SlideShare a Scribd company logo
1 of 26
A comparison of Random Effects
meta-analysis methods when study
effects are non-normally distributed
Evan Kontopantelis &
David Reeves
NPCRDC
How meta-analysis works
• A search for papers relevant to
the research question is
conducted. Unsuitable papers
are filtered out
• In each paper…
– for each outcome measure that
is directly relevant to the RQ,
or a good enough proxy, we
calculate an effect (of
intervention vs control) and its
variance
– An overall effect and variance
is selected
• Effects and their variances are
combined to calculate an
overall effect
Chronic disease - Risk factors
effect
-.4 0 .4 .8
Combined
Woolard(B), 1995
Woolard(A), 1995
Eckerlund, 1985
Moher, 2001
Cupples, 1994
Campbell, 1998
Van Ree, 1985
Heterogeneity
• Heterogeneity can be attributed to clinical and/or
methodological diversity
• Clinical heterogeneity: variability that arises from
different populations, interventions, outcomes
and follow-up times
• Methodological heterogeneity: relates to
differences in trial design and quality
• Detecting (usually with Cochran’s Q test)
quantifying and dealing with heterogeneity can
be very hard
Absence of heterogeneity
• Assumes that the true
effects of the studies
are all equal and
deviations occur
because of imprecision
of results
• Analysed with fixed-
effects method
i iY e 
Presence of heterogeneity
• It is assumed that
there exists variation
in the size of the true
effect among studies
(in addition to the
imprecision in results)
• Analysed with
random-effects
methods
i i iY e 
Random-effect MA methods
• Estimate the between-study variance and
use it in estimating the overall effect
• Parametric:
– DerSimonian-Laird (1986)
– Maximum & Profile likelihood (1996)
• Non-parametric:
– Permutations method (1999)
– Non Parametric Maximum Likelihood (1999)
2


“Potential” problems?
• Heterogeneity is common & the FE model is
under fire
• Parametric RE models assume that both the
effects and errors are normally distributed
• Almost all RE models (except PL) do not
take account of uncertainty in
• DL is usually the preferred method of
analysis because it is easy to implement
and is available in all software packages
2
ˆ
So far…
• The number of studies and the amount of
heterogeneity have been found to affect
method performance
• Performance comparisons usually focus on
coverage and ignore power or have not
included some important methods (e.g. PL,
PE)
• Evaluations were based on normal data:
method robustness has not been assessed
with non-normal data
Our bit
In a nutshell
• Simulated various non-normal distributions
for the true effects: skew normal, bimodal,
beta, uniform, U and others
• Created datasets of 10000 meta-analyses
for various numbers of studies k and
different degrees of heterogeneity, for each
distributional assumption
• Compared FE, DL, ML, PL and PE methods
(along with a simple t-test) in terms of
coverage and power across all datasets
Generating the data
• For a single study we simulated the effect size estimate
and the within-study variance estimate of a binary
outcome
• The variance was assumed to be a realisation from a
distribution, multiplied by .25 and restricted to the
(.009, .6) interval
• involves two components
– where
–
• Four values were used: .01, .03, .07 & .1
• Number of studies (MA size) varied from 2 to 35
iY
i
2
ˆ
2
1
iY i i iY e( ) 
2
(0, ˆ )i ie 
2

k
i i    2
?(0, )i 
Details on the MA methods
• Fixed effects (FE)
• DerSimonian-Laird (DL)
• Q method (Q)
• Maximum Likelihood (ML)
• Profile Likelihood (PL)
• Permutations method (PE)
• T-test method (T)
Performance
• For each simulated meta-analysis case we
calculated confidence intervals for the overall
effect estimate , for all the methods
• Coverage: % of confidence intervals that
contain the true overall effect in a sample of
10000 meta-analyses
• Power: % of CIs that do not contain the 25th
centile of the population distribution of the
10000 effect sizes
ˆ

Results
Zero
2

Normally distributed i
Skew normal i
Bimodal i
PL performance across various
distributions
Drawing conclusions
Summary
• Within any given method, the results were
consistent across all types of distribution shape
• This can give researchers confidence that
methods are highly robust against even the most
severe violation of the assumption of normally
distributed effect sizes
• If it is reasonable to assume that the effect size
does not vary between studies, the FE, Q and
ML methods all provide accurate coverage
coupled with good power
In the presence of heterogeneity…
• However, zero between study variance is the
exception rather than the norm and the
presence of even a moderate amount of
alters the picture considerably
• FE, Q and ML quickly lose coverage as
heterogeneity increases
• DL rapidly goes from providing a coverage that
is overly high, to one that is overly low
• PE, and to a lesser extend PL, now provide the
best coverage, even with very small sample
sizes
2

Which method then?
• If priority is given to maintaining an accurate
Type I error rate then the simple t-test is the
best method. But its power is very low, making
it a poor choice when control of the Type II
error rate is also important
• PE gives accurate coverage in all situations
and has better power than T, but the method is
more difficult to implement and cannot be used
with less than 6 studies
• PL has ‘reasonable’ coverage in most
situations, giving it an edge over other methods
Current & future work
• Created a freely available Excel add-in
that implements all the described MA
methods and various measures of
heterogeneity
• Working on a STATA module that will do
the same
• Investigate performance of heterogeneity
measures under non-normally distributed
data
Main references
• Brockwell SE, Gordon IR. A comparison of statistical methods for
meta-analysis. Stat.Med. 2001; 20(6):825-840
• Engels EA, Schmid CH, Terrin N, Olkin I, Lau J. Heterogeneity and
statistical significance in meta-analysis: an empirical study of 125
meta-analyses. Stat.Med. 2000; 19(13):1707-1728
• Follmann DA, Proschan MA. Valid inference in random effects meta-
analysis. Biometrics 1999; 55(3):732-737
• Hardy RJ, Thompson SG. A likelihood approach to meta-analysis
with random effects. Stat.Med. 1996; 15(6):619-629
• Micceri T. The Unicorn, the Normal Curve, and Other Improbable
Creatures. Psychological Bulletin 1989; 105(1):156-166
• Ramberg JS, Dudewicz EJ, Tadikamalla PR, Mykytka EF. A
Probability Distribution and Its Uses in Fitting Data. Technometrics
1979; 21(2):201-214
Thank you for listening

