The economics of schizophrenia care in Europe: the EPSILON study

Societal Costs of Schizophrenia in Denmark: A Nationwide Matched Controlled Study of Patients and Spouses Before and After Initial Diagnosis

BACKGROUND

Information on welfare cost of patients with schizophrenia and spouses is limited.

AIM

The main aim of this study to investigate factual societal mean annual costs per individual during 5 years before and after the initial diagnosis of schizophrenia.

METHOD

A register-based cohort study of 12 227 patients with incident schizophrenia (International Classification of Diseases, Tenth Revision F20-F20.99) with spouses and 48 907 matched controls in Denmark during 2002-2016.

RESULTS

The total annual costs of health care and lost productivity were Euro 43 561 higher for patients with schizophrenia and health care costs and costs of lost productivity were increased during 5 years before the initial diagnosis. The total annual direct health care and indirect costs of lost productivity were Euro 21 888 higher for spouses to patients with schizophrenia than spouses of individuals with no diagnosis of schizophrenia. Also before initial diagnosis, health care costs and lost productivity were increased among spouses of patients with schizophrenia.

CONCLUSION

Patients with schizophrenia differed from the general population with respect to all included costs. The study documented a significant burden on spouses. The excess health care costs of schizophrenia are further increased by psychiatric and somatic comorbidity, and the societal costs are 4-10 times higher than chronic neurological disorders such as epilepsy and multiple sclerosis. Early onset of schizophrenia implies that patients are affected before finishing school and before entrance to labor market. Cost savings could be achieved by investments in preventive interventions reaching young people's needs; in initiatives to reduce hospital admissions caused by medication side effects, substance misuse, and lifestyle factors; and in occupational training.

Does future resource input reflect need in first-episode psychosis: Examining the association between individual characteristics and 5-year costs

AIM

Coupling data on future resource consumption with baseline characteristics can provide vital information of future consumption patterns for newly diagnosed patients. This study tested whether higher need (as measured by severity of illness) and other baseline characteristics of newly diagnosed patients were associated with higher future service costs.

METHOD

Five hundred forty-seven patients between 18 and 45 years randomized to the OPUS trial was analysed in the study. Multiple regression analysis was applied to estimate the impact of the explanatory variables on mean total costs, which consisted of total health care costs and costs of supportive living facilities.

RESULTS

Lower age, higher level of symptoms (global assessment of functioning), alcohol or cannabis misuse, and being homeless were associated with higher total costs over 5 years, whereas sex, duration of untreated psychosis, and educational level did not show any impact on future resource consumption.

CONCLUSION

The association between future costs and severity of illness suggests that higher needs among patients were associated with higher resource input level. Our results also indicate that other factors than need might affect future costs, for example, parents who serve as advocates for young patients had impact on future health costs. We also found indications of potential barriers among patients with other citizenship in access to health-care services. The strength of the study is that resource data were extracted from official Danish registers and interviewers collected information on clinical characteristics. The results are likely to be context-specific but can be generalized to settings with similar treatment practices.

Adherence therapy for people with schizophrenia. European multicentre randomised controlled trial

BACKGROUND

There is equivocal evidence of the effectiveness of adherence therapy in improving treatment adherence and clinical outcomes for people with schizophrenia.

AIMS

To evaluate the effectiveness of adherence therapy in improving quality of life for people with schizophrenia.

METHOD

A 52-week, single-blind, multicentre randomised controlled trial of the effectiveness of adherence therapy. Participants were individually randomised to receive eight sessions of adherence therapy or health education. Assessments were undertaken at baseline and at 52-week follow-up.

RESULTS

Adherence therapy was no more effective than health education in improving quality of life.

CONCLUSIONS

This effectiveness trial provides evidence for the lack of effect of adherence therapy in people with schizophrenia with recent clinical instability, treated in ordinary clinical settings.

