Objectives Economic evaluations of interventions for people with mental–physical multimorbidity, including a depressive disorder, are sparse. This study examines whether such interventions in adults are cost-effective.

Design A systematic review.

Data sources MEDLINE, CINAHL Plus, PsycINFO, Cochrane CENTRAL, Scopus, Web of Science and NHS EED databases were searched until 5 March 2022.

Eligibility criteria We included studies involving people aged ≥18 with two or more chronic conditions (one being a depressive disorder). Economic evaluation studies that compared costs and outcomes of interventions were included, and those that assessed only costs or effects were excluded.

Data extraction and synthesis Two authors independently assessed risk of bias in included studies using recommended checklists. A narrative analysis of the characteristics and results by type of intervention and levels of healthcare provision was conducted.

Results A total of 19 studies, all undertaken in high-income countries, met inclusion criteria. Four intervention types were reported: collaborative care, self-management, telephone-based and antidepressant treatment. Most (14 of 19) interventions were implemented at the organisational level and were potentially cost-effective, particularly, the collaborative care for people with depressive disorder and diabetes, comorbid major depression and cancer and depression and multiple long-term conditions. Cost-effectiveness ranged from £206 per quality-adjusted life year (QALY) for collaborative care programmes for older adults with diabetes and depression at primary care clinics (USA) to £79 723 per QALY for combining collaborative care with improved opportunistic screening for adults with depressive disorder and diabetes (England). Conclusions on cost-effectiveness were constrained by methodological aspects of the included studies: choice of perspectives, time horizon and costing methods.

Conclusions Economic evaluations of interventions to manage multimorbidity with a depressive disorder are non-existent in low-income and middle-income countries. The design and reporting of future economic evaluations must improve to provide robust conclusions.

PROSPERO registration number CRD42022302036.