Background: Hispanics are 1.7 times more likely than non-Hispanic white patients to have been diagnosed with diabetes. Due to cultural and socioeconomic circumstances, they are at risk for poor health literacy. We evaluated a cohort of diabetic ED patients to test the hypothesis that Hispanic race would be associated with poor disease knowledge and higher prevalence of long-term complications.
Methods: We enrolled a convenience sample of consenting adult ED patients with prior diagnosis of diabetes. Patients provided answers to basic demographic/historical information on a written survey. They were subsequently asked to provide yes/no answers on a previously validated diabetes knowledge assessment tool for Mexican-Americans (DKQ-24), including risk factors, disease management, and complications. Study authors conducted a structured EMR review for evidence of long-term complications within our 6-hospital system that includes more than 70% of hospital-based ED visits in a 12-county region. Categorical data analyzed by chi-square. Continuous data analyzed by t-tests. Multivariate logistic regression performed to control for confounding.. The primary outcome measure was to compare Hispanic vs. non-Hispanic DKQ scores.
Results: There were 184 patients in the study group; 76% Hispanic. Hispanics and non-Hispanics similar with respect to mean age (59+/-12.9 vs.61+/-12.8 years; p=0.35), % female (40% vs. 42%; p=0.90), income < $20,000 (61% vs 46%; p=0.08), private insurance (31% vs. 26%; p=0.48), established primary care (PCP) (88% vs. 91%; p=0.61). Hispanics had lower dkq scores than non-Hispanics 14.0 vs 15.3+/-3.1 (p=0.03) and more commonly had education level less than high school graduate (45% vs. 16%; p=0.001). There were no significant differences for most diabetic related disorders: macrovascular disease (50% vs. 56%;p=0.53), ocular disease (5% vs. 0%;p=0.11), and neurovascular disease, (28% vs. 14%;p=0.07). Hispanics had higher renal disease prevalence (31% vs. 14%; p=0.03). Multivariate regression revealed that only ≤ high school graduate associated with DKQ score (p=0.02), while female gender (p=0.03) and PCP were associated with renal disease (p=0.04).
Conclusions: Controlling for confounding factors within our ED diabetic cohort, Hispanics were similar to non-Hispanics in terms of disease knowledge and diabetic related disorders.