More Related Content

What's hot

Power, Effect Sizes, Confidence Intervals, & Academic Integrity
Power, Effect Sizes, Confidence Intervals, & Academic IntegrityPower, Effect Sizes, Confidence Intervals, & Academic Integrity
Power, Effect Sizes, Confidence Intervals, & Academic IntegrityJames Neill
 
Correlational research
Correlational researchCorrelational research
Correlational researchJijo G John
 
Relational and correlational research
Relational and correlational researchRelational and correlational research
Relational and correlational researchNieLeeXin
 
Survey and correlational research (1)
Survey and correlational research (1)Survey and correlational research (1)
Survey and correlational research (1)zuraiberahim
 
Correlation research design presentation 2015
Correlation research design presentation 2015Correlation research design presentation 2015
Correlation research design presentation 2015Syed imran ali
 
Network meta-analysis & models for inconsistency
Network meta-analysis & models for inconsistencyNetwork meta-analysis & models for inconsistency
Network meta-analysis & models for inconsistencycheweb1
 
Correlational Research
Correlational ResearchCorrelational Research
Correlational Researchirshad narejo
 
Power, effect size, and Issues in NHST
Power, effect size, and Issues in NHSTPower, effect size, and Issues in NHST
Power, effect size, and Issues in NHSTCarlo Magno
 
Survey and correlational methods of research: Assumptions, Steps and Pros and...
Survey and correlational methods of research: Assumptions, Steps and Pros and...Survey and correlational methods of research: Assumptions, Steps and Pros and...
Survey and correlational methods of research: Assumptions, Steps and Pros and...Michael J Leo
 
Exploratory factor analysis
Exploratory factor analysisExploratory factor analysis
Exploratory factor analysisJames Neill
 
STATISTICAL TOOLS USED IN ANALYTICAL CHEMISTRY
STATISTICAL TOOLS USED IN ANALYTICAL CHEMISTRYSTATISTICAL TOOLS USED IN ANALYTICAL CHEMISTRY
STATISTICAL TOOLS USED IN ANALYTICAL CHEMISTRYkeerthana151
 

What's hot (20)

Overview and Objectives of the Workshop
Overview and Objectives of the WorkshopOverview and Objectives of the Workshop
Overview and Objectives of the Workshop
 
Power, Effect Sizes, Confidence Intervals, & Academic Integrity
Power, Effect Sizes, Confidence Intervals, & Academic IntegrityPower, Effect Sizes, Confidence Intervals, & Academic Integrity
Power, Effect Sizes, Confidence Intervals, & Academic Integrity
 
Correlational research
Correlational researchCorrelational research
Correlational research
 
Correlational research
Correlational researchCorrelational research
Correlational research
 
Relational and correlational research
Relational and correlational researchRelational and correlational research
Relational and correlational research
 
Correlational research
Correlational research Correlational research
Correlational research
 
Survey and correlational research (1)
Survey and correlational research (1)Survey and correlational research (1)
Survey and correlational research (1)
 
Seminar in Meta-analysis
Seminar in Meta-analysisSeminar in Meta-analysis
Seminar in Meta-analysis
 
Correlation research design presentation 2015
Correlation research design presentation 2015Correlation research design presentation 2015
Correlation research design presentation 2015
 
Meta analysis
Meta analysisMeta analysis
Meta analysis
 
Network meta-analysis & models for inconsistency
Network meta-analysis & models for inconsistencyNetwork meta-analysis & models for inconsistency
Network meta-analysis & models for inconsistency
 
Correlational Research
Correlational ResearchCorrelational Research
Correlational Research
 
Network meta analysis
Network meta analysisNetwork meta analysis
Network meta analysis
 
Power, effect size, and Issues in NHST
Power, effect size, and Issues in NHSTPower, effect size, and Issues in NHST
Power, effect size, and Issues in NHST
 
Survey and correlational methods of research: Assumptions, Steps and Pros and...
Survey and correlational methods of research: Assumptions, Steps and Pros and...Survey and correlational methods of research: Assumptions, Steps and Pros and...
Survey and correlational methods of research: Assumptions, Steps and Pros and...
 
Experimental
ExperimentalExperimental
Experimental
 
Exploratory factor analysis
Exploratory factor analysisExploratory factor analysis
Exploratory factor analysis
 
Meta analysis
Meta analysisMeta analysis
Meta analysis
 
Desres final
Desres finalDesres final
Desres final
 
STATISTICAL TOOLS USED IN ANALYTICAL CHEMISTRY
STATISTICAL TOOLS USED IN ANALYTICAL CHEMISTRYSTATISTICAL TOOLS USED IN ANALYTICAL CHEMISTRY
STATISTICAL TOOLS USED IN ANALYTICAL CHEMISTRY
 

Viewers also liked

Công ty tổ chức sự kiện chuyên nghiệp tại hcm, cần thơ, đà nẵng
Công ty tổ chức sự kiện chuyên nghiệp tại hcm, cần thơ, đà nẵngCông ty tổ chức sự kiện chuyên nghiệp tại hcm, cần thơ, đà nẵng
Công ty tổ chức sự kiện chuyên nghiệp tại hcm, cần thơ, đà nẵngHoàng Tuấn
 
