Triggers and Targeted Injury Detection Systems (TIDS) Expert Panel Meeting: Conference Summary Report

You Are Here: AHRQ Home > Quality & Patient Safety > Medical Errors & Patient Safety > Triggers and Targeted Injury Detection Systems (TIDS) > Trigger Tool Implementation Experiences in Kaiser Permanente

Trigger Tool Implementation Experiences in Kaiser Permanente

Dot Snow, M.P.H.^a

Background

Kaiser Permanente is the Nation's largest not-for-profit health plan, serving almost 9 million people. It is noted for being an integrated system with a focus on prevention: medical services are integrated with wellness activities, and care delivery processes are integrated with health plan operations. Kaiser Permanente has local markets in 8 geographical areas of the United States and operates with 32 hospitals, over 400 medical office buildings, approximately 13,000 physicians, and over 159,000 employees.^H1 Representing the world's largest civilian deployment, Kaiser Permanente's electronic medical record is called KP HealthConnect™. It provides a complete health care system by integrating clinical care, practice management, ancillary service systems, and online access to a personal health record.

Trigger Tools

For this panel discussion, three experiences with trigger tools within Kaiser Permanente will be addressed: an initial pilot of the IHI (Institute for Healthcare Improvement) Global Trigger Tool (GTT) in 2006; a focal trigger tool pilot of IV (intravenous) heparin therapy in 2007; and our current Automated Adverse Event Monitoring Program (AAEMP) project, which seeks to automate the IHI Global Trigger Tool.

Initial pilot of IHI GTT: The methodology was pilot tested at six medical centers across Northern and Southern California regions in 2006. Reported data representing 2,363 patient days across 400 patients came from 2 medical centers, 1 from each region, where data were collected during the same time period (January-October 2006). Consistent with previous studies, harm was detected by the trigger tool in over 30 percent of the charts reviewed.^H2 Approximately 20 percent of the patients experienced more than one harm event. Fifty-three percent of the adverse events fell in four categories:

Medication events, 22 percent.
Infection, 15 percent.
Surgical complications, 8 percent.
Blood pressure management (primarily associated with treatment of hypertension that resulted in hypotension and volume resuscitation or other complications), 8 percent.

Focal trigger tool pilot of IV heparin therapy: The methodology was piloted at one medical center. Twenty patient charts representing 50 days of treatment were reviewed by a multidisciplinary team comprised of an inpatient pharmacist, risk director, quality director, and nurse. Fifty percent of patients had positive triggers, including a drop in platelets, unplanned readmissions within 30 days, red cell transfusions, antifactor Xa >1.4, and administration of Narcan. Twenty-five percent experienced harm, with 35 percent of patients experienced thrombocytopenia. While the pilot confirmed a high rate of compliance with the double-check policy, several improvement opportunities were identified.^H2

Automated Adverse Event Monitoring Program (AAEMP): This foundation-funded project represents a collaborative effort between Kaiser Permanente and Computer Sciences Corporation to automate the IHI GTT methodology utilizing data from the electronic medical record (EMR). The project involves both the development of an application that searches the EMR of hospitalized patients for positive triggers and a pilot test to integrate the AAEMP into operations. The project is currently in the process of developing the application and preparing the first pilot site for implementation.

Challenges and Benefits

What have been the one or two most significant challenges your team/organization has faced when implementing triggers/TIDS at the initial development site? Beyond the initial site?

Paper-based methodology: Resourcing and a perceived lack of actionability of the findings from trigger tool reviews have been the two main challenges. Implementing paper-based trigger tools does not replace carrying out existing surveillance processes, so additional time and personnel to undertake the activity must be identified. More importantly, a basic value proposition must emerge from the trigger tool review activity—what new information is it telling us? Some medical centers discontinued the trigger tool activity because sampling across several categories led to small numbers of adverse events associated with individual triggers, and improvement activity based on one or two cases was not compelling.

