Miriam E. Tucker

May 22, 2013

Stroke patients receive better care if they have their stroke outside the hospital than in the hospital. These worrisome findings come from a study based on data from the American Heart Association's Get With the Guidelines program.

"Maybe the emergencies we are least prepared for are the emergencies that happen in our very own backyard," Ethan Cumbler, MD, said here at Hospital Medicine 2013.

Approximately 35,000 to 75,000 in-hospital strokes are reported annually in the United States, but even this might be an underestimate because cases are often under-reported, noted Dr. Cumbler, who is program director for the National Stroke Association In-Hospital Stroke Quality Improvement Initiative.

In an interview with Medscape Medical News, Dr. Cumbler said that these data point to the need for improvements in rapid recognition of stroke symptoms by hospital providers and systematic changes to allow inpatients to receive appropriate thrombolysis in a timely manner.

"Patients who suffer their stroke while under our care deserve the same consideration for treatment, which could reduce their deficits and improve their outcome, as any patient who has a stroke at home," he emphasized.

"I definitely agree that the data clearly show an opportunity for improvement with regard to inpatient stroke," said session moderator Eduard Vasilevskis, MD, from Vanderbilt University Medical Center in Nashville, Tennessee.

He added that "hospitalized patients who are already sick and then have a stroke are going to do worse. The question is, if we improve quality, can we at least diminish how much worse they're going to do?"

Worse Outcomes

Data for the study came from 1280 hospitals reporting at least 1 in-hospital stroke to the Get With the Guidelines program. Investigators compared patient characteristics, comorbid illnesses, medications, quality-of-care measures, and outcomes between the 21,349 in-hospital ischemic strokes and 928,885 community-onset ischemic strokes.

The in-hospital stroke patients had more thromboembolic risk factors, including atrial fibrillation, prosthetic heart valves, carotid stenosis, and heart failure (P < .001). However, they were less likely to have had a previous stroke, have hypertension, or use tobacco (P < .0001).

Median National Institutes of Health Stroke Scale scores indicated that in-hospital strokes were more severe than community-onset strokes (9.0 vs 4.0; P < .001). In addition, in-hospital stroke patients were significantly less likely to achieve all 7 achievement Get With the Guidelines quality metrics and 3 of the 8 quality measures.

Stroke education didn't differ between the 2 groups, and both rehabilitation assessment and intensive statin treatment were better for the in-hospital stroke patients.

The gap in defect-free care was larger than I expected.

Defect-free care, defined as the proportion of patients who received all of the achievement-measure interventions for which they were eligible, was significantly worse for in-hospital stroke than for community-onset stroke (60.8% vs 82.0%; P < .0001).

"The gap in defect-free care was larger than I expected," Dr. Cumbler told Medscape Medical News.

However, of the 11% of in-hospital stroke patients who did receive appropriate thrombolysis, multivariate analysis showed a lower rate of intracranial hemorrhage than in community-onset stroke patients (odds ratio, 0.80; P = .049).

"Many patients with in-hospital stroke are candidates for aggressive intervention with thrombolysis and do not appear to have higher rates of hemorrhage when treated," Dr. Cumbler pointed out.

More in-hospital stroke patients died during hospitalization than community-onset stroke patients (14% vs 5%; odds ratio, 2.72; P < .0001). In addition, in-hospital stroke patients were less likely to be discharged home (OR, 0.37) and less able to ambulate independently at discharge (OR, 0.42).

"It makes sense that 2 problems are worse than 1 problem. If you come in with a heart attack, you will do worse than if you only had a stroke. However, there's also the disquieting hypothesis that some of these discrepancies in outcomes are related to discrepancies in the performance of quality metrics," Dr. Cumbler said.

Dr. Vasilevskis noted that most patients who have a stroke in the community end up in the emergency department, which is prepared to provide rapid stroke care because "there's a stroke alert, there's a stroke team, everyone comes, there's an order set.... We just haven't worked it out for the inpatient setting because it's much less common," he explained.

He added that family members might be better than hospital personnel at detecting sudden differences in a person's usual behavior that might signal a stroke.

The Specialist Advantage

Dr. Cumbler told Medscape Medical News that beyond basic differences in workflow that affect the ability of the emergency department and the wards to respond to time-critical emergencies, strokes from the community are often cared for on dedicated neurology or stroke services. Patients who are already hospitalized could be on cardiology, cardiothoracic surgery, or medicine services staffed by personnel who are "less attuned to the elements of quality processes for stroke care."

The solution, he said, is for individual providers to learn to better recognize new neurologic deficits and for systems to be improved so that evaluation and treatment decisions can occur within 60 minutes of symptom recognition.

"As individuals, we need to get the stroke program involved at the onset of stroke symptoms, so that the quality metrics can be met and reported appropriately. As systems, we need to create bundles of these interventions, which can be an overlay onto existing orders for patients who experience stroke during hospitalization," he told Medscape Medical News.

There is evidence that in-hospital strokes are under-reported. "We should be reporting in-hospital strokes just as we do community-onset strokes to quality databases so that we can accurately understand the quality of care we are providing, identify gaps, and perform process improvement to close any discrepancies."

"It seems that there's definitely room for us to improve process," Dr. Vasilevskis told Medscape Medical News.

Dr. Cumbler is on the Quality Speakers Bureau for the American Stroke Association Western/Pacific Region, is a course director for the National Stroke Association, and has a research collaboration with the Colorado Stroke Alliance. Dr. Vasilevskis has disclosed no relevant financial relationships.

Hospital Medicine 2013: Society of Hospital Medicine (SHM) Annual Meeting. Presented on May 18, 2013.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.