Population attributable fraction of esophageal squamous cell carcinoma due to smoking and alcohol in Uganda

Abstract

Background: Despite the high rates and regional variation of esophageal squamous cell carcinoma (ESCC) in East Africa, the contributions of smoking and alcohol to the ESCC burden in the general population are unknown. Methods: We conducted a case-control study of patients presenting for upper gastrointestinal endoscopic examination at Mbarara Regional Referral Hospital, Uganda. Sociodemographic data including smoking and alcohol intake were collected prior to endoscopy. Cases were those with histological diagnosis of ESCC and controls were participants with normal endoscopic examination and gastritis/duodentitis or normal histology. We used odds ratios associated with ESCC risk to determine the population attributable fractions for smoking, alcohol use, and a combination of smoking and alcohol use among adults aged 30 years or greater who underwent upper gastrointestinal endoscopy. Results: Our study consisted of 67 cases and 142 controls. Median age was 51 years (IQR 40–64); and participants were predominantly male (59 %). Dysphagia and/or odynophagia as indications for endoscopy were significantly more in cases compared to controls (72 % vs 6 %, p < 0.0001). Male gender and increasing age were statistically associated with ESCC. In the unadjusted models, the population attributable fraction of ESCC due to male gender was 55 %, female gender - 49 %, smoking 20 %, alcohol 9 % and a combination of alcohol & smoking 15 %. After adjusting for gender and age, the population attributable fraction of ESCC due to smoking, alcohol intake and a combination of alcohol & smoking were 16, 10, and 13 % respectively. Conclusion: In this population, 13 % of esophageal squamous cell carcinoma cases would be avoided if smoking and alcohol use were discontinued. These results suggest that other important risk factors for ESCC in southwestern Uganda remain unknown