Accuracy of Self-Reported Cancer-Screening Histories: A Meta-analysis

• Published in: Cancer epidemiology, biomarkers & prevention Vol. 17; no. 4; pp. 748 - 757
• Main Authors: Rauscher, Garth H, Johnson, Timothy P, Cho, Young Ik, Walk, Jennifer A
• Format: Journal Article
• Language: English
• Published: United States American Association for Cancer Research 01-04-2008
• Subjects: Breast Neoplasms - diagnosis; cancer; Colonoscopy - statistics & numerical data; Colorectal Neoplasms - diagnosis; Female; Humans; Male; Mammography - statistics & numerical data; Mass Screening - statistics & numerical data; meta-analysis; Occult Blood; Prostate-Specific Antigen - blood; Prostatic Neoplasms - diagnosis; Reproducibility of Results; screening; Self Disclosure; Sensitivity and Specificity; Social Class
• ISSN: 1055-9965; 1538-7755
• DOI: 10.1158/1055-9965.EPI-07-2629

Background: Survey data used to study trends in cancer screening may overestimate screening utilization while potentially
underestimating existing disparities in use.
Methods: We did a literature review and meta-analysis of validation studies examining the accuracy of self-reported cancer-screening
histories. We calculated summary random-effects estimates for sensitivity and specificity, separately for mammography, clinical
breast exam (CBE), Pap smear, prostate-specific antigen testing (PSA), digital rectal exam, fecal occult blood testing, and
colorectal endoscopy.
Results: Sensitivity was highest for mammogram, CBE, and Pap smear (0.95, 0.94, and 0.93, respectively) and lowest for PSA
and digital rectal exam histories (0.71 and 0.75). Specificity was highest for endoscopy, fecal occult blood testing, and
PSA (0.90, 0.78, and 0.73, respectively) and lowest for CBE, Pap smear, and mammogram histories (0.26, 0.48, and 0.61, respectively).
Sensitivity and specificity summary estimates tended to be lower in predominantly Black and Hispanic samples compared with
predominantly White samples. When estimates of self-report accuracy from this meta-analysis were applied to cancer-screening
prevalence estimates from the National Health Interview Survey, results suggested that prevalence estimates are artificially
increased and disparities in prevalence are artificially decreased by inaccurate self-reports.
Conclusions: National survey data are overestimating cancer-screening utilization for several common procedures and may be
masking disparities in screening due to racial/ethnic differences in reporting accuracy. (Cancer Epidemiol Biomarkers Prev
2008;17(4):748–57)