The efficacy of complicated grief therapy for <scp>DSM‐5‐TR</scp> prolonged grief disorder

The American Psychiatric Association recently announced the inclusion in the DSM-5-TR of a new category for prolonged grief disorder (PGD)1, 2, following introduction of this category in the ICD-11. Our group previously demonstrated the efficacy of a targeted treatment (complicated grief therapy, CGT) for complicated grief, a condition corresponding in many respects to PGD. We examined now the performance of that treatment among people who met the DSM-5-TR criteria for PGD. CGT is a manualized 16-session intervention developed when we observed that treatments for depression did not appear to be effective for complicated grief3. We considered loss of a loved one to be a major life stressor4 and understood grief from an attachment theory perspective5. We conceptualized grief after attachment loss as typically emerging in an acute form and becoming integrated over time as the reality of the loss is accepted and the capacity for well-being is restored. We understood complicated grief as a condition in which the initial intense form of grief persisted and interfered with functioning. A body of research informed our understanding of impediments to adapting to the loss. We developed a treatment that focused on facilitating adaptation to loss and addressing impediments, drawing upon strategies and techniques from prolonged exposure, motivational interviewing, positive psychology, interpersonal psychotherapy, and psychodynamic psychotherapy. CGT was tested in three randomized controlled trials funded by the US National Institute of Mental Health6-8. For the present report, we analyzed data from one of these trials6, in which participants (N=395) were people with a score of 30 or higher on the Inventory of Complicated Grief (ICG) who underwent a clinical interview confirming that grief was the primary problem. People with current substance use disorder, or a lifetime history of psychotic disorder, bipolar I disorder, active suicidal plans requiring hospitalization, or a Montreal Cognitive Assessment score less than 21 were excluded. These patients were evaluated through the Structured Clinical Interview for Complicated Grief (SCI-CG), an instrument that can be used to identify DSM-5-TR criteria for PGD9. The evaluation was available for 307 study participants, 77 (25.1%) of whom were bereaved between 6 and 12 months and therefore did not meet the DSM-5-TR criteria solely due to time considerations. Of the remaining 230, 194 (84.3%) met DSM-5-TR criteria for PGD and 36 (15.7%) did not. All patients recruited for the parent study were randomized either to citalopram or to placebo, with or without CGT6. Among patients meeting criteria for PGD (N=194), we compared study outcomes at endpoint (week 20) for those who received CGT (N=96) versus those who did not receive it (N=98). The main outcome was treatment response measured as a rating of “much improved” or “very much improved” on the Clinical Global Impression (CGI) Improvement. We further used several grief symptom measures: the ICG, the Grief-Related Avoidance Questionnaire (GRAQ), the Typical Beliefs Questionnaire (TBQ), and the Grief-Related Work and Social Adjustment Scale (WSAS). Chi-squared tests were used for binary outcomes and two sample t-tests for continuous outcomes. All hypothesis tests were two-sided with a 5% level of significance. All analyses were performed in R (v1.4.1717). The parent study had been approved by the relevant institutional review board6. Written informed consent had been obtained from all participants before baseline assessment. The sample of patients with PGD was not significantly different with respect to demographic and clinical variables from the parent study sample. Most patients were female (79.9%), white (80.9%), completed at least partial college (90.2%), and were bereaved of a parent or spouse (68.6%) by illness (65.5%) for 4-5 years on average. The sample had an average age of 52.7±14.2 years. Patients had high rates of current depression (69.6%), current post-traumatic stress disorder (46.4%), and suicidal ideation since the loss (61.9%) (see also supplementary information). Treatment response for the sample with PGD closely reflected that of the parent study. Specifically, response rates for those randomized to CGT vs. no CGT were 88.2% vs. 60.9% (p<0.001) for the DSM-5-TR PGD group compared to 82.9% vs. 63.4% for all participants in the parent study. Also comparable to the parent study, average post-treatment scores on grief-related symptoms and impairment were significantly lower for those who received CGT vs. no CGT (ICG: 17.7 vs. 25.4, p<0.001; WSAS: 7.9 vs. 13.4, p=0.001; GRAQ: 9.4 vs. 14.6, p=0.01; TBQ: 3.9 vs. 7.1, p<0.001) (see also supplementary information). Our results indicate that study participants who met DSM-5-TR criteria for PGD showed no significant demographic or clinical differences from the full parent study sample. Those diagnosed with PGD showed significantly greater response rates to CGT vs. no CGT, with results nearly identical to the parent study. These findings are limited by the need to apply retrospectively the DSM-5-TR criteria for PGD, and diagnosis may have been less accurate than if made using a validated instrument1. Additionally, those diagnosed with PGD for these analyses represented only half of the originally randomized sample. However, almost half (43.8%) of the omitted participants simply did not receive the assessment needed to diagnose PGD, and another 38% were excluded because it was too soon (six months to one year since the loss) to receive a PGD diagnosis. Further, those assessed showed no differences in demographic or clinical characteristics from participants in the parent study. We endorse continued study of effective treatments for PGD. In the meantime, we believe that clinicians will benefit from knowing that CGT, a strongly validated intervention6-8, can be appropriately re-labeled as prolonged grief disorder therapy (PGDT).

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.