CC - Quality Management Part 2

Reference interval

A pair of medical decision points that span the limits of results expected for a defined healthy population.

Applications of reference intervals

• Diagnosis
• Monitoring physiologic conditions
• Monitoring therapeutic drugs

Establishing a reference interval

A new RI is established when there is no existing analyte or methodology in the lab. It may require from 120-700 study individuals.

Transferring and validating a reference interval

RI will be transferred from a similar instrument or laboratory and validated using the method comparison and patient population and/or using a smaller sample of 40 healthy individuals.

Selection of individuals

To establish a RI, selection of individuals requires defining: • Inclusion criteria
• Exclusion criteria
• Partitioning individuals into subgroups
• Well-written confidential questionnaire and consent form

Exclusion criteria

Age, sex, fasting or non-fasting, stage of pregnancy, diet.

Partitioning into subgroups

Age, sex, fasting or non-fasting, stage of pregnancy, diet, tobacco use.

Pre-analytic variables

Variables that can affect specific tests: subject preparation, medications, collection time, food or beverage ingestion, sample storage.

Analytic variables

Variables that can affect specific tests: precision, accuracy, interferences, linearity, lot-to-lot reagents, recovery.

Calculating a reference interval

RI is calculated statistically using parametric or non-parametric methods. Prepare a histogram to evaluate the distribution of data.

The middle 95 %

RI is defined as the middle 95 % of values, from the 2.5th to the 97.5th percentile.

Method evaluation

Systematic execution of laboratory experiments to collect and analyze objective data, to characterize the analytical performance of a laboratory method, and to ensure it is fit for its intended clinical purpose.

Precision

Estimates the random error associated with the test method and detects any problems affecting its reproducibility.

Precision

Defined as the closeness of agreement between independent results of measurements obtained under stipulated conditions. The degree of precision is derived from SD or CV.

Accuracy

Difference between a measured value and its actual value. Due to the presence of a systematic error.

Estimating accuracy

Estimated using three types of studies: recovery, interference, and comparison of methods.

Recovery studies

Determine how much of the analyte can be detected in the presence of all the other compounds in the matrix.

Recovery studies

Used to estimate proportional systematic error, whose magnitude increases as the concentration of analyte increases.

Interference studies

Designed to determine if specific compounds affect the accuracy of laboratory tests.

Common interferences

Hemolysis, icterus (high bilirubin), and turbidity (particulate matter or lipids).

Interference studies

Used to estimate constant systematic error, whose magnitude of change is constant and not dependent on the amount of analyte.

Recovery against interference

Recovery studies estimate proportional systematic error, whose magnitude increases as the concentration of analyte increases. Interference studies estimate constant systematic error, whose magnitude of change is constant and not dependent on the amount of analyte.

Comparison of methods studies

Test method is compared with a reference method. Performed to estimate inaccuracy or systematic error.

Comparison of methods protocol

Westgard et al and CLIA recommend that 40 to 100 specimens be run by each method on the same day, within 4 hours, over 8 to 20 days.

Comparison of methods sampling

Samples should cover the entire clinical or analytical range: • 25 % below the normal (reference) range
• 50 % within the normal range
• 25 % above the normal range

Comparison of methods specimens

Include samples near medical decision levels, such as borderline high or low values. Specimens should be analyzed in duplicate.

Random error

Present in all measurements and can be either positive or negative. Due to instrument, operator, reagent, and environmental variation.

Random error

Calculated as the SD of the points about the regression line (Sy/x).

Sy/x

Refers to the average distance of the data from the regression line.

Constant systematic error

Continual difference between the test method and the comparative method values, regardless of the concentration.

Proportional error

Differences between the test method and the comparative method values are proportional to the analyte concentration. Present when the slope is 1.

Diagnostic sensitivity

Ability of a test to detect a given disease or condition.

Diagnostic sensitivity

Sensitivity = True Positives / (True Positives + False Negatives).

Diagnostic specificity

Ability of a test to identify the absence of a given disease or condition.

Diagnostic specificity

Specificity = True Negatives / (True Negatives + False Positives).

Positive predictive value

Probability of an individual having a given disease or condition if the test result is positive.

Positive predictive value

PPV = True Positives / (True Positives + False Positives).

Negative predictive value

Probability that an individual does not have a given disease or condition if the test result is negative.

Negative predictive value

NPV = True Negatives / (True Negatives + False Negatives).

The four denominators

Sensitivity and NPV both sit over a false negative term. Specificity and PPV both sit over a false positive term.

Lean Six Sigma

Combination of Lean principles and Six Sigma methodology. It uses a problem-cause-solution methodology to improve any process through waste elimination and variation reduction.

Lean

Manufacturing strategy pioneered by Toyota Motor Company Management. It works to eliminate the waste, such as streamlining a process to reduce wait times or modifying a process to reduce cost.

Six Sigma

Business management strategy developed by Motorola, Inc. Seeks to improve the performance of a process by identifying and eliminating causes of defects and errors, resulting in eliminating variation in the process.

Sigma

A statistical concept that represents how much variation there is in a process relative to customer specifications. The Sigma value is based on defects per million opportunities.

Six Sigma, defined

3.4 Defects Per Million Opportunities.

DMAIC

A structured, problem-solving method used in Lean Six Sigma to improve processes. • Define
• Measure
• Analyze
• Improve
• Control

Define

DMAIC step 1. Clearly identify the problem, goals, and customer requirements.

Measure

DMAIC step 2. Collect data to understand the current performance.

Analyze

DMAIC step 3. Identify the root cause or causes of the problem.

Improve

DMAIC step 4. Develop and implement solutions to eliminate root causes.

Control

DMAIC step 5. Monitor the improved process to maintain gains.