Healthcare expenditures are steadily growing worldwide. Between 2004 and 2014, the World Bank reckons that the average share of health expenditures of high-income countries has increased from 10.9 to 12.3 per cent of gross domestic product (GDP). Singapore is no exception as its share of health expenditures has increased from 3.3 to 4.6 per cent between 2011 and 2015.

There are multiple drivers of healthcare inflation: spending can rise due to an increase in price, or in quantity consumed.

An important factor is the development of new and expensive health technologies such as biologics - extremely costly drugs revolutionising the treatment of rheumatoid arthritis, cancers and diabetes. Another factor is rising wages, fuelled in part by competition between private providers to attract the best healthcare professionals.

The quantity of care consumed in Singapore is also increasing. While population ageing is often cited as a cause, lifestyle risk factors, such as eating too much and exercising too little, also contribute to the rise of costly chronic conditions.

Another factor worth mentioning is "physician-induced demand" where some doctors take advantage of patients' lack of medical knowledge to provide them with more care than necessary, or to charge a higher fee.

Health insurance plays a central role in influencing healthcare consumption. By significantly decreasing the price of care at the point of consumption, insurance plans increase the quantity of care consumed. Part of this increase is actually a good thing as it reflects better access to healthcare, especially for lower-income individuals who could not afford treatment otherwise.

However, by distorting the price of care, health insurance may induce overconsumption. Such moral hazard is not the only adverse effect as health insurance also reduces patients' incentive to shop around for the best-priced treatments and providers in the market. Physician-induced demand is also facilitated, as patients have little financial incentive to question the care they receive when it is covered by their health insurance.

This illustrates the subtle balancing act of health insurance: While some coverage is good, too much is detrimental. Overconsumption of care puts a financial burden on insurance companies which, in turn, need to increase their insurance premiums.

Patients who pay with insurance may feel that their care is "free" but it is of course not so - the insured will ultimately foot the bill through higher insurance premiums.

The seminal study of health insurance conducted more than 30 years ago in the United States by Professor Willard G. Manning, my former mentor, showed that introducing insurance co-payments substantially decreased health expenditures with only very few adverse health effects. Since that study, all public insurance schemes have included a combination of payout limits.

In Singapore, the balancing act is achieved by setting the level of insurance benefits, imposing a deductible, insisting on co-payment by the patient, and introducing claim limits per year.

Singapore has Medisave, which is workers' individual health savings account, which can be used for inpatient and some outpatient care. With Medisave in place, the role of basic health insurance, MediShield Life, has historically been to protect against catastrophic conditions, such as hospital treatment due to disease or accidents. Payouts are pegged to subsidised rates in public hospitals.

That is why MediShield Life coverage includes inpatient care; and only selected outpatient services, such as chemotherapy and kidney dialysis. There is a deductible ($1,500 to $3,000), which is the amount the insured must pay before the insurance payout kicks in, while the co-insurance rate (3 to 10 per cent ) is the fraction of the bill that the insured needs to co-pay after the deductible is reached. MediShield Life also has a claim limit of $100,000 per policy year.

Integrated Shield Plans, which cover non-subsidised healthcare expenditures, are also required to have deductibles and co-insurance. The purpose of these features is to reduce the moral hazard problem by having patients pay a share of the hospital bill, and in this way manage health expenditure.

Extending coverage, lowering the deductible and co-insurance rates, and raising claim limits in an effort to improve access to the health system would also result in higher levels of health expenditure.

SETTING LIMITS

The hard truth is that there is very little data on how to optimally set the level of these parameters. These parameters vary internationally considerably. For instance, France's deductible lies below $2 per sickness episode, while Switzerland's maximum annual deductible of $3,500 is higher than Singapore's.

For co-insurance rates, Germany's ranges from 5 to 10 per cent, while France's ranges from 30 to 40 per cent.

Lastly, most European countries do not have a claim limit but, instead, cap the amount of cost sharing. In other words, the risk exposure of the insured person (instead of the insurer) is capped.

However, even in the absence of relevant data and precise international guidance, there are some guiding principles.

PRINCIPLES OF GOOD INSURANCE DESIGN

• First, co-payments are features of good insurance design

Given that policyholder co-payment encourages prudent healthcare decisions, their optimal level is unlikely to be zero. Having an insurer pay 100 per cent of a patient's bill is thus hardly conducive to encouraging prudent spending.

In this respect, I would question the wisdom of having private insurers offer "rider" plans that pay out claims on an "as charged" basis. These in effect fully insure individuals against the out-of-pocket co-payment requirement of Integrated Shield Plans claims.

Co-payments tie patients' interest to a lower healthcare bill and act as a check to promote prudent health spending. Permitting insurance to remove that cash co-payment component reduces that check on prudence. Not surprisingly, claims under Integrated Shield Plans have soared in recent years, sparking sharp increases in premiums.

• More generous for essential services

Another principle is that insurance coverage can be relatively more generous for essential healthcare services, as the risk of moral hazard is lower for non-elective treatments. For instance, co-payments might be lower for outpatient care for chronic conditions or some inpatient care. But more research is needed to determine if lower co-payments result in overconsumption.

•Assessing value-for-money care

With its newly founded Agency for Care Effectiveness, the national health technology assessment centre, Singapore will be better able to determine the value for money of healthcare, in terms of life and quality of life gained.

Knowing the health benefits from health services relative to their cost will open the possibility of modulating MediShield Life's co-insurance rates according to the value of the services covered. Such value-based insurance design would make it possible to strategically determine the right level of coverage that maximises the health benefits for the population.

Singapore is in an enviable position, spending less on healthcare than other rich countries, yet enjoying good outcomes. However, addressing the challenge of increasing healthcare costs and health insurance premiums will require finding innovative solutions. By spending wisely on the most cost-effective care, value-based insurance is a promising one.

• Marcel Bilger is assistant professor, Health Services and Systems Research Programme at the Duke-NUS Medical School.