Association between health literacy and self-care behaviors among patients with chronic kidney disease

This was a retrospective cross-sectional study using baseline data from the Kidney Awareness Registry and Education (KARE) study, a pilot randomized controlled trial (NCT01530958) examining the impact of a language-concordant and culturally sensitive CKD self-management support (CKD-SMS) program on blood pressure control among diverse patients with CKD who received primary care in urban safety-net clinics [18]. As this was an ancillary study to KARE; all pre-randomization data from the 137 KARE participants related to demographics, health literacy, and self-reported participation in self-care behaviors were used. Details of KARE participant eligibility and recruitment have been previously reported. In brief, the San Francisco Health Network’s electronic patient registry was used to identify eligible patients, who included adults with CKD, defined by an estimated glomerular filtration rate (eGFR) 15–60 ml/min/1.73 m² or albumin-to-creatinine ratio of > 30 mg/g or presence of at least two ≥ 1+ dipstick albuminuria separated by at least 90 days, who had contact with their primary health care team at least once within the prior 2 years and spoke English, Spanish or Cantonese. Exclusion criteria included: kidney transplant recipients, pregnant women, individuals on dialysis, and presence of co-morbid conditions that impeded meaningful communication between providers and patients or limited the usefulness of a self-management support program: prevalent dementia, impaired cognition or severe mental illness; expected life expectancy < 12 months; self-reported hearing impairment or severe visual impairment preventing use of a touchtone telephone keypad (Fig. 1). After the baseline visit, KARE participants were randomized to one of two groups: usual care without intervention or participation in a comprehensive self-management support program in addition to usual care. The study was approved by the UCSF Institutional Review Board (Protocol # 11–07399).

Baseline demographic data were self-reported and included: age, gender, race, preferred language, education attainment, and health insurance. Comorbid conditions (hypertension, diabetes, mean eGFR, and proteinuria) were ascertained from the electronic health record. Health literacy and self-care behaviors (medication adherence, tobacco use, activity level, fast food consumption, and sugary beverage intake) were ascertained with validated questionnaires [19‐21].

Health literacy, the primary explanatory variable, was determined by a validated 3-item screening questionnaire that assesses the patient’s ability to comprehend health information. Questions include: 1. “How often do you have someone help you read hospital materials?” 2. “How confident are you filling out medical forms by yourself?” 3. “How often do you have problems learning about your medical condition because of difficulty understanding written information?” Patients indicated their responses to each question on a 5-point scale from 0 to 4. The score for each item was summed and the total divided by 3. An individual was considered to have limited health literacy if he/she had a score ≥ 6 out of a maximal score of 12 [19].

Outcomes included measures of selected self-care behaviors: medication adherence, tobacco use, physical activity, fast food consumption, and sugary beverage intake. Medication adherence was assessed with a 4-item validated, subjective scale of medication adherence [20]: 1. “Do you ever forget to take your medicine?” 2. “Are you careless at times about taking your medicine?” 3. “When you feel better, do you sometimes stop taking your medicine?” 4. “Sometimes if you feel worse when you take the medicine, do you stop taking it?” Responses were either Yes or No to each question, which a score of 1 for each “Yes”. Responses were categorized as “high adherence” – 0 points; “intermediate adherence” – 1-2 points; or “low adherence” – 3-4 points. Current tobacco use was determined using 2 validated questions from the 2009 Behavioral Risk Factor Surveillance Survey [21]: 1. “Have you smoked at least 100 cigarettes in your entire life? (5 packs=100 cigarettes)” (response options were “yes”, “no”, and “don’t know/not sure”) and 2. “Do you now smoke cigarettes every day, some days, or not at all?” (response options were “every day”, “some days”, “not at all”, and “don’t know/not sure”). Responses to both questions were used to create a binary variable depicting current tobacco use. Participation in physical activity was also determined using three items from the Behavioral Risk Factor Surveillance System (BRFSS): (1) “When you are at work, which of the following best describes what you do?” Answer options were categorized as none/light activity [mostly sitting or standing or mostly walking] or moderate/vigorous activity [mostly heavy labor or physically demanding work]; (2) Thinking about the moderate activities you do when you are not working in a usual week, do you do moderate activities for at least 10 min at a time, such as brisk walking, bicycling, vacuuming, gardening, or anything else that causes some increase in breathing or heart rate? Answers options were “yes”, “no”, or “don’t know/not sure”; and (3) Thinking about the vigorous activities you do when you are not working in a usual week, do you do vigorous activities for at least 10 min at a time, such as running, aerobics, heavy yard work, or anything else that causes large increases in breathing or heart rate? Answer options were “yes”, “no”, or “don’t know/not sure” [21]. Fast food and sugary beverage consumption were also assessed using questions from BRFSS. Individuals were asked to indicate the number of times in the past week they had eaten from a fast food restaurant (“During the past 7 days, how many meals did you eat from a fast-food restaurant or pizza place?”) and the number of times per day/week/month/year that they consumed sugary beverages (“How often do you drink soda that contains sugar? Do not include diet soda” and “How often do you drink sweetened fruit drinks, such as Kool-Aid and lemonade? Include fruit drinks you make at home and add sugar to.” Responses for fast food intake were categorized as “none”, “once”, “twice” or “three or more”. Responses for sugary beverage intake were categorized into: “never”, “monthly/yearly” or “daily/weekly” [21].

