Catastrophic health expenditure on acute coronary events in Asia: a prospective study

Abstract Objective To estimate out-of-pocket costs and the incidence of catastrophic health expenditure in people admitted to hospital with acute coronary syndromes in Asia. Methods Participants were enrolled between June 2011 and May 2012 into this observational study in China, India, Malaysia, Republic of Korea, Singapore, Thailand and Viet Nam. Sites were required to enrol a minimum of 10 consecutive participants who had been hospitalized for an acute coronary syndrome. Catastrophic health expenditure was defined as out-of-pocket costs of initial hospitalization > 30% of annual baseline household income, and it was assessed six weeks after discharge. We assessed associations between health expenditure and age, sex, diagnosis of the index coronary event and health insurance status of the participant, using logistic regression models. Findings Of 12 922 participants, 9370 (73%) had complete data on expenditure. The mean out-of-pocket cost was 3237 United States dollars. Catastrophic health expenditure was reported by 66% (1984/3007) of those without insurance versus 52% (3296/6366) of those with health insurance (P < 0.05). The occurrence of catastrophic expenditure ranged from 80% (1055/1327) in uninsured and 56% (3212/5692) of insured participants in China, to 0% (0/41) in Malaysia. Conclusion Large variation exists across Asia in catastrophic health expenditure resulting from hospitalization for acute coronary syndromes. While insurance offers some protection, substantial numbers of people with health insurance still incur financial catastrophe.


Introduction
Acute coronary syndromes are caused by sudden, reduced blood flow to the heart muscle. These conditions are a major cause of mortality and morbidity in the Asia-Pacific region and account for around half of the global burden from these conditions, i.e. around seven million deaths and 129 million disability-adjusted life years (DALYs) annually from 1990 to 2010. [1][2][3][4][5] Significantly, during this period associated mortality and morbidity accounted for nearly two-thirds of all DALYs and over half of deaths from acute coronary syndromes occurring in low-and middle-income countries. 5 The management of acute coronary syndromes varies widely between countries in Asia. In this area, hospital admission can create significant financial hardships for participants as treatment costs in many settings are borne largely out-of-pocket. 1,2,[6][7][8] In India, for example, survey data indicate that household expenditure on health care is 16.5% higher in households where one or more adults has cardiovascular disease. 9 In China, outof-pocket costs for medications treating high blood pressure -a risk factor for acute coronary syndrome -limit the adherence to the treatment. 10 In addition, out-of-pocket costs for health care act as a barrier to improvements in the treatment of acute coronary syndromes in hospitals. 11 This economic burden on households due to out-ofpocket costs is commonly defined as catastrophic health ex-penditure when out-of-pocket expenditure over a year exceeds a certain threshold. This threshold has been defined in many ways: one of the most common definitions is out-of-pocket costs exceeding 30% of annual household income. [12][13][14][15] There is a lack of evidence on the household economic burden associated with acute coronary syndromes in Asia. Studies that have considered the issue have been small-scale and localized 1 or have studied cardiovascular disease in general 14 rather than specific diseases. 16 In China, India and the United Republic of Tanzania, more than 50% of people admitted with cardiovascular disease experienced catastrophic health expenditure. 14 Among 210 people with acute coronary syndromes in Kerala, India, 84% (176/210) experienced catastrophic health expenditure based on a slightly different threshold of > 40% of disposable income (income minus expenditure on food). 2 While health insurance potentially provides protection from the burden of out-of-pocket costs, the extent of such protection will vary across different health-care systems. Catastrophic health expenditure was reported to be more frequent in uninsured than insured participants with: stroke in China; 13 cardiovascular disease in India; 14 and injury in Viet Nam. 17 There is some contrary evidence that health insurance coverage is associated with catastrophic health expenditure in China and Viet Nam. 7,18 This may be attributed to insurance-based

Costs of acute coronary events in Asia
Stephen Jan et al.
funding making treatment available to groups who may otherwise have not sought care but who, because of limited reimbursement, also incur high levels of out-of-pocket costs. 19 Here we examine the out-of-pocket costs of hospitalization for acute coronary syndromes in China, Hong Kong Special Administrative Region (SAR) of China, India, Malaysia, the Republic of Korea, Singapore, Thailand and Viet Nam. We also assess the incidence of catastrophic health expenditure associated with such hospitalization and the influence of health insurance and other background characteristics on such outcomes.

