Jiashan county is a rural area in China with very high incidence and mortality rates of colorectal cancer. From 1980 to 1984, we conducted a mass screening for rectal neoplasm in the 11 people's communes in this county. Of the 72,879 individuals who were eligible for screening, 60,496 participated, representing an overall participation rate of 83.0%. Both the fecal occult blood (OB) test and rectoscopy were performed on 47,560 of the participants, and rectoscopy only was performed on the remaining 12,936 individuals. From the mass screening, 15 rectal cancers, 3 carcinoids, 899 polyps, 98 ulcers, and 7 inflammatory tumors were detected, a total of 1022 cases. Eight of the 15 detected cases of rectal cancer were in Dukes' stage A. We did not find any evidence for an association between schistosomiasis and colorectal cancer. Results from the OB test were found to have a negative correlation with results from rectoscopy (odds ratio = 0.8, 95% CI = 0.6, 1.1). The OB test had a high false positive rate of 32.7%, and a poor positive predictive value of 3.5%. Both its positive and negative predictive powers were below 1.0, in other words, its positive likelihood ratio was below 1.0 and negative likelihood ratio was above 1.0, indicating that the OB test is not informative and has a poor predictive accuracy for rectal neoplasm. In addition, there were a number of practical problems concerning the use of the OB test for mass screening in the rural community.(ABSTRACT TRUNCATED AT 250 WORDS)