iiinova — where your digital self keeps evolving

Reposted by Miguel Hernan

CAUSALab @causalab.org · 22/06/2026

The 2026 CAUSALab Summer Courses on Causal Inference have come to a close 📝 CAUSALab was excited to host 4 unique courses @hsph.harvard.edu. This year's 330+ participants represented: 🏢 180+ organizations 🌎 48 countries 📍 30 U.S. states Thank you for an exciting two weeks of #causalinference!

042

Miguel Hernan @miguelhernan.org · 02/04/2026

Two common misconceptions when repurposing data for #causalinference: 1) the target trial is an ideal trial 2) the target trial protocol can be prespecified Our new paper examines how the target trial protocol depends on the causal question AND the available data. journals.lww.com/epidem/abstr...

1209

Reposted by Miguel Hernan

CAUSALab @causalab.org · 09/03/2026

Interested in using health databases for #causalinference research? Target Trial Emulation (TTE) covers the target trial emulation framework in increasingly complex settings. 📆 June 8-12, 2026 Taught by Babra Dickerman, Joy Shi, @miguelhernan.org Apply now: hsph.harvard.edu/research/cau...

Target Trial Emulation (TTE). Dates: June 8-12, 2026. Taught by Barbra Dickerman, Joy Shi, Miguel Hernán

041

Miguel Hernan @miguelhernan.org · 13/01/2026

Having trouble with time zero when using healthcare databases to emulate a #TargetTrial? See our review of procedures to align eligibility and treatment assignment in observational emulations. We use 3 target trials of increasing complexity and provide a decision diagram www.bmj.com/content/392/...

lnkd.in

Starting right: aligning eligibility and treatment assignment at time zero when emulating a target trial

This article provides methodological guidance when emulating a target trial with longitudinal observational data by showing how to align eligibility criteria and treatment assignment at the start of f...

0191

Miguel Hernan @miguelhernan.org · 16/12/2025

New study: Small benefits and risks of COVID-19 vaccines in children in Madrid. Hospitalization risk very low in unvaccinated, lower in vaccinated. 6-11 years old: no myocarditis cases 12-17 years old: myocarditis risk very low in vaccinated, lower in unvaccinated journals.lww.com/pidj/fulltex...

0154

Reposted by Miguel Hernan

CAUSALab @causalab.org · 09/10/2025

You're invited! ✉️ 20th Kolokotrones Symposium: “Acetaminophen During Pregnancy and Autism: What Does Causal Inference Take?" Details in comments. In-person limited to Harvard ID holders due to space restrictions. Online attendance free & public. Register: www.eventbrite.com/e/acetaminop...

173

Reposted by Miguel Hernan

Daniela van Santen @dvansanten.bsky.social · 04/09/2025

Free Causal Inference Consulting Available at Harvard T.H. Chan School of Public Health! Take advantage of expert advice for your research projects. Learn more and help spread the word! :)

092

Miguel Hernan @miguelhernan.org · 03/09/2025

When using observational data for #causalinference, the choice isn’t between emulating or not emulating a #TargetTrial, but between reporting or not reporting the target trial that we are emulating. For those who prefer to be explicit about what they do, we have developed the TARGET Statement 👇

1237

Reposted by Miguel Hernan

CAUSALab @causalab.org · 08/07/2025

SER 2025 @societyforepi.bsky.social included a session spotlighting James M. Robins⭐ "Celebrating James M. Robins Contributions to Epidemiology" explored Robins' impact, including his landmark 1986 paper. It concluded with his comments on progress still to come in #causalinference research.

0111

Reposted by Miguel Hernan

CAUSALab @causalab.org · 22/05/2025

See you in Madrid? CAUSALab is partnering w/ @cemfi.es for the course, Causal Inference for Health and Social Scientists. 📆 Aug 25-29, 2025 Taught by @miguelhernan.org, CEMFI course introduces 2 step causal framework for experimental & non-experimental data. www.cemfi.es/programs/css...

CEMFI Summer School Course, "Causal Inference for Health and Social Scientists" (Aug 25-29, 2025)

074

Miguel Hernan @miguelhernan.org · 24/04/2025

If you're wondering about differences between publicly-funded research in non-profit universities and privately-funded research in for-profit companies, watch this: www.youtube.com/watch?v=Ar0z... The topic is the "de-extinction of the dire wolf", but the message applies beyond it. (Think "AI".)

youtube.com

They Didn't Make Dire Wolves, They Made Something…Else

YouTube video by hankschannel

060

Miguel Hernan @miguelhernan.org · 02/04/2025

Barbra Dickerman, @joy-shi.bsky.social, and I have a new online course for anyone who wants to learn the basics of confounding adjustment for time-fixed treatments. A must if you are considering CAUSALab's "Advanced Confounding Adjustment" course for time-varying treatments in the Summer.

