Cluster randomization trial of sequence mass screening for colorectal cancer

Abstract

Purpose: Colorectal cancer is a major cause of death worldwide. To reduce the incidence and mortality from rectal cancer, an individual quantitative risk-assessment model (hereafter referred to as the Attributive Degree Value) and reverse passive hemagglutination fecal occult blood test were used in a randomized, controlled, population-based trial that was conducted in Jiashan County, People's Republic of China.

Methods: All residents of Jiashan County aged 30 years or older were enrolled in the study, and 21 townships in the county were randomized to either a screening (n = 10 townships) or control (n = 11 townships) group. Participants in the screened group submitted a one-article-per-slide stool sample and completed a structured risk-assessment questionnaire from which their attributive degree value was computed. According to study protocol, 4,299 participants were defined as high risk and underwent diagnostic evaluation with 60-cm flexible sigmoidoscopy and, in some cases, an additional screening with colonoscopy.

Results: From 1989 to 1996, cumulative mortality from colon cancer was 90 (95 percent confidence interval, 83-97) per 100,000 in the screened group and 83 (95 percent confidence interval, 76-90) per 100,000 in the control group (log-rank P = 1.49, P = 0.222). Mortality from rectal cancer during this time was 110 (95 percent confidence interval, 102-118) per 100,000 in the screened group, which differed significantly from the control group mortality rate of 161 (95 percent confidence interval, 152-170) per 100,000 (log-rank P = 0.003).

Conclusion: Mass screening with a reverse passive hemagglutination fecal occult blood test along with an individual attributive degree value score was effective in reducing mortality from rectal cancer but not in reducing mortality from colon cancer or the incidence of colorectal cancer.