Out-of-pocket health expenditure differences in Chile: Insurance performance or selection?

Highlights

• Out-of-pocket health expenditure is high in Chile, reflecting a poor financial protection of the health insurance system.

• Segmentation creates two populations and gaps in terms of their health expenditure.

• The structure of out-of-pocket expenditure has remained constant during the past years, with expenditure in drugs representing 30% of the total expenditure.

• The private insurance scheme is regressive, offering better financial protection to the richest households in the country.

• Most of the gap in expenditure in the health insurance system is explained by the selection process present in the private scheme.

Abstract

Chile has a mixed health system with public and private actors engaged in provision and insurance. This dual system generates important differences in health expenditure between private and public insurances. Selection is a preeminent feature of the Chilean insurance system. In order to explain the role of the insurance in out-of-pocket expenditures between households for different insurance schemes, decomposition methods are applied to disentangle the effect of household ‘composition and insurance’ degree of financial protection on health expenditures. Health expenditure patterns have not changed in the last 10 years with drugs, outpatient care, and dental health representing 60% of the health expenditure. Health expenditure/income is similar for different income groups in the public insurance, but decreases with income in households with private coverage, reflecting regressivity in health expenditure. On the other hand, health expenditure as share of expenditure increases with income for both groups.

Per capita health expenditure in households with private coverage is four times the expenditure of households with public insurance; this gap is mostly explained by differences in households’ expenditure and demographics. Roughly 80% of the difference in expenditure is explained by the model, showing the role of selection in understanding the expenditure gap between insurance schemes.

Introduction

Since the 1980s Chile has a mixed health system with participation of public and private actors both in provision and insurance. Currently 75% of the population is covered by the public health insurer, the National Health Fund (FONASA, Fondo Nacional de Salud), while 18% is covered by private insurers, ISAPREs (Instituciones de Salud Previsional). The remaining 7% is mostly insured in alternative systems, such as the health insurance for the Armed Forces (CAPREDENA, Caja de Previsión de la Defensa Nacional) or the scheme for victims of human rights violations (PRAIS, Programa de Reparación en Atención Integral en Salud) [1].

Both schemes of insurance differ in several aspects, but two are particularly relevant in terms of their implication on the health system performance and impact on the Chilean citizens: a high level of out-of-pocket (OOP) expenditure and segmentation of private and public insurance schemes – with different rules and pools – resulting in poor financial protection [2].

First, using the World Health Organization’s framework for analyzing universal coverage [3], health insurance coverage is large in terms of population covered (breadth) – 98% of the population is insured [4] –, but small in terms of percentage of services and cost coverage (depth and height): OOP expenditure as share of total health expenditure is 33% – one of the highest among OECD countries – far from the OECD average of 20% [5].

Second, contribution to health is mandatory. Every employee (contribution will be mandatory for independent workers starting in 2018) must pay 7% of her salary to a health insurer (either public or private); the main difference between both schemes is that ISAPREs are allowed to charge premiums over the 7%, offering individual health plans to their affiliates. On the other hand, FONASA’s premium is equal to the 7% contribution, offering a benefit package with the same services but different coverage, according to four income groups: FONASA A offers coverage to people classified as indigent, with no obligation to contribute and whose healthcare – provided within the public network – is fully subsidized by the government (no copayment); the rest of the groups contribute with the mandatory 7%, but receive differentiated subsidies related to their monthly income: FONASA B have a 100% subsidy, while coverage is 90% for group C and 80% for group D. Finally, people in groups B, C and D can seek healthcare services with a private provider; in this case, they receive a voucher that partially covers its cost, according to type of service and type of provider (for more details, see [6] and [7]). In 2014, 25.5% of FONASA affiliates were classified in group A, 35.2% in FONASA B, 17.1% in FONASA C, and 23.1% of the affiliates were in group D [1]. These features explain the pooling in both schemes and the segmentation in the Chilean insurance system, where FONASA ends up covering the riskier (in terms of age and gender) and poorer, while ISAPRE offer insurance to those with less risk and more income (Fig. 1, Fig. 2).

The Chilean dual system generates important differences in health expenditure for people covered by private and public insurances, mainly due to legal differences that allow selection in the private market. The aim of the paper is to explore and explain these differences in order to understand the role of both insurance schemes in providing financial protection to the population, as well the challenges of balancing universal coverage, public-private participation and inequalities in health.

The document is organized as follows. Section 2 describes the methods and data used to analyze the Chilean OOP expenditures. Section 3 shows the results of the analysis, describing health financial statistics, discussing how they have changed over time, and explaining differences in expenditure patterns between populations covered by public and private insurance. Finally, Section 4 presents the conclusions of the study and discusses future implications.