Conceptual framework

As health care is meant to alleviate pain and suffering of the medically needy and consequently can enhance the wellbeing of individuals, health policy, which concerns the mobilization and distribution of the health care resources contributes to the welfare improvement of the society. Health policy viewed as one aspect of public policy can hence be analyzed through adopting the process framework developed for public policy. In this study, we have borrowed some concepts from the studies of public policy and proposed a conceptual framework for examining the evolution of Taiwan’s health care system through an iterative evolutionary process with population health problems as inputs and health care reform decision and action as outputs (Figure 1). In this framework, we specifically emphasize four external forces, which bear significant influence on the development of health policies, namely, knowledge, socio-cultural context, economic resources and the political system.

Figure 1

Analytical framework for health policy.

Input – ‘Population health problems’: As depicted in Figure 1, the recognition of ‘population health problems’ triggers the health policy process and, in the health sector, an epidemiological transition of disease pattern signifies the birth of different health problems to be resolved through an iterative process.

Process – ‘Policy process’: Recognition of population health problems is followed by the policy process box, where phases of policy agenda setting, policy formulation, and policy adoption are undertaken (Anderson, Reference Anderson1994).

Policy agenda: The magnitude and urgency of the population health problems identified will be assessed and prioritized, upon which the policy agenda will be set.

Policy formulation: Once the policy agenda is set, a feasible work proposal that delineates courses of action to resolve the health problems should be developed.

Policy adoption: The last phase of the policy process focuses on legitimizing and authorizing the proposed work plan and getting the government to accept the proposed solution.

In addition, there are four external forces, which facilitate the policy process: knowledge, socio-cultural context, economic resources and political system.

Knowledge: Knowledge is essential to problem solving in the policy process, which can be generated not only from systematically collecting domestic data to produce scientific evidence, but also from exploiting valuable lessons abroad. The value of the knowledge bestowed upon the process depends on the group of domestic and international scholars involved.

Socio-cultural context: Socio-cultural context is important as it reflects the beliefs and ideology held by the members of a society; consequently it often exerts significant influence on the policy process in terms of reaching a consensus in policy making, which is acceptable to the society.

Economic resources: The strength of the economy is essentially the backbone of any reform policy proposal as it determines the resources available to eventually facilitate the policy implementation. In particular, the size of the resources in the private sector signifies the significance of the role that the private sector can play in initiating changes.

Political system: Political force is a vital determinant of the feasibility and operability of any policy proposal as it is closely associated with the executive arms of the government. In the context of social policy, political and social environment are regarded as crucial to the practicability of the design and execution of the social policy to be launched (Hill and Bramley, Reference Hill and Bramley1986).

Output – ‘Reform decisions and actions’: The outcome of the policy process is phased into ‘reform decisions and actions’ stage. In the governmental policy making process, two overriding questions often asked tend to focus on the efficacy of the policy making in solving problems and on its responsiveness to popular control (Lindblom, Reference Lindblom1980). Consequently, reform decisions and actions will lead to the iterative evolutionary process of the policy making loop in response to constantly changing public demand or unsatisfactory reform outcomes.

In the following section, an account of the health policies implemented in Taiwan from 1970 onward illustrates the conceptual framework proposed in this section.

Pursuing health care for all

In the early 1970s, the development of health policy in Taiwan moved into a new era characterized by the government’s policy focus on the supply of health care professionals and hospital capacities, and then tackling an uneven distribution of health care resources and quality issues.

1970–1985: Period of expansion

The establishment of the DOH as a Directorate General in charge of health affairs directly under the Executive Yuan in 1971 marked a momentous step in the expansion of the health care industry in Taiwan. Despite government reports, which claimed no physician shortage in the late 1960s (Hsu, Reference Hsu1965; Wu, Reference Wu1980), the number of physicians per 100,000 population declined from 45 in the 1950s to 41 in the 1970s and a quarter of health stations were often staffed without physicians in rural areas (Chiang, Reference Chiang1995a; Table 1). With an aim to enhance access to health care, the government no longer adopted a laissez-faire approach, but rather took up, a much more aggressive policy to expand the supply of health professionals and tackle the issue of mal-distribution of health resources through various policies.

