Chapter 2 Quality Assurance Cycle PDF

Chapter 2 The Quality Assurance Cycle Learning Objectives Upon completion of this chapter the student will be able to: 1. Identify parameters of pre-analytical performance 2. Discus method selection and evaluation Chapter 2 Outline The Quality assurance cycle  Pre analytical phase...

Chapter 2 The Quality Assurance Cycle Learning Objectives Upon completion of this chapter the student will be able to: 1. Identify parameters of pre-analytical performance 2. Discus method selection and evaluation Chapter 2 Outline The Quality assurance cycle  Pre analytical phases of quality assurance  Method selection and evaluation  Method selection  Method evaluation  Objectives of method evaluation  Establishing a working plan Chapter 2 Outline The Quality assurance cycle  Linearity check  Replicate experiment  Recovery studies  Interference experiment  Check for Precision  Linear regression and correlation  Comparison of methods when there is reference method  Correlation co-efficient Lecture Outline  Upon completion of this lecture the student will be able to:  Describe the pre- analytical phase of quality assurance  Discuss method selection and evaluation  Discuss objectives of a method evaluation  Explain establishing a working plan The Quality Assurance Cycle Patient/Client Prep Sample Collection Personnel Competency Reporting Test Evaluations Data and Lab Management Safety Customer Service Sample Receipt and Accessioning Record Keeping Quality Control Sample Transport Testing Pre-analytical Quality Assurance Pre-analytic laboratory procedures and processes need to be standardized in order to have  reproducible laboratory results and  uniform activities in the laboratory.  This is achieved through Standard Operating Procedures (SOP) which should be strictly followed by all staff involved in laboratory services, which constitute and impact the overall quality of services to patients and public as a whole. Pre-analytical Quality Assurance Activities: Specimen collection and transport conditions Pre-Analytical Quality Assurance Activities: Specimen Receipt and Processing Specimen Transport to  Testing lab  Referral lab Pre-analytic Quality Assurance Involves Specimen  Collected properly Identification and label Container and additive  Storage and transport temperature  Processing such as adding preservative centrifugation or protein free filtrate Equipment Reagents  Receipt  Referral or delivery to testing Method Evaluation Definition: systematic test of an analytical method to assess a procedure or instrument performance ERROR ASSESSMENT Objectives of Method Validation When getting a new method of testing or a new instrument  Precision and accuracy must be tested Replicant studies - Replicant studies involve repeating the same experiment multiple times to ensure consistency and reliability of results. Linearity - Linearity refers to how well the response (like a measurement) changes in a consistent, proportional manner with the concentration or amount of the analyte. It’s usually tested to ensure that the method can accurately measure across a range of concentrations.  Interference - Interference testing checks if other substances in the sample might affect or distort the measurement of the analyte. Compare with previous method 2 Types of Error 1. Random (imprecision) - Random errors are unpredictable fluctuations that can happen during any measurement, often due to small, uncontrollable factors. - These errors lead to small, random variations in your measurements each time you repeat the test, under the same conditions. 2. Systematic (inaccuracy) Systematic errors are consistent and repeatable inaccuracies that affect every measurement in the same way. typically arise from issues with the measurement system itself, like poorly calibrated instruments, biases in sample handling, or errors in the procedure. Method Selection A new method or instrument should be chosen based on:  Medical usefulness Screen Diagnostic/Prognostic  Financialissues  Technical issues Application Methodology Performance Method Selection Technical Issues  Application Issues Instrumentation involved -Refers to the specific instruments used for the analysis and their capabilities. Equipment - Other tools or devices used in conjunction with instrumentation, like sample containers, pipettes, or consumables. Stat capability (random access)  Required personnel training -Ensuring that staff operating the equipment and conducting the analysis have the right skills and knowledge. Safety consideration -Identifying and mitigating risks associated with the testing process. Sample type and size -The nature of the sample being analyzed, including its physical and chemical properties. Anticipated workload and run size -The expected volume of samples to be processed over a given time period. Routine turnaround time - The time it takes to complete analysis Method Selection Technical issues  METHODOLOGY CHARACTERISTICS : chemical specificity of new method -This refers to the ability of the method to accurately and exclusively measure the target analyte optimum reaction conditions -The specific conditions under which the chemical reaction for the analysis occurs most efficiently and reliably. This includes factors like temperature, pH, solvent, and time.  manner of calibration -this is the process by which the method is standardized, ensuring that the results from the test are accurate and reproducible. determination of results -The process of interpreting the data obtained from the analysis. relative stability and expense of reagents and controls -refers to the shelf life and cost of the chemicals used in the method. Method Selection Technical issues  PERFORMANCE CHARACTERISTICS : Upper limit of linearity of new method: - The upper limit of linearity refers to the highest concentration or amount of analyte that the method can measure with a direct, linear relationship between the signal (e.g., absorbance, voltage) and the concentration of the analyte. Lower limit of sensitivity of new method: -The lower limit of sensitivity is the smallest amount or concentration of the analyte that the method can reliably detect.  