A systematic review of the association of depression and psychological distress with cancer incidence and mortality risk in people with chronic diseases: evidence primarily from diabetes populations

Abstract

Objective: This systematic review examined the association between depression, anxiety, psychological distress, and the risks of cancer incidence and mortality in adults with chronic diseases, to inform integrated chronic disease management. Methods: PubMed, Embase, Web of Science, and the Cochrane Library were searched from inception to April 2026. Eligible studies included adults (≥18 years) with chronic diseases, or those allowing subgroup extraction for such populations. Exposures of interest were depression, anxiety, or psychological distress assessed before cancer outcomes. Outcomes included overall and site-specific cancer incidence and mortality. Core evidence from chronic disease populations was synthesized separately from supplementary general-population evidence. Results: All directly relevant core evidence was confined to diabetes populations. In a Taiwanese cohort, diabetes with comorbid depression was associated with a mildly increased overall cancer incidence (adjusted hazard ratio [HR] = 1.08, 95% confidence interval [CI]: 1.05–1.11). A US National Health Interview Survey–National Death Index (NHIS-NDI) study showed that higher psychological distress in diabetes was linked to increased all-cause mortality (HRs: 1.39–1.98), with cancer mortality showing a similar but non-significant interaction. A United Kingdom (UK) Biobank study reported that comorbid diabetes and depression increased cancer mortality risk (HR = 1.62, 95% CI: 1.35–1.93), potentially exceeding additive effects. Pancreatic cancer evidence suggested new-onset depression, anxiety, and diabetes were associated with increased risk, but mainly driven by treatment initiated within 3 months prediagnosis, indicating reverse causation. Supplementary general-population evidence generally supported associations but cannot substitute for chronic-disease-specific data. Conclusion: In chronic disease populations, particularly diabetes, depression and psychological distress may serve as useful prognostic signals, with more consistent evidence for cancer mortality than incidence. Causal effects on incidence remain uncertain due to confounding, behavioral mediation, surveillance bias, and reverse causation. Evidence for other chronic diseases—including cardiovascular disease, chronic obstructive pulmonary disease, chronic pain, and multimorbidity—remains insufficient.