Exploring the relationship between health concerns and high‐risk behaviours in Medical Sciences' students

Abstract Aim The need to cope with life concerns may drive an individual to resort to high‐risk behaviours. This study aimed to determine the relationship between health concerns and high‐risk behaviours. Design A cross‐sectional study. Methods We sampled 926 Medical Sciences' students from the North of Iran from September–December 2017 using the stratified sampling method. Results The most and less common high‐risk behaviour was physical inactivity (97.5%) and high‐risk sexual behaviour (15.7%), respectively. In multivariate logistic regression analysis, participants' concerns about human sexuality (AOR: 1.39; CI: 1.22, 1.57), injury prevention and control (AOR: 1.12; CI: 1.01, 1.20), nutrition (AOR: 1.13; CI: 1.02, 1.26) and emotional health (AOR: 1.08; CI: 1.02, 1.15) increased the odds of risky behaviours. Among Medical Sciences' students, health concerns are linked with risky behaviours. The result of this study can be used to develop relevant interventions targeting mental health to reduce risky behaviour among youth.


| INTRODUC TI ON
High-risk behaviours among youth have grown into one of the leading and most widespread concerns of society despite attempts over the past three decades to address these problems (Duell et al., 2018;Hayashi et al., 2018;Stormshak et al., 2019). The literature elucidates 71% of the causes of death among 10-24 years old, which can lead to chronic diseases in youth (Saffier et al., 2017;West, 2017).
Intentional and unintentional injuries, tobacco use, alcohol and drug use, high-risk sexual behaviours, unhealthy dietary behaviours and physical inactivity are some examples of such behaviours (Ahmadi-Montecalvo et al., 2019;Sorush et al., 2018).
A cross-sectional study among 977 Iranian university students showed that during the last year alone, Hookah use (16.1%) and cigarette smoking (10%) had increased dramatically among youth. The frequency of alcohol and illicit drug use in the same group during the lifetime was found to be 11.9% and 8.8%, respectively (Afrashteh, Ghaem, Abbasi-Ghahramanloo, & Tabatabaee, 2017). Although the prevalence of high-risk sexual behaviours is controversial in the Iranian context, some studies reported extramarital sexual relationships among Iranian youth to be about 8%-41% (Mahmoodi et al., 2019;Rahmati-Najarkolaei & Kamalikhah, 2014;Vakilian, Mousavi, & Keramat, 2014).
High-risk conduct may appeal to individuals as a means of coping with concerns in their life (Kim, Kim, & Park, 2020;Noel et al., 2013;Soleimani et al., 2017). This issue highlights the need to examine the association between health concerns and high-risk behaviours, as it can improve our understanding of the stimulants of risky conduct (Baheiraei, Khoori, Ahmadi, Rahimi Foroushani, & Ghofranipour, 2014;Teychenne et al., 2015). Although high-risk behaviours are relatively common in the general population with psychological distress (Harakeh, Engels, Monshouwer, & Hanssen, 2010;Murphy et al., 2001;Tavolacci et al., 2013), less is known about the association between psychological distress and high-risk behaviour among Medical Sciences' students.
There are several reasons why this issue should be given special attention in university students. According to the existing literature, emerging adults (people between 18-25 years of age), who are represented widely by university students, are believed to have higher tendency to adopt unhealthy behaviours than the other age groups (Laska et al., 2009;Tavolacci et al., 2013). Also, it is stated that following entering to the university and the more freedom from dayto-day parental control, the more participation in the peer groups, the more being in diverse social situations, the university students' worries and distresses increase and predispose them to high-risk behaviours (Basic & Erdelez, 2015;Darling, McWey, Howard, & Olmstead, 2007;Griggs & Crawford, 2019). Finally, it showed that university students experiencing multiple forms of distress simultaneously as a result of multifactorial changes like the transition from leaving home and high school and adjusting to student life (Dyrbye et al., 2011;Hope & Henderson, 2014;Reis et al., 2019). Although it is anticipated that Medical Sciences' students adopt healthier practices as a result of better information concerning health issues (Chourdakis, Tzellos, Papazisis, Toulis, & Kouvelas, 2010), the literature is controversial (Arroyo et al., 2004;Yahia et al., 2008). Also, according to existing studies, medical students are more concerned due to reasons such as perfectionism and some aspects of medical education that are aimed to graduate knowledgeable, skilful and professional physicians. Studies suggest that mental health worsens after students begin medical school and remain poor throughout the training which consequently can predispose them to risky behaviours (Dyrbye et al., 2005;Jönsson & Ojehagen, 2006). So, the research in this group has some implication for health policy makers.
Investigating the related factors to risky behaviours early in life may help decrease the burden of non-communicable diseases in adult life and thus decrease the pressure on society and the healthcare system. Therefore, this study was implemented to assess the relationship between health concerns and high-risk behaviours in the students of Mazandaran University of Medical Sciences (MAZUMS).

