27 Nov 2009 Dar es Salaam 1

CEM analysesCEM analyses

27 Nov 2009 2Dar es Salaam

What are we trying to What are we trying to measure?measure?

First, the risks associated with each First, the risks associated with each ARV regimenARV regimen Each regimen will be treated like a Each regimen will be treated like a

single drugsingle drug

Second, the risks associated with Second, the risks associated with individual drugsindividual drugs

27 Nov 2009 3Dar es Salaam

27 Nov 2009 4Dar es Salaam

ApproachApproach

Keep it simpleKeep it simple most of the important results can be most of the important results can be

revealed by simple measuresrevealed by simple measures

Keep it accurateKeep it accurate clean, quality dataclean, quality data controlled at inputcontrolled at input Example: variation in spellingExample: variation in spelling

27 Nov 2009 5Dar es Salaam

Small things 1Small things 1 Rates Rates

What to do with drop-outs?What to do with drop-outs? Choosing the denominatorChoosing the denominator

total cohort includes drop-outstotal cohort includes drop-outs demographicsdemographics

total follow-up questionnaires returnedtotal follow-up questionnaires returned general rates eg rates of events; reactions & general rates eg rates of events; reactions &

incidentsincidents Sites / regionsSites / regions

Total per question answeredTotal per question answered Rate per data element eg traditional meds, TBRate per data element eg traditional meds, TB

Small things 2Small things 2Examples of rates Overall reporting rate (response rate)

Total cohort enrolled = 8432 Total follow-up forms received = 6998 Reporting rate = 6998/8432 = 83%

Rates of events Total follow-up forms received = 6998 Total patients with oral candidiasis = 843 Rate of oral candidiasis = 843/6998 = 12%

27 Nov 2009 6Dar es Salaam

Small things 3Small things 3Examples of ratesExamples of rates Rates per total question answered for Rates per total question answered for

use of traditional medicinesuse of traditional medicines Total follow-up forms received = 6998Total follow-up forms received = 6998 Total TM questions answered = 5432Total TM questions answered = 5432 Total answered ‘Yes’ = 3954Total answered ‘Yes’ = 3954 Rate of use of traditional medicines = Rate of use of traditional medicines =

3954/5432 = 73%3954/5432 = 73%

27 Nov 2009 7Dar es Salaam

Small things 4Small things 4Calculation of riskCalculation of risk The rate of occurrence of an event in The rate of occurrence of an event in

the exposed cohort is a measure of the exposed cohort is a measure of absolute riskabsolute risk

Attributable riskAttributable risk: this is a measure of : this is a measure of the increased risk associated with the the increased risk associated with the medicine. It is calculated as follows:medicine. It is calculated as follows: rate in the exposed cohort minus rate in the exposed cohort minus rate before exposure (in the control period)rate before exposure (in the control period)

27 Nov 2009 8Dar es Salaam

27 Nov 2009 9Dar es Salaam

Small things 4Small things 4 DosesDoses

Analyses are done on the total daily doseAnalyses are done on the total daily dose Total daily dose x weekly e.g.Total daily dose x weekly e.g.

300mg 3 x weekly300mg 3 x weekly Or 100mg 7 x weeklyOr 100mg 7 x weekly

DurationDuration Duration in daysDuration in days

1 m = 30 days1 m = 30 days 1 y = 365 days1 y = 365 days Between x & y –take mid-pointBetween x & y –take mid-point

27 Nov 2009 10Dar es Salaam

Data manipulationData manipulationCollationCollation

summary of reporting rates for summary of reporting rates for males, females and totals;males, females and totals;

age/sex profiles of the cohort; age/sex profiles of the cohort; patient numbers by region or site;patient numbers by region or site; event profiles by clinical category;event profiles by clinical category;

within a clinical categorywithin a clinical category reactions & incidentsreactions & incidents

