Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
ShawnE.Simpson DavidMadigan
ABayesianself‐controlledmethodfordrugsafetysurveillanceinlarge‐scalelongitudinaldata
Introduc?on• Ensuringdrugsafetybeginswithextensivepre‐approval
clinicaltrials
• Thisprocesscon:nuesa"erapprovalwhendrugsareinwidespreaduse:post‐marke?ngsurveillance
• Drugstakenbymorepeople,forlongerperiodsof:me,andindifferentwaysthaninpre‐approvaltrials
• Mayiden:fyadversehealthoutcomesassociatedwithdrugexposurethatwerenotpreviouslydetected
1997 2004
Sta:s:calObjec:ves• Iden:fydrug‐condi:onpairsthatmaybelinked
• Finddruginterac:onslinkedwithcondi:ons• Es:matethestrengthoftheseassocia:ons
• FundamentalDifficul:es- Largesize:Millionsofpeople,10000’sofcondi:ons
- Highdimension:10000’sofdrugs,millionsofinterac:ons
CurrentSystem:FDAAERS
LongitudinalHealthcareDatabases• Sen?nelIni?a?ve‐FDAplanstoestablishanac've
surveillancesystemusingdatafromhealthcareinforma:onholders
• Observa?onalMedicalOutcomesPartnership(OMOP)‐Researchingmethodsforanalyzinghealthcaredatabasestoevaluatesafetyprofilesofdrugsonthemarket
Self‐ControlledCaseSeries
!" !" !" !"
MI! VIOXX!
!" !" !" !"
MI!
!" !" !" !"
#$%&'(")"
#$%&'("*" VIOXX! VIOXX! VIOXX!
+,-"
+.."
• Personiobservedforτidays;(i,d)istheirdthday• yid=#ofeventsobservedon(i,d)• xid=1ifexposedtodrugon(i,d),0otherwise
• Methoddevelopedtoes:materela:veincidenceofAEstoassessvaccinesafety[Farrington,1995]
• Onedrug,oneadverseevent(AE)
• Eventsariseaccordingtoanon‐homogeneousPoissonprocess,exposuremodulatestheeventrate
• Intensityon(i,d)=
Method mAPscore22PRR 0.2251486
22OR 0.2280057
23BCPNN 0.209197
22EBGM 0.2173618
23CHI‐SQ 0.2144175
22PRR05 0.2046662
22ROR05 0.2046221
12BCPNN05 0.1832317
12EB05 0.1860902
SCCS(1AE,1drug) 0.2216072
BayesianSCCS,Normalprior,precision1(1AE,1drug)
0.26065568
BayesianLogisticRegression,Normalprior,precision1(1AE,multipledrugs)
0.2665139
Case‐Control 0.186743
• Advantages- Automated- Be`ertemporaldata
• Disadvantages- Li`lebaselinedata- NoOTCinforma:on
Yes No
Yes a b
No c d
AEj
Drugi
Total:N
2. Condi:onalindependenceofeventsandexposures
€
yid
€
yid '
€
x i
€
for d ≠ d'
€
yid
€
xid '
€
xid
€
for d ≠ d'
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Single drug
SCCS: Poisson event rate on each day
Assume that events arise according to a non-homogeneous Poissonprocess, where drug exposure modulates the baseline event rate
Poisson intensity on day (i , d) = e !i + "xid
wheree !i = fixed individual intensity ratee " = multiplicative e!ect of drug exposure
yid | xid ! Poisson( e !i+"xid )
The likelihood contribution of (i , d) is
P( yid | xid ) =e!e{!i +"xid}(e !i+"xid )yid
yid !
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Single drug
SCCS: Likelihood and Assumptions
Joint probability of events for i over all observed days
Li = P( yi1, . . . , yi!i | xi1, . . . , xi!i ) = P( yi | xi ) =!i!
d=1
P( yid | xid )
= exp{!"
d
e "i+#xid} "!i!
d=1
(e "i+#xid )yid
yid !
