Performance Monitoring of External Quality Assurance (Current)
Description
This project is investigating if an early warning system can be established to help minimise risk to patients and assist laboratories in addressing external quality assurance (EQA) concerns early. The Royal College of Pathologists of Australasia Quality Assurance Programs Pty Ltd (RCPA QAP) previously ran a RCPA – Quality Assurance Programs Key Indicator Project trial which will be revisited for chemical pathology (data had been collected over four years). KPIs are now being developed for other pathology disciplines, notably anatomical pathology and transfusion serology, and evaluated as indicators. This project will harmonise with current Key Incident Monitory & Management Systems (KIMMS) initiatives.
to develop pathology laboratory performance from a number of perspectives.
Objectives
to examine pathology laboratory performance from a number of perspectives
to establish a working party with the Therapeutic Goods Administration (TGA) to develop ways to identify sub-standard tests and kits that can have a direct effect on pathology laboratory performance
to review four years of KPI data and correlate with National Association of Testing Authorities (NATA) accreditation visits
to develop novel protocols to use ongoing EQA results to monitor performance
to establish mechanisms to use EQA data to help monitor quality of tests.
This project is current and the aim and objectives are in the process of being achieved.
Outcomes
As at December 2011 the following had been achieved:
KPI systems and benchmarks for acceptable performance had been developed and finalised for transfusion medicine, cytopathology and anatomical pathology
a second round consultation phase on the project’s draft framework had been completed
software development included definition of overall project specification documentation, scoping of anatomical, transfusion and cytopathology program specifications and production of project design documentation
finalisation of the design of Performance Summary Reports for anatomical pathology, cytopathology and transfusion medicine
a cumulative score for transfusion medicine developed to reflect performance over six surveys
the RCPA QAP is currently exploring the issue of sending participant reports directly to NATA.
NATA File Audit – Risk Analysis of Assessment Non Conformances Identified in Pathology Laboratory Assessment Accreditation (Current)
Description
The National Association of Testing Authorities (NATA) is conducting an analysis of non-conformances identified during its assessments of pathology laboratories against the accreditation materials over the past three to six years. This project complements other QUPP funded initiatives, such as the Performance Monitoring of External Quality Assurance projects, as part of the pathology risk minimisation strategies to improve patient safety following the 2009 Budget initiative to continue the development of key performance indicators and other risk identifiers in pathology service provision.
Grant Recipient
NATA
Aims and Objectives
to identify common issues and trends in pathology laboratory assessment
to establish trends and patterns in poorly performing pathology laboratories
to examine corrective actions that are currently being undertaken by laboratories to address deficiencies documented in assessment reports
to develop a basis to articulate strategies that will address causes of poor performance.
This project is current and these aims and objectives are in the process of being achieved.
Outcome
NATA has developed risk matrices to assist with their examination of common issues and trends in poorly performing laboratories, and is continuing to progress their review of files.
Key Incident Monitoring & Management Systems (Current)
Description
The Key Incident Monitoring and Management Systems (KIMMS) of Pathology Laboratories assist pathology laboratories to identify, measure, monitor and investigate incidents in pathology which have the potential to cause harm, and investigate the root cause of these incidents so they are minimised, which in turn will minimise potential harm to patients. It also addresses the quality assurance (QA) issues in the areas of both analytical and non-analytical errors. The pre-analytical phase focuses on identification/labeling of the sample and the sample rejection rates, while the post-analytical of testing examines results corrected because of laboratory error and results reported to the wrong doctor. The project analysed the 2007-2008 KIMMS pilot data collection.
Grant Recipient
Royal College of Pathologists of Australasia Quality Assurance Programs Pty Ltd (RCPA QAP)
Objectives
to assist laboratories to measure and monitor incidents in the pre-and post-analytical phases in the pathology request-test-report cycle
to provide data collection and reporting in four areas:
pre- and post-analytical phases
how each incident was identified (by the laboratory or by a complaint)
the outcome of the incident (actual or potential patient harm)
overall incident rate as a percentage of workload each quarter
to raise awareness of safe work practices which will in turn reduce errors and increase patient safety
to reduce the number of incidents in pathology laboratories which will directly impact on patient safety.