Table 1

Basic and health care indicators: Taiwan, 1960–2006

*1961; **1971; ***1982; na = not available.

Source: Chiang, T. L. (Reference Chiang2003); Council for Economic Planning and Development (2007); Department of Health (2007).

First, to cope with the imminent demand for physician services and the retirement issue of military doctors (most of them were not trained in medical school), the government granted practicing licenses to 2724 retired military doctors through seven special license examinations from 1972 to 1976 and was able to recruit some to practice in the rural health stations (Lin, Reference Lin1980). Although this measure somewhat ameliorated the shortage problem, critiques on quality of care rendered were not uncommon.

Second, to achieve the target of one physician per 1000 population, the government took multi-faceted action to increase medical student enrollment, which had reached roughly 1200 per year by the mid-1980s (Chiang, Reference Chiang1995a). The enrollment expansion was made possible through opening up two new public medical schools (in 1975 and 1983), scaling up the enrollment class size at existing medical schools, and launching in 1982 several new five-year post-graduate programsFootnote 1 (similar to the most prevailing medical school system in US) parallel to its concurrent seven-year program (Chiang, Reference Chiang2003).

Third, the DOH also aimed to expand hospital capacities to achieve the target of four hospital beds per 1000 population with joint efforts contributed by the private sector (Chiang, Reference Chiang1997). With a growing economy, the private sector was able to mobilize more funds to invest in hospitals and launch large-scale hospital construction projects. One of the most significant examples is the hospital investment project undertaken by the Chang Gung Memorial Hospital group in 1976, which now operates more than 9000 beds in five different geographical locations and has become the largest hospital chain in Taiwan. This continuing endeavor by private investment (mainly in large-scale proprietary hospitals managed by non-for-profit foundations)Footnote 2 was reflected by the steadily declining market share taken up by public hospitals and increasing average hospital scale since 1960 (Table 1).

Fourth, while the expansion in the supply of physicians and hospital beds seems to have achieved the target, the mal-distribution of health care resources remained a real issue to tackle. The most significant policy intervention to address the mal-distribution of physician workforce was the ‘group practice center’ (GPC) program introduced in 1983 as part of a series of policies laid out to enhance the infrastructure of rural communities and to improve farmers’ income levels. The GPCs were normally set up in under-served areas and shared capital facilities with local health stations (Chiang et al., Reference Chiang, Yaung and Lew-Ting1991). The DOH, for its part, subsidized large teaching hospitals to send their physicians to practice in the centers and concurrently allowed the staff physicians to retain 80% of net profits as a bonus, which greatly enticed physicians to practice in rural areas. In addition, starting in 1984, the government assigned public-sponsored medical graduatesFootnote 3 to practice in the GPCs, which assured a steady flow of staff physicians for the GPC’s (Chiang, Reference Chiang2003).

Moving toward a high performing health-care system

Despite government’s efforts to enhance the functions of the delivery system through various intervention programs since the 1970s, fragmented health care financing mechanisms received much less due attention. In 1984, exploiting the fruits of recent socio-economic development and in response to the call for a widespread social security system, the government charged the CEPD with organizing a planning task force to study Taiwan’s social security programs (Chiang, Reference Chiang1997). Two years later, the task force completed an integrated planning report and concluded with a recommendation for a universal health insurance scheme by the year 2000 and a supply target of 1.5 physicians and five beds per 1000 population (CEPD, 1986a). These recommendations later became a chapter of ‘Long-term Economic Perspectives in Taiwan, 1986–2000’, which eventually formed Executive Yuan’s policy guidelines for Taiwan’s economic development (CEPD, 1986b). In fact, Premier Kuo-hwa Yu had made the policy announcement of ‘health insurance for all by 2000’ in his report to the Legislative Yuan in early 1986.

