Objective: To estimate the effectiveness of the primary screening strategy for liver cancer in rural areas to provide basic information for the optimization and perfection of the technical program for the early detection and treatment of liver cancer. Methods: Residents including males aged 35-64 and females aged 45-64 from 9 counties in rural China between 2013 and 2015 were selected as the target population. The participant was classified into a high-risk and non-high-risk group based on the standardized questionnaire or HBsAg, and the Chi-squared test was applied to compare differences between the two groups. The Cox proportional hazard regression models were applied to assess hazard ratio (HR) and its 95% confidence interval (CI). Results: 358 348 participants were recruited from 2013 to 2015. 1 196 individuals were identified with liver cancer until December 31, 2021, with an incidence density of 52.0/10(5) person-years. Of the participants, 54 650 were assessed as high risk (15.3%) based on the questionnaire and the status of HBsAg. The high-risk population had a higher incidence density (168.3/10(5) person-years vs 31.5/10(5) person-years) and higher risk of developing liver cancer (HR=2.98, 95% CI=2.64-3.35), compared to the non-high-risk group. Based on the questionnaire-based high-risk assessment system, 47 884 (13.4%) individuals were identified as high risk, who showed statistical differences in terms of incidence density and incidence risk, in comparison to the low-risk population (all P<0.05). HBsAg can screen out a higher proportion of high-risk individuals who are women, non-smokers, non-drinkers, and individuals without a family history of liver cancer (all P<0.05). The sensitivity analysis of the effectiveness of the whole primary screening method is stable, and high-risk individuals still had a higher risk of liver cancer. Conclusions: The primary screening method of the questionnaire-based risk assessment system and HBsAg can achieve satisfactory effectiveness. The questionnaire-based risk assessment system could identify high-risk individuals to some extent, however, it still needs to be improved to meet the actual requirements.