Out-of-pocket health expenditure and fairness in utilization of health care facilities in Cambodia in 2005 and 2010

Introduction

Globally, the main goal of a health care system is to have appropriate funding mechanisms for individuals to acquire care for preventive and curative health needs without deepening into poverty. There is a general consensus that ease of access to health care facilities has the potential to improve the health and wellbeing of individuals and increase the life expectancy of the entire population^1,2. However, the pluralistic nature and differences in financing mechanisms of most health systems around the world means varied methods will be adopted by policy makers for health care provision. Revenue collection through general taxation and compulsory social health insurance, as well as individual private health insurance schemes, are among the most commonly used methods for funding health care provisions in high income nations³. Sadly, in spite of the successes of these funding strategies in high income nations, inadequate planning and lack of political will has given rise to non-existence of universal financing mechanisms in low-and-middle-income countries (LMICs). In these countries, many individuals continue to rely on out-of-pocket (OOP) payments each time they access care, and as a result, are not shielded from economic hardship due to huge health care expenditures. Catastrophic health expenditure through OOP payments is a global phenomenon and can exist regardless of health care funding mechanisms. However, OOP payments constitute a significant component of health spending in most LMICs. According to a study investigating global estimates of catastrophic spending and OOP payments⁴, approximately 150 million people suffer financial burden each year due to health care payments, and about 100 million are pushed into poverty because of OOP payments.

Paying for health care through OOP may prevent patients from seeking medical care when needed due to economic constraints. This is particularly common for people with lower incomes⁵. Low income groups sometimes risk debt due to paying for health care⁶. In Cambodia, as is the case with most countries in South East Asia^6–9, excessive reliance on OOP payment for care is pervasive due to lack of universal health coverage¹⁰. Cambodia is a country with a turbulent history, enduring genocide and societal collapse, but is now rapidly evolving and has managed to decrease poverty significantly¹¹. The repositioning of the Cambodian health system is aimed at preventing and controlling communicable and non-communicable diseases^12,13. Although the health care system has also experienced progress, it still is in need of further development, as the majority of health expenditure comes from OOP payments and this usually benefits unregulated private health care¹⁴. There was a health care system reform in 1996, and the National Charter on Health Financing was introduced allowing user fees in health care facilities¹⁵. In conjunction with this charter on health financing, a health equity fund was introduced in 2009 to aid access to health care for the poor. However, the health equity fund has not been implemented on a national scale, but rather has been used in different ways in different districts^16,17. Examining the impact of catastrophic health care spending on household impoverishment during the introduction of a new funding mechanism, such as health equity fund, over a period of time is of policy relevance. While several studies have been conducted to assess the impact of catastrophic expenditure on household impoverishment over a period of time in other countries in South East Asia^6–10, no such study have been done in Cambodia.

The present study was conducted to bridge this research gap and examine, for the first time, trends in catastrophic OOP payment burden and equity in use of health care facilities among a nationally representative sample of Cambodian adults aged 15–49 across 611 communities. Specifically, this study aimed to: first, examine fairness in the distribution of cost burden among wealth strata and compare trends between 2005 and 2010; and second, characterise sources of OOP payment for treatment and compare the trends between 2005 and 2010. Understanding such comparisons would aid in further planning and future implementation of various health benefit packages aimed at alleviating high OOP payments for healthcare use in Cambodia.

Methods

Results

A total of 5989 participants participated in the 2010 study. Overall, 92% sought health services for a first-time treatment in 2010 compared to 90% in 2005. For services sought for second treatments, there was a decrease from 29% in 2005 to 24% in 2010. Similar results were shown for third treatments; 40% in 2005 to 33% in 2010.

Utilization of public and private health care facilities

Overall, there was an increase in utilization of both public and private health care facilities between 2005 and 2010, as can be seen in Figure 1. There was an increase in usage of private facilities as the preferred place for service utilization at 1st instance i.e. from 38.6% in 2005 to 61% in 2010. There was also an increase, albeit a small one, in use of public facilities, 26.5% in 2005 to 32.6% in 2010. For utilization of health care at the 2^nd instance, similar results can be found; increase in usage of private health care from 37.7% in 2005 to 63.7% in 2010 and increase in usage of public health care from 26.7% in 2005 to 29.4 in 2010. This is also the same for the 3^rd instance; an increase in utilization of private health care facilities from 37% in 2005 to 64.8% in 2010. There was been a decrease in usage of informal health care facilities (such as shop/market, magician, religious leader, all characterised as others), for all three different instances from 2005 to 2010. These results who that, generally, a higher percentage of Cambodians have a strong preference for private health care facilities.

Figure 1. Utilization of public and private health care facilities in different instances.

Households most likely to engage in catastrophic spending

Mode of out-of-pocket payment. Figure 2 displays a breakdown of percentages of different modes of OOP. Briefly, the mode of OOP was divided into the following categories: wages/pocket money, savings, sale of assets, interest loan, non-interest loan, and other sources, e.g. such as help from an NGO. In 2005, 44% of all OOP payments came from wages/pocket money, followed by 33% from savings. In 2010, 31% of OOP payments were from wages/out of pocket money and 41% from savings. Payments from loans generally remained the same between 2005 and 2010, around 5-7%. Sale of assets as means to pay for OOP payments for health care increased from 3% in 2005 to 6% in 2010. Gifts from relatives and friends (5% in 2005 and 6% in 2010) were also OOP sources. The health equity fund introduced in 2009 seems to have effect on payments and accounted for 2% of the OOP payments in 2010.

Figure 2. Mode of out-of-pocket payments for health care utilizations: comparison between 2005 and 2010.

