Prevalence of substance abuse and its associated factors among medical students: a cross-sectional study [version 2; peer review: 1 approved]

Background: Substance abuse by medical students poses a significant risk not only on their personal life but also on their professional life by putting their patients at risk. Our study aims to find out the prevalence of substance abuse and factors associated with it among medical students. Methods: An online-questionnaire based, cross-sectional, analytical study was carried out at Gandaki Medical College, a medical school in Gandaki Province of Nepal. Whole sampling technique was used. SPSS-25.0 software was used for statistical analysis. Descriptive statistics were used for analyzing demographic and substance abuse related variables. Chi square and Fishers exact test was used to find out if any differences existed in substance abuse by demographic characteristics. Results: Among 420 eligible participants, 219 participated in the survey (response rate: 52.2%). Alcohol was the most commonly used substance with overall lifetime prevalence of 58% followed by cigarette smoking and illegal drug abuse at 21.9% and 13.7%, respectively. The abuse of substances was significantly higher (p<0.05) in males, who were offered substances by friends, and amongst study participants who worked with an intoxicated colleague. The most common reason for substance abuse was “For pleasure, curiosity or to go along with friends” (23%, n=18). Major and minor dysfunction due to alcohol abuse was reported by 8(3.7%) and 33(15.1


Introduction
Substance abuse refers to the harmful or hazardous use of psychoactive substances, including alcohol and illicit drugs and is one of the major problems globally. 1 According to the Global Burden of Disease Study 2017, there were 585,000 deaths and 42 million years of "healthy" life lost as a result of the substance abuse. 2 The World Drug Report 2019 estimated 271 million people, or 5.5% of the global population aged 15-64 years, had used drugs in the previous year, while 35 million people are estimated to be suffering from drug use disorders. 3 Medical students, the future healthcare professionals to be involved in the task of diagnosing and treating substance abuse are not immune to drug abuse, and the history of substance abuse within the medical fraternity is not new. 4 Substance abuse by medical students poses risks and can also have serious consequences on their effectiveness and fitness to practice as good clinicians. 5 Substance abuse by health care professionals can seriously influence their professional behavior, compromising the standard of delivered health care services and in turn, placing their patients at risk. 6 It is believed that substance abuse among physicians starts early in their careers, and the importance of studying the lifestyles of medical students to detect substance abuse is well recognized. 7 Researchers have suggested that substance abuse in medical school may be the root of the ongoing problem of increased substance abuse in practicing physicians. [8][9][10] Data concerning substance abuse by medical students is scarce in Southeast Asia, especially from Nepal. This study was conducted with objectives to find out the prevalence of various types of substance abuse and factors associated with it, including consequences among Nepalese medical students of a medical college in western Nepal. The findings from our observation could be useful for developing strategies to minimize the use of substance by the medical students.

Study design and setting
A cross-sectional analytical study was carried out among medical students (1 st to 5 th year) enrolled at Gandaki Medical College Teaching Hospital and Research Center (GMCTHRC), Nepal from 1 st June 2020 to 1 st July, 2020. Whole sampling technique was used. Data was collected from the study participants through a questionnaire sent via Google form which was sent to their email through class representatives of each year. Students who were not available through email contact or did not consent for their participation in the digital consent form were not included in the study. The participation was completely voluntary and no incentives were offered to the study participants. Anonymity was maintained throughout.

Study sample
All the medical students enrolled in the Bachelor of Medicine and Bachelor of Surgery (MBBS) program at GMCTHRC (n = 420) were included in the study. The study objectives and its implications were explained to the participants through the questionnaire sent via Google form on their email. The online method was preferred as physical classes had completely been cut down by the College administration in line with the COVID-19 protocols set up by the Government of Nepal.

Study instrument
Data was collected using a modified, online version of the Health Professional Questionnaire developed by Kenna and Wood. 11 Permission was taken from Elsevier for using the questionnaire in our research. This tool was preferred as it had already been used successfully in a similar study among other healthcare students in Nepal. 12 The final version of the survey did not require major changes and contained demographic information as well as the details and effects of substance abuse. We did not include the questions about university of study, faculty, course undertaken and religious beliefs that were originally present in the questionnaire.

Demographic characteristics
The study participants were asked about their age, gender, year of study, place of stay and if there was a problem of alcohol and drug abuse in their family.