Xưởng may, Cho thuê trang phục sự kiện giá rẻ nhất tại Tp.HCM
Xưởng may, Cho thuê trang phục sự kiện giá rẻ nhất tại Tp.HCMXưởng may, Cho thuê trang phục sự kiện giá rẻ nhất tại Tp.HCM
Xưởng may, Cho thuê trang phục sự kiện giá rẻ nhất tại Tp.HCMHoàng Tuấn
 
Linear Local Brand Optimization PowerPoint
Linear Local Brand Optimization PowerPointLinear Local Brand Optimization PowerPoint
Linear Local Brand Optimization PowerPointSean Brown
 
식물세포 디펜스
식물세포 디펜스식물세포 디펜스
식물세포 디펜스snugdc
 
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCMCho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCMHoàng Tuấn
 
MC tiếng hoa, MC song ngữ tiếng Hoa, MC chuyên nghiệp tại Tp.HCM
MC tiếng hoa, MC song ngữ tiếng Hoa, MC chuyên nghiệp tại Tp.HCMMC tiếng hoa, MC song ngữ tiếng Hoa, MC chuyên nghiệp tại Tp.HCM
MC tiếng hoa, MC song ngữ tiếng Hoa, MC chuyên nghiệp tại Tp.HCMHoàng Tuấn
 
Beverages & Purchacing
Beverages & PurchacingBeverages & Purchacing
Beverages & PurchacingKelly Hoins
 
Decor, trang trí không gian tết chuyên nghiệp tại tp.hcm
Decor, trang trí không gian tết chuyên nghiệp tại tp.hcmDecor, trang trí không gian tết chuyên nghiệp tại tp.hcm
Decor, trang trí không gian tết chuyên nghiệp tại tp.hcmHoàng Tuấn
 
Gabriel Garcia Márquez
Gabriel Garcia MárquezGabriel Garcia Márquez
Gabriel Garcia Márquezjimenabravo
 
Công ty tổ chức sự kiện chuyên nghiệp nhất tại tp.hcm
Công ty tổ chức sự kiện chuyên nghiệp nhất tại tp.hcmCông ty tổ chức sự kiện chuyên nghiệp nhất tại tp.hcm
Công ty tổ chức sự kiện chuyên nghiệp nhất tại tp.hcmHoàng Tuấn
 
Презентация Юлии Бычковой
Презентация Юлии БычковойПрезентация Юлии Бычковой
Презентация Юлии БычковойYulya Kalsina
 
Cho thuê mc, cung cap mc chuyen nghiep tai HCM
Cho thuê mc, cung cap mc chuyen nghiep tai HCMCho thuê mc, cung cap mc chuyen nghiep tai HCM
Cho thuê mc, cung cap mc chuyen nghiep tai HCMHoàng Tuấn
 
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCMCho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCMHoàng Tuấn
 
Decor, trang trí không gian tết chuyên nghiệp tại Bình Dương
Decor, trang trí không gian tết chuyên nghiệp tại Bình DươngDecor, trang trí không gian tết chuyên nghiệp tại Bình Dương
Decor, trang trí không gian tết chuyên nghiệp tại Bình DươngHoàng Tuấn
 
Công ty tổ chức tiệc trà giá rẻ, teabreak chuyên nghiệp giá rẻ nhất tại tp.hcm
Công ty tổ chức tiệc trà giá rẻ, teabreak chuyên nghiệp giá rẻ nhất tại tp.hcmCông ty tổ chức tiệc trà giá rẻ, teabreak chuyên nghiệp giá rẻ nhất tại tp.hcm
Công ty tổ chức tiệc trà giá rẻ, teabreak chuyên nghiệp giá rẻ nhất tại tp.hcmHoàng Tuấn
 
The best Event Company in HCMC , Vietnam
The best Event Company in HCMC , VietnamThe best Event Company in HCMC , Vietnam
The best Event Company in HCMC , VietnamHoàng Tuấn
 
Cho thuê mc dẫn chương trình tiếng hoa chuyên nghiệp tại tp.hcm
Cho thuê mc dẫn chương trình tiếng hoa chuyên nghiệp tại tp.hcmCho thuê mc dẫn chương trình tiếng hoa chuyên nghiệp tại tp.hcm
Cho thuê mc dẫn chương trình tiếng hoa chuyên nghiệp tại tp.hcmHoàng Tuấn
 

Viewers also liked (20)

Công ty tổ chức sự kiện chuyên nghiệp tại hcm, cần thơ, đà nẵng
Công ty tổ chức sự kiện chuyên nghiệp tại hcm, cần thơ, đà nẵngCông ty tổ chức sự kiện chuyên nghiệp tại hcm, cần thơ, đà nẵng
Công ty tổ chức sự kiện chuyên nghiệp tại hcm, cần thơ, đà nẵng
 
Netiquette
NetiquetteNetiquette
Netiquette
 
Xưởng may, Cho thuê trang phục sự kiện giá rẻ nhất tại Tp.HCM
Xưởng may, Cho thuê trang phục sự kiện giá rẻ nhất tại Tp.HCMXưởng may, Cho thuê trang phục sự kiện giá rẻ nhất tại Tp.HCM
Xưởng may, Cho thuê trang phục sự kiện giá rẻ nhất tại Tp.HCM
 
Linear Local Brand Optimization PowerPoint
Linear Local Brand Optimization PowerPointLinear Local Brand Optimization PowerPoint
Linear Local Brand Optimization PowerPoint
 
식물세포 디펜스
식물세포 디펜스식물세포 디펜스
식물세포 디펜스
 
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCMCho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
 
