Kieke Okma, Associate Professor at New York University, reviews the core elements of health systems to highlight key differences between countries. —Report from a presentation at the 2009 conference of the MUHC-ISAI

It is hard to compare countries. But identifying the basic elements of health systems can help us to see where important differences lie, identify general trends and extract lessons from particular experiences. An analytical model that works very simply for understanding health systems is that of the Roman house. It starts with a foundation, has three pillars, a stone over those pillars and then a roof. These are the six core elements of any given health care system in the world.


The foundation represents the underlying values or cultural orientations of society. They do not usually change very rapidly. There have been some changes, in Europe for example, where countries have moved away from strong notions of solidarity and equity to somewhat looser notions of markets. It would be a mistake to think these countries have suddenly embraced market competition, but there has been a slight change in values. Canada’s principles are expressed in the Canada Health Act, and are pretty solid (even Quebec’s famous Chaoulli case has not really changed that). Such principles do not tell governments what they should be doing, but limit their scope of action.

Funding, contracting and provision

The three pillars are the funding, contracting and provision of health care and here we see different mixes from one country to the next.


There are only so many options for funding health care: general taxation, social health insurance, private health insurance and out-of-pocket payments. In Europe, we have two dominant models. In the traditional Bismarckian model, which started in the late 19th century as employment-based social health insurance, employees and employers pay into the system. And because the contribution is linked to taxable income, there is implicit cross-subsidization. The other model, typified by Britain’s National Health Service (NHS), was introduced after the Second World War. Lord Beveridge realized that the Bismarckian model left out the self-employed and farmers and was somewhat complicated. The NHS model extended coverage to the entire population and nationalized hospitals, though not doctors. Cross-subsidization takes place via general taxation.

Those two basic models have proven extraordinarily stable over time. But we have also seen processes of hybridization, for example in Germany and Holland, where employment-based social insurance was expanded with population-wide schemes for long-term care. There have been a couple of efforts to change the models, notably when Margaret Thatcher tried to replace the NHS with private health insurance. Within three months, however, she realized that the NHS was the third rail of British politics and had to admit defeat. She left the basic model alone and instead experimented with internal markets, purchaser-provider splits: that was the beginning of the whole generation of new ideas we are seeing around the world, including Canada.

There are two countries in Europe that have really changed their funding model in recent years. Spain replaced Bismarckian employment-based insurance with a tax-based insurance, and adopted a regionalized model. The Netherlands, in 2006, adopted a model for which there is still no legal definition. Citizens are sort of mandated to take basic insurance with insurers who sort of have to be non-profit or a little bit for-profit. The government is now facing embarrassment as the number of uninsured has almost tripled over two years (though the actual number is still very small). The Dutch government started out presenting its plans as privatizing insurance, but realized that international treaties, e.g. those of the International Labour Organization (ILO) and the Council of Europe and the European Union, prohibit member countries from privatizing social security and require governments to cover at least 60-70% of the population with social health insurance. The Netherlands then decided to call the new scheme social insurance. Interestingly, the only authority in the world that can really determine whether The Netherlands now has social or private insurance is the European Court of Justice, which will only rule after someone has gone to court to challenge the Dutch model. No one has done that so far, but eventually someone will.

The biggest constraint on the model is that Dutch citizens, like Canadians, feel that basic health care is a social right. They want equal or at least fair access without financial barriers. This strongly restricts market activity, extra payments, co-payments or preferential care for certain groups — all of that remains in the margin. The basic bloc of health care in The Netherlands is firmly based in notions of social solidarity. Dutch citizens expect government to act if there is a problem.

The single payer systems (either based on general taxation or social insurance with administrative rules that apply to all) have proven to be much better in cost control than any other alternative. Taiwan, after shopping around extensively, decided that national health insurance worked better and was cheaper than quasi-privatized health insurance systems. There are still problems, but these can be solved within national health insurance. The main appeal is that single payer systems enable government, if it is willing to do so politically, to use its collective muscle to deal with the other main actors in the playing field, e.g. the pharmaceutical industry, doctors and hospitals. The use, or threat of using, that power, is one of the best cost-saving devices in the world.


The second pillar contains the contracting and payment models. In the last 20 years we have had extensive debate about shifts in payment models, with proposals to move from fee-for-service to capitation, from capitation to payment for performance. It zigzags all over the place and the lesson is that we all end up with mixed payment models: some notion of capitation, plus some fees for certain services, plus some extra money for doctors to acquire technology. Mixed models appear inevitable and they have emerged all over the world. Moreover, there is always a trade-off between health professionals’ level of income, mode of payment (billing or salaried) and professional autonomy, and it has proven all but impossible to squeeze physicians on all three fronts.

There are a couple of universal shifts. One is the shift to case-based payment, and the emergence of the whole industry of disease-related groups (DRGs). Holland decided to create its own DRG system, which now involves 30,000 DRGs and still only covers about 20% of hospital costs. After 10 years and millions of euros invested in consultancies, the Health Ministry is looking for ways to simplify that system, and plans to aggregate the 30,000 DRGs into 400. The whole industry of designing and implementing DRGs is fraught with difficulties.


Provision is the third pillar of our health care systems. Europe has a very long tradition of non-government health care and ever since the 10th century, churches, monasteries, local communities and lay charities, all non-state organized, were the organizations that set up local hospitals and other community care institutions. These private not-for-profit hospitals were then incorporated into a public system of funding. The local voluntary boards that run these hospitals also have a very long history.

England, the Scandinavian countries and some of the southern European countries nationalized health facilities. Others, like Germany, nationalized some and kept some private. In general, one can argue, it does not much matter whether the provision of care is public, private not-for-profit or even for-profit, as long as they are firmly embedded in regulation under collective funding.

Administration and governance

The big stone on top of those pillars is administration and governance of health care services. In the last few decades, private business notions have crept into the governance of public agencies and hospitals. Until the 1950s, hospitals were run mainly by nurses or by nuns. Then doctor-directors became popular, followed by economic directors, and now tri-board membership (managers, physicians and nurses) has been adopted in many places, though it has not necessarily improved the quality of management. There has been a marked change in the vocabulary of management. Hospital directors have become CEOs and their salaries have tripled. Hospital divisions have become cost centres or profit centres. And administrative costs have skyrocketed. Voluntary boards are being replaced by expensive professional boards. Hospitals (like government departments) now have mission statements.


The roof of the house, finally, is government regulation. There is no dichotomy here between state and markets. The issue with regulation is not whether to regulate, but how. There is no dispute over the need to regulate professionals, safety, quality, and the subsidies to ensure that low-income families have access to health care. Cross-subsidization is perhaps the most important role of government, to make sure that people have access to health care. Some form of tax-based or insurance system is required and there are more and less complicated ways to administer this cross-subsidization. When The Netherlands moved from social insurance to a mandate to purchase semi-private insurance, the tax department had to hire 600 people to verify which families were eligible for low-income family cross-subsidies: 40% of the population counts as low income for this particular subsidy.

Is there a difference?

To summarize, there are a limited number of options in how countries provide health care, and most changes occur not in fundamental principles but in the mechanisms of payment and provision. Despite the volatile rhetoric surrounding these debates, no single model has emerged as the “best” model: most countries end up with somewhat mixed models and compromises are made along the way.