Self-reported race as a social determinant of stroke risk in observational versus clinical trial datasets

Abstract

Background

Self-reported Black (SRB) Americans are approximately twice as likely to have a stroke as self-reported White (SRW) Americans. While social determinants of health and vascular risk factors account for some of the disparity, half the increased risk remains unexplained and may be related to unmeasured real-world factors of the racialized experience.

Methods

In this cohort study, we compared SRB and SRW participants in the Systolic Blood Pressure Intervention Trial (SPRINT) to the same groups in the observational Atherosclerosis Risk in Communities (ARIC) study to evaluate if clinical trial participation mitigates disparities in stroke risk. We set the ARIC baseline at visit 4 and excluded participants with prior stroke to create an ARIC cohort similar in age to SPRINT participants. The study outcome was incident primary stroke. We report hazard ratios from Cox models and inverse-probability weighted Cox models with propensity score matching on participant age, sex, diabetes, atrial fibrillation, and smoking.

Results

We included 10,094 patients from ARIC and 8,869 from SPRINT, of which 26.1% were SRB. The risk of stroke between SRW participants in SPRINT versus ARIC was not significantly different (IPW-Weighted HR 0.78 [0.52-1.19]). SRB ARIC participants were twice as likely to have a stroke as SRW ARIC participants (IPW-Weighted HR = 1.96 [1.41 – 2.71]). However, SRB SPRINT participants did not have higher stroke risk compared to SRW SPRINT or ARIC participants (IPW-Weighted HR 0.99 [0.68-1.77] and 0.95 [.57-1.59], respectively). SRB SPRINT participants in the intensive BP control group had a lower risk of stroke compared to SRB ARIC participants (IPW-Weighted HR = 0.39 [0.20 – 0.75]).

Conclusions

SRB race, compared to SRW race, is associated with an increase in primary stroke risk in the ARIC study but not in the SPRINT trial. The absence of the racial disparity in stroke incidence in SPRINT indicates that aspects of the disparity are modifiable. Population-based interventions that test this hypothesis deserve further attention.

Introduction

Addressing health disparities requires understanding their root causes and how to alleviate them. Self-reported Black (SRB) Americans are twice as likely to have a stroke as self-reported White (SRW) Americans.¹ Apart from self-reported race, other social determinants of health (SDOH), such as environmental exposures, income, education, and insurance status, are central components to this disparity,² which in turn create differences in stroke risk factors, including higher rates of hypertension, hyperlipidemia, and diabetes in SRB patients. ^3,4 Moreover, the treatment and control of vascular risk factors varies by race, with higher rates of uncontrolled hypertension and hyperlipidemia among SRB patients.^3,4 Notably, one analysis indicated that beyond increased prevalence, some vascular risk factors may have a differential impact by race; elevated blood pressure disproportionately increased stroke risk for SRB patients compared to SRW participants.⁵ While sociodemographic and vascular risk factors account for some of the disparity in stroke risk, half the increased risk remains unexplained.²

Materials and Methods

In this cohort study, we compared SRB and SRW participants in the Systolic Blood Pressure Intervention Trial (SPRINT) to the same groups in the observational Atherosclerosis Risk in Communities (ARIC) study to evaluate if clinical trial participation mitigates disparities in stroke risk. We obtained the publicly available, deidentified SPRINT and ARIC datasets from NHLBI, which were exempt from IRB review. We included 10,094 patients from ARIC and 8,869 from SPRINT, of which 26.1% were SRB. To create a cohort of ARIC participants comparable in age to SPRINT, we set the ARIC baseline at visit 4 (1996-1998) and excluded participants with prior stroke. The study outcome was incident primary stroke, both ischemic and hemorrhagic. We report hazard ratios from Cox models and inverse-probability weighted Cox models with propensity score matching on participant age, sex, diabetes, atrial fibrillation, and smoking. To ensure balance, standardized differences were observed before and after weighting.

