Medical education in a foreign language and history-taking in the native language in Lebanon – a nationwide survey

Background With the adoption of the English language in medical education, a gap in clinical communication may develop in countries where the native language is different from the language of medical education. This study investigates the association between medical education in a foreign language and students’ confidence in their history-taking skills in their native language. Methods This cross-sectional study consisted of a 17-question survey among medical students in clinical clerkships of Lebanese medical schools. The relationship between the language of medical education and confidence in conducting a medical history in Arabic (the native language) was evaluated (n = 457). Results The majority (88.5%) of students whose native language was Arabic were confident they could conduct a medical history in Arabic. Among participants enrolled in the first clinical year, high confidence in Arabic history-taking was independently associated with Arabic being the native language and with conducting medical history in Arabic either in the pre-clinical years or during extracurricular activities. Among students in their second clinical year, however, these factors were not associated with confidence levels. Conclusions Despite having their medical education in a foreign language, the majority of students in Lebanese medical schools are confident in conducting a medical history in their native language.


Background
Effective doctor-patient communication is essential for delivering quality healthcare [1][2][3][4][5]. Nevertheless, language barriers remain some of the most reported hindrances that physicians need to overcome in a healthcare setting [6][7][8]. While approximately one third of international medical schools outside the United States and Canada offer education in English, only one-fifth of the countries, within which these institutions are located, list English as an official language [9]. The broad adoption of the English language in medical education largely follows the international medical community that has adopted English as the official language for the majority of medical journals and international conferences and allows greater access to specialty training in Englishspeaking countries [10,11].
Lebanon is a good example of the divergence of spoken and educational languages. Out of the seven registered medical schools in the country, four use English as their language of medical instruction, while the remaining use French; two languages that are not primary speaking languages in the country. The official and most widely spoken language in Lebanon is Arabic, and the majority of patients in academic and community healthcare centers speak it. Lebanon is also unique in having a high proportion of bilingual and trilingual medical students [12]. Both these factors create a unique population of medical students for evaluating the influence of medical education in a foreign language on patient communication in the native language. The present study investigates the association between medical education in a foreign language and medical students' selfperceived confidence in their history-taking skills in their native language.

Participants
All registered medical schools in Lebanon were invited to participate (n = 7). Students rotating in clinical clerkships were included (defined as either Med III or Med IV in schools adopting the American model or Med VI or Med VII in schools adopting the French model). In brief, 3 medical schools incorporate the American model (4-year program), 2 incorporate the French model (7-year program), and 1 incorporates the British model (6year program). English is the language of instruction in schools that follow the American and British models (n = 4), whereas French is the language of instruction in schools that follow the French model (n = 3). Medical schools of the American model include 2 years of preclinical education followed by 2 years devoted for clinical rotations. In contrast, medical schools adopting the French or the British model typically devote 1 or 2 years of clinical observerships followed by 2 final years of clinical training.

Ethics, consent, and permissions
All participants were informed that their participation was voluntary. The study objective was explained to all participants, and verbal consent was obtained. All inperson and online surveys were anonymized and untraceable. No individual subject identifiers were collected. The study protocol was approved for exemption by the local institutional review board a the Lebanese American University and was in accordance with the Helsinki Declaration and its later amendments.

Survey
Students were invited to anonymously fill a standardized 17-question survey. Surveys were administered in a mixed-format approach either in-person or using an online tool based on the requirements for surveys per medical school over a 4-month data collection period. The survey was developed based on a prior study and focused on the linguistic aspects of the students' education and daily life, as well as their self-perceived confidence in their history-taking skills (Table 1) [13]. Questions in the survey were either open-ended, categorical, or ordinal. Open-ended questions addressed the spoken language at home and languages of high school, undergraduate, and medical education. Categorical questions addressed academic level, gender, Arabic as native language, bilinguality, and pre-clinical curricular or extracurricular exposure to conducting a medical history in Arabic. Ordinal questions very. An initial version of the survey was piloted among a subgroup of Lebanese medical students, and questions were either added or modified based on subject feedback.

