Barry Kramer

NIH Office of Disease Prevention

PLCO (Learn more about this study)

2006-0010

Jul 1, 2006

Cumulative Incidence of False-Positive Screening Test and Resulting Risks of Adverse Events in the PLCO Trial

Multiple cancer screening tests have been advocated for the general population. However, providers and patients are not always well-informed of the burdens incurred by screening. We sought to determine the cumulative risk of a false positive (FP) screen, and resulting diagnostic procedure risk, for an individual in a multiple modality screening program. Methods Data were analyzed for a 3-year period from the intervention arm of the randomized PLCO Cancer Screening Trial. Eligible individuals were 55 to 74 years of age with no prior history of prostate, lung, colorectal, or ovarian cancer. Additional exclusion criteria were: 1) death before/refusal to take/missing all screens, and 2) less than 3-year follow-up. 68,415 participants were included. The control arm received "usual care." Women randomized to screening were offered annual CA-125, transvaginal ultrasound, CXR, and baseline plus 3- or 5-year sigmoidoscopy (FSG). Men randomized to screening were offered annual digital rectal exam, PSA, CXR, and baseline plus 3- or 5-year FSG. Fourteen exams were possible for each sex over the study period. We defined a FP as a positive screen with no target cancer diagnosis after at least 3-years follow-up. Advanced adenomas (villous histology, severe cellular dysplasia, or =1cm in diameter) were considered true positives. Results 42.6% (N=29,152) of participants had at least one FP. Using a geometric distribution model, the cumulative risk of a FP after 1 screen was 5.7% for men, and 2.2% for women. After 14 tests, the risk was 62.2% for men and 50.7% for women. Sensitivity analysis for FP risk after 14 tests by Kaplan-Meier method demonstrated a cumulative FP risk of 61.5% for men and 54.5% for women. The consequent cumulative risk of a diagnostic procedure was about 27% for men and 18% for women after 14 tests. Conclusions For an individual participating in a multiple modality cancer screening trial, the risk of a FP exceeded 50% by the 14th test. Physicians should educate patients about the risks of FP tests and resulting diagnostic interventions when counseling about cancer screening regimens.