Prevalence and correlates of low back pain among undergraduate medical students in Serbia, a cross-sectional study

Study setting

This study was carried out at the Faculty of Medical Sciences, University of Kragujevac, a state university in central Serbia. The study programme at the integrated academic studies for doctor of medicine is organized in semesters; there are 12 semesters in total during the 6 years of studies. The teaching process is realized through lectures, laboratory sessions, other forms of lecturing and six-year long professional practice. The teaching process is held by the use of interactive programmes oriented toward the students, providing conditions for a greater engagement of the students during lectures, with constant checking of their knowledge. The curriculum provides the means of knowledge evaluation according to the law on higher education: at least 30% of a grade is accomplished through the pre-final examination activities. The contents of the courses of the study programme are based on the latest knowledge in adequate disciplines, and oriented towards patients as the center of interest of theoretical and practical medicine. The methods of teaching are provided problem-oriented lectures and small-group work. A relatively small number of medical students (a total of 536 freshmen, from 2013 to 2018) are enrolled in the programme at the integrated academic studies of medicine, with the purpose of succeeding in fulfilling the set standards of quality. All teaching modalities (lectures, seminars, exercises) are mandatory, which encompasses regular students’ names call-outs to determine attendance to all academic activities. The research was conducted during November–December, 2018. The questionnaire was filled out by each participant during their classes at the medical faculty. All classes were visited by the researchers during the data collection period, in order to inform students about the goal of research and provide assistance during the filling out of the questionnaire, which included providing clarifications to any questions which participants might have had about the items in the questionnaire.

Study design

The research was carried out using a cross-sectional design using a self-administered questionnaire. Only those respondents who gave voluntary, informed written consent for participation in the study were included.

Study population

The research involved students of medicine at only one of the faculties of medicine in Serbia. All students who met the inclusion criteria were eligible to participate in the study.

All full-time medical students were considered for inclusion in the study. Eligible medical students were aged over 18 years, were fluent in spoken and written Serbian language. Also, the inclusion criteria included voluntary agreement to participate in the study. Reasons for ineligibility were age <18 at the time of recruitment, not speaking or writing in Serbian language, any other reason that did not allow or hindered participation in the study or refusal to participate in the study.

Responses to the questionnaire were received from 508 medical students (response rate was 92.9%). Responses which were not valid or fully complete (total of nine questionnaires) were not analyzed due to a lack of data. The reason for non-response was well documented in the study, and this was mostly lack of interest (notified by returning a form included in the letter with the initial study invitation). Finally, the participation rate was 91.2% (499/547). The nonparticipant group comprised 39 medical students who did not attend the classes at the time of the study, and about whom there is no data due to the anonymity of participants.

Data collection

A self-reported, anonymous questionnaire was distributed to each participant. All study participants provided written informed voluntary consent prior to taking part in the study. After accepting to participate in the study, medical students completed the questionnaire, which took approximately 20 min.

For students, LBP was defined as ‘pain in the area on the posterior aspect of the body from the lower margin of the twelfth ribs to the lower glutaeal folds, that lasts for at least one day, with or without pain referred into one or both lower limbs’ (Duthey, 2013; Hoy et al., 2012). The definition of LBP was given verbally to medical students. In our survey, the case of LBP was defined as a positive response to the appropriate question, which was as follows: Do you NOW have lower back pain? No/Yes.

Questionnaire collected basic information such as age, gender, the study year, total overall grade (from 6 to 10), level of study, completed secondary school (Statistical Office of the Republic of Serbia, 2020). Further on, the questionnaire contained questions on the exposure to factors potentially associated with LBP, such as the habit of cigarette smoking, history of motor vehicle accidents, studying conditions (injuries at faculty, prolonged standing at faculty, prolonged sitting at faculty, incorrect body posture, lifting/moving heavy loads, stress during classes, conflicts at faculty), circumstances outside the faculty such as recreational sports, incorrect sleeping position, housework (window washing, etc.), house chores (ironing, etc.), other domestic tasks, weather (e.g. cold, humidity, …), draft, LBP in family history, other factors. The body mass index (BMI) was calculated based on self-reported weight and height. Degree of obesity was estimated as follows: BMI less than 18.5kg/m2 denoted underweight, 18.5–24.9 meant normal or healthy weight, 25.0–29.9 denoted overweight, while subjects with BMI 30.0 and above were considered obese (World Health Organization, 2000). As ‘other domestic tasks,’ students most often reported vacuum cleaning and dusting, hanging curtains and dishwashing. As ‘other factors,’ students most often reported computer usage, studying, watching television. Data concerning personal medical history included information on depression, anxiety, some other psychiatric illness, rheumatic disease, arterial hypertension, ischemic heart disease, diabetes mellitus, thyroid disease, asthma, anemia, migraine, sinusitis, myopia. Personal medical history was estimated as positive in the presence of either one or multiple diseases (such as depression, anxiety, chronic rheumatic diseases, hypertension, diabetes mellitus, thyroid disease, surgical interventions, etc.). Comorbidity represented the presence of more than one disease in personal medical history.

