1.2 Quality Systems, Quality Control, and Westgard Rules Interpretation

Quality Assurance (QA) and Quality Control (QC) are the bedrock of clinical laboratory testing, ensuring that the results reported are accurate, reliable, and clinically useful. While these terms are sometimes used interchangeably, they represent distinct concepts in laboratory management.

Quality Assurance (QA) is a comprehensive, overarching system that monitors all aspects of the laboratory's operations. It encompasses the entire testing pathway: the pre-analytical phase (patient preparation, specimen collection, transport), the analytical phase (instrument calibration, reagent integrity, maintenance), and the post-analytical phase (result reporting, critical value notification, data interpretation). QA is proactive, aiming to prevent errors before they occur.

Quality Control (QC) is a specific, quantitative subset of QA focused solely on the analytical phase. It involves the regular assaying of control materials alongside patient samples to verify that an analytical process is functioning correctly. Control materials must mimic patient samples in matrix and have known, predetermined target values.

Statistical Foundations of QC

To interpret QC data, laboratorians rely on fundamental statistical metrics derived from a Gaussian (normal) distribution.

• Mean: The arithmetic average of a set of data points, representing the target value of the control.
• Standard Deviation (SD): A measure of the dispersion or spread of data around the mean. In a normal distribution, approximately 68.2% of values fall within +/- 1 SD, 95.5% within +/- 2 SD, and 99.7% within +/- 3 SD. Clinical labs universally establish a +/- 2 SD range as the acceptable limit for a control result, representing a 95% confidence interval.
• Coefficient of Variation (CV%): An index of precision, normalizing the standard deviation relative to the mean. It is calculated as: CV% = (SD / Mean) x 100. A lower CV% indicates greater precision. Ideally, laboratory assays strive for a CV of less than 5%.

Accuracy vs. Precision

Understanding the distinction between accuracy and precision is a classic exam concept, often visualized using a target analogy.

• Accuracy: How close the measured value is to the true or accepted value (hitting the bullseye). Poor accuracy reflects systematic error.
• Precision: How closely repeated measurements of the same sample agree with each other (hitting the same spot repeatedly, even if not the bullseye). Poor precision reflects random error.
• Reliability: The combination of both accuracy and precision. An assay must be reliable to be clinically valid.

Types of Analytical Errors

1. Random Error: Fluctuations that occur without a predictable pattern. Affects precision. Causes include air bubbles in reagent lines, electrical voltage surges, improper pipetting technique, or fluctuations in temperature.
2. Systematic Error: A continuous, unidirectional shift away from the true value. Affects accuracy. It can be seen as a shift or a trend on a Levey-Jennings chart. Causes include degrading reagents, a failing light source, or a calibration shift.
   • Shift: An abrupt, sudden change in the mean that becomes continuous (e.g., installing a new, miscalibrated lot of reagent).
   • Trend: A gradual, progressive drift of QC values in one direction over several days (e.g., gradual deterioration of a reagent or a slowly failing lamp).

Levey-Jennings Charts and Westgard Multirule System

QC data is visually tracked using a Levey-Jennings (L-J) chart, which plots daily control values against the established mean and SD limits. To interpret L-J charts systematically and decide whether to accept or reject an analytical run, clinical laboratories utilize the Westgard Multirule System.

Dr. James Westgard developed these rules to optimize error detection while minimizing false rejections. It is crucial for the exam to differentiate between a warning rule and a rejection rule, and to identify whether a rule indicates random or systematic error.

Exam Trap: The R-4s rule only applies within a single run. It cannot be applied across different days or different runs. If the low control is -2.1 SD and the high control is +2.2 SD in the morning run, the R-4s rule is violated.

When a rejection rule is violated, patient results cannot be reported. The technologist must troubleshoot the assay, document the corrective action (e.g., recalibration, changing reagents), run the controls again, and verify they are within acceptable limits before releasing patient data.

Diagnostic Efficacy (Sensitivity and Specificity)

Beyond internal QC, a laboratory must evaluate the clinical utility of a test.

• Diagnostic Sensitivity: The ability of a test to correctly identify individuals who have the disease (True Positives). High sensitivity assays are used for screening because they have few false negatives. Formula: TP / (TP + FN) x 100.
• Diagnostic Specificity: The ability of a test to correctly identify individuals who do not have the disease (True Negatives). High specificity assays are used for confirmation because they have few false positives. Formula: TN / (TN + FP) x 100.
• Positive Predictive Value (PPV): The probability that a positive test result actually means the patient has the disease. Strongly influenced by disease prevalence.