Assumptions
1 Conditionally independent events
yid ## yid! | xi for d $= d !
2 Events are conditionally independent of exposures
yid ## xid! | xid for d $= d !
Condi:ontoremoveφi
• CoulduseMLtogetes:mates,butdrugeffectβisofinterestandtheφi’sarenuisanceparameters
• Condi:ononsufficientsta:s:cni=Σyid
• Condi:onallikelihoodfori
• Maximize togetβCMLEconsistent,asympto:callyNormal[CameronandTrivedi,1998]
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Single drug
SCCS: Condition to remove !i
Could use ML to get estimates, but the drug e!ect ! is of interestand the "i ’s are nuisance parameters.
Condition on su"cient statistic ni =P
d yid to remove "i
ni | xi ! Poisson(!
d
e !i+"xid )
Conditional likelihood for i
Lci = P( yi | xi, ni ) =
P( yi | xi )
P( ni | xi )"
#i"
d=1
#e "xid
$d! e "xid!
%yid
Maximize lc =$
log Lci to get !̂CMLE
#$ consistent, asymptotically normal [Cameron and Trivedi, 1998]
If i has no events (yi = 0) then Lci = 1 $ only need cases (ni % 1)
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Single drug
SCCS: Condition to remove !i
Could use ML to get estimates, but the drug e!ect ! is of interestand the "i ’s are nuisance parameters.
Condition on su"cient statistic ni =P
d yid to remove "i
ni | xi ! Poisson(!
d
e !i+"xid )
Conditional likelihood for i
Lci = P( yi | xi, ni ) =
P( yi | xi )
P( ni | xi )"
#i"
d=1
#e "xid
$d! e "xid!
%yid
Maximize lc =$
log Lci to get !̂CMLE
#$ consistent, asymptotically normal [Cameron and Trivedi, 1998]
If i has no events (yi = 0) then Lci = 1 $ only need cases (ni % 1)
• Ifihasnoevents(yi=0)thenLiC=1,soweonlyneedcases(i.e.ni≥1)intheanalysis
• SCCSdoeswithin‐personcomparisonofeventrateduringexposuretoeventratewhileunexposed(‘self‐controlled’)
BayesianExtensionofSCCS
• Longitudinaldatabaseshave10000’sofpoten:aldrugs• Intensitymodel:e(maineffects)+(2‐wayinterac:ons)
highdimensionalitywithmillionsofpredictors
• StandardMLleadstooverfilng;needtoregularize
• Ourapproach–putaprioronβparameterstoshrinkthees:matestowardzero,smoothoutes:ma:on,andreduceoverfilng
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Single drug
SCCS: Poisson event rate on each day
Assume that events arise according to a non-homogeneous Poissonprocess, where drug exposure modulates the baseline event rate
Poisson intensity on day (i , d) = e !i + "xid
wheree !i = fixed individual intensity ratee " = multiplicative e!ect of drug exposure
yid | xid ! Poisson( e !i+"xid )
The likelihood contribution of (i , d) is
P( yid | xid ) =e!e{!i +"xid}(e !i+"xid )yid
yid !
Mul:pleDrugsandInterac:ons• WeextendthemodeltooneAEandmul?pledrugs
!"!" !"
!"
MI!!" !"
!" !"!" !"
MI!
MI!
#$%&'(")"
#$%&'("*"
!" !"
!" !"
QUETIAPINE!
!" !"!" !"
OLANZAPINE!
VIOXX! VIOXX!
VIOXX! VIOXX!
+,-"
.*/"
!"!"!" !"
OLANZAPINE!CELECOXIB!
QUETIAPINE!
• Intensityon(i,d)=
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Multiple drugs
SCCS: Multiple drugs
We extend the model to one adverse event and multiple drugs:
!"!" !"
!"
MI!!" !"
!" !"!" !"
MI!
