Background
Ensuring financial protection for the population against the cost of ill-health is one of the fundamental objectives of the health system [
1]. It has been estimated that approximately 100 million people worldwide are forced into poverty, and around 150 million people face catastrophic expenditure as a consequence of high out-of-pocket (OOP) payments for health care each year [
2]. Additionally, a large number of people abstain from utilizing health care resources due to the financial costs.
The literature has shown that a higher proportion of OOP payments of health expenditure is related to a higher percentage of households that suffer from catastrophic health expenditures and are forced into poverty [
3,
4]. The World Health Report 2010 proposed a strategy to improve or modify country-specific health financing systems so as to achieve universal health coverage (UHC), which aims to ensure that everyone has access to comprehensive and health services of acceptable quality without incurring financial hardship [
2]. Furthermore, UHC has become one of the overarching health targets of the Sustainable Development Goals by 2030, approved by the UN General Assembly in September 2015 [
5].
As in many other countries, health financing reforms in Mongolia are guided by the concept of UHC. Before 1990, Mongolia had a centralised health system where the government was wholly responsible for both health service delivery and financing. It provided access to universal “free” health care (i.e. no patient cost-sharing) [
6]. However, during the socio-economic transition from a centrally planned to a market-oriented economy that started in 1990, health financing reforms aimed to expand funding sources beyond the government budget [
7]. This policy change encouraged household contribution to health financing through user fee and co-payment initiatives. At the same time, reforming the health care financing system with user fees/co-payments needed a strategy and evidence-based actions where available, so as to preclude people from incurring financial difficulties due to the introduction of co-payments.
In 1994, the government of Mongolia successfully introduced a new social health insurance system (SHI) [
8,
9] . The main purpose of SHI introduction and implementation was not only to promote equitable access to health care through prepayment, i.e. reducing negative effects of user fee policies, but also to provide financial protection for the population from excessive financial hardships, especially for low income and vulnerable population groups [
7,
9]. This policy focus is still valid, and remains as an issue of priority in Mongolia’s medium- and long-term health and SHI development policies [
6,
9‐
12].
Currently, SHI is one of the main sources of health financing along with the government health budget. As of 2014, almost the entire population is covered by SHI and entitled to the same health service benefits regardless of their socio-economic characteristics [
9].
According to the health insurance law, the insurance premium for employees in public and private sectors is 4 % of their monthly salaries, which is shared equally between the employer and employee The premium for children under 18 years and students is equal to 1 % of the national minimum wage per month. For all other groups, the premium rate is 2 % of the national minimum wage. It is worth mentioning that the premiums for some vulnerable and specific groups, including children younger than age 18, pensioners, mothers caring for new-born children up to the age of two, military personnel and people on low incomes are fully subsidized by the government [
12].
The benefit package offered by SHI includes major inpatient services at both secondary and tertiary level hospitals with a patient cost-sharing at around 10 to 15 % and a limited number of outpatient services [
8,
13]. In addition, SHI covers 50 to 100 % of the cost of essential medicines prescribed by family physicians and medical doctors working at the primary level of health centres [
8]. The Government health budget funds primary health care services provided through family health centres and
soum (district) health centres; and some specific services including treatment of chronic and infectious diseases provided at secondary and tertiary level hospitals [
8,
13].
In the past, several efforts have been made to assess the effect of direct out-of-pocket payments on household income and expenditure. The World Bank report, based on the Mongolian Household Socio-Economic Survey (HSES) 2007/2008, stated that the incidence of catastrophic health expenditure was 3.3 %, when the threshold was set at health expenditures at 40 % or more of non-food expenditure. It also reported that the rate of impoverishment due to out-of-pocket payments was 2.5 % [
14]. These results indicated that the incidence of catastrophic health expenditure and impoverishment is smaller in Mongolia compared to some other developing countries [
14].
