Journal of Mental Health Policy and Economics最新文献

Introduction: The authors are health scientist administrators at the National Institute of Mental Health (NIMH). The mission of NIMH is "to transform the understanding and treatment of mental illnesses through basic and clinical research, paving the way for prevention, recovery, and cure." As part of its portfolio, NIMH supports research on mental health economics, and mental health services research.

Method: In this perspective article, the authors comment on two papers presented at the NIMH-sponsored Mental Health Services Research Conference in 2018 and subsequently published in the September 2019 issue of the Journal of Mental Health Policy and Economics. Two important areas are highlighted in this review: (i) the impact of insurance and labor markets on the delivery of high-quality mental health services, and (ii) the need for advancements in method development and design in future studies.

Discussion: The complexity of health insurance markets created some unintended consequence of the mental health insurance parity legislation. Mental health provider shortages in local labor markets are a barrier to successful implementation and sustainment of innovative and evidence-based mental health service-delivery models for people with serious mental illness.

Implications for research: Data-capture techniques that seamlessly integrate insurance claims with clinical outcomes (e.g., from electronic health records) will better equip health economists and other end-users with rigorous research findings to inform public health policy and practice recommendations. Despite early signals of success, larger sample sizes and more rigorous research designs are needed to refine predictive models of functional outcomes of evidence-based service-delivery models (e.g., coordinated specialty care model including supported education, and supported employment) for people with first-episode psychosis.

Background: Prescription of antidepressant drugs (ADs) has increased in recent decades, with rising costs for patients as well as for the health care system. There is sparse evidence of which factors explain the high economic costs and financial burden for the general population.

Aims of the study: The aim was to assess individual-level determinants of out-of-pocket and total health care costs of AD use in the Swedish general population.

Methods: We randomly sampled 400,000 individuals aged 18+ from Statistics Sweden's population register from 2010 to 2013. Two-part regression models were used for our two primary outcome variables: (i) total health care costs for AD use per year and individual, and (ii) total out-of-pocket costs of AD use per year and individual.

Results: Women, the unemployed, unmarried people and residents of big cities have both higher use of ADs and higher associated total health care and out-of-pocket costs. Today, ADs are relatively inexpensive and average cost differences among all groups are therefore minor. The elderly have higher use of ADs, but are more commonly low-volume users and do not have higher total health care or out-of-pocket costs.

Discussion and limitations: Groups with relatively low socioeconomic status are at risk of higher costs for antidepressant use. However, given the Swedish system of drug subsidies, differences in financial burden for individuals are minor. The limitations of this study included that we lacked data on diagnosis and could therefore not categorize the reasons for AD consumption. Furthermore, our results may not be generalized to other countries with a lower AD prevalence then Sweden's, since our estimates are dependent on the point prevalence of antidepressant use in the population.

Implications for health care provision and use: Groups with higher AD consumption and economic costs may suffer from more severe depression owing to more risk factors and less social support in their surroundings, and may be in greater need of additional treatment and support than other groups.

Implications for health policies and further research: Our results offer insight at an aggregate level, and more information on the underlying causes of higher costs is needed to discern the policy implications.

Background: There are many potential pathways in the income health relationship. Problems arise in examining the effect of income loss primarily because of difficulties in disentangling income from the effect of the labour market, but also because of the heterogenous nature of health variables. Psychological, rather than physical, health is important because younger populations are unlikely to manifest clinical evidence of physical disease in the short term. However, biological pathways of stress indicate that this can result in increased mortality and morbidity in the longer term.

Aims of the study: The study follows the example of work that has harnessed the Great Recession to examine income loss, but in contrast to previous work, this study examines the relationship of disposable income and mental health. The study exploits disposable, rather than gross, income, because of economic theory relating disposable income to consumption and research demonstrating the role of consumption in wellbeing. Data from the period of the Great Recession in Ireland allows the examination of changes in disposable income due to government policies that reduced public expenditure and increased taxation.

Methods: Using three waves of panel data from the Growing Up in Ireland study a fixed effects approach is taken to examine disposable income and the mental health of the mothers of young children. A balanced panel is used which results in 6821 individuals being studied over the three waves. The primary dependent variable of interest is depression, scored using a short form of the Centre for Epidemiological Depression Scale (CES-D), although additional outcomes of interest include treatment for a mental health condition, and measures of parental stress using the Parental Stress Scale.

Results: There is a statistically significant relationship between changes in depression score and disposable income loss over the three waves. This relationship is independent of labour market loss during that time. The effect of income loss is predominantly seen for those who are homeowners. Subjective reports of being in mortgage or rent arrears are also associated with an increase in depression score.

