Study design and recruitment of study participants

This is an observational, population-based study. The target population is Lebanese adults, eighteen-to-sixty five years old. The sample was selected from a convenience population of bank employees in Lebanon. The choice of bank employees was based on the rationale that they represent a significant percent of the private working force. Bank employees also provide a sufficient wide age range of a cohort of working adults in Lebanon. The selected banks were located in major cities across the country: Beirut, the capital, Tripoli and Byblos, in the North, Zahle and Chtaura in the Bekaa, and Saida, Tyre and Nabatieh in the South of Lebanon. All major banks in Beirut were contacted initially by e-mail to request the administration approval to conduct the survey among their employees in the main headquarters in the city of Beirut, as well as branches in other cities. The selected banks were representative of the Lebanese banking sector, as they constituted the major banks in the country. The majority (95%) of the contacted banks agreed to participate.

The data collection extended from January 2016 to April 2016. The employees were informed about the survey by the administration and asked for voluntary participation in the study. The questionnaires were distributed and collected by the study team during working hours (Additional file 1). The consenting participants were asked to return the filled questionnaire in a sealed envelope and deposited in an assigned box for confidentiality. The response rate in different banks varied from 70% to 80%. Bank employees were excluded if they had a history of Crohn’s disease, or treated for peptic ulcer diseases. A screening question was used to exclude the non-eligible participants. The research team approached the consenting bank employees and inquired: Have you ever been diagnosed with Crohn’s disease? Are you currently taking medication for peptic ulcer disease?. Any respondent who responded yes to any of these two questions was not eligible to participate in the study. All eligible employees in the selected banks completed a fifteen minutes anonymous self-administered questionnaire. The questionnaire included a consent form on its cover page, all participants completing the questionnaire were considered to be indirectly providing their informed consent. The Lebanese American University Institutional Review Board (IRB) committee reviewed and approved the bank request formal letter as well as the study questionnaire.

Data to ascertain prevalence and diagnosis of IBS were based on the Rome III criteria for Functional Gastrointestinal Disorders. IBS is defined as recurrent abdominal pain or discomfort in the last three months for at least 3 days per month, associated with at least two of the following: relief after defecation; changes in bowel movement frequency, and occurrence of symptoms associated with changes in stool form [20]. The questionnaire collected data on socio-economic, demographic and behavioral characteristics (sex, age, education and regional distribution), smoking patterns (cigarettes, waterpipe, cigars and pipe), alcohol consumption physical activity and food intolerance.

Sample size

The sample size calculation based on the probability that the prevalence of IBS is 20% in the population and the error in the estimate of ±3.4% with a 95% confidence interval yielded a required sample size of 532 individuals.

Statistical analysis

The data was analyzed using the Statistical Package for Social Sciences (Version 23.0. IBM Corporation, Arnouk, USA). The association between IBS and socio-demographic characteristics, smoking, alcohol consumption, physical activity and food intolerance was assessed by Chi-square. Logistic regression was used to predict the independent association of demographic factors, smoking, alcohol consumption, and physical activity on the odds of having IBS. A p-value less than 0.05 indicated statistical significance.

Prevalence and bivariate association of IBS by risk factors

(Table 2 to be inserted here)

Cigarettes smoking

The prevalence of current cigarette smoking in the total sample was 31.3%, with 12.8% being past smokers, and the remaining 55.9% having never smoked. The prevalence of cigarette smoking was higher (P˂ 0.001) among males (56.3%) compared to females (33.2%). Smoking was more prevalent (P˂ 0.003) among those who had a secondary education or a technical school degree (57.9%) compared to those with a university degree or higher studies 42.4% and 39.4% respectively. No significant difference in IBS prevalence was observed between those who ever smoked and those who never smoked cigarettes, (Table 2).

Waterpipe smoking

The prevalence of current waterpipe smoking was 36.6%, 10.1% were past smokers, while 53.3% never smoked. No sex or educational level differential was noted among waterpipe smokers. Table 2 show that IBS occurred more frequently among ever waterpipe smokers (24.0%) than those who never smoked water pipe (16.6%) (P˂0.01).

Alcohol consumption

The prevalence of reported current alcohol consumption was 31.4% with 6.9% being past-drinkers and 61.7% never consumed alcohol. Males reported a greater (P˂0.01) alcohol consumption (47.6%) compared to females (34.8%). Alcohol consumption was more prevalent(P˂0.0001) among those who had a university or completed higher studies (58.2%) compared to those with secondary education or technical school (38.1%). Table 2 showed that IBS was more prevailing (P˂0.001) among alcohol drinkers (26.5%) compared to never alcohol drinkers (16.1%).

