Related Experiment Videos
A clinical evaluation of various delta check methods
Clinical Chemistry
|January 1, 1981
Summary
Delta check techniques can identify laboratory errors, but often flag normal changes due to disease or treatment as false positives. Careful review is needed to distinguish true errors from physiological variations.
Area of Science:
- Clinical Chemistry
- Laboratory Medicine
- Medical Diagnostics
Background:
- Delta checks are used to detect errors in laboratory testing by comparing sequential results.
- Evaluating the performance of different delta check methods is crucial for improving laboratory quality assurance.
Purpose of the Study:
- To assess the effectiveness of three distinct delta check techniques in identifying potential errors in continuous-flow laboratory test results.
- To quantify the rate of true positives versus false positives generated by these delta check methods.
Main Methods:
- Analysis of 707 unselected pairs of continuous-flow test results.
- Application of three different delta check methods, including specific analytes like urea nitrogen/creatinine ratio and anion gap.
- Investigation of delta check failures through subsequent tests, specimen retesting, and patient chart review to classify as true or false positives.
Main Results:
- The percentage of true positive delta check failures ranged from 5% to 29% across the evaluated methods.
- All three methods exhibited variability, with certain test types yielding higher percentages of true positives than others.
- A significant number of false positives were observed, often due to physiological changes from disease processes or therapeutic interventions.
Conclusions:
- Delta check methods are valuable for detecting laboratory errors that might otherwise be missed.
- The clinical utility of delta checks is tempered by a substantial rate of false positives, necessitating thorough investigation.
- Understanding the impact of patient-specific factors like disease and therapy is essential when interpreting delta check results.