Table 2.

Estimated self-reported screening rates per recommendations: estimated parent study self-reported rates and corrected rates using the PPV and NPV

Self-report rate*Corrected rate (to medical record)ΔReport-to-records ratio§
Baseline (%)
    FOBT (n = 1,697)21.813.97.81.57
    Flexible sigmoidoscopy (n = 1,695)33.227.55.71.21
    Colonoscopy (n = 1,690)27.716.710.91.66
    Barium enema (n = 1,691)14.410.53.91.37
    Overall adherence (n = 1,698)54.846.68.21.18
Year 1 follow-up (%)
    FOBT (n = 1,558)26.613.613.01.96
    Flexible sigmoidoscopy (n = 1,556)36.629.27.41.25
    Colonoscopy (n = 1,544)31.622.88.81.39
    Barium enema (n = 1,544)14.110.73.41.32
    Overall adherence (n = 1,553)61.349.711.71.23
  • * Estimated self-reported screening rates adjusted for parent study eligibility and nonresponse.

  • Screening rates “corrected” (adjusted) for PPV and NPV of self-reported screening relative to medical records; predictive values were unadjusted for validity study nonresponse due to small sample size.

  • Reduction in screening rate after “correction” (adjustment) = Δ = (estimated self-reported screening rate − medical record screening rate).

  • § Because our validation study sample was stratified by survey-reported screening adherence status, the report-to-records ratio was not derived directly from the 2 × 2 table of validation study counts [e.g., as in Warnecke et al. (23)]. Instead, it is the estimated self-report rate divided by the corrected rate (to medical record), which incorporates adjustment.

  • Overall screening adherence via any one of the four tests per recommendations.