Change in avoidance and negative grief-related cognitions mediates treatment outcome in older adults with prolonged grief disorder

Abstract

Objective: The present study investigated the role of the two theoretically derived mediators in the treatment of Prolonged Grief Disorder (PGD). Mediators were changes in avoidance and maladaptive cognitions. An additional hypothesis tested whether these candidate mediators are specific to CBT-based Complicated Grief Treatment (CGT) compared to Interpersonal Therapy (IPT). Method: We performed secondary analyses with assessment completers (n = 131) from a randomized-controlled trial with older adults with PGD. Patients received 16 sessions of CGT or IPT. Outcomes were treatment response and reductions in grief symptoms and grief-related related impairment. Results: Reductions in avoidance between baseline and week 16 mediated reductions in grief symptoms and grief-related impairment. Reductions in maladaptive grief-related cognitions over the same period mediated treatment response, reductions in grief symptoms and grief-related impairment. There were no significant treatment-mediator interactions. We could not establish that mediators changed before the outcomes. Conclusion: Results are consistent with theoretical models of PGD, including the CGT treatment model. Despite different therapeutic procedures, we found no significant interaction effect, but CGT produced larger effects. Future research needs to establish a timeline of change through the use of multiple measurements of mediators and outcomes.Trial registration: ClinicalTrials.gov identifier: NCT01244295.

Keywords: avoidance; complicated grief/prolonged grief disorder; maladaptive cognitions; mediation; older adults; treatment.

Conflict of interest statement

Disclosure of interest: No potential conflict of interest was reported by the authors.

Figures

Figure 1.

Mediation model for hypotheses 1 and 2 (a; top) and analyses of temporal precedence of the mediators (b; bottom). Path c represents the total effect of the treatment on the outcomes. Path a presents the effect of the treatment on the mediators. Path b represents the effect of the mediators on the outcomes. Path c′ represents the effect of treatment on the outcomes when the effect of the mediator is taken into account (direct effect). The indirect effect of treatment on outcomes via the mediator is represented by the product of the coefficients a and b and noted as ab.