A meta-analysis based on Bayesian statistics for combining results of multiple studies that have different research designs, such as 
randomized controlled trials and 
observational studies, by adjusting the results of each individual study for its methodological biases before combining the results into a probability distribution for the parameter(s) of interest.