Quality Assessment & Control in the Lab

• Quality Assessment (QA) or Quality Assurance constitute a comprehensive system for establishing and adhering to procedures and policies.
• The aims are to provide reliable patient laboratory results, reduce errors in the preanalytical, analytical, and postanalytical phases.

Preanalytical Factors

• Includes everything that precedes the performance of a laboratory test.

Analytical Phase

• Includes all aspects directly related to the assay itself.

Postanalytical Factor

• Everything occurring after the analysis of the test is complete.

Quality Control

• Emphasizes maintaining the validity of laboratory operations over time.

Method of Evaluation

• Used to verify the acceptability of new methodologies before reporting results for patients.

Errors in Testing Phases

• Errors may occur in three phases: preanalytical, analytical, and postanalytical.
• Examples that occur in the preanalytical phases are test orders, patient preparation, patient identification, specimen collection or processing
• Examples of analytic errors are test analysis, qc reagent, preventive maintenance
• Examples of postanalytic errors are verification of calculations, reference range, interpretation of results

Descriptive Statistics

• Descriptive statistics, including mean, help to statistically describe data
• Mean calculation is the sum of all observations divided by the number of observations.
• The mean is most applicable for measuring the center.

Median

• The median defines the middle point of the data.
• The median is useful with skewed data.
• To find it: rank ordered values from least to greatest, the middle value is selected.
• Median is calculated by (n+1)/2 term

Mode

• The mode is the value that appears most frequently in a dataset.
• Not commonly used to describe datasets.
• Instead, it's a reference to the shape of data.

Measures of Dispersion (Measures of Spread)

Range

• The range is the simplest expression of dispersion.
• The range of a dataset is the difference between the highest and lowest observations.
• The range is effective for small samples, but is susceptible to extreme values and outliers.

Standard Deviation (SD)

• The standard deviation is the dispersion of values from the mean
• You solve for the square root of the variance to find the standard deviation

Coefficient of Variation

• The coefficient of variation (CV) serves as an index of precision.
• CV is standard deviation expressed as % of the mean
• Increase CV = decrease precision

Variance

• Variance, or SD squared, signifies the difference between each value and the average of the data.

Parametric vs. Nonparametric Analysis

• Parametric tests assume observed values in a data set.
• Nonparametric tests do not make specific assumptions about the distribution of data

T Test vs. F Test

• T-tests are used for assessing accuracy and mean.
• F-tests are used for precision.

Method Evaluation

• It is used to verifying the acceptability of new methods prior to reporting patient results.
• Determine imprecision and inaccuracy
• Compare the errors (total errors) with the maximum allowable error.
• Total error>maximum allowable error= unacceptable and must be modified and reevaluated or rejected.

Measurement of Imprecision/Random Analytic Error

• Imprecision is the dispersion of repeated measurement around a mean.

Estimation of Inaccuracy/Systematic Analytic Error

• Inaccuracy is the difference between measured value and its actual value due to the presence of an systematic error.

Three Types of Study

• Recovery study, interference study, and COM study.

Quality Control

• Quality control ensures that tests remain reproducible and accurate.
• This is done by utilizing control specimens or validating methods over time.
• Quality control ensures accuracy and precision in the laboratory by utilizing quality of the reagents.
• This control is to verify stability of machine, quality of reagents, and check for technical errors.
• A control sample is chemically and physically like unknown specimens and is tested in the same manner.

Control Limit

• The control limit is a threshold at which the value is statistically unlikely (upper and lower limits).

Control Material

• Control material is exclusively analyzed for QC purposes.
• Should be clinically important range of the analyte at appropriate decision levels.

Two Kinds of Quality Control

Internal QC (Intralaboratory QC)

• Ensures daily accuracy and precision in the laboratory.
• This is achieved by including quality control reagents and analyzing them together with patient specimens.

External QC (Interlaboratory QC)

• Involves periodic proficiency testing programs.
• These programs provide samples of analytes with unknown concentrations to participating laboratories.

Quality Control Statistics

Calibration

• Calibration establishes the correlation between measured and actual concentrations and was formerly known as standard.
• A calibrator is a reference material with a known concentration.
• Calibration verification involves testing materials of known concentrations to ensure result accuracy throughout reportable result range.

Delta Check

• Delta checks compare individual patient results throughout the day or week, using computer detection to identify changes from earlier results.

Proficiency Testing

• Involves sending a series of unknown samples to the lab from a program.
• These samples analyzed to yield a true indication of accuracy.
• Acceptability criteria is provided by the proficiency program.
• Laboratories must incorporate proficiency testing into routine workflow
• Proficiency samples should be tested by bench tech staff who conduct patient testing.

Reference Interval Studies (Normal Range)

• Reference interval studies include the data points that define the range of observations.
• Therapeutic range is the reference interval applied to a therapeutic drug.

Three Main Categories of Application of Reference Intervals

• Diagnosis of a disease or condition.
• Monitoring of a physiologic condition.
• Monitoring therapeutic drugs.

Establishing a Reference Interval

• Is needed when no analytes exist in the lab to conduct comparative studies.
• May require conducting 120-700 individual studies.

Verifying a Reference Interval (Transference)

• Confirms validity of an existing reference interval for an analyte that uses the same type of analytic system.
• This is common and can require as few as 20 individual studies.

Characteristics of the Ideal QC Material

• Resembles human sample.
• Inexpensive and stable for long periods.
• If Human samples
• Control samples = Bovine (1 year)
• If Reagent
• Once opened = can expire
• Powdered form more stable
• No communicable diseases.
• No matrix effects / known matrix effects.
• With known analyte concentrations.
• Convenient packaging for easy dispensing and storage.

