What's next? There are still many interesting questions at the trait level. For example, heritability estimates for BMI vary substantially across methods and suggest that more complex environmental interactions may be at play (an echo of prior work: pubmed.ncbi.nlm.nih.gov/28692066/).

So there you have it, twin study estimates were greatly inflated, and molecular data sets the record straight. I walk through possible counter-arguments, but ultimately the uncomfortable truth is that genes contribute to traits much less than we always thought.

Is it true that for RCT & the very common continuous-level measurements, mean (CI) treatment effect estimation is exactly the same whether one uses change scores or follow-up measurements as the outcome - as long as ANCOVA with baseline covariate is used?

Read more about RCTs and why they’re so important: ourworldindata.org/randomized-c...

Find the full paper here 🔗 doi.org/10.1123/ijsn...