Primary care delivery, risk pooling and economic efficiency

See Missialos and Dixon [16] for discussion on health-care funding in Europe and Greβ, Delnoij and Groenewegen [13] for the effects of financial incentives on behavior on primary care providers.

In the UK, the primary care trusts contract services of self-employed general practitioners who act as gatekeepers and commission services from specialists in the publicly funded NHS system. In Germany, health care is funded by social health insurance where patients are, in principle, free to choose providers, but the family doctor model is encouraged. In Switzerland, individual’s health-care expenses are covered by compulsory health insurance, and patients are free to choose providers in either the public or private sectors. In Australia, health care is funded by a universal system of health insurance, known as Medicare, where health care is delivered mainly by self-employed providers. In Singapore, a medical savings system is adopted; the government subsidizes health care, and patients are free to choose providers within the government or private health-care delivery system. Taiwan adopts a National Health Insurance system for funding health care [6]. In Hong Kong, primary care is mainly provided in the self-regulated private market funded by private spending, while hospital care is mainly provided by the public sector which is heavily subsidized by public funding. In the United States, private health insurance is a primary source of funding for health care. Gaynor [12] and Dranove and Satterthwaite [8] discuss the theoretical and empirical studies on the organization of the health-care market.

The use of primary care here differs from the usual one, which refers to the services provided by general practitioners and other allied health-care professionals by excluding specialist care, as adopted by many medical professionals. To define primary care as the first point of contact of consumers with health-care services does not exclude consumers from directly consulting specialists and receiving non-hospital-based care. Blomqvist and Leger [3] consider the choice of consumers between outpatient care (defined as primary care) and inpatient specialist care (defined as hospital-based care) under the influence of insurance without the endogenous specialization of service providers.

See McGuire [15] for more discussions on provider agency problems.

To alleviate the problem of demand-side ex ante moral hazard, co-payment and deductibles as well as supply-side measures are commonly built in health insurance, as Feldstein [10] and Pauly [19] describe. Zweifel and Manning [27], on the other hand, provide a survey on the literature of moral hazard. Medical savings accounts are sometimes considered as an alternative to limit ex ante moral hazard, according to Hsiao [14]. If health services are publicly funded and provided, then user changes may play a dual role for generation of revenue and alleviating the demand-side moral hazard problem, opines Robinson [21].

Blomqvist and Leger [3] consider the case in which a patient chooses the quantity of care based on his subjective belief of the illness condition and the recommendation of the doctor.

This contrasts with the analysis of moral hazard by Pauly [18] in which consumers decide how much they shall be receiving from providers. Pauly also models that the extent of the moral hazard problem is limited by the illness conditions that determine the price elasticity of demand for treatment. In particular, if the price elasticity of demand for treatment is small, then the problem of moral hazard will be largely reduced. On the other hand, if the price elasticity of demand for the treatment is large, then the severity of the moral hazard problem will be increased.

This view of demand inducement differs from the classical view of Dranove [7] and Folland et al. [11], which models supplier-induced demand similar to the effect of advertisement that raises demand at every price of health care.

Pauly and Ramsey [20] assume that consumers know their illness condition, and the provider might supply what consumer demand for analyzing the demand-side moral hazard problem. The setup here follows Chalkley and Khalil [4] in which providers know more about the illness condition and then allowed for the provision of unnecessary treatment and stinting by providers.

Recently, Liu [25] considered a situation in which conscientious profit maximizing and pure profit maximizing providers co-exist and shows interestingly that consumer welfare depends non-uniformly on the fraction of the conscientious providers.

In the presence of search cost, Salop and Stiglitz [23] show that if prices charged are not observable, then all suppliers will charge at monopoly price or there will be multiple equilibria. To avoid complications, this paper assumes observable prices.

Medical boards, associations, councils and colleges are the regulatory authorities overseeing service providers’ conduct by accepting consumer’s complaints, investigating and exercising disciplinary actions, although they are sometimes questioned by the general public. In the UK, there are seven such medical councils that regulate different specialties of services providers, and the General Medical Council regulates doctors. In the US, medical boards of different states accept complaints, investigate and impose disciplinary actions. In Australia, medical boards play a similar role. In Canada, colleges of physicians and surgeons in different provinces have the power to suspend and revoke licenses, as well as to charge fines, if an allegation is substantiated after hearings. In Queensland, Australia, for example, there were 80 complaints received concerning inadequate treatment in 2007–2008 according to the Annual Report of the Medical Board of Queensland.

The parameter k can also be interpreted as a search cost, as in Wolinsky [26]. One can introduce another parameter s to replace k as a measure of disutility of illness if k is interpreted as a search cost on each visit. As a measure of disutility from illness, the assumption of a parameter k invariant to illness severities is necessary if a consumer can discern its level. If there are high and low ks corresponding to the two types of illness severity, then there is no asymmetrical information between consumers and service providers. Hence, the first-best treatment and full insurance will always result if providers can be punished. If the consumer cannot tell whether a disutility is of high or low level, then k can change with illness severities, which has the same implication as having an invariant k.

The choice of modeling can be rationalized by arguing that a third-party wishes to ensure the accessibility and affordability of health care by balancing the affordability of health insurance and the out-of-pocket cost at the point of services received, which is often the concern of governments and private health insurers. Consumer’s welfare maximization is naturally a plausible benchmark for modeling the objective of a third party.

For example, the set of measures may include networking, case and disease management, patient education, preventive programs, and utilization management and review, etc. The paper considers several that mainly tackle the supply-side and demand-side moral hazard problems.

Most third-party payers engage in negotiation with service providers for better prices and charges because service providers do have market power because of the imperfect flow of information in the provider markets, which are discussed in the “Salary payment model.”

Wolinsky [26] describes a special case in which there is no insurance. By setting ϕ _H  = ϕ _L  = 1, then whether condition (8) to be discussed below is satisfied will determine which equilibrium will prevail. However, consumer welfare will be lowest among other equilibria discussed below due to a lack of insurance.