Optimizing smartphone psychotherapy for depressive symptoms in patients with cancer: Multiphase optimization strategy using a decentralized multicenter randomized clinical trial (J-SUPPORT 2001 Study)

Abstract

Aim: Patients with cancer experience various forms of psychological distress, including depressive symptoms, which can impact quality of life, elevate morbidity risk, and increase medical costs. Psychotherapy and pharmacotherapy are effective for reducing depressive symptoms among patients with cancer, but most patients prefer psychotherapy. This study aimed to develop an efficient and effective smartphone psychotherapy component to address depressive symptom.

Methods: This was a decentralized, parallel-group, multicenter, open, individually randomized, fully factorial trial. Patients aged ≥20 years with cancer were randomized by the presence/absence of three cognitive-behavioral therapy (CBT) skills (behavioral activation [BA], assertiveness training [AT], and problem-solving [PS]) on a smartphone app. All participants received psychoeducation (PE). The primary outcome was change in the patient health questionnaire-9 (PHQ-9) total score between baseline and week 8. Secondary outcomes included anxiety.

Results: In total, 359 participants were randomized. Primary outcome data at week 8 were obtained for 355 participants (99%). The week 8 PHQ-9 total score was significantly reduced from baseline for all participants by -1.41 points (95% confidence interval [CI] -1.89, -0.92), but between-group differences in change scores were not significant (BA: -0.04, 95% CI -0.75, 0.67; AT: -0.16, 95% CI -0.87, 0.55; PS: -0.19, 95% CI -0.90, 0.52).

Conclusion: As the presence of any of the three intervention components did not contribute to a significant additive reduction of depressive symptoms, we cannot make evidence-based recommendations regarding the use of specific smartphone psychotherapy.

Keywords: decentralized clinical trial; depressive symptoms; multiphase optimization strategy; neoplasm; smartphone‐based CBT.

Fig. 1

Participant flow diagram (CONSORT diagram).

Fig. 2

Effect size of patient health questionnaire‐9 change scores and their differences for each component at week 8. The effect size of change scores was calculated using the baseline standard deviation (SD), and the effect size for the differences between presence and absence was calculated using the week 8 SD.