Factors Associated With Treatment Response in an Internet-Based Intervention for Prolonged Grief Disorder After Cancer Bereavement

Abstract

Internet-based psychological interventions have proven effective in the treatment of prolonged grief disorder (PGD). Yet, some patients do not benefit from treatment in a clinically significant way. We aimed to examine predictors of symptom reduction in an Internet-based intervention for PGD after cancer bereavement, in order to identify possible treatment mechanisms and discern directions for future intervention design. A secondary analysis of data from a randomized wait-list controlled trial on an Internet-based intervention for PGD after cancer bereavement was conducted. Multiple regression models were used (1) to test for the influence of pretreatment PGD, working alliance, avoidance and gender on PGD symptom reduction; and (2) to explore further predictors of treatment success with a best subset selection protocol. The regression models explained 18% (Model 1) and 34% (Model 2) of variance in symptom reduction. Participants with more favorable symptom change had more severe pretreatment PGD scores and better working alliance. Those with lower social support and less posttraumatic growth experienced more PGD symptom change. In conclusion, therapeutic alliance is an important factor that should be monitored and fostered. Findings regarding social support and posttraumatic growth need further replication and clarification.