Health Data Literacy as a Public Health Intervention: Improving the Public’s Ability to Understand and Use Health Information

Abstract

Today, people receive health information mainly in numerical form: risk percentages, laboratory results, screening probabilities and online dashboards. The ability to understand and use these numbers, referred to here as health data literacy, is therefore a public health issue. Population evidence on who has trouble with health statistics, and on how this difficulty relates to everyday information tasks, is still limited. We conducted a secondary cross-sectional analysis of the Health Information National Trends Survey 6 (HINTS 6, 2022; n = 6,252 U.S. adults). The primary outcome was self-reported difficulty understanding medical statistics (hard or very hard). Design-based analyses used the final person weight and 50 jackknife replicate weights. We fitted survey-weighted logistic and proportional-odds models, calculated adjusted predicted probabilities and examined dose–response relationships with five information outcomes. Among 6,011 adults with a valid answer, 26.8% (95% CI 24.7–28.8) reported difficulty. The adjusted odds of difficulty increased by 42% for each step down in education (adjusted odds ratio [aOR] 1.42, 95% CI 1.21–1.66). Not using the Internet (aOR 1.66, 1.20–2.29) and fair or poor health (aOR 2.82, 2.03–3.92) were also independent correlates. Among adults with recent health care visits, those whose clinicians did not always explain things clearly had higher odds of difficulty (aOR 2.72, 2.22–3.32). Difficulty was associated with low confidence completing medical forms (aOR 2.34), low confidence finding online health resources (aOR 2.60), difficulty understanding cancer information (aOR 2.75) and difficulty understanding patient-portal information (aOR 5.30). Portal-information difficulty rose from 2.6% to 44.4% across the four response levels. About one in four U.S. adults report difficulty understanding medical statistics. This difficulty is socially patterned and strongly associated with practical barriers involving forms, online information, cancer information and patient portals. The results identify priority groups and delivery channels for health data literacy interventions. Because the design is observational, these interventions require controlled evaluation.