Procedure to follow if test sample exceeds limit of linearity: -When a sample’s analyte concentration exceeds the upper limit of linearity, the standard procedure is to dilute the sample so that the concentration falls within the linear range, then re-test. Expected precision from the new method: Expected accuracy from the new method Establishing a working Method Evaluation Plan A plan for method evaluation needs to be set up  These steps must be part of the method evaluation 1) Precision Check -To assess the reproducibility and consistency of the method when repeated measurements are taken on the same sample under the same conditions.  2) Linearity check -To determine whether there is a linear relationship between the concentration of the analyte and the measurement (e.g., absorbance, signal, etc.).  3) Interference experiment -To test if other substances in the sample, aside from the target analyte, could interfere with the measurement and affect accuracy. 4) Comparison of methods when there is reference method -To evaluate the performance of the new method by comparing its results with those from a reference method  Chapter 2 Quality Assurance Cycle Part 2 Method Evaluation Objectives Upon completion of this lecture, the student will be able to: Describe how accuracy, precision and total error is measured Discuss what statistical parameters are used in accuracy studies Describe a linearity check Discuss what statistical parameters indicate the precision of a method Learning Objectives Upon completion of this lecture, the student will be able to: Describe recovery and interference studies. Discuss Linear regression and correlation Describe a study for comparison of methods when there is reference method Method Validation Precision and accuracy must be tested  Firsttest precision to test random errors  Second perform accuracy studies to test systematic errors Part of the process to evaluate method performance 2 Types of Error Random (precision error) Systematic (accuracy error) Total error (TE) is the sum of the random (RE) and the systematic error (SE) TE = RE + SE  Represents the sum of the variability of the measurement process (imprecision) and the shift from a true value (inaccuracy) Systematic Error Error that occurs in one direction only, increasing or decreasing results by the same amount Due to factors such as erroneous values for standards, incomplete calibration of shifts in reagent baseline Results in inaccuracy  Linearity Example (Cholesterol) 600.00 500.00 Measured Value 400.00 300.00 200.00 100.00 y = 1.001x - 0.289 0.00 0 100 200 300 400 500 600 Bottle Value Decision Levels 1 2 3 4 Cholesterol (mg/dL) 90 240 260 350 Linearity Studies %Recovery= Analyte Found​ X 100 Theoretical Amount Where as - Analyte Found: The amount of the analyte measured or detected in the sample after the recovery procedure (usually in the same units as the theoretical amount). - Theoretical Amount: The known amount of analyte that was added (spiked) to the sample.  Should be between 90% - 110% Perform linearity study to confirm analytic range. Linearity/Dilution Studies Example Std Result Result Result Average % Recovery mg/dL 1 2 3 50 49 50 49 49.33 98.67 100 99 101 102 100.67 100.67 150 151 149 149 149.67 99.78 200 202 199 200 200.33 100.17 250 249 251 248 249.33 99.73 300 298 302 300 300.00 100.00 400 398 400 401 399.67 99.92 500 502 500 500 500.67 100.13 Average % Recovery 99.88 Precision Error: Random Error Error that occurs unpredictably or randomly. Due to factors such as instability in instruments, measurement, temperature or reagents or imprecision in pipeting. Results in imprecision Increase causes test results to be more variable What do we calculate to measure random error? %CV and SD Replicant Experiment Random error Series of aliquots of the same test samples within a specified period of time  Short-term (within an analytical run) Intra-precision -Intra-precision refers to the consistency or reproducibility of measurements when the same sample is analyzed multiple times within the same analytical run or by the same operator.  Long-term (over a period of a month) Inter-precision (day to day; between analytical runs) -Inter-precision refers to the consistency or reproducibility of results when the same sample is analyzed under different conditions (e.g., different days, different analysts, or different equipment). Short-Term Replication Experiment INTRA PRECISION Select at least 2 different control materials that represent low and high medical decision pts. Analyze 20 samples of each material within a run or with a day to obtain an estimate of short- term imprecision. Long-Term Replication Experiment INTER PRECISION Select at least 2 different control materials that represent low and high medical decision pts. Analyze 1 sample of each of the 2 materials on 20 different days to estimate long-term imprecision. Recovery Studies – Calculation of Recovery In recovery studies, you test the ability of an analytical method to measure the correct amount of analyte in a sample by spiking the sample with a known quantity of the analyte. The recovery percentage indicates how much of the added analyte is successfully detected by the method. The recovery is calculated by comparing the measured amount of analyte in the sample (after spiking) to the theoretical amount of analyte that was added to the sample. Cont.. Percent Recovery=(measured amount of analyte/ theoretical amount of analyte) x 100 Where: -Measured Amount of Analyte: The amount of analyte detected in the sample after analysis. -Theoretical Amount of Analyte: The known amount of analyte that was added (spiked) to the sample. Example of Recovery Calculation: 1. Spiking the Sample: Let’s say you are testing the recovery of glucose in a blood serum sample. You know the serum has a low concentration