Traditionally used with individual-level data to generate the PRS and combined using one-sample MR methodology (e.g., two-stage least squares) but summary statistics from the PRS-exposure and PRS-outcome associations can be obtained and used to derive MR estimates with two-sample MR methodology (e.g., Wald ratio). The univariate score must be a valid instrument. More robust to weak instrument bias than two-stage least squares (TSLS). Hard to test for instrument validity.
- Burgess S, Dudbridge F, Thompson SG. Combining information on multiple instrumental variables in Mendelian randomization: comparison of allele score and summarized data methods. Statistics in medicine 2016;35:1880-1906.