Skip to main content
. Author manuscript; available in PMC: 2023 Apr 1.
Published in final edited form as: Drug Alcohol Depend. 2022 Mar 2;233:109384. doi: 10.1016/j.drugalcdep.2022.109384

Table 2.

GEE modeling alcohol problems at follow-up (3, 6, 12 month) predicted by baseline discrimination, controlling for treatment condition, time-varying alcohol use, baseline alcohol problems, and the discrimination by treatment condition interaction.

Variable B SE p 95% Wald CI
Within participant
 Time −.069 .048 .145 −.162 .024
 Time-varying alcohol use .011 .004 .001 .004 .018
Between participants
 MI (vs. TAU) .072 .171 .675 −.264 .408
 Baseline discriminationa .063 .021 .002 .022 .103
 Baseline alcohol problems .080 .006 .000 .067 .092
 Baseline drinks per week −.004 .005 .480 −.014 .007
Interaction
 Discrimination by MI −.063 .029 .029 −.119 −.007

Note. MI = motivational interviewing treatment. TAU = HIV treatment as usual assessment only control condition. Discrimination by MI reference group for MI = TAU.

a

Given the interaction of discrimination and MI, this equals the effect of discrimination in participants receiving TAU.