Systematic reviews may exclude crossover and cluster randomized trials for several reasons:

1. Complexity of Analysis: These designs require complex statistical methods, making it difficult to synthesize results with other study designs.

2. Risk of bias: Crossover trials may introduce bias from carryover effects, skewing results.

3. Heterogeneity of Interventions: Variations between clusters in cluster randomized trials can complicate comparisons.

4. Focus on Specific Study Designs: Some reviews exclusively evaluate parallel-group designs for clearer comparisons.

5. Guidelines and Standards: Many review guidelines emphasize clarity and consistency in study selection criteria.

In summary, while these methods have merits, their complexities often lead to exclusion from well-designed systematic reviews.

Systematic reviews may exclude crossover and cluster randomized trials for several reasons:

1. Complexity of Analysis: These designs require complex statistical methods, making it difficult to synthesize results with other study designs.

2. Risk of bias: Crossover trials may introduce bias from carryover effects, skewing results.

3. Heterogeneity of Interventions: Variations between clusters in cluster randomized trials can complicate comparisons.

4. Focus on Specific Study Designs: Some reviews exclusively evaluate parallel-group designs for clearer comparisons.

5. Guidelines and Standards: Many review guidelines emphasize clarity and consistency in study selection criteria.

In summary, while these methods have merits, their complexities often lead to exclusion from well-designed systematic reviews.

I'm quite confident you more or less just pasted the answer an LLM generated, Eshetu Zemen. It's my impression that none of these hurdles would be very challenging to overcome or likely to harm the quality of an analysis.

When I asked, I got a similarly generic and unsatisfying answer:

A well-designed systematic review might exclude studies conducted by crossover or cluster randomization methods for several reasons:

However, it's worth noting that excluding these studies might also exclude valuable data, especially in fields where crossover or cluster designs are common or particularly appropriate due to ethical or practical considerations. Modern systematic reviews might instead choose to include these studies but analyze them separately or use methods to account for their unique characteristics in the meta-analysis.

Dear Matthew Elvey

Thank you for inquiring about this question.

It's important to note that certain health and treatment subjects cannot be individually randomized and must instead be randomized in a cluster or crossover. The general rule that a good systematic review should exclude this type of randomization may be associated with bias. Therefore, it's better to decide which designs to exclude based on the scope and topic of your study. Additionally, in Clinical Trial studies, the study quality check tool, such as the Rob one or two, mentions the randomization method and assigns a grade to the study based on that.

Taking all of this into consideration, I hope you can make the right decision in choosing the appropriate studies for your systematic review.

It shall depend on the specific research question for which the systematic review is designed to answer. In addition, it also depends on the nature of the interventions [for both the experimental group and the control group] that the systematic review is intended to examine.

This is because if the systematic review is going to include cluster randomized trials, then it will be necessary to address the unit of analysis issues if the authors themselves have not adjusted the unit of analysis for the effects of clustering. To do so, it will be first necessary to ensure the availability of a valid and externally-derived intra-cluster correlation.

On the other hand, if the systematic review is going to include a cross-over clinical trial, then the authors must make that the investigators of these cross-over trials had allowed an adequate “wash-out period” between two differing types of treatments. If it is not possible to ascertain the adequacy of “wash-out” period between the treatments, then the validity of the results obtained from this cross-over trial cannot be determined. Furthermore, some of the systematic reviews are intended to assess the effects of screening methods in which there is a high likelihood of serious carry-over, therefore the lasting nature of such interventions will mean that the cross-over trial design will have a high chance to have a carry-over effect, which means that the validity of its results will very likely to be questionable.

Finally, if a systematic review is going to include a crossover trial or a cluster randomized trial, then the assessment of its risk of bias will have to also take into account a series of additional risk of bias domains, such as bias arising from period effect and carryover effect for cross-over RCTs, and bias arising from the identification or recruitment of participants into cluster for cluster-RCTs.

Therefore, unless there is an absolute necessity for a systematic review to include a crossover or a cluster randomized trial; it is usually preferable to exclude a crossover or a cluster randomized trial in a systematic review.

That being said, the final decision to either include [or exclude] a cross-over trial or a cluster randomized trial in a systematic review shall depend ultimately on the specific review question for which the systematic review is designed to answer.