Ethical considerations

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (the Ethics Committee of the Faculty of Medical Sciences, University of Kragujevac, Ref. No.: 01-13459) and with the 2013 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Data analyses

A descriptive analysis was performed for categorical variables by using absolute and relative frequencies. The association between factors potentially associated with LBP and LBP in medical students was investigated using univariate logistic regression. Multivariate logistic regression was used to determine the independent factors associated with LBP. Results from the logistic regression analyses were expressed as odds ratios (OR) with 95% confidence intervals (CI). Multivariate logistic regression models were made for all variables that were related to LBP in univariate analyses at a p value of <0.10. This analysis used the `enter` method (default with the menu system). Models fit were assessed by the Hosmer-Lemeshow test of goodness of fit and Cox and Snell’s and Nagelkerke’s Pseudo R square measures. The whole model was statistically significant (Omnibus goodness of fit test: Chi-square = 47.09, df = 8, p <0.001; Hosmer-Lemeshow goodness of fit test: Chi-square = 5.70, df = 8, p = 0.681). Cox and Snell’s and Nagelkerke’s Pseudo R square measures were 0.09 and 0.14, respectively. Among the observed variables, multicollinearity was examined with Variance Inflation Factors (VIF). Our results show that all independent variables had a VIF near 1, indicating that there is no correlation between this independent variable and any others. A test for linear trend in risk was based on the logistic regression model.

Statistical significance was accepted at the level of p < 0.05. All statistical analyses were conducted with SPSS 20.0 (SPSS, Chicago, IL, USA).

Participant characteristics and prevalence of LBP

Based on the previously described inclusion and exclusion criteria, 508 medical students were recruited into the study group. Of that number, nine questionnaires were excluded from the data analysis due to invalid or incomplete answers. Finally, a total of 499 eligible participants were included in the study. The participation rate was 91.2% (499/547). Four hundred ninety nine medical students were evaluated: of the participants, 161 (32.3%) were male and 338 (67.7%) were female, while the mean age was 22.0±2.2 (range 18–34) years (Table 1). The proportion of students at clinical level of study (from 7th to 12th semester) was higher than the proportion of students on preclinical level - those in the first 6 semesters (54.5% vs. 45.5%). More than half of medical students (64.3%) previously completed secondary medical school. The proportion of students with overall grade ≥8 is higher (69.0%) than the proportion of students with lower overall grade (15.8%), with note that freshmen have not been graded yet. Overall, 20.8% (95% CI [17.6–24.0]) of medical students reported suffering from LBP at the time of answering the questionnaire.

Factors associated with prevalence of LBP

Tables 2 and 3 demonstrate the results of the univariate analyses concerning the factors associated with prevalence of LBP. No significant differences were observed in the LBP prevalence in medical students by age, gender, completed secondary school, study year, cumulative overall average grade, level of study (Table 2).

Cigarette smoking (p = 0.001), prolonged sitting at faculty (p = 0.018), incorrect body posture (p = 0.008), stress during classes (p = 0.001), incorrect sleeping position (p = 0.008), and positive family history of LBP (p = 0.006) were the variables associated with prevalence of LBP at the time of the study (according to the univariate logistic regression analysis) (Table 3). Also, according to the univariate logistic regression analysis, obesity and draft were associated with prevalence of LBP at p < 0.1.

In the multivariate logistic regression model (Table 4), only cigarette smoking (OR = 2.5, 95% CI [1.5–4.2], p = 0.001), stress during classes (OR = 1.8, 95% CI [1.1–3.0], p = 0.039), incorrect sleeping position (OR = 1.9, 95% CI [1.2–3.1], p = 0.006) and family history of LBP (OR = 1.6, 95% CI [1.1–2.7], p = 0.047) were independently associated with high prevalence of LBP at the time of the study.

Strengths and limitations of the study

To the best of our knowledge, our study is one of the first studies on LBP in medical students in Serbia. Even though the response rate was satisfactory, this study has several possible limitations. Data in this study were based on anonymous self-reports, but information bias cannot be ruled out. Limitation of the present study includes sampling bias, because only students who attended lectures were involved. Further, there remains the possibility of participant over- or underreporting their LBP. Even though surveys using self-report data can be inaccurate, medical students were in a position to understand both the definitions we used in the study and the importance of giving accurate and true responses to health survey due to their acquired medical knowledge. Also, since the study was conducted at one faculty, the results may not be representative of all medical students in Serbia. Finally, one important limitation of our study is the cross-sectional design, with all known shortcomings of such studies in identifying factors predicting LBP.