Classifying risk level of gastric cancer: Evaluation of questionnaire-based prediction model

This study aimed at evaluating the efficacy of the questionnaire-based prediction model in an independent prospective cohort. A cluster-randomized controlled trial was conducted in Changsha, Harbin, Luoshan, and Sheyang in eastern China in 2015-2017. A total of 182 villages/communities were regarded as clusters, and allocated to screening arm or control arm randomly. Face-to-face interview through a questionnaire interview, including of relevant risk factors of gastric cancer, was administered for each subject. Participants were further classified into high-risk or low-risk groups based on their exposure to risk factors. All participants were followed up until December 31, 2019. Cumulative incidence rates from gastric cancer between high-risk and low-risk groups were calculated and compared using the log-rank test. Cox proportional hazard regression models were applied to estimate hazard ratio (HR) and 95% confidence interval (95% CI). Totally, 89,914 residents were recruited with a mean follow-up of 3.47 years. And 42,015 (46.73%) individuals were classified into high-risk group and 47,899 (53.27%) subjects were categorized into low-risk group. Gastric cancer was diagnosed in 131 participants, of which 91 were in high-risk group. Compared with the low-risk participants, high-risk individuals were more likely to develop gastric cancer (adjusted HR=2.15, 95% CI, 1.23-3.76). The sensitivity of the questionnaire-based model was estimated at 61.82% (95% CI, 47.71-74.28) in a general population. Our questionnaire-based model is effective at identifying high-risk individuals for gastric cancer.