Cigarette, waterpipe and e-cigarette use among an international sample of medical students. Cross-sectional multicenter study in Germany and Hungary

Despite numerous efforts to stop the tobacco epidemic, tobacco use is still the single most preventable cause of morbidity and mortality worldwide [1]. This risk factor is responsible alone for the death of more than seven million people across the world each year [2]. Among regions of the World Health Organization, Europe has the highest prevalence of tobacco smoking in the adult population and some of the highest prevalence of tobacco use by adolescents [3]. Health professionals, especially physicians, have a unique role while fighting tobacco smoking. Besides providing smoking cessation plans to their patients, they are role models for society at large. Since physicians who smoke are less likely to advise their patients to quit [4, 5], the smoking behavior of future physicians (i.e. medical students) does influence not only their personal health but also the health of their future patients.

Previous studies have detected differences in smoking prevalence among medical students with regard to gender and country of origin [6‐8], however they did not consider socio-economic status and religious involvement which are known to influence smoking behavior in young adults [9‐11].

New and emerging tobacco products, waterpipe and e-cigarette use are gaining popularity among the youth and young adults worldwide including dual waterpipe and cigarette use [12, 13]. Waterpipe smoking is believed by many users to be less detrimental to health than traditional cigarette consumption [14]. On the contrary, accumulating evidence shows that waterpipe smoking is harmful and addictive [15]. There is not yet enough evidence on the safety of e-cigarettes. Although electronic nicotine delivery systems release generally fewer toxic chemicals than traditional cigarettes, and are less harmful than cigarettes, their use among youth and young adults is far from being safe. E-cigarettes may trigger nicotine dependence [16]. Though cigarette smoking has been studied repeatedly among medical students in Europe [6‐8, 17‐20], data is limited on waterpipe and e-cigarette use in this specific population [21, 22].

A total of 2935 medical students from 65 different nations (response rate: 56.2%) at four universities participated in this multicenter study. Gender was reported in 2925 cases. Socio-demographic characteristics of the participants are shown in Table 1. German, Hungarian and Norwegian students were the most represented nationalities in the sample; the remaining students formed the multinational group. Among 324 German and 147 Norwegian students studying in Hungary only two (0.4%) specified Hungary as country of origin. Most students in the multinational group (89.6%) reported that the country in which they had spent the most of their life was one other than the country of their current medical school. These findings show that most of the students of other nationalities were temporary residents in Germany and Hungary. The composition of the multinational group was heterogeneous including students from European (54.6%), Asian (32.2%), American (7.1%) and African (6.1%) countries. Data on cigarette smoking was available in 2883 cases, waterpipe use in 2725, and e-cigarette use in 2688 cases. Proportion of cases with missing data on cigarette, waterpipe and e-cigarette use was less than 10%.

Marked variation in smoking prevalence was observed among medical students by gender and nationality. Our findings of male predominance in smoking rates corresponds with previous studies in general populations [28, 29], as well as with specific studies of medical students [6, 8, 17].

Overall, the prevalence of smoking among medical students in our cross-sectional study was lower than that of the comparable young adult populations in Germany, Hungary and Norway. This is a favorable trend as compared to a previous European survey that found the smoking prevalence to be higher among medical students than the general population [8]. In our study, prevalence of cigarette smoking was 20% among German male and 14% among German female students, whereas smoking prevalence was 31% among German males and 21% among German females aged 18–24 years as reported by the 2015 Epidemiological Survey of Substance Abuse [30]. According to the European Health Interview Survey 2014, smoking rates in Hungary were 42% in males and 29% in females in the 18–34 age group [29], which were higher than prevalence of smoking among Hungarian medical students in our study (24% in males, 16% in females). Norway had relatively low smoking rates in 2014 in the young adult population (20% in males and 14% in females aged 16–24 years [31]). Accordingly, we found very low smoking rates among Norwegian medical students (6.5% in females and 5.8% in males). Moreover, none of the Norwegian students were daily smokers, while about 7% of German and Hungarian students smoked cigarettes every day.

