Cost analysis and policy implications in psychiatric care

Abstract

The economic burden of mental health constitutes a substantial part of the total costs of illnesses. Most estimates of cost-of-illness focus on somatic illnesses and corresponding studies of mental health are few. In Sweden there has been one Cost of Illness (CoI) study for psychiatric disorders which was carried out in the early 1980s and two studies that estimate the economic burden of all diseases. The relationships between direct and indirect costs and their respective development over time are also less studied. The aims of the present thesis were to analyze costs and regional cost differences due to psychiatric disorders: the economic burden of psychiatric disorders to society, the analysis of regional differences of direct costs and productions losses (due to sickness absence and disability pension) and the potential determinants of these differences, and the analysis of trends and costs of hospital admissions for bipolar disorders.

The total economic cost of psychiatric disorders in Sweden in 2001 (Study I) was estimated at MSEK 85 774 (M€ 9 426, €1= SEK9.1). Indirect costs were totally predominant and constituted about 80% of the total cost, i.e. production losses due to early retirement, sickness absence and premature death. In Study II, the total direct costs were found to be influenced by sociodemographic factors especially the proportions of people aged 65 years and the proportion of women. Moreover, population density (rural/urban) influences the differences in direct costs. There are differences between county councils in how mental healthcare resources are organized which may partly account for differences, developmental trends and allocations in direct costs. In Study III, total production losses due to psychiatric disorders were estimated based on sickness absence and disability pension. Both types of production losses were significantly predicted by four determinants among which three (the proportion of people born abroad, living alone and having a low income, respectively) had a positive and one (the amount of psychiatric care provision) a negative influence on production losses. Psychiatric care provision seems to be a determinant for the variation in production losses due to sickness absence, disability pension and the sum of these; i.e. the lower the number of employees the higher the production losses. In Study IV, more than half of the patients with their first bipolar disorder admission in the year 2000 had no bipolar readmission during the five-years of follow-up. On the other hand, 15% of the first diagnosed bipolar patients together had 66% of the admissions, with a yearly hospitalization cost of MSEK 230 (M€ 25).

In summary, the indirect cost of psychiatric disorders seems to have increased relative to the direct costs and constitutes the greater part of the societal costs for psychiatry. In regions where the provision of psychiatric care was greater, the indirect costs were smaller, thus indicating that psychiatric health care provision is central for diminishing the social and work disabling consequences of psychiatric illnesses. Furthermore, women and elderly patients had a decreasing effect on direct costs despite an age- and gender-related increase in the incidence of mental health problems, especially depression. More research and better data is needed to provide a better insight into the utilization of healthcare resources in psychiatric diagnosis groups and differences in regions where health economic aspects should be included.