of glucose (e.g., 50 mg/dL), and you add (spike) 100 mg/dL of glucose into the sample. Theoretical Amount of Glucose: 100 mg/dL (this is the amount you added to the sample). Cont … Analyzing the Sample: After spiking the sample, you analyze it using your glucose assay method, and the measured amount of glucose is 190 mg/dL. Measured Amount of Glucose: 190 mg/dL (this is what the method detected in the spiked sample). Percent Recovery=(190mg/dL/ 190mg/dL​ )×100=190% Interpretation of Results: A recovery of 190% suggests that the method detected more glucose than expected. This could indicate interference or a problem with the measurement process. Ideally, 100% recovery is desired, meaning the method should accurately measure the amount of spiked glucose without overestimating or underestimating. If recovery is significantly higher or lower than 100%, you may need to investigate the cause, such as issues with sample handling, matrix interference, or calibration errors. What if the Measured Amount is Lower than the Theoretical Amount? If the measured amount is lower than the theoretical amount, this suggests loss of analyte during the analysis, which can occur due to: Sample degradation: Some analytes may degrade over time, especially if stored improperly. Cont.. Matrix interference: Substances in the sample (e.g., proteins, lipids, salts) may interfere with the detection of the analyte. Inefficient extraction: The process of extracting the analyte from the sample could be incomplete, leading to a lower measured concentration. Interference Studies Measures constant systematic error  Checks for errors caused by other substances  Caused by the lack of specificity of the method Similar to analyte Test for errors from common interfering substances such as bilirubin, lipemia and hemolysis or other interfering analytes Percent interference calculation Percent Interference=(Measured Value with Inter fering Substance−Measured Value without Interf ering Substance​) ×100 Measured Value without Interfering Substance Where: - Measured Value with Interfering Substance is the analyte concentration measured in the sample after adding the interfering substance. - Measured Value without Interfering Substance is the analyte concentration measured in the original (unspiked) sample. Example of Interference Study Calculation: Let’s say you’re testing a glucose assay and you want to check if bilirubin (a potential interferent) affects the glucose measurement in serum. Original (Unspiked) Sample:  Measured Glucose Concentration: 90 mg/dL (this is the concentration of glucose in the unspiked sample). Sample Spiked with Bilirubin:  Measured Glucose Concentration with Bilirubin: 85 mg/dL.  % Interference = (85mg/dL−90mg/dL​) * 100/ 90mg/dL = −5.56% Interpretation: The result is -5.56%, indicating that bilirubin caused a 5.56% decrease in the measured glucose concentration. A negative interference means that the interfering substance decreased the analyte concentration. If this interference is significant (e.g., greater than 10%), further action may be required, such as modifying the method or using a different analytical approach. Linearity Determination with Graph and Linear Regression What is Linearity? Linearity means that the relationship between the concentration of the analyte and the instrumental response is directly proportional over a specified range. In other words, if the concentration of the analyte increases, the signal detected by the instrument should increase proportionally. Cont.. 2. How is Linearity Determined? Linearity is determined through a calibration curve, which is constructed by plotting the instrument response (y-axis) against the concentration of the analyte (x-axis) for a series of known concentrations. The goal is to find out if the plotted points form a straight line. Steps for Determining Linearity: 1. Prepare Standard Solutions:  Prepare a set of known standard solutions with varying concentrations of the analyte. For example, if you're testing glucose in serum, prepare standard solutions with known glucose concentrations (e.g., 0 mg/dL, 10 mg/dL, 20 mg/dL, 50 mg/dL, etc.). 2. Measure Instrument Response:  For each standard solution, measure the response using the analytical instrument (e.g., absorbance in spectrophotometry, peak area in chromatography). This gives you the instrument's response values (y-values) corresponding to each concentration (x-values). 3. Plot the Calibration Curve: On a graph, plot the concentration of the analyte (x-axis) versus the instrumental response (y-axis). The points should ideally form a straight line if the method is linear. 4. Perform Linear Regression: Apply linear regression to the data points to obtain the best-fit line. The linear regression equation will be of the form: y=mx+by  Where: y = instrument response (e.g., absorbance, signal intensity) x = concentration of the analyte m = slope of the line (which indicates sensitivity or the change in response per unit concentration) b = y-intercept (which ideally should be 0 for a perfect linear method) Example of Linearity Determination: Let’s say you are measuring glucose in blood serum using a spectrophotometric method. You prepare standard glucose solutions with known concentrations and measure their absorbance using the spectrophotometer. Concentration (mg/dL) Absorbance (Instrument Response) 0 0.00 10 0.10 20 0.20 50 0.50 100 0.100  Plot the concentration (mg/dL) on the x-axis and the absorbance (instrument response) on the y- axis. Linear Regression Equation: From the plot, you perform linear regression and obtain the equation of the line: y=0.01x+0 Summary of Part 2 Method Evaluation Precision is measured with replicant studies. Mean, standard deviation and %CV is measured. Accuracy is measured with linearity, recovery and interference studies. Linearity is determined with graph and linear regression. Total error is the sum of random error (from precision study) and systematic error (from linearity or comparison of methods).