Neighborhood Deprivation Index Is Not Predictive of Patient's Perceived Barriers to Healthcare in an Urban Setting

Historically, the Southern United States has reported a high prevalence of healthcare inequities. This region has consistently underperformed in both access to healthcare and quality of care. Additionally, the patients in this area have a higher percentage of underrepresented minorities and are more likely to fall below the poverty line. The goal of our study was to determine the relationship between access to healthcare and economic disadvantage in a large Southern urban healthcare system. Patients presenting to a vascular surgery clinic for evaluation were asked to interview with a member of our research team. Enrollment was limited by availability of the research team; oral interviews were required to eliminate any literacy bias. Our researchers recorded patient demographics, transportation resources, income, neighborhood, and the patient’s perceived barriers to healthcare access (BHA). The primary outcome variable was the patients self-reported BHA. The primary predictor variable study was the area deprivation index (ADI) obtained using the patient’s home address. Descriptive statistics, χ2, Fisher exact test, analysis of variance, t test, and logistic regression were used to analyze the data as appropriate for the data type. Multivariable logistic regression analysis was used for potential confounding variables using a backward elimination technique, forcing the ADI to remain in the analysis as a variable of interest. Between November 2021 and August 2022, 268 patients participated. The mean age of the study population was 61.6 years, 50.7% (136/268) of our population was Black, and there were 129 (48.1%) women. ADI was associated with race (P < .0001), income (P = .0035), educational level (P < .0001), and insurance type (P = .0015). BHA was reported by 42.2% of study participants (113/268). Variables associated with BHA were race (P = .0026), age (P = .036), employment status (P = .0106), additional expenses associated with the visit (P = .0006), and the distance travelled to the clinic (P = .0007). The ADI was not associated with BHA in the unadjusted analysis. In the multivariable regression, BHA was predicted by unemployment status (odds ratio [OR], 2.2; 95% confidence interval [CI], 1.2-4.1), expense related to the appointment (OR, 3.7; 95% CI, 1.5-9.0), and younger age (OR, 0.97; 95% CI, 0.95-0.99). The ADI did not predict BHA. Barriers to healthcare in vascular surgery are associated with unemployment, younger age, and clinic visit associated expenses. The ADI alone does not predict barriers to healthcare. Future studies need to focus on patient-specific variables.