From: Incorporating published univariable associations in diagnostic and prognostic modeling
 |  | No meta-analysis | Greenland/Steyerberg | Improved adaptation method | |
---|---|---|---|---|---|
 |  |  | adaptation method | Variant 1 | Variant 2 |
Step 1 | Estimate associations in IPD | Â | |||
 | Implemented | Yes | Yes | Yes | Yes |
 | Association type | m | u+m | u+m | u+m |
 | Prior distribution | none | none | none | weakly informative |
Step 2 | Summarize univariable associations | Â | |||
 | Implemented | No | Yes | Yes | Yes |
 | Source | - | I+L | I+L | I+L |
 | Pooling Method | - | random effects | random effects | random effects |
Step 3 | Estimate adaptation from univariable to multivariable association | Â | |||
 | Implemented | No | Yes | Yes | Yes |
 | Assumptions | - | (1)+(2) | (1) | (1) |
 | Estimation procedure | - | analytic | bootstrap | bootstrap |
 | Prior distributions | - | none | none | weakly informative |
Step 4 | Apply adaptation to summary estimate from the literature and estimate β m|L |  | |||
 | Implemented | No | Yes | Yes | Yes |