Digital and AI-supported psychosocial interventions modestly reduce distress and improve quality of life in women with breast cancer.

This systematic review and meta-analysis synthesized data from randomized controlled trials to assess the impact of digital and AI-supported psychosocial interventions on women with breast cancer. The study population comprised approximately 7,551 participants, though the specific setting was not reported in the available data. The interventions evaluated included cognitive behavioral therapy, mindfulness, psychoeducation, AI-assisted formats, and blended delivery methods. These were compared against static or minimally interactive platforms. The primary outcomes assessed were stress, anxiety, depression, and health-related quality of life (HRQoL). Follow-up periods were generally short, and the evidence certainty is rated as moderate, constrained by significant methodological limitations.

The meta-analysis found that digital and AI-supported interventions yielded modest but consistent reductions in stress compared to control conditions. Effect sizes were expressed as standardized mean differences (Hedges' g). Similarly, anxiety levels showed modest but consistent reductions with the same metric. Depressive symptoms also demonstrated modest but consistent reductions across the included trials. In terms of health-related quality of life, the interventions resulted in measurable improvements. While specific absolute numbers and precise confidence intervals were not detailed in the provided data, the direction of effect was consistent across outcomes.

Safety and tolerability data were not reported for adverse events, serious adverse events, discontinuations, or general tolerability. Consequently, no specific rates or details regarding safety profiles could be extracted from the current evidence. The lack of reported safety data represents a notable gap in the comprehensive assessment of these interventions.

When compared to prior landmark studies in supportive oncology, these findings suggest that digital and AI-supported tools may serve as viable adjuncts to conventional care. The potential exists to improve access to supportive services, particularly for populations facing barriers to traditional face-to-face therapy. However, the current evidence base relies heavily on self-reported outcomes and small single-centre samples, which limits the robustness of the conclusions drawn from these trials.

Several key methodological limitations must be considered when interpreting these results. There was substantial heterogeneity in intervention design, making direct comparisons difficult. The reliance on self-reported outcomes introduces potential bias, and the short follow-up periods prevent conclusions regarding long-term sustainability of benefits. Furthermore, the studies exhibited minimal cultural adaptation and lacked representation from low- and middle-income countries, raising concerns about generalizability.

Clinical implications suggest that digital and AI-supported psychosocial interventions may be integrated as adjuncts to standard oncology care. This approach could potentially expand access to supportive services for women with breast cancer. However, clinicians should exercise caution regarding the generalizability of findings, the sustainability of effects beyond short follow-up, and the feasibility of clinical integration given the current heterogeneity in study designs.

Several critical questions remain unanswered. The long-term durability of symptom reduction and quality of life improvements has not been established due to short follow-up periods. The optimal format for delivery, particularly the comparative efficacy of AI-assisted versus purely human-led or blended formats, requires further investigation. Additionally, the lack of diversity in cultural adaptation and geographic representation means that the applicability of these interventions to diverse patient populations remains uncertain. Future research must address these gaps to provide a more definitive evidence base for clinical practice.