290

Reposted by Miguel Hernan

Harvard Epidemiology @harvardepi.bsky.social · 26/02/2025

Join us on Wednesday, March 5th at 1:00pm EST for the Department's seminar series with Miguel Hernan speaking on "How to make people immortal and why it is not a good idea: Improving the causal analyses of healthcare databases" ➡️ Go to event page to register: hsph.harvard.edu/epidemiology...

0163

Miguel Hernan @miguelhernan.org · 18/02/2025

Roger: You’ve been ridiculing my posts for years. However, you've never written a paper that presents a thoughtful criticism of our work. Would you consider engaging in a scientific exchange? Also, a piece of advice: Stop embarrassing yourself and read our papers before posting about them. Prou.

050

Miguel Hernan @miguelhernan.org · 18/02/2025

2/ The #TargetTrial framework is a structured procedure to operationalize good practices for study design, data analysis, and reporting. It avoids design-induced biases but not biases arising from data limitations, such as measurement error and insufficient information to adjust for confounding.

0101

Miguel Hernan @miguelhernan.org · 18/02/2025

1/ When using observational data for #causalinference, emulating a target trial helps solve some problems... but not all problems. In a new paper, we explain why and when the #TargetTrial framework is helpful. www.acpjournals.org/doi/10.7326/... Joint work with my colleagues @causalab.bsky.social

34612

Miguel Hernan @miguelhernan.org · 03/02/2025

3. "Why use methods that require proportional hazards?" @amjepi.bsky.social 2025 doi.org/10.1093/aje/... The proportional hazards assumption is generally superfluous. We encourage the use of survival analysis methods that produce absolute risks and that don't require constant hazard ratios.

1153

Miguel Hernan @miguelhernan.org · 03/02/2025

2. "Why test for proportional hazards?" @jama.com 2020 jamanetwork.com/journals/jam... Several examples show that hazards aren't expected to be proportional because either the effect isn't constant or the selection bias isn't constant. An exception: null effect of treatment (hazard ratio=1) ...

130

Miguel Hernan @miguelhernan.org · 03/02/2025

1. "The hazards of hazard ratios" EPIDEMIOLOGY 2010 journals.lww.com/epidem/fullt... Hazard ratios have a built-in selection bias because of depletion of susceptibles. Also, reporting only hazard ratios is insufficient because we also need (adjusted) absolute risks for sound decision making. ...

120

Miguel Hernan @miguelhernan.org · 03/02/2025

In a recent commentary, Mats Stensrud and I argue that the proportional hazards assumption is not only implausible but also unnecessary. doi.org/10.1093/aje/... Easy-to-implement survival analysis methods that don't rely on proportional hazards are typically preferred. The argument in 3 steps 👇

160

Miguel Hernan @miguelhernan.org · 03/02/2025

1/ If you were taught to test for proportional hazards, talk to your teacher. The proportional hazards assumption is implausible in most #randomized and #observational studies because the hazard ratios aren't expected to be constant during the follow-up. So "testing" is futile. But there is more 👇

47819

Miguel Hernan @miguelhernan.org · 06/01/2025

2/ Immortal time may occur when individuals 1) are assigned to treatment strategies based on post-eligibility information or 2) determined to be eligible based on post-assignment information. #TargetTrial emulation prevents it by synchronizing eligibility and assignment at the start of follow-up.

060

Miguel Hernan @miguelhernan.org · 06/01/2025

1/ That "immortal time" is so frequent in survival analyses for #causalinference is fascinating. Because "immortal time" doesn't exist in the data, *we* create it when misanalyzing the data. Our new paper pubmed.ncbi.nlm.nih.gov/39494894/ summarizes why immortal time arises & how to prevent it.

35118

Miguel Hernan @miguelhernan.org · 23/12/2024

Upgrade your #causalinference arsenal. A revision of our book "Causal Inference: What If" is available at miguelhernan.org/whatifbook Thanks to everyone who suggested improvements, reported typos, and proposed new citations and material. Enjoy the #WhatIfBook plus code and data. Also, it's free.

10374113

Miguel Hernan @miguelhernan.org · 26/11/2024

Agree. Stephen Senn's "Seven myths of randomisation in clinical trials" pubmed.ncbi.nlm.nih.gov/23255195/ is a good place to start. And the work by Jamie Robins and colleagues helped us understand "the curse of dimensionality" in high-dimensional settings (references in Chapter 10 of "What If").

pubmed.ncbi.nlm.nih.gov

Seven myths of randomisation in clinical trials - PubMed

I consider seven misunderstandings that may be encountered about the nature, purpose and properties of randomisation in clinical trials. Some concern the practical realities of clinical research on pa...

0265

Miguel Hernan @miguelhernan.org · 26/11/2024

In Chapter 10 of "Causal Inference: What If", we describe arguments for adjustment in randomized trials and refute some fallacies used to advise against adjustment. www.hsph.harvard.edu/miguel-herna... A practical challenge is how to incorporate adjustment into the design of #randomizedtrials.

