UHC as proposed by the World Health Organization (WHO) is that ‘all’ people and communities are able to access essential health services of sufficient quality, while the government ensures that the use of such services does not expose the users to financial hardship [11]. This study clearly confirms that Thailand’s UHC achieved a high level of financial risk protection against catastrophic health spending and impoverishment from health payments by households and reaffirms the negative correlation between public health insurance coverage and incidence of catastrophic payments [6]. The percentage of households in Thailand encountering catastrophic health spending and healthcare impoverishment was on par with several high-income countries in Europe, North America and Oceania; for instance, Austria, France and Germany [6, 12].
Prior to UHC in 2002, all Thai citizens, including the rural poor, had adequate access to health services at the district level which provided primary and secondary services. As a result of successive governments’ investment in the health delivery system since the 1970s, Thailand had achieved full geographical coverage of a district health system nation-wide by the mid 1990s. The district health system consists of a district hospital of 10-120 beds and a network of 10-15 health centres. A health centre serves a catchment area of five thousand people. District health systems are the backbone for equitable access to services for all populations [13]. In parallel to the health delivery systems expansion, successive governments introduced financial risk protection schemes through targeting different population groups such as low-income households since 1975, the informal sector since 1984 and private sector employees since 1991 [14], until UHC was achieved in 2002. These financial risk protection schemes prior to and after UHC in 2002 provided a comprehensive benefit package [15] with minimum co-payments which reduced household out-of-pocket payment significantly from 34% of Current Health Expenditure (CHE) in 2000 prior to UHC to 11% of CHE in 2017 [16]. Access to care through district health systems, the so called “closed-to-client setting” [17] foster equity in access and benefit incidence [18]. One study also shows equity in maternal and child health services coverage as a result of district health systems [19].
Economic growth is the main driver for poverty reduction in Thailand. Per capita Gross Domestic Product (GDP) increased from US$ 1,084 in 1986 to 3,415 in 2013 (2005 constant US$). Extreme poverty as measured by the international extreme poverty line (US$ 1.90 per day, 2011 PPP) is no longer a concern, as it fell from 14.3% in 1988 to 0.1% in 2012. Using a national poverty line (in 2013, approximately US$ 6.20 per day 2011 PPP), the poverty head count also fell from 67% in 1986 to 10.5% in 2014, with 26.8 million Thai citizens moving out of poverty (Figure 6) [20]. Evidence shows that since 2000, economic growth has continued to play the dominant role in reducing poverty. Although redistribution has also helped, nearly 85% of poverty reduction was attributable to growth while the remaining 15% was attributable to improvements in income distribution. Further analysis between 2006 and 2013 shows that growth has been highly pro-poor, with redistribution playing a larger role, primarily through the introduction of elderly pensions and UHC [20]. This evidence explains why poverty levels prior to household out-of-pocket payments are low. Further, the low level of out-of-pocket payment for health by households, as a result of comprehensive benefit package and zero co-payment, explains the low level of additional poor due to medical payments. This helps explain the phenomena in Figure 4, 5 and 6.
Several factors synergistically contributed to the financial protection of households against catastrophic health spending and impoverishment.
Firstly, all three public health insurance schemes provide full financial coverage to their members and cover the full cost of services to healthcare facilities; this does not allow any co-payment or balanced billing from service users. Full financial coverage for health services therefore reduced OOP in households. Also, general taxation, the sole source of financing for UCS and CSMBS, is the most progressive source of health financing as the rich pay a higher direct tax in monetary terms than the poor [1]. This is a redistribution tool between the rich and poor. Only public sources of financing, and not a reliance on external donors, can sustain UHC in the long term. Full financial coverage is reflected in the percentage of domestic general government health expenditure (GGHE-D) to current health expenditure which increased from 65% in 2002 (when the UCS was launched) to 78% in 2016; while the percentage of OOP on current health expenditure reduced from 28% to 12% during the same period [21]. The lower the proportion of OOP in financing health services, the lower the incidence of catastrophic health spending and impoverishment [22].
