metlit 12 diagnosis-ppt- juli 2011
TRANSCRIPT
Diagnostic TestDiagnostic Test
Partini PudjiastutiPartini PudjiastutiChild Health Department Faculty of Medicine Child Health Department Faculty of Medicine
University of Indonesia JakartaUniversity of Indonesia Jakarta
Tests as diagnostic aids and screening tools - key Tests as diagnostic aids and screening tools - key element of clinical medicine and public health.element of clinical medicine and public health.
Electrocardiogram, cardiac enzymes for diagnosis Electrocardiogram, cardiac enzymes for diagnosis of myocardial infarctionof myocardial infarction
Murphy’s sign (right upper abdominal tenderness Murphy’s sign (right upper abdominal tenderness on inspiration) in diagnosis of acute cholecystitison inspiration) in diagnosis of acute cholecystitis
Pap smear for detection of cervical cancerPap smear for detection of cervical cancer
Also essential in many epidemiologic studies where Also essential in many epidemiologic studies where diagnostic criteria and/or tests are used to diagnostic criteria and/or tests are used to establish exposure, outcome status.establish exposure, outcome status.
Diagnostic TestsDiagnostic Tests
Properties of diagnostic Properties of diagnostic teststests
Diagnostic tests include lab tests, Diagnostic tests include lab tests, radiologic tests, tissue diagnosis and radiologic tests, tissue diagnosis and history and physical examination history and physical examination maneuversmaneuvers
An accurate diagnostic test can An accurate diagnostic test can determine if a disease or condition is determine if a disease or condition is truly presenttruly present
Evaluating a diagnostic test means Evaluating a diagnostic test means looking at the relationship between the looking at the relationship between the test result and the true diagnosistest result and the true diagnosis
Reference (“Gold”) standardReference (“Gold”) standard One compares the result of a diagnostic test to One compares the result of a diagnostic test to
the “reference” or “gold” standard for the the “reference” or “gold” standard for the diagnosisdiagnosis
““Gold standard” may be expensive, inappropriate Gold standard” may be expensive, inappropriate (e.g. autopsy based) or unsuitable (e.g. clinical (e.g. autopsy based) or unsuitable (e.g. clinical follow-up when immediate decision required)follow-up when immediate decision required)
There may be no gold standard test and may There may be no gold standard test and may need to use long term outcome of patients with need to use long term outcome of patients with suspected disease or autopsysuspected disease or autopsy
Cannot determine characteristics of test without Cannot determine characteristics of test without this comparisonthis comparison
Designs of a diagnostic testDesigns of a diagnostic test
Pts with & w/o disease
New Test
+
-
Disease
No diseaseDisease
Gold std
• Observational study and consists of:Predictor variable (test result)Outcome variable (presence / absence of the disease)
• Specific “cross sectional”• Variable scale: binominal/dichotomous
Architecture of diagnostic researchArchitecture of diagnostic research
Phase I questionsPhase I questions: Do patients : Do patients with the target with the target disorderdisorder have have different test resultsdifferent test results from normal from normal individuals? individuals?
Phase II questionsPhase II questions: Are patients : Are patients with certain test with certain test resultsresults more more likely to have the target disorderlikely to have the target disorder than than patients with other test results?patients with other test results?
Phase III questionsPhase III questions: Among patients in whom it is : Among patients in whom it is clinically sensible to suspect the target disorderclinically sensible to suspect the target disorder, , does the test result does the test result distinguish those with and distinguish those with and withoutwithout the target disorder? the target disorder?
Phase IV questionsPhase IV questions: Do patients who : Do patients who undergo the undergo the diagnostic testdiagnostic test fare betterfare better (in their ultimate health (in their ultimate health outcomes) than similar patients who do not? outcomes) than similar patients who do not?