Although it cannot be answered directly from our results, knowledge about health risks of smoking as well as personal beliefs about roles as future physicians may have contributed to the lower smoking prevalence among medical students as compared to the general population. Based on European studies, the majority of medical students believed that health professionals are role models for patients [8], and almost 70% believed to know the health risks of smoking [19]. It is also possible that other factors such as parental smoking and peer influences may explain the difference in smoking rates between medical students and the general population.

Prevalence of cigarette smoking among medical students reported by former studies at German universities (males 22–42%, females 13–22% [6, 8, 17]) was similar to or higher than that of German medical students in our study (males 20%, females 14%). Previous surveys among Hungarian medical students showed wide variation in the prevalence of smoking (from 18 to 40% [18, 20]); our study mirrors those with the lower smoking rate. Surveys of Norwegian medical students found that smoking prevalence was 20–29% among males and 8–19% among females [7, 32], whereas in our study only 5.8% of Norwegian male and 6.5% of Norwegian female students were current smokers. Summarizing the existing data on smoking rates among medical students, the most obvious difference between previously published results and our findings was observed in Norwegian students which showed a remarkable decrease in the prevalence of cigarette smoking.

Taken together, similar differences in smoking rates could be found between nationalities both in the general population and among medical students. Norway had the lowest smoking rates in both settings and an ongoing decrease in the prevalence of cigarette smoking among Norwegian young people has been reported recently [33]. Lower prevalence of cigarette smoking in Norway may be explained by Norway’s long standing ban on all tobacco advertising, other preventive and control measures against tobacco use, and partly by a recent shift to snuff use [34].

In the last decades several tobacco control measures were introduced in Germany and Hungary as well. In Germany, numerous measures (increases in cigarette taxes, raising the minimum age for purchasing tobacco products, ban on advertising, laws on the protection of non-smokers and warnings on cigarette packages) have been implemented to reduce tobacco consumption. Nevertheless, there is still room for improvement in many areas of tobacco control policy [35]. According to the results of the 2013 Tobacco Control Scale, in which 34 European countries were compared with regard to their efforts to achieve effective tobacco prevention and control, Germany took the 33th place [36]. In recent years Hungary has also adopted or strengthened a series of tobacco control measures. The most important of these are the indoor smoking ban in public places and some outdoor prohibitions, the significant tax increase on cigarettes, the combined warnings (text and pictures) on cigarette packages, and the drastic reduction in the number of stores selling tobacco products [37]. Thanks to these efforts, Hungary has been ranked above average (11th place) on the European tobacco control scale in 2013 [36]. In that year, Norway ranked fourth on the same scale [36]. Although a causal relationship between a country’s tobacco strategy and smoking rates among its medical students cannot be established, the promotion of a comprehensive smoke-free legislation with a consequent social rejection of smoking may contribute to a decrease in smoking rates among medical students.

In our sample about 5% of the respondents were current waterpipe smokers. An even higher prevalence of current waterpipe smoking (11%) was reported in a previous study involving medical students in London [21]. The lack of awareness about the harmful health effects of waterpipe smoking, the social nature of its use and the presence of waterpipe cafés near educational establishments may contribute to the growing use of waterpipe among medical students [12, 21]. In our study, the majority of waterpipe smokers used tobacco in this way only 1–3 times a month, which is lower than the frequency shown to induce nicotine dependence. Median waterpipe smoking frequency was 6 days per month when the first symptoms of nicotine dependence emerged [38]. Nevertheless, concurrent use of waterpipe and cigarettes, which was reported by about 50 students (1.9%) in our study, increases the risk for nicotine exposure and other smoking related health effects.

Prevalence of e-cigarette smoking was relatively low (0.9%) in our sample, even lower than that (3.5%) found in a recent study at the Medical University of Silesia, Poland [22]. However, data concerning prevalence of e-cigarette use should be interpreted with caution because of the low number of e-cigarette users. Further studies are required to monitor e-cigarette use among medical students.

