Study Sheds Light on the Process Metrics Driving Rural Inequity in Stroke Care

In the United States, rural hospitals use thrombolytic treatment for patients with ischemic stroke less often than urban hospitals do and have slower average time to treatment. Additionally, rural stroke centers match urban stroke centers and exceed rural non-stroke centers in the provision of secondary stroke treatment. These are among key findings of a retrospective cohort study of patients treated for acute ischemic stroke at Get With The Guidelines (GWTG)-Stroke hospitals.

The study was published in Stroke by a Cleveland Clinic-led research team.

“This analysis provides the first comprehensive assessment of thrombolytic treatment, secondary stroke prevention and other guideline-recommended stroke care metrics in rural hospitals,” says corresponding author Shumei Man, MD, PhD, Director of the Thrombectomy-Capable Stroke Center at Cleveland Clinic Fairview Hospital. “Focused efforts are needed to help rural hospitals provide timely thrombolytic treatment for eligible patients, and our findings can help guide these efforts in a targeted way.”

Study rationale and design

Previous data have established that U.S. stroke incidence is higher in rural than in urban areas and that stroke patients in rural areas receive less thrombolytic therapy and have higher case fatality compared with their urban counterparts. However, previous studies have not adequately factored in time to presentation following stroke symptom onset and have not evaluated modifiable process metrics in acute stroke care.

“The questions are whether rural hospitals provide less care and what processes can be improved,” Dr. Man says. “We undertook this study to systematically assess rural hospitals in providing guideline-recommended stroke treatments and to identify specific opportunities to improve care.”

She and colleagues from Cleveland Clinic and UCLA Medical Center used the national GWTG-Stroke database to identify all adults treated for acute ischemic stroke at hospitals participating in the database from January 2017 through December 2019.

They set out to determine the following primary outcomes according to hospital location (urban vs. rural) and whether or not hospitals were stroke centers:

• Thrombolytic treatment rates among patients arriving within the thrombolytic therapy time window (4.5 hours of stroke symptom onset)
• Proportion of patients receiving thrombolysis with door-to-needle times within the guideline-recommended windows of 30, 45 and 60 minutes

Also assessed were additional speed-of-treatment metrics, various safety and efficacy metrics, and use of secondary stroke prevention measures (i.e., antiplatelets and statins, anticoagulation for atrial arrhythmias and smoking cessation interventions). Multivariable mixed-effect logistic regression was used to adjust for stroke severity and patient- and hospital-level characteristics.

Key findings

Among 1.1 million eligible patients identified, 692,839 had sufficiently complete data for inclusion in the analysis. Of these, 31,492 patients (4.5%) were treated at 378 rural hospitals (81 stroke centers and 297 non-stroke centers) and 661,347 patients (95.5%) were treated at 1,820 urban hospitals (966 stroke centers and 854 non-stroke centers). Median values on the NIH Stroke Scale were comparable between patients treated at rural versus urban hospitals.

The main results of the analysis included the following:

• Patients presenting within 4.5 hours of symptom onset at rural hospitals received thrombolytic treatment at only about half the rate of their counterparts at urban hospitals.
• Among patients arriving within 4.5 hours, rates of thrombolytic therapy utilization in rural stroke centers were lower than those in urban stroke centers (31.7% vs. 43.5%; adjusted odds ratio [aOR] = 0.72; 95% CI, 0.68-0.76) but higher than those in rural non-stroke centers (31.7% vs. 22.1%; aOR = 1.26; 95% CI, 1.15-1.37).
• The likelihood of a door-to-needle time ≤ 45 minutes was lower at rural stroke centers than at urban stroke centers (33.0% vs. 44.7%; aOR = 0.86; 95% CI, 0.76-0.96) but was higher than at rural non-stroke centers (33.0% vs. 28.9%; aOR = 1.24;95% CI, 1.04-1.49). Similar differences were seen with door-to-needle time thresholds of 30 and 60 minutes.
• On metrics of secondary stroke prevention, rural stroke centers performed comparably to urban stroke centers and exceeded the performance of rural non-stroke centers.
• Risk-adjusted in-hospital mortality was comparable between rural and urban stroke centers and among rural hospitals overall.

The disparities in process metrics between rural and urban hospitals were greater before multivariable adjustment, with hospital characteristics — particularly hospital size — having the greatest impact on comparisons in the adjusted model. The study authors note that a greater share of rural hospitals are small and nonteaching hospitals, which they write “may explain the significant attenuation of rural-urban differences after adjusting for hospital characteristics.”

Implications

“Our study has yielded novel data on modifiable gaps in processes of stroke care between rural and urban hospitals and also among different rural hospitals,” Dr. Man observes. “Unlike prior investigations, this study evaluated differences in the care processes that lead to the previously established inequity in stroke care in rural communities. These include missed opportunities for timely thrombolytic treatment, interventions for secondary prevention and inpatient rehabilitation.”

She and her co-authors write that these care gaps can be overcome by focused quality initiatives, as has been demonstrated by GWTG-Stroke hospitals in Target: Stroke national quality initiatives. They note that the American Heart Association recently started a three-year rural initiative to grant cost-free access to GWTG quality programs to as many as 700 rural hospitals.

“Our study provides specific and actionable data to guide these and other efforts to enhance rural stroke care and health equity in the U.S. and beyond,” Dr. Man concludes.