The European Schizophrenia Cohort (EuroSC): a naturalistic prognostic and economic study

BACKGROUND

Schizophrenia has a variety of clinical profiles, disabilities and outcomes requiring responsive management and the devotion of considerable resources. The primary objective of the European Schizophrenia Cohort (EuroSC) is to relate the types of treatment and methods of care to clinical outcome. Secondary objectives include the assessment of treatment needs in relation to outcome, the calculation of resource consumption associated with different methods of care, and the identification of prognostic factors.

METHOD

EuroSC is a naturalistic follow-up of a cohort of people aged 18 to 64 years, suffering from schizophrenia and in contact with secondary psychiatric services. The study was done in nine European centres, in France (N = 288), Germany (N = 618), and Britain (N = 302). Participants were interviewed at 6-monthly intervals for a total of 2 years. This initial paper describes the methods used and presents clinical and social baseline data.

RESULTS

The clinical and socio-demographic differences between patients from the different countries were small. However, patients from Britain were considerably more likely than their continental counterparts to have a history of homelessness, rooflessness or imprisonment, even when social and clinical differences between the samples were controlled.

CONCLUSIONS

The samples were largely similar in clinical terms. Thus, the social differences between the samples seem likely to be due more to the societal context and may reflect relatively benign situations in the continental locations of our study.

The personal impact of schizophrenia in Europe

The personal impact of schizophrenia is poorly described in the scientific literature. The European Psychiatric Services: Inputs Linked to Outcome Domains and Needs (EPSILON) study compared representative treated prevalence cohorts of patients with schizophrenia in five European countries, to assess unmet needs, impact on caregivers, quality of life, satisfaction with services, symptoms and disability. Of the 404 patients, 79% undertook no work of any kind, and 65% were single. Low quality of life was associated with: anxiety, depression, psychotic symptoms, more previous psychiatric admissions, alcohol abuse, having no reliable friends nor daily contact with family, being unemployed, and having few leisure activities. The most frequently occurring unmet needs among the patients were: daytime activities, company and intimate relationships, psychotic symptoms, psychological distress, and information. The most common worries of relatives were about the patient's health, and their own future, safety and financial position. Psychiatric services were therefore largely ineffective in managing the personal impact of schizophrenia, especially upon work, home and family life. Research, clinical practice and disability policy developments need to address a wider range of consequences of the disorder than symptoms alone.

The effect of previous psychiatric history on the cost of care: a comparison of various regression models

OBJECTIVE

Many studies conducted in Europe have reported high variance of costs estimated at the individual level. The objectives of the present study were to estimate costs of patients and to evaluate the performance of various regression models.

METHOD

All patients who in the period 1992-1995 had at least one contact with the psychiatric services in South Verona were included in the study and were followed for 1 year after the index contact (n = 1725). Four different groups, classified on the basis of their previous services utilization, were analysed.

RESULTS

First-ever patients and patients with a new episode of care after 3 years were less costly than patients with an ongoing episode of care and patients having a new episode, after an interval between 3 months and 3 years.

CONCLUSION

The regression models allow us to predict cost for a patient with a given set of characteristics. Great care in the interpretation of the values of the individual coefficients should be taken in order to have a broader understanding of the expenditure dynamic.

Describing, explaining or predicting mental health care costs: a guide to regression models. Methodological review

BACKGROUND

Analysis of the patterns of variation in health care costs and the determinants of these costs (including treatment differences) is an increasingly important aspect of research into the performance of mental health services.

AIMS

To encourage both investigators of the variation in health care costs and the consumers of their investigations to think more critically about the precise aims of these investigations and the choice of statistical methods appropriate to achieve them.

METHOD

We briefly describe examples of regression models that might be of use in the prediction of mental health costs and how one might choose which one to use for a particular research project.

CONCLUSIONS

If the investigators are primarily interested in explanatory mechanisms then they should seriously consider generalised linear models (but with careful attention being paid to the appropriate error distribution). Further insight is likely to be gained through the use of two-part models. For prediction we recommend regression on raw costs using ordinary least-square methods. Whatever method is used, investigators should consider how robust their methods might be to incorrect distributional assumptions (particularly in small samples) and they should not automatically assume that methods such as bootstrapping will allow them to ignore these problems.