We are writing a protocol for PROSPERO for a pairwise meta-analysis comparing comparing two cardiovascular treatments in terms of their effects on a dichotomous endpoint, such as death. We expect to include 5 o 6 homogeneous trials, eventually (thus most likely statistical inconsistency will be minimal, e.g. I-squared will be 0).

We are uncertain as to whether opt for the Peto method to compute odds ratios, or the Mantel-Haenszel fixed-effect method.

It appears the Peto method might be less robust, as it relies on more assumptions (eg Khera et al), but the Cochrane Collaboration considers both Peto and Mantel-Haenszel approaches as fine (see for instance section 9.4.4 Meta-analysis of dichotomous outcomes of the Cochrane Handbook for Systematic Reviews of Interventions).

Any suggestion of which method to prefer for the primary analysis in the protocol (leaving the other one for sensitivity analysis)?

  • 1
    $\begingroup$ I assume you are analyzing the occurrence of some type of cardiovascular events. Do you only have aggregate data or do you have (or can get) individual patient data (so that you could do a time-to-event analysis)? Are results such as (log) hazard ratios with SE or 95% CI available for each trial (in which case it may be best to meta-analyze those using e.g. inverse variance, unless e.g. the events are rare)? Are these outcomes rare (with some trials having very few or no events occurring) so that you need methods that are robust in that setting? Are the trials of identical duration? $\endgroup$
    – Björn
    Mar 18, 2016 at 9:15
  • $\begingroup$ Mine is more a strategic/theoretical question, as currently state of the art meta-analyses need to be designed and registered online before starting data collection or analysis. Yet, I would expect to have only aggregate data, reported as raw numbers (events per patients at risk, so n/N), with some variability in follow-up duration (eg 1 vs 3-4 years). The reasonable rate of events would be between 5% and 20%. $\endgroup$ Mar 18, 2016 at 23:41
  • $\begingroup$ Suppose Peto and M-H give different results? What will you do then? $\endgroup$
    – mdewey
    Mar 19, 2016 at 15:23
  • $\begingroup$ Again, the scenario of most traditional meta-analyses, in which you can play around with many different approaches, is not that difficult to face. If two or more methods agree, then it is not a key issue. Conversely, if they disagree, that's casting a shadow of doubt on the results. Yet, the key point is: when drafting a meta-analysis protocol, is there any reason to choose Peto over MH? Or the other way around? $\endgroup$ Mar 19, 2016 at 18:50

1 Answer 1


For a start, using time-to-event methods on the individual patient data is the gold standard approach and one should try to get it. It is important to realise that assuming a binomial distribution means that the losses to follow-ups in the arms being compared need to follow the exact same distribution. If the number of losses to follow-up are similar and there is no suspicion of different time patterns in drop-outs, a binomial distribution may be reasonable and okay for detecting an effect. Conventional wisdom 10 or so years ago was that

  1. The Peto-Yussuf one-step odds ratio has a relatively good reputationfor rare events (with some potential problems for unbalanced randomization).
  2. For slightly less rare events, the Mantel-Haenszel fixed-effects odds-ratio (however, note: zero-cell corrections are a difficult topic, because you do not want to ignore trials with some zero cell, but a 0.5 correction biases the result too much towards no effect) and logistic regression (standard, exact or Firth penalized likelihood logistic regression) have also been reported to be relatively good methods.

This is, I believe what the Cochrane handbook bases its recommendations on. However, this is an area of ongoing research and papers are regularly appearing proposing new methods in particular with respect to trials with no (or very few events) e.g. this one last year. For a particular problem, I would have an in-depth discussion with people familiar with the disease area and statisticians specializing in meta-analysis to pick a reasonable method that should perform decently for the particular question.

  • $\begingroup$ None of the links in this answer work for me $\endgroup$
    – mdewey
    Mar 19, 2016 at 13:49
  • $\begingroup$ They should work now. $\endgroup$
    – Björn
    Mar 20, 2016 at 6:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.