At the same time, Taiwan’s political environment and climate had gone through major transitional changes. The political participation by the opposition power substantially intensified and the government lifted the Emergency Decree Law (martial law) in 1987. In the process of democratization, Taiwan’s first opposition party, the Democratic Progressive Party (DPP) was officially formed on 28 September 1986. To mobilize public support and raise its political capital, DPP often seized headlines to fiercely criticize the government on issues such as political freedom and social security. Although, the ruling KMT party had strategically implemented a health insurance demonstration project to cover farmers in 1985 (fully implemented in 1989), the mounting political pressure compelled an earlier enactment of National Health Insurance (NHI) to appeal to the public. On 28 February 1989, Premier Yu finally announced to advance the new target year for implementing the NHI program to 1995 (CEPD, 1990).

In 1988, CEPD had assembled a task force composed of Taiwanese public health scholarsFootnote 8 to draft the basic principles of the new health insurance program. Recognizing the importance of international experiences, Minister of CEPD Frederick Chien engaged Professor William Hsiao of Harvard University as chief advisor to the task force in 1989. As a result of a cabinet reshuffle in the middle of 1990, the CEPD task force was dismissed and a task force under the DOH was created after the NHI planning report was submitted to the Executive Yuan (CEPD, 1990). After nearly five years of planning efforts (1988–1993) and a two-year legislative marathon, the NHI program was inaugurated on 1 March 1995. As the task force emphasized the values of learning from the past and abroad, the Taiwan NHI program has been described by the former deputy Health Minister and CEO of Bureau of NHI (BNHI), Dr Hong-Jen Chang, as ‘a car that has been domestically designed and produced, but with many components imported from over ten other countries’ (Cheng, Reference Cheng2003).

Before the introduction of NHI in 1995, Taiwan’s government had established four major social insurance programs,Footnote 9 which closely modeled the social insurance approach of Germany and Japan, namely Labor Insurance (in 1950), Government Employee Insurance (in 1958), Farmers Insurance (in 1985) and Low-Income Household Insurance (in 1990) (Lu and Hsieh, Reference Lu and Hsieh2000; Cheng, Reference Cheng2003; Lu and Hsiao, Reference Lu and Hsiao2003). Collectively, those existing social insurance programs had covered about 57% of the total population by 1995, compared to 16% by 1980, which left most of the high-risk population uninsured (Table 2, Figure 2). Recognizing the administrative deficiency and concurrent large deficits in the existing multiple payer social insurance system, then President Teng-Hui Lee along with major policy makers took the advice from the task force and decided on a single-payer model. BNHI, a quasi-governmental agency that by law is the only administration that operates the insurance program was then inaugurated in January 1995, nearly six months after the Legislative Yuan passed the NHI Act.

Table 2

Financing mix in Taiwan, 1991–2006

Source: Department of Health (2008).

Figure 2

Insurance coverage expansion in Taiwan, insured people as a percentage of population 1950–2006. Sources: For 1950–1994, data from Lu and Hsieh (Reference Lu and Hsieh2000), ‘The National Health Insurance Program in Taiwan’, in Health Economics, Taipei: Pro-Ed Publishing Company; for data after 1995, Department of Health (2008), Health Statistics in Taiwan, 2006, Taipei: Department of Health, Executive Yuan, R.O.C. (Taiwan).