MI!
#$%&'(")"
#$%&'("*"
!" !"
!" !"
QUETIAPINE!
!" !"!" !"
OLANZAPINE!
VIOXX! VIOXX!
VIOXX! VIOXX!
+,-"
.*/"
!"!"!" !"
OLANZAPINE!CELECOXIB!
QUETIAPINE!
Intensity on (i , d): e !i + !Txid = e !i + "1 xid1 + ··· + "p xidp
xidj =
!1 exposed to drug j on (i , d)0 otherwise
xid = ( xid1, . . . , xidp )T
! = ( !1, . . . , !p )T
Intensity with drug interactions, time-varying covariates
e {!i + !Txid +P
r !=s #rs xidr xids + "Tzid}
• xidj=1ifexposedtodrugj,0otherwise• Intensitywithdruginterac:ons,:me‐varyingcovariates:
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Multiple drugs
SCCS: Multiple drugs
We extend the model to one adverse event and multiple drugs:
!"!" !"
!"
MI!!" !"
!" !"!" !"
MI!
MI!
#$%&'(")"
#$%&'("*"
!" !"
!" !"
QUETIAPINE!
!" !"!" !"
OLANZAPINE!
VIOXX! VIOXX!
VIOXX! VIOXX!
+,-"
.*/"
!"!"!" !"
OLANZAPINE!CELECOXIB!
QUETIAPINE!
Intensity on (i , d): e !i + !Txid = e !i + "1 xid1 + ··· + "p xidp
xidj =
!1 exposed to drug j on (i , d)0 otherwise
xid = ( xid1, . . . , xidp )T
! = ( !1, . . . , !p )T
Intensity with drug interactions, time-varying covariates
e {!i + !Txid +P
r !=s #rs xidr xids + "Tzid}
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Multiple drugs
SCCS: Multiple drugs
We extend the model to one adverse event and multiple drugs:
!"!" !"
!"
MI!!" !"
!" !"!" !"
MI!
MI!
#$%&'(")"
#$%&'("*"
!" !"
!" !"
QUETIAPINE!
!" !"!" !"
OLANZAPINE!
VIOXX! VIOXX!
VIOXX! VIOXX!
+,-"
.*/"
!"!"!" !"
OLANZAPINE!CELECOXIB!
QUETIAPINE!
Intensity on (i , d): e !i + !Txid = e !i + "1 xid1 + ··· + "p xidp
xidj =
!1 exposed to drug j on (i , d)0 otherwise
xid = ( xid1, . . . , xidp )T
! = ( !1, . . . , !p )T
Intensity with drug interactions, time-varying covariates
e {!i + !Txid +P
r !=s #rs xidr xids + "Tzid}
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Multiple drugs
SCCS: Multiple drugs
We extend the model to one adverse event and multiple drugs:
!"!" !"
!"
MI!!" !"
!" !"!" !"
MI!
MI!
#$%&'(")"
#$%&'("*"
!" !"
!" !"
QUETIAPINE!
!" !"!" !"
OLANZAPINE!
VIOXX! VIOXX!
VIOXX! VIOXX!
+,-"
.*/"
!"!"!" !"
OLANZAPINE!CELECOXIB!
QUETIAPINE!
Intensity on (i , d): e !i + !Txid = e !i + "1 xid1 + ··· + "p xidp
xidj =
!1 exposed to drug j on (i , d)0 otherwise
xid = ( xid1, . . . , xidp )T
! = ( !1, . . . , !p )T
Intensity with drug interactions, time-varying covariates
e {!i + !Txid +P
r !=s #rs xidr xids + "Tzid}
1. Normalprior(ridgeregression)
2. Laplacianprior(lasso)
3.4 Shrinkage Methods 71
TABLE 3.4. Estimators of !j in the case of orthonormal columns of X. M and "are constants chosen by the corresponding techniques; sign denotes the sign of itsargument (±1), and x+ denotes “positive part” of x. Below the table, estimatorsare shown by broken red lines. The 45! line in gray shows the unrestricted estimatefor reference.