Automated methodology: The project is still in the application development phase. It is anticipated that the automation of the trigger search across a hospitalized population will identify greater numbers of positive triggers to be evaluated for harm. Larger numbers may yield consistent patterns that direct improvement efforts. It is expected that the AAEMP will be more effective at detecting harm than existing surveillance efforts; hence, it may replace instead of augment current efforts at detection and measurement, thus reducing the challenge associated with resourcing the activity.

Discussion:With the paper-based trigger tool methodology, implementation beyond the initial site can be colored by the experience of the initial site. If the results were perceived to be inactionable or too resource intensive to integrate into operations, some centers may elect to forgo further trigger tool review. With the automated trigger tool methodology, it is expected that implementation challenges will be greatest at the initial site as the operational workflow processes are worked out. Subsequent site implementations should be able to benefit from the experience of the initial site. An automated trigger tool approach will yield more reliable measures of harm and shine the light on areas of improvement opportunity.

One region has incentivized medical center leadership to put trigger tool reviews of high-alert medications in place by the end of 2008. In order to receive full bonuses, the medical center must demonstrate that a high-alert medication trigger tool process is in place and that learnings from the activity are put to use in performance improvement processes.

What are the incremental benefits of TIDS compared with current systems for drug management and surgical quality initiatives? What have been your experiences using TIDS-like systems as a separate surveillance system vs. alerts going directly to the patient care team?

Kaiser Permanente saw several benefits emerge from the focal review of IV heparin therapy. The approach of bringing in a multidisciplinary team to review the patient charts and focus on that particular therapy led to the establishment of new relationships among the clinicians, risk manager, and quality director. There was also problem-solving and learning about both the trigger tool review as well as the clinical care processes. The improvement opportunities that were identified led to clearer logic for the regional heparin algorithm, improved documentation of double-checks in the emergency department, and improved training of nurses in the overall heparin protocol.

There is value derived from focusing on one category of harm at a time. Medical center leadership can use shortcuts and a strategic approach to utilizing trigger tools in the measurement of harm experienced in their medical center. Certain categories in the GTT can be eliminated, either by the absence of that service within their medical center or by known competent performance. Remaining categories can be prioritized for focal review based on known significant events and/or vulnerabilities.

Conclusions

Perhaps it is not an "either/or" proposition but rather "and/both" for targeted injury detection systems (TIDS) and quality-of-care initiatives. Compliance with evidence-based clinical processes is the heart of the Surgical Care Improvement Program (SCIP), and a focal trigger tool study of surgical care could yield information about outcomes.

The initial GTT activity identified the areas where harm is occurring in the Kaiser Permanente delivery system, and it appears to be consistent with the areas of harm reported in a recent Medicare study on medical errors, particularly infections and postoperative complications.^H3 Focal trigger tool studies of the individual categories would generate larger samples, reveal consistent patterns, and provide understanding of that terrain. The data would be taken to the appropriate committees to decide if performance is acceptable or whether improvement work is needed.

The long-term view would be to systematically conduct focal reviews for all relevant harm categories and work the improvement opportunities as identified. Automating the process should lead to more reliable triggers to capture harm. Once a cycle of all categories is completed, the delivery system could use the GTT to keep a finger on the pulse, maintaining the level of performance achieved from the improvement efforts. While it is undeniable that the category-by-category approach for reducing harm involves lots of time and lots of work, people are willing to participate as long as they feel the work is actionable and leads to real improvements in the quality of care delivered to patients.

References

H1. Kaiser Permanente. Fast facts about KP. Available at http://xnet.kp.org/newscenter/aboutkp/fastfacts.html. Accessed June 12, 2008.

H2. Schilling L, Rubino M, Golubjatnikov K. Trigger tools. Presentation at Kaiser Permanente National Quality & Brand Conference, 2007 May 7-10; Long Beach, CA.

H3. Medical errors costing U.S. billions. Washington Post 2008 Apr 4.

^a National Patient Safety, Department of Care and Service Quality, Kaiser Permanente Foundation Health Plan.

Previous Section Contents Next Section