Secondly, the benefits package covered by all schemes is comprehensive, with no maximum limit of financial coverage and no co-payment at point of service, resulting in a massive reduction of OOP for households. The benefits package also applies a negative list approach; that is, all interventions are covered except a few exclusions such as infertility, aesthetic surgery and treatment under research or a pilot study [1]. In 2006, when the national capacity to conduct health technology assessments improved, more cost-effective interventions were included in the benefits package, which further boosted financial risk protection [15]. Curative services included medicines on the national list of essential medicines (NLEM) and in 2004 the NLEM was scaled up from the minimum ‘essential medicine list’ (with reference to the WHO model list) to a ‘reimbursement list’ for all three public health insurance schemes[15]. As of 2017, there are 849 drug items on the current NLEM [23], Table 4.
Table 4: Number of drugs in the national list of essential medicines, by 17 groups
Group no.
|
Category
|
No. of drugs
|
1
|
Gastrointestinal
|
39
|
2
|
Cardiovascular
|
72
|
3
|
Respiratory
|
30
|
4
|
Central nervous systems
|
102
|
5
|
Infections
|
133
|
6
|
Endocrine systems
|
43
|
7
|
Obstetrics and gynaecology
|
22
|
8
|
Malignant diseases and immuno-suppression
|
56
|
9
|
Nutrition
|
93
|
10
|
Musculoskeletal and joint diseases
|
24
|
11
|
Eye
|
41
|
12
|
Ears, nose, oropharynx and oral cavity
|
42
|
13
|
Skin
|
47
|
14
|
Immunological products and vaccines
|
24
|
15
|
Anaesthesia
|
31
|
16
|
Antidotes
|
33
|
17
|
Contrast media and radiopharmaceuticals
|
17
|
Total
|
849
|
Source: Food and Drug Administration (FDA), Thailand [23]
Thirdly, closed-end provider payment, notably the dominance of capitation for OP care and Diagnostic Related Groups under the global budget for IP care, is applied by the three schemes (except the fee for service for CSMBS OP services). This results in cost containment which frees up budget for the extension of the benefits package to further strengthen financial risk protection [1]. The UCS covers certain high-cost life-saving interventions such as antiretroviral treatment for HIV in 2006 and renal replacement therapy in 2009 (chronic dialysis is not cost effective, but the cost of dialysis is prohibitively high and can be catastrophic to households) [24, 25]. The UCS also covers long-term community interventions such as treatment for psychotic diseases, certain items in Thai traditional medicine, and seasonal influenza vaccinations [26]. Figure 6 describes the chronological events of the extension of the UCS benefits package to high-cost interventions, which were all subject to rigorous health technology assessment.
Fourthly, Thailand has developed local capacities to generate evidence on health technology assessments. Health technology assessments help improve the efficiency of resource use and minimize waste from spending on interventions that are not cost-effective. It was rigorously applied to the annual review for the inclusion of new health interventions into the UCS benefits package. The benchmark for including cost-effective interventions is the incremental cost-effectiveness ratio equal to one GDP per capita for one Quality Adjusted Life Year gained from the intervention. Alongside value for money, which is measured by the cost effectiveness ratio, other criteria for decision-making are equally important. They include budget impact assessments, which should meet the criteria of being within the state’s fiscal capacity to fund new interventions and the readiness of health systems to deliver the interventions equitably [27]. Assessment of these criteria ensures smooth implementation of new interventions.
Fifthly, capitation payment requires UCS members to register with a primary health care network, which comprises 1 district hospital and 10-12 sub-district health centres, serving about 50,000 people in the district catchment area [1, 3]. The gatekeeping function of a primary healthcare contractor network gains efficiency and provides better continuity of care for non-communicable diseases (NCD) in particular. Better access to a primary health care network, with assured referral to provincial tertiary care hospitals when clinically indicated, results in adequate use of services and low level of OOP and transport cost by households [3, 25].
Lastly, the full geographical coverage of over 9,800 sub-district health centres in all 8,860 sub-districts, and 780 district hospitals and 116 provincial/regional hospitals in all 998 districts and 77 provinces is the solid platform for equitable access to the comprehensive benefits package which results in favourable financial risk protection at sub-national level [3, 13].
Certain limitations remain. Firstly, data on OOP paid by households is an aggregate figure which does not identify types of health facilities; this hampers further detailed breakdown analysis by types of health facility. Secondly, as the unit of analysis is ‘household’ not ‘individual’, per capita expenditure was estimated from total household OOP divided by the number of household members without adjustment; this cannot perfectly represent the real data collected from the individual household member. Lastly, the interview survey is prone to recall bias, which may undermine the accuracy of reported data by household members. Also, there was a possibility that the head of a household, who is the respondent to NHSO surveys, may not catch up with the real health spending by other household members.