. Author manuscript; available in PMC: 2024 Mar 1.

Published in final edited form as: Psychoneuroendocrinology. 2022 Nov 24;149:105989. doi: 10.1016/j.psyneuen.2022.105989

Table 3.

Multilevel Model Coefficients for Significant Interactions Between Communication Patterns, Coded Behaviors, and Gender Across Immune, Emotional, and Relational Outcomes

Outcome	b (SE)	p
Baseline IL-6
Husband demand/wife withdraw*Gender	0.01 (0.002)	<0.001
Roles in demand/withdraw*Gender	−0.005 (0.002)	0.004
Post-Discussion Negative Emotion
Negative behavior*Gender	0.04 to 0.05 (0.02)	<0.008
Post-Discussion Positive Emotion
Mutual discussion avoidance*Gender	−0.56 (0.27)	0.037^a
Mutual constructive communicationNegative behaviorGender	−0.02 (0.01)	0.02
Husband demand/wife withdrawPositive behaviorGender	−0.01 (0.01)	0.045^a
Post-Discussion Evaluation
Husband demand/wife withdraw*Negative behavior	0.003 (0.001)	0.047^a
Wound healing (TEWL AUC)
Mutual discussion avoidance*Positive behavior	0.35 (0.11)	0.002

Note. Only significant interactions are shown; all interaction effects are reported in text. Negative behavior coefficent ranges reflect effects across models with each self-reported communication pattern.

This effect remained significant after a Benjamini–Hochberg false discovery rate correction of .15 but not .10.