Building bridges between health economics research and public policy evaluation

Abstract

The Institut de Recherche et Documentation en Economie de la Santé (IRDES) Workshop on Applied Health Economics and Policy Evaluation aims at disseminating health economic research’s newest findings and enhancing the community’s capacity to address issues that are relevant to public policy. The 2010 program consisted of 16 articles covering a vast range of topics, such as health insurance, social health inequalities and health services research. While most of the articles embedded theoretical material, all had to include empirical material in order to favor more applied and practical discussions and results. The 2010 workshop is to be the first of a series of annual workshops in Paris gathering together researchers on health economics and policy evaluation. The next workshop is to be held at IRDES in June 2011.

Keywords::

• econometrics
• health insurance
• health services research
• health services utilization
• policy evaluation
• social health inequalities

The use of research evidence to produce pertinent and efficient tools for health policymaking has been emphasized by research as well as regulation bodies for more than a decade, and is now a unanimously shared opinion [101,102]. Nevertheless, behind consensus and easy talking, the gaps between research and action remain and are sometimes very wide. When addressing research results, knowledge transfer experts and observers often refer to clinical practice guidelines and health technology assessments [1]. Economic science, in dealing with the production, distribution and consumption of rare resources, clearly has something additional to say about the design, monitoring and evaluation of health policies. No doubt health, health providers and the health system, and public health insurance funding represent, today more than ever, scarce and inequitably distributed resources.

The 2010 Institut de Recherche et Documentation en Economie de la Santé (IRDES) Workshop on Applied Health Economics and Policy Evaluation was designed to disseminate health economic research’s newest findings and to enhance the community’s capacity to address issues that are relevant to public policy. The IRDES workshop aimed at stimulating scientific interaction on new topics, methods and issues in health economics, with an emphasis on how results can help public decisions or evaluate public policies. The program consisted of 16 articles covering a vast range of topics, such as health insurance, social health inequalities and health services research. While most of the articles embedded theoretical material, all had to include empirical material in order to favor more applied and practical discussions and results. The 2010 workshop is to be the first of a series of annual workshops in Paris gathering together researchers on health economics and policy evaluation.

In this article, we report the main issues presented and discussed during the workshop, grouped into three main themes: health insurance, social health inequalities and the supply of health services.

Research on health insurance issues

Kesternich examined the impact of the introduction of Medicare Part D in the USA [103]. She analyzed data from a hypothetical choice experiment conducted with a sample of the eligible population and developed a joint model for consumers’ revealed and stated choices. Results suggest an underestimation of welfare when relying on estimates based on the current users and not on the whole potential market.

Buckley, Cameron, Hurley, McLeod and Mestelman presented a new approach – incentivized revealed choice experiments – to test impacts of alternative healthcare financing arrangements on health system outcomes [104]. Results were broadly consistent with theoretical predictions, especially regarding the impact of changing the three experimental parameters of concerns: the public-sector allocation rule, the supply of the healthcare resource and the size of the public budget. They demonstrated that subjects are systematically willing to pay more than predicted for private insurance.

Debrand and Sorasith used a microsimulation model in an attempt to measure the impact of introducing out-of-pocket (OOP) maximum rules [105]. The authors estimated the redistributive effects of fixed- versus income-based OOP maximum rules. They showed that when reducing the risk of being faced with a very high OOP burden, uniform thresholds level out the heterogeneity of situations and suit individuals with high-risk aversion. Inversely, income-related thresholds increase the heterogeneity of OOP burden but have a less-regressive redistributive effect.

Van Ophem and Berkhout analyzed the decision to choose a deductible in the case of health insurance in The Netherlands [106]. They constructed a simultaneous model, specifying both the choice for a deductible and the number of doctor visits, and they estimated it using Dutch data from 2007. Results indicated that the choice for a deductible neither depends on the health status of individuals nor the expected demand for healthcare.

Grignon and Kambia-Chopin examined the demand for complementary health insurance (CHI) among the nongroup contracts market in France [107]. They analyzed the reasons why a policy subsidizing the purchase of CHI for individuals just above the poverty line did not produce any significant effect. They developed a theoretical model, and using their estimations of the relationship between income and spending on CHI, they simulated the effect of various levels of price subsidies on the demand for CHI among those with incomes around the poverty level. Their simulations indicated that no policy of price subsidy can significantly increase the uptake of CHI among those who are just above the poverty line, since any increase in the level of subsidy generates a windfall benefit for richer households.

Cheng and Vahid examined the determinants of the length of hospital stay, the decision to seek hospital care and the decision to purchase private hospital insurance using data from Australia [108]. They used a simultaneous equation model that accommodates the nature of the count data of the length of hospital stay and binary outcomes in patient characteristics and insurance decisions. They did not find any significant moral hazardous effects of private hospital insurance on the length of private hospital care.