Estimator Formula
Best subset (size M) !̂j · I[rank(|!̂j | ! M)
Ridge !̂j/(1 + ")
Lasso sign(!̂j)(|!̂j |" ")+
(0,0) (0,0) (0,0)
|!̂(M)|
"
Best Subset Ridge Lasso
!^ !^2. .!
1
! 2
!1 !
FIGURE 3.11. Estimation picture for the lasso (left) and ridge regression(right). Shown are contours of the error and constraint functions. The solid blueareas are the constraint regions |!1| + |!2| ! t and !2
1 + !22 ! t2, respectively,
while the red ellipses are the contours of the least squares error function.
3.4 Shrinkage Methods 71
TABLE 3.4. Estimators of !j in the case of orthonormal columns of X. M and "are constants chosen by the corresponding techniques; sign denotes the sign of itsargument (±1), and x+ denotes “positive part” of x. Below the table, estimatorsare shown by broken red lines. The 45! line in gray shows the unrestricted estimatefor reference.
Estimator Formula
Best subset (size M) !̂j · I[rank(|!̂j | ! M)
Ridge !̂j/(1 + ")
Lasso sign(!̂j)(|!̂j |" ")+
(0,0) (0,0) (0,0)
|!̂(M)|
"
Best Subset Ridge Lasso
!^ !^2. .!
1
! 2
!1 !
FIGURE 3.11. Estimation picture for the lasso (left) and ridge regression(right). Shown are contours of the error and constraint functions. The solid blueareas are the constraint regions |!1| + |!2| ! t and !2
1 + !22 ! t2, respectively,
while the red ellipses are the contours of the least squares error function.
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Bayesian Extension of SCCS
Overfitting
Priors on model parameters !
1 Normal prior (ridge regression)
!j ! N(0, "2!)
max likelihood subject top!
j=1
!2j " s
2 Laplacian prior (lasso)
!j ! Laplace(0, 1/#)
max likelihood subject top!
j=1
|!j | " s
Posterior modes via cyclic coordinate descent [Genkin et al, 2007].Handles millions of predictors in logistic case (BBR).
Log likelihood is concave # convex optimization problem. Each stepincreases log posterior, “climb to top of the hill”
maxliksubjectto
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Bayesian Extension of SCCS
Overfitting
Priors on model parameters !
1 Normal prior (ridge regression)
!j ! N(0, "2!)
max likelihood subject top!
j=1
!2j " s
2 Laplacian prior (lasso)
!j ! Laplace(0, 1/#)
max likelihood subject top!
j=1
|!j | " s
Posterior modes via cyclic coordinate descent [Genkin et al, 2007].Handles millions of predictors in logistic case (BBR).
Log likelihood is concave # convex optimization problem. Each stepincreases log posterior, “climb to top of the hill”
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Bayesian Extension of SCCS
Overfitting
Priors on model parameters !
1 Normal prior (ridge regression)
!j ! N(0, "2!)
max likelihood subject top!
j=1
!2j " s
2 Laplacian prior (lasso)
!j ! Laplace(0, 1/#)
max likelihood subject top!
j=1
|!j | " s
Posterior modes via cyclic coordinate descent [Genkin et al, 2007].Handles millions of predictors in logistic case (BBR).
Log likelihood is concave # convex optimization problem. Each stepincreases log posterior, “climb to top of the hill”
maxliksubjectto
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Bayesian Extension of SCCS
Overfitting
Priors on model parameters !
1 Normal prior (ridge regression)
!j ! N(0, "2!)
max likelihood subject top!
j=1
!2j " s
2 Laplacian prior (lasso)
!j ! Laplace(0, 1/#)
max likelihood subject top!
j=1
|!j | " s
Posterior modes via cyclic coordinate descent [Genkin et al, 2007].Handles millions of predictors in logistic case (BBR).
