Download - Screening and Diagnostic Test10[1]
-
8/14/2019 Screening and Diagnostic Test10[1]
1/22
Sc re eni ng a ndDia gnos ticTest
-
8/14/2019 Screening and Diagnostic Test10[1]
2/22
Definition of Screening:
Th e pre sump tiv e ide ntificatio n ofunrecogniz ed disease or defect byth e app lica tion of te sts,exa mina tions or other procedure swhich ca n be ap plie d rapi dly.
-
8/14/2019 Screening and Diagnostic Test10[1]
3/22
Screening program
Goa l: secon da ry p reven tion det ec t di sea se befo re clini calpoi nt for cure or improvedoutcome ge t peop le with di sease int oappropriate treatment
-
8/14/2019 Screening and Diagnostic Test10[1]
4/22
1. Design of screening test
gol d stan dard Se lect ion of objecti ves Samp le siz e
cas es:
controls
Asse ssing th e validity an dreliability
2
2
1
)1(
d
SeSezn =
2
2
2
)1(
d
SpSpz
n
=
-
8/14/2019 Screening and Diagnostic Test10[1]
5/22
2. Assessment of Screening Test valid ity reliabil ity predictive
valu e
-
8/14/2019 Screening and Diagnostic Test10[1]
6/22
-
8/14/2019 Screening and Diagnostic Test10[1]
7/22
(1) Validity of screening test
The abi lity of a tes t todisting uish bet ween who has adisease an d wh o does not haveth e di sease.
Two m ain comp on en ts:sen sitiv ity an d sp ecif icity
-
8/14/2019 Screening and Diagnostic Test10[1]
8/22
Sensitivity (Se):Pro portion of positiv e testwh o have th e disease.
Se = a / (a+c ) Specificity (Sp) :
Pro portion of neg ative testwh o do n ot h ave th e diseas e. Sp = d / (b+ d)
-
8/14/2019 Screening and Diagnostic Test10[1]
9/22
False positive rate (FP):
FP = b / (b +d) = (1 - S p) False negative rate (FN):
FN = c / (a+c) = (1 -Se )
Youdens index (YI):
YI = (Se + Sp) - 1 Likelihood ratio (LR):
Sp
Se
dbb
caaLR
=
+
+=+
1)/(
)/(
Sp
Se
dbd
cacLR
=
+
+=
1
)/(
)/(
-
8/14/2019 Screening and Diagnostic Test10[1]
10/22
Agreement rate:
Diagnostic index (DI):
Diagnostic availability (DA):
db
d
ca
aSpSeDI
+
+
+
=+=
1)(2 22 += SpSeDA
%100rateAgreement +++
+=
dcba
da
-
8/14/2019 Screening and Diagnostic Test10[1]
11/22
ROC c urve a nd i tsapplications ROC cur ve(Recei ve Oper at or
Char act er isti c cu rve) can show therel ati ons hi p between sens it ivi tyand spec ifi city i nt ui ti vel y.
Ap pli cation s: Se lect ion of cut off value (cu t poi nt ) Diagno st ic val ue co mpar is on amo ngsever al di agno st ic tes ts
-
8/14/2019 Screening and Diagnostic Test10[1]
12/22
-
8/14/2019 Screening and Diagnostic Test10[1]
13/22
-
8/14/2019 Screening and Diagnostic Test10[1]
14/22
-
8/14/2019 Screening and Diagnostic Test10[1]
15/22
(2) reliability ( Repeatability )
The test is reli abl e orrep ea table.
Intra-subject variation (var iati onwi thin i ndivi du al sub jec ts) Variance , sta nda rd dev iati on ,C.V. etc .
Inter-observer variation (var iati onbetween th ose rea di ng th e testresu lts) agreement, Kap pa val ue
-
8/14/2019 Screening and Diagnostic Test10[1]
16/22
Dr. APositive Negative
Dr. B Positive aagreement
b
Negative c dagreement
Agreement = (a+d) / (a+b+c+d)
Kappa va lue: (sel f readi ng ) Kappa = 2(ad-bc)/(a+b)(b+d)(a+c)(c+d)
-
8/14/2019 Screening and Diagnostic Test10[1]
17/22
(3) predictive value (PV)
Positive PV (+ PV):
The likel ih ood th at a pati entwi th a pos itive test resu lt reallyhas th e diseas e.
+ PV = a / (a+b) Negative PV (- PV) :
The likel ih ood th at a pati entwi th a nega ti ve test result reallydoes not ha ve th e di sea se.- PV = d / (c +d)
-
8/14/2019 Screening and Diagnostic Test10[1]
18/22
PV and disease prevalence
-
8/14/2019 Screening and Diagnostic Test10[1]
19/22
P : disease prevalence
-
8/14/2019 Screening and Diagnostic Test10[1]
20/22
3.Two-stage screening
To increase sen siti vity orspecificity:Two-stage screening
Series Test:mo re sp ec ifi ci ty( eg. di abe tes, HIV
inf ect ion ) Parallel Test: mor e sensi tivi ty
( eg. blood donorsc ree ni ng )
-
8/14/2019 Screening and Diagnostic Test10[1]
21/22
Multiple test
multiple test Test 1 Test 2 Result+ + +
in parallel + - +- + +- - -+ + +
in series + - -- Not necessary -
-
8/14/2019 Screening and Diagnostic Test10[1]
22/22