The diagnostic research question to be answered has to be carefully formulated, and determines the appropriate research approach:
Basic principles (1)Basic principles (1) Ideal diagnostic tests – right answers:Ideal diagnostic tests – right answers:
(+) results in everyone with the disease and(+) results in everyone with the disease and ( - ) results in everyone else( - ) results in everyone else
Usual clinical practice:Usual clinical practice: The test be studied in the same way it would The test be studied in the same way it would
be used in the clinical settingbe used in the clinical setting
Basic principles (2)Basic principles (2) Sensitivity, specificitySensitivity, specificity Prevalence, prior probability, predictive valuesPrevalence, prior probability, predictive values Likelihood ratiosLikelihood ratios Dichotomous scale, cutoff points (continuous Dichotomous scale, cutoff points (continuous
scale)scale) Positive (true and false), negative (true and Positive (true and false), negative (true and
false)false) ROC (receiver operator characteristic) curveROC (receiver operator characteristic) curve
General Structure: 2 X 2 General Structure: 2 X 2 tabletable
Condition Condition (by gold standard)(by gold standard)
PresentPresent AbsentAbsent TotalTotal
TestTest
PositivePositive True True positive (a)positive (a)
False False positive (b)positive (b)
a + ba + b
NegativeNegative False False negative negative (c)(c)
True True negative (d)negative (d)
c + dc + d
TotalTotal a + ca + c b + db + d a + b a + b + c + d+ c + d
d
What a study should tell youWhat a study should tell you Goal of clinical studies of diagnostic tests:Goal of clinical studies of diagnostic tests:
Provide information on all four cells in Provide information on all four cells in the 2X2 tablethe 2X2 table
Include information on negative testsInclude information on negative tests Information on test results in non-Information on test results in non-
diseased (especially if test to be used for diseased (especially if test to be used for screening)screening)
Must know population in which test Must know population in which test studied as this affects properties of test – studied as this affects properties of test – want to know results in patients across want to know results in patients across broad clinical spectrumbroad clinical spectrum
SensitivitySensitivity
Sensitivity – proportion of people with the disease who Sensitivity – proportion of people with the disease who have a positive test have a positive test
Usually positive in the presence of disease Usually positive in the presence of disease A sensitive test will rarely miss disease in those who have A sensitive test will rarely miss disease in those who have
itit Sensitive tests rule out disease “SnOut”Sensitive tests rule out disease “SnOut” Sn = a / (a +c)Sn = a / (a +c)
presenpresentt
absenabsentt
PositivPositivee
aa bb a+ba+b
negativnegativee
cc dd c+dc+d
a+ca+c b+db+d a+b+c+a+b+c+dd
SpecificitySpecificity
Specificity refers to proportion of patients without the disease with a Specificity refers to proportion of patients without the disease with a negative testnegative test
Rarely positive in the absence of diseaseRarely positive in the absence of disease A specific test will rarely identify disease in someone who does not A specific test will rarely identify disease in someone who does not
have ithave it A specific test rules in disease – “SpIn”A specific test rules in disease – “SpIn” Sp = d / b+dSp = d / b+d
presenpresentt
absenabsentt
PositivPositivee
aa bb a+ba+b
negativnegativee
cc dd c+dc+d
a+ca+c b+db+d a+b+c+da+b+c+d
Example:Example:
A researcher develops a new saliva pregnancy A researcher develops a new saliva pregnancy test. She collects samples from 100 women test. She collects samples from 100 women known to be pregnant by blood test (the gold known to be pregnant by blood test (the gold standard) and 100 women known not be standard) and 100 women known not be pregnant, also based on the same blood test.pregnant, also based on the same blood test.
The saliva test is “positive” in 95 of the pregnant The saliva test is “positive” in 95 of the pregnant women. It is also “positive” in 15 of the non-women. It is also “positive” in 15 of the non-pregnant women. What are the sensitivity and pregnant women. What are the sensitivity and specificity?specificity?
PregnantPregnant Non-pregnantNon-pregnantTotalsTotals
Saliva +Saliva + 95 95 1515 110110Saliva -Saliva - 5 5 8585 9090TotalsTotals 100 100 100100 200200
Sensitivity = TP/(TP+FN) = 95/100 = 95%Sensitivity = TP/(TP+FN) = 95/100 = 95%Specificity = TN/(TN+FP) = 85/100 = 85%Specificity = TN/(TN+FP) = 85/100 = 85%
Is it more important that a test be sensitive or Is it more important that a test be sensitive or specific?specific?
It depends on its purpose. A cheap mass It depends on its purpose. A cheap mass screening test should be sensitive (few cases screening test should be sensitive (few cases missed). A test designed to confirm the missed). A test designed to confirm the presence of disease should be specific (few presence of disease should be specific (few cases wrongly diagnosed).cases wrongly diagnosed).
