340 - Parent Health Literacy and Medication Comprehension for Children with Medical Complexity

Publication Number: 4333.340

Lauretta A. Oseni, Children's National Health System, Silver Spring, MD, United States; Salma Sadaf, Comer Children's Hospital at University of Chicago Medical Center, Chicago, IL, United States; Apurva Devaraj, 11 hoyt, BROOKLYN, NY, United States; Sheetal Sriraman, Columbia University Irving Medical Center, New York, NY, United States; Emily Acker, State University of New York Downstate Medical Center College of Medicine, Brooklyn, NY, United States; Saema H. Khandakar, NYC Health + Hospitals/Kings County, Albertson, NY, United States; Melvyn Braiman, State University of New York Downstate Medical Center College of Medicine, Brooklyn, NY, United States; Risa Bochner, New York City Health and Hospitals Harlem, New York, NY, United States

Background: Children with medical complexity (CMC) are high-risk for medication errors due to polypharmacy, multiple caregivers, and frequent transitions between care facilities. Low parental health literacy is associated with medication dosing errors, adherence errors and discharge plan comprehension errors.

Objective: To determine whether parental health literacy is associated with discharge medication comprehension among CMC.

Design/Methods: A cross-sectional study was conducted of English and Spanish speaking parents (n=60) of CMC < 18 years admitted to the pediatric critical care or acute care unit of two affiliated hospitals. Subjects completed a questionnaire about the child's discharge medications and sociodemographic measures. Subject responses were compared to medication data extracted from the medical record to determine accuracy. Medication comprehension was assessed within 6 domains (name, indication, dose, frequency, duration, side effects). Each domain was scored 1 if subject response matched the medical record and 0 if it did not match. The outcome measure, "total comprehension," was calculated as a continuous variable by summing the domains scores (minimum 0, maximum 6). The primary predictor was parental health literacy, measured by the Newest Vital Sign tool, and dichotomized into low (score 0-3) and adequate (score 4-6) health literacy. Bivariate analyses using the chi-square test and t-tests compared subjects with low versus adequate health literacy on sociodemographic and medication comprehension measures. Simple and multiple linear regression models assessed the association between health literacy and comprehension, accounting for covariates.

Results: Of 83 patients approached, 70 (84.3%) met eligibility and 60 (85.7%) were enrolled. Sociodemographic variables stratified by health literacy are presented in Table 1. Subjects with limited health literacy were less likely to recall medication names (33% versus 67%, P=0.01), frequencies (25% versus 54%, P=0.02) and side effects (8% vs 29%, P=0.03), and had lower total comprehension scores (mean 1.6 (standard deviation 1.2) versus 2.8 (1.4), P< 0.01) (Table 2). Unadjusted comprehension was 1.16 higher (standard error 0.33) in caregivers with adequate compared to limited health literacy (P < 0.01). Once adjusting for income, comprehension was 0.76 (SE 0.38) higher, but not statistically significant (P=0.05) (Table 3).

Conclusion(s): Low parental health literacy is associated with worse comprehension of discharge medications for parents of CMC but the relationship is confounded by income.

Sociodemographic factors by health literacy status


Comprehension Percent by Medication Domain and Total Comprehension (Summation of Correct Domains)
The total comprehension score is the summation of correct responses for the 6 domains. A higher score is better (minimum score 0, maximum score 6).

Simple regression health literacy association with total comprehension; Multiple regression health literacy association with total comprehension controlling for income
The table shows the results of a simple regression analysis with predictor variable health literacy (appropriate versus limited) and outcome variable total comprehension score. For comprehension score (summation of correct responses for the 6 domains), a higher score is better (Minimum score 0, maximum score 6). The intercept represents the predicted value of the comprehension score when health literacy is at the reference level (limited health literacy).

The table shows the results of a multivariable regression analysis with predictor variables health literacy (appropriate versus limited) and income ( <$25,000, $25,000+), and outcome variable, comprehension score. For comprehension score (summation of correct responses for the 6 domains), a higher score is better (Minimum score 0, maximum score 6). The intercept represents the predicted value of the comprehension score when health literacy is at the reference level (limited health literacy) and income is at the reference level ( <$25,000).