These objectives were achieved by this project, although the final objective is a longer-term deliverable.
Outcomes
The RCPA QAP made a submission under Part VC of the Australian Health Insurance Act 1973 for the KIMMS program to be formally declared as a Quality Activity to protect laboratories from litigation due to data collection on adverse incidents, especially those who voluntarily disclose this information.
The RCPA QAP has employed three Information Technology (IT) staff to move software development in house. The new software developed will be housed on RCPA QAP servers under RCPA QAP control, and will be a repository for the aggregated KIMMS data.
KIMMS data will raise awareness of unsafe work practices and provide information to the relevant sections of the medical community so targeted solutions can be implemented to improve patient safety.
KIMMS will introduce graded classification of patient harm in 2010.
KIMMS data can be utilised to set achievable national benchmarks for good pathology practice.
Findings
KIMMS data confirmed their quality system captures most incidents (97%) allowing the pathology service to correct the problem before they cause patient harm.
KIMMS 2007 & 2008 pilot showed that between 70-80% of sample misidentification problems were due to specimens not meeting minimum labeling requirement of providing two identifiers (e.g. name and date of birth).
The overall incident rate was 0.7% in 2007 (fourth cycle) and 0.9% in 2008. Although these rates are low, the numbers (114,000 incidents annually from only 24 laboratories is a small fraction of Australian pathology) indicate the staggering resources used to stop non-laboratory errors affecting patient safety. These simple errors are all preventable.
Overall, pre-analytical incidents accounted for the majority of incidents (85-90% in 2008) which is consistent with the literature.
Post-analytical phase of testing incident rates were 0.01% of all accessions. These include results retracted due to an error by the laboratory and then re-issued, and very low levels where results were sent to the wrong doctor.
The incident rates in the pre- and post-analytical phase of testing is under-reported as many laboratories do not currently record these incidents in a way that allows easy consolidated reporting.
The absolute rate at which pre-analytical problems occurred exceeded the rate at which analytical problems occurred as measured in the details framework for quality assurance programs (QAPs).
There is a need to address poor sample collection technique.
Incident rates increased during this study as participants improved data extraction and capture techniques, but should eventually start to fall as:
various sections of the medical community are educated in safe work practices.
Laboratory information systems (LIS) needed to be configured to capture the data at the time of entry of patient information into the system.
Recommendations
Standardising key definitions and terms for incident monitoring and reporting is an issue that needs to be addressed to accommodate comparison of data.
Contact LIS providers to discuss incorporating KIMMS data requirements into future versions of LIS software.
Key Project Learnings
The significant challenge that pathology services experience is the majority of errors (75%) are not caused by pathology. Most occur when a sample is collected by a non-pathology person and are due to patient misidentification, incorrect sample collection and the sample being mislabeled. These inappropriate practices can jeopardise patient safety, particularly for irreplaceable samples (neonates, biopsy material and spinal fluids) where the laboratory has no option but to process the sample and provide a (qualified) result.
Many participants in the pilot study disliked the KIMMS classification of actual and potential harm to patients.
Feedback suggested a more accurate measure for sample recollections would be sample rejections as laboratories could capture this data.
Evaluating the outcome of an incident caused confusion and concern for many participants.
The pilot phase of this project highlighted the difficulty laboratories experienced in extracting or collecting of data from current LIS which are not set up to record and report/extract this data. A 2006 questionnaire sent to all pilot participants highlighted that 45% or participants collected incident data manually (in an exercise book) and a further 30% recorded data in a spreadsheet or a database.
Both pilot data and feedback indicated that pathology services were not systematically collecting data on incidents, while others who do were not doing so in a manner that allowed comparison or peer review.
Follow on Initiatives and Projects
Promotion of successful strategies on the RCPA QAP website.
Areas for Future Consideration
Standardise key definitions and terms for incident monitoring and reporting.
Consider strategies to address the issue of errors occurring when a non-pathology person collects a pathology sample.