Discussion: This group, comprising the mothers of young children, is particularly interesting in view of the credit constraints experienced by younger households during the financial crisis in Ireland. Both sets of results are consistent with qualitative studies which have shown that mortgage difficulties can lead to depression, anxiety and poor mental health, and that high status groups experience shame and self-blame when they experience a financial loss. It remains to be seen if this will have a long-term effect on the mental health of either the mothers or their children.

Background: The Treatment Inventory Cost in Psychiatric patients (TIC-P) instrument is designed to measure societal costs in patients with psychiatric disorders and to be applied in economic evaluations. Efforts have been made to minimize respondents' burden by reducing the number of questions and meanwhile retaining the comprehensiveness of the instrument. Previously, a TIC-P Mini version and a TIC-P Midi version were developed and tested in a predominantly inpatient patient population.

Aims of the study: The aims of this study are to examine the comprehensiveness of the abridged questionnaires in estimating the societal costs for patients with anxiety or depressive disorders and to assess the impact of productivity costs on the total costs.

Methods: The comprehensiveness of the abridged versions of the TIC-P was assessed in four populations: a group of primary care patients with anxiety disorders (n=175) and three groups of patients with major depressive disorders in various outpatient settings (n=140; n=125; and n=79). Comprehensiveness was measured using the proportion of total health care costs and productivity costs covered by the abridged versions compared to the full-length TIC-P. Costs were calculated according to the guidelines for costing studies using the Dutch costing manual.

Results: Our results showed that the TIC-P Mini covered 26%-64% of health care costs and the TIC-P Midi captured 54%-79% of health care costs. Health care costs in these populations were predominantly dispersed over primary care, outpatient hospital care, outpatient specialist care and inpatient hospital care. The TIC-P Midi and TIC-P Mini captured 22% and 0% of primary care costs respectively. In contrast, inpatient hospital care costs and outpatient specialist mental health care costs were almost fully included in the abridged versions. Costs due to lost productivity as measured by the full-length TIC-P were substantial, representing 38% to 92% of total costs.

Discussion: A reduction of the number of items resulted in a substantial loss in the ability to measure health care costs compared to the full-length TIC-P, because these outpatient populations consumed health care from a variety of health care providers. Two limitations of the study need to be stressed. Firstly, the number of patients in each of the four studies was relatively small. However, results were consistent over the four studies despite the small number of patients. Secondly, we did not take costs of medication into account.

Implications for health policies: In developing mental health policy, it is important to include considerations on cost-effectiveness. Increasing the evidence on instruments to measure costs from a societal perspective may support policymakers to adopt a broader perspective.

Implications for further research: The TIC-P

Background: Beginning in late 2010, private health insurance plans were required to allow dependents up to age 26 to remain on a parent's plan. Known as the dependent coverage or young adult mandate, this provision increased coverage substantially within the group of 19-25 year-olds affected by the policy change. Subsequent work evaluating whether increased coverage had a positive effect on mental health found mild improvements in self-reported mental health. This work focused exclusively on average effects among young adults in the years after the policy change, leaving open the question of how young adults fared depending on where they reside in terms of the distribution of risk for mental health issues.

Aims of the study: We assess the effects of the dependent coverage mandate on young adult mental well-being focusing on the distribution of mental health issues. We seek to understand how potential improvements (or degradations) differ across the entire risk profile. Gains among individuals who are at low risk for severe mental health issues may send a far different signal than gains among those with higher risks.

Methods: Using MEPS data from 2006 through 2013, we use quantile regression within a difference-in-differences design to compare pre/post outcomes across the distribution of risk for young adults ages 23-25 affected by the mandate to 27-29 year-olds not affected by the mandate. Further, we evaluate differences in the effect of the mandate by sex, given well-known disparities in incidence and prevalence of mental illness between men and women. To gauge the effects of the mandate on mental health, we use the Mental Component Score measure within the MEPS, ideal for our quantile regression given the broad range of scores. The key premise in our evaluation is that individuals with higher risks for mental health problems due to biological or socioeconomics factors are more likely to rank at locations of the mental health score distribution indicating worse outcomes.

Results: We find significant improvements in self-reported mental health in the 23-25 year-old group following the mandate. However, the gains were not equal across the risk distribution. For individuals at the 0.1 quantile (worse self-reported mental health), the improvement in MCS scores was significant, a 6.1% increase compared to the pre-mandate baseline at that quantile. Effects were smaller but still significant at the median but there was no apparent effect for those that were at higher levels of self-reported mental health. Our results also suggest improvements for women (+9% relative to baseline at the 0.1 quantile, e.g.) but limited evidence of an effect for men.

Implications for future research: The finding that increased insurance coverage led to improved self-reported mental health foremost for young adults with the highest risk of mental health problems