Disparities in Low-Cost Preventive Cardiovascular Disease Screening: The Interaction between Health Insurance Status and Blood Pressure Measurement

The American Heart Association estimates that cardiovascular disease (CVD) caused one in every 2.8 deaths in the US in 2005. Because CVD is the number one cause of mortality in the US, preventive care should be a high priority for health care providers. In the context of over 46 million Americans without insurance, we sought to compare rates of non-receipt of blood pressure measurement − a low-cost CVD preventive screening test − between uninsured and insured populations. We conducted multivariate logistic regression analyses of nationally weighted, cross-sectional Medical Expenditure Panel Survey data from 2005, limiting the data to non-institutionalized adults who had either an office-based, inpatient, outpatient, or emergency room visit in the past year (N=15,922). The outcome variable was self-reported non-receipt of blood pressure measurement. Controls included age, race, gender, income, education, metropolitan statistical area, self-reported health status, usual source of care, diabetes diagnosis, smoking status, weight, hypercholesterolemia, and previous hypertension diagnosis. Cut-off for statistical significance was set at p<0.05. Logistic regression analyses revealed that uninsured patients were more likely to have non-receipt of blood pressure measurement for all medical visits combined (19.7% uninsured v. 8.0% insured; adjusted odds ratio [AOR] 1.77, 95% confidence interval [CI] 1.46-2.14), as well as for office-based visits only (19.5% uninsured v. 7.6% insured; AOR 1.89, 95% CI 1.53-2.32). Differences for inpatient, outpatient, and emergency room visits were not significant. To control for population differences, hypothetical analyses were conducted in which uninsured individuals were ″given″ insurance by changing their insurance indicator variable from 0 to 1. Predictions of the change in rates of non-receipt of blood pressure measurement for the uninsured were then made based on the regression model. These analyses showed significant decreases in rates of non-receipt of blood pressure measurement after insurance acquisition for every visit category: all visits (7.0% decrease; 95% CI 6.7-7.2%), office-based (7.5% decrease; 95% CI 7.2-7.8%), outpatient (1.5% decrease; 95% CI 1.2-1.7%), emergency room (2.1% decrease; 95% CI 1.9-2.3%), and inpatient (1.4% decrease; 95% CI 1.1-1.6%). Based on these data, we conclude that being uninsured makes an individual more likely to have non-receipt of blood pressure measurement during medical visits, especially office-based visits. Health care workers must be made aware that low-cost CVD preventive screening care is being offered at lower frequency to the population of uninsured individuals. Further study to explain this discrepancy is merited.