• David G Coughlin, Monisha A Kumar, Neha N Patel, Rebecca L Hoffman, and Scott E Kasner.
• Department of Neurology, University of Pennsylvania, 3400 Spruce St, 3W Gates Pavilion, Philadelphia, PA, 19104, USA. David.Coughlin@uphs.upenn.edu.
• Neurocrit Care. 2018 Apr 1; 28 (2): 175-183.

BackgroundEarly unplanned readmissions of "bouncebacks" to intensive care units are a healthcare quality metric and result in higher mortality and greater cost. Few studies have examined bouncebacks to the neurointensive care unit (neuro-ICU), and we sought to design and implement a quality improvement pilot to reduce that rate.MethodsFirst, we performed a retrospective chart review of 504 transfers to identify potential bounceback risk factors. Risk factors were assessed on the day of transfer by the transferring physician identifying patients as "high risk" or "low risk" for bounceback. "High-risk" patients underwent an enhanced transfer process emphasizing interdisciplinary communication and rapid assessment upon transfer during a 9-month pilot.ResultsWithin the retrospective cohort, 34 of 504 (4.7%) transfers required higher levels of care within 48 h. Respiratory failure and sepsis/hypotension were the most common reasons for bounceback among this group. During the intervention, 8 of 225 (3.6%) transfers bounced back, all of who were labeled "high risk." Being "high risk" was associated with a risk of bounceback (OR not calculable, p = 0.02). Aspiration risk (OR 6.9; 95% CI 1.6-30, p = 0.010) and cardiac arrhythmia (OR 7.1; 95% CI 1.6-32, p = 0.01) were independent predictors of bounceback in multivariate analysis. Bounceback rates trended downward to 2.8% in the final phase (p for trend 0.09). Eighty-five percent of providers responded that the pilot should become standard of care.ConclusionPatients at high risk for bounceback after transfer from the neuro-ICU can be identified using a simple tool. Early augmented multidisciplinary communication and care for high-risk patients may improve their management in the hospital.

29 keywords...

hide…