Cognitive decline impairs financial and health literacy among community-based older persons without dementia.

Literacy is an important determinant of health and well-being across the lifespan, and lower literacy poses a major public health challenge in the United States. According to the United Nations Educational, Scientific, and Cultural Organization, literacy refers to the ability to identify, understand, interpret, and use written materials to function effectively in varying contexts (Kutner, Greenberg, Jin & Paulsen, 2003). Attainment of literacy involves a continuum of learning to acquire the knowledge needed to achieve goals, maximize potential, and function optimally in society. Low literacy affects about 90 million Americans and is a major problem among persons over the age of 65, the majority of which fall below basic competency levels (Dewalt, Berkman, Sheridan, Lohr & Pignone, 2004; Lusardi & Mitchell, 2007). Many older persons also perform poorly in two specific domains of literacy, health and financial literacy, which are critically necessary for the many complex and influential health and financial decisions older persons face (e.g., end of life healthcare choices, retirement planning, distribution of accumulated wealth) (Baker, Gazmararian, Sudeno & Patterson, 2000; Saha, 2006; Sudore et al., 2006; Wolf, Gazmarian & Baker, 2005; Wood et al., 2011). Among older persons, lower health literacy is associated with an increased risk of mortality and disability, higher health care expenditures, and less frequent use of preventive services (Baker et al., 2002; Baker et al., 2007; Baker, Wolf, Feinglass & Thompson, 2008; Peterson et al., 2011; Schillenger et al., 2002; Scott, Gazmarian, Williams & Baker, 2002; Sudore et al., 2006; Wolf, Gazmararian & Baker, 2005), and lower financial literacy is associated with poor savings and investment behaviors, limited wealth, and mental health complaints (Agarwal, Driscoll, Gabaix & Laibson, 2009; Bennett, Boyle, James & Bennett, 2012; Condelli, 2006; Lusardi & Mitchell, 2007a and b). Despite increasing awareness that low levels of health and financial literacy threaten the health, economic security and well-being of literally millions of older Americans, to date, relatively little is known about the determinants of lower literacy in advanced age. Older persons may exhibit lower literacy due to longstanding influences such as limited education or word knowledge (Baker, 2006; Baker et al., 2007). However, aging often is accompanied by cognitive decline, and an emerging body of research suggests that cognitive function is related to literacy in old age, although a limitation of most studies is that they have not excluded persons with dementia who may be driving the association (Baker et al., 2008; Bennett et al. 2012; Wolf et al., 2012). Moreover, other individual resources (e.g., education, word knowledge, decision making style) contribute to literacy and are related to cognition, yet studies have not attempted to disentangle the effects of aging versus cognition on literacy in the context of related resources (Baker, 2006; Baker et al., 2008; Dewalt et al., 2004). An understanding of the complex interplay between aging, cognition and related resources that contribute to literacy is essential for the development of targeted interventions to improve literacy and consequently health and financial outcomes among the already large and rapidly growing population of older persons. In this study, we present a conceptual model of how aging, cognition and other individual resources influence literacy and test several specific hypotheses about the pathways linking these resources to literacy (Figure 1). Our conceptual model was guided by existing literature and is based on the fundamental premise that attainment of health and financial literacy requires multiple inter-related skills that enable one to access, comprehend and process health and financial materials, understand and recall information, weigh alternatives, make inferences and decisions, communicate as needed in institutional settings, and engage in a life long process of learning (Baker, 2006, Baker et al., 2008, Institute on Medicine, 2004, Wolf et al., 2012). Thus, our model includes resources that are typically acquired early in life and well-established determinants of literacy such as education and word knowledge. Additionally, the model incorporates cognition, but we focus specifically on the domains of executive function and episodic memory since these abilities are widely known to be vulnerable to age-related decline and are fundamental to the acquisition of domain-specific literacy such as that studied here (Prull, Gabrielli & Bunge, 2000, Salthouse, 2000, Wolf et al., 2012). Finally, the model includes an indicator of decision making style (i.e., risk aversion). Although decision making style is rarely studied in the context of literacy, we included it because attainment of literacy requires interaction with the social environment and decision making under conditions of ambiguity (Baker, 2006), and risk aversion specifically can alter both the frequency and effectiveness of such interactions and decisions. Specifically, individuals who are highly risk averse may avoid exposure to new or uncertain educational or occupational experiences and tend to be less educated and prefer more secure but lower wages over less secure but potentially much higher wages; they also are less likely to become entrepreneurs (Bonin, Dohmen, Falk & Huffman, 2007; Cramer, Hartog, Jonker & Van Praag, 2002; Orazem & Mattilla, 1991). Thus, risk aversion may limit opportunities to gain new knowledge and experience as needed for the attainment of literacy, particularly domain-specific literacy, over the lifespan. Finally, risk aversion is related to cognitive function and the two may operate in tandem or interact to influence decision making and attainment of literacy in old age (Boyle, Yu, Buchman & Bennett, 2012; Boyle, Yu, Buchman, Laibson & Bennett, 2011; Mata, Josef, Samanez-Larkin & Hertwig, 2011). The approach used here extends prior work by directly examining the complex interplay between multiple individual resources that contribute to literacy, quantifying the effect of aging on literacy, and determining the extent to which the effect of aging is direct or indirect via cognition. Based on our conceptual model, we used path analysis to test the following specific hypotheses regarding the pathways linking age, cognition and other resources to literacy: Figure 1 Proposed conceptual model linking age, cognition and other individual resources with literacy. Aging has a direct effect on literacy, such that older age is associated with lower health and financial literacy. The effect of age on literacy is partially due to an indirect effect of age on executive function and episodic memory. The effect of executive function on literacy is partially due to an indirect effect of executive function on decision making style (i.e., risk aversion). Additional resources (i.e., education and word knowledge) have independent effects on literacy. Finally, because the hypotheses above were tested using cross sectional data and therefore preclude an understanding of whether cognitive decline is associated with reduced literacy, we next used longitudinal cognitive data collected for up to 14 years prior to the literacy assessment to test the hypothesis that declines in executive function and episodic memory are associated with decreased literacy among older persons without dementia.