| Study setting and data collection
This cross-sectional study was conducted at MAZUMS located in Mazandaran province, North of Iran, from September-December 2017. Initially, in this three-stage stratified proportional to size sampling method, the sample size of each school of MAZUMS (Medicine, Dentistry, Pharmacy, Public Health, Nursing, and Allied Medical Sciences), as main strata, was obtained by dividing the total number of the students in that school by the total number of students in the university. Then, the number of samples allocated to each school was categorized based on the number of the male and female students in each school, alongside with their respective academic year. In turn, some classrooms in each school were selected randomly with the assistance of the random number table. Finally, in each classroom, all of the volunteer students who were present on the day of the research team's visit were invited to complete the self-administered instruments. The survey was conducted by a trained reproductive health researcher. The sample size was determined to be 970 subjects with a 95% confidence level (α = 0.05), 90% power (β = 0.01) and 0.13 correlation coefficient between mental health and high-risk sexual behaviour of a former study (Rosario et al., 2006) and with the assistance of G Power software. To be eligible for participation, students had to meet the following inclusion criteria: studying at the undergraduate level at the time of the survey and being in the age range of 18-25 years old. No exclusion criteria were applied.

| Study instruments
The original version of the Youth Risk Behavior Surveillance System questionnaire was developed in 1990 by the Centers for Disease Control and Prevention to monitor priority health risk behaviours that contribute markedly to the leading causes of death, disability and social problems among youth and adults in the United States (Kann, 2016). This self-report questionnaire contains 84 questions in six domains, including intentional and unintentional injuries (7 and 12 questions, respectively), tobacco use (12 questions Prevention (Kann, 2016). For example, students were considered high-risk if he/she had at least one or more involvement in unintentional and intentional injury behaviours (during the past 30 days before the survey), had ever tried cigarette smoking (even one or two puffs), had at least one drink of alcohol even once in a lifetime (i.e. ever drank alcohol), had used illegal drugs even once in a lifetime, had sexual intercourse with four or more persons during their lifetime or had any kind of high-risk sexual behaviours during their last intercourse. Unhealthy dietary behaviours were defined as if did not consume milk, vegetables, fruit or 100% fruit juices or had drunk a can, bottle or glass of soda or pop (not counting diet soda or diet pop) one or more times per day during the 7 days before the survey.
Insufficient physical activity was defined as not participating in at least 60 min of physical activity on at least one day during the 7 days before the study (Baheiraei, Hamzehgardeshi, Mohammadi, Nedjat, & Mohammadi, 2013;Baheiraei, Khoori, et al., 2013).
For measuring health concerns of Medical Sciences' students, Adolescent Health Concerns Inventory that was developed by Weiler in 1990 was used (Weiler, Sliepcevich, & Sarvela, 1993). This inventory contains 150 items and 12 subscales regarding concerns about: substance use and abuse, such as use of alcohol, cigarettes, diet pills (15 items); diseases and disorders such as diabetes, cancer, sleep problems (17 items); the environment such air pollution, water pollution, littering (11 items); health services such as the cost of medical care, health advertising, healthcare confidentiality (7 items), human sexuality such as close and intimate relationship with the opposite sex, talking about sex (11 items); personal health such as acne, attractiveness to others, body shape (13 items); nutrition such as eating a balanced diet, eating fast foods (8 items); injury prevention and control such as auto accidents, preventing sport injuries (13 items); social health such as discrimination, military conflict, unemployment (16 items); relationships such as falling in love, getting married, having a role model (14 items); emotional health such as dying, having faith in a religion, loneliness (17 items); and concern about the future such as being successful, getting good grade (8 items). Each item is scored dichotomously as either zero or one, and the overall score ranges from 0-150. Higher scores indicate more health concerns.
The internal consistency of the Persian version of the Adolescent Health Concerns Inventory, as measured by Cronbach's α, ranged between 0.68-0.89 for 12 subscales and 0.96 for the total inventory. The Kappa coefficients of all items were between 0.40-0.75 or above 0.75 which indicated good to excellent agreementions of the inventory (Baheiraei, Hamzehgardeshi, et al., 2013;Baheiraei, Khoori, et al., 2013).
The reliability and the consistency of these instruments in the present project were assessed with participation of 20 university students. In this way, Cronbach's α 0.76 and CCI 0.88, (p < .001), were established for the Youth Risk Behavior Surveillance System questionnaire and Cronbach's α 0.94 ICC 0.74, (p < .001), for the Adolescent Health Concerns Inventory as well.

| Data analysis
Descriptive findings were reported using proportions and means with standard deviations. To identify whether high-risk behaviours (as dependent variables) are more frequent in Medical Sciences' students who experienced health concerns, we first conducted the bivariate analysis using the chi-square test and independent t test.
Then, a multivariate logistic regression model was administrated with the variables that were related to the high-risk behaviours.
Statistical significance was based on a p < .05. Analyses were performed using Statistical Package for Social Sciences for Windows version 16.0 (SPSS Inc.).

