Framing our observations of four countries

We conceptualise a health system composed of three distinct domains: institutions consisting of formal and informal rules, norms, customs, inspired by the institutional economics theory popularised by Douglass North4 ; organisations that are the formal structures through which individuals and groups work towards common objectives; and individuals and teams who work within organisations to carry out those objectives (figure 1). The dynamic relationships that exist between the domains underscore that MOHs both influence and are themselves influenced by reform initiatives.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Governance Domains in a Health System

For our analysis, we drew on three commonly known health systems frameworks—the WHO’s ‘building blocks’,5 the 2000 World Health Report’s ‘functions’ of a health system6 and the ‘control knobs’ framework7—to develop a stylised rubric of selected core responsibilities for health finance and provision that are typically distributed across health organisations within a country (figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

The distribution of key health system roles and responsibilities in four case study countriesGP, general practitioner; MOH, ministry of health.

We reviewed literature on the health system of each country, focusing on service delivery and financing reforms for UHC and conducted semistructured interviews with three to four researchers and policymakers in each country in late 2017 and early 2018. We held webinar discussions between subject matter experts from each country and senior officials from the Malaysian MOH to validate our understanding of the organisational transformations and provide insights on the extent to which health sector transformations were affected by the sociopolitical context, the sequencing of organisational change, the building of capacity to meet these organisational changes and the subsequent changes in accountability structures. Our inquiry focused on understanding organisational transformations in the MOHs accompanying reforms in finance and delivery.

The countries we examined experienced major reforms to the financing and delivery of healthcare over one or more decades. Although each country established a health purchasing agency, the control the MOH exerts over the purchaser varied in formal structure and in practice. Independent purchasers in a nominal or de jure sense may function in close concert with the leadership of MOH because of structural linkages such as the location of rights to appoint directors or governing boards or rules regarding MOH oversight. But how this distribution actually functions, in terms of actual roles, responsibilities, authorities and accountabilities, may not be well reflected in the nominal or de jure structures and may be modulated by institutional factors. We found that the sequence of steps taken to put in place reforms varies even if two countries had the same programmatic goal—and in most cases goals vary considerably. The sociopolitical context of each country affects the trajectory of implementation and change significantly.

Figure 2 shows the variations found in our identified set of core responsibilities in service delivery and financing across health system organisations after reform implementation in the case study countries. The conceptual starting point is a government-funded, owned and operated health system in which all responsibilities were embedded within the MOH—the classic ‘national health service’ model. Three of the case countries implemented reforms to move away from a unitary system, while Australia never had a ‘national’ health service.

Chile

In the 1970s and early 1980s, the military government in Chile fundamentally altered how government health services were financed and delivered. The funding and purchasing of health services was given over to a new government organisation, FONASA. A legal framework was established to promote the participation of private health insurers, known as ISAPREs, and the market share and power of private providers subsequently increased. Over time, it became apparent that private insurers enrolled healthier citizens and thus the burden on the government-financed system did not decrease, also partially due to the epidemiological and demographic changes in the population. The two-tiered system exacerbated equity and financing gaps, which the democratically elected government of the 2000s addressed by standardising the quality and service criteria by which any provider or insurer would be reimbursed. This was done by mandating a national health benefits package through the Universal Access with Explicit Guarantees (AUGE) reforms of the mid-2000s. A Superintendency of Health (SOH) was established to oversee the AUGE benefits package purchased by FONASA, the government purchaser, and by private insurers.

The relationships between MOH, SOH and FONASA illustrate the complexities of formal governance structures as well as informal relationships. On paper, MOH roles have narrowed greatly: government funding for healthcare now flows through FONASA, not through the MOH budget, and the direct management of healthcare delivery is the responsibility of subnational organisations. The MOH has less vertical authority but has a very strong role in overseeing the system and can exert a high degree of influence. For example, FONASA is structurally independent of the MOH, but the director of FONASA is appointed by the President in agreement with the Minister of Health. The SOH is an independent government organisation linked to the President through the MOH. It regulates the government purchaser, FONASA, and private insurers. The MOH has responsibility for overall system performance and health sector policy. Although the three organisations are operationally independent, decisions about their leadership, with consequences for their operations, reflect considerable influence from the MOH. One downside to this arrangement is lack of certainty about the policy direction agencies may take since their leaders could change quickly for political reasons. Effective health governance in Chile is therefore partially dependent on personal and political relationships in addition to the formal organisational governance structures.

