Role of Smoking in Functional Dyspepsia and Irritable Bowel Syndrome

Three Random Population-based Studies

Nicholas J. Talley; Nicholas Powell; Marjorie M. Walker; Mike P. Jones; Jukka Ronkainen; Anna Forsberg; Lars Kjellström; Per M. Hellström; Pertti Aro; Bengt Wallner; Lars Agréus; Anna Andreasson


Aliment Pharmacol Ther. 2021;54(1):32-42. 

In This Article

Abstract and Introduction


Background: It is uncertain if functional dyspepsia (FD) or irritable bowel syndrome (IBS) are linked to smoking, and smoking cessation is not part of the routine advice provided to these patients.

Aim: To assess if smoking is an independent risk factor for FD and IBS.

Methods: Three population-based endoscopy studies in Sweden with 2560 community individuals in total (mean age 51.5 years, 46% male). IBS (14.9%), FD (33.5%), and associated symptoms were assessed using the validated abdominal symptom questionnaire, and smoking (17.9%) was obtained from standardised questions during a clinic visit. The effect of smoking on symptom status was analysed in an individual person data meta-analysis using mixed effect logistic regression, adjusted for snuffing, age and sex.

Results: Individuals smoking cigarettes reported significantly higher odds of postprandial distress syndrome (FD-PDS) (OR 10–19 cig/day = 1.42, 95% CI 1.04–1.98 P = 0.027, OR ≥20 cig/day = 2.16, 95% CI 1.38–3.38, P = 0.001) but not epigastric pain. Individuals smoking 20 or more cigarettes per day reported significantly higher odds of IBS-diarrhoea (OR = 2.40, 95% CI 1.12–5.16, P = 0.025), diarrhoea (OR = 2.01, 95%CI 1.28–3.16, P = 0.003), urgency (OR = 2.21, 95%CI 1.41–3.47, P = 0.001) and flatus (OR = 1.77, 95%CI 1.14–2.76, P = 0.012) than non-smokers. Smoking was not associated with IBS-constipation or IBS-mixed.

Conclusion: Smoking is an important environmental risk factor for postprandial distress syndrome, the most common FD subgroup, with over a twofold increased odds of PDS in heavy smokers. The role of smoking in IBS-diarrhoea, but not constipation, is also likely important.


Globally, functional gastrointestinal disorders (FGIDs) are highly prevalent, and two of the most well recognised are the irritable bowel syndrome (IBS) and functional dyspepsia (FD).[1] These disorders are important because they are common, and can substantially impact on quality of life, work and relationships.[2,3] IBS can present with predominant diarrhoea (IBS-D), constipation (IBS-C) or a mixed pattern (IBS-M), while FD may present with early satiety or postprandial fullness (postprandial distress syndrome, PDS) or epigastric pain syndrome (EPS), or a mixed pattern.[1–3]

The aetiopathogenesis of both IBS and FD is uncertain but environmental and genetic factors are likely relevant. Infections of the gastrointestinal tract represent one important environmental factor, and in post-gastroenteritis there is an increased risk of IBS, FD or IBS and FD overlap.[4] Other environmental factors have been less well studied in the FGIDs but are potentially important to evaluate as modification of risk may impact disease burden.[5]

Older population-based and outpatient studies have suggested smoking is not a risk factor for FD but the major subgroup of PDS was not evaluated.[6,7] More recent work suggests smokers may have an increased risk of post-infectious FD which is unexplained.[8] A population-based study from Italy reported that smoking is a risk factor for FD,[9] but it remains unclear whether this risk is of importance, and whether it applies to all of FD or a subgroup. Studies of current smoking status in IBS have been conflicting and it remains uncertain if this environmental exposure is relevant or not.[10] No studies have evaluated whether snuff use, common in Scandinavia, is also associated with FGIDs.

Smoking cessation is not currently part of the management recommendations for FD or IBS. Our aim was to investigate the association between smoking and IBS or FD (including subtypes) in three similar large general population-based samples that had undergone endoscopy and pool the data to ensure there was sufficient study power to detect even weak associations. We also evaluated snuff use as a risk factor.