Public Reporting Tied to Better Care, but Similar Outcomes, in Stroke Centers

Voluntary participation in public reporting is associated with a slightly greater likelihood of delivering guideline-based care to patients with acute ischemic stroke, but with no major impact on clinical outcomes, according to an analysis of hospitals in the Get With The Guidelines (GWTG)-Stroke program.

Although seven ischemic stroke quality measures were more likely to be used after centers opted into public reporting, there were no differences in the odds of discharge to home or in-hospital mortality, researchers led by Michael Mullen, MD (Temple University, Philadelphia, PA), report in a paper published online this week in JAMA Network Open.

Additionally, there were only minimal differences in the likelihood of independent ambulation at discharge or a composite of in-hospital mortality or discharge to hospice.

The lack of much difference in clinical outcomes “was a little bit of a surprise in that we expected for hospitals that were participating [in public reporting]—and we expected them to be high-performing hospitals at baseline—that their clinical outcomes may be different, Mullen told TCTMD

He pointed to two main factors to explain why quality of care, but not clinical outcomes, would differ based on public reporting.

For one, “part of the real success of Get With The Guidelines is that adherence to the quality measures is very high overall,” Mullen said. “That’s great in a way because that tells you that Get With The Guidelines works. But it may be that because quality of care from a process measure standpoint is high across the board, that already means you’re less likely to see clinical outcome differences.”

And second, many of the performance and quality measures included in GWTG-Stroke are not necessarily geared toward improving the outcomes examined in this analysis. “It may be that those measures, although important—making sure people are on antiplatelet therapy, patients with A-fib get anticoagulation, and statins are being used, etc—may not be as directly relevant for the outcomes of short-term mortality or independent ambulation at hospital discharge,” he suggested.

Public Reporting in GWTG-Stroke

Participation in the GWTG-Stroke program, a hospital-based quality-improvement initiative and clinical registry developed by the American Heart Association, has been associated with better quality of care and clinical outcomes. In 2019, organizers announced a voluntary public reporting program, and the current study delved into which centers opted into that initiative and the impact.

The analysis included registry data from the year 2021 on 501,763 patients (mean age 69.8 years; 51.5% men) who were admitted for acute ischemic stroke at 2,423 hospitals.

Nearly two-thirds of centers in GWTG-Stroke decided to participate in the public reporting initiative. Centers with a higher volume of stroke admissions per year (adjusted OR 2.07; 95% CI 1.43-2.99 for the highest- vs lowest-volume hospitals) and those considered high performers based on receipt of a silver or gold GWTG-Stroke quality award in 2018 (OR 3.32; 95% CI 2.63-4.20) were more likely to start public reporting.

That makes intuitive sense, Mullen said. “The hospitals that are doing a great job and know they’re doing a great job are maybe going to more readily share their data and their outcomes.”

On the other hand, hospitals that lacked stroke center certification, were located in the South, and were owned privately were less likely to join the voluntary initiative.

The hospitals that are doing a great job and know they’re doing a great job are maybe going to more readily share their data and their outcomes. Michael Mullen

The primary metric for quality in the study was defect-free care, a composite of seven measures: IV thrombolysis for patients arriving by 3.5 hours and treated by 4.5 hours, early antithrombotic use within 48 hours of admission, venous thromboembolism prophylaxis, antithrombotics at hospital discharge, anticoagulation for atrial fibrillation or flutter, smoking cessation counseling, and intensive statin therapy at discharge. This was met by more than 90% of hospitals, irrespective of whether they did or did not participate in public reporting (95.2% vs 92.3%, respectively).

In a fully adjusted model, however, patients treated at hospitals with public reporting were more likely to receive defect-free care (OR 1.31; 95% CI 1.27-1.35). This was observed generally for all components of that endpoint, albeit with small absolute differences between groups.

The primary clinical outcome was independent ambulation at discharge, which was significantly more likely when public reporting was in place (OR 1.02; 95% CI 1.01-1.04). There were no differences in the likelihood of discharge to home or in-hospital mortality, with slightly greater odds of a composite of in-hospital mortality or discharge to hospice in the public reporting hospitals (OR 1.05; 95% CI 1.02-1.08).

Taken together, Mullen said, “the qualitative takeaway for me is that the outcomes are very similar” in the two groups.

He indicated that future research should focus on whether differences would emerge between hospitals with and without public reporting for longer-term outcomes after discharge and on the potential impact of changes to the structure of public reporting programs, including what measures are reported.