Transport cost, treatment cost and total cost of treatment. Degree of fairness in terms of spending on transport, treatment and total cost of treatment was examined using Lorenz curves (Figures 3a–c). As depicted in Figure 3a, for all categories of treatment instances, the transportation cost is the most equal across nearly all the economic strata, with individuals in the poorest quintiles of the economic strata being more able to cope with spending. Compared to 2005, in 2010 individuals in the 3^rd, 4^th and 5^th strata are in a relatively good condition to bear the cost of transportation. Regarding the treatment cost, Figure 3b indicates that inequality was more pronounced in 2010 compared to 2005. For instance, people at the 4^th & 5^th quintile of the economic strata are in relatively good position to afford the cost of treatment than in 2005 for all the instances of treatment. Figure 3c shows that for all instances of treatments, the total cost is most unequal in 2005 compared to 2010. For instance, those at the 4^th & 5^th quintiles of the economic strata are more able to bear total treatment costs relative to those at the lower strata.

Figure 3.

Lorenz curves for Transport cost (3a), Treatment cost (3b) and Total cost of treatment (3c): comparison between 2005 and 2010.

Discussion

This study provides the first detailed trend analysis of catastrophic OOP health expenditure and fairness of healthcare facilities utilization in Cambodia in 2005 and 2010. The findings show that the trend of catastrophic spending and fairness in utilization of health care facilities is not improving. The finding that more people from the poorer households are seeking treatment in all three instances compared to richer households over the two periods (2005 and 2010) is not new. The reason for this may be because individuals from lower strata of the economic index tend to patronize hawkers (here, a kind of quack without any medical degree selling medicine), as these providers are known to charge nominal fees^24,25. Care seeking from hawkers and unqualified health care professionals has been shown to be a precursor for treatment failures and switching of care providers at private and public health facilities^24,26,27. The present finding supports a report from elsewhere in South East Asia, i.e. that poorer individuals have large healthcare burdens²⁸.

The present results displayed a huge increase in utilization of private facilities compared to public facilities from 2005 to 2010. Hence, one may argue that costs are an important determinant of choice of place of treatments. People would use more public facilities to avoid healthcare burden at private facilities. In this study, however, the trend in use of health care facilities has been shown to tilt consistently towards the use of private facilities for all instances of service use for the two time periods. This finding is not new, but is consistent with what has been reported before in other LMICs, and South East Asia in particular^4,28. Countries in South East Asia have witnessed an increase in the proliferation of private health care facilities in the last decade. One probable reason for the present observed finding may be one of the unique features of poor health systems, where private and public care facilities always substitute each other, mostly in LMICs²⁹. In addition to this, inefficiency in service delivery in public health systems has been shown to be another contributory factor. For example, long waiting times, nonchalant attitudes of health care workers and perennial lack of essential medicines, coupled with informal fees charged by medical personnel to survive poor wages^30–32.

Financing health care through formal and informal OOP payments have proven to affect people’s behavior in seeking care, especially those from poor households^27,33. In this analysis, the pattern of OOP has been shown to be consistent and comprised of loans both with and without interests, wages/pocket money and savings in both 2005 and 2010. This finding supports what has been reported in several studies conducted across different countries^34–36. The present analysis documents a wide variation in the mode of payment for health care between the two periods in relation to use of personal savings and wages/pocket money. Evidence has consistently shown that an individual from a poorer household faces a huge economic burden from health care costs than their richer counterpart³⁶. People saved more in 2010 and relied less on their wages and pocket money to offset payments for their treatment compared to 2005. Another explanation for this observation could be related to people’s awareness of the importance of setting aside a smaller portion of their income for health care emergencies to avoid selling and borrowing.

Health equity funds, exemption schemes and other subsidizations have been introduced to address inequity of health care utilization, which target the poor, in several LMICs. It can be difficult to identify and target the intended people for these exemption schemes and other subsidizations; however, studies in Cambodia found that the health equity fund to be rather successful in this country¹⁷. In contrast, in the present study, the health equity fund only accounted for 2% of the OOP payments being made. Possession of health insurance has been shown to hold promise for financing individual health care costs in LMICs³⁷. The use of health insurance as a mode of payment for health service utilization was reported in 2010, but not in 2005. This finding is in line with the results from other studies^38,39.

Economic hardship experienced by people from poorer households when accessing health care is immeasurable. As in other studies, the present analysis further documents the existence of inequity and unfairness in cost burden between the poor and the rich in Cambodia when accessing care. In summary, the distribution of cost burden and catastrophic spending among Cambodian adults was more inequitable over the two time periods. However, there is a growing understanding of the need to bridge poor-rich inequalities in access to health care through adoption of several coping mechanisms among the entire populace.

A health system should, according to the WHO, improve the health of the population they serve, respond to people’s expectations and provide financial protection against the costs of ill-health². With the present reported result, Cambodian health systems, which heavily rely on OOP payments, do not meet the last objective. Furthermore, according to the Strategic Plan 2008–2015 from the Ministry of Health, one of five working principles are “social health protection, especially for the poor and vulnerable groups”¹⁵. The findings from this analysis echoes the need to ensure that those from poorer households are protected against hardship spending when seeking care. As such, more work is needed to guarantee access to sufficient health care for the poor and vulnerable in Cambodia. One of the ways to mitigate financial hardships constantly being faced by those from poorer households and reduce poverty drift through catastrophic health care spending is to adopt point-of-care health financing mechanisms targeted at those in need⁴⁰.

Data availability

The DHS Program owns data used in this study. The DHS for Cambodia 2005 and 2010 are available for researchers interested in further analyses. Researchers should contact the DHS Program and to get permission to use the required data (https://dhsprogram.com/data/Access-Instructions.cfm).