Results from the inequality sessions

Keese and Schmitz studied the impact of household overindebtedness and debt-related financial distress on obesity [109]. Controlling for reverse causality, they showed that household debts cause deterioration of physical and mental health, but they found no causal effect on being obese.

De Chaisemartin and Geoffard evaluated the impact of the 2007 French workplace smoking ban Act using original data on patients who consulted tobacco cessation services [110]. Although the ban had no measurable effect on overall prevalence in the general population, the ban caused an increase in the demand for tobacco cessation services, and increased the number of successful attempts to quit smoking.

Erreygers, Clarke and Van Ourti addressed new issues in social health inequality measurement through an exploration of the extended concentration index and its properties and implications for the measurement of inequalities in health [111]. The authors proposed and examined alternative indicators. They incorporated attitudes towards inequality into the calculation of the index of socioeconomic inequality in health, and showed how it may eventually affect the ranking of countries in terms of inequality.

Dourgnon and Afrite addressed socioeconomic discrimination in overweight management by general practitioners [112]. Based on an exploratory French survey, they showed that physicians do categorize a patient’s compliance and demands for health information according to a patient’s socioeconomic status. Doctors who categorize patients tend to provide more advice and recommendations to all patients. Whether these categorizations correlate with actual differences in patients’ treatments remains unclear.

Lordan, Brown, Jimenez Soto and Greene studied how the distribution of household health resources can be influenced by bargaining within the household, with a particular focus on gender biases across children in four countries: Armenia, Azerbaijan, Kyrgyzstan and Tajikistan [113]. Results indicated a pro-boy bias with respect to resource allocation across children in all countries. In addition, with the exception of Kyrgyzstan, the pro-boy bias appeared greater in male-led households. Since a blanket increase in healthcare access may simply aid the advantaged group – in this case, boys – the authors recommended that schemes to offset this bias should provide free access to girls only, with particular attention being paid to the 4–16-year-old female age group, where these biases are certainly larger in magnitude.

Berchet and Jusot evaluated the contribution of social capital to health inequalities between the native and immigrant populations in France [114]. The study used a decomposition method in order to identify the main factors, explaining how much social capital and socioeconomic status contributed in accounting for social health inequalities between migrants and non-migrants. Results emphasized the essential role of social capital in widening the health gap between migrants and non-migrants. They suggested that public policies should focus on the improvement of social capital and tackling social isolation in order to reduce health inequalities related to migration.

Results from the health services sessions

Kruse, Pradhan and Sparrow examined the marginal effects of decentralized public health spending by incorporating estimates of behavioral responses to changes in public health spending through benefit incidence analysis [115]. They analyzed the situation in 207 Indonesian districts from 2001 to 2004. They found no evidence that public expenditures crowded out the utilization of private services or household health spending. Their analysis suggested that increased public health spending improves access for the poor through behavioral changes in public healthcare utilization.

Lezzi, Lippi Bruni and Ugolini studied the effects of incentive schemes on quality of care [116]. They investigated the impact of financial incentives in Regional and Local Health Authority contracts for primary care in the Italian region Emilia Romagna for the years 2002–2005. Their results supported the hypothesis that financial transfers may contribute to the improvement of the quality of care, even when they are not based on performance assessments.

Yilmaz and Raynaud attempted to measure the influence of social deprivation on the length of hospitalization [117]. They used a French sample based on 27 hospitals, including public and private hospitals (for-profit and not-for-profit), representing 57,175 stays, 6800 of which concerned patients with social deprivation. After adjusting for age and severity of illness, the authors found that there was a longer length of stay for patients with social deprivation, and, in particular, for patients living in social isolation and for patients with inadequate housing.

Dormont and Milcent compared the productivity of public, private for-profit and private not-for-profit hospitals in France [118]. They evaluated the respective impacts of differences in efficiency, of patient characteristics and production characteristics, based on a French database. Their results showed that medium- and large-sized public hospitals did not display a higher level of inefficiency with respect to private hospitals, once production and patients’ characteristics were taken into account. Moreover, smaller public hospitals appeared relatively more inefficient than private ones.

IRDES Committee members

The following individuals are members of the IRDES Committee: Thierry Debrand (IRDES, Paris, France), Paul Dourgnon (IRDES), Brigitte Dormont (Paris Dauphine University, France), Alberto Holly (HEC Lausanne, Switzerland), Jeremiah Hurley (McMaster University, Canada) and Lise Rochaix (HAS, St Denis, France).

Financial disclosure

The workshop has benefited from a grant from the Risk Foundation (a French foundation funded by private insurance providers)[119]and has obtained the financial support of IRDES. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.