Figure 3.

Effects and noninferiority ranges one week after post-test counseling for women who were tested. The between-group intervention differences were estimated using linear models together with one-sided 97.5% cluster bootstrap confidence intervals. One-sided 97.5% confidence intervals were used because the goal was to test if telephone counseling was not unacceptably worse than in-person counseling. IP-GC = in-person genetic counseling; TEL-GC = telephone genetic counseling.