The Science Journal of the American Association for Respiratory Care

2001 OPEN FORUM Abstracts

Results of a Ventilator Management Improvement Program

C Haas RRT, M Konkle RRT,R Dechert RRT, L Folk RRT, P Loik RRT, L Stapp RRT, V Stevenson RRT, A AndrewsRRT, C Lane RRT, S Lawrence RRT, L Young RRT, A Benschoter RRT, K Marecle RRT,T Behm RRT, S Mack, S Gay MD, JG Weg MD Respiratory Care Department, Universityof Michigan Health System, Ann Arbor Michigan

Background: A performanceimprovement team was formed in 1999 to determine improvement opportunities foradult mechanical ventilation (MV). The purpose of this abstract is to presenta before and after comparison of changes to the MV process.

Methods: Eight major timepoints of the MV process were identified: 1) MV initiated; 2) ventilator settingssuggest a wean assessment; 3) initial wean assessment done; 4) patient readyto begin weaning; 5) wean begins; 6) patient ready for extubation or MV to stop;7) patient extubated or support stopped; and 8) time of reintubation or initiationof other support within 48 hours. Data was collected by the bedside respiratorytherapist. Thresholds were established for the following segments: 1) 4-hrsfrom when ventilator settings suggest an assessment until it was done (Gap 1);2) 4-hrs from when patient was ready to begin weaning until it started (Gap2); and 3) 1-hr from when patient was ready to be extubated/stop MV until itwas accomplished (Gap 3). Unit specific reports of the initial findings wereshared with each ICU medical director and nurse managers. Strategies were developedto improve each major segment of the MV process, including: a generic wean protocol,multiple MV and extubation policy changes, development of a RC care-plan tool,and RCP inservices. Before implementing the changes, 12 months of data was collected.Six months were allowed to implement changes between measurement periods. Theresults presented compare the same 6-month period: January through June of 2000(pre) vs 2001 (post).

Results: Data was gatheredon 1102 pre-change and 1196 post-change patients. The duration of MV was 75.6+ 135.6 hr (median=22.1 hr) for pre and 66.8 + 128.6 hr (median=18.1hr). The table shows the duration (in hrs) of various segments of the MV processand the time points used to calculate each segment in parenthesis.

Segments ofthe MV process
Pre(n=1102)
Post(n=1196)
 MeanSDMedianMeanSDMedian

1-Begin MV to ready for parameters(2-1)

13.1 40.22.1 13.962.31.9

2-Ready for parameters untildone (3-2)

20.948.1 7.816.741.96.2

3-Initial parameters to readyto wean (4-3)

9.132.20 8.327.40

4-Ready to wean until beginwean (5-4)

3.116.7 02.6 13.40

5-Begin wean until ready toextubate (6-5)

19.065.80.618.069.0 0.4

6-Ready to extubate until done(7-6)

4.817.30.84.115.10.8

Thepercent of patients exceeding the thresholds for Gaps 1-3 (pre vs post) are:Gap 1=54.3% vs 51.6%; Gap 2 =7.6% vs 5.9% and Gap 3=32.0% vs 30.6%.

Discussion: The median timeof segments 3&4 of 0 hr indicates that when initial parameters are taken,a majority of patients are immediately ready to begin weaning and the wean processis initiated. Although there appears to be a trend toward reducing the overallduration of MV and the duration of most segments of the process, we are unsurehow these actual times relate to predicted times based on the severity of illness.Comparing the actual times against predicted times, based on a predictive modelfor duration of MV, would help clarify whether improvement has indeed been made.The wide variety and severity of patient illness results in a large SD, whichmay obscure significant changes. Further analysis of sub-populations may identifysuch changes.

OF-01-171

You are here: RCJournal.com » Past OPEN FORUM Abstracts » 2001 Abstracts » Results of a Ventilator Management Improvement Program