Long-term effects on depression and anxiety of an internet-based stepped care intervention for patients with cancer and symptoms of depression and anxiety. The U-CARE AdultCan trial

BACKGROUND: Cancer is a serious disease that commonly causes significant psychological distress. The internet-based intervention (iCAN-DO), utilizing a stepped care approach for the treatment of anxiety and depression in individuals with cancer, has been shown to have favorable results for symptoms of depression at the primary endpoint, 10 months after randomization compared to standard care (SC). The aim of the present study was to evaluate the long-term effects of the intervention 18 and 24 months after randomization. METHODS: Patients with breast, colorectal, or prostate cancer and a score > 7 on either of the Hospital Anxiety and Depression Scale (HADS) subscales (n = 245) were recruited to the study in conjunction with a regular hospital visit. They were randomized to access to the stepwise iCAN-DO intervention for 24 months or to SC. Step 1 of the intervention comprised psycho-educative online material. In Step 2, internet-based cognitive-behavioral therapy with individual online support from a therapist was added. Step 2 was offered to those who reported persistent anxiety or depression symptoms (>7 on HADS), also at 1, 4, and/or 7 months after randomization. Missing data were imputed using the last rank carried forward method and used for the main analyses according to the intention-to-treat approach. Effects sizes (Cohen's d), and minimally clinically important difference (MCID) were calculated. Linear mixed models were used to analyze intervention effects over time. RESULTS: Symptoms of depression decreased significantly (p < 0.05) in the iCAN-DO group compared with the SC group from baseline to 18 months (d = 0.29), but not to 24 months (d = 0.27). Even though the average iCAN-DO group participant surpassed a MCID in symptoms of anxiety (>2 p) at both long-term follow-ups, the differences did not reach statistical significance, either at 18 months (p = 0.10) or 24 months (p = 0.09). Positive effects of iCAN-DO compared with the SC were also shown for some secondary HRQoL-outcomes; social functioning at 18 months (p = 0.02) and 24 months (p = 0.001), and sleep problems at 24 months (p = 0.01). CONCLUSION: A stepped-care internet-based intervention that has previously shown positive results for symptoms of depression at 10 months did show similar positive long-term effects also at 18 months. For symptoms of anxiety, no effect could be shown. The internet may provide an effective format for interventions to reduce symptoms of depression after cancer at patients' own choice of time, regardless of distance to a psycho-oncology clinic.