FDA and EMA Guidance on Adjusting for Covariates in Randomized Clinical Trials

Last week, FDA issues its draft guidance for industry titled 'Adjusting for Covariates in Randomized Clinical Trials for Drugs and Biologicals with Continuous Outcomes'. The guidance is short and sweet and gives five recommendations:

• Sponsors can use ANCOVA to adjust for differences between treatment groups in relevant baseline variables to improve the power of significance tests and the precision of estimates of treatment effect. 
• Sponsors should not use ANCOVA to adjust for variables that might be affected by treatment. 
• The sponsor should prospectively specify the covariates and the mathematical form of the model in the protocol or statistical analysis plan. 
• Interaction of the treatment with covariates is important, but the presence of an interaction does not invalidate ANCOVA as a method of estimating and testing for an overall treatment effect, even if the interaction is not accounted for in the model. The prespecified primary model can include interaction terms if appropriate. 
• Many clinical trials use a change from baseline as the primary outcome measure. Even when the outcome is measured as a change from baseline, the baseline value can still be used advantageously as a covariate. 

Not sure why the guidance is only for clinical trials 'with continuous outcomes' and ANCOVA. Adjusting for covariates is also applicable for studies with other types of outcomes: time to event outcomes analyzed using Cox regression, categorical outcomes analyzed using logistical regression,... It is also important to follow the same rules in these studies when dealing with the covariates.

The second recommendation essentially said that the post-randomization or post-treatment variables should not be used as covariates in analyses - which is consistent with what was laid out in EMA's guidelines (see below). If there is a time-dependent covariate in longitudinal studies, the best way is to come up with a pre-adjustment formula instead of using the time-dependent covariate as a covariate in the model. There is no discussion of whether or not the post-treatment covariates can be used in the imputation model when multiple imputation method is used to impute the missing data. There was a misperception that the post-treatment variables could be used in the imputation model, but not in the analysis model. 

EMA had a similar guidance "Guideline on adjustment for baseline covariates in clinical trials" - it was in effect more than three years earlier than FDA's guidance and it was much longer (11 pages versus 3 pages in FDA's guidance) with more details about its recommendations. The pre-specification of the covariates to be used and avoidance of using the post-randomization variables as covariates was also emphasized. Here are executive summaries:

• Stratification may be used to ensure balance of treatments across covariates; it may also be used for administrative reasons (e.g. block in the case of block randomisation). The factors that are the basis of stratification should normally be included as covariates or stratification variables in the primary outcome model, except where stratification was done purely for an administrative reason.
• Variables known a priori to be strongly, or at least moderately, associated with the primary outcome and/or variables for which there is a strong clinical rationale for such an association should also be considered as covariates in the primary analysis. The variables selected on this basis should be pre-specified in the protocol. 
• Baseline imbalance observed post hoc should not be considered an appropriate reason for including a variable as a covariate in the primary analysis. However, conducting exploratory analyses including such variables when large baseline imbalances are observed might be helpful to assess the robustness of the primary analysis.
• Variables measured after randomisation and so potentially affected by the treatment should not be included as covariates in the primary analysis. 
• If a baseline value of a continuous primary outcome measure is available, then this should usually be included as a covariate. This applies whether the primary outcome variable is defined as the ‘raw outcome’ or as the ‘change from baseline’.
• Covariates to be included in the primary analysis must be pre-specified in the protocol. 
• Only a few covariates should be included in a primary analysis. Although larger data sets may support more covariates than smaller ones, justification for including each of the covariates should be provided. 
• In the absence of prior knowledge, a simple functional form (usually either linearity or categorising a continuous scale) should be assumed for the relationship between a continuous covariate and the outcome variable. 
• The validity of model assumptions must be checked when assessing the results. This is particularly important for generalised linear or non-linear models where mis-specification could lead to incorrect estimates of the treatment effect. Even under ordinary linear models, some attention should be paid to the possible influence of extreme outlying values. 
• Whenever adjusted analyses are presented, results of the treatment effect in subgroups formed by the covariates (appropriately categorised, if relevant) should be presented to enable an assessment of the model assumptions.
• Sensitivity analyses should be pre-planned and presented to investigate the robustness of the primary analysis. Discrepancies should be discussed and explained. In the presence of important differences that cannot be logically explained – for example, between the results of adjusted and unadjusted analyses – the interpretation of the trial could be seriously affected. 
• The primary model should not include treatment by covariate interactions. If substantial interactions are expected a priori, the trial should be designed to allow separate estimates of the treatment effects in specific subgroups. • Exploratory analyses may be carried out to improve the understanding of covariates not included in the primary analysis, and to help the sponsor with the ongoing development of the drug. • In case of missing values in baseline covariates the principles for dealing with missing values as outlined e.g. in the Guideline on missing data in confirmatory clinical trials(EMA/CPMP/EWP/1776/99 Rev. 1) applies. 
• A primary analysis, unambiguously pre-specified in the protocol, correctly carried out and interpreted, should support the conclusions which are drawn from the trial. Since there may be a number of alternative valid analyses, results based on pre-specified analyses will carry most credibility.