The accelerated implementation timeline of NHI (five years ahead of its original schedule) was mostly a product of political compromise to the increasing public demand for universal health coverage. Given the scale of the program, there was chaos when NHI was first introduced. However, with some precautionary measures taken by the government, there was hardly any major unintended undesirable consequence. First, the design of NHI was a path-dependent social insurance model, which the previously insured 57% of the population was familiar with. Second, Prime Minister Lien has formed an intergovernmental task force in September 1993 to coordinate NHI related affairs among the ministries involved within Executive Yuan. In addition, the NHI Preparatory Office was officially established in December 1993 to prepare for the inauguration of the NHI program in 1995. Third, the premium contribution rate was politically set with consideration of insurance shock effect. Hence, NHI was free from financial distress at the initial implementation stage. Fourth, the BNHI set up many hot lines to accommodate call-in for questions and complaints, and several Ministers without portfolio toured around the country to engage face-to-face communication with the public, followed by improved operations of the NHI. With the considerably concerted efforts, the public responded with a satisfaction rate increase from 30% in April 1995 to 50% in September 1995, and to 55% in January 1996 (Department of Health, 1996).

The government planned its new NHI system to achieve three essential objectives: providing equal access to health care for all citizens, controlling total health spending within a reasonable level, and promoting efficient use of health care resources. The NHI provides a comprehensive benefits package that covers preventive and medical services, prescription drugs, dental services, Chinese medicine and home nurse visits. To constrain growing utilization rates due to moral hazard, the NHI also incorporated a co-payment of US$2 for each outpatient visit to clinics, US$5 for each visit to medical center outpatient clinics, and a 10% co-insurance for inpatient services, but capped the total amount that a patient has to pay for each admission at 6%, and for each year at 10% of the average national income per person (Lu and Hsiao, Reference Lu and Hsiao2003).

On the delivery side, Taiwan maintained a market-driven system, which reflects Taiwan’s free-enterprise economy. Both patients and providers have free choice. Hospitals and physicians are paid mainly on a fee-for-service basis and abide by a uniform fee schedule established by BNHI for its contracted providers, which was adopted from the one originally implemented by the Labor Insurance program. Hospital physicians are often paid on a salaried basis with bonus payments based on productivity. Private practitioners remain mainly reimbursed on a fee-for-service basis. For the same type of treatment rendered, the medical institutions are reimbursed contingent upon their accreditation status, which bestows the medical institutions incentives to ‘upgrade’ through means of expanding capacities. As a result, the average bed size of medical institutions has gone up from 82 beds in 1986 to 171 beds in 2000 (Lu and Hsieh, Reference Lu and Hsieh2003). Evidently, the market share of NHI expenditures has been unevenly absorbed by the small number of large-scale hospitals.

NHI is mainly financed through premiums (in the form of payroll tax), supplemented with direct government funding. The premium, calculated as a percent of total payroll was to be shared between employers, employees and the government, at the rate of 60%, 30% and 10%, respectively. At its infancy, the NHI wage base was rather narrow as contributions are capped at the upper end of income (where the top earning base was only 5.5 times that of the lowest earning base), but later the wage base was gradually expanded to resolve the financial distress. To avoid the possibility that employers may discriminate against workers with large families, the employers would only have to pay for the worker plus 0.7 dependent. The worker has to pay the premium for himself/herself plus up to three dependents. The government also allotted a reserve fund for NHI at its inauguration and would pay for the annual operating cost, which is currently less than 2% (by law, it is capped at 3.5%) of its total expenditures. In other words, the collected premiums are only used to cover the medical expenditures incurred. Of the total premium income in 2006, the insured is the largest contributor, paying for 37.9%, followed by 33.9% from the government (25.4% as subsidiesFootnote 10 and 8.5% paid as an employer) and the remaining 28.1% is borne by the employers (BNHI, 2008).

However, the NHI’s revenue has not kept pace with gross domestic product growth, as the annual growth rate for health expenditure was 6%, compared to 4% for premium income. In 1998, three years after the NHI’s implementation, the actual annual cash flow of the NHI program began to run a deficit. To counter financial pressure born by steady growth in expenditure outlay and a relatively lagged increase in premium income, BNHI gradually set up separate global budgets for dental services, Chinese medicines, and primary care services (outpatient services delivered in clinic settings). In 2002, Taiwan created a separate global budget for hospital outpatient and inpatient services, which provoked vociferous opposition by the hospitals.