Log likelihood is concave # convex optimization problem. Each stepincreases log posterior, “climb to top of the hill”
References
Results:OMOPMethodsEvalua?on• Methodsevalua:on:
- Chose10drugs,10condi:onsofinterest- 9drug‐condi:onpairswithatrueassocia'on- Pairsdeterminedtohavenolinkserveasnega'vecontrols
• Evalua:onisbasedonmeanaverageprecision(mAP)score:measuresthedegreetowhichamethodmaximizes‘trueposi:ves’whileminimizing‘falseposi:ves’
MSLRdatabase(1.5Mpeople)
FurtherWork
• CameronandTrivedi(1998)RegressionAnalysisofCountData.CambridgeUniversityPress.
• Farrington(1995)“Rela:veincidencees:ma:onfromcaseseriesforvaccinesafetyevalua:on,”Biometrics,Vol.51,No.1,pg.228‐235.
• Genkinetal.(2007)“Large‐scaleBayesianlogis:cregressionfortextcategoriza:on,”Technometrics,Vol.49,No.3,pg.291‐304.
• CurrentapproachtosurveillanceisbasedontheFDA’sAdverseEventRepor?ngSystem(AERS)
• Anyonecanvoluntarilysubmitareportonadverseevents(AEs)thatmayberelatedtodrugexposures
• Difficul:eswithAERS
- Messy–spellingerrors,etc.- Bias–underrepor:ng,duplicatereports,media- Unreliabletemporalinforma:on
• Mul:pledrugsandAEsmaybelistedononereport
• 15000drugs×16000AEs=240milliontables
• MostAEsdonotoccurwithmostdrugs;smallcountsina
• FDAuses2×2summaries,appliesBayesianshrinkagemethodstodealwithvariabilityduetosmallcounts
• Limita:ons
- Noadjustmentforconfoundingdrugs- Ignoresinterac:ons- Maynotu:lizetemporalinforma:on
xi1 xi2 xi3 xiτ...
...
yi1 yi2 yi3 yiτ...• Relaxindependenceassump:onstoallowdependencebetweenevents
• Alloweventstoinfluencefutureexposures
β1x1+…+βjxj+…+βkxk+…+βpxpeN(μ[1],σ[1]2) N(μ[D],σ[D]2)...
drugclass[1] drugclass[D]...
β1x1+...+βpxpe
β1x1+...+βpxpe
...
...
ym
y1~
~
...
N(μ1(c),σ(1)2) N(μp(c),σ(c)2)...
cond
itio
ncl
ass
(c)
• Hierarchicalmodelingofdrugsintodrugclasses
• Hierarchicalmodelingofcondi:onsintoclasses
• Convexop:miza:on:Posteriormodesviacycliccoordinatedescent[Genkinetal,2007]
• Handlesmillionsofpredictorsinlogis:ccase(BBR)
A Bayesian self-controlled method for drug safety surveillance in large-scale longitudinal data
Self-controlled case series method
Single drug
SCCS: Condition to remove !i
Could use ML to get estimates, but the drug e!ect ! is of interestand the "i ’s are nuisance parameters.
Condition on su"cient statistic ni =P
d yid to remove "i
ni | xi ! Poisson(!
d
e !i+"xid )
Conditional likelihood for i
Lci = P( yi | xi, ni ) =
P( yi | xi )
P( ni | xi )"
#i"
d=1
#e "xid
$d! e "xid!
%yid
Maximize lc =$
log Lci to get !̂CMLE
#$ consistent, asymptotically normal [Cameron and Trivedi, 1998]
If i has no events (yi = 0) then Lci = 1 $ only need cases (ni % 1)
⌃
DataReduc:ontoCasesOnly
• Poten:alanalysistechniques:maxSPRT,cohortmethods,casecontrol,case‐crossover,self‐controlledcaseseries…