In the entire sample, prevalence of cigarette smoking was lower in students who reported high religiosity. Religious involvement was preventive in other populations as well [10, 11]. There is evidence that religiosity is associated with a lower prevalence of smoking initiation [11]. Furthermore, it has been demonstrated that very religious smokers were more likely to be interested in quitting within the next 6 months than their less religious counterparts [27]. However, Bailey et al. found that religiosity was not associated with successful smoking cessation in middle-aged adults [11].

Tobacco use was not related to students’ financial status, perhaps because most respondents reported a stable financial situation. Of course, our results may not be generalizable to other populations with a wider range of socio-economic backgrounds.

Although the whole sample had a narrow age range, a statistically significant difference in age between smokers and non-smokers could be detected. Cigarette smokers had an older mean age than non-smokers while prevalence of smoking was not higher in students further along in their studies. This apparent contradiction may be explained by the finding that the age difference between smokers and non-smokers was only significant in the 1st academic year, which had the largest sample size among study years, leading to a significant difference in mean age between smokers and non-smokers of the whole sample. Thus, it seems that being older than peers in the same class cohort, rather than age, was associated with cigarette smoking. Presumably, students who were older than their peers may have started university later or repeated a grade. However, the exact reason for the age difference was not investigated in the present study.

Cigarette smokers were older than non-smokers, whereas waterpipe users were younger than non-users. Similar findings concerning waterpipe smoking and age have been shown previously in the general European population as well: the highest prevalence of waterpipe tobacco use was detected among the youngest respondents [39]. Further studies are needed to determine whether this trend will herald an increase in waterpipe use and a decrease in cigarette smoking among next generations of medical students.

Cigarette smoking was associated with worse self-reported health status in our study. Cigarette smokers were less likely to rate their health status as very good or excellent, which is in line with previous observation of lower health-related quality of life scores in smoking university students compared to non-smokers [40]. These findings are alarming given the young age of the respondents. An even stronger negative association between self-reported health status and e-cigarette use was found. Slightly more than one-third of the e-cigarette users considered their health status to be very good or excellent. Further research is needed to confirm and clarify this finding. On the other hand, waterpipe use did not affect self-reported health status, probably because the majority of waterpipe smokers used this form of tobacco only 1–3 times a month.

Young smokers may recognize their addiction to tobacco when they experience withdrawal symptoms during their first attempt to quit [41]. In our study 288 young cigarette smokers (56% of current smokers) have already tried to quit smoking, but failed. Smokers who had one or more recent failed cessation attempts are less likely to quit successfully [42]. Furthermore, young adults tend to underutilize evidence-based cessation treatments, although all interventions effective for the general adult population are also effective for young adults [43].

University years provide a unique opportunity for interventions to change harmful health behaviors of future physicians, which is of particular interest not only for improving their personal health, but also because of their impending role in health promotion. Therefore, medical schools need to regularly collect data on the patterns of tobacco use including alternative tobacco products as well. Medical students who smoke should be encouraged to seek cessation treatment and medical faculties need to offer cessation programs to help their students to quit. Information about the health effects of alternative tobacco products and cessation support for their users should also be provided. Currently none of the medical faculties participating in our study provide smoking cessation programs for students. Strengthening of tobacco-free university policies, as well as tobacco control laws within the countries where the medical schools are located and/or the countries from which medical students are from, may also contribute to a decrease in tobacco use among future physicians.

Although the harm from tobacco use is undisputed, many medical students continue to smoke. Nevertheless, some positive trends were observed in smoking rates among future physicians in the recent years. We found a significant difference in smoking rates across different nations. There were fewer cigarette smokers among Norwegian students when compared to other countries. Prevalence of e-cigarette use was low, whereas waterpipe tobacco smoking was popular in our sample. Therefore, tobacco prevention and cessation programs for medical students should address not only the health risks of cigarette smoking but also discourage waterpipe use as well. Finally, medical schools should extend smoke-free areas from university buildings to the whole campus, which could have the dual advantage of creating a healthy environment and reducing the perceived social acceptance of smoking.