Note that sensitivity and specificity are two Note that sensitivity and specificity are two distinct properties. Where classification is distinct properties. Where classification is based on an cutpoint along a continuum, there based on an cutpoint along a continuum, there is a tradeoff between the two.is a tradeoff between the two.
Example:Example:The saliva pregnancy test detects The saliva pregnancy test detects
progesterone. A refined version is progesterone. A refined version is developed.developed.
Suppose you add a drop of indicator solution Suppose you add a drop of indicator solution to the saliva sample. It can stay clear (0 to the saliva sample. It can stay clear (0 reaction) or turn green (1+), red (2+), or reaction) or turn green (1+), red (2+), or black (3+).black (3+).
(For purposes of discussion we will ignore (For purposes of discussion we will ignore overlapping colors)overlapping colors)
The researcher conducts a validation study and The researcher conducts a validation study and finds the following:finds the following:
PregnantPregnant Non-pregnantNon-pregnantTotalsTotals
Saliva 3+Saliva 3+ 85 85 55 9090Saliva 2+Saliva 2+ 10 10 1010 2020Saliva 1+Saliva 1+ 3 3 1717 2020Saliva 0Saliva 0 2 2 6868 7070
TotalsTotals 100 100 100100 200200
The sensitivity and specificity of the saliva test will The sensitivity and specificity of the saliva test will depend on the definition of “positive” and depend on the definition of “positive” and “negative” used.“negative” used.
If “positive” If “positive” 1+, sensitivity = (85+10+3)/100 = 1+, sensitivity = (85+10+3)/100 = 98%98%specificity = 68/100 = 68%specificity = 68/100 = 68%
If “positive” If “positive” 2+, sensitivity = (85+10)/100 = 95% 2+, sensitivity = (85+10)/100 = 95%specificity = (68+17)/100 = 85%specificity = (68+17)/100 = 85%
If “positive” = 3+, sensitivity = 85/100 = 85%If “positive” = 3+, sensitivity = 85/100 = 85%specificity = (68+17+10)/100 = 95%specificity = (68+17+10)/100 = 95%
The choice of cutpoint depends on the relative The choice of cutpoint depends on the relative adverse consequences of false-negatives vs. adverse consequences of false-negatives vs. false-positives.false-positives.
If it is most important not to miss anyone (FN), If it is most important not to miss anyone (FN), use use sensitivity and sensitivity and specificity. specificity.
If it is most important that people not be If it is most important that people not be erroneously labeled as having the condition erroneously labeled as having the condition (FP), use (FP), use sensitivity and sensitivity and specificity. specificity.
Sensitivity, and specificity Sensitivity, and specificity don’t help to make a don’t help to make a
diagnosisdiagnosis What you need to know What you need to know
is, given a positive or is, given a positive or negative result what is negative result what is the chance the patient the chance the patient has the diseasehas the disease
NOT…. If they have NOT…. If they have disease what is the disease what is the chance the patient has chance the patient has a positive (sensitivity) a positive (sensitivity) or negative (specificity) or negative (specificity) testtest
Positive Predictive valuesPositive Predictive values
Positive predictive value – probability of Positive predictive value – probability of disease in a patient with a positive (abnormal) disease in a patient with a positive (abnormal) test (i.e. that a positive test is a true positive)test (i.e. that a positive test is a true positive)
Highly specific diagnostic tests have high PPVHighly specific diagnostic tests have high PPV
PPV = a / a + bPPV = a / a + b
presenpresentt
absenabsentt
PositivPositivee
aa bb a+ba+b
negativnegativee
cc dd c+dc+d
a+ca+c b+db+d a+b+c+da+b+c+d
Negative Predictive valuesNegative Predictive values
