Studies on health insurance coverage often rely on measures self-reported by respondents and this is the case in both developed and developing countries. The presumed accuracy of survey reports of health insurance enrolment influences how these data are used for health policy evaluations, yet the accuracy of such measures outside the United States has not been thoroughly validated. This paper aims to fill that gap in the literature by presenting the first evidence on the extent and factors associated with accuracy of private health insurance (PHI) coverage reporting in an Australian context.
This paper uses linked Australian National Health Survey and administrative population tax data to explore the accuracy of self-reported private health insurance coverage in survey data. We find that 9% of individuals misreport their PHI coverage status, with 5% of true PHI holders reporting that they are uninsured and 16% of true non-insured persons self-identifying as insured. Our results show reporting errors are systematically correlated with individual and household characteristics, including age, migration status, education, marital status, employment status and income. We additionally find that most of these characteristics influence the probability of giving a false negative or a false positive report very differently. Our evidence on the determinants of errors is supportive of common reasons for misreporting. We directly investigate biases in the determinants of PHI enrolment using survey data. We find that, as compared to administrative data, survey data depict a quantitatively different picture of PHI enrolment determinants, especially those capturing age, language proficiency, labour force status or the number of children. We also show that PHI coverage misreporting is subsequently associated with misreporting of reasons for purchasing PHI, type of cover and length of cover.
This study finds that reporting accuracy of PHI coverage is quite high in a nationally representative health survey in Australia, providing some good news for studies using such survey data to document PHI coverage. Our evidence of the factors associating with PHI misreporting may provide useful insights for the constructors of surveys to consider in order to improve accuracy of responses to PHI-related questions. Our finding of a substantial relationship between PHI coverage misreporting and a range of explanatory variables indicates that reporting errors of PHI enrolment in survey data are non-classical. These non-classical errors suggest complicated biases in other studies that use self-reported PHI enrolment as an independent variable in regressions, including those evaluating effects of PHI enrolment on health care utilization and health outcomes.