================================================================= COMPARE2 Version 1.90 Tuesday, 13th January 2009, 15:50. ----------------------------------------------------------------- Measures with number-of-individuals denominators [module B] ================================================================= DATA: A: Yes, 2 No, 115 Denominator, 117 B: Yes, 0 No, 862 Denominator, 862 RESULTS: Proportions (of "Yes"): A, 0.0171 B, 0.0000 If inverse sampling was used, see results at end of output. [Campbell (2007) recommends use of Fisher's exact two-tailed test for this table.] Exact tests: Fisher's P: One-tailed: P = 0.014 or 1.000 Two-tailed: P = 0.014 Double one-tailed: P = 0.028 Mid-P: One-tailed: P = 0.007 [ 7.1E-3 ] or 0.993 Two-tailed: P = 0.007 [ 7.1E-3 ] Double one-tailed: P = 0.014 Overall's continuity-corrected P: One-tailed: P = 0.002 [ 1.7E-3 ] or 1.000 Two-tailed: P = 0.002 [ 1.7E-3 ] Double one-tailed: P = 0.003 [ 3.4E-3 ] Tocher's test P > 0.001 {one-tailed) Upton's "N - 1" chi-sq. = 14.750 P = 0.000 [ 1.2E-4 ] Pearson's chi-square = 14.765 P = 0.000 [ 1.2E-4 ] with Yates's correction = 7.571 P = 0.006 [ 5.9E-3 ] Haber's adjusted chi-sq. = 7.571 P = 0.006 [ 5.9E-3 ] WARNING: 2 cells have an expected frequency of <5. 1 cell has an expected frequency of <1. Log-likelihood chi-square with Yates's correction = 4.267 P = 0.039 DIFFERENCE [A minus B] = 0.017 S.E. = 0.012 Large-sample method (Fleiss), continuity-corrected: 90% C.I. = -0.007 to 0.042 95% C.I. = -0.011 to 0.045 99% C.I. = -0.019 to 0.053 Wilson's score method: Not continuity-corrected (Newcombe's method 10): 90% C.I. = 0.005 to 0.050 95% C.I. = 0.004 to 0.060 99% C.I. = 0.001 to 0.083 Continuity-corrected (Newcombe's method 11): 90% C.I. = 0.003 to 0.056 95% C.I. = 0.002 to 0.067 99% C.I. = -0.000 to 0.089 RATIO [A:B] = infinity Low-bias estimator of ratio in population = 14.752 ODDS RATIO = infinity. Fisher's exact confidence intervals: 90%: 2.1321 to infinity 95%: 1.3899 to infinity 99%: 0.5629 to infinity Mid-P exact confidence intervals: 90%: 3.4253 to infinity 95%: 2.1321 to infinity 99%: 0.8221 to infinity Cornfield's confidence intervals: 90%: 2.4404 to infinity (approx. C.I.) 95%: 1.8249 to infinity (approx. C.I.) 99%: 1.1277 to infinity (approx. C.I.) S.E. of log O.R. = 1.552 Low-bias indicator of O.R. in the population = 14.86 Adjusted O.R. (0.5 added in each cell) = 37.34 Yule's Q = 1.00 Phi = 0.12 Lambda (for prediction of Yes-No distribution) = 0.00 NUMBER NEEDED: Subjects needed in B to avoid 1 event*: 58.5 95% C.I.: 15.0 to 520 Assuming a risk of R per 1 in group B, number needed in group B to avoid 1 event* (if the risk ratio seen in this study sample is appropriate) = 1/R *If this is a therapeutic trial where A = "treated" and B = "controls", this is the NNTH (number needed to treat to produce one episode of harm). MEASURES OF IMPACT (if Group A is exposed, and Group B not exposed, to a risk factor] Attributable fraction in exposed = 100.0% Attributable fraction in population = 100.0% INVERSE SAMPLING The following results are applicable if, in each group (A and B), subjects were added until a prespecified number of cases (subjects with "Yes") were found. Large-sample test: one-tailed P = 0.030 two-tailed P = 0.060 DIFFERENCE [A minus B] = 0.017 S.E. = 0.003 Unbiased estimate = 0.010 Confidence intervals: 90% conf. interval = 0.007 to 0.012 95% conf. interval = 0.007 to 0.013 99% conf. interval = 0.006 to 0.014 RATIO (A:B) = infinity ADJUSTED RATIO (A:B) = 36.838 (0.5 added to each numerator) Unbiased estimate = 22.197 Confidence intervals: 90% conf. interval = 2.886 to 466.787 95% conf. interval = 1.773 to 759.797 99% conf. interval = 0.684 to 1968.867 ODDS RATIO = 37.34 [reciprocal = 0.03] (adjusted by adding 0.5 in each cell) 90% C.I. = 2.91 to 478.35 95% C.I. = 1.79 to 779.71 99% C.I. = 0.69 to 2025.99 ======================================================*===========