GASTRIC CANCER PREVALENCE AND LIFESTYLE OBSERVATIONS AMONG ADULTS WITH DYSPEPSIA: A RETROSPECTIVE REVIEW AT A NATIONAL REFERRAL HOSPITAL IN KENYA

Abstract

Background: Gastric cancer (GC) remains a significant global health burden, with dietary and lifestyle factors playing major roles in its pathogenesis. There is limited data from East Africa on the lifestyle factors of patients with dyspepsia.
Study Objectives: To determine the prevalence of gastric cancer and describe lifestyle factors observed among adult dyspeptic patients undergoing upper gastrointestinal endoscopy at a national referral hospital.
Methodology: This was a retrospective descriptive study analyzing 221 medical records of adults who underwent upper gastrointestinal endoscopy and biopsy for dyspepsia between June and December 2024. Systematic sampling was applied, and data were extracted using a structured proforma and the Rome IV checklist. Variables of interest included histologically confirmed GC, lifestyle factors (diet, smoking, alcohol), and clinical characteristics. Statistical analysis (SPSS v26) used Pearson’s Chi-square/Fisher’s exact tests for associations and logistic regression for exploratory predictors.
Results: Of the 221 evaluated records, 12 (5.4 %) showed histological evidence of GC. Observed associations included frequent consumption of fried foods (p = 0.001), salty foods (p < 0.001), spicy foods (p = 0.003), and processed meats (p < 0.001). Other factors noted were dyspepsia duration > 6 months (p = 0.016), prior endoscopy (p = 0.011), hospital admission (p = 0.007), non-prescribed medication use (p = 0.024), and herbal remedy use (p = 0.041). Helicobacter pylori infection prevalence was 58.8 %, but this was not significantly associated with GC. No independent predictors were identified on multivariable analysis, likely due to the limited sample size.
Conclusions: Findings manifest a noteworthy prevalence of gastric cancer
among adults with dyspepsia and possible links between dietary habits and GC occurrence. However, these represent observational associations rather than causal inferences. Broader, prospective studies are warranted to confirm these trends and guide targeted prevention strategies.