Do air pollution and neighborhood greenness exposures improve the predicted cardiovascular risk?. (October 2017)

• Record Type: Journal Article
• Authors: Yitshak-Sade, Maayan; Kloog, Itai; Novack, Victor
• Is Part Of: Environment international. Volume 107(2017)
• Page Start: 147
• Page End: 153
• Publication Date: 2017-10
• Subjects: Air pollution -- Neighborhood greenness -- Cardiovascular risk -- Stroke -- Myocardial infarction; Environmental protection -- Periodicals; Environmental health -- Periodicals; Environmental monitoring -- Periodicals; Environmental Monitoring -- Periodicals; Environnement -- Protection -- Périodiques; Hygiène du milieu -- Périodiques; Environnement -- Surveillance -- Périodiques; Environmental health; Environmental monitoring; Environmental protection; Periodicals; 333.705
• DOI: 10.1016/j.envint.2017.07.011
• Languages: English
• ISSNs: 0160-4120

Abstract

Abstract: Background: Numerous studies show associations between exposure to Particulate Matter and Cardiovascular disease (CVD). Current cardiovascular equations incorporate the major risk factors for CVD. The patients' environment, however, is not incorporated in these equations. Methods: In a retrospective analysis, we assessed the contribution of neighborhood greenness and particulate matter (coarse-PM and PM < 2.5 μm–PM2.5 ) to the development of CVD by analyzing the change in prediction abilities. We included members of the largest health-care provider in Southern-Israel, who had at least one cardiovascular risk factor (dyslipidemia, diabetes, hypertension or smokers). PM exposure and neighborhood greenness (Normalized Difference Vegetation Index-NDVI) were assessed by satellite-based models. We used pooled logistic mixed regressions to obtain the CVD risks including conventional risk factors (i.e. age, gender, blood-pressure, etc.) and measured the model performance with and without PM and NDVI. Results: We included 23, 110 subjects, of whom 12% had CVD. Coarse-PM exposure was associated with stroke and Myocardial-Infarction (MI) (OR 1.02, p < 0.01 for both). NDVI was associated with MI: OR 0.72( p < 0.01) for NDVI 0.1–0.2; and OR 0.52( p = 0.270) for NDVI > 0.2. The c-statistics slightly improved from 77.30%–77.40% for the prediction of MI ( p = 0.004) and from 75.60%–75.76% for the prediction of stroke ( p = 0.027). Calibration was fair in all models. The associations were partially mediated through the patients' comorbidities. Conclusion: The negligible improvement in the prediction performance, despite significant associations with PM and NDVI, may be due to partial mediation of these associations through the conventional cardiovascular risk factors, suggesting the importance in assessing the environmental effects on more basic physiological pathways when addressing the contribution to the cardiovascular risk. Highlights: Stroke and MI were positively associated with a higher chronic exposure to coarse PM. Larger amount of neighborhood greenness was associated with a decreased MI risk. The incorporation of PM and NDVI did not improve the cardiovascular risk prediction.