Does reporting heterogeneity bias the measurement of health disparities?

Abstract

Heterogeneity in reporting of health by socio-economic and demographic characteristics potentially biases the measurement of health disparities. We use anchoring vignettes to identify reporting heterogeneity in self reports on health for Indonesia, India and China. Correcting for reporting heterogeneity tends to reduce estimated disparities in health by age, sex (not Indonesia), urban/rural and education (not China) and to increase income disparities in health. Overall, while homogeneous reporting by socio-demographic group is significantly rejected, the results suggest that the size of the reporting bias in measures of health disparities is not large.

Subjective socioeconomic status and health: Relationships reconsidered

“…For example, mental health and negative affect are likely predictors of both subjective socioeconomic status and physical health outcomes, particularly those that are self-reported and represent perceptions of health versus biological indicators of functioning (Bago d’Uva et al 2008; Gabarski 2010; Powadthee, 2007). Nevertheless, several studies have shown that the correlation between subjective socioeconomic status and reported health, if lessened in magnitude, persists in the presence of depression and affect controls (e.g.…”

“…The population’s health profile shares many patterns with countries around the world. Inequality in health conditions follow class lines; an expenditure and education gradient exists for numerous outcomes, including self-rated health and ADL indicators of functioning (Bago d’Uva et al 2008; Stoddard 2006). As in other populations, self-rated health predicts later-life morbidity and mortality (Frankenberg and Jones 2004).…”

“…Alternative explanations reference difficult-to-measure characteristics of families and individuals, such as educational quality and asset networks that may be captured more fully in subjective assessments. Other alternatives include reverse causation, running from health to perceived status, and personality-based response bias that may drive both perceptions of status and health outcomes (Bago d’Uva et al 2008; Garbarski 2010; Singh-Manoux, Marmot, and Adler 2005). Schnittker and McLeod summarized this perspective, writing: “researchers cannot fully understand the mechanisms behind these effects without adequately understanding from where individuals derive their senses of status” (2005:84).…”

Socioeconomic Inequalities in Health in China: A Reassessment with Data from the 2010–2012 China Family Panel Studies

“…For example, in a regional sample of three Chinese provinces surveyed during 2000–2001, Bago d’Uva, Van Doorslaer and colleagues (2008) found evidence of reporting heterogeneity in self-rated cognition, pain, self-care, and daily activity by education, income, and rural-urban residence. In a nationally representative sample interviewed in 2012, Xu and Xie (2015) found systematic variation by socioeconomic and demographic characteristics in thresholds used by respondents in rating their overall health status.…”

Slipping Anchor?

“…Applications of the methodology are increasing rapidly and now cover a wide range of topics including political efficacy (King et al 2004), work disability (Kapteyn, Smith, and Van Soest 2007), job satisfaction (Kristensen and Johansson 2008), life satisfaction (Christensen et al, 2006), health (Bago d’Uva et al 2008; Bago d’Uva, O’Donnell, and Van Doorslaer 2008) and health system responsiveness (Rice, Robone, and Smith 2010). These studies typically claim to reveal substantial reporting heterogeneity and therefore important impacts of vignette corrections on the comparisons of interest.…”