Cost of Schizophrenia to UK Society: An Incidence-Based Cost-of-Illness Model for the First 5 Years Following Diagnosis

Objective: This study estimated the cost to UK society of an annual cohort of newly diagnosed patients with schizophrenia over the first 5 years following diagnosis, using an incidence-based cost-of-illness framework. Design and setting: A discrete event model of the course of schizophrenia was constructed, based on a literature review and interviews among a panel of healthcare professionals (n = 7). Seven discrete disease states were defined within the model. Patients' movements between these disease states enabled 10 disease courses to be identified. In each disease state, the model estimated resource use and corresponding costs borne by the National Health Service (NHS), Local Authorities, the Home Office and society as a result of lost productivity. Patients and participants: The model simulated patients' movements between disease states over the first 5 years following diagnosis. Since there are 7500 new cases of schizophrenia per year in the UK, the model was run for 7500 patient simulations. Main outcome measures and results: The total discounted cost to society attributable to an annual cohort of newly-diagnosed patients with schizophrenia over the first 5 years following diagnosis was estimated at Lstg 862 million (range: Lstg 788 million to Lstg 926 million in sensitivity analysis). The discounted mean 5-year cost was estimated to be approximately Lstg 115 000 (range: Lstg 105 000 to Lstg 124 000) per patient or approximately Lstg 23 000 (range: Lstg 21 000 to Lstg 25 000) per patient per year. The NHS accounted for 38% of the total cost, Local Authorities for 12% and the Home Office for 1%. Indirect costs due to lost productivity accounted for 49%. Of the NHS costs, hospital admissions accounted for 69% and hospital visits (outpatient, day ward and day centre attendances) for a further 26%. Drugs (antipsychotics and adjunctive medications) accounted for 2%. Conclusions: NHS expenditure and lost productivity costs predominated, irrespective of disease course. This indicates that treatments that reduce hospitalisation and potentially enable patients to return to active employment could significantly reduce the societal burden of schizophrenia.