27 Nov 2009 11Dar es Salaam

Profile of Ages at First Prescription

213543

1477

3325

5827 5969

6552

3601

503468

945

1939

3426

4704

4254 4297

2289

280

0

5

10

15

20

25

< 20 20-30 30-39 40-49 50-59 60-69 70-79 80-89 90 plus

% o

f tot

al k

now

n ag

es

Celecoxib Rofecoxib

27 Nov 2009 12Dar es Salaam

IMMP example –COX-2IMMP example –COX-2Age & sexAge & sex CelecoxibCelecoxib RofecoxibRofecoxib

MeanMean 6363 5858

ModeMode 5959 5353

<40 years<40 years 6.9%6.9% 12.6%12.6%Highly significantHighly significant

70+ years70+ years 32.7%32.7% 25.7%25.7%Highly significantHighly significant

WomenWomen 61.6%61.6% 60.5%60.5%

27 Nov 2009 13Dar es Salaam

Celecoxib dose mg/no./%Celecoxib dose mg/no./%

100 6,622 8.1 200 65,591 80.5 300 274 0.3 400 8,927 11.0 600 46 800 30

27 Nov 2009 14Dar es Salaam

27 Nov 2009 15Dar es Salaam

Event profilesEvent profilesCelecoxibCelecoxib RofecoxibRofecoxib Odds RatioOdds Ratio

CirculatoryCirculatory 315 (17%)315 (17%) 214 (20%)214 (20%) NSNSAlimentaryAlimentary 302 (17%)302 (17%) 218 (20%)218 (20%) 0.8 (0.66,0.97)0.8 (0.66,0.97)

DiedDied 293 (16%)293 (16%) 179 (16%)179 (16%)SkinSkin 160 (9%)160 (9%) 59 (5%)59 (5%) 1.7 (1.24-2.30)1.7 (1.24-2.30)

RespiratorRespiratoryy

108 (6%)108 (6%) 54 (5%)54 (5%)

PsychiatricPsychiatric 88 (5%)88 (5%) 49 (5%)49 (5%)RenalRenal 72 (4%)72 (4%) 46 (4%)46 (4%)

27 Nov 2009 Dar es Salaam 16

Events collationEvents collation

27 Nov 2009 17Dar es Salaam

Event collationsEvent collationsLook Carefully

27 Nov 2009 Dar es Salaam 18

Understanding the Understanding the dictionarydictionary

27 Nov 2009 19Dar es Salaam

ExamplesExamples Annex 8 events collationAnnex 8 events collation Annex 12 eye events with signalAnnex 12 eye events with signal Annex 11 deathsAnnex 11 deaths Annex 13 concomitant medicinesAnnex 13 concomitant medicines

27 Nov 2009 20Dar es Salaam

RiskRisk Risk calculationRisk calculation Relative risk (rate ratio)Relative risk (rate ratio)

Rate of A / rate of B = relative riskRate of A / rate of B = relative risk Confidence intervalsConfidence intervals

Risk factors eg ageRisk factors eg age RRRR Multiple logistic regressionMultiple logistic regression

Risks of individual drugsRisks of individual drugs Compare events before and after drug Compare events before and after drug

substitutions with the patients acting as substitutions with the patients acting as their own controls eg their own controls eg d4t(30)-3TC-NVP to d4t(30)-3TC-EFVd4t(30)-3TC-NVP to d4t(30)-3TC-EFV

Compare events between regimens where Compare events between regimens where substitutions have taken placesubstitutions have taken place

Use multiple logistic regression to test each Use multiple logistic regression to test each drug as a risk factor for events of interest drug as a risk factor for events of interest

Large database needed for these analyses Large database needed for these analyses (pooled international data)(pooled international data)

27 Nov 2009 21Dar es Salaam

27 Nov 2009 22Dar es Salaam

Life table (survival) analysisLife table (survival) analysis Helps to characterise a reactionHelps to characterise a reaction

time to onsettime to onset spread of onset timesspread of onset times

Testing a possible signalTesting a possible signal

Statistical programmesStatistical programmes CemFlowCemFlow

automated outputsautomated outputs

MedCalcMedCalc free for 25 sessionsfree for 25 sessions

27 Nov 2009 23Dar es Salaam

27 Nov 2009 24Dar es Salaam

Don’t be afraid!

of statistics