Abstract
The interaction between smoking and genetic predispositions in the pathogenesis of ulcerative colitis (UC) remains to be determined. Also the effect of smoking on UC in Asian populations is not well-studied.
Our aim was to evaluate the association of smoking with UC and to assess the interaction between smoking and family history in UC development.
Using the National Health Insurance and National Health Screening Database which covers the entire Korean population(N=50 million) and includes data on family relationships, smoking status and other lifestyle factors, we identified 7 million individuals, comprising 3.7 million families. To examine the association of UC with current and former smoking, Cox proportional hazard regression was used and hazard ratios (HRs) were computed with a 95% confidence interval (CI). We calculated the familial risk of UC using incidence risk ratios (IRRs), and adjusted IRRs for lifestyle risk factors, including smoking and body mass index. Interactions between family history and current and former smoking were assessed on additive and multiplicative scales.
A total of 7 million people were initially included in our study and among them, we identified 1.2 million current smokers, 0.7 million former smokers and 5.1 million non-smokers. Our findings show that current smoking was associated with a decreased risk of disease (HR 0.6), while former smoking was associated with an increased risk (HR 1.2), compared to non-smokers. In the separate analysis among familial and non-familial groups, the protective effect of current smoking was less pronounced in the familial group (HR 0.5 vs 0.8) and the risk associated with former smoking also diminished in the familial group (HR 1.5 vs 1.2). In interaction analysis, the joint effect of current smoking and a positive family history of UC was lower than the product of their individual risks (p<0.05), suggesting a negative interaction. Also, the joint effect of former current smoking and family history was lower than the product of their individual risks with statistically significant interaction.
This large scale cohort study found that current smoking was associated with a decreased risk of UC and former smoking was associated with an increased risk. The impact of current and former smoking was different between family history positive and negative individuals. These results provide evidence for gene-smoking interaction in UC disease risk.
