A French-Language Web-Based Intervention Targeting Prolonged Grief Symptoms in People Who Are Bereaved and Separated: Randomized Controlled Trial.

Abstract

Losing a loved one, through death or separation, counts among the most stressful life events and is detrimental to health and well-being. About 15% of people show clinically significant difficulties coping with such an event. Web-based interventions (WBIs) are effective for a variety of mental health disorders, including prolonged grief. However, no validated WBI is available in French for treating prolonged grief symptoms. This study aimed to compare the efficacy and adherence rates of 2 WBIs for prolonged grief symptoms following the loss of a loved one through death or romantic separation. LIVIA 2.0 was developed relying on theoretical and empirical findings on bereavement processes and WBIs, and is compared with LIVIA 1, which has already demonstrated its efficacy. We conducted a randomized controlled trial and provided on-demand guidance to participants. Outcomes were assessed through web-based questionnaires before the intervention, after the intervention (12 weeks later), and at follow-up (24 weeks later). Primary outcomes were grief symptoms, depressive symptoms, and well-being. Secondary outcomes were anxiety symptoms, grief coping strategies, aspects related to self-identity, and program satisfaction. In total, 62 participants were randomized (intent-to-treat [ITT] sample), 29 (47%) in LIVIA 2.0 (active arm) and 33 (53%) in LIVIA 1 (control arm). The dropout rate was 40% (37/62), and 10 participants were removed due to exclusion criteria, leading to a final per-protocol sample of 27 (44%) completers who differed from noncompleters only based on reporting fewer anxiety symptoms (t60=3.03; P=.004). Participants who are separated reported more grief symptoms (t60=2.22; P=.03) and attachment anxiety (t60=2.26; P=.03), compared to participants who are bereaved. There were pre-post within-group differences for both programs in the ITT sample, with significant reductions in grief (Cohen d=-0.90), depressive symptoms (Cohen d=-0.31), and centrality of the loss (Cohen d=-0.45). The same pattern was observed in the per-protocol sample, with the exception that anxiety symptoms also significantly diminished (Cohen d=-0.45). No difference was found in efficacy between the 2 programs (all P>.33). Participants (ITT sample) reported overall high levels of program satisfaction (mean 3.18, SD 0.54; over a maximum of 4). Effect stability was confirmed at the 6-month follow-up for all outcomes, with an improvement in self-concept clarity. The 2 grief-related WBIs were effective in reducing grief, depressive and anxiety symptoms for participants who are bereaved or separated. The analyses did not reveal any pre-post between-group differences, suggesting that the innovations brought to LIVIA 2.0 did not significantly affect the outcome. However, caution is warranted with the interpretation of the results given the limited power of the sample, which only allows the detection of medium effect sizes. ClinicalTrials.gov NCT05219760; https://clinicaltrials.gov/study/NCT05219760. RR2-10.2196/39026.