Health Care Cost-Sharing Produces Different Outcomes For Different Populations
Higher health insurance cost-sharing can reduce the use of medical services without affecting health outcomes for most workers, but people at high medical risk are negatively impacted by high co-insurance rates, especially if they are low-income, according to a study released by the Kaiser Family Foundation.
The report, by Jonathan Gruber of MIT, is based on an analysis of the findings of the RAND Health Insurance Experiment (HIE), as well as more recent evidence on the impacts of patient co-insurance. The HIE was a large-scale social experiment started by the federal government in 1974, involving 2,000 non-elderly families. These families were randomly assigned health insurance plans with widely varying co-insurance and maximum out-of-pocket expenditure amounts. Participants in the experiment were followed for up to five years after enrollment.
Results of the study showed that the co-insurance rate of a given plan has a strong effect on the likelihood that participants will use any medical services, Gruber noted. The number of physician visits and total outpatient expenditures fall as co-insurance rates rise, though there is no clear correlation between inpatient utilization and co-insurance rates, Gruber said. Compared with free care, total medical expenditures were found to be 15% lower among participants in the 25% co-insurance plan, and 30% lower among enrollees in the 95% co-insurance plan.
According to Gruber, higher co-payment rates appear to reduce the use of both effective and ineffective care by the same amount. However, he added, the study also found significant reductions in preventive care.
On average, the health outcomes of participants in the various plans were roughly the same over the period studied, though differences were observed in specific areas of health, such as blood pressure, vision, and hearing. "This result is quite powerful," Gruber said. "It suggests that, at least at the time of the experiment, the typical enrollee in the study was on the 'flat of the medical effectiveness curve,' the portion where additional care was not buying medically effective care. Thus, care could fall significantly without adverse health consequences for the average person."
Differences do, however, emerge when the HIE data are divided into samples of those with high and low risk of illness, Gruber said. Results showed that people at high risk of developing illnesses are more likely to have negative health outcomes when enrolled in the high co-insurance plans. This effect was found to be especially strong when participants are low-income.
Gruber cautioned that these findings are not directly applicable to current health plan design for several reasons: The plans in the study imposed maximum limits on out-of-pocket costs depending on the income of enrollees, the three- to five-year time frame of the study was not sufficient to determine the long-term effects of varying degrees of preventive care utilization, and the treatments available in the 1970s were less effective than those offered today.
However, Gruber said, the findings suggest that co-insurance policies can be actively targeted to promote effective and efficient health care use. Results indicate that caps on service utilization can do more harm than good, and that high-deductible plans are inappropriate for low-income consumers.
