Testing the unique effect of educational quality on BMI and central adiposity in adulthood and hypothesized mediators
Major disparities in health by socioeconomic status persist in the Unites States despite enormous health care expenditures and decades of codified commitment to reducing health disparities (e.g., Healthy People 2010, 2020, and 2030; National Center for Health Statistics, 2010). Education is one indicator of socioeconomic status that has been consistently associated with multiple indicators of health, including cardiometabolic disease and risk factors for poor cardiometabolic health such as greater BMI and central adiposity (Hamad et al., 2018; Hamad et al., 2019). However, common measures of education used to inform policy and interventions on health outcomes may vastly underestimate the importance of education. The present study examines individual-level differences in educational quality and physical health outcomes beyond traditional measures of education in two diverse community samples. Additionally, we tested multiple mechanisms hypothesized to explain why increased education may result in better physical health.
Does Educational Quality Matter?
Although empirical evidence has largely examined educational attainment (e.g., years of education or highest degree attained) in relation to health, educational quality is likely a better measure of the construct of education in health research. Preliminary evidence suggests incremental predictive utility of educational quality above and beyond educational attainment (Glymour & Manly, 2018; Mirowsky & Ross, 2003). Educational quality might be a stronger predictor of health compared to assessments of educational attainment because it more directly measures cognitive and skill-based benefits of both formal and informal education that are thought to confer benefits to health (e.g., Sisco et al. 2015). For example, education is hypothesized to influence cardiovascular risk in part through the acquisition of skills and abilities that lead to better income and employment, increased health literacy and general knowledge, and greater self-efficacy and perceived control (Feinstein, 1993; Ross and Wu, 1995; Matthews & Gallo, 2011). Not all individuals that attain a higher level of education also experience the same gains in such skills and abilities. Further, educational quality may be a more useful measure for understanding educational disparities in health across race and socioeconomic status in the United States, given that similar quantities of educational attainment (e.g., years in school) tend to result in lower educational quality for racial and ethnic minority students (Kirsh, Jungeblut, Jenkins, & Kolstad, 2015). As Sentell and Zhang (2015, p.244) put it, “Many minority adults thus get less ‘education’ for their years of schooling,” suggesting that measures of educational attainment do not fully capture racial and ethnic disparities in schooling that may influence health.
To date, much of the small empirical literature on educational quality and health has focused on cognitive health outcomes rather than physical health, despite the well-established association between education and physical health (e.g., Dotson, Kitner-Triolo, Evans & Zonderman, 2009; Manly, Touradji, Tang, & Stern, 2003). For example, improvements in educational quality at the state-level (e.g., student-teacher ratio) have been associated with reduced risk for dementia (Nguyen et al., 2016), and educational quality assessed using an individually assessed word-reading subtest has been associated with memory decline even after controlling for educational attainment (Manly et al., 2003). Current research also tends to focus on differences in educational quality between groups as broad context or policy differences (e.g., funding, length of school days, racial segregation) or examines the implementation of high-quality early childhood education such as Head Start. While these studies inform whether educational quality may be associated with different outcomes across groups of learners, they cannot speak to individual level outcomes or the psychobiological processes that may help explain how educational quality can protect against poor health, perhaps even over and above educational attainment. An individual and objective measure of educational quality might fill such gaps.
Theoretical Mechanisms Connecting Educational Quality to Cardiovascular Health
Education is thought to influence cardiovascular health through multiple pathways including perceived control, executive functioning, and health literacy (Cohen & Syme, 2013; Mirowsky & Ross, 2003). Perceived control, the belief that one’s actions can directly influence life outcomes, has been consistently linked to better health, physical functioning, and lower mortality risk (Infurna et al., 2013; Infurna et al., 2011; Turiano et al., 2014). Further, perceived control appears to mediate associations between education and poor health, including outcomes of mortality, incident disease, and biological risk factors (Bosma, 2005; Orton et al., 2019; Zilioli, Imami, & Slatcher, 2017). Education is hypothesized to increase perceived control through the development of analytic and communication skills (Mirowsky & Ross, 1998). Self-reported perceptions of control also are more predictive of outcomes than actual control, perhaps because feeling in control is associated with lower psychological stress (Lachman et al., 2015; Ross & Wu, 1995).
Education may also influence cardiovascular health through executive functioning. Executive functioning is a set of higher order cognitive processes that involve reasoning, problem solving, and the planning, organization, and execution of behavior (Miyake et al., 2000; Suchy, 2009; Williams & Thayer, 2009). Executive functioning is associated with educational quality (St Clair-Thompson & Gathercole, 2006; Thorell, Veleiro, Siu, & Mohammadi, 2013), and educational quality has been shown to be a better predictor of executive function than educational attainment in an ethnically diverse cohort (Manly, Schupf, Tang, & Stern, 2005). Deficits in executive functioning have also been implicated in the onset and progression of multiple diseases, including cardiovascular and metabolic diseases. Weight and the behavioral complexity and inhibition necessary to manage weight in American society has been suggested as a likely mediator of the association between executive function and cardiometabolic health (Gettens & Gorin, 2017).
Lastly, educational quality may also influence cardiometabolic health, in part, through health literacy. Health literacy is the ability to obtain and understand basic health information and services (National Center for Health Statistics, 2009). General literacy and analytic skills are fundamental to the development of health literacy. Greater health literacy is associated with more education and higher educational quality (e.g., Martin, et al., 2009; Mõttus et al., 2014) and lower morbidity including from cardiometabolic diseases (e.g., Koh et al., 2012; Marzec et al., 2015). Poorer health literacy is thought to contribute to poorer health, in part, because it is associated with poorer comprehension of health care services and outcomes as well as poorer health behaviors (DeWalt et al., 2004). Additionally, educational quality appears to be more closely related to objective measures of health literacy than educational attainment (Dotson et al., 2009; Manly et al., 2003).
The Current Study
The current study tests whether educational quality is associated with BMI and two measures of central adiposity (visceral fat, waist circumference), all of which are risk factors for poor cardiometabolic health (Klein et al., 2007; Sironi et al., 2012). We examine whether effects remain after adjustment for demographic factors as well as educational attainment, and we test whether perceived control, executive function, or health literacy mediate the association between educational quality and our health outcomes.
To validate our findings, two independent samples were used. An exploratory analysis used data from the Alabama Brain Study on Risk for Dementia (ABSoRD; McDonough, Letang, & Stinson, 2019), which had more relevant measures including visceral fat, waist circumference, perceived control, and health literacy. The confirmation analysis used data from the Human Connectome Project-Aging (HCP-A; Bookheimer et al., 2019) and had many overlapping measures (e.g., BMI, education quality, and executive function).