Pediatric Hospitalist Co-Management of Neurosurgical Patients Leads to Decreased Length of Stay

Sunday, October 27, 2013
Regency Ballroom S (Hyatt Regency Orlando, formerly the Peabody)
Catherine B. Sullivan, MD1, Christine M. Skurkis, MD1, Kathleen McKay, PhD1, Zhu Wang, PhD1, Paul M. Kanev, MD2 and Anand K. Sekaran, MD, FAAP1, (1)Connecticut Children's Medical Center, Hartford, CT, (2)Neurosurgery, Connecticut Children's Medical Center, Hartford, CT

Purpose: Hospitalist co-management of sub-specialty patients has become increasingly common, yet there are few published studies describing patient outcomes. Two studies have reported decreased length of stay (LOS) or decreased cost after hospitalist co-management. No study has examined the benefits of pediatric hospitalist co-management in a pediatric neurosurgery population. The objective of this study is to determine whether pediatric hospitalist co-management of pediatric neurosurgical patients results in decreased LOS.

Methods: Prior to initiation of the co-management program, the pediatric hospitalists and neurosurgeons met to clarify processes. Routine ventriculo-peritoneal shunt revision patients were excluded from automatic co-management due to their typically limited need for complex medical care. The neurosurgeons provided education to hospitalists and pediatric residents prior to implementation. Hospitalist co-management consisted of care in the areas of nutrition, pulmonary clearance, pain management, rehabilitation, and discharge planning. A survey of the neurosurgeons assessed their overall satisfaction with the program.

A retrospective analysis of the LOS data was performed for the 2 years prior to initiation of the program (March 2009-March 2011), and was compared to LOS for the first year of the program (March 2011 to May 2012).

Results: Charts were reviewed for 347 neurosurgical patients in the 2 years prior to initiation of the co-management program and 179 patients who were cared for after initiation. Hospitalists co-managed 80% of all neurosurgical patients. Mean LOS decreased from 8.6 days (SD 20) to 5.6 days (SD 7.2). Median LOS decreased from 4 days to 3 days, (p< 0.01). Variability in LOS decreased as demonstrated by a reduction in the standard deviation. Differences in LOS persisted even when controlling for individual surgeons (p < 0.001). Neurosurgeons reported a high degree of satisfaction with hospitalist co-management. There were no differences in readmission rates between the two groups. Ongoing review of incident reports revealed no increase in complications pre and post hospitalist involvement.

Conclusion: Hospitalist co-management of pediatric neurosurgery patients was associated with a significant decrease in LOS and decreased variability in LOS. The shorter LOS did not result in increased readmission or complication rates. Future studies should further explore both efficiency and quality outcomes related to hospitalist co-management of sub-specialty patients.

   Graph: Neurosurgery (except shunt replacement) LOS, before and after transition to IMT management