Poland

Poland underwent multiple health system reforms over two decades, starting in the 1990s, in step with broader administrative reforms for increased local autonomy and liberal governance in the public sector as the country transitioned from communist party rule to a market economy. Poland’s centrally planned government healthcare system evolved to a social health insurance model with a mixed government-private delivery system. New payment methods such as capitation and diagnosis-related groups were introduced. Dissatisfaction with underfunding of public services which characterised the initial decentralisation led to the recentralisation of purchasing. Today, Poland is the only case study country that has a single national health insurance fund, the National Health Fund (NFZ), which is a hybrid of an internal government ‘purchaser’ and a more independent ‘social health insurance’ agency. Contributions are required for formally employed workers meeting certain requirements and the government subsidises care for other population groups through additional tax-based funding.

Under the centrally planned model, the MOH was the funder, organiser and regulator for healthcare. Resource allocation was input based rather than performance or outcomes based. The establishment of the independent NFZ eliminated the MOH’s role as funder of health services. Some functions previously implemented by the MOH were externalised into separate organisations reflecting capacity and financing issues or inadequate geographical reach of a centralised system. For example, an independent agency for health technology assessment was established to introduce evidence-based analysis in developing the benefits package for the NFZ because the MOH lacked the technical capacity for this function. Decision makers perceived that compensation levels above civil service levels would attract a workforce with the necessary competencies. However, the MOH remains legally responsible and most of these organisations remain formally accountable to the MOH. This allows the MOH to continue its strong role in system governance. Even as the MOH’s official functions were reduced during the reforms, over time the MOH has expanded the scope of its regulatory role. The MOH approves NFZ annual plans and determines key financing and purchasing aspects, including geographical distribution of funds, the benefits package, tariffs, broad purchasing guidelines and quality controls.

Thailand

In the early 2000s, Thailand introduced a new government health insurance, the Universal Coverage Scheme (UCS), for the 75% of the population that was previously uninsured or underinsured, including people not formally employed, or those covered by the Medical Welfare Scheme as well as voluntary community-based health insurance. The National Health Security Office (NHSO) was created as the new government purchaser for the UCS. Technocrats within the Ministry of Public Health (MOPH) and political leaders in the Thai Rak Thai Party, which won the 2001 general election, believed that UHC was technically feasible and that the funding gap could be filled by the government. After reform, the MOPH exercised three types of power: official power as the steward of the overall health system of Thailand; managerial power as the owner and operator of the MOPH healthcare service delivery system itself; and power of influence as a strategic partner for the other government insurance schemes. The establishment of NHSO represented the largest change to traditional roles and responsibilities in the government health sector. The NHSO manages funds and purchase services on behalf of UCS beneficiaries through strategic purchasing instead of the internal management mechanisms used by the two other ministries overseeing existing government health insurance programmes covering civil servants and formal employees.

The Minister of Public Health chairs the independent board that oversees the NHSO, with authority over leadership functions such as determining the annual UCS budget and the benefits package. But there are measures that limit the previous direct management authority of the Minister of Public Health. To encourage external accountability and transparency of management, the NHSO board has 30 external members, including five members representing civil society organisations, seven experts, MOPH officials, and permanent secretaries from other line ministries. The Minister of Public Health cannot use traditional command authority with this blended group.

Australia

Unlike the other case study countries, Australia never had a national health service system under central government purview. Its complex system differs in financing and organisation for primary care and hospital service. Primary care is largely a unitary, universal and central government-funded system, although largely privately delivered. Most general practitioners are reimbursed according to fee schedules through the Medicare programme. For hospital services, a mixed government-private insurance approach results in a state-based system. States have authority to manage their hospital systems. Most hospitals are owned and operated by state governments and jointly funded by the federal and state governments. Alongside public insurance, robust policies and financial incentives from the federal government, such as tax penalties against the wealthy for non-participation, encourage enrolment in private health insurance for hospital care.

