Objective: To reduce the hospital-acquired pneumonia rate in stroke patients.
Background: The hospital-acquired pneumonia rate for stroke patients at our institution was 6.2% in 2014. This is the 39th percentile for the Joint Commission Comprehensive Stroke Centers in the University HealthSystem Consortium. Our goal was to achieve top decile performance by reducing our hospital-acquired pneumonia rate for stroke patients to 3.0%.
Methods: A multidisciplinary team was assembled with representation from administration, coding, nursing, resident physicians, and attending physicians from stroke, neurocritical care and infectious disease. The team analyzed benchmarking data and decided to use chart review to validate the data. The team decided to use the CDC’s definition and clinical criteria for hospital-acquired pneumonia. Further chart review was performed to find information on swallow evaluations, incentive spirometer use, oral care, and head of bed elevation. A coding manager reviewed the documentation and the coding decision for each patient. After results were presented, the team decided on a course of action.
Results: Review of stroke patients with hospital-acquired pneumonia showed 44% of the patients coded with pneumonia did not meet the clinical definition on physician review. When reviewed by a coding manager, all coding was found to be correct based on the documentation in the chart. To solve this documentation problem, attending physicians discharging stroke patients were informed of these results about unclear pneumonia documentation and were asked to clarify documentation if the diagnosis was not appropriate. After this change, the hospital acquired pneumonia rate in stroke patients fell from 6.2% prior to the changes from April 2014 to March 2015 to 1.9% after the changes were implemented from April to June 2015.
Conclusion: Our facility was able to reduce the rate of hospital-acquired pneumonia in stroke patients from 6.2% to 1.2% through a multidisciplinary approach to achieve more accurate documentation and coding. The widespread over-documentation showed it is clinically difficult to make a diagnosis of pneumonia in this population. Therefore, the rate of hospital-acquired pneumonia is not a valid measure of quality of care.