REVISED Amendments from Version 1
We have revised this as per reviewer comments. We have added a short paragraph about the factors associated with substance abuse in the discussion section. In addition, we have also added further scope and recommendations from our study.
Any further responses from the reviewers can be found at the end of the article

Substance abuse related characteristics
We asked the study participants about alcohol, cigarette and illegal as well as prescription drug use. The questions "how many times, if ever, have you smoked cigarettes?" and "recognize yourself as: Nondrinker/infrequent drinker/light drinker/moderate drinker/heavy drinker" were used, respectively, for lifetime cigarette smoking and alcohol use. The participants were asked about their lifetime experience and tendency of use of both prescription as well as illegal drugs, e.g. "how many times, if ever, have you smoked marijuana or hashish?" Participants were asked to rate the frequency of their use, varying from "0" times to ">30 times". The prescription drugs under consideration were major opiates, minor opiates, stimulants, sedative-hypnotics, tranquilizers, anxiolytics and other prescription pain medicines. Illegal drugs included marijuana, cocaine and designer drugs such as ecstasy, LSD, Meth and Ketamine.
The study participants were also asked about how often they were offered alcohol and drugs by friends and how often they worked with colleagues who were intoxicated by such substances. We also asked about the reasons for substance abuse and the dysfunction it had on their daily life using a series of questions. The list of dysfunctions included 1) falling behind in work; (2) calling in sick or being late; (3) having trouble getting along with people; (4) worrying that you might be using too much or too often; (5) seriously considered suicide; (6) having an automobile accident or other type of accident; (7) seeing a psychiatrist, psychologist, or a counselor. 12 The initial four effects were grouped as "minor dysfunction" and the remainder as "major dysfunction". Subsequently, the minor and major dysfunction due to alcohol and drug abuse was calculated.

Statistical methods
The data collected through Google forms were extracted to Microsoft Excel-13. Any response with missing data was excluded from the study. After data cleaning, it was imported and analyzed by using SPSS (Statistical Package for Social Sciences) 25. Demographic variables and substance abuse were analyzed using descriptive statistics. The frequency of use was recorded as 0 = never used and 1 = any use. Chi square test was used to examine any differences in substance abuse by demographic characteristics. Fisher's exact test was applied for cell count less than 5. P value for statistical significance was set as less than 0.05.

Ethical consideration
This study was approved by the Institutional Review Committee of GMCTHRC (Ref no: 023/2076/2077). All the participants were informed about the study through the online form attached in the questionnaire sent to them and informed consent was obtained. Anonymity of the study participants was maintained throughout. Table 1 shows detailed demographic and general characteristics of study participants. Out of 420 medical students, 219 students participated in the study (response rate = 52.2%). There were considerably less participants from nonclinical years (n = 56, 35%) of medical school when compared to clinical years (n = 163, 63%). The mean age of study participants was 21.32 years, ranging from 18-30 years. Majority of respondents were females (n = 115, 52.5%) and most of them lived out of a rented property (n = 118, 53.9%). Table 2 shows the detailed prevalence of various substances abused by medical students. Alcohol was the most commonly abused substance with overall lifetime prevalence of 58% (n = 127), followed by cigarette smoking and illegal drug abuse at 21.9% (n = 48) and 13.7% (n = 30), respectively.

Correlates of substance abuse
When analyzing the distribution of substance use with various demographic variables, it was shown that the abuse of substances was significantly higher (p < 0.05) in males, if substances were offered by friends and amongst study participants who worked with an intoxicated colleague (Table 3).

Reasons for substance abuse
Out of all the study participants, 78 respondents stated the reason for using substances. The most common reasons included: 1) for pleasure, curiosity or to go along with friends (n = 29, 37%); 2) supervising self for medical condition (n = 18, 23%); and 3) originally prescribed but now using on my own (n = 14, 18%). Table 4 shows the details regarding other reasons for substance abuse among the study participants.   To stay awake, perform better or lose weight 5 6 Can't quit their use 2 3 Originally prescribed but now using on my own 14 18 Effect of alcohol and drug abuse As stated earlier, we grouped the harmful effects of substance abuse into "major" and "minor" dysfunction. Details of these dysfunctions due to alcohol and drug abuse have been presented in Table 5. Alcohol caused dysfunction in 41 respondents while drugs caused dysfunction in 15 respondents respectively.