Different organisations govern different aspects of the health system, but all are linked through a network of policies set by the Federal Department of Health (DOH) for the services outside hospitals, and by the State Departments of Health for services within public hospitals. The DOH sets pricing and other policies for the universal care programme, including the amounts set in the Medicare Benefits Schedule; partially funds service delivery; and provides the policies and funding that maintain enrolment in private insurance through incentives and penalties. The Department of Human Services administers funding and claims in accordance with DOH policies. States fund and provide services within the limits of agreements and policies set by the DOH, directly operating most public hospitals and funding many public and community health services. Recent reforms around quality and integration of primary care have allowed Primary Care Networks to purchase services independently although in accordance with policies set by the federal government.

Learnings relevant to other countries

From this comparative study of organisational transformation within the health sector in four upper and middle-income countries, we identified four observations relevant to other countries. First, no single model emerges for how responsibilities are distributed in health systems after reforms. The changes that result from health system reforms are more complex than the simple reduction or shifting away of MOH authority and responsibilities. There are several explanations for why a country may choose different distributions of roles between organisations or between teams within a single organisation, such as the capacity of MOH or other entities to perform a given responsibility; the need to realign accountability and power with new incentive structures, organisational relationships and political preferences; or the perceived need for independence and transparency. As one example, as Chile decentralised much authority for healthcare delivery to municipal governments, its MOH did not simply reduce its responsibilities, but rather had to develop its capacity to guide, supervise and enforce rules within a system of devolved responsibilities.

Second, the exercise of responsibilities in practice in a health system may differ substantially from the de jure, nominal, or official distribution of those responsibilities. The configuration of responsibilities, power and risk arising from reforms is often not accurately discerned from the formal structures that are in place. Health system reforms often formally incorporate significant institutional and organisational changes such as the movement of financing to a third-party payer or decentralising authorities for healthcare delivery to local governments or individual facilities. However, our case studies suggest that an MOH can retain substantial power and influence in financing and delivery even when formal reforms remove these powers. For example, in Poland the MOH still holds primary legal responsibility for performance of organisations in the sector, even where those organisations are independent of the MOH.

Third, the ability of an MOH to act alone to achieve fundamental system restructuring is limited when such restructuring involves significant organisational transformation. Centrally funded and operated national health service bureaucracies reflect long-standing conditions which impede endogenously driven change. Without political impetus and external buy-in, it is likely that an MOH will rely on management-level reforms within its own authority to try to implement large system changes. The system-wide reforms we examined in this comparative study were not purely managerial or technical in nature and required stakeholder mobilisation from outside the MOH. For example, as seen in the Thailand case, an MOH is unlikely to be able to design or implement institutional change for financing reforms without the ministry of finance.

Fourth, the MOH may use a variety of strategies to shape organisational change that accompanies health systems reform. These include:

1. Establish a high-level interministerial team to create political commitment to UHC-related reforms as a countervailing force to bureaucratic resistance to change. Particularly in countries where an MOH may be perceived as a ministry with relatively lower power and influence,8 an interministerial platform, with technical guidance from the MOH, could envision the future health system and alleviate institutional barriers to organisational changes in MOH or other agencies.

2. Establish a ‘change team’9 in the MOH responsible for translating the strategic systems-level decisions into implementable steps for internal organisational change. A change team may enable change in the domain of individuals and teams, by accessing different parts of the organisation and developing a roadmap to navigate and sequence potentially disruptive changes within the bureaucratic structure. In strongly vertical organisations such as an MOH-led national health service, a change team can drive organisational change at the lower levels of the organisation.

3. Pursue legislation to modify the institutional environment to enable the changes needed in the domain of organisations. When Poland, Chile and Thailand set up third-party purchasers, legislation served to define the functions of new entities and the separation of powers from the MOH. Legislation can preserve key reform components through future political fluctuations.

4. Capitalise on political ‘windows of opportunity’10 when political change makes it more possible to successfully introduce health reforms and accompanying organisational change. Although this is a more difficult mechanism, it is arguably more successful because comprehensive government reform packages reflect political commitment. Ideally, a change team in MOH would prepare for and identify such political windows and pursue a legislation change to underpin the health system reforms.