AAN: Stroke Outcomes Better in Patients Treated by Neurologists

By John Gever
CHICAGO, 21 april 2008 -- Stroke patients whose care was managed primarily by a neurologist were less likely to die or need a second hospitalization, a researcher said here. Neurologist-treated stroke patients had a 14% lower rate of 30-day mortality and a 7% lower rate of 30-day readmission, according to a review of Medicare records reported by Judith Lichtman, Ph.D., M.P.H., of Yale, at the American Academy of Neurology meeting. On the other hand, primary care by a neurologist appeared to extend hospital stays significantly (P<0.0001), Dr. Lichtman said.
Dr. Lichtman also reported that in-hospital mortality rates were no different for patients treated by neurologists than for those treated by other physicians.
Dr. Lichtman and colleagues reviewed 328,400 cases of ischemic stroke in the fee-for-service Medicare database during 2002. Neurologists were the primary physicians in 22,960 cases; other physicians supervised care for 305,440 patients.
Dr. Lichtman said the fee-for-service patients are about 85% of the total covered by Medicare.
The "other" physicians included cardiologists (30.3% of the total), internal medicine specialists (37.8%), and general practice physicians (20.4%).
Cases involving transfers, discharges from facilities providing non-acute care, length of stay under one day, or no information on primary physician were excluded.
The analysis yielded an adjusted relative risk of 30-day mortality of 0.86 (95% CI 0.82 to 0.91) for cases supervised by neurologists versus other physicians.
For 30-day readmission, the adjusted relative risk was 0.93 (95% CI 0.90 to 0.97) for neurologist-led cases.
Adjustments included presence of comorbid conditions, race, age, gender, admission source, previous hospitalizations, and provider type.
In absolute terms, the 30-day mortality rate was 9.41% in neurologist-treated cases versus 11.10% for all others (P<0.0001).
Thirty-day readmission rates were 13.61% for patients treated by neurologists and 14.43% for the others (P=0.001).
In-hospital mortality was virtually identical at 5.42% for neurologist-led cases and 5.33% for those treated by other specialists (P=0.5624).
Risk-adjusted length of stay was 1.54 days (95% CI 1.52 to 1.57) in neurologist-led cases versus 1.40 days (95% CI 1.38 to 1.41) for others.
Stroke patients treated by neurologists tended to be male, black, and younger. They also had significantly fewer comorbid conditions, Dr. Lichtman said.
Perhaps as a result, the data showed that patients treated primarily by neurologists were significantly less likely to be readmitted for pneumonia, myocardial infarction, or coronary artery disease, relative to cases supervised by other physicians.
Dr. Lichtman said differences in comorbidities could have accounted for at least some of the apparently improved outcomes with neurologist-led treatment.
"It's possible that [the patients cared for by other specialists] may be more complex cases, maybe they had other acute events that were going on at the same time. That we haven't been able to look at very closely," she said.
Another limitation was the lack of information in the database about whether patients were "treated aggressively with medications at discharge," Dr. Lichtman said. The study also could not determine what other care might have been provided after discharge.
But the lack of detail in the Medicare records is offset by the large numbers and diversity of institutions represented, she pointed out.
She said additional research was needed to explore reasons for the apparently improved outcomes with neurologist-led treatment.
The study was funded by the National Institute of Stroke and Neurological Disorders and the Centers for Disease Control and Prevention.
Dr. Lichtman reported no potential conflicts of interest. One co-author on the study reported relationships with Bayer, Bristol-Myers Squibb, Cordis, Pfizer, AstraZeneca, Organon, GlaxoSmithKline, Merck, and Boehringer Ingelheim.
Primary source: NeurologySource reference:Lichtman J, et al "Stroke outcomes better for patients treated by neurologists" Neurology 2008; 70: A204.