Cross-over trials may, in principle, be combined with parallel group trials in the same meta-analysis. Consideration should be given to the possibility of important differences in other characteristics between the different types of trial. For example, cross-over trials may have shorter intervention periods or may include participants with less severe illness. It is generally advisable to meta-analyse parallel-group and cross-over trials separately irrespective of whether they are also combined together.
Authors should explicitly state how they have dealt with data from cross-over trials and should conduct sensitivity analyses to investigate the robustness of their conclusions, especially when correlation coefficients have been borrowed from external sources (see Chapter 9, Section 9.7). Statistical support is recommended.