Skip to main content
. 2008 Aug 19;10(1):172–186. doi: 10.1093/biostatistics/kxn025

Table 1.

Baseline (pretest) and augmented (posttest) models for risk of hearing impairment based on a nested case–control substudy of the Neonatal Audiology for test B. Shown are log odds ratios for P(D = 1|X) and P(D = 1|X, Y) and coefficients for covariates in the corresponding model for logDLRX(Y). Age is centered at 35 weeks. p-values in parentheses are calculated by comparing estimates with standard errors

Factor Pretest risk Posttest risk log DLRX(Y)
Intercept – 0.05 (0.67) – 0.58 (< .001) – 0.54 (1.00)
Age (weeks) – 0.04 (0.04) – 0.04 (0.05) 0.002 (0.27)
Location (booth versus room) 0.10 (0.49) 0.15 (0.34) 0.04 (0.72)
Age × location 0.07 (0.02) 0.07 (0.03) – 0.005 (0.47)
Test result (Yes versus No) 1.06 (< 0.001) 1.06 (< 0.001)