Distribution of imposter syndrome among medical students of Bangladesh: a cross-sectional study [version 1; peer review: awaiting peer review]

Background: Imposter syndrome (IS), suffering from self-doubt and fear, despite clear accomplishment and competencies, is often detected in medical students and adversely affects the well-being of the student. This study aimed at assessing the prevalence of IS among public and private medical students in Bangladesh. Methods: This study was a cross-sectional design among medical students in Bangladesh. Data were collected between February to July 2020 through snowball sampling technique across medical colleges in Bangladesh. Relative risk ratios (RRRs) with 95% CI were calculated to investigate the magnitude of association between imposter syndrome exposure and explanatory variables. Results: A total of 500 students participated in this study with approximately 47% and 53% of students studying at public and private medical colleges, respectively. Around 32% of medical students were exposed to IS (47% of public and 53% of private medical college students). Medical students were the most significantly associated with IS for third (RR: 1.487, CI: 1.068-2.071) and fourth-year students (RR: 1.493, CI: 1.043-2.136). Overall, we found that respondents aged 22 to 25 were 3.6% (RR:1.036, CI:0.801-1.339) more likely to be suffering from IS than their younger counterparts. Conclusion: Third and fourth-year medical students, in particular, Open Peer Review Reviewer Status AWAITING PEER REVIEW Any reports and responses or comments on the article can be found at the end of the article. Page 1 of 11 F1000Research 2021, 10:1059 Last updated: 20 OCT 2021


Introduction
Imposter syndrome (IS) is characterized by a feeling of not belonging, out-of-placeness, and the belief that one's perceived competence and success by others is undeserved. Usually, this is considered an individual problem that should be addressed by keeping a record of accomplishments as a reminder of progress (Breeze, 2018;Mak et al., 2019;Hendriksen, 2015). In late 1978, clinical psychologists Pauline Clance and Suzanne Imes first cited the IS which arbitrated the relationship between perfectionism and anxiety and partially influences perfectionism and depression (Clance & Imes, 1978) A more recent systemic review conducted in the U.S. in 2020 found that the prevalence of IS ranged from 9% to 82% in the general population (Bravata et al., 2020), while another review conducted in the U.S. in 2020 found that it ranged from 22% to 60% among physicians and from 33% to 40% among trainee physicians (Gottlieb et al., 2020). In the United States, recent studies on IS have found that 57% of pharmacy students (Sullivan & Ryba, 2020) and 15% of medical students have IS (Holliday et al., 2020). Indeed, IS is becoming a growing public health issue both globally and regionally. For instance, the prevalence of IS among medical students has been found to be 30% in the US (Villwock et al., 2016), 45.7% in Malaysia (Ikbaal & Salim Musa, 2018, and 47% in Pakistan (Qureshi et al., 2017).
The evidence indicates that medical students experienced moderated-to-strong IS (Levant et al., 2020), which has psychological and academic consequences (Ikbaal & Salim Musa, 2018;Qureshi et al., 2017;Villwock et al., 2016). The move to the clinical phases of study can be particularly challenging and cause students to suffer from low confidence during this time (Burstein et al., 1980;Niemi & Vainiomäki, 2006). In addition, many studies confirmed that IS often negatively affects medical students' physical and academic well-being (Ikbaal & Salim Musa, 2018;Levant et al., 2020;Villwock et al., 2016). Thus, they may miss out on opportunities because they do not recognize their potential.
There have been many studies on IS among medical students conducted across the world. However, no study has yet been published on IS among medical students in Bangladesh. Therefore, this study aims to draw attention to the growing need for more mental health research within Bangladesh. The current study focused on evaluating the prevalence of IS, exploring the frequency distribution of IS in different institutions, and assessing the factors contributing to IS among public and private medical college students. The survey provides insight into areas of commonality and differences between IS in students.

Ethical approval
Ethical approval was obtained from the Institutional Review Board (IRB) at North South University, Dhaka, Bangladesh. The study objectives were explained to each participant and confidentiality, and anonymity was assured, and written informed consent was obtained from respondents before proceeding.