Results

Baseline demographics, stroke rates and mean follow-up are in Supplementary Table 1, and standardized differences after matching are in Supplementary Table 2. In ARIC, 1.5% of SRW and 3% of SRB participants had an incident stroke during the follow-up period, while 1.9% of SRW and 1.6% of SRB SPRINT participants had an incident stroke. Although ARIC predates SPRINT by 12 years, the risk of stroke between SRW participants in SPRINT versus ARIC was not significantly different (Figure 1). SRB ARIC participants were twice as likely to have a stroke as SRW ARIC participants (IPW-Weighted HR = 1.96 [1.41 – 2.71]). However, SRB SPRINT participants did not have higher stroke risk compared to SRW SPRINT or ARIC participants (IPW-Weighted HR 0.99 [0.68-1.77] and 0.95 [.57-1.59], respectively, Figure 1). SRB SPRINT participants in the intensive BP control group had a lower risk of stroke compared to SRB ARIC participants (IPW-Weighted HR = 0.39 [0.20 – 0.75]).

Figure 1.

Hazard ratio for stroke.

Hazard ratio (HR) for stroke, showing unadjusted and adjusted Cox models, and an inverse probability weighted (IPW) matched model. Cox models were adjusted for age, sex, diabetes, smoking, and atrial fibrillation.

Discussion

Our data suggest that standardized and cost-free hypertension care may reduce the disparity in incident stroke risk between SRB and SRW patients. SPRINT participants had improved access to antihypertensive medication, which would overcome an important economic SDOH and could increase adherence.⁶ Additionally, SPRINT staff addressed participant-specific adherence issues at monthly follow-up visits for six months and quarterly visits thereafter, and participants had protocolized medication adjustments if the SBP goal was not reached.

The standardized approach to medication adjustments may have helped alleviated any real-world differences in treatment approaches by race; SRB SPRINT participants had on average a slightly higher number of antihypertensives prescribed compared to SRW SPRINT participants in both the standard and intensive BP control groups.⁷ SPRINT excluded participants expected to have adherence issues, including those without a primary care provider, suggesting that SPRINT participants may be more likely to have controlled risk factors due to pre-existing support structures. At baseline, 21.2% of SPRINT patients self-reported low medication adherence as measured by the Morisky Medication Adherence Scale (MMAS-8), while 40% reported medium adherence, and 38.8% reported high adherence, highlighting that enrolled SPRINT participants did not have uniformly high medication adherence prior to study initiation.⁸

Despite a well-established higher risk of stroke in SRB community dwelling participants, SRB SPRINT participants did not demonstrate an increased stroke risk when compared to SRW ARIC or SPRINT participants. This suggests that some of the SDOH unique to SRB participants may have been overcome with the resources provided in the SPRINT trial. Other explanations are possible for these hypothesis-generating results, including that the SDOH in SRB SPRINT participants were themselves unique. While there may be uncaptured baseline differences in health, education level, or socioeconomic status between the two study populations, the Black patients in ARIC and the SPRINT control group had comparable stroke risk (IPW-Weighted HR = 0.96 [0.42 – 2.22]), suggesting that our propensity score matching did mitigate confounding. Limitations to our study include not being able to measure SDOH, apart from self-reported race, and the 12 years separating the trials during which medical management evolved. However, because the the risk of stroke in SRW patients was not significantly different in ARIC vs. SPRINT, we can infer that the 12-year difference did not meaningfully impact our analysis. SPRINT also excluded patients with SBP above 180 mm Hg, which limits generalizability of results to patients with uncontrolled hypertension and high SBP, especially considering the differential impact of hypertension on stroke risk in SRB compared to SRW patients.⁵

Conclusions

Our data suggest that the standardized, intensive, and cost-free hypertension care in SPRINT may reduce the disparity in incident stroke risk between SRB and SRW patients. Understanding the drivers of real-world disparities that may not persist in the clinical trial setting may be an opportunity to find public health interventions that could narrow disparities in stroke prevention.