Table 4.

Statistical comparison of differences in the overall sensitivity of claims according to case definition and event type

ComparisonAggregate Sensitivity, p1_hat (x1/n1)Aggregate Sensitivity, p2<_hat (x2/n2)Z StatisticaP
Part A or B (1) versus Part A alone (2), all events, ±30-d ascertainment window90.9% (160/176)82.4% (145/176)2.350.009b
Part A or B (1) versusPart B alone (2), all events, ±30-d ascertainment window90.9% (160/176)84.6% (149/176)1.790.037b
Procedures (1) versus diagnoses (2), Part A or B claims, up to ±90-d ascertainment window100% (40/40)91.2% (124/136)1.950.05c
Claims ascertainment window up to ±90 d (1) versus ±30 d (2), using Part A or B claims
    all events93.2% (164/176)90.9% (160/176)0.790.22b
    atrial fibrillation and venous thromboembolism96.0% (48/50)90.0% (45/50)1.180.12b
    all other events92.1% (116/126)91.3% (115/126)0.220.41b
Adapted “Hebert method” (1) versus Part A or B claims (2) for diagnoses88.2% (120/136)89.0% (121/136)−0.190.43b
  • a Z = (p1_hat − p2_hat)/{[pp_hat(1 − pp_hat)]1/2 *[(1/n1) + (1/n2)]1/2 }, wherein the proportions compared are p1_hat = (x1/n1) and p2_hat = (x2/n2), and pp_hat= (x1+ x2/n1+ n2). Z >1.645 is significant at P < 0.05 when using a one-tailed test. Z >1.960 is significant at P < 0.05 when using a two-tailed test.

  • b Hypotheses with only one possible direction of difference were examined with a one-tailed test. Specifically, the addition of claims data for event ascertainment by inclusion of both Medicare parts or by an expanded capture window could increase sensitivity or produce no change but not reduce sensitivity; the requirement for additional confirmatory diagnosis claims could reduce or not alter sensitivity achieved with single claims but not increase sensitivity.

  • c The difference in sensitivity between procedures and diagnoses was examined with a more conservative two-tailed test.