Negative predictive value – probability Negative predictive value – probability that a patient with a negative test that a patient with a negative test (normal) does not have disease (normal) does not have disease
More sensitive tests have higher NPVMore sensitive tests have higher NPV NPV = d / c +dNPV = d / c +d
presenpresentt
absenabsentt
PositivPositivee
aa bb a+ba+b
negativnegativee
cc dd c+dc+d
a+ca+c b+db+d a+b+c+da+b+c+d
Predictive valuesPredictive values Predictive value influenced by prevalence, Predictive value influenced by prevalence,
therefore, not independent of situation in therefore, not independent of situation in which it is usedwhich it is used
Positive tests in a low prevalence Positive tests in a low prevalence population likely to be false positivespopulation likely to be false positives
As prevalence approaches 0, the PPV also As prevalence approaches 0, the PPV also approaches 0approaches 0
In order to know the PPV in an individual In order to know the PPV in an individual patient, need to know or estimate patient, need to know or estimate prevalence (likehood of disease) in such a prevalence (likehood of disease) in such a patientpatient
Predictive valuesPredictive values Sensitivity and specificity are fixed characteristics Sensitivity and specificity are fixed characteristics
of a diagnostic testof a diagnostic test Do no depend on the prevalence of disease in a Do no depend on the prevalence of disease in a
populationpopulation PPV/NPV useful for diagnosis PPV/NPV useful for diagnosis
Probability of disease after a + or – testProbability of disease after a + or – test Predictive values do depend on prevalencePredictive values do depend on prevalence e.g. a highly sensitive test applied in a high e.g. a highly sensitive test applied in a high
prevalence population will have a greater PPV prevalence population will have a greater PPV than the same test in a low prevalence populationthan the same test in a low prevalence population
Likelihood RatiosLikelihood Ratios Ratio of the chance of finding that test Ratio of the chance of finding that test
result in patient with disease compared result in patient with disease compared to the chance of that same result in to the chance of that same result in patients without diseasepatients without disease
A way to incorporate the A way to incorporate the sensitivitysensitivity and and specificityspecificity of a test into a of a test into a single single measuremeasure
probability of an individual with the condition having the test result
probability of an individual without condition having the test result LR =
Likelihood ratiosLikelihood ratios
Expresses how many times more (or less) Expresses how many times more (or less) likely a test result is found in diseased likely a test result is found in diseased versus non-diseased peopleversus non-diseased people
Likelihood ratio = probability of test result Likelihood ratio = probability of test result in diseased / probability of test result in in diseased / probability of test result in non-diseasednon-diseased
Positive LR =Positive LR = a / (a + c) a / (a + c) b / (b +d)b / (b +d)
presenpresentt
absenabsentt
PositivPositivee
aa bb a+ba+b
negativnegativee
cc dd c+dc+d
a+ca+c b+db+d a+b+ca+b+c+d+d
LR+ LR+ = sensitivity / (1-specificity) = sensitivity / (1-specificity) = [a/(a+c)) / (b/(b+d)]= [a/(a+c)) / (b/(b+d)]
LR- LR- = (1-sensitivity) / specificity = (1-sensitivity) / specificity = [c/(a+c)) / (d/(b+d)]= [c/(a+c)) / (d/(b+d)]
Impact on Disease Impact on Disease LikelihoodLikelihood
LR >10 or <0.1 cause large changes LR >10 or <0.1 cause large changes in likelihoodin likelihood LR 5-10 or 0.1-0.2 cause moderate LR 5-10 or 0.1-0.2 cause moderate
changeschanges LR 2-5 or 0.2-0.5 cause small changesLR 2-5 or 0.2-0.5 cause small changes LR between <2 and 0.5 cause LR between <2 and 0.5 cause little or no changelittle or no change
Likelihood ratios - why do Likelihood ratios - why do we need them?we need them?