Participant characteristics were compared by level of health literacy by chi-square and ANOVA tests. Multivariable logistic regression models were used to determine odds ratios and 95% confidence interval for the presence, direction, strength and independence of an association of literacy level with current tobacco use and low physical activity. Ordinal logistic regression models were used to determine the association of health literacy status with lower medication adherence, greater fast food intake and higher sugary beverage consumption. Each model was adjusted for confounders statistically associated with both the explanatory variable and the outcomes in this study: age, diabetes, and educational attainment, as well as race and gender to be consistent with prior literature. Sample size calculations suggested that with estimated prevalence of limited health literacy and tobacco use both around 40% for this study population (unpublished data), a sample size of 137 participants would provide 80% power to detect a 2.8-fold increased odds of tobacco use among those with limited health literacy compared to those with adequate health literacy. All statistical analyses were performed using STATA software, Version 14.0 (StataCorp, College Station, Texas, USA).

Among the 137 study participants, 101 (74%) had adequate health literacy and 36 (26%) had low health literacy. Individuals with low health literacy were older than those with adequate health literacy (mean age 59.4 years vs. 53.9 years, p = 0.02), but were similar with respect to gender, race/ethnicity, language preference and health insurance. The overall study population was racially diverse (6% White, 36% Hispanic, 43% Black, and 15% Asian) and nearly 40% had a non-English language preference. Nearly 25% had no health insurance, 45% had Medi-Caid, and 27% had Medi-Care. Individuals with low health literacy had less educational attainment than those with adequate health literacy, with higher proportions of individuals reporting only completing primary school and fewer reporting attending college (p = 0.003). A greater proportion of individuals with low literacy had diabetes compared to those with adequate health literacy (75% vs. 44%, p < 0.001), though the two groups of participants did not differ with respect to the prevalence of hypertension, prevalence of albuminuria or severity of CKD. Overall, 39% of the study population had hypertension, 69% had albuminuria 33% had Stage 1 or 2 CKD, and 67% had Stage 3 or 4 CKD (Table 1).

Over two-thirds of study participants (70%; n = 95) reported lower medication adherence (51%, n = 70 with intermediate adherence; 18.2%, n = 25 with low adherence) and nearly one-third (32.6%, n = 44) reported current tobacco use, neither of which differed by health literacy level. Relatively few study participants reported engaging in no or light activity (11.8%, n = 16), again with similar levels of engagement by health literacy status (p = 0.50). Overall, a small number of study participants (14.6%) reported frequent fast food consumption, though healthier patterns were patient-reported among those with limited health literacy (p = 0.04). Similarly, while 60% of all study participants reported daily or weekly sugary beverage consumption, healthier patterns were reported by individuals with low health literacy (p = 0.01) (Table 2).

Among KARE study participants, health literacy was not associated with lower medication adherence (adjusted odds ratio [aOR] = 0.84; 95% CI 0.38–1.89) or physical activity (aOR = 2.39; 0.54–10.53), independent of age, gender, race/ethnicity, educational attainment and diabetes status (Fig. 2). However, patients with lower health literacy had non-statistically significant higher tobacco use compared to those with adequate health literacy (aOR = 2.33; 0.90–6.06). Additionally, individuals with low health literacy reported substantially lower consumption of sugary beverages (aOR = 0.50; 0.20–1.23) and statistically significant decreased fast food intake (aOR = 0.38; 0.16–0.93) compared to those with high health literacy.