Amputation Rates and Associated Social Determinants of Health in the Most Populous US Counties

Abstract

Social Determinants of Health (SDoH) are the environmental conditions that affect individuals’ health, functional status, and quality of life, and have been demonstrated to contribute to health outcome disparities in patients with peripheral artery disease (PAD). However, the impact of specific components that comprise the SDoH are not well understood. We evaluated how the components of SDoH and related demographic factors impact amputation rates in the most populous counties of the United States (U.S.). The Healthcare Cost and Utilization Project (HCUP) State Inpatient Database (SID) was used to determine the rates of discharge following lower extremity amputation for circulatory system disorders across the 100 largest counties of the U.S. in 2017. County demographic, hospital and SDoH data were matched using the U.S. Census, Dartmouth Atlas of HealthCare, and University of Wisconsin Population Health Institute County Health Rankings & Roadmaps data sources. Counties were divided into quartiles (Qs) based on amputation rates, and linear regression analysis was performed to assess associations between county-level amputation rates and SDoH factors. Amputation rates in the most populous U.S. counties assessed in the study varied widely from an average of 5.5 per 100,000 in Q1 to 15 per 100,000 in Q4. Compared with Q1, counties in Q4 had a higher proportion of African Americans (27% vs 7.9%; P < .05), diabetics (11% vs 7.9%; P < .05), smokers (17% vs 13%; P < .05), higher rates of unemployment (5.8% vs 4.6%; P = .01), poverty (16% vs 10%; P < .05), food insecurity (17% vs 13%; P < .05), single-parent households (42% vs 29%; P < .05), and physical inactivity (21% vs 17%; P < .05) (Table I). A significant association was found between amputation rate and county diabetes mellitus prevalence (ß = 0.68; 95% confidence interval [CI], 1.3-2.1; P < .05), mental distress (ß = 5.3; 95% CI, 3.7-6.9; P < .05), adult smokers (ß = 0.69; 95% CI, 0.46-0.92; P < .05), poverty (ß = 0.46; 95% CI, 0.32-0.60; P < .05), unemployment (ß = 1.2; 95% CI, 0.59-1.73; P < .05), homicide rate (ß = 0.61; 95% CI, 0.45-0.77; P < .05), physical inactivity (ß = 0.74; 95% CI, 0.57-0.90; P < .05), and food insecurity (ß = 0.51; 95% CI, 0.30-0.72; P < .05) (Table II). Amputation rates in the most populous U.S. counties are associated with several population characteristics and components of the SDoH, particularly rates of diabetes mellitus, physical inactivity, mental and physical distress, income/wealth, and food access. These findings indicate potential individual and community-level targets for improvement in the disparities present in the management of PAD.Table IPatient characteristics by amputation quartile (Q)1Q2Q3Q4QP valueAmputation rate per 100,000 Average5.57.910.415<.05Demographic data Age 65 and over13141314.16 African American7.912.01427<.05 Asian12.611.46.04.4<.05 Hispanic25203120.20 Non-Hispanic White51534747.54 Uninsured13121516.13 Diabetic7.98.99.411<.05 Current smokers13131517<.05 Physical inactivity17202121<.05 Rural2.93.03.23.6.91 Unemployed4.65.25.75.8.01 Poverty rate10111516<.05 Single-parent household29283742<.05 Housing insecurity21202221.73 Food insecurity13131517<.05 High school graduate81858179.12 Some college71706463<.05 Not proficient in English6.75.78.15.3.15Data are presented as percentages.Boldface P values indicate statistical significance.Quartiles corresponding to amputation rate per 100,000 in 76 US counties. 1Q (lowest amputation rate) to 4Q (highest amputation rate).Percentages reported correspondent to county averages. Open table in a new tab Table IIAssociation between social determinants of health components and amputation rateHealth care access and qualityß coefficient; [95% CI]P valueEducation access and qualityß coefficient; [95% CI]P valueSocial and community contextß coefficient; [95% CI]P valuePrimary care provider ratio0.0021 [0.0001-0.004]<.05English proficiency−0.027 [−0.23 to 0.18].798Racial segregation Black/White0.19 [0.13-0.26]<.05Health care cost0.00090 [0.00033-0.0014].002High school graduate−0.67 [−1.39 to 0.048].067Racial segregation non-white/White0.19 [0.13-0.25]<.05Diabetes mellitus prevalence1.7 [1.3-2.1]<.05Some college−0.18 [−0.27 to −0.098]<.05Limited access to healthy foods0.29 [0.074-0.50]<.05Diabetes mellitus monitoring−0.04 [-0.29-0.21].755Economic stabilityß coefficient; [CI]P valueFood insecurity0.51 [0.30-0.72]<.05Uninsured adults0.18 [0.035-0.32].015Poverty (%)0.46 [0.32-0.60]<.05Single-parent household0.29 [0.22-0.36]<.05Drug overdose mortality rate0.14 [0.015-0.27]<.05Median household income−0.0013 [−0.001 to −0.0009]<.05Mental distress5.3 [3.7-6.9]<.05Unemployment (%)1.2 [0.59-1.73]<.05Physical distress1.2 [0.86-1.6]<.05Neighborhood and built environmentß coefficient; [CI]P valueObesity0.51 [0.35-0.66]<.05Homicide rate (per 100,000)0.61 [0.45-0.77]<.05Adult smokers0.69 [0.46-0.92]<.05Severe housing problem0.13 [−0.058 to 0.31].174Total beds per county0.0019 [−0.0007 to 0.00004].158Air pollution particulate matter0.10 [−0.32 to 0.52].630Major teaching institutions per county0.41 [−0.090 to 0.90].107Physical inactivity0.74 [0.57-0.90]<.05ß coefficient, Percent difference in amputation rate; CI, confidence Interval. Open table in a new tab