MC tiếng hoa, MC song ngữ tiếng Hoa, MC chuyên nghiệp tại Tp.HCM
MC tiếng hoa, MC song ngữ tiếng Hoa, MC chuyên nghiệp tại Tp.HCMMC tiếng hoa, MC song ngữ tiếng Hoa, MC chuyên nghiệp tại Tp.HCM
MC tiếng hoa, MC song ngữ tiếng Hoa, MC chuyên nghiệp tại Tp.HCM
 
Embedded
EmbeddedEmbedded
Embedded
 
Beverages & Purchacing
Beverages & PurchacingBeverages & Purchacing
Beverages & Purchacing
 
Decor, trang trí không gian tết chuyên nghiệp tại tp.hcm
Decor, trang trí không gian tết chuyên nghiệp tại tp.hcmDecor, trang trí không gian tết chuyên nghiệp tại tp.hcm
Decor, trang trí không gian tết chuyên nghiệp tại tp.hcm
 
Gabriel Garcia Márquez
Gabriel Garcia MárquezGabriel Garcia Márquez
Gabriel Garcia Márquez
 
Công ty tổ chức sự kiện chuyên nghiệp nhất tại tp.hcm
Công ty tổ chức sự kiện chuyên nghiệp nhất tại tp.hcmCông ty tổ chức sự kiện chuyên nghiệp nhất tại tp.hcm
Công ty tổ chức sự kiện chuyên nghiệp nhất tại tp.hcm
 
Презентация Юлии Бычковой
Презентация Юлии БычковойПрезентация Юлии Бычковой
Презентация Юлии Бычковой
 
Cho thuê mc, cung cap mc chuyen nghiep tai HCM
Cho thuê mc, cung cap mc chuyen nghiep tai HCMCho thuê mc, cung cap mc chuyen nghiep tai HCM
Cho thuê mc, cung cap mc chuyen nghiep tai HCM
 
Human Resource Management
Human Resource ManagementHuman Resource Management
Human Resource Management
 
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCMCho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
Cho thuê ca sỹ , cung cấp ca sỹ chuyên nghiệp tai Tp.HCM
 
Decor, trang trí không gian tết chuyên nghiệp tại Bình Dương
Decor, trang trí không gian tết chuyên nghiệp tại Bình DươngDecor, trang trí không gian tết chuyên nghiệp tại Bình Dương
Decor, trang trí không gian tết chuyên nghiệp tại Bình Dương
 
Công ty tổ chức tiệc trà giá rẻ, teabreak chuyên nghiệp giá rẻ nhất tại tp.hcm
Công ty tổ chức tiệc trà giá rẻ, teabreak chuyên nghiệp giá rẻ nhất tại tp.hcmCông ty tổ chức tiệc trà giá rẻ, teabreak chuyên nghiệp giá rẻ nhất tại tp.hcm
Công ty tổ chức tiệc trà giá rẻ, teabreak chuyên nghiệp giá rẻ nhất tại tp.hcm
 
The best Event Company in HCMC , Vietnam
The best Event Company in HCMC , VietnamThe best Event Company in HCMC , Vietnam
The best Event Company in HCMC , Vietnam
 
Cho thuê mc dẫn chương trình tiếng hoa chuyên nghiệp tại tp.hcm
Cho thuê mc dẫn chương trình tiếng hoa chuyên nghiệp tại tp.hcmCho thuê mc dẫn chương trình tiếng hoa chuyên nghiệp tại tp.hcm
Cho thuê mc dẫn chương trình tiếng hoa chuyên nghiệp tại tp.hcm
 

Similar to RSS 2008 - meta-analyis when assumptions are violated

Meta Analysis in Agriculture by Aman Vasisht
Meta Analysis in Agriculture by Aman VasishtMeta Analysis in Agriculture by Aman Vasisht
Meta Analysis in Agriculture by Aman VasishtAman Vasisht
 
Meta analysis
Meta analysisMeta analysis
Meta analysisSethu S
 
Bio-Statistics in Bio-Medical research
Bio-Statistics in Bio-Medical researchBio-Statistics in Bio-Medical research
Bio-Statistics in Bio-Medical researchShinjan Patra
 
Cochrane Collaboration
Cochrane CollaborationCochrane Collaboration
Cochrane CollaborationNinian Peckitt
 
BASIC STATISTICS AND THEIR INTERPRETATION AND USE IN EPIDEMIOLOGY 050822.pdf
BASIC STATISTICS AND THEIR INTERPRETATION AND USE IN EPIDEMIOLOGY 050822.pdfBASIC STATISTICS AND THEIR INTERPRETATION AND USE IN EPIDEMIOLOGY 050822.pdf
BASIC STATISTICS AND THEIR INTERPRETATION AND USE IN EPIDEMIOLOGY 050822.pdfAdamu Mohammad
 
1. complete stats notes
1. complete stats notes1. complete stats notes
1. complete stats notesBob Smullen
 
David Moher - MedicReS World Congress 2012
David Moher - MedicReS World Congress 2012David Moher - MedicReS World Congress 2012
David Moher - MedicReS World Congress 2012MedicReS
 
Meta analysis
Meta analysisMeta analysis
Meta analysisJunaidAKG
 
Overview of systematic review and meta analysis
Overview of systematic review and meta  analysisOverview of systematic review and meta  analysis
Overview of systematic review and meta analysisDrsnehas2
 
2010 smg training_cardiff_day1_session4_harbord
2010 smg training_cardiff_day1_session4_harbord2010 smg training_cardiff_day1_session4_harbord
2010 smg training_cardiff_day1_session4_harbordrgveroniki
 
