Copayments and the evidence-base paradox

When is evidence merely opinion? The evidence-base paradox is particularly relevant to the GP copayment

The proposed introduction by the Australian Government of a copayment for visits to general practitioners has received much attention. In academic journals and op-ed pages, evidence is cited that supposedly shows that the copayment is a “bad idea”.1 However, is the evidence cited relevant to this particular policy of a copayment of $5 to $7?

The RAND Health Insurance Experiment

From 1971 to 1982, the RAND Health Insurance Experiment (HIE) in the United States randomly allocated families to health insurance plans with zero, 25%, 50% or 95% copayments.2 The study found that “Cost sharing in general had no adverse effects on participant health”. This finding might support a copayment. However, the study also provides evidence against copayments by emphasising the finding that “The poorest and sickest 6 percent of the sample at the start of the experiment had better outcomes under the free plan for 4 of the 30 conditions measured”.

Participants in the RAND HIE were also grouped by the maximum percentage of their income that they were expected to contribute. Thus, even the poorest in the RAND study could be charged copayments ranging from 5%–15% of their gross income, capped at up to US$4000 per year in today's money (whichever was less). So, is this finding about the poor from the RAND study relevant to a proposed $7 copayment, capped at $70 for the year for holders of concession cards?

Longer term consequences of the copayment

It has been theorised that a copayment will cost the federal government more money in the long run; that it would prevent people from getting necessary outpatient care, leading to a need for more expensive inpatient care. To support this, a 2010 study is often quoted.3 However, this study was limited to people aged over 65 years, and participants already paid copayments which thereafter increased for both primary and specialist care by between 33% and 150% with no yearly cap.

Importantly, a 2014 study specifically looked at the impact of cost-sharing by patients on “low-income populations”.4 The results of this study “largely confirm the conclusions of the RAND Health Insurance Experiment” in regard to the impact of copayments on demand for health care, but specifically found “no statistically significant evidence for ‘offset effects' that would indicate that reduced use of outpatient services led to increased demand for hospital services”.4 Therefore, the data are mixed. Offset effects might then depend on other policy settings.

Impact on affordability and access to health care

Another widely quoted finding is that 16% of Australians are already unable to see a doctor because of the cost. This implies that people are already going without necessary care, and that this will inevitably be exacerbated by the copayment. This claim is based on a survey of health consumers in 11 countries,5 and to be included in that Australian 16%, individuals had to have, once in the past year, not filled a prescription, not visited a doctor while experiencing a medical problem, and/or not received recommended care because of cost. But is this finding relevant to the currently proposed copayment?

Twenty-five per cent of Australian survey participants already had to pay at least $1000 out-of-pocket for health-related expenses for the year. Therefore, the 16% would include private patients baulking at an out-of-pocket expense. Furthermore, this same survey showed that 14% of Australian participants waited 6 days or more for an appointment to see a doctor or nurse when they were sick or needed care, and half waited more than a month and 18% more than 2 months to see a specialist. Presumably this percentage would be higher for public patients.

Waiting times are a product of a system that is overwhelmed. Waiting leads to loss of economic productivity and quality of life and even death.6 It could be argued that the waiting times listed above indicate a failure to have a truly universal health care system in Australia; more so than would the introduction of a copayment. A copayment might reduce demand2,4 and allow resources to be allocated to reducing waiting time. Thus, accusations that a copayment is immoral7 are unjustified.

Impact on outcomes

A Consumers Health Forum of Australia report acknowledges that “there is very little robust research on the impact of co-payments on the longer term health outcomes of consumers. This makes it very difficult to draw any conclusions about the overall impact of co-payments on health status as the relationship between access to health care, the provision of care and health outcomes is very complex”.8 Particularly relevant then is the Oregon Health Insurance Experiment, in which low-income adults with no health insurance at all (which would require an extreme degree of copayment) were randomly selected (from a state ballot) to receive government health care with “relatively comprehensive benefits with no consumer cost sharing”.9 Those who received free health care predictably increased their use of health services, including primary and preventive care, but this did not lead to actual improvements in basic objective health indicators such as blood pressure, cholesterol levels or diabetes control.10 Despite increased access to primary and preventive care, those with insurance had a 35% increase in hospital admissions10 and a 40% increase in emergency department visits.11 These data are also relevant to studies such as the RAND study cited above, in which it was found that both necessary and unnecessary health care were reduced;2 that is, the RAND HIE demonstrates the complexity of predicting what really is necessary care in relation to actual health outcomes.

Where lies the truth?

Health care is complex, entwined with biology, clinical medicine, politics, ideology, sociology, psychology and patient belief systems. The copayment is an example of what has been described as the fact-checking paradox (or evidence-base paradox);12 that is, the more something really needs the scrutiny of fact-checking, the less it is possible to impartially decipher the facts free from ideological bias.