Palliative care and coping in patients with acute myeloid leukemia receiving intensive induction therapy: A mediation analysis of data from a randomized trial.
LeBlanc, TW; Kavanaugh, A; Webb, J; Jackson, VA; O'Connor, N; Bhatnagar, B; Fathi, AT; Brunner, AM; Greer, JA; Temel, JS; El-Jawahri, A
Published in: Journal of Clinical Oncology
12007 Background: Patients with acute myeloid leukemia (AML) receiving intensive induction chemotherapy experience significant symptoms, quality of life impairment, anxiety, and depression. We previously showed that integrated palliative care improves quality of life (QOL), anxiety and depression during induction, but the mechanism of these benefits is unexplained. Methods: We conducted a non-blinded, multi-site randomized trial of integrated palliative and oncology care (IPC) (n = 86) vs usual care (n = 74) for hospitalized patients with AML receiving induction therapy. IPC patients were seen by palliative care clinicians at least twice weekly while hospitalized. Patients completed the Functional Assessment of Cancer Therapy-Leukemia scale, the Hospital Anxiety and Depression Scale, and the Brief COPE questionnaire, to assess QOL, mood, and coping at baseline and weeks 2, 4, 12, and 24. To facilitate analysis, we categorized coping strategies into “approach-oriented” (active coping, positive reframing, and acceptance) or “avoidant” (denial, or self-blame), per prior studies. We used linear regression models adjusting for baseline scores to assess the effect of the intervention on coping at week 2. We used causal mediation regression models to examine whether changes in coping during the initial hospitalization mediated intervention effects on QOL, depression and anxiety symptoms at week 2, when patients feel worst during induction. Results: We enrolled 160 of 235 (68.1%) eligible patients. Those randomized to IPC reported more approach-oriented coping (B = 1.85, 95%CI 0.62-0.38, P = 0.004) and less avoidant-oriented coping (B = -0.70, 95%CI -1.28, -0.11, P = 0.020) at week 2. Intervention effects on approach-oriented coping were sustained up to week 24 (B = 0.36, 95%CI 0.68, 0.09, P = 0.010), but not on avoidant-oriented coping (B = -0.01, 95%CI -0.28- 0.05, P = 0.163). Changes in approach-oriented coping and avoidant-oriented coping during hospitalization partially mediated the intervention effects on week 2 QOL (indirect effect = 6.58, 95%CI 2.14, 13.63), depression (indirect effect = -1.08, 95%CI -2.05, -0.27), and anxiety symptoms (indirect effect = -0.52, 95%CI -1.25, -0.04). Changes in approach- and avoidant-oriented coping accounted for 78% of the total palliative care intervention effect on QOL; 66% of the intervention effect on depression; and 35% of the intervention effect on anxiety. Conclusions: IPC as part of induction chemotherapy for AML facilitates active coping strategies for patients, while reducing avoidant coping. Improvement in coping skills accounts for a substantial proportion of the effect from a palliative care intervention on QOL, depression, and anxiety symptoms. These findings offer important insights into the mechanism by which palliative care may enhance patient-reported outcomes in patients with AML. Clinical trial information: NCT02975869.