Study design and settings
This study was conducted using a cross-sectional survey assessing the presence of IS among medical college students in Dhaka city. Data were collected from medical colleges (public and private) in Dhaka between February 2020 and July 2020. Our study did not include any students studying in medical colleges located outside of Dhaka city, Bangladesh. We collected more samples than the recommended minimum sample size for better accuracy and results. So, throughout our survey, we collected a sample size of 500 students selected from public and private medical colleges in Dhaka. Our survey was conducted using a snowball sampling method (a non-probability sampling method) to achieve the desired sample size. The medical colleges chosen for the study are attended by students from across the country, allowing this study to provide an understanding of the prevalence of IS among medical students across Bangladesh. Participants of the study were the students of both public and private medical colleges ranging from 1 st to 5 th year of study. Initially, we identified some potential IS cases among our respondents and asked them to recommend other people for the study. After this, we contacted them via email or mobile phone and collected data through face-to-face interviews. The eight items from the Young Imposter Syndrome (YIS) scale (Villwock et al., 2016) were used to assess whether the participant had IS or not. The scale was in the form of eight questions, and a student was considered to have IS if they answered 5 or more questions as "Yes".
Public and private medical colleges represent the whole community of medical students, and our study aimed to find out the underlying risk of having IS among medical college students. This condition is less discussed in Bangladesh, so we aimed to reveal such mental conditions and associated risks within the study group.

Dependent variable
In our study, we used IS as the dependent variable. The dependent variable contains two categories, "Yes" (has IS) and "No" (does not have IS). We defined them as follows: In our survey, we used Young Imposter Syndrome (YIS) scale to assess the presence of IS. The scale was in the form of eight questions. A student was considered to have IS if they answered 5 or more questions as "Yes" in the YIS scale, otherwise considered "No".

Independent variables
Participants' sociodemographic factors and some academic data were considered as independent variables. The sociodemographic variables included gender (male, female), age group (18-21, 22-25 years), smoking status (current, past, never), and living with family (yes, no). Body Mass Index (BMI) was categorized using the WHO BMI standard scale (Nuttall, 2015). The economic condition was determined by the monthly family income (MFI) of the respondents. The MFI was categorized as ≤20,000 BDT (approximately US$237), 21,000-30,000 BDT, 31,000-40,000 BDT, ≥41,000 BDT. This MFI interval was a self-imposed category. The academic year was categorized as first, second, third, fourth, or fifth, and the reasons for studying in medical college were classified as own preference, family preference, failing to qualify for another department or better job opportunities. Each response was dichotomized (yes, no). All relevant information was directly collected from each participant.

Survey procedures
In our study, we collected the desired data through snowball sampling. This sampling method was used because of the lack of a sampling frame and the unavailability of targeted samples. Using the (Israel, 1992) formula (i), we obtained the minimum sample size required for the study.
Where, n 0 is the initial sample size, z is the standard normal variate, and p is the (estimated) proportion of the population with the attribute in the question, i.e., p + q = 1, d is the allowed maximum error in estimating a population proportion. Considering the degree of accuracy, d = 0.05 and p as the approximate proportion of YIS (p = 0.47), taken from a related study conducted in Pakistan (Qureshi et al., 2017). So, the calculated sample size was 382; however, we targeted more subjects to be included than the required sample size. Finally, we collected complete information from 500 unique samples for our study, and incomplete questionnaires were not considered for analysis.
Design of the study Data were collected through a face-to-face interview. Participants were reassured that all the information collected would be kept strictly confidential and would not be used for anything other than research purposes. Written consent was taken for the study from the participants. After this, we briefly introduced the student to our research and presented them with the questionnaire. The response was scribed immediately. The collected data was compiled for further processing and analysis. After cleaning the raw data, we had 500 complete observations for our study.

Statistical analysis
Statistical analysis was performed with Statistical Package for Social Science (SPSS) version 26 (IBM Corporation, Armonk, NY, USA). And Stata 16.0 for windows (Stata Corp LP, 4905 Lakeway Drive College Station, USA). Throughout our study, we collected only the completed questionnaire and discarded any incomplete questionnaire. To observe the background of the study, we used frequency distribution which is presented in Table 3. For the bivariate study, we used a relative risk ratio (RRR) with a 95% confidence interval to determine the risk of having IS, as presented in Table 4.

Results
We selected 500 students among the medical colleges, and the questionnaires were collected from each individual during face-to-face meetings. Therefore, the response rate was 100%. Out of the 500 participants, 233 (46.6%) were public, and 267 (53.4%) were private medical college students. Detailed information on how respondents addressed all eight questions (Villwock et al., 2016) is displayed in Table 1.
The students who answered "Yes" for five or more out of eight questions were considered positive for IS. In total, 61 (32.2%) out of 500 students answered "Yes" for five or more questions, and 339 (67.8%) answered "No" according to the definition of the YIS scale. A breakdown of YIS scores in relation to the institution is given in Table 2.

