The Validity and Reliability of Dichotomized Self-rated Health Under Different Cutpoints

Abstract Self-rated health is a widely used indicator of overall health status. It is most often reported on a Likert scale of three to five values in surveys. To facilitate presentation and interpretation, it is common practice to simplify the variable by dichotomizing it; however, there has been little documented reflection on how this should be done. This paper explores all four possible dichotomizations of self-reported health, taken from three years of the Canadian Community Health Survey and reported by a Likert scale. We evaluated each dichotomization stratified by sociodemographic variables. We use regression analysis to explore the validity and reliability of all four possible dichotomizations by mapping them to the Health Utility Index. We found that lower cutpoints of dichotomization capture more pronounced differences in health status and are more consistent across sociodemographic variables. However, higher cutpoints of dichotomization should be considered for small data sets. About the Research Department The Saskatchewan Health Authority Research Department leads collaborative research to enhance Saskatchewan’s health and healthcare. We provide diverse research services to SHA staff, clinicians, and team members, including surveys, study design, database development, statistical analysis, and assistance with research funding. We also spearhead our own research programs to strengthen research and analytic capability and learning within Saskatchewan’s health system. About the UPHN The Urban Public Health Network (UPHN) is a national organization established in 2004 which today includes the Medical Officers of Health in 24 of Canada’s large urban centres. Working collaboratively and with a collective voice, the network addresses public health issues that are common to urban populations. Research operations of the UPHN are conducted in partnership with the University of Saskatchewan. Disclaimer This working paper is for discussion and comment purposes. It has not been peer-reviewed nor been subject to review by Research Department staff or executives. Any opinions expressed in this paper are those of the author(s) and not those of the Saskatchewan Health Authority. Suggested Citation Charles Plante, Sharalynn Missiuna, and Cordell Neudorf. 2024. “The Validity and Reliability of Dichotomized Self-rated Health Under Different Cutpoints.” medRxiv. Extended Abstract Introduction Self-rated health is a widely used indicator of overall health status. It is most often reported on a Likert scale of three to five values in surveys. To facilitate presentation and interpretation, it is common practice to simplify the variable by dichotomizing it; however, little documented reflection has been done on how this should be done. Methods We use regression analysis to explore the validity and reliability of all four possible dichotomizations of self-reported health in the Canadian Community Health Survey in 2013-2015 by mapping them to a validated health measure: the Health Utility Index Mark 3 (HUI). We posit that more valid cutpoints in self-rated health are associated with larger changes in HUI. We posit further that more reliable cutpoints are associated with similar changes across sociodemographic variables, including age, sex, education, marital status, geography and income. We also provide descriptive statistics to contextualize our analysis. Results The greatest proportion of respondents reported having “very good” health, although the proportion of the population reporting “excellent” or “very good” health decreased with age. Similarly, Canadians tend to score highly in HUI. Our regression results suggest that HUI tends to be higher for younger, richer, married, educated and urban populations. However, these associations are muted as the cutpoint used to dichotomize self-reported health is raised. The model with the lowest cutpoint, distinguishing between poor health and all other health statuses, was associated with the greatest and most consistent negative changes in HUI among different sociodemographic groups. Conclusions Dichotomizing self-rated health using lower cutpoints captures more pronounced differences in health status measured by HUI and tends to capture more consistent differences across sociodemographic variables. That is, lower cutpoints produce more valid and reliable results. However, lower cutpoints isolate less commonly reported health levels and may lead to less accurate results in smaller populations. Key Points This article addresses the knowledge gap concerning the most accurate way to dichotomize self-rated health data reported using a Likert scale. This paper explores the validity and reliability of all four possible dichotomizations of self-reported health reported by a Likert scale. Lower cutpoints of dichotomization capture more pronounced differences in health status and are more consistent across sociodemographic variables. Higher cutpoints of dichotomization should be considered for small data sets.