When to use, What Statistical Test for data Analysis modified.pptx
When to use, What Statistical Test for data Analysis modified.pptxWhen to use, What Statistical Test for data Analysis modified.pptx
When to use, What Statistical Test for data Analysis modified.pptxAsokan R
 
Research 101: Inferential Quantitative Analysis
Research 101: Inferential Quantitative AnalysisResearch 101: Inferential Quantitative Analysis
Research 101: Inferential Quantitative AnalysisHarold Gamero
 
scope and need of biostatics
scope and need of  biostaticsscope and need of  biostatics
scope and need of biostaticsdr_sharmajyoti01
 
Meta-analysis when the normality assumptions are violated (2008)
Meta-analysis when the normality assumptions are violated (2008)Meta-analysis when the normality assumptions are violated (2008)
Meta-analysis when the normality assumptions are violated (2008)Evangelos Kontopantelis
 
Non parametric study; Statistical approach for med student
Non parametric study; Statistical approach for med student Non parametric study; Statistical approach for med student
Non parametric study; Statistical approach for med student Dr. Rupendra Bharti
 
Introduction to Data Management in Human Ecology
Introduction to Data Management in Human EcologyIntroduction to Data Management in Human Ecology
Introduction to Data Management in Human EcologyKern Rocke
 
PPT on Sample Size, Importance of Sample Size,
PPT on Sample Size, Importance of Sample Size,PPT on Sample Size, Importance of Sample Size,
PPT on Sample Size, Importance of Sample Size,Naveen K L
 

Similar to RSS 2008 - meta-analyis when assumptions are violated (20)

Meta Analysis in Agriculture by Aman Vasisht
Meta Analysis in Agriculture by Aman VasishtMeta Analysis in Agriculture by Aman Vasisht
Meta Analysis in Agriculture by Aman Vasisht
 
Meta analysis
Meta analysisMeta analysis
Meta analysis
 
Bio-Statistics in Bio-Medical research
Bio-Statistics in Bio-Medical researchBio-Statistics in Bio-Medical research
Bio-Statistics in Bio-Medical research
 
Cochrane Collaboration
Cochrane CollaborationCochrane Collaboration
Cochrane Collaboration
 
BASIC STATISTICS AND THEIR INTERPRETATION AND USE IN EPIDEMIOLOGY 050822.pdf
BASIC STATISTICS AND THEIR INTERPRETATION AND USE IN EPIDEMIOLOGY 050822.pdfBASIC STATISTICS AND THEIR INTERPRETATION AND USE IN EPIDEMIOLOGY 050822.pdf
BASIC STATISTICS AND THEIR INTERPRETATION AND USE IN EPIDEMIOLOGY 050822.pdf
 
1. complete stats notes
1. complete stats notes1. complete stats notes
1. complete stats notes
 
David Moher - MedicReS World Congress 2012
David Moher - MedicReS World Congress 2012David Moher - MedicReS World Congress 2012
David Moher - MedicReS World Congress 2012
 
Meta analysis
Meta analysisMeta analysis
Meta analysis
 
Overview of systematic review and meta analysis
Overview of systematic review and meta  analysisOverview of systematic review and meta  analysis
Overview of systematic review and meta analysis
 
Chi-Square Test Non Parametric Test Categorical Variable
Chi-Square Test Non Parametric Test Categorical VariableChi-Square Test Non Parametric Test Categorical Variable
Chi-Square Test Non Parametric Test Categorical Variable
 
2010 smg training_cardiff_day1_session4_harbord
2010 smg training_cardiff_day1_session4_harbord2010 smg training_cardiff_day1_session4_harbord
2010 smg training_cardiff_day1_session4_harbord
 
When to use, What Statistical Test for data Analysis modified.pptx
When to use, What Statistical Test for data Analysis modified.pptxWhen to use, What Statistical Test for data Analysis modified.pptx
When to use, What Statistical Test for data Analysis modified.pptx
 
Research 101: Inferential Quantitative Analysis
Research 101: Inferential Quantitative AnalysisResearch 101: Inferential Quantitative Analysis
Research 101: Inferential Quantitative Analysis
 
scope and need of biostatics
scope and need of  biostaticsscope and need of  biostatics
scope and need of biostatics
 
Meta-analysis when the normality assumptions are violated (2008)
Meta-analysis when the normality assumptions are violated (2008)Meta-analysis when the normality assumptions are violated (2008)
Meta-analysis when the normality assumptions are violated (2008)
 
UNIT 5.pptx
UNIT 5.pptxUNIT 5.pptx
UNIT 5.pptx
 
Non parametric study; Statistical approach for med student
Non parametric study; Statistical approach for med student Non parametric study; Statistical approach for med student
Non parametric study; Statistical approach for med student
 
Introduction to Data Management in Human Ecology
Introduction to Data Management in Human EcologyIntroduction to Data Management in Human Ecology
Introduction to Data Management in Human Ecology
 
Statistics
StatisticsStatistics
Statistics
 
PPT on Sample Size, Importance of Sample Size,
PPT on Sample Size, Importance of Sample Size,PPT on Sample Size, Importance of Sample Size,
PPT on Sample Size, Importance of Sample Size,
 

More from Evangelos Kontopantelis

Investigating the relationship between quality of primary care and premature ...
Investigating the relationship between quality of primary care and premature ...Investigating the relationship between quality of primary care and premature ...
Investigating the relationship between quality of primary care and premature ...Evangelos Kontopantelis
 
Re-analysis of the Cochrane Library data and heterogeneity challenges
Re-analysis of the Cochrane Library data and heterogeneity challengesRe-analysis of the Cochrane Library data and heterogeneity challenges
Re-analysis of the Cochrane Library data and heterogeneity challengesEvangelos Kontopantelis
 
