Principles Of Diagnostic Testing PDF

PRINCIPLES OF DIAGNOSTIC TESTING PEDRO BITTENCOURT, DVM, MSC, PHD PRINCIPLES OF INFECTIOUS DISEASES HOW DO WE KNOW WHAT IS GOING ON? DIAGNOSTIC TESTS Learning Objectives Describe the importance of diagnostic tests Understand the basis of diagnostic test and sample selection Understand the indicators...

PRINCIPLES OF DIAGNOSTIC TESTING PEDRO BITTENCOURT, DVM, MSC, PHD PRINCIPLES OF INFECTIOUS DISEASES HOW DO WE KNOW WHAT IS GOING ON? DIAGNOSTIC TESTS Learning Objectives Describe the importance of diagnostic tests Understand the basis of diagnostic test and sample selection Understand the indicators of diagnostic test validity Sensitivity, Specificity, Predictive Values Why are diagnostic tests important? Confirm/exclude diagnosis Determine treatment strategies Epidemiological surveillance Prevention, control and eradication strategies Identification of new pathogens Phases of Testing 3prasesottesting analytical occur most errors stage during testing ofdiagnostic pre Pre-analytical Test selection; sampling; storage; transportation ii qq.fi ifiiif.i Analytical pertormingtneactuaitest Handling and analysis of the specimen Post-analytical Report results; interpretation in in I sottestresnits msn.tn Test Selection Determined by: Type of pathogen Type sample needed bloodurineteresetc Test characteristics Phase of the disease acutevschronic Availability Cost Considerations About Samples Samples must be correctly: Selected Blood? Tissues? Feces? Urine? Sputum? Collected Timing? Sterile? Multiple? Stored & transported Tubes? Transport medium? Temperature? Determined by the type of test, pathogen, and phase of disease Inadequate samples = Inadequate results Understanding Test Results: The result of a diagnostic test can be divided into 4 categories: outcomes Four possible Cut-off: value that divides results between positive or negative hhkiihd ididMi mMddddMdk iii.it iii iii.in I Understanding Test Results: SICK HEALTHY Diseased Non-diseased Positive test result True Positive (TP) False Positive (FP) Negative test result False Negative (FN) True Negative (TN) Diagnostic Sensitivity & Specificity: How Reliable is a Test? nowwenatestpicksuptruepositives.IM Diagnostic Sensitivity jpjifiÉÑonly Measures the capacity of a test to correctly identify the positive individuals (True positives). TP × 100 Determined by the equation: TP + FN Diagnostic Specificity HEALTHY POPULATIONONLY Measures the capacity of a test to correctly identify the negative individuals (True negatives) TN 0 × 100 Determined by the equation: TN + FP Diagnostic Sensitivity & Specificity: How Reliable is a Test? Diseased Non-diseased atest and winaisonarea positive negative result iiiiiiiiiiiiiii iii iiii specific avery test Positive test result True Positive (TP) Negative test result False Negative (FN) Sensitivity: TP TP + FN False Positive (FP) will have few very toalmost taiserositives True Negative (TN) truenegative Specificity: TN TN + FP Notice the cut-off change from A to B. What are the impacts of this change? B A Notice the cut-off change from figure A to B. What are the impacts of this change? A SENSITIVITY = TP × 100 TP + FN B SPECIFICITY = TN × 100 TN + FP Notice the cut-off change from figure A to B. What are the impacts of this change? tradeoffbetweensensitivityandspecificity A B wanttominimize ifynumber ou negatives offalse the sensitivity increase want you to the sensitivity situation depending the on more whereas important might be aaitterentsination s pecificity in more could be important SENSITIVITY = TP × 100 TP + FN Sensitivity ↑ More FPs Fewer FNs SPECIFICITY = TN × 100 TN + FP Specificity ↓ Test Predictive Value In clinical practice, what you need to know: What is the probability that a test-positive animal from the population you are testing is truly diseased? SENSITIVITY What is the probability that a test-negative animal from the population you are testing is truly healthy? SPECIFICITY Test Predictive Value Positive Predictive Value Probability of a positive test being a TP TP truepositives × 100 Determined by: truepositives falsepositives TP + FP Negative Predictive Value Probability of a negative test being a TN TN × 100 Determined by: TN + FN t.ieaie asenaaiive Prevalence has a strong impact on predictive values Prevalence O O Proportion of a population affected by a disease at a specific time A “photograph” of a specific moment Not to be confused with Incidence, which refers to the number os new cases of a disease in a O time interval includes diseased O all individuals at a giventimenot justthenewcases likefor incidence prevalance Predictive Values: How Reliable is a Test Result? Diseased Wip Non-diseased comes ii in Positive test result 0 willonlyhave Negative test result True Positive (TP) False Negative (FN) negative sickindividuals esbothhealthyand False Positive (FP) True Negative (TN) Positive Predictive Value: TP × 100 TP + FP Negative Predictive Value: TN × 100 TN + FN 0 Impact of prevalence in the PPV in tests with different levels of diagnostic sensitivity and specificity gasspecificity sensitivity 99s tinprovalence pro i uf T.itfie'mY i usi iiii specificity Inefimbaedtesttnattestsposiiivef 0 Impact of prevalence in the NPV in tests with different levels of diagnostic sensitivity and specificity prevalence themoreiikeiyanegativ f.ir TeYIlbetonetora pprevacence.tnelessiikeiy nogativeresuitistobet.ru Diseased Non-diseased Positive Predictive Value: Positive test result True Positive (TP) False Positive (FP) TP × 100 TP + FP Negative Predictive Value: Negative test result False Negative (FN) True Negative (TN) Sensitivity: Specificity: TP × 100 TP + FN TN × 100 TN + FP TN × 100 TN + FN Accuracy: TP + TN TP + FP + TN + FN × 100 Based on the following data, calculate: Sensitivity, Specificity, PPV and NPV. Diseased Positive test result Negative test result 105 profsitive Fegative 25 Non-diseased faffsitive 23 Tegative 847 Based on the following data, calculate: Sensitivity, Specificity, PPV and NPV. Diseased Non-diseased Positive test result 78 88 Negative test result 5 829 QUESTIONS? THANK YOU FOR ATTENDING! ©2021 Ross University School of Veterinary Medicine. All rights reserved.