2010 smg training_cardiff_day1_session2b (2 of 2) herbison
-
Upload
rgveroniki -
Category
Education
-
view
557 -
download
0
Transcript of 2010 smg training_cardiff_day1_session2b (2 of 2) herbison
![Page 1: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/1.jpg)
Meta-analysis of count data
Peter HerbisonDepartment of Preventive and
Social MedicineUniversity of Otago
![Page 2: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/2.jpg)
What is count data
• Data that can only take the non-negative integer values 0, 1, 2, 3, …….
• Examples in RCTs– Episodes of exacerbation of asthma– Falls– Number of incontinence episodes
• Time scale can vary from the whole study period to the last few (pick your time period).
![Page 3: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/3.jpg)
Complicated by
• Different exposure times– Withdrawals– Deaths– Different diary periods
• Sometimes people only fill out some days of the diary as well as the periods being different
• Counts with a large mean can be treated as normally distributed
![Page 4: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/4.jpg)
What is large?
0
.1
.2
.3
.4P
roba
bilit
y
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15n
Mean 1 Mean 2 Mean 3Mean 4 Mean 5
![Page 5: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/5.jpg)
Analysis as rates
• Count number of events in each arm• Calculate person years at risk in each arm• Calculate rates in each arm• Divide to get a rate ratio• Can also be done by poisson regression
family– Poisson, negative binomial, zero inflated
poisson– Assumes constant underlying risk
![Page 6: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/6.jpg)
Meta-analysis of rate ratios
• Formula for calculation in Handbook• Straightforward using generic inverse
variance• In addition to the one reference in the
Handbook:– Guevara et al. Meta-analytic methods for
pooling rates when follow up duration varies: a case study. BMC Medical Research Methodology 2004;4:17
![Page 7: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/7.jpg)
Dichotimizing
• Change data so that it is those who have one or more events versus those who have none
• Get a 2x2 table• Good if the purpose of the treatment is to
prevent events rather than reduce the number of them
• Meta-analyse as usual
![Page 8: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/8.jpg)
Treat as continuous
• May not be too bad• Handbook says beware of skewness
![Page 9: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/9.jpg)
Time to event
• Usually time to first event but repeated events can be used
• Analyse by survival analysis (e.g. Cox’s proportional hazards regression)
• Ends with hazard ratios• Meta-analyse using generic inverse
variance
![Page 10: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/10.jpg)
Not so helpful
• Many people look at the distribution and think that the only distribution that exists is the normal distribution and counts are not normally distributed so they analyse the data with non-parametric statistics
• Not so useful for meta-analyses
![Page 11: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/11.jpg)
Problems with primary studies
• Can be difficult to define events– Exacerbations of asthma/COPD– Osteoporosis fractures
• Need to take follow-up time into account in analysis
• Failure to account for overdispersion
![Page 12: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/12.jpg)
So what is the problem?
• Too many choices• All reasonable things to do• So what if studies choose different
methods of analysis?
![Page 13: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/13.jpg)
Rates and dichotomizing
• From handbook– “In a randomised trial, rate ratios may often be
very similar to relative risks obtained after dichotomizing the participants, since the average period of follow-up should be similar in all intervention groups. Rate ratios and relative risks will differ, however, if an intervention affects the likelihood of some participants experiencing multiple events”.
![Page 14: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/14.jpg)
Combining different analyses
• So is it reasonable to combine different methods of analysing the data?
• A colleague who works on the falls review and I have been thinking about this for a while
• Falls studies are quite fragmented partly because of the different methods of analysis
![Page 15: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/15.jpg)
Simulation
• Simulated data sets with different characteristics• Analysed them many ways
– Rate ratio– Dichotomized– Poisson regression (allowing for duration)– Negative binomial (allowing for duration)– Time to first event– Means– Medians
![Page 16: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/16.jpg)
Data sets
• Two groups (size normal(100,2))• Low mean group (Poisson 0.2, 0.15)• Medium mean group (2, 1.5)• High mean group (7, 5)• Overdispersion built in by 0%, 20% and
40% from a distribution with a higher mean• 20% not in the study for the full time
(uniform over the follow up)
![Page 17: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/17.jpg)
Low mean no overdispersion
• Rate ratio = 0.7977
Not possiblemedian0.03200.0028mean0.1072-0.0058hazard0.00890.0006neg binom0.00010.0000poisson0.0839-0.0121binary
SDMean difference
![Page 18: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/18.jpg)
Low mean some overdispersion
• Rate ratio = 0.7521
Not possiblemedian0.03040.0008mean0.1114-0.0252hazard0.01240.0019neg binom0.00010.0000poisson0.0955-0.0376binary
SDMean difference
![Page 19: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/19.jpg)
Low mean high overdispersion
• Rate ratio = 0.7205
Not possiblemedian0.0293-0.0008mean0.0959-0.0170hazard0.01220.0015neg binom0.00010.0000poisson0.0852-0.0417binary
SDMean difference
![Page 20: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/20.jpg)
Medium mean no overdispersion
• Rate ratio = 0.7578
Not possiblemedian0.02830.0008mean0.1081-0.0616hazard0.00690.0024neg binom0.00010.0000poisson0.0723-0.1433binary
SDMean difference
![Page 21: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/21.jpg)
Medium mean some overdispersion
• Rate ratio = 0.7789
0.1770-0.0758median0.02910.0008mean0.1211-0.0499hazard0.01080.0032neg binom0.00010.0000poisson0.0752-0.1362binary
SDMean difference
![Page 22: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/22.jpg)
Medium mean high overdispersion
• Rate ratio = 0.7979
0.1400-0.0743median0.02990.0008mean0.1236-0.0508hazard0.01260.0030neg binom0.00010.0000poisson0.0739-0.1318binary
SDMean difference
![Page 23: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/23.jpg)
High mean no overdispersion
• Rate ratio = 0.7154
0.05620.0064median0.02720.0000mean0.1400-0.2769hazard0.02110.0096neg binom0.00010.0000poisson0.0493-0.2783binary
SDMean difference
![Page 24: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/24.jpg)
High mean some overdispersion
• Rate ratio = 0.7101
0.0588-0.0005median0.0271-0.0001mean0.1522-0.2841hazard0.02920.0125neg binom0.00010.0000poisson0.0490-0.2846binary
SDMean difference
![Page 25: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/25.jpg)
High mean high overdispersion
• Rate ratio = 0.7068
0.05030.0018median0.02710.0000mean0.1552-0.2912hazard0.03090.0148neg binom0.00010.0000poisson0.0504-0.2891binary
SDMean difference
![Page 26: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/26.jpg)
Conclusions 1
• When you have a low mean you should be able to combine almost anything
• As the mean increases then dichotomizing events increasingly underestimates treatment effects
• Time to first event underestimates but to a lesser extent than dichotomizing– Allowing for multiple events may help
![Page 27: 2010 smg training_cardiff_day1_session2b (2 of 2) herbison](https://reader033.fdocuments.us/reader033/viewer/2022052905/55840a97d8b42af00a8b5428/html5/thumbnails/27.jpg)
Conclusions 2• Adjusting for overdispersion by using
negative binomial has only a small effect even for a quite a bit of overdispersion– In spite of neg bin being a better fit
• Means (at least ratio of means) is surprisingly good
• Ratio of medians is astonishing, but inefficient– Problems with SE