³⁷ China³⁸ Malaysia,³⁹ Korea,⁴⁰ Iran,⁴¹ Germany,⁴² Poland,⁴³ and Turkey.⁴⁴ There are two trends in the studies on the CSAS. Like the original CSAS,³⁴ some studies stick to two factorial structures,^{44, 45} and some studies have used a different subscale dimension.^{46, 47} For example, Anvik et al.⁴⁸ applied the CSAS differently. They identified three factors—learning, importance, and respect—that influence medical students' attitudes toward learning communication skills. The learning factors refer to the student's feelings about the way communication skills are taught. The “importance” factors describe students' attitudes and values toward the importance of having communication skills for doctors. On the other hand, “respect” factors indicate students' feelings about whether good communication skills may help them respect patients and colleagues.

2 METHODS

3 ADAPTATION OF CSAS

The original CSAS was in English (see Annex-I), and it was adapted into Bangla by following standard forward and backward translation. The forward translation from English to Bangla was performed by an expert translator. The expert had a master's degree in English language and literature and had more the 15 years of experience in professional translation. He was not aware of the research. Another forward translation was performed by a medical student. He was also fluent in both English and Bangla. Later, both the forward translated versions were compared and contrasted by the paper's two authors—one was an expert in communication in healthcare, and had bachelor's and master's degrees in communication, and another was a trained physician. After addressing and accommodating differences, inconsistencies, and variances, they compiled and merged both translations into a single Bangla-translated version. Then another medical graduate and public health expert back-translated the tool into English. All the translators are native speakers of Bangla and fluent in English. The translators were instructed to use simple language and to capture the meaning of the item rather than perform a literal word-by-word translation. The back-translation version was compiled by following similar procedures to the forward translation. Later, all four versions were evaluated by an expert committee formed for this study. The expert committee reviewed and suggested the final adaptation of the instrument (see Annex II). The reviewed version was used for the pretest among 52 respondents of the study.

4 RESULTS

5 ATTITUDINAL SCORES IN ALL ITEMS OF THE CSAS QUESTIONNAIRE

Data in Table 2 presents information about attitudinal scores for both male and female participants in all items of the CSAS questionnaire. In the PAS, the attitude of the participants indicates that most of the medical students strongly believe that they need to have good communication skills to be a good doctor (Mean 4.50, and SD 1.04), and communication skills teaching should be more like a science subject (Mean 3.85, SD 1.19). From the table, it is evident that overall female medical students are more willing to learn communication skills compared male students as shown in the item-wise average scores of the PAS scale. On the other hand, the NAS indicate that the students are reluctant to learn communication as they think that they do not need good communication skills to be a doctor (Mean 1.61, SD 1.11). Meanwhile, the item-wise average scores in the NAS scale indicate that male students tend to have more negative attitude toward learning communication skills compared with female students. Details are shown in Table 2.

6 SAMPLE ADEQUACY AND INTERNAL CONSISTENCY

Overall, the KMO value is 0.940 which corresponds to excellent sample adequacy. The internal consistency of the items is good since the overall value of α is 0.882. For PAS, the internal consistency is 0.933. While, for NAS, the value is 0.719. Details are shown in Table 3.

7 CORRELATION BETWEEN THE FACTORS

The items are not correlated with each other within or without the groups (p value: <0.001). Details are shown in Table 4.

Figure 1 displays a steep curve in the scree plot of the Eigenvalues, indicating that the top two or three components adequately describe the data's essence. Details are shown in Figure 1.

Table 5 shows the Eigenvalues for different factors utilized in Principal Component Analysis. Details are shown in Table 5.

Figure 2 shows the relationship between the factors used in PCA through scatter plots. As per Figure 2 and Table 5, Factor 1 is most strongly correlated with each other component. Details are shown in Figure 2.

8 DISCUSSIONS

This study translated and adapted the CSAS into the Bangla language and validated it in a cohort of medical students in Bangladesh. It found that the Bangla version of the scale is feasible, valid, and internally consistent.

The overall internal consistency of the CSAS-Bangla is 0.882, and the values of Cronbach α of the two subscales are 0.933 and 0.719. In terms of internal consistency, it has a higher degree of similarity with some previous studies. For example, the overall internal consistency of the Polish version was α 0.853,⁴⁹ and the Iranian version 0.84.⁴¹

Similar to the original CSAS and some other studies that had two factorial structures, the Bangla version of the CSAS also has two components. However, several other studies in different contexts yielded more than to two subscales. For example, the Norwegian version had three subscales—“learning,” “importance,” and “respect”⁴⁸; and the Hungarian version had seven—respect and interpersonal skills, learning, the importance of communication in the medical profession, excuse, counter, exam and overconfidence, and Persian version of the CSAS had four components—important in the medical context, excuse, learning and overconfidence.⁴¹ The difference in the factorial structures may be due to socioeconomic, geographic, cultural, and linguistic differences of the participants.

To the best of our knowledge, this is the first endeavor to validate the psychometric properties of CSAS in the Bangla language, and from the context of a developing country like Bangladesh. While most of the previous validation of the scale included students from a single medical college, we collected data from various medical colleges.

The results of our study have both theoretical and practical implications. Theoretically, it will contribute to the understanding of the attitude of medical students toward learning communication skills in the context of a developing country. Moreover, it will help to improve the curriculum and delivery of medical education. Thus, it will contribute to preparing future physicians equipping them with the art and craft of communication and ensuring better healthcare delivery.

However, acknowledge that this study has some limitations. First, we collected data from only a region of the country as the medical colleges we selected for data collection were located only in the northern region of Bangladesh. Second, we collected data mainly from undergraduate level medical students. Third, we did not put emphasis on collecting data from students of dental colleges. So, future studies be conducted in a large scale with a focus on medical and dental students from different levels and different medical colleges across the country. Future studies also should focus on postgraduate level students.

9 CONCLUSION

We conclude that CSAS-Bangla is a valid and reliable tool for assessing communication skill attitudes among Bangla speaking medical students in Bangladesh. The psychometric properties of this tool are a novel approach to understanding the communication dynamics of preparing future healthcare professionals in the context of a developing country, Bangladesh. This scale can be used in future studies in measuring the attitude toward learning communication skills among medical Bangladesh, designing and evaluating communication skills training programs in medical colleges in the country.

AUTHOR CONTRIBUTIONS

Mohammad Aminul Islam contributed to conceptualizing the study, designing the study, preparing data collection tools, data collection, data analysis, writing manuscript and finalizing the manuscript, and coordinating the project. Maskura Benzir contributed to designing the study and data collection. Md. Kaoser Bin Siddique contributed to writing the draft manuscript and data collection. Md. Abdul Awal contributed to conceptualizing the study and data collection. Mohiuddin Ahsanul Kabir Chowdhury contributed to designing the study, data analysis, and writing the manuscript. Taufique Joarder contributed to conceptualizing the study, writing and finalizing the manuscript, and supervising the project.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

TRANSPARENCY STATEMENT

The lead author Mohammad Aminul Islam affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.