Modification of Associations Between PM2.5 and Blood Pressure by Anti-Hypertensive Medication Usage in Heart Failure Patients

Background: Fine particulate matter (PM2.5) is associated with cardiovascular morbidity and mortality. Medications that target similar pathophysiologic pathways as PM2.5 may modify PM2.5-related health risks. Here, we used EPA CARES, a collection of electronic health records (EHRs) linked to environmental data, to examine whether anti-hypertensive medications modify associations between PM2.5 and blood pressure (BP). Methods: For this study, we used EHRs from 27,953 heart failure patients observed from 2014-2016. Daily PM2.5 was measured at the nearest US EPA monitor to each study participant’s primary residence. Linear mixed models adjusted for age, sex, race, season, relative humidity, temperature, and a natural spline term for time since study start were used to estimate associations between systolic and diastolic BP and daily PM2.5 on the day of measurement and up to 4 days before measurement as well as the 5-day average. Associations were stratified on use of anti-hypertensive medications and a multiplicative interaction term was used to estimate the interaction between PM2.5 and medication usage. Results are presented as the change in BP (mmHg) per 10 µg/m3 PM2.5 and the associated 95% confidence interval (CI). Results: The pattern of associations was consistent for all time periods examined so we present here just the 5-day average PM2.5 associations. For BP assessed when on anti-hypertensive medications we observed negative associations for systolic (-0.30, CI= -0.42, -0.18) and diastolic (-0.19; CI= -0.27, -0.12) BP. Conversely, for time-periods not on anti-hypertensive medications associations were positive for systolic (0.42, CI= 0.27, 0.57; interaction P= 5.3x10-12) and diastolic (0.13; CI= 0.04, 0.23; interaction P= 2.3x10-9) BP. Conclusions: Anti-hypertensive medication usage likely has interactions with short-term PM2.5 and medication usage in general should be accounted for when possible and explored for its ability to modify PM2.5-related health risks. This abstract does not necessarily reflect the policies of the US EPA.