Setting
The seven countries included in the study have a combined population of around 2.8 billion people (64% of the overall Asian population and 40% of the global population) and represent a mix of income categories and healthcare systems. China, Hong Kong SAR, Republic of Korea and Singapore have high incomes; China, Malaysia and Thailand are upper-middle income countries; while India and Viet Nam are lower-middle income countries. Table 1 provides basic demographic, economic and disease indicators for each of the seven countries included in this study.
Hong Kong SAR of China, Malaysia, the Republic of Korea, Singapore and Thailand have achieved universal health coverage, albeit through a varied combination of financing mechanisms. 22,23 The health services are mainly provided by the public sector and health insurance generally plays a supplementary role in which participants access coverage mainly for private sector services or elective treatments. In China, 19 India 24 and Viet Nam 25 there are known to be gaps in financial protection and heavy reliance on out-of-pocket payments in access to health care.

EPICOR Asia study
The EPICOR Asia study 26 is a prospective observational study of consecutively recruited participants surviving hospitalization for acute coronary syndromes, enrolled in 218 hospitals in seven countries in Asia between June 2011 and May 2012.

Research
Costs of acute coronary events in Asia Stephen Jan et al.
Participants were eligible for inclusion in the study if they were 18 years or older; hospitalized within 48 hours of symptom onset of the index event; with a discharge diagnosis of an acute coronary syndrome; provided written informed consent at discharge; and completed a contact order form agreeing to be contacted for regular follow-up interviews after discharge.
Participants were excluded if their acute coronary event was caused by, or was a complication of, surgery, trauma, gastrointestinal bleeding or post-percutaneous coronary intervention; hospitalization for other reasons; a condition or circumstance arose that in the opinion of the investigator could significantly limit follow-up; they were participating in a randomized interventional clinical trial; or they had concomitant serious/severe comorbidities, which at the discretion of the investigator might have limited short-term life expectancy. 26 Participants were followed-up via telephone interviews at six weeks and three months after the index event, and subsequently every three months until 24 months following hospital discharge. Only baseline and six-week data are reported here as the focus of the study is on the economic burden associated with hospitalization for a relevant acute episode of acute coronary syndromes. Baseline data were collected through interviews with participants on: (i) demography; (ii) index event type (ST elevation myocardial infarction, non-ST elevation myocardial infarction, or unstable angina); and (iii) health insurance status (government, private, employer-provided, other or none). Further details of the data collection have been published previously. 26 The study was conducted in compliance with the principles of the Declaration of Helsinki, International Conference on Harmonisation Good Clinical Practice guidelines and applicable legislation on non-interventional studies in participating countries. The protocol, including the informed consent form, was approved in writing by the applicable ethics committee of the participating centres according to local regulations in each country. The ethics committee also approved any other non-interventional study documents, according to local regulations. A list of participating centres is available from the corresponding author.
Health insurance status was defined as a binary variable based on whether individuals nominated any one of the listed forms of health insurance or none. Treatment costs associated with hospitalization, amount reimbursed and out-of-pocket costs were assessed at the follow-up interviews at six weeks after discharge and converted into United States dollars (US$) based on exchange rates in March 2013. The primary outcome, catastrophic health expenditure, was assessed on the basis of whether a participant had incurred out-of-pocket treatment costs greater than 30% of annual baseline household income. [12][13][14][15] A multivariable logistic regression model was used to assess associations between catastrophic health expenditure and age, sex, the type of index event and health insurance status. The results are presented as odds ratios (OR) and corresponding 95% confidence intervals (CI).
Analyses were undertaken using SAS version 8.2 or later (SAS Institute, Cary, United States of America).

Results
Overall, 9370 out of 12 922 participants (73%) had complete economic data and were included in the analysis. Background characteristics of participants by health insurance status are shown in   (Fig. 2). Catastrophic health expenditure was reported by 56% (5280/9373) of participants; of these, there was a significantly greater proportion in occurrence of this outcome amongst the uninsured (66%; 1984/3007) compared with insured (52%; 3296/6366; P < 0.05). The occurrence of catastrophic expenditure ranged from 80% (1055/1327) in uninsured and 56% (3212/5692) of insured participants in China, to 0% (0/41) in Malaysia (Fig. 3). In short, there was a broad range in the level of out-of-pocket costs and occurrence of catastrophic health expenditures experienced across the seven countries.
In the adjusted analysis, catastrophic health expenditure was significantly less likely for those with health insurance compared to those without insurance; OR: 0.563 (95% CI: 0.51-0.62). Catastrophic health expenditure was significantly more likely among older participants (Table 3).

