Share this article on:

Measuring Success: Stroke Registry Shows Care is Improving: What Works? What Needs Improving?

Shaw, Gina

doi: 10.1097/01.NT.0000398692.78443.0c


Hospitals in seven states participating in the Centers for Disease Control and Prevention's Paul Coverdell National Acute Stroke Registry (PCNASR) demonstrated significant improvement in compliance with the 10 quality measures monitored by the CDC over the course of a 5-year period from 2004 to 2009.

In a Morbidity and Mortality Weekly Report, published in the April 27 Journal of the American Medical Association, CDC authors noted that compliance rates for nine of the ten measures improved, with five measures showing average improvements of at least 6 percent. The sole measure that showed no improvement, provision of anti-thrombotic therapy at discharge, was already at such a high rate of compliance — 98 percent — that further significant improvement is unrealistic, the report authors wrote.

The measures for which hospitals demonstrated the most improvement included adherence to the use of thrombolytic therapy (IV tPA), at 11 percent; counseling on smoking cessation, at 9.2 percent, and lipid testing and/or treatment (7.6 percent).

Overall, the proportion of patients receiving “defect-free care” — that is, they were given all the quality-of-care measures for which they were eligible — also increased significantly (p=0.0001) in all four categories studied: 21 percent in patients with TIA; 17 percent for inpatient measures in patients with ischemic stroke; 17 percent in patients with hemorrhagic or ill-defined stroke; and 11 percent for outpatient measures in patients with ischemic stroke.

Over the same time period, there were mixed effects seen on inpatient mortality measures. While in-hospital mortality for hemorrhagic stroke increased (from 21 to 23.7 percent, p=0.02), in-hospital mortality for ischemic stroke decreased significantly, from 6.2 percent to 5.1 percent (p=0.001), and the in-hospital mortality rate for ill-defined stroke also decreased, although not significantly (from 6.2 percent to 4.9 percent).

Back to Top | Article Outline


“Like with anything else, when you shine a light on a process, people tend to pay attention to it,” said Marilyn Rymer, MD, founder and director of Saint Luke's Mid-America Brain and Stroke Institute in Kansas City, MO. “Institutions have now developed standing order sets, clinical pathways, and nursing and physician education programs to ensure that these things don't fall through the cracks. In our center, beginning in the mid-1990s, we started to embed most of these measures in our admission orders and clinical pathways, and we continue to revise them based on new evidence.”

The CDC established PCNASR — commonly referred to as “Coverdell” — in 2001 to track and monitor hospital-based stroke care, but initially it was only in a prototype phase, with registries led by CDC-funded investigators in academic medical institutions. It was only in 2005 that statewide registries, led by CDC-funded state health departments, began to be fully implemented.

Back to Top | Article Outline


Coverdell isn't the only stroke registry out there. The American Heart Association's “Get With The Guidelines” collaborative performance program has a stroke component, which tracks a much larger set of achievement, quality, reporting, and descriptive measures in stroke care. Participating hospitals receive a Web-based Patient Management Tool™, decision support, a robust registry, real-time benchmarking capabilities and other performance improvement methodologies to support them in reaching their quality-of-care goals.

But the “Get With The Guidelines” program primarily involves larger academic institutions and other major hospitals with primary stroke centers.

The improvements seen in the Coverdell reporting surprised and encouraged S. Claiborne (“Clay”) Johnston, MD, PhD, professor of neurology and epidemiology, associate vice chancellor of research and director of the Stroke Service at the University of California-San Francisco, because they came from a broader range of hospitals, including county and community hospitals, across seven states.

“It's actually a little more improvement than we've seen in studies of centers that have specifically committed to improving stroke care, like those enrolled in ‘Get With The Guidelines,’” Dr. Johnston said. “Now, probably part of the reason that the numbers are greater is because the baseline for many of these measures is lower at these institutions, so there's more room for improvement. But it's very heartening that a much broader range of hospitals than we've seen reported in other studies can improve process measures in stroke care so dramatically.”

“This is great news,” Dr. Rymer agreed. “The more we plug away at each of these measures, the better we're getting. The more we can put routines and standards in place, the better the outcome is for each patient.”

Back to Top | Article Outline


The Joint Commission in conjunction with the Centers for Medicare and Medicaid Services also has stroke measures, she noted. “They are not quite the same, but many of them overlap. We're getting to agreement on the metrics that are evidence-based and matter in outcomes, and we need to comply with them all the time.”

That's easier said than done when it comes to some measures. The one stroke measure that everyone agrees is a particular challenge involves the administration of thrombolytic therapy (IV tPA) to all eligible patients.