Other studies found that the poverty rate in Mongolia has decreased from 35.2 to 27.4 % (between 2008 and 2012) [
15,
16]. In the same period, SHI coverage increased from 83.2 to 98.6 % [
17,
18]. Despite these positive trends, our previous studies showed that the degree of income-related inequalities in health care utilization increased between 2008 and 2012 [
19]. In addition, the poor tended to use primary health care rather than secondary or tertiary hospital care, although they had a lower health status and greater needs [
19], i.e. high-income groups reported significantly better health [
20]. In contrast, high-income groups were more likely to bypass primary health care and directly choose more costly health services at the higher levels of hospitals or in the private sector [
19]. A referral system has been built into the health sector; nonetheless, gatekeeping at primary health care is weak and unnecessary self-referral to the upper level hospitals is common [
18].
Another emerging fact is that the share of OOP payments in the total health expenditure reached 41 % in 2011 [
9]. As the international evidence indicates, when OOP payments exceed 20 % of total health expenditure, it is difficult to reach UHC, and a country may need to improve financial protection policies [
3,
21].
Based on these grounds, this paper aims to contribute to policy discussions by estimating the incidence of catastrophic health expenditure and the rate of impoverishment with the latest available household income and expenditure survey data, and to analyse the overall trends and effects of SHI on financial protection in Mongolia. The paper also intends to promote uniform measurement and regular monitoring of household catastrophic health expenditure and impoverishment in Mongolia as part of national efforts to reform health financing in order to achieve UHC.
Conclusion and discussion
Measuring and monitoring OOP impact is critical for countries aiming at UHC. Since there is an increasing OOP concern, this study using the most recent data provides new evidence on catastrophic health expenditure and its impoverishment effect in. In the past, Mongolia made efforts to estimate OOP impacts by using different methodologies and thresholds, which made it difficult to compare and follow-up policy reforms. Some studies estimate the impoverishment effects of health care payments at the individual level using ADePT software, and some do it at the household level [
14,
26]. The methodology used in the current study is described in the World Bank guidelines which are more relevant to health financing reforms aimed at improving financial protection [
23].
In this study, we estimated the rate of catastrophic health expenditure and impoverishment due to the OOP payments for health care using the HSES 2012. The study reveals several interesting points.
First, 5.5 % of total households suffered from catastrophic health expenditures based on an OOP threshold at 10 % of total household expenditure. At the threshold of 40 % of capacity to pay, 1.1 % of the total household incurred catastrophic health expenditures.
Bredenkamp et al. analysed the incidence of catastrophic health expenditure in Mongolia using the total household consumption as the living standard indicator based on the HSES 2007/2008 data [
14]. They found that the incidence of catastrophic health expenditure was 10 and 3.3 % at the threshold of 10 % of the total household expenditure and 40 % of capacity to pay. Considering the different choices of the living standard indicator between the studies, we cannot make a direct comparison between the results. However, it is worth mentioning that the government of Mongolia provided a one-time subsidy to the uninsured groups from the Human Development Fund (a special stabilization fund from mining revenue) in 2011 [
6]
Consequently, between 2008 and 2012 the SHI coverage increased from 83.2 to 98.1 %. This implied that specifically among the poor and vulnerable groups, financial protection was extended.
Second, to our knowledge, this study provides the first evidence of intensity of catastrophic health payments in Mongolia. Intensity of catastrophic health payments was relatively low, for instance, it was 0.58 % (0.17 %) at the threshold of 10 % (40 %) of total household expenditure (capacity to pay).
Third, the result demonstrates that the richer households (or households with a higher capacity to pay) are more likely to incur catastrophic health payments. Similar results were reported in the World Bank’s study in Mongolia as well as in other developing countries [
4,
14].
This pattern is likely to be a reflection of the health system structure in the country. The ADB (2008) reported that the health system has a risk to become a dual system in where the poor use public facilities and the rich use private facilities [
27].
In a previous study, we found that the poor and low-income groups were more likely to use primary health care, regardless of their health needs, which in general are greater than in higher income groups [
19,
20]. This may be explained by the fact that primary health care is free of charge and more easily accessible than other upper level hospitals. Today, nearly the entire population has SHI coverage regardless of their socio-economic characteristics; however, the low-income groups are substantially less likely to access specialized health care services at the higher referral levels due to both direct costs, including co-payments, medicines, consultations, and indirect costs, such as for transport and meals.