Univariate analysis
For univariate analysis, we used frequency distributions and percentages to compare the variables within the study, as presented in Table 3.  In our study, 161 (32.2%) out of 500 observations were suffering from YIS, while the rest of the respondents (339; 67.8%) were students without YIS. This indicates that around one-third of our total respondents were experiencing symptoms of YIS. Approximately 53% of our respondents were from private, and 47% were from public medical colleges. Most of the respondents were young; 58% were between 18 and 21 years old, and 42% were between 22 and 25 years old. Overall, 269 (53.8%) out of 500 respondents were male, and 231 (46.2%) were female. The majority of participants (191; 38.2%) of our study had a healthy BMI (between 18.5 and 24.9). Most of our respondents (208; 41.6%) replied that they study medical science because of their own preference, and 197 (39.4%) study because their family wanted them to. Most of our respondents (361; 72.2%), were non-smokers, and the rest were smokers or had smoked in the past. Most of our respondents (306; 61.2%) lived without family, and the rest lived with the family.

Bivariate analysis
In order to find out the risk of having YIS, we calculated relative risk along with 95% confidence interval, and this is presented in Table 4.
In our study, respondents who study in public medical colleges were 1.21 times or 21% (RR: 1.219, CI: 0.946-1.572) more at risk of having IS than private medical colleges. Respondents aged 22 to 25 years are 3.6% (RR: 1.036, CI: 0.801-1.339) more at risk of having IS than their younger counterparts. Our study found that female respondents were 4.1% (RR: 1.041, CI: 0.807-1.343) more at risk of having IS than male respondents. Surprisingly, we found that students who had a lower than healthy BMI score (between 18.5 and 24.9) were 9.6% (RR: 1.096, CI: 0.649-1.849) more at risk of having IS. We also found that students whose BMI was higher than the healthy range were at lower risk , of having IS than those who took medical science as their desired field of study. Students whose MFI was between 31000-40000 BDT had a 10% (RR: 0.902, CI: 0.486-1.675) reduction in risk of IS compared to those who had a MFI between 21000-30000 BDT. Furthermore, students whose MFI was ≥41000 BDT had a 23% (RR: 0.777, CI: 0.425-1.420) reduction in risk compared to those who had a MFI ranging from 21000-30000 BDT. In our study, we also found that students who smoked were 13% (RR: 1.133, CI: 0.816-1.575) more susceptible to IS than those who had never smoked, but past smokers had relatively low risk (RR: 0.982, CI: 0.650-1.484) of having IS, compared to those who have never smoked. Finally, respondents who lived with family were 4.1% (RR: 1.041, CI: 0.803-1.349) more at risk of having IS than those who didn't live with the family.