| RE SULTS
Of the 970 completed questionnaires, 44 were eliminated due to incomplete and inadequate replies to the questions (response rate = 95%). Among the respondents, 56.2% were female, and the mean age of the participants was 19.19 ± 1.97 years. Most of the participants (52.1%) were studying for Bachelor's degree in Science.
Most of them were single (92.1%), and 45.9% lived in dormitories.
The socio-demographic characteristics of the participants are shown in Table 1.
The results indicated that physical inactivity was the most common high-risk behaviour that the students engaged in (97.5%, CI 95%: 96.5%, 98.5%), followed by unintentional injuries (65.4%, CI 95%:62.45%, 68.4%) and intentional injuries (46.4%, CI 95%: 43.2%, 49.6%). Intentional and unintentional injuries, smoking and tobacco use, alcohol and drug use and high-risk sexual behaviours more frequently occurred among males (p < .05), whereas physical inactivity was reported more among females (Figure 1). According to Table 2, the mean and standard deviation of the total score for the students' health concerns was 35.64 ± 24.66 (CI 95%: 34.05, 37.23). The mean and standard deviation of the subscales of Adolescent Health Concerns Inventory showed that the maximum and minimum mean score related to concern about the future (4.39 ± 4.04; CI 95%: 4.04, 4.39) and health services(1.60 ± 1.77; CI 95%: 1.50, 1.72), respectively.
In bivariate analysis as showed in Table 3, by considering potential covariates such as participants' age and gender and their parents' education and occupation status, university students' concerns increased the odds of risky behaviours (crude odds ratio ranged between 0.89-1.25, p < .05). As the goal in any data analysis is to extract from raw information the accurate estimation (Alexopoulos, 2010    showed that worry about being successful, choosing a job, getting a good grade, the future in the years to come was the most concerns of the young people (Baheiraei et al., 2016;Firouzkouhi et al., 2006).

| D ISCUSS I ON
Our finding about the relationship between concern about human sexuality (e.g. concerns about having sexual intercourse,  (Bi, Ma, Yuan, & Zhang, 2016). In contrast, it is showed that people who were more concerned with body image and appearance had less confidence in establishing relationships, which in turn resulted in the occurrence of fewer high-risk sexual behaviours (Littleton, Axsom, & Pury, 2005).
The more intentional injury in the university students who have concerns about emotional health may indicate such students seek relief from the consequent stress and tension by resorting to risky behaviours (Palti, Halevy et al., 1995;Tavolacci et al., 2013). Also, association between university students' concerns about injury prevention and control (such as being murdered and driving accidents) with substance use indicated that both improving safe environment and reducing the negative psychological effects of unsafe conditions are strongly needed (Smalley et al., 2017). Also, the predisposing role of students' concerns about the future (such as concern in relation to academic and professional achievements) to unhealthy dietary behaviours may suggested that intense feeling of insecurity about their financial and occupational rehabilitation, manifested by symptoms such as eating disorders (Eisenberg et al., 2007;Fradelos et al., 2019).
Due to limited studies on the relationship between health concerns and high-risk behaviours in medical students, it was not possible to compare some findings of the present study with the other studies.
The more high-risk sexual behaviours in university students with the more concerns about diseases and disorders (such as concern about chronic diseases and sexually transmitted diseases) are an example.
Although it seems diseases can affect people not only physically, but also mentally, and finally, can alter the affected person's perspective on life and involving in risky behaviours.
In the present study, Medical Sciences' students who had more concerns about nutrition were more likely to exhibit unhealthy dietary behaviours. A study on Spanish students showed the importance of body weight and fitness and being female were some of the significant predictors of unhealthy methods employed to lose weight (Blow, Taylor, Cooper, & Redfearn, 2010). Also, it is indicated that with increasing concerns about self-esteem, eating disorders and unhealthy nutritional behaviours generally will be increased (Sassaroli & Ruggiero, 2005
Employed As there is no significant relationship between the some variables in above Ref. -
Employed of smoking and tobacco use (Harakeh, Engels, Monshouwer, & Hanssen, 2010;Palti et al., 1995). However, on the contrary, it is reported that students who were concerned about nutrition and calorie intake may exhibited fewer high-risk dietary behaviours (Sun, 2008). The results of the present study may indicate that concerns about certain health issues can have an important role in preventing some high-risk behaviours. For example, concern about disease and disorders was associated with reduced high-risk sexual behaviours, and the students who were concerned about the future were less likely to have unhealthy dietary behaviours.
These findings may be evidence that high-risk behaviour may decrease as knowledge and worries increase (Arroyo et al., 2004;Yahia et al., 2008).

| S TRENG TH S AND LIMITATI ON S
As entering to the Iranian university is followed by a national en- all of the volunteer students who were present on the day of the research team's visit were invited to complete the self-administered instruments. This not-randomly participation of the students may have increased the likelihood of selection bias, but this was inevitable as if we asked students to answer these culturally sensitive questionnaires according to the randomized selection in each class, it might increase the information bias, because they might be wondering why I was chosen to answer? However, based on the randomly selection of the classrooms in each strata, the students have an equal chance of being included and the selection bias can be limited. high-risk behaviours and health concerns with a qualitative approach and through in-depth interviews with university students in the future. In this way, the roles of new emerging social determinants of health such as social media are recommended to be in the spotlight.

ACK N OWLED G EM ENTS
Authors would like to thank Mazandaran University of Medical Sciences for funding this project and all participants