Sophia K Smith, Eric Kuhn, Tamara J Somers, Eric Laber, Cole Manschot, Anthony D Sung, Sarah A Kelleher, Catherine Majestic, Karen L Syrjala, Deborah Mayer, Sergio Giralt, Andrea Pires, Rebecca R Gebert, Jason E Owen, Jean C Yi, Allison J Applebaum
Although a self-guided digital intervention did not outperform Usual Care initially, adaptive stepped-care strategies that triage early responders and escalate non-responders to therapist-delivered CBT-PTSD were associated with the greatest reductions in symptoms. Digital tools may function as entry points within stepped-care models to allocate higher-intensity psychotherapy.
OBJECTIVE: Cancer-related posttraumatic stress disorder (PTSD) is prevalent among hematopoietic cell transplantation (HCT) survivors, yet access to evidence-based care is limited. This study evaluated adaptive, stepped-care treatment sequences for PTSD symptoms using a Sequential Multiple Assignment Randomized Trial (SMART), comparing digital and therapist-delivered interventions.
METHODS: HCT survivors (N = 477), 1-5 years post-transplant, with probable or subthreshold PTSD, were randomized to the Cancer Distress Coach (CaDC) app or Usual Care. At Week 4, non-responders (i.e., < 5-point PCL-5 reduction) were re-randomized to CaDC + Coaching or therapist-delivered cognitive behavioral therapy for PTSD (CBT-PTSD). PTSD symptoms (primary outcome), depression, and anxiety were assessed at baseline, Week 4, Month 3 (primary endpoint), and Month 6.
RESULTS: Both initial treatment groups demonstrated symptom reductions at Week 4 with no significant between-group differences; approximately 60% met early response criteria. By Month 3, outcomes reflected treatment escalation rather than direct comparison of initial interventions. Regimen-level analyses demonstrated differences across adaptive sequences for PTSD at Months 3 and 6, and for depression and anxiety at Month 3 (global p-values ≤ 0.018). Regimens escalating to CBT-PTSD were associated with greater and more sustained reductions than those escalating to CaDC + Coaching. The sequence initiating with CaDC and escalating to CBT-PTSD produced the largest effects.
CONCLUSIONS: Although a self-guided digital intervention did not outperform Usual Care initially, adaptive stepped-care strategies that triage early responders and escalate non-responders to therapist-delivered CBT-PTSD were associated with the greatest reductions in symptoms. Digital tools may function as entry points within stepped-care models to allocate higher-intensity psychotherapy.
TRIALS REGISTRATION: ClinicalTrials.gov, NCT04058795, registered 8/16/2019.