Patients with acute stroke who were transferred to a hospital participating in a stroke registry in Michigan had significantly higher rates of in-hospital mortality and in-hospital complications compared with those who were not transferred.

A multivariable logistic regression analysis found that independent predictors of being transferred included younger age, hemorrhagic stroke and higher stroke severity. However, having a previous stroke decreased the likelihood of being transferred.

Lead researcher Adrienne V. Nickels, MPH, of the Michigan Department of Community Health in Lansing, and colleagues published their findings online in Circulation: Cardiovascular Quality and Outcomes on April 12.

“As stroke systems of care continue to evolve, registries need to expand their data collection efforts to provide a more complete understanding of the relative benefits and risks of transferring patients for acute stroke care,” the researchers wrote. “Our results also suggest that it is prudent to account for patient transfer status when comparing hospital outcomes for hospital profiling.”

The researchers noted that approximately 795,000 people in the U.S. each year have a new or recurrent stroke, which is the fourth leading cause of death in the U.S. and a major cause of long-term disability.

In this study, the researchers evaluated 16,202 acute stroke admissions to 36 hospitals that participated in the Michigan Stroke Registry from 2009 to 2011. Of the strokes, 13,453 were ischemic strokes, 631 were subarachnoid hemorrhage strokes and 2,118 were intracerebral hemorrhage strokes.

Of the 36 hospitals, 17 were primary stroke centers, 21 were teaching hospitals and 14 were both primary stroke centers and teaching hospitals. In addition, 16 hospitals had more than 300 beds, 7 had between 100 and 300 beds and 13 had fewer than 100 beds.

Hospitals in the registry entered data using the Get With The Guidelines-Stroke Patient Management Tool, which identifies patients who arrived at the hospital after being transferred from another hospital.

The mean age of patients was 69.2 years old, while 51 percent were female and 68 percent were white. In addition, 19 percent of patients were transferred to a registry hospital. Further, 53.9 percent of patients with subarachnoid hemorrhage and 32.7 percent of patients with intracerebral hemorrhage were transferred.

The researchers mentioned there was a significant increase in hospital transfers from 16.0 percent in the first quarter of 2009 to 22.4 percent in the last quarter of 2011. They added that patients who were 60 years old or younger, men, whites and hemorrhagic stroke cases were more likely to be transferred, whereas patients living in nursing homes and those with medical comorbidities were less likely to be transferred.

The in-hospital mortality rate was 7.4 percent, including 12.0 percent in transferred patients, 6.4 percent in those who were not transferred, 14.5 percent for acute transferred patients and 9.4 percent for delayed transferred patients.

In addition, 14 percent of patients suffered in-hospital complications, including 18.4 percent in transferred patients, 12.8 percent of patients who were not transferred, 20.8 percent for acute transferred patients and 15.4 percent for delayed transferred patients. The complications measured were deep vein thrombosis, urinary tract infection and pneumonia.

After adjusting for confounding variables, being transferred remained a significant predictor of in-hospital mortality and in-hospital complications.

The researchers cited a few limitations, including that the registry does not collect data on patient outcomes post-discharge such as disability, quality of life, stroke recurrence, hospitalization or survival. Hospitals also volunteered to participate in the registry, so the results may not be generalizable to other hospitals. In addition, they did not have information on the hospitals that transferred patients to the registry hospitals and did not have patient-level information on the timing and reasons for transferring patients.

“The lack of patient and hospital information relevant to the period between stroke onset and the arrival of the patient at the second (receiving) hospital is critical because it impairs our ability to quantify the benefits and potential risks of transferring patients between hospitals,” the researchers wrote. “Factors such as the severity of stroke at onset, time since stroke onset to first (and second) hospital admission, patient response to prior treatments, and the underlying reasons that led to the decision to transfer can all act as important confounding variables when attempting to compare outcomes between transferred and regular admissions.”