2351

Miguel Hernan @miguelhernan.org · 26/11/2024

When risk factors are imbalanced for non-chance reasons in #observational studies, we call it #confounding. An interesting point is that, regardless of whether the imbalance results from chance or confounding, we are better off ADJUSTING for prognostic factors that are imbalanced between groups.

1321

Miguel Hernan @miguelhernan.org · 26/11/2024

Unsurprising. By definition, the 95% confidence interval of 5% of (perfect) trials isn't expected to include the true value of the effect. Again: Of 20 randomized trials in which treatment truly has a null effect, the 95% CI of one of them isn't expected to include the null value. Just by chance.

2171

Miguel Hernan @miguelhernan.org · 26/11/2024

Vass M (PhD Thesis). Prevention of functional decline in older people. Faculty of Health Sciences, U of Copenhagen 2010, p.120. (Thanks to Mikkel Zöllner Ankarfeldt for bringing this example to my attention.) What happened? By chance, some risk factors were more common in the intervention group.

1130

Miguel Hernan @miguelhernan.org · 26/11/2024

Does #randomization ensures balance of risk factors between groups? Consider this: In Denmark 860 individuals were randomly allocated to either intervention or control. Individuals were unaware of their allocation. No intervention took place. Mortality was higher in the intervention group (p=0.003)

712347

Miguel Hernan @miguelhernan.org · 20/11/2024

It's always a good time to remember Brandolini's principle @ziobrando.bsky.social

Alberto Brandolini presents a slide with the text "The amount of energy necessary to refute bullshit is an order of magnitude bigger than to produce it"

47117

Reposted by Miguel Hernan

CAUSALab @causalab.org · 02/10/2024

Congrats to Roger Logan on his retirement! Roger has worked as a CAUSALab Senior Research Scientist @harvardchanschool.bsky.social for 23 years. He has been a valuable team member & made major contributions to #causalinference research. Wishing Roger all the best in this new chapter! #publichealth

Roger Logan, CAUSALab Senior Research Scientist, retirement collage

141

Miguel Hernan @miguelhernan.org · 02/10/2024

I’m so thankful that I could work side-by-side with Roger Logan for over two decades. He was a cornerstone of CAUSALab. I learned a lot from him. Roger’s expertise and generosity will be very much missed.

020

Reposted by Miguel Hernan

Peter Tennant @pwgtennant.bsky.social · 27/09/2024

Really enjoyed @miguelhernan.bsky.social's talk on the promises and limitations of AI for health data research at the #WCE2024! A fair and critical dose of reality that the wider health research sphere desparately needs to hear! #EpiSky

Photo of Miguel Hernan in front of a slide explaining the three tasks of data science

1221

Reposted by Miguel Hernan

CAUSALab @causalab.org · 24/06/2024

New research published in Annals of Internal Medicine challenges past scholarship on metformin. CAUSALab collaborator Yu-Han Chiu identified no increased risk for childbirth with major birth defects when compared w/ women who discontinued the drug. CNN article: www.cnn.com/2024/06/17/h...

121

Miguel Hernan @miguelhernan.org · 23/06/2024

This week I discussed methods for health technology assessment at HTAi. My main point: "Observational data (#RWD) can often be used to emulate a #TargetTrial, but we need more research to characterize questions that can only be answered by randomized trials." Let's learn the limits of #RWE.

060

Miguel Hernan @miguelhernan.org · 11/06/2024

Did you know that the LATE estimator was independently described in 1994 by Imbens & Angrist in Econometrica and Baker & Lindeman in Statistics in Medicine? onlinelibrary.wiley.com/doi/10.1002/... A delightful historical overview of LATE is now available www.tandfonline.com/doi/full/10....

0185

Miguel Hernan @miguelhernan.org · 18/05/2024

With #TargetTrial emulation becoming increasingly popular, it's important to understand what it can and can't do. In this podcast I discuss how target trial emulation can improve causal inference from observational data and extend inferences from randomized trials edhub.ama-assn.org/jn-learning/...

edhub.ama-assn.org

Target Trial Emulation for Causal Inference From Observational Data With Dr Hernán

Miguel A. Hernán, MD, DrPH, professor of epidemiology, Harvard T.H. Chan School of Public Health, discusses Target Trial Emulation: A Framework for Causal Inference From Observational Data with JAMA Statistical Editor Roger J. Lewis, MD, PhD.

0119

Reposted by Miguel Hernan

CAUSALab @causalab.org · 16/04/2024

First FEP-CAUSAL collab paper is out in AJE! pubmed.ncbi.nlm.nih.gov/38576166/ Target trial emulation findings support aripiprazole & paliperidone as first-line therapy in first episode psychosis treatment. Led by CAUSALab researchers Alejandro Szmulewicz & @miguelhernan.bsky.social.

111

Reposted by Miguel Hernan

CAUSALab @causalab.org · 20/03/2024

CAUSALab has a new collaboration with Broad Institute of MIT & Harvard! @miguelhernan.bsky.social received a Multidisciplinary University Research Initiative (MURI) award as a project collaