October 28, 2008
More on Computerized Trigger Tools
We’ve done several columns on use of trigger tools in patient safety. Our October 30, 2007 Patient Safety Tip of the Week “Using IHI’s Global Trigger Tool” is a good starting point to learn about what trigger tools can do for your organization. Trigger tools have consistently demonstrated an ability to identify far more adverse events than are identified by typical voluntary hospital reporting systems and formal incident reporting systems. However, we talked about the real potential of using hi-tech tools to enhance the trigger tool concept in our April 15, 2008 Patient Safety Tip of the Week “Computerizing Trigger Tools”.
Two excellent articles on computerized trigger tools have been published recently by the group at Duke University. Ferranti and colleagues (Ferranti 2008a) reported on their dual systems for reporting adverse drug events in adult inpatients, one a voluntary reporting system and the other a trigger-tool based computerized system. The authors feel that Duke’s non-punative culture of safety makes voluntary reporting likely to be utilized. Yet the adverse drug event rate detected by the computerized trigger tool detected substantially higher rates than did the voluntary reporting system (6.93 ADE’s per 1000 patient days vs. 1.96 ADE’s per 1000 patient days). The overlap between the computerized surveillance system and the voluntary reporting system was only 5.6%. Yet both systems contributed significantly to the overall patient safety program at Duke. The voluntary system provided insight into the system nature of medication errors and also identified a wider spectrum of events. The latter, of course, provide the opportunity to develop new rules to add to the computerized trigger system. Also, some types of events (such as drug omissions) may be picked up in the voluntary system but are not detected by the computerized system. The computerized system probably provides a better overall estimate of the frequency of ADE’s and is probably more accurate for trending ADE’s over time. The computerized system also detected some ADE’s that had occurred prior to admission. None of those were captured in the voluntary system. Overall, the dual approach to ADE detection appears to be very useful. Their study also pointed out the need for standardizing taxonomy in all systems.
The second paper (Ferranti 2008b) looked at the two systems in the pediatric population. The ADE rate was 1.8 per 1000 patient days in the voluntary reporting system and 1.6 ADE’s per 1000 patient days in the computerized system. Note that the computerized surveillance system did not perform as well in the pediatric population as in the adult inpatient population. The authors suggest the need for development of more rules specifically-tailored to the pediatric population. Nevertheless, the two systems were complementary. The voluntary system was better at identifying system failures and failures in the medication administration processes, the computerized system appeared better at identifying events related to high-risk medications.
These two articles provide great insight into the potential capabilities of computerized systems for ADE surveillance. Their description of some of the rules in the trigger system and their system for severity scoring are also well worth reading about. Yet they underscore the importance of using more than one method for detecting ADE’s and highlight the strengths and weaknesses of each, while demonstrating synergy.
And there are also two recent articles by David Classen, an original architect of the trigger tool concept. Classen and colleagues (Classen 2008) have refined the IHI Global Trigger methodology. They have used a methodology based on the old Harvard Medical Practice Study for chart review and used a 2-stage approach in which 4 clinical reviewers perform and initial review and then 2 physician reviewers perform a second review, followed by a consensus review. All reviewers reviewed a set of 15 training records, then a set of 50 test records. They were able to demonstrate substantial improvement in interrater agreement after training, both for the presence of an adverse event and also for the classification of the severity of the adverse event. They did limit the individual chart reviews to a maximum of 20 minutes, as commonly recommended by IHI.
Griffin and Classen (Griffin 2008) summarized the results of IHI’s Perioperative Safety Collaboborative. At the 11 hospitals participating, surgical adverse events were identified in 14.6% of patients and the rate of surgical adverse events was 16 per 100 patients. Almost half the events resulted in increased length of stay and 8.7% resulted in harm or required life-saving intervention. As in most other studies using trigger tools to identify adverse events, many more were identified than had been reported in voluntary incident reporting systems.
Trigger tools are here to stay. They provide great ways to identify and trend a whole variety of patient safety events and undoubtedly will be expanded as we develop new rules for triggers. Their potential to also identify events at a stage where intervention may prevent harm to patients is exciting. But they’ll never be the sole tools you’ll use in your patient safety programs. They will, however, nicely complement many of your more conventional patient safety and quality activities.
Ferranti J, Horvath M, Cozart H, Whitehurst J, Eckstrand J, Pietrobon R, Rajgor D, Ahmad A. A Multifaceted Approach to Safety: The Synergistic Detection of Adverse Drug Events in Adult Inpatients. Journal of Patient Safety. 2008; 4(3):184-190
Ferranti J. Horvath MM. Cozart H. Whitehurst J. Eckstrand J. Reevaluating the safety profile of pediatrics: a comparison of computerized adverse drug event surveillance and voluntary reporting in the pediatric environment. Pediatrics 2008; 121(5):e1201-e1207 http://pediatrics.aappublications.org/cgi/reprint/121/5/e1201
Classen, David C. MD, MS *; Lloyd, Robert C. PhD +; Provost, Lloyd PhD +; Griffin, Frances A. RRT, MPA +; Resar, Roger MD + Development and Evaluation of the Institute for Healthcare Improvement Global Trigger Tool. Journal of Patient Safety. 4(3):169-177, September 2008