Effect of a neurohospitalist service on outcomes at an academic medical center

Objective: To study the effect of a neurohospitalist service on length of stay, cost, patient satisfaction, and education at an academic medical center. Methods: This was a retrospective cohort study using administrative data, educational surveys, and standardized patient satisfaction surveys to compare outcomes in the 21 months before (n = 343) and 27 months after (n = 436) the introduction of a neurohospitalist service in October 2006 at a single tertiary care academic medical center. Results: The most common diagnoses treated in both periods were demyelinating disease, neuromuscular disease, seizure, CNS infection, and cerebrovascular disease. Mean length of stay was reduced during the neurohospitalist period compared with that during the preneurohospitalist period (4.6 days vs 6.3 days; p < 0.001), but there was no difference in median cost ($6,758 vs $7,241; p = 0.25) or in-hospital mortality (1.6% vs 1.2%; p = 0.61). After adjustment for diagnosis, admission source, and severity of illness, both length of stay (coefficient -1.85, 95% confidence interval [CI] -2.47 to -1.24) and cost (coefficient -1,558, 95% CI -2,645 to -470) were reduced during the neurohospitalist period. Thirty-day readmission rates were not different between the 2 periods in adjusted analysis. There were no differences in patient satisfaction or resident trainee satisfaction between the 2 periods, but medical student satisfaction was higher on standard educator evaluations after the neurohospitalist program was introduced. Conclusions: The introduction of a neurohospitalist service at an academic medical center coincided with a reduction in length of stay and cost and a nonsignificant trend toward improvement in medical student satisfaction without affecting mortality, readmission rates, or patient satisfaction.