Send to

Choose Destination
Prev Med Rep. 2019 Jul 13;15:100953. doi: 10.1016/j.pmedr.2019.100953. eCollection 2019 Sep.

Identifying neighborhood characteristics associated with diabetes and hypertension control in an urban African-American population using geo-linked electronic health records.

Author information

Urban Health Collaborative, Dornsife School of Public Health, Drexel University, Philadelphia, PA, USA.
Department of Epidemiology & Biostatistics, Dornsife School of Public Health, Drexel University, Philadelphia, PA, USA.
Family Practice & Counseling Network, Philadelphia, PA, USA.
Health Federation of Philadelphia, Philadelphia, PA, USA.
Department of Health Management and Policy, Dornsife School of Public Health, Drexel University, Philadelphia, PA, USA.


For health care providers, information on community-level social determinants of health is most valuable when it is specific to the populations and health outcomes for which they are responsible. Diabetes and hypertension are highly prevalent conditions whose management requires an interplay of clinical treatment and behavioral modifications that may be sensitive to community conditions. We used geo-linked electronic health records from 2016 of African American patients of a network of federally qualified health centers in Philadelphia, PA to examine cross-sectional associations between characteristics of patients' residential neighborhoods and hypertension and diabetes control (n = 1061 and n = 2633, respectively). Hypertension and diabetes control were defined to align with the Health Resources and Services Administration (HRSA) Uniform Data System (UDS) reporting requirements for HRSA-funded health centers. We examined associations with nine measures of neighborhood socioeconomic status (poverty, education, deprivation index), social environment (violent crime, perceived safety and social capital, racial segregation), and built environment (land-use mix, intersection density). In demographics-adjusted log-binomial regression models accounting for neighborhood-level clustering, poor diabetes and hypertension control were more common in highly segregated neighborhoods (i.e., high proportion of African American residents relative to the mean for Philadelphia; prevalence ratio = 1.27 [1.02-1.57] for diabetes, 1.22 [1.12-1.33] for hypertension) and less common in more walkable neighborhoods (i.e., higher retail land use). Neighborhood deprivation was also weakly associated with poor hypertension control. An important consideration in making geographic information actionable for providers is understanding how specific community-level determinants affect the patient population beyond individual-level determinants.


African Americans; Diabetes; Electronic health records; Hypertension; Neighborhoods; Social determinants of health

Supplemental Content

Full text links

Icon for Elsevier Science Icon for PubMed Central
Loading ...
Support Center