Attikon 2014 - Software and model selection challenges in meta-analysis
Attikon 2014 - Software and model selection challenges in meta-analysisAttikon 2014 - Software and model selection challenges in meta-analysis
Attikon 2014 - Software and model selection challenges in meta-analysisEvangelos Kontopantelis
 
RSS 2013 - A re-analysis of the Cochrane Library data]
RSS 2013 - A re-analysis of the Cochrane Library data]RSS 2013 - A re-analysis of the Cochrane Library data]
RSS 2013 - A re-analysis of the Cochrane Library data]Evangelos Kontopantelis
 
Faculty showcase 2013 - Opening up clinical performance
Faculty showcase 2013 - Opening up clinical performanceFaculty showcase 2013 - Opening up clinical performance
Faculty showcase 2013 - Opening up clinical performanceEvangelos Kontopantelis
 
SAPC 2013 - general practice clinical systems
SAPC 2013 - general practice clinical systemsSAPC 2013 - general practice clinical systems
SAPC 2013 - general practice clinical systemsEvangelos Kontopantelis
 
Internal 2013 - General practice clinical systems
Internal 2013 - General practice clinical systemsInternal 2013 - General practice clinical systems
Internal 2013 - General practice clinical systemsEvangelos Kontopantelis
 
Internal 2012 - Software and model selection challenges in meta-analysis
Internal 2012 - Software and model selection challenges in meta-analysisInternal 2012 - Software and model selection challenges in meta-analysis
Internal 2012 - Software and model selection challenges in meta-analysisEvangelos Kontopantelis
 
Internal 2012 - individual patient data meta-analysis
Internal 2012 - individual patient data meta-analysisInternal 2012 - individual patient data meta-analysis
Internal 2012 - individual patient data meta-analysisEvangelos Kontopantelis
 
Amsterdam 2012 - one stage meta-analysis
Amsterdam 2012 - one stage meta-analysisAmsterdam 2012 - one stage meta-analysis
Amsterdam 2012 - one stage meta-analysisEvangelos Kontopantelis
 
NIHR School for primary care showcase 2012 - financial incentives
NIHR School for primary care showcase 2012 - financial incentivesNIHR School for primary care showcase 2012 - financial incentives
NIHR School for primary care showcase 2012 - financial incentivesEvangelos Kontopantelis
 
RSS local 2012 - Software challenges in meta-analysis
RSS local 2012 - Software challenges in meta-analysisRSS local 2012 - Software challenges in meta-analysis
RSS local 2012 - Software challenges in meta-analysisEvangelos Kontopantelis
 
SAPC north 2010 - provider incentives for influenza immunisation
SAPC north 2010 - provider incentives for influenza immunisationSAPC north 2010 - provider incentives for influenza immunisation
SAPC north 2010 - provider incentives for influenza immunisationEvangelos Kontopantelis
 
HSRN 2010: incentivisation and non-incentivised aspects of care
HSRN 2010: incentivisation and non-incentivised aspects of careHSRN 2010: incentivisation and non-incentivised aspects of care
HSRN 2010: incentivisation and non-incentivised aspects of careEvangelos Kontopantelis
 

More from Evangelos Kontopantelis (20)

Primary Care data signposting
Primary Care data signpostingPrimary Care data signposting
Primary Care data signposting
 
Investigating the relationship between quality of primary care and premature ...
Investigating the relationship between quality of primary care and premature ...Investigating the relationship between quality of primary care and premature ...
Investigating the relationship between quality of primary care and premature ...
 
Re-analysis of the Cochrane Library data and heterogeneity challenges
Re-analysis of the Cochrane Library data and heterogeneity challengesRe-analysis of the Cochrane Library data and heterogeneity challenges
Re-analysis of the Cochrane Library data and heterogeneity challenges
 
Attikon 2014 - Software and model selection challenges in meta-analysis
Attikon 2014 - Software and model selection challenges in meta-analysisAttikon 2014 - Software and model selection challenges in meta-analysis
Attikon 2014 - Software and model selection challenges in meta-analysis
 
Internal 2014 - data signposting
Internal 2014 - data signpostingInternal 2014 - data signposting
Internal 2014 - data signposting
 
Internal 2014 - Cochrane data
Internal 2014 - Cochrane dataInternal 2014 - Cochrane data
Internal 2014 - Cochrane data
 
RSS 2013 - A re-analysis of the Cochrane Library data]
RSS 2013 - A re-analysis of the Cochrane Library data]RSS 2013 - A re-analysis of the Cochrane Library data]
RSS 2013 - A re-analysis of the Cochrane Library data]
 
Faculty showcase 2013 - Opening up clinical performance
Faculty showcase 2013 - Opening up clinical performanceFaculty showcase 2013 - Opening up clinical performance
Faculty showcase 2013 - Opening up clinical performance
 
SAPC 2013 - general practice clinical systems
SAPC 2013 - general practice clinical systemsSAPC 2013 - general practice clinical systems
SAPC 2013 - general practice clinical systems
 
Internal 2013 - General practice clinical systems
Internal 2013 - General practice clinical systemsInternal 2013 - General practice clinical systems
Internal 2013 - General practice clinical systems
 
Internal 2012 - Software and model selection challenges in meta-analysis
Internal 2012 - Software and model selection challenges in meta-analysisInternal 2012 - Software and model selection challenges in meta-analysis
Internal 2012 - Software and model selection challenges in meta-analysis
 
Internal 2012 - individual patient data meta-analysis
Internal 2012 - individual patient data meta-analysisInternal 2012 - individual patient data meta-analysis
Internal 2012 - individual patient data meta-analysis
 
