Introduction:
Hypertension and diabetes control are leading modifiable risk factors for cardiovascular disease (CVD) risk. Contemporary epidemiological research suggests that individual and health system characteristics are associated with control. Yet, there is little evidence examining how the neighborhood environment influences hypertension or diabetes control among patients engaged in primary care.
Methods:
We analyzed data of adult patients (n = 5,711) with hypertension or diabetes who received primary care at 1 of 3 clinics of a statewide health care organization in Baltimore, MD and lived within the Baltimore city limits. The neighborhood environment exposures were tertiles of neighborhood socioeconomic status (SES), crime, and healthy food availability index (HFAI) assessed at the census tract level. Separate multivariate logistic regression models were constructed to estimate the odds ratio (OR) for each of these neighborhood environment characteristics and hypertension and diabetes control, adjusting for patient demographics, and health behaviors; and for physician demographic characteristics.
Results:
The overall mean age was 57.8 ± 14.3 years, 67.2% were female, and 89.6% were Black. There were 5,325 patients with hypertension and 2,094 patients with diabetes (not mutually exclusive). In bivariate analyses, high neighborhood SES (p= 0.030) and low crime (p= 0.006) were associated with hypertension control, and none of the neighborhood exposures were associated with diabetes control. In fully adjusted analyses, being Black was significantly associated with decreased odds of hypertension control in all models (OR range: 0.63 – 0.65), independent of neighborhood exposures. Among those with hypertension, living in a low or moderate SES neighborhood or a high crime neighborhood was associated with lower odds of hypertension control; however, these associations were attenuated in adjusted models. None of the neighborhood exposures were significantly associated with odds of diabetes control in models that adjusted for patient characteristics. Yet, compared to high SES neighborhoods, living in a low or moderate SES neighborhood (OR=0.74, 95% CI: 0.57 - 0.97 and OR=0.75, 95% CI: 0.57 - 0.98, respectively) was associated with reduced diabetes control after adjusting for both patient and physician characteristics.
Conclusion:
Exposure to neighborhood disadvantage may contribute to poor diabetes and hypertension control among patients in primary care, independent of patient and physician characteristics. Patient-centered risk assessments including measures of social need and preventive interventions adapted to neighborhood environments could be useful for optimizing hypertension and diabetes control in clinical settings.
Hypertension control and cardiovascular disease – Authors' reply
