OBJECTIVES: In England, nearly two-thirds of older people in acute hospital care suffer from co-morbid physical and mental health conditions. This study investigated the health and social care costs for a group of older (70+) people identified with a mental health condition after hospital admission. METHODS: The Better Mental Health (BMH) study recruited 247 patients at hospital admission in Nottingham, England. Electronic administrative records were sought for six months post-admission from health (services: general practices, hospitals, ambulance transport, intermediate and mental healthcare) and social care. The cohort was characterised by one or more aspects of mental health: cognitive impairment, depression, delirium, and neuropsychiatric health. Differences in mean cost between groups were assessed using t-tests; association between mental health and service-level cost was investigated using GLM regression. RESULTS: Health and social care costs were derived for all 247 participants, except primary care, derived for 122 (subset) participants due to GP recruitment. In the subset, mean (95% CI, median, range) total cost was £9842 (8573-11256, 7717, 715-48795). Mean cost (95% CI) for mental healthcare was significantly (p<0.05) higher for patients: with depression than without (£194 (106-322) Vs. £55 (17-111)); bottom-50% on the neuropsychiatric health scale (£202 (124-298) Vs. £55 (16-118)). Patients with delirium, compared to without, had significantly lower costs for GP consultations (£316 (196-492) Vs. £552 (429-701)) and hospital outpatient visits (£333 (253-444) Vs. £497 (400-621)). The GLM did not identify a significant association between aspects of mental health and service-level costs. CONCLUSIONS: This study suggests a person’s mental health affects consumption of some, but not all, services evaluated. In general, these patients are costly, high resource-users, of health and social care services; however, this consumption pattern cannot be attributed to one particular aspect of mental health. Future work should investigate the impact of physical and mental health comorbidities on resource-use.