Amsterdam 2012 - one stage meta-analysis
Amsterdam 2012 - one stage meta-analysisAmsterdam 2012 - one stage meta-analysis
Amsterdam 2012 - one stage meta-analysis
 
SAPC 2012 - exception reporting
SAPC 2012 - exception reportingSAPC 2012 - exception reporting
SAPC 2012 - exception reporting
 
NIHR School for primary care showcase 2012 - financial incentives
NIHR School for primary care showcase 2012 - financial incentivesNIHR School for primary care showcase 2012 - financial incentives
NIHR School for primary care showcase 2012 - financial incentives
 
RSS 2012 - ipdforest
RSS 2012 - ipdforestRSS 2012 - ipdforest
RSS 2012 - ipdforest
 
RSS local 2012 - Software challenges in meta-analysis
RSS local 2012 - Software challenges in meta-analysisRSS local 2012 - Software challenges in meta-analysis
RSS local 2012 - Software challenges in meta-analysis
 
SAPC north 2010 - provider incentives for influenza immunisation
SAPC north 2010 - provider incentives for influenza immunisationSAPC north 2010 - provider incentives for influenza immunisation
SAPC north 2010 - provider incentives for influenza immunisation
 
HSRN 2010: incentivisation and non-incentivised aspects of care
HSRN 2010: incentivisation and non-incentivised aspects of careHSRN 2010: incentivisation and non-incentivised aspects of care
HSRN 2010: incentivisation and non-incentivised aspects of care
 
GPRD 2010
GPRD 2010GPRD 2010
GPRD 2010
 

Recently uploaded

Digital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxDigital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxMohamedFarag457087
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learninglevieagacer
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bSérgio Sacani
 
Conjugation, transduction and transformation
Conjugation, transduction and transformationConjugation, transduction and transformation
Conjugation, transduction and transformationAreesha Ahmad
 
Stages in the normal growth curve
Stages in the normal growth curveStages in the normal growth curve
Stages in the normal growth curveAreesha Ahmad
 
Pulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceuticsPulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceuticssakshisoni2385
 
COMPUTING ANTI-DERIVATIVES (Integration by SUBSTITUTION)
COMPUTING ANTI-DERIVATIVES(Integration by SUBSTITUTION)COMPUTING ANTI-DERIVATIVES(Integration by SUBSTITUTION)
COMPUTING ANTI-DERIVATIVES (Integration by SUBSTITUTION)AkefAfaneh2
 
Module for Grade 9 for Asynchronous/Distance learning
Module for Grade 9 for Asynchronous/Distance learningModule for Grade 9 for Asynchronous/Distance learning
Module for Grade 9 for Asynchronous/Distance learninglevieagacer
 
Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.Silpa
 
Zoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfZoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfSumit Kumar yadav
 
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....muralinath2
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsSérgio Sacani
 
Introduction of DNA analysis in Forensic's .pptx
Introduction of DNA analysis in Forensic's .pptxIntroduction of DNA analysis in Forensic's .pptx
Introduction of DNA analysis in Forensic's .pptxrohankumarsinghrore1
 
Use of mutants in understanding seedling development.pptx
Use of mutants in understanding seedling development.pptxUse of mutants in understanding seedling development.pptx
Use of mutants in understanding seedling development.pptxRenuJangid3
 
GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)Areesha Ahmad
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)Areesha Ahmad
 
An introduction on sequence tagged site mapping
An introduction on sequence tagged site mappingAn introduction on sequence tagged site mapping
An introduction on sequence tagged site mappingadibshanto115
 
Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Silpa
 
Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .Poonam Aher Patil
 

Recently uploaded (20)

Digital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxDigital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptx
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learning
 
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 bAsymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
Asymmetry in the atmosphere of the ultra-hot Jupiter WASP-76 b
 
Conjugation, transduction and transformation
Conjugation, transduction and transformationConjugation, transduction and transformation
Conjugation, transduction and transformation
 
Stages in the normal growth curve
Stages in the normal growth curveStages in the normal growth curve
Stages in the normal growth curve
 
Pulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceuticsPulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceutics
 
COMPUTING ANTI-DERIVATIVES (Integration by SUBSTITUTION)
COMPUTING ANTI-DERIVATIVES(Integration by SUBSTITUTION)COMPUTING ANTI-DERIVATIVES(Integration by SUBSTITUTION)
COMPUTING ANTI-DERIVATIVES (Integration by SUBSTITUTION)
 
Module for Grade 9 for Asynchronous/Distance learning
Module for Grade 9 for Asynchronous/Distance learningModule for Grade 9 for Asynchronous/Distance learning
Module for Grade 9 for Asynchronous/Distance learning
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.
 
Zoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfZoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdf
 
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....
Human & Veterinary Respiratory Physilogy_DR.E.Muralinath_Associate Professor....
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
 
Introduction of DNA analysis in Forensic's .pptx
Introduction of DNA analysis in Forensic's .pptxIntroduction of DNA analysis in Forensic's .pptx
Introduction of DNA analysis in Forensic's .pptx
 
Use of mutants in understanding seedling development.pptx
Use of mutants in understanding seedling development.pptxUse of mutants in understanding seedling development.pptx
Use of mutants in understanding seedling development.pptx
 
GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)
 
An introduction on sequence tagged site mapping
An introduction on sequence tagged site mappingAn introduction on sequence tagged site mapping
An introduction on sequence tagged site mapping
 
Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.
 
Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .
 

