Guest Essay
Invitation for Patient Percentile Monitoring
Dr. Dietmar Stöckl and Dr. Linda Thienpont are embarking on an ambitious new project of performance monitoring. This is an open invitation for the pilot stage of an exciting new effort at using patient percentiles as a way to fulfill quality indicator requirements of the ISO 15189 examination phase.
INVITATION FOR PARTICIPATION
Utility of patient percentile monitoring as ISO 15189 quality indicator for the examination phase
Dietmar (
Project
Continuous monitoring of outpatient percentiles (50th percentile or median at the start; in a later phase also the 10th and 90th percentiles) across laboratories and manufacturers by STT Consulting/Laboratory for Analytical Chemistry (Ghent University) (IQC monitoring in the 2nd phase); peer group sizes of 20 or more laboratories are intended.
Participants: high volume laboratories using homogeneous systems from the 5 main manufacturers (instrument, reagent, calibrator from the same manufacturer).
Laboratory contribution
IT surrounding that is capable of “automatic background generation” of daily out-patient percentiles (IQC data in the 2nd phase) and transfer of those data to an external receiver (STT Consulting).
An exemplary solution (e-mail transfer, no costs) has been worked out for laboratories that use the GLIMS laboratory information system from MIPS, Belgium (see below).
STT/UGent contribution
Development of data visualization and interpretation software based on graphical presentation of moving medians and summary statistics; regular individual reports to laboratories; regular, general reports about assay quality and stability. After a successful pilot study, a user interface shall be developed that gives the individual laboratory access to the peer group comparisons.
Costs
Participation in the percentile/IQC monitoring project will be free of charge during the pilot phase.
Timeline
- Single laboratory participation: immediately
- Peer Group recruitment: March 2014
- Establishment of MySQL database and application: December 2013
- Reporting: sporadic reports on 1st experiences: October 2013
- Regular reports: March 2014
- Duration of overall pilot phase: December 2015
- Development of user interface to the database: January 2016 (when pilot is successful).
Data Ownership
Data ownership will be addressed in a contract.
Data use
Every laboratory can use its own data for its own purposes.
Use of data comprising different laboratories will be addressed in a contract.
Benefits for the laboratory
Direct, internal benefits
Patient percentile monitoring: gives the laboratory a direct, real-time quality indicator for the analyses of patient samples, as requested by the ISO 15189. Added values of the quality indicator are:
- Monitoring the effect of analytical (in)stability on a surrogate medical outcome, for example, the frequency of “flagged results”;
- Comparison of patient data among laboratories belonging to a network;
- Monitoring is done by an independent third party.
IQC monitoring: shows the laboratory’s performance in its peer group, as offered nowadays by commercial combined IQC/EQA programs.
Combination of patient-and IQC data monitoring: creates evidence about mid- to long-term variation of the instrument, calibrator, and reagent of the manufacturer (lot-to-lot consistency, drifts). This evidence is backed-up by information from other laboratories using the same assay.
Linking patient data with IQC information generally strengthens the laboratory’s quality management/quality assurance system (see ISO 15189).
Longer-term, internal benefits
- Creation of evidence about the reasons for assay variation: own performance or manufacturer performance (for example, lot-to-lot-variation).
- Strengthening of the laboratory’s position in claims versus the manufacturer.
- Creation of a tool for developing realistic quality goals and a tool for strengthening the physician/laboratory interface by more transparent communication of performance.
- Provision of data about assays from other manufacturers, which might be helpful in decisions for the acquisition of new instruments.
External (“Profile”) benefits
Laboratories can develop a “high-profile” and set peer-performance standards.
- Can document long-term performance when applying for contracts for clinical studies and when applying for government contracts for creating epidemiological data.
- Can act as a group in the improvement of the laboratory/clinician interface, in the relationship with manufacturers, in the relationship with health-care authorities, and ministry of economics.
- Show fit for future tasks, such as contribution to the development, implementation, and control of global health-care policies. Its high profile may be an advantage in the ever increasing battle for money.
Example: How Data will be transmitted for patient percentile monitoring
e-mail to
______________________________________________________
PERC_STUDY
______________________________________________________
UZG;29/05/2013;COBAS8000C;POL;ALB;g/L;47.50
UZG;29/05/2013;COBAS8000C;POL;ALKFOS;U/L;61.70
UZG;29/05/2013;COBAS8000C;POL;ALT;U/L;19.00
***
***
***
UZG;29/05/2013;COBAS8000C;POL;PROT;g/L;69.50
UZG;29/05/2013;COBAS8000C;POL;UREUM;mg/dL;37.00
UZG;29/05/2013;COBAS8000C;POL;URINZ;mg/dL;5.64
______________________________________________________
Codes
Lab identifier;Date(DD/MM/JJJJ);Instrument ID;Outpatient code;Analyte;unit;median (or 50th percentile).
20 Analytes (serum or plasma)
Albumin; Alkaline Phosphatase; Alanine aminotransferease (ALT); Aspartate aminotransferase (AST); total-Bilirubin; Calcium; total-Cholesterol; Chloride; C-reactive protein (CRP); Gamma-glutamyl transferase (GGT); Glucose; Potassium; Creatinine; Lactate dehydrogenase (LDH); Magnesium; Sodium; Inorganic phosphor (phosphate); total-Protein; Urea; Uric acid (urate)