The tale of out-of-pocket spending on health care

It is the best of times and the worst of times. Bulk-billing rates for non-referred attendances (principally general practitioners) have reached over 82%.1 Yet there has been renewed attention focused on the growing financial burden that out-of-pocket (OOP) payments impose on patients.2 The apparent contradiction can be reconciled, but to do that we need to get beyond the headline figures.

All health care financing, whether provided through government, social agencies or private providers, aims to ensure that individuals are not excluded from receiving costly health care when they need it. When patients face charges, their use of health services is lowered, with OOP expenditures having a greater impact on the use of health care by those with less financial means. So at first glance it seems inconsistent to impose copayments in a system set up to reduce barriers to use.

This would indeed be the case if all health care were equally appropriate and equally valuable. But it is not. Copayments are widely used, and not just in health care, to provide a price signal that will reduce demand for those services that are less valued. The problem in health care is that consumers are not always good judges of what is most effective and may reduce their use of needed care. This finding, established by the RAND Health Insurance Study some three decades ago,3 has not been seriously challenged. Further, copayments that reduce costs in the short run may increase them in the longer term, as shown, for example, in the United States, where charges for pharmaceuticals resulted in patients not purchasing essential drugs, subsequently leading to hospital care.4

Another approach is to develop a process for assessing the social value of health care interventions, and subsidising those services that provide more value relative to the societal willingness to pay. The Pharmaceutical Benefits Advisory Committee process is a good example. New drugs are assessed for their safety, effectiveness and cost-effectiveness, and those with a cost for a quality-adjusted life-year that falls below the value threshold are more likely to be included in the Pharmaceutical Benefits Scheme. Individuals who judge a particular drug to be of greater value to them are still (in the Australian system) free to buy it, but that purchase is not subsidised by the taxpayer. Clearly then, it is important to understand who is paying and for what in the Australian health care system. And the article by Yusuf and Leeder in this issue helps answer that question.5

The largest component of Australians’ OOP costs is private health insurance (PHI) premiums. PHI is already subsidised by the taxpayer and, at least for hospital treatment, duplicates free cover. It is hard to argue then that this is something other than a discretionary expenditure, particularly when it is cheaper to buy subsidised PHI than to pay the Medicare levy surcharge.6 Prescription (subsidised) medicines are a relatively small component of OOP costs. Costs of other medicines and supplies are relatively large, but it is hard to determine the level of discretion. This category covers various items, from over-the-counter asthma medications to vitamin pills and herbal tonics. Dental fees are also large, so not surprisingly, low-income groups use less dental care overall.7 But these services range from major corrective work (high on appropriateness) to purely cosmetic benefits (high on discretion). Similarly, high OOP costs for the services of specialists and inadequate public outpatient services work to exclude poorer households from this type of care.8

The results reported by Yusuf and Leeder still need to be interpreted with caution. The estimates are imprecise for many items. It is not clear whether the OOP costs exclude rebates, or whether respondents have counted the cash they outlaid before claiming from Medicare. The results do not show distributions of costs or the number of households with zero or minimal OOP costs, but high averages can be driven by a few big spenders.9

So this paper throws some light on those services for which Australians are paying from their own pockets, but there are still plenty of dark corners. Further, the policy response to reduce inappropriate OOP costs is not straightforward. Simply providing higher subsidies may have some effect on patients’ expenses but it can also lead to higher charges from groups that have the market power to raise fees — as experience with the Extended Medicare Safety Net has shown.10

The headline figures on rising OOP payments by Australians for their health care certainly are important in drawing community and policymaker attention to a phenomenon, but it is important to recognise that this is a phenomenon, not necessarily a problem. As Australia’s health care spending is predicted to continue to rise, more funds will be needed to support the health system. No country can afford to finance all the health care that is possible for everyone. So universal access is no longer envisaged as unlimited access for all to everything, but rather as appropriate access for all to cost-effective and socially valued services. We need a policy debate about what should be in that cost-effective and valued minimum package.