Allow us to summarize information Allow us to summarize information from studies over a range of values from studies over a range of values i.e. the degree of abnormality, not i.e. the degree of abnormality, not just presence or absencejust presence or absence
Allows us to apply results to patients Allows us to apply results to patients with varying pre-test probabilities of with varying pre-test probabilities of diseasedisease
Likelihood ratiosLikelihood ratios Use same information from diagnostic Use same information from diagnostic
test studies test studies Used to determine how much a particular Used to determine how much a particular
test result changes the probability that a test result changes the probability that a patient has a particular disease (how patient has a particular disease (how much it increases the posttest much it increases the posttest probability compared to pretest probability compared to pretest probability)probability)
Whether a test result convinces you to Whether a test result convinces you to go ahead and treat a patient depends on go ahead and treat a patient depends on how much it changes the probability the how much it changes the probability the patient has the diseasepatient has the disease
Likelihood ratiosLikelihood ratios Use LR to convert pre-test probability to Use LR to convert pre-test probability to
posttest probabilityposttest probability First convert pretest probability to pretest First convert pretest probability to pretest
odds odds Then multiply pretest odds by likelihood ratio Then multiply pretest odds by likelihood ratio
to get posttest oddsto get posttest odds Convert posttest odds back to posttest Convert posttest odds back to posttest
probabilityprobability Odds = prob of event / 1- prob of eventOdds = prob of event / 1- prob of event Probability = odds / 1 + oddsProbability = odds / 1 + odds
LR+ LR+ = sensitivity / (1-specificity) = sensitivity / (1-specificity) = [a/(a+c)) / (b/(b+d)]= [a/(a+c)) / (b/(b+d)]
LR- LR- = (1-sensitivity) / specificity = (1-sensitivity) / specificity = [c/(a+c)) / (d/(b+d)]= [c/(a+c)) / (d/(b+d)]
Pre-test odds = pre-test probability / (1-Pre-test odds = pre-test probability / (1-pre-test probability)pre-test probability)
Post-test odds = pre-test odds * LR Post-test odds = pre-test odds * LR Post-test probability = post-test odds / Post-test probability = post-test odds /
(post test odds+1)(post test odds+1)
Test & Treatment Test & Treatment Thresholds in DiagnosisThresholds in Diagnosis
Probability of Disease
0% 100%
Treatment Threshold
Test Threshold
Further testing
No test
Treat
Test
CA B
pretest probability
0 .10 .20 .30 .40 .50 .60 .70 .80 .90 1
do not test
do nottreat
do not test
get on with treatment
Likelihood ratio
posttest probability
Test
+ = Se/(1-Sp)(1-Se)/Sp= -
PreTest odds x LR
pretest probability
Iron deficiency anemiaIron deficiency anemiaTotalsTotalsPresentPresent AbsentAbsent
DiagDiagnostic nostic
testtest result result (Serum (Serum ferritin)ferritin)
(+)(+)<65 <65
mmol/Lmmol/L731731
aa270270
bb10011001 a+ba+b
(-)(-)>65 >65
mmol/Lmmol/L7878cc
15001500dd
15781578 c+dc+d
TotalsTotals 809809 a+ca+c
17701770 b+db+d
25792579
a+b+ca+b+c+d+d
Sensitivity=a/a+c=90%Specificity =d/b+d=85%
Pos predictive value=a/a+b=73%Neg predictive value=d/c+d=95%
LR + = se/(1-sp)=90/15=6
Prevalence=(a+c)/(a+b+c+)= 32%
PredictorOutcome
Pretest probability
Likelihood ratio
Posttest probability
Diagnostic properties X-test for DHF Diagnostic properties X-test for DHF
DHFDHFYesYes
DHFDHFNoNo
X-test Result X-test Result PositivePositive
3030 1515 4545
X-Test Result X-Test Result NegativeNegative
1010 4545 5555
4040 6060 100100
Sensitivity = 30/40 = 0.75Specificity = 45/60 = 0.75
PPV = 30/45 = 0.67NPV = 45/55 = 0.82
Prevalence = 0.40LR+ = 0.75/0.25 = 3LR- = 0.25/0.75 = 0.33
Diagnostic properties X-test for DHF Diagnostic properties X-test for DHF DHFDHFYesYes
DHFDHFNoNo
X-test Result X-test Result PositivePositive
66 2323 2929
X-Test Result X-Test Result NegativeNegative
22 6969 7171
88 9292 100100
Sensitivity = 6/8 = 0.75Specificity = 69/92 = 0.75
PPV = 6/29 = 0.20NPV = 69/71 = 0.97
Prevalence = 0.08
LR+ = 0.75/0.25 = 3LR- = 0.25/0.75 = 0.33
Trade-offs between Sn and Trade-offs between Sn and SpSp
Not all tests are positive/negativeNot all tests are positive/negative For some there is a continuum of possible For some there is a continuum of possible
results for a test where cut-off set a some results for a test where cut-off set a some levellevel
Generally, Sn and Sp are increased or Generally, Sn and Sp are increased or decreased at each other’s expensedecreased at each other’s expense