Discussion
This study shows that the burden of out-of-pocket costs associated with treatment for acute coronary syndromes in Asia can be substantial, reflecting the limited financial protection available for hospitalization for these conditions. It further highlights large variation across countries in rates of catastrophic health expenditure resulting from hospitalization for acute coronary syndromes, and high rates of financial catastrophe incurred, particularly in China and India. With data from over 9000 respon-

Research
Costs of acute coronary events in Asia Stephen Jan et al. dents, we have conducted the largest ever prospective observational study of household economic burden associated with chronic disease. 27 In terms of variation across index diagnoses, the out-of-pocket costs for ST elevation myocardial infarction are, on average, slightly higher than non-ST elevation myocardial infarction and unstable angina. This is related to differences in the complexity of treatments for these events and is reflected in the higher risk of catastrophic health expenditure in participants with ST elevation myocardial infarction relative to those with non-ST elevation myocardial infarction or unstable angina.
Out-of-pocket costs were lower in those with health insurance. Notably, in China, India, Thailand and Viet Nam, participants with health insurance have significantly lower risks of financial ca-tastrophe than those without insurance. In China, although health insurance was protective, more than 50% of insured participants nonetheless incurred catastrophic health expenditure. This is most likely due to a lower threshold for financial catastrophe resulting from a combination of gaps in insurance coverage, high co-payments and low incomes. Such findings are consistent with a previous study in China, in which 53% (1798/3384) of stroke survivors who reported catastrophic health expenditure had health insurance. 13 Current initiatives to roll out social insurance programmes in China -particularly to rural populations through rural cooperative medical schemes -may go some way to addressing this problem. 28 However, the high costs of increasingly prevalent disease events associated with chronic noncommunicable diseases will continue to present challenges to policymakers over coming years.
Given substantial differences in the types of health-care systems involved in this study, health insurance has different roles across the various settings. In health-care systems such as those used in Malaysia, the Republic of Korea, Singapore and Thailand, where there is universal health coverage, health insurance is largely supplementary cover. This may explain why, in the Republic of Korea for instance, there were higher rates of catastrophic health expenditure in insured compared to uninsured participants. Here, participants who are insured may be opting for care through the private system, for reasons such as the avoidance of waiting lists for nonurgent procedures, choice of provider and superior hotel services. They may incur private insurance co-payments and treatment costs not covered under their policies. In contrast, uninsured participants may have been receiving free or heavily subsidized services through the public system.
Financial catastrophe appears to increase with age up to 75 years, whereupon participants experience a lower risk. The greater complexity of illness with increasing age may increase costs associated with an event. The fact that this association is reversed in participants from the highest age groups may be due to higher mortality and an emphasis on more conservative treatment and palliation in this group.
There are three main limitations of the study. First, being clinic-based, the

Costs of acute coronary events in Asia
Stephen Jan et al.
study assessed only the costs incurred by those participants who were able to afford treatment. The study therefore lacks an account of this aspect of financial burden where individuals are excluded from treatment due to cost. Second, only 73% of the participants were able to complete income and cost data. In some countries, participants were recruited from a few centres only. This limitation mirrors well-recognized difficulties in eliciting economic data 29 and limits the generalizability of the findings. Third, the relatively small numbers of participants from countries other than China and India limited the ability to conduct adjusted analyses for individual countries. The study should not be viewed as a series of individual country sub-studies but a multi-country analysis. The fact that there were smaller numbers of participants enrolled in smaller countries does not detract from the generalizability of the findings since financial catastrophe -a standardized outcome indicator -was used.
Ensuring that treatment for acute coronary syndromes is provided and financed affordably in future years will be a major challenge. In particular, this will entail the funding of financial protection programmes that adequately offset the high costs of conditions such as acute coronary syndromes in low-resource settings. While these findings relate to the acute phase of hospitalization, the need to provide affordable, long-term treatment and rehabilitation presents further significant challenges. ■

Research
Costs of acute coronary events in Asia Stephen Jan et al.