SESSION TYPE: Respiratory Support Posters
PRESENTED ON: Wednesday, October 24, 2012 at 01:30 PM - 02:30 PM
PURPOSE: A Cleveland Clinic health system initiative to improve processes, enhance efficiency and optimize outcomes during weaning and liberation from mechanical ventilation using short cycle Business Intelligence(BI) techniques
METHODS: At the Cleveland Clinic, a dedicated electronic charting system [Medilinks™, Mediserve Corporation] is used for respiratory therapy documentation. Documentation items are stored in a relational database [SQL Server™, Microsoft Corporation] as observation items. Clinical protocols are implemented in Mediserve as templates using branching logic. We modified our existing mechanical ventilation weaning protocols to capture key performance indicators (KPI) from the therapists’ documentation. These KPIs were analyzed using BI dashboards to characterize spontaneous breathing trials (SBT), sedation level and extubation in the medical intensive care unit (MICU). Reasons for failure were examined and iterative improvements were made in the weaning and extubation process. Using short-cycle BI, persistent sedation was identified as the leading reason for failure of spontaneous breathing trials, resulting in modification of sedation protocols.
RESULTS: A new daily weaning and extubation protocol was implemented in the 43 bed MICU in January, 2011. Weaning and extubation data were analyzed and interpreted every month. The data was also shared with nursing, physicians, and respiratory therapy staff. Continuous process improvement resulted in progressive improvement in rate of extubation and reduced duration of mechanical ventilation from an average of 6.3 to 4.4 days.
CONCLUSIONS: Techniques of short cycle business intelligence that have been used in retail and production industries can be adapted to clinical processes such as weaning from mechanical ventilation and extubation with significant improvements in successful outcomes.
CLINICAL IMPLICATIONS: Electronic medical records can be utilized as sources for KPI and short cycle business intelligence methods can be extended to clinical processes to improve outcomes and reduce costs.
DISCLOSURE: The following authors have nothing to disclose: Vince Roberts, Rory Mullin, Martha Lemin, Edward Hoisington, Madhu Sasidhar
No Product/Research Disclosure InformationCleveland Clinic, Cleveland, OH