“TPA requires a lot of processes to be in place to work well, such as specialized staff and availability of acute imaging,” said Dr. Johnston. “It's going to take much more effort over a longer period of time to see substantial improvements in tPA rates. Given that, the kind of improvements they showed were encouraging. But there's much further that we can go. We should be able to treat the vast majority of patients eligible for tPA at the hospitals the way they want to receive stroke patients.”

Some hospitals, he said, may never get there. “They may not have a CT scanner available 24-7 to the emergency department. They may not see enough stroke patients in a year to justify having a neurologist always on call. Ultimately, we need a system where we can identify that some hospitals just can't do this, and have ambulances bypass those hospitals with suspected stroke patients and go to stroke centers. That's already happened in states like Massachusetts and New York.”



Another challenging measure is dysphagia screening. “It's an unusual measure, in that it is not one of the performance measures used by the Joint Commission for Primary Stroke Center certification, so for some hospitals it may fall off to the side,” said Irene Katzan, MD, director of the primary stroke center at the Cleveland Clinic. “Clinically we feel that it's important, but it's tough to make sure that staff remember to do it and be careful about their documentation, to be clear that screening is done prior to any oral intake.”

As stroke care gets better, that can paradoxically cause challenges for the implementation of some measures. “Stroke education is a big issue for us, because our length of stay on the acute care stroke center is about 4.9 days,” said Dr. Rymer. “That's good — thanks to improvements in care, stroke patients are able to get home sooner. But that means there's less time for education.“

“During the time they're with us, there's a lot going on for patients and their families: they're going for diagnostic studies, getting therapy initiated, and the time for in-depth education is minimal,” Dr. Rymer said. “We have a patient education booklet and a risk modification sheet, and we set goals for them and try to educate them during their stay about each of those goals, such as lowering blood pressure and lipids and smoking cessation. But in the time crunch to get it all done, sometimes we forget to document our education efforts.”

Back to Top | Article Outline


Implementation of electronic health record systems can improve compliance with stroke measures, Dr. Katzan said. “Hospitals that are up and running with EHRs are beginning to use those systems to provide reminders to help improve performance and optimize care.”

Stroke performance measures are also part of the meaningful use clinical quality requirements that hospitals must meet if they want to receive Medicare and Medicaid reimbursements under the Health Information Technology for Economic and Clinical Health (HITECH) Act.

But during the transition period from an old-school paper facility to a 21st-century electronic health record system, compliance with guidelines such as stroke measures may actually deteriorate. The Cleveland Clinic learned this the hard way when it implemented an electronic health record. The stroke center's rate of compliance with dysphagia screening suddenly plummeted.

“We weren't doing a worse job. But the documentation was different, and we had to learn about things like time-stamping,” Dr. Katzan said. “The EHR may timestamp when you document performing a measure, but maybe you're documenting it an hour and a half after you actually did it. With dysphagia screening, that can make a huge difference. You have to pay a lot of attention to how your documentation is changing.”



The measures now in place should actually be considered just a good start, said Dr. Johnston. “There are still other areas that need more work. Blood pressure is also a major indicator of stroke and stroke recurrence, and we don't have a measure for that. That's a hole in our knowledge,” he noted.

“And we don't have good assessments of what it means to have adequate physical therapy and occupational therapy, and an adequate discharge plan for patients with stroke,” he continued. “Where are they going? Are they getting adequate support to make sure they have the greatest chance for independence? That's not part of what we assess at this point. We don't have measures looking at what's happening to patients after discharge. It's great to provide wonderful care during hospitalization, but if you aren't continuing to treat with the right drugs 3-6 months later, you aren't doing nearly as much good as you could be. We need to ultimately make health systems responsible for the care of stroke patients beyond just the acute hospitalization.”

Back to Top | Article Outline


The CDC assessed these 10 performance measures:

  1. Screening for dysphagia (ischemic and hemorrhagic stroke).
  2. Education on stroke — all stroke and transient ischemic attack (TIA).
  3. Receipt of intravenous tissue plasminogen activator (ischemic stroke).
  4. Lipid measurement and/or lipid-lowering medication prescribed at discharge (ichemic stroke and TIA).
  5. Nonambulatory patients given deep venous thrombosis/venous thromboembolism prophylaxis by end of second hospital day (ischemic and hemorrhagic stroke).
  6. Assessment for rehabilitation (ischemic and hemorrhagic stroke).
  7. Anticoagulation medication prescribed at discharge for patients with atrial fibrillation (ischemic stroke and TIA).
  8. Antithrombotic medication within 48 hours or by the end of the second hospital day.
  9. Antithrombotic medication provided at discharge (ischemic stroke and TIA).
  10. Counseling on smoking cessations (all stroke and TIA).
Back to Top | Article Outline


Use of a Registry to improve acute stroke care—Seven states, 2005–2009. JAMA 2011; 305 (16): 1649–1653.
    ©2011 American Academy of Neurology