Discussion
IS affects medical students from all different medical colleges and circumstances. In this study, we found nearly one-third (32.2%) of the respondents have IS, 85 out of 184 were male, and 76 out of 155 were female students. Whereas, in a study conducted by Qureshi et al., in 2017 in Pakistan, almost half (47.5%) of the participants had IS, among 143 students where 23 were males, and 45 were females (Qureshi et al., 2017). Generally, female students were more likely to had IS. In some research, women were twice as likely to suffer from it than men (Farzana et al., 2017;Maqsood et al., 2018;Qureshi et al., 2017). Our observations were also similar, and the female respondents had more IS (32.9%) than males (31.6%). The leading cause was that women tended to address IS by confronting their concerns and working harder to address the concerns, whereas men tended to avoid areas where they felt uneasy (Villwock et al., 2016).
According to our study, fourth-year students were more likely to suffer from IS, with 39.7% meeting the criteria, the largest among all academic years. Most likely, fourth-year students faced more pressure than others because of the increased academic curriculum, worries about study and performance, and concerns about the future (Farzana et al., 2017;Villwock et al., 2016). A prior study of IS conducted in 2017 showed that among Pakistani medical college students, 23.1% of respondents had an IS score of 1 (IS score <5 means does not possess IS, ≥5 means they might possess IS), with males making up 27.1%, and females 20.2%. In the same study, 4.9% of all the students had an IS score of 4, among which 5.1% were male, and 4.8% were female (Qureshi et al., 2017). Similarly, in our study, the most frequented IS score was 4, with 30.8% of respondents scoring 4. Of those participants, 27.9% were public, and 33.3% were private medical college students. Only 0.8% of students scored 8 (meaning they might possess IS), with 0.9% from public and 0.7% from private medical schools.
According to Eva et al., private medical college students were more relaxed and confident than public medical college students (Eva et al., 2015). However, our study suggested that both public and private medical college students lacked confidence and suffered from IS. Indeed, in our study, 60.7% of private and 52.4% of public medical college students answered "yes" to the question "Do you live in fear of being found out, discovered, or un-masked?".
An earlier study conducted in the U.S. in 2016 showed that almost 25% of male and nearly 50% of female students analysed had IS and that IS was significantly associated with burnout or dropout (Villwock et al., 2016). Every year, many students drop out or burn out and suffer from anxiety and other psychological problems (Eva et al., 2015).
According to the findings of Maqsood et al. (2018) the fourth-year medical students suffered more severely with IS than others, reflecting our own findings. They also found that 27.41% of the third year and 46.31% of fourth-year medical students had moderate IS and that 64.51% of the third year and 47.36% of fourth-year students had severe IS (Maqsood et al., 2018). In 2003, a survey of the mental health of medical students in Iran examined the age and health status of the respondents and using the 12-item General Health Questionnaire (GHQ-12) (Montazeri et al., 2003), 30.6% of first-year and 30.6% of fourth-year medical students were suffering psychological distress (Hendriksen, 2015).
The generalisability of our findings on IS among medical college students studying in Dhaka, to other city students is unknown. Studying only medical college students in Dhaka, may have resulted in an underestimation of the coverage of IS. Similarly, investigating the underlying causes was beyond the scope of our study. Our study presents the risk of IS at the time of our assessment, however, the situation in Bangladesh has likely further deteriorated because of the COVID-19 pandemic. Given that the COVID-19 situation has forced Bangladesh's education system to close for nearly two years, there is a greater need that ever to conduct surveys to monitor mental health issues such as IS amongst students.
Our presented data and our risk analysis showed higher BMI, and higher MFI coincided with lower risk of IS. So, ensuring appropriate diet and proper socio-economic improvement, may help reduce the risk of IS and other mental health problems among medical students. The hope is that this study will encourage more research on this topic and a greater understanding of the impact of IS on various aspects of the social, academic, and daily life of those who suffer from it.

Strength and limitations
This study advances knowledge in several ways. This research is among the first to examine the risk of IS among medical college students in Bangladesh, which is not a commonly discussed syndrome within this country. This research also addresses further need for inquiries into the psychological predictors of IS, helping to understand the reasons and the causes behind IS. IS is an increasingly prevalent psychological syndrome. However, it's possible, when looking for participants for a mental health study, potential volunteers might be hesitant to participate due to the personal nature of the topic. As a result, snowball sampling was utilized in this study to reach the population, because it is difficult to collect samples using other sampling methods.
While this study makes several contributions to IS research, there are some limitations. This study used the snowball sampling method, which is a non-probability type sampling method. The researchers had little control over the sampling approach. We only interviewed those who were recommended by the previous respondents. Therefore, the sampling approach was limited and subjective. There might be some sampling bias present in this study because of the nature of the sampling method. Because of the sampling bias, our samples might not represent the true distribution of the population. Similarly, our study only sampled Dhaka city; it would've been better if we'd studied the entire country. Consequently, we would have been able to get a better picture of the distribution of IS and the underlying factors that influence it.

Conclusion
Our analysis revealed a serious public health issue for medical college students in Bangladesh. As a result, students may suffer from depression and, in some cases, even commit suicide (Qureshi et al., 2017). Public and private medical institutions should take that matter seriously and establish various programs, organizations, counseling teams, and other activities to help identify students suffering from IS and support them. It is evident that students, especially those in their third and fourth year of medical college, need more care than others to ensure they stay focused on their goals. Faculty and fellow students should guide, mentor, and encourage IS suffers to produce self-confident and self-sufficient students.  (Shahjalal, Md et al., 2021b) This project contains the following extended data:

Data availability
-Copy of questionnaire used in study Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

The benefits of publishing with F1000Research:
Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com