Depending on desired characteristic of test Depending on desired characteristic of test (to maximize sensitivity or specificity) is (to maximize sensitivity or specificity) is where one sets cut-offwhere one sets cut-off
Can also express this relationship with a Can also express this relationship with a “receiver operating characteristic curve”“receiver operating characteristic curve”
ROC curvesROC curves Obtained by plotting the true positive rate Obtained by plotting the true positive rate
(sensitivity) against the false positive rate (sensitivity) against the false positive rate (1-specificity) over a range of cut-off values(1-specificity) over a range of cut-off values
Values on the axes run from 0 to 1.0Values on the axes run from 0 to 1.0 Tests that discriminate well crowd the Tests that discriminate well crowd the
upper left corner of the curveupper left corner of the curve Generally best cut-off is at the shoulder of Generally best cut-off is at the shoulder of
the curvethe curve Accuracy of test equal to AUCAccuracy of test equal to AUC
T4 T4 valuevalue
HypoHypothyroidthyroid
EuEuthyroidthyroid
5 or less5 or less 1818 115.1 – 7.05.1 – 7.0 77 17177.1 – 9.07.1 – 9.0 44 36369 or 9 or moremore
33 3939
TotalsTotals 3232 9393
T4 valueT4 value HypoHypothyroidthyroid
EuEuthyroidthyroid
≤ ≤ 5 5 1818 11
> 5> 5 1414 9292
TotalsTotals 3232 9393
T4 valueT4 value HypoHypothyroidthyroid
EuEuthyroidthyroid
≤ ≤ 77 2525 1818
> 7> 7 77 7575TotalsTotals 3232 9393
T4 valueT4 value HypoHypothyroidthyroid
EuEuthyroidthyroid
≤ ≤ 99 2929 5454
> 9> 9 33 3939
TotalsTotals 3232 9393
Cutoff Cutoff pointpoint
SensSens SpecSpec
55 0.560.56 0.990.9977 0.780.78 0.810.8199 0.910.91 0.420.42
T4 level in suspected hypo-thyroidism in children
For tests / predictors with continuous values result , cutoff points should be determine to choose the best value to use in distinguishing those with and without the target disorder
Continuum Scale Data: example
Cutoff Cutoff pointpoint
SensSens SpecSpec
55 0.560.56 0.990.9977 0.780.78 0.810.8199 0.910.91 0.420.42
Cutoff Cutoff pointpoint
SensSensTPTP
1-Spec1-SpecFPFP
55 0.560.56 0.010.0177 0.780.78 0.190.1999 0.910.91 0.580.58
RECEIVER OPERATING RECEIVER OPERATING CHARACTERISTIC CURVECHARACTERISTIC CURVE
Accuracy of the testAccuracy of the test
The accuracy of the test depends on how well the test separates the group being tested into those with and without the disease in question
Accuracy is measured by the area under the ROC curve. An area of 1 represents a perfect test; an area of 0.5 represents a worthless test (AUC)
0.90-1.00 = excellent (A) 0.80-0.90 = good (B) 0.70-0.80 = fair (C) 0.60-0.70 = poor (D) 0.50-0.60 = fail (F)
An ROC curve demonstrates several An ROC curve demonstrates several things: things:
It shows the tradeoff between sensitivity and It shows the tradeoff between sensitivity and specificity specificity
any increase in sensitivity will be accompanied by a decrease in any increase in sensitivity will be accompanied by a decrease in specificityspecificity
The closer the curve follows the left-hand border and The closer the curve follows the left-hand border and then the top border of the ROC space, the more then the top border of the ROC space, the more accurate the test. accurate the test.
The closer the curve comes to the 45-degree diagonal The closer the curve comes to the 45-degree diagonal of the ROC space, the less accurate the test. of the ROC space, the less accurate the test.
The slope of the tangent line at a cutoff point gives The slope of the tangent line at a cutoff point gives the likelihood ratio (LR) for that value of the test. the likelihood ratio (LR) for that value of the test.
Summary- 1Summary- 1 What do you want to DO with a test?What do you want to DO with a test?
Rule Rule in in disease? Rule disease? Rule out out disease?disease? Think about the pre-test probability Think about the pre-test probability
beforebefore you order a testyou order a test Compare the operating Compare the operating
characteristics of tests characteristics of tests beforebefore you you select oneselect one
Think about what you will do with Think about what you will do with the results of the test (implications)the results of the test (implications)
Summary- 2Summary- 2 If a serious outcome if the disease is If a serious outcome if the disease is
missed, you want a high ________missed, you want a high ________ If the treatment is invasive or risky, If the treatment is invasive or risky,
you want a high _________you want a high _________ Predictive value is influenced by Predictive value is influenced by
underlying underlying prevalenceprevalence of disease of disease Likelihood ratios are not influenced Likelihood ratios are not influenced
by prevalence of diseaseby prevalence of disease
Sensitivity
Specificity