The interplay between partners’ responsiveness and patients’ need for emotional expression in couples coping with cancer

We also examined the models including gender, and morbidity as covariates because of their correlation with patients’ depressive symptoms. Patients’ were asked to indicate whether they had additional health complaints or not on a checklist of 26 chronic medical condition. Morbidity was coded as present if patients indicated one or more medical conditions other than cancer. Including these covariates did not change the results. Therefore, we excluded them from the final model to assure enough power for testing our hypotheses. We further examined the associations between depressive symptoms and several other demographic, and medical variables but none of these variables showed a significant correlation with depressive symptoms. Therefore they were not included in the final analysis. Finally, we have examined the association between several patient behaviors during the interaction (i.e., self-disclosure, frequency of speech turns, and total duration of patients’ speech) and other variables included in the models. Except for patient’s self-disclosure that was found to be related with patients’ need for emotional expression, none of the other behaviors were related to any of the predictors or outcome and therefore only patient’s self-disclosure was included in the final models.

We also conducted one hierarchical regression analysis for the aggregated component of partners’ responsiveness (i.e., the average score of understanding, validation, and caring) and found similar results as for the models of understanding and validation R ² = .34 for step 1; ∆R ² = .10 for step 2 (p = .08); ∆R ² = .06 for step 3 (p < .05).