The Effect of Coinsurance on the Health of Adults: Results from the RAND Health Insurance Experiment
Jan 1, 1984
Do children whose families bear a percentage of their health care costs reduce their use of ambulatory care compared with those families who receive free care? If so, does the reduction affect their health? To answer these questions, the author randomly assigned (for a period of three or five years) 1844 children aged 0-13 to one of 14 insurance plans. The plans differed in the percentage of their medical bills that families paid. One plan provided free care. The others required up to 95 percent coinsurance subject to a $1000 maximum. Children whose families paid a percentage of costs reduced use by up to a third. For the typical child in the study, this reduction caused no significant difference in either parental perceptions of a child's health or in physiologic measures of health. Confidence intervals are sufficiently narrow for most measures to rule out the possibility that large true differences went undetected. Nor were statistically significant differences observed for children at risk of disease due to prior conditions. Wider confidence intervals for these comparisons, however, mean that clinically meaningful differences, if present, could have been undetected in certain subgroups.