REACH Center Research Project 2

SUMMARY - RESEARCH PROJECT 2: Landscape fire PM2.5, air conditioning, and mortality among
dialysis patients
Patients with end-stage renal disease (ESRD) would benefit from more knowledge about health risks
posed by climate-sensitive hazards. One risk that is not fully explored for these patients is the long-term
impact of exposure to fine particulate matter (PM2.5) from landscape fire smoke. Large wildfires, an
important portion of landscape fires, have increased in frequency and size and models project that the
risk of these fires will increase in coming years. Though research has shown that short-term exposure to
landscape fire PM2.5 (e.g., 1-30 days) increases risk of all-cause mortality among patients receiving
dialysis, we do not yet know if long-term exposure to landscape fire PM2.5 increases mortality among
ESRD dialysis patients. In addition, residential factors that may mitigate environmental risks have not
been fully explored for ESRD patients. One of these residential factors is residential air conditioning (AC),
which may provide multiple benefits, including potentially reducing exposure to both heat and PM2.5
(from all outdoor sources). Recent research has shown variations in intra-urban distribution of
residential AC prevalence across metropolitan areas in the US. We will examine whether residential AC
prevalence may influence mortality among dialysis patients. In coordination with the Community
Engagement Core (CEC), we will work with a patient advisory board from the American Association of
Kidney Patients (AAKP) to discuss planned and future research, disseminate results, and plan for
continued engagement. The US Renal Data System (USRDS) maintains a nationwide registry of patients
who require dialysis. We will use the USRDS data of approximately 1.5 million patients who initiated
dialysis between 2006-2018 and follow them until 2019, creating a retrospective cohort to examine
whether landscape fire PM2.5 or AC prevalence influence mortality in dialysis patients. For Aim 1, our
exposure is the number of days exposed to elevated landscape fire PM2.5, defined as days when
landscape fire PM2.5 exceeded the current National Ambient Air Quality Standards 24-hr standard for
PM2.5 of 35 μg/m3. We will use publicly available estimates of 24-hr mean landscape fire PM2.5 (and
control for non-fire PM2.5), which uses spatially interpolated PM2.5 observations, available on a 15 x 15
km grid, for the full contiguous US each day. We will interpolate the data to the ZIP code level, as
residential ZIP codes are available for dialysis patients. We will use a semiparametric multi-level mixedeffect
Cox hazards regression for mortality and estimate the hazard ratios of our exposure (landscape
fire PM2.5 or prevalence of AC), adjusting for both individual- and community-level covariates. For Aim
2, AC prevalence was previously estimated for 115 metropolitan areas. We will restrict our analyses to
examine these 115 metropolitan areas and employ a similar model and confounders. This research will
provide information critical to help guide preventive strategies for dialysis patients, a population
vulnerable to health effects of environmental and climate-sensitive risks. Through the CEC and AAKP
engagement, we will design present and future research, disseminate findings, and engage end-users.

Grant Number: 5P20ES036775-02
NIH Institute/Center: NIH
Principal Investigator: Katie Applebaum