Effects of health literacy on health status and health service utilization amongst the elderly

Abstract

Amid increased concerns about the adverse consequences of low health literacy, it remains unclear how health literacy affects health status and health service utilization. With a sample of 489 elderly Medicare patients in a Midwestern city in the USA, we explored the intermediate factors that may link health literacy to health status and utilization of health services such as hospitalization and emergency care. We expected to find that individuals with higher health literacy would have better health status and less frequent use of emergency room and hospital services due to (1) greater disease knowledge, (2) healthier behaviors, (3) greater use of preventive care, and (4) a higher degree of compliance with medication. Using path analysis, we found, however, that health literacy had direct effects on health outcomes and that none of these variables of interest was a significant intermediate factor through which health literacy affected use of hospital services. Our findings suggest that improving health literacy may be an effective strategy to improve health status and to reduce the use of expensive hospital and emergency room services among elderly patients.

Introduction

Over the past decade, many studies have reported linkages between health literacy and health outcomes, such as health status, chronic illnesses, and hospitalization. For example, Baker et al. showed that among 2659 patients from emergency care centers and walk-in clinics, those with low health literacy were more likely to report their health as poor and had a higher rate of hospitalization (Baker et al., 1997, Baker et al., 1998, Williams et al., 1998). Using a sample of Medicare beneficiaries enrolled in a managed care organization, Gazmararian et al. (1999) found that individuals who rated their health as “fair/poor” were twice as likely to have low health literacy compared with individuals who rated their health as “good/excellent.” Baker et al. (2004) found that participants with inadequate and marginal health literacy were more likely to have an emergency department visit. Similarly, Friedland (1998) discovered that patients with lower health literacy tended to have longer hospital stays but fewer outpatient physician visits.

Findings of these studies raise the issue of how health literacy affects individual health and health care utilization. Experts suggest that the effects of health literacy on health status and utilization may be indirect, possibly through conditions such as disease knowledge, health behavior, and use of preventive care (Ad Hoc Committee on Health Literacy for the Council on Scientific Affairs, AMA, 1999, Neilsen-Bohlman et al., 2004). As Berkman et al. (2004) has cautioned, potential confounders may lie in the causal pathway between health literacy and health outcomes. However, prior research tended to examine the consequences of low health literacy in a singular fashion and did not consider the inter-relationships among the various health outcomes (Lee, Arozullah, & Cho, 2004).

In this study, we explored four potential intermediate factors that may link health literacy and health status and utilization: (1) disease knowledge, (2) health behavior, (3) preventive care, and (4) compliance. The relevance of these potential intermediary variables is reviewed below.

Research has shown that individuals with lower health literacy are less knowledgeable about diseases and less capable of properly caring for themselves (Arnold et al., 2001, Davis et al., 1996, Gazmararian et al., 2003, Kalichman et al., 1999, Kalichman et al., 2000, Knight, 1999, Lindau et al., 2002, Miller et al., 2003, Schillinger et al., 2002, Williams et al., 1998, Williams et al., 1998, Wolf et al., 2005). The areas of medical knowledge and self-care documented to be associated with health literacy include the knowledge of tobacco effects, diabetes, hypertension, chronic heart failure, asthma, HIV/AIDS, and the knowledge of prostate cancer screening and management, mammography screening, and cervical cancer prevention. Disease knowledge may be treated as a subcomponent of health literacy, which may include general health background knowledge (Neilsen-Bohlman et al., 2004). However, since it is not clear what constitutes “background” health knowledge and to what extent health literacy should encompass disease knowledge, we are treating disease knowledge as a separate construct in this study.

Research evidence linking health behavior specifically to health literacy is limited. Several studies, however, have reported associations between general literacy level and substance use (e.g., Arnold et al., 2001, Fredrickson et al., 1995, Hawthorne, 1996). It is conceivable that individuals with lower health literacy are more likely to engage in negative health behaviors, such as smoking, drinking, abuse of illegal substances, and living a sedentary lifestyle. This may be in part because of their limited access to and ability to understand health and medical information.

Limited ability to comprehend information about the importance and methods of early disease detection and treatment may lead to a lower rate of preventive care utilization among people with lower health literacy. Having problems following physician instructions and understanding information on an appointment slip or referral form (Davis et al., 2002, Williams et al., 1995) also may limit access to preventive care and routine physician visits. Some empirical evidence supports these arguments. Scott, Gazmararian, Williams, and Baker (2002) demonstrated a positive relationship between health literacy and preventive health care use among enrollees of a managed care organization's Medicare plan. Friedland (1998) found a significant association between lower health literacy and fewer physician visits.

Lorenc and Branthwaite (1993) conducted a study to understand factors leading to better medication compliance. Five of seven factors were potentially related to patients' health literacy level: accurate knowledge of regimen, belief in taking tablets exactly as prescribed, less fear of illness, ability to read the label on the bottle, and understanding what the doctor had said. Low health literacy was a significant predictor of two-day treatment adherence among HIV patients, after controlling for factors such as age, education, ethnicity, income, HIV symptoms, substance abuse, and emotional distress (Kalichman et al., 1999). A more recent study, however, found that low literacy was not associated with adherence to antiretroviral therapy among the HIV-infected patients (Paasche-Orlow et al., 2006). Among patients with cardiovascular risk factors, Gazmararian et al. (2006) found that patients with inadequate health literacy had an odds ratio of 1.4 for low medication refill adherence compared to those with adequate health literacy.

We anticipated that individuals with lower health literacy would have poorer disease knowledge, worse health behaviors, less preventive care, fewer physician visits, and poorer compliance with routine clinical visits and medications. These factors, in turn, may delay seeking timely and appropriate care, leading to worse health status and increased use of emergency and hospital services.