Performance Measures for Acute Ischemic Stroke Care: A Welcome Update

Stroke is a leading cause of death and disability in the United States and worldwide.¹ Clinical practice guidelines, such as those produced by the American Heart Association/American Stroke Association (AHA/ASA) and other organizations, provide detailed evidence-based recommendations for the appropriate investigation and treatment of patients with stroke, but their uptake may be suboptimal in the absence of interventions to promote their use. The use of performance measures can enhance adherence to guidelines and promote interventions known to improve outcomes.² It is with this goal in mind that the AHA/ASA Stroke Performance Measures Oversight Committee has released its latest report on clinical performance measures for adults hospitalized with acute ischemic stroke.³

A performance measure is defined by the American College of Cardiology (ACC)/AHA Task Force on Performance Measures as “any objective measure that has been developed to support self-assessment and quality improvement at the provider, hospital and /or health care system level” and that has “attributes that render them suitable for public reporting, and other forms of accountability, including direct comparisons between different institutions and health care providers, and possibly pay for performance”.⁴ Thus, performance measures are important to patients, clinicians and organizations, as they will often drive quality improvement initiatives, will require resources for monitoring and reporting, may prompt the diversion of resources from other clinical areas to improve performance on selected measures, and may be linked to funding and accreditation. For all of these reasons, it is critical that performance measures be reliable, accurate, evidence-based and clinically relevant, that data collection be feasible and that providers be able to directly influence the measure.

Prior efforts have been made to harmonize the existing stroke performance measures endorsed by major US stroke quality improvement organizations.⁵ Such harmonization is important to avert the frustration that can occur among provider organizations if different measures are required by different agencies, or if measures or definitions change over time, resulting in the need for new data collection, programming or analyses, or in difficulties in comparing results across organizations or eras. The current report builds on this prior work by reviewing existing performance measures from key organizations and identifying potential new or modified indicators. Consistent with the framework recommended for the development of performance measures,⁴ the report’s working group collaborated with other stakeholder organizations and provided an opportunity for public comment and peer review.

Fifteen performance measures are proposed. Of these, eight are existing measures which are endorsed by the National Quality Foundation (NQF) and which are generally supported by a strong evidence base: prophylaxis for venous thromboembolism, discharge on antithrombotic therapy, anticoagulation for atrial fibrillation/flutter, thrombolysis in eligible patients, initiation of antithrombotic therapy within two days of hospitalization, discharge on statins, assessment for or receipt of rehabilitation services, and initiation of thrombolytic therapy within 60 minutes of hospital presentation. Two measures (stroke education and tobacco use counseling) are existing indicators endorsed by many major stroke quality improvement organizations but not by the NQF because of difficulty in measuring these processes of care as well as uncertainty about the link between stroke education and outcomes; because of these concerns, the report recommends further evaluation and potential future modification of the measures. Five new measures are proposed: dysphagia screening performed within 24 hours of admission, documentation of passing a dysphagia screen prior to first oral intake, documentation of a National Institutes of Health Stroke Scale (NIHSS) score on arrival, use of continuous cardiac monitoring during the first 24 hours of admission, and early carotid imaging.

The current report is a success on many levels. It is methodologically rigorous and clearly incorporates input from clinicians with an understanding of why measures might or might not reflect best practice. There is a thoughtful review of both existing and candidate indicators, a discussion of issues of feasibility and accountability, and a rationale for why indicators were or were not included in the final list of performance measures. The measures are operationalized with clearly-defined numerators and denominators and with helpful recommendations about potential case ascertainment options. Importantly, the report recognizes the need to obtain data on either all consecutive patients or on a large random sample from each participating organization/jurisdiction so as to avoid the selection bias inherent in reporting results on a convenience sample of patients or institutions. Of the new measures, documentation of the NIHSS is of particular interest. Stroke severity is the strongest predictor of death and disability following stroke, and is therefore required for adequate risk adjustment when comparing outcomes after stroke among institutions or providers.⁶ The hope is that endorsing the NIHSS as a performance measure may increase rates of its documentation, which are currently below 50%, even at Get With The Guidelines® institutions, and may ultimately permit reporting on outcomes after stroke.⁷

Some caveats regarding the use of these performance measures are warranted. For some indicators, the denominator used for calculating the proportion of eligible patients who receive the intervention excludes those with contraindications; however, the specific contraindications are not specified. This ambiguity, while likely unavoidable, could lead to variations in interpretation of what constitutes a contraindication, as well as the potential for gaming by excluding specific patient groups from the denominators. Clinicians and institutions may have less enthusiasm for some performance measures than for others. For example, cardiac monitoring during the first 24 hours of admission, while included in many guidelines for the management of acute ischemic stroke,⁸ is supported by a relatively weak evidence base and could require significant institutional resources for its implementation. As noted by the report writing committee, this, as well as the other new indicators, will require pilot testing for feasibility and reliability, and the committee plans to review and update the measures at regular intervals. It is also worth noting that the measures evaluate processes of stroke care delivery, but not outcomes or issues related to organizational structure or efficiency, only relate to the acute hospitalization phase rather than the continuum of stroke care, are designed for use in adult patients with ischemic stroke, and may or may not be applicable to those with transient ischemic attack or hemorrhagic stroke. Benchmarks and targets for most indicators are still under development.⁹ Finally, some of the measures require data that may not be routinely collected at all institutions.¹⁰

This report represents an impressive step forward in our efforts to improve the care and outcomes of people with stroke and will be of tremendous relevance to organizations involved in the measuring and monitoring of stroke care delivery. Future efforts should focus on developing sustainable models of data collection for stroke performance monitoring and expanding performance measures across the continuum of stroke care.