O brother how art thou: Propensity to report self-assessed unmet need

Abstract

Research investigating self-assessed unmet need (SUN) has taken the reports from surveys as given and subsequently attempted to discover patterns in inequality of access to healthcare. This requires the yet untested assumption that, given a certain level of care and demand, the likelihood of reporting unmet need does not vary across socioeconomic/demographic status (SEDS), be satisfied. Using an administrative dataset spanning 2001 to 2011 comprised of sufferers of a set of conditions that suggest unmet need (n = 3300) we evaluate the proposition that, given health status and care received, the propensity to report unmet need does not vary along SEDS. The results are further validated using the Canadian Community Health Surveys between 2001 and 2013 (n = 237,483). We find that the assumption of independence between reporting SUN and SEDS is not satisfied. Many of the groups found to have less access in previous studies may simply be more prone to interpret/answer the survey questions about unmet need in a certain way. The results of this research suggest that, in its present incarnation, survey data on self-assessed unmet need does not accurately measure what much of the academic literature has assumed it does.

Keywords: Inequality; Response bias; Unmet need.