Statistical analysis
All statistical analyses were performed using Stata® 14 (StataCorp LP, TX, USA). Data was analyzed for all subjects who had an available answer for the outcome (i.e. confidence level in conducting a medical history in Arabic). Characteristics of the study population were evaluated using descriptive statistics. Ordinal variables of interest, including levels of confidence in conducting a medical history in Arabic, were first recorded as 4 categories according to the Likert scale, and were then dichotomized for subsequent analyses. The association between categorical variables was performed using Chisquare test. Cochran-Mantel-Haenszel test was performed to evaluate the association between categorical variables when stratification by clinical year was conducted. To evaluate the odds and independent association with the outcome in each clinical year, mixedeffects logistic regression was performed. Logistic regression models were fit via maximum likelihood and were clustered at the school level (i.e. random school effect) to account for the correlated outcomes for students within the same medical school. All variables that were associated with the outcome on univariate analysis with a p-value ≤ 0.10, as well as variables of historical interest, were included in the logistic regression models. All pvalues were two-sided. A p-value < 0.05 was considered statistically significant.  Table 2 summarizes the characteristics of the study participants. Although all participants had their medical education in a foreign language, 88.5% of participants whose native language was Arabic were either extremely or moderately confident that they could conduct a medical history in Arabic during their clinical years (first clinical year: 87.5% vs. second clinical year: 90.3%, p = 0.38). In contrast, this proportion was significantly lower among participants whose native language was not Arabic (64.9%, p < 0.001). When further subgrouped by clinical year, this difference remained statistically significant among students in their first clinical year (87.5% vs. 57.7%, p < 0.001), but was not significant when participants in the second clinical year were evaluated alone (90.3% vs. 81.8%, p = 0.37). Despite varying results between both clinical years, clinical year did not significantly modulate the effect of Arabic being the native language on Arabic history-taking confidence levels (pinteraction = 0.23).

Results
To evaluate the factors that were independently associated with high confidence levels in conducting a medical history in Arabic in each clinical year, a separate multiple logistic regression model was constructed for each clinical year. Among participants enrolled in the first clinical year, high confidence in Arabic historytaking was independently associated with Arabic being the student's native language and exposure to Arabic history-taking in either pre-clinical years or during extracurricular activities. In contrast, the impact of all these factors on confidence levels was lost when participants enrolled in the second clinical year were evaluated alone (Table 3).
Interestingly, confidence in conducting a medical history in Arabic was irrespective of the participants' confidence in history-taking in the foreign language of medical education. Approximately 87% of participants who were not confident in conducting English/French history-taking were still confident in conducting a medical history in Arabic. This difference was consistent across both clinical years (first clinical year: 80% vs. second clinical year: 95.2%).
Subsequent analysis demonstrated that when stratified by clinical year, pre-clinical curricular exposure to Arabic history-taking was associated with a higher likelihood of a student being engaged in extracurricular activities involving Arabic history-taking (e.g. volunteer services) (first clinical year: 32.1% vs. 21.0%; second clinical year: 31.2% vs. 18.8%, p = 0.005) (Fig. 1). Although the vast majority of participants (82.9%) believe their future patients will speak Arabic, only 64.5% of those perceived that adding courses for communication in Arabic will be beneficial

Discussion
This study demonstrates that during the clinical years, the vast majority (88.5%) of medical students in Lebanon are confident in conducting a medical history in their native language (Arabic) despite having their medical education a foreign language (in either English or French). This proportion was substantially higher in this study as compared with other populations, where confidence levels have previously been reported to be as low as 30% [13]. Although the exact etiology of this contradistinction is unclear, it may be attributed, at least in part, to the unique bilingual nature of the Lebanese population in general, which is not universally common among other studied populations [13][14][15][16].
This study further highlights that effective doctorpatient communication among medical students is largely dependent on their exposure to real-life patients in the hospital setting. Although higher confidence level in Arabic history-taking during the first clinical year was associated with Arabic being the native language, as well as prior curricular and extracurricular exposure to Arabic history-taking, this association was evidently not present among students in their second clinical year. Although the clinical year did not significantly modulate the effect of Arabic being the native language on Arabic history-taking confidence levels, the results from Table 3 suggest that as medical students transition from their first clinical year to their second clinical year, the impact of external factors becomes less relevant in predicting confidence in history-taking skills in their native language. These findings are consistent with prior studies, where medical students perceived that a foreign language of medical instruction may be an obstacle only during the first year of medical school, but was no longer important during the subsequent medical years [17]. Accordingly, this study demonstrates that in Lebanon, medical education in a foreign language does not represent a gap in effective communication between graduating medical students and their patients.
This is the largest representative study among medical students to correlate medical education in a foreign  language to the confidence in conducting a medical history in the native language. Nonetheless, this study is limited by its cross-sectional nature and differences in the sample sizes and characteristics between first and second clinical year students. Although the overall response rate was lower than the majority of in-person surveys, it was higher than most online nationwide surveys conducted among medical students. Finally, the unique characteristics of the Lebanese population, i.e. the majority of whom are raised bilingually, limits the generalizability of the results to other populations.

Conclusions
Despite receiving their medical education in a foreign language, the majority of students in Lebanese medical schools report being confident in conducting a medical history in their native language. In the first clinical year, high confidence levels in conducting a medical history in Arabic was associated with having Arabic as the student's native language, as well as engagement in Arabic history-taking in the pre-clinical years and during extracurricular activities. These factors were not associated with the students' confidence in the second clinical year, suggesting that the experience accrued in the first clinical year overcomes the external elements that may limit their confidence in conducting a medical history in their native language.
Abbreviations OSCE: Objective structural clinical examination