Given its financial condition and significant impact on the welfare of the Taiwanese people, NHI was never short of reform proposals over the years (Figure 3). In 1997, just two years after the NHI was introduced, the KMT government proposed a new ‘multiple carriers’ structure in providing health insurance, often misperceived as ‘privatization of the NHI scheme’, and the draft amendment essentially died in the Health and Welfare Committee of the Legislative Yuan (Wong, Reference Wong2003; Chiang, Reference Chiang2005). When the government changed hands in the 2000 presidential election, the newly elected DPP government first charged the National Health Research Institutes to critically review the performance of NHI (NHI Review Committee, 2001). After the NHI Review Committee completed the assessment report in 2001, the DOH decided to form the ‘2nd generation NHI Planning (NHI 2G) Task Force’ in July 2001 to develop a reform proposal. Before the reform proposal could be presented in 2002, the government decided to immediately raise the contribution rate from 4.25% to 4.55% and co-payment for services rendered at medical centers and regional hospitals, as the NHI reserve fund had less than one-month expenditure outlay.Footnote 11 This move, termed ‘NHI Double Jeopardy’ by the media, consequently created a perilous political storm.

Figure 3

National Health Insurance (NHI) reform policies, 1995–2006.

Facing a nearly depleted reserve fund in 2005,Footnote 12 NHI quickly promulgated a series of ‘multi-facet micro-tuning’ programs to expand the revenue base and further contain costs, but deliberately circumvented further increases in contribution rates across the board. Specifically, the contribution cap was raised so that the highest earning base became 8.3 times that of the lowest; tobacco tax was doubled from NT$5 to NT$10; and new budgets from relevant government agencies were appropriated to finance preventive interventions, infection disease control, and medical education in teaching hospitals. On cost containment, NHI instituted primary care gate keeping by increasing co-payments for hospital services without a referral, and strengthening physician and patient profiling to reduce duplicate and unnecessary tests and prescriptions. Figure 3 depicts NHI policy reforms from 1995 to 2006.

Nevertheless, ‘multi-facet micro-tuning’ was only a temporary measure to ameliorate the imminent threat of insolvency. In 2006, amidst much controversy, the NHI 2G task force unveiled a reform package and submitted it to the Executive Yuan for consideration. This legislative amendment proposed to change NHI premium contribution from a payroll tax base to a household income base thereby broadening the contribution pool, and to limit (by explicitly capping contributions at a fixed percentage of the total revenue base) the financial exposure of employers and government. In particular, the current system classifies the insured into 6 categories and 14 classes based on occupation by which the level of government subsidy varies rather than by income. By moving away from this complicated occupation-based classification to a household income contribution mechanism, NHI anticipates that government subsidies would be more targeted to the socially indigent for greater equity, although it could prove administratively costly in terms of premium assessment and collection. Due to its controversy, the draft legislative amendment never made it to the first reading.

Currently, Taiwan reported spending approximately 6.1% of GDP in health care, averaging to a per capita health expenditure of NT$31,661 (US$934) in 2006, of which NHI, household out of pocket payment, and government, accounts for 55.9%, 34.3% and 6.6%, respectively (Table 2).Footnote 13 An average household in Taiwan devoted 3.74% of total household consumption expenditures to health care and the distribution of out-of-pocket sharesFootnote 14 across income quintiles (indicated by concentration index of −0.1065) showed a modest pro-poor tendency, namely, that the less well-off were burdened with a higher out-of-pocket share than the better-off (Van Doorslaer et al., Reference Van Doorslaer, O’Donnell, Rannan-Eliya, Somanathan, Adhikari, Garg, Harbianto, Herrin, Huq, Ibragimova, Karan, Lee, Leung, Lu, Ng, Pande, Racelis, Tao, Tin, Tisayaticom, Trisnantoro, Visasvid and Zhao2007).