Primary health care centre have a gatekeeping role in the health sector. In practice, the gatekeeping is weak and cannot control self-referrals to the upper level hospital admissions [
6]. In addition, the SHI covers a larger part of inpatient services at the upper level hospitals where the proportion of unnecessary admission is high. These factors lead to higher OOP payments for households.
For instance, in order to obtain inpatient services, the insured person has to pay 10 and 15 % co-payment for inpatient services at secondary and tertiary level hospitals. Moreover, a majority of the inpatients get meals from their home daily. About 40 % of the inpatients buy drugs and injections themselves during their hospitalization [
28]. The HSES 2009 reported that 71 % of household OOP payments tend to be made for medications, which were bought from private pharmacies [
6,
15].
Fourth, we found that the poverty head count before accounting for OOP payments was 22.26 %, which is lower that the NSO’s estimation (27.4 %) [
16]. It can be explained by a choice of living standard indicator. After accounting for OOP for health care from household expenditures, the poverty rate increased by 0.78 percentage points (relative change is 3.51 %). This indicates that about 20,000 people were forced into poverty due to paying for health services based on the Mongolian national poverty line. A similar previous study was conducted by the World Bank team using the HSES 2007/2008, reported that OOP for health payments increased the poverty head count by 2.5 % and that its relative change was 7 % [
14]. In general, this impoverishment effect of OOP payments for health expenditures is lower compared to other developing countries, specifically those countries where prepaid health financing mechanisms is less developed [
4,
22,
29,
30].
In accordance with the findings of the study, we emphasize some potential policy implications. First, in spite of the developments of financial protection policies and a high SHI coverage, the country still has barriers to reaching UHC. Our results suggest that intensity and incidence of catastrophic health expenditure is relatively low; however, the OOP share of total health expenditures is still 41 %. International and regional evidence showed that the higher share of OOP expenses, in particular above 20 %, of total health expenditures in countries, the more people suffer from catastrophic health expenditure [
3,
21]. Additionally, the current way to report the catastrophic health expenditure and impoverishment effect is ad-hoc and is insufficient to support and sustain UHC. Hence, monitoring progress towards UHC in Mongolia requires more frequent studies.
Second, the results also demonstrate that rich households are more likely suffer from catastrophic payments. In general, they tend to bypass primary health care services and seek more expensive upper health care services with self-referral, regardless of their health needs [
19]. In this scenario, direct and indirect costs are usually much higher for users owing to the current weak gatekeeping system.
Hence, a more effective referral system may be beneficial, including stronger gatekeeping, at the primary health care level of the health sector. It would not only reduce households’ risk of incurring catastrophic OOP payments for health expenditures, but it would also lead the health sector to better efficiency. For instance, the WHO concluded that primary health care is essential for better health outcomes of those health care systems, in which primary health care plays a main role in i) providing balanced preventive and promotive services regarding the social determinants, and ii) a referral service to higher level of hospitals [
31].
Third, it is difficult to control cost escalation and unnecessary treatments in the health sector without improving the regulation of the private health sector, including hospitals and pharmacies. The growth of the private health sector with a weak regulation in the country leads to the unmet health needs among the population and increases duplications of health services in the private and public sectors [
9,
27]. Fourth, in 2011, the SHI coverage reached 98.6 % as a result of a one-time subsidy from mining revenues, which was a political promise during the parliament election. Hence, keeping the current high SHI coverage is a vital issue for the future financial protection of the poor and workers in the informal sector.
Importantly, a recent study showed that high SHI coverage alone (breadth) cannot lead to UHC and improve financial protection in the country [
9]. There is a need of policy actions to improve other dimensions of UHC, including health service coverage (depth) and cost coverage (height).
Additionally, it is known that people sell their apartments or borrow money from others in order to afford some specific and expensive health care services regardless of the SHI coverage in the country. In this case, intensity of catastrophic health expenditure and impoverishment is much deeper among certain specific groups. Thus, we emphasize that the HSES questionnaire should be extended with questions which focus on how people make health care payments, for example, from either savings or by loan, etc. Further studies aiming to acquire this information will provide more specific policy messages to the decision makers.
Finally, we should note the limitations of this study. The study did not address the distribution or structure of catastrophic health expenditures and its poverty effect. It may call for further studies in this field. In addition, it is a cross-sectional study, thus unable to make a causal analysis.