Three QC Program as a Three-Stage Process

• Establish allowable statistical limits of variation for each analytic method.
• Use these limits as criteria for evaluating the QC data generated for each test.
• Taking action to remedy errors when indicated
• Assaying control specimens and standards along with patient specimens provides following functions
• Provides a guide to the functioning of equipment, reagents, and individual technique.
• Confirms the accuracy of testing when compared with reference values.
• Detects an increase in the frequency of both high and low minimally acceptable values (dispersion).
• Detects any progressive drift of values to one side of the average value for at least 3 days (trends).
• Demonstrates an abrupt shift or change from the established average value for 3 days in a row (shift).

QC Oversees

• QC oversees each procedure for an established protocol to ensure the quality of the results.
• Usually normal and abnormal control samples are analyzed at the same time patient specimens are analyzed.

QC Control Charts

Cumulative Sum Graph

• Gives earliest indication of shift or trend.

Youden/Twin Plot

• Best plot for comparison of accuracy and precision among clinical labs.

Shewhart Or Levey-Jennings Chart

• Best plot indicating all types of quality control errors
• Most widely used QC Chart

Gaussian Bell-Curved

• Bell-shaped curve, normal distribution, median=mean=mode

Multirule Procedures

• Establishes the criteria for judging whether an analytic process is out of control.
• Example: 13s = one data point cannot exceed 3 SDs

Shift

• Six consecutive values on the same side of the mean (upward).
• It's most common cause is improper calibration of instrument.

Trend (Drift)

• Trend (Drift) are Control values increasing and decreasing for 6 consecutive runs.
• Common cause: control samples and reagents were deteriorated

Dispersion

• Dispersion is observed when random error of lack of precision increases.

Outlier

• A control result outside the established limits.

False Rejection

• Rejection of run because QC results indicate that a problem when none is present.

Accuracy

• Accuracy refers to the closeness of the result obtained to the true value.

Precision (Reproducibility)

• Precision refers to obtaining the same value in subsequent tests on the same sample.
• A procedure can be precise but not accurate.

Diagnostic Efficiency

Diagnostic Sensitivity (Screening Test)

• The diagnostic sensitivity identifies the proportion of cases with the disease that tests positive.
• Requires True Positive assays.
• High sensitivity and low specificity is appropriate for screening tests.

Diagnostic Specificity (Confirmatory Test)

• Identifies the proportion of cases with a specific disease or condition that give a negative result.
• Requires true Negative assays.
• Low sensitivity and high specificity is appropriate for confirmatory tests

Analytical Specificity

• Analytical specificity tests for the lowest concentration of a substance that can be detected by test method.

Analytical Sensitivity

• Measures for the analyte of interest.

Positive Predictive Value

• Number of abnormal results who have the disease

Negative Predictive Value

• Number of normal test results who do not have the disease

Analytic Methods

• Analytic methods characterize radiant energy from short wavelengths to long wavelengths X-Rays, UV rays, microwave, radio waves, gamma, cosmic, infrared
• Wavelength is the distance traveled by one complete wave cycle measured in nanometers (nm). Is inversley proportional to the frequency and energy

Spectroscopy and Spectrometry

Instruments

• Includes spectrophotometry, atomic absorption, and mass spectrometry

Luminescence

• Includes Fluorescence and chemiluminescence

Electroanalytic

• Includes Electrophoresis, potentiometry, and amperometry

Chromatography

• Include Gas chromatography, liquid chromatography, and thin-layer chromatography

I. Photometry

• Spectrophotometry is an optical instrument that measures light transmitted to measure concentration of a light-absorbing analyte.

Types of Spectroscopy/Spectrometry

Emission Spectroscopy

• Includes Fluorometry, Flame photometry and Atomic emission spectroscopy

Absorption Spectroscopy

• Includes UV/ visible spectroscopy, infrared spectroscopy, and nuclear magnetic resonance spectroscopy (NMR).

Beer's Law

• Notes the relationship between the absorption of light by a solution and the concentration of that solution.
• Describes solution concentrations: a substance concentration is directly proportional to the amount of light absorbed and inversely proportional to the logarithm of the transmitted light.

Absorbance

• Is defined as A = 2 - log %T or A = log (100 / %T).

Single Beam Spectrophotometer

• It is the simplest type of absorption spectrometer, making one measurement at a time at one specified wavelength.

Double Beam Spectrophotometer

• Splits the monochromatic light into two components.
• Absorbance of the sample can be recorded directly.

Types

• In space: uses two photodetectors
• In time: one photodetector with one chopper [rotating mirror]

Spectrophotometer Components

• Light Source : Produces an intense, reproducible, constant beam of polychromatic light.
• Entrance Slit- Minimizes extraneous from entering monochromator
• Monochromator: Disperses light into isolated wavelengths
• Exit slit: Select wavelength that asses through cuvette to the detector
• Cuvette: Holds test solution. Increase sensitivity by 10x increase
• Bandpass/Spectral bandwidth: a range of wavelengths transmitted between 2 points.

Photodetector

• Photodetector: Converts transmitted light into photoelectric energy.
• More light transmitted = greater electrical signal measured

Quality Control Tests for Spectrophotometers

• Check For: Wavelength accuracy, Stray light and Linearity

B. Flame Emission Photometry

• It measures light emitted by a since atom burned in a flame.
• It's used Measurement of excited ions, and principle is excitation

Volumetric (Titrimetric)

• An unknown sample is made to react with a known solution in the presence of an indicator (Chloride test/ EDTA titration method)

V. Electrophoresis

• It is the migration of charged particles in an electric field.
• Is used for the separation, quantitation, and determination of molecular weights of proteins and peptides, Analysis of PCR products.

Electrophoresis vs Isoelectric Focusing

• Electrophoresis separates molecules by macromolecules and their migration, molecules by PH gradients.