Physical activity

The prevalence of physical activity was 63.9%, where 24.4% reported exercising less than once per week, 25.5% exercised 2 to 3 times per week and 14.0% exercised more than 3 times per week. A higher proportion of males (P˂0.0001) were physically active (77.9%) compared to females (53.8%), while no difference in activity level was noted between those with different levels of education. There was no change in IBS occurrence by reported physical activity, (Table 2).

Food intolerance

The respondents were also asked to report if they experience any food intolerance. Those who reported experiencing food intolerance (n = 28) had a greater (P˂ 0.05) occurrence of IBS (35.7%) compared to those who did not report food intolerance (19.3%).

Logistic regression analysis

(Table 3 to be inserted here)

Key findings

This study assessed the prevalence of IBS among a sample of Lebanese employees as well as major lifestyle associations. The prevalence of 20.1% among the sampled adult participants was comparable to two studies of university students, one in Lebanon [17] and a recent one in neighboring Syria [21]as well as estimates reported in developed countries [6]. However the occurrence of IBS was much higher than what has been reported in the West [2]. The reported sex and age differential among those with IBS was consistent with results reported in the literature [7, 8]. Various hypotheses have been proposed to explain the higher prevalence of IBS in females, i.e. higher serotonin synthesis in the brain [22],female sex hormones’ effect on gastrointestinal motility [23], and probable association of IBS with an anti-nociceptive mechanism diminishing pain related to pelvic events such as pregnancy and delivery [24].The association of IBS with educational level is inconsistent in the reported literature, where some studies show a higher prevalence of IBS among educated compared to less educated [25–27].Gwee et al. [28] reported that IBS is significantly more prevalent among those with more than 6 years of post-secondary education. Yet, lower education was associated with a higher IBS prevalence in one study [29]. There was minimal variability in the educational level of the participants in our study; therefore such a statistical association could not be analyzed adequately.

Behavioral (lifestyle)risk factors

Individuals who consumed alcohol in our study were twice as likely to suffer from IBS compared to those who did not. The association between alcohol consumption and IBS has been inconsistent in the literature, some studies report no effect [14, 15], while others show that alcohol consumption was associated with a higher IBS prevalence [16]. Certainly, more studies are required to reach a better understanding of the association between alcohol and IBS.

Our study is the first to investigate the relationship between IBS and waterpipe smoking. Waterpipe smokers were found to have significantly more IBS compared to non-smokers. Waterpipe smoking (Hookah or Shisha) is becoming popular in coffee shops across the Middle East and North Africa and its prevalence in Lebanese youth (13-15 years) has reached 64.5% of males and 54.6% of females; whereas its prevalence in Lebanese adults (18+ years) is about 25.8% for males and 23.3% for females. Unfortunately, Lebanese women have the highest female reported water pipe smoking rate in the region [30]. This social habit seems to have spread to Western countries namely in Europe [31, 32].

Cigarette smoking was significantly more prevalent in males (56.3%) compared to females (33.2%), yet no sex difference was noted in waterpipe smoking. One probable explanation for this difference in sex prevalence seen in cigarette but not waterpipe smoking can be related to the wide social acceptance of waterpipe in the Lebanese culture and the limited knowledge about its risks.

Physical activity has been shown to be an effective measure in relieving gas-related symptoms, and is currently recommended for people who suffer from abdominal bloating as it improves impaired gas clearance related to altered small bowel activity [33].Although Costanian et al. [17], Kim et al. [18] and others [34]have consistently shown higher IBS prevalence in those with low physical activity and that activity improved IBS symptoms, our study showed that physical activity was not significantly associated with IBS occurrence in this adult sample of the Lebanese population.

The role of food intolerance in provoking or exacerbating IBS symptoms have been well described and constitute the basis of the low FODMAP diet [35] . The significant relationship observed in our study of IBS in patient reporting food intolerance, is well established and in agreement with what has been reported in the literature. However, the respondents were not asked what type of food induces the food intolerance, which is a study limitation.

Study limitation

There are several limitations that should be noted in this study. The data was collected through self-administered questionnaires that may lead to a higher percentage of incomplete data than interview surveys. Selection bias is also possible as we limited the study population to bank employees that might already have a better socio-economic status and educational level compared to the general Lebanese population. This was evident in the finding that 65% of our sample had a university degree compared to 20% in the general population in Lebanon. Moreover, other factors known to be associated with IBS were not addressed, such as depression and anxiety. In addition, the amount of tobacco use, alcohol consumption, and water pipe use was not quantified.