Skip to main content
. Author manuscript; available in PMC: 2014 Nov 1.
Published in final edited form as: Prev Med. 2013 Sep 9;57(5):10.1016/j.ypmed.2013.08.032. doi: 10.1016/j.ypmed.2013.08.032

Table 4.

Proportion of analyzed participants returning mailed FOBT, by randomization group, Group Health, Washington State 2010-2012a

OC-Auto® InSure® Hemoccult SENSA® Omnibusb Posthoc pairwise testsc
N = 739 N = 741 N = 754 Gray's test OC-Auto® vs. InSure® OC-Auto® vs. Hemoccult SENSA® InSure® vs. Hemoccult SENSA®
Any FOBT returned P < 0.001 P = 0.005 P < 0.001 P = 0.176
    by 3 months 0.64 (0.60, 0.67) 0.60 (0.56, 0.63) 0.55 (0.52, 0.59)
    by 6 months 0.69 (0.66, 0.72) 0.64 (0.61, 0.68) 0.61 (0.58, 0.65)
Study FOBT returned P < 0.001 P = 0.016 P < 0.001 P = 0.008
    by 3 months 0.61 (0.57, 0.64) 0.57 (0.54, 0.61) 0.50 (0.46, 0.53)
    by 6 months 0.65 (0.61, 0.68) 0.61 (0.57, 0.64) 0.54 (0.51, 0.58)
Any CRC testing completed P < 0.001 P = 0.006 P < 0.001 P = 0.155
    by 3 months 0.65 (0.62, 0.69) 0.61 (0.57, 0.64) 0.56 (0.53, 0.60)
    by 6 months 0.71 (0.68, 0.74) 0.67 (0.64, 0.70) 0.64 (0.60, 0.67)

Abbreviations: CRC, colorectal cancer; FOBT, fecal occult blood test.

a

Data are proportion (95% confidence interval).

b

Omnibus Gray's test is for differences in cumulative incidence estimates across all three none, with follow-up truncated at 6 months. Estimates of cumulative incidence accommodate competing risks, as described in the text.

c

Posthoc pairwise Gray's tests are significant when P is below 0.05/3 = 0.0167, based on a Bonferroni correction.