RSS 2008 - meta-analyis when assumptions are violated

  • 1. A comparison of Random Effects meta-analysis methods when study effects are non-normally distributed Evan Kontopantelis & David Reeves NPCRDC
  • 2. How meta-analysis works • A search for papers relevant to the research question is conducted. Unsuitable papers are filtered out • In each paper… – for each outcome measure that is directly relevant to the RQ, or a good enough proxy, we calculate an effect (of intervention vs control) and its variance – An overall effect and variance is selected • Effects and their variances are combined to calculate an overall effect Chronic disease - Risk factors effect -.4 0 .4 .8 Combined Woolard(B), 1995 Woolard(A), 1995 Eckerlund, 1985 Moher, 2001 Cupples, 1994 Campbell, 1998 Van Ree, 1985
  • 3. Heterogeneity • Heterogeneity can be attributed to clinical and/or methodological diversity • Clinical heterogeneity: variability that arises from different populations, interventions, outcomes and follow-up times • Methodological heterogeneity: relates to differences in trial design and quality • Detecting (usually with Cochran’s Q test) quantifying and dealing with heterogeneity can be very hard
  • 4. Absence of heterogeneity • Assumes that the true effects of the studies are all equal and deviations occur because of imprecision of results • Analysed with fixed- effects method i iY e 
  • 5. Presence of heterogeneity • It is assumed that there exists variation in the size of the true effect among studies (in addition to the imprecision in results) • Analysed with random-effects methods i i iY e 
  • 6. Random-effect MA methods • Estimate the between-study variance and use it in estimating the overall effect • Parametric: – DerSimonian-Laird (1986) – Maximum & Profile likelihood (1996) • Non-parametric: – Permutations method (1999) – Non Parametric Maximum Likelihood (1999) 2  
  • 7. “Potential” problems? • Heterogeneity is common & the FE model is under fire • Parametric RE models assume that both the effects and errors are normally distributed • Almost all RE models (except PL) do not take account of uncertainty in • DL is usually the preferred method of analysis because it is easy to implement and is available in all software packages 2 ˆ
  • 8. So far… • The number of studies and the amount of heterogeneity have been found to affect method performance • Performance comparisons usually focus on coverage and ignore power or have not included some important methods (e.g. PL, PE) • Evaluations were based on normal data: method robustness has not been assessed with non-normal data
  • 10. In a nutshell • Simulated various non-normal distributions for the true effects: skew normal, bimodal, beta, uniform, U and others • Created datasets of 10000 meta-analyses for various numbers of studies k and different degrees of heterogeneity, for each distributional assumption • Compared FE, DL, ML, PL and PE methods (along with a simple t-test) in terms of coverage and power across all datasets
  • 11. Generating the data • For a single study we simulated the effect size estimate and the within-study variance estimate of a binary outcome • The variance was assumed to be a realisation from a distribution, multiplied by .25 and restricted to the (.009, .6) interval • involves two components – where – • Four values were used: .01, .03, .07 & .1 • Number of studies (MA size) varied from 2 to 35 iY i 2 ˆ 2 1 iY i i iY e( )  2 (0, ˆ )i ie  2  k i i    2 ?(0, )i 
  • 12. Details on the MA methods • Fixed effects (FE) • DerSimonian-Laird (DL) • Q method (Q) • Maximum Likelihood (ML) • Profile Likelihood (PL) • Permutations method (PE) • T-test method (T)
  • 13. Performance • For each simulated meta-analysis case we calculated confidence intervals for the overall effect estimate , for all the methods • Coverage: % of confidence intervals that contain the true overall effect in a sample of 10000 meta-analyses • Power: % of CIs that do not contain the 25th centile of the population distribution of the 10000 effect sizes ˆ 
  • 19. PL performance across various distributions
  • 21. Summary • Within any given method, the results were consistent across all types of distribution shape • This can give researchers confidence that methods are highly robust against even the most severe violation of the assumption of normally distributed effect sizes • If it is reasonable to assume that the effect size does not vary between studies, the FE, Q and ML methods all provide accurate coverage coupled with good power
  • 22. In the presence of heterogeneity… • However, zero between study variance is the exception rather than the norm and the presence of even a moderate amount of alters the picture considerably • FE, Q and ML quickly lose coverage as heterogeneity increases • DL rapidly goes from providing a coverage that is overly high, to one that is overly low • PE, and to a lesser extend PL, now provide the best coverage, even with very small sample sizes 2 
  • 23. Which method then? • If priority is given to maintaining an accurate Type I error rate then the simple t-test is the best method. But its power is very low, making it a poor choice when control of the Type II error rate is also important • PE gives accurate coverage in all situations and has better power than T, but the method is more difficult to implement and cannot be used with less than 6 studies • PL has ‘reasonable’ coverage in most situations, giving it an edge over other methods
  • 24. Current & future work • Created a freely available Excel add-in that implements all the described MA methods and various measures of heterogeneity • Working on a STATA module that will do the same • Investigate performance of heterogeneity measures under non-normally distributed data
  • 25. Main references • Brockwell SE, Gordon IR. A comparison of statistical methods for meta-analysis. Stat.Med. 2001; 20(6):825-840 • Engels EA, Schmid CH, Terrin N, Olkin I, Lau J. Heterogeneity and statistical significance in meta-analysis: an empirical study of 125 meta-analyses. Stat.Med. 2000; 19(13):1707-1728 • Follmann DA, Proschan MA. Valid inference in random effects meta- analysis. Biometrics 1999; 55(3):732-737 • Hardy RJ, Thompson SG. A likelihood approach to meta-analysis with random effects. Stat.Med. 1996; 15(6):619-629 • Micceri T. The Unicorn, the Normal Curve, and Other Improbable Creatures. Psychological Bulletin 1989; 105(1):156-166 • Ramberg JS, Dudewicz EJ, Tadikamalla PR, Mykytka EF. A Probability Distribution and Its Uses in Fitting Data. Technometrics 1979; 21(2):201-214
  • 26. Thank you for listening