REGARDS is a longitudinal study of United States (US) participants aged 45 years and older. The REGARDS study was designed to investigate reasons underlying the higher rate of stroke mortality among blacks, compared with non-Hispanic whites, and among residents in the Southeastern United States, compared with other US regions. At baseline, 56% of the participants were residing in the stroke belt (an area of the US with high stroke mortality in the Southeast including GA, NC, SC, AL, TN, LA, AR, MS), with the remaining 44% from the rest of the contiguous 48 US. Participants from the stroke buckle (an even higher stroke mortality region comprised of the coastal plains of NC, SC, and GA) comprised 21% of the REGARDS population. The cohort population at baseline was 42% African-American/58% white and 45% male/55% female. Further details on the study are available elsewhere. The REGARDS study and the current analysis were approved by the Institutional Review Boards of participating institutions.
At baseline, a telephone interview was conducted during which informed consent was obtained and the participant's self-reported demographic and behavioral factors, and medical history were recorded. Participants were then visited in their homes by a trained health professional who collected blood pressure, height, weight, blood, urine, and conducted electrocardiograms (ECGs), and obtained written informed consent. Blood and urine were sent to a central repository at the University of Vermont and ECG data were read at Wake Forest University. The examiner also left a residential history form to fill and mail.
Assessment of Sunlight Exposure and Temperature
We used data from the National Aeronautics and Space Administration (NASA) – National Oceanic and Atmospheric Administration (NOAA) NLDAS-2 dataset to determine sunlight radiation and temperature. The NLDAS-2 dataset is based on model reanalysis data and remotely-sensed and ground observations, and consists of a grid surface with ~14 km resolution over North America. NLDAS-2 solar radiation that was assessed at one-hour intervals was used to calculate a daily total referred to herein as daily "insolation". For this study, we merged daily insolation and maximum air temperatures with data from REGARDS' residential history form, which consists of locations where the participant had lived prior to enrollment into REGARDS, along with age when relocating. Each location the participant recorded was matched to a feature in the US Geological Survey's Geographic Names Information System using ArcGIS 9.3. For participants who had a period of missing residential data, due to having an unidentifiable location or residence outside of the contiguous 48 United States, we used only the existing residential history to compute environmental exposure averages. We assumed participants moved during July of the indicated moving year.
As in our previous studies, we calculated each month's average daily insolation and temperature exposure at each participant's residential location to estimate each participant's average exposure for the year previous to baseline.[20,27] We then categorized insolation and temperature exposure into quartiles. In order to capture extreme exposures, we also categorized insolation and temperature exposure using cutpoints at the 5th and 95th percentiles.
Blood pressure was measured during the REGARDS in-home visit by a trained technician using a standard protocol and regularly tested aneroid sphygmomanometer and was calculated as an average of two measurements taken after the participant was seated for five minutes. Hypertension was defined as present with self-reported use of antihypertensive medications, systolic blood pressure (SBP) ≥ 140 mm Hg, or a diastolic blood pressure (DBP) ≥ 90 mm Hg. Blood was collected during the in-home visit, and shipped to the central laboratory at the University of Vermont using standard protocols. Standard assays were used to determine lipid levels and high-sensitivity C-reactive protein (CRP) assays were used to determine the level of CRP, which was log transformed due to a skewed distribution. CRP levels were categorized into low/medium risk (CRP ≤ 3 mg/dL) and high risk (CRP > 3 mg/dL). Dyslipidemia was defined as present with self-reported use of lipid-lowering medication, total cholesterol ≥ 240 mg/dL, low-density lipoprotein (LDL) ≥ 160 mg/dL, or high-density lipoprotein (HDL) ≤ 40 mg/dL. Kidney function was determined by the estimated glomerular filtration rate (eGFR) computed using the CKD EPI equation (participants with eGFR < 60 classified as having impaired kidney function).
Because some confounders had missing data for large numbers of participants, we attempted to minimize selection bias by creating a separate "missing" category for any variable that had >1,000 participants missing data. Participants were excluded due to data anomalies, stroke or coronary heart disease at baseline, missing residential history, and missing confounder data (for those will <1,000 participants missing data). Of the 30,239 participants enrolled at baseline, 17,773 participants were available for analyses (Figure 1).
Eligibility flowchart for participants included in the analysis of sunlight and cardiovascular risk factors.
To perform a split-sample replication analysis we randomly assigned the eligible participants into one of two samples of equal size. In the first "exploratory" sample, we ran multivariable logistic or linear regression models adjusting for temperature, age, race, region (stroke belt, stroke buckle, or non-stroke belt), gender, education (less than high school, high school graduate, some college, or college graduate), income (<$20,000, $20,000 to $34,999, $35,000 to $74,999, or ≥ $75,000), quartiles of vitamin D intake, exercise (none, 1 to 3 times/week, or ≥ 3 times/week), alcohol use (none, moderate: ≤ 1 drink per day for women or ≤ 2 drinks per day for men, or heavy: > 1 drink per day for women and > 2 drinks per day for men), smoking status (current, past, or never) and body mass index (<18.5, 18.5 to 24.9, 25 to 29.9, or ≥ 30). We also adjusted for statin use in models with cholesterol, HDL, or LDL as the outcome, and adjusted for antihypertensive medication use in the models with SBP as the outcome. Using interaction terms in fully adjusted models, we tested for multiplicative interactions between insolation and each of race or impaired kidney function on our outcome variables. In the second "confirmatory" sample, we re-ran all models for outcomes with significant associations in the explanatory sample.
This study was approved by Institutional Review Boards of participating institutions.
BMC Neurol. 2014;14(133) © 2014 BioMed Central, Ltd.