Please do give our paper a read -- it's a fairly short one (7 pages). Thank you for your time and your attention to our research! n/n

For example, the landmark WHI trial reported additional risks of invasive breast cancer of 9 per 10,000 women per year. In contrast, we estimate that an extra year of MHT use increases the risk of invasive breast cancer by 13 per 10,000 women per year. 6/n

We find that the risks of MHT (increased risk of invasive breast cancer, pulmonary embolism), as well as the benefits of MHT (decreased risk of endometrial cancer and hip fractures), are 22% higher using IV methods relative to previously reported ITT estimates. 5/n

Note for those designing clinical trials: we can apply IV because the WHI team (including my coauthor Marcia, one of the WHI investigators) collected data on adherence to MHT in the treatment group, *AND* those randomized to placebo who started MHT outside of the trial. 4/n

Our paper estimates the effect of menopausal hormone therapy (MHT) on health outcomes using individual-level clinical trial data from the landmark WHI trial combined with IV methods, which account for deviations from randomized assignment. 3/n

Most trials report only intent-to-treat (ITT) estimates, which measure the effect of being randomized to treatment vs. control on outcomes. Prior WHI studies have also reported only ITT estimates. However, ITT does not account for imperfect adherence to randomized assignment. 2/n

This milestone would not have been possible without the support of my wonderful advisors: David Cutler (@davidmcutler.bsky.social), Jason Abaluck (@jabaluck.bsky.social), Zack Cooper, Mark Shepard, and my JMP coauthor Samantha Burn. Thank you all—I couldn’t have done it without you! 3/3

I feel so fortunate to continue studying health and health care and how to improve them at an institution as amazing as the LBJ School and UT Austin. Thank you for welcoming me into your community. 2/3