Difference between IPTW and MAIC? Could someone describe the difference between IPTW and MAIC methodologies for indirect treatment comparison?
 A: Matching-adjusted indirect comparison (MAIC) is a misnomer; the comparison is actually weighting adjusted. To perform MAIC, you estimate a version of propensity score weights using a form of logistic regression that relies on method of moments rather than maximum likelihood. This is done because individual patient data exists only for those in one sample, and your logistic regression models the probability of being in the sample. These weights are then applied to the sample with individual patient data to estimate the expected outcome in the weighted focal treatment group, which after weighting is comparable to the other treatment group (for which only aggregate outcomes are available). The method of estimating the MAIC weights is almost identical to CBPS and entropy balancing.
Inverse probability of treatment weighting (IPTW) is used when you have individual patient data for all members of your study sample. For indirect comparison, this would mean having individual patient data for both studies. You could then estimate the probability of being in one study and use that to weight the focal treatment group to be comparable to the other. 
Both methods rely on estimating weights to balance two groups. With MAIC, you only have individual patient data for one of the groups. When you have data for both groups, you can use IPTW. 