Inappropriate use of substances
In addition to all the other questions pertaining substance abuse, we also asked the study participants if they felt they used alcohol or drugs more than they would consider appropriate (Table 6). A total of 84% of the study participants (n = 184) responded that they didn't feel they did, while 15.1% (n = 33) participants felt they used alcohol inappropriately. Only one participant each felt they used drugs and both substances inappropriately.

Discussion
From our study, the prevalence of substance abuse was found to be 58% for alcohol, 21.9% for cigarette smoking and 13.7% for illegal drug use. Alcohol was the most commonly abused substance. Peer influence and sharing the workspace with an intoxicated colleague were significant predictors of substance abuse by medical students in our study. The most common reason for substance abuse was for pleasure, curiosity or to go along with friends. Our study participants also reported both major and minor dysfunctions due to substance abuse. The dysfunctions were higher among the alcohol using group and a majority of study participants admitted that they used alcohol inappropriately.
The overall prevalence of substance abuse among medical students varies from as low as 3% to as high as 84%. 13 The prevalence varies widely between the geographic regions with higher rates in the more developed regions like Europe and North America. An extensive study carried out in medical students across the south-east Asian region reported the prevalence of smoking to be 31.7%. 14 A multi-centric cross-sectional study carried out in India states the prevalence of substance abuse among medical students to be 31.5%. 15 The precise prevalence of substance abuse is currently not known in Nepal for both the general population as well as the healthcare students. Among various single-institution studies in Nepal, the prevalence of substance abuse by medical students varies from 28% to about 64%. 12,14,[16][17][18] In a multi-centric study carried across three universities of Nepal, the overall prevalence of substance abuse among healthcare students was stated to be 42.8% 12 but this study included only paramedical and allied science students. Nevertheless, comparison with available data showed that prevalence of substance abuse was higher in our study population.
In our study, male gender, peer influence and working with intoxicated colleagues were significant predictors of substance abuse among the study participants. Males are usually more commonly associated with abusing substances and this has been reported both in studies from Nepal and abroad. [15][16][17] Peer influence and sharing the workplace with abusive colleagues have also been reported predictive of substance abuse. 12 These findings are important and interventions targeting substance abuse prevention and control should primarily focus on male students and friends/colleagues of such students.
Medical students are the future doctors of the nation. That the high rate of substance abuse and its effects among medical students may significantly impair their learning ability and clinical judgement as physicians, affecting patient outcomes, cannot be denied. Further, substance abuse by medical students tarnishes the image that physicians have held for a long time as healers. Now is high time that this problem is seriously looked upon by stakeholders and policymakers for production of efficient and competent physicians both locally and globally. Further studies should be conducted at more extensive level, if possible including all the medical schools of Nepal to exactly find out the magnitude of this important problem among Nepalese medical students. We recommend that peer-support and awareness programs against substance abuse be conducted regularly in medical schools. Provision and strict implementation of a punishment system may also act to control substance abuse. For those already addicted to the use of substances, psychosocial interventions, self help groups and pharmacotherapy may be effective interventions and medical schools should try to provide these facilities for prevention and control of substance abuse among their students.

Limitations
Due to the COVID-19 pandemic, the study was carried out through an online questionnaire. In a country where the internet penetration is only 57% and majority of students were stuck at home due to travel restrictions imposed by the Government of Nepal, this may have led to a decreased response rate in our study. 19 Further, ours is a sensitive topic and there is a risk of response bias in our study. Another limitation is that our study is a single center study and the results from our study may not be generalizable across other parts of the country.

Conclusion
The prevalence of substance abuse was high among medical students and was found to be associated with male gender, if offer of substances were made by friends and amongst those who worked with an intoxicated colleague. Substance abuse by medical students may affect their professional career as independent physicians and significantly alter patient outcomes. Peer support programs and reward-punishment system may be effective interventions to curb this problem.

Data availability
Underlying data In this study, participant-level data were collected. No data are available publicly because consent for publication of raw data was not obtained and the dataset could in theory pose a threat to confidentiality. Researchers interested in accessing the data will need to submit an official letter of request for the data to Gandaki Medical College Institutional Review Committee, and will be asked to confirm that they will not violate the ethical standards of the ethical committee and protect the anonymity of the participants. Researchers can contact the corresponding author, who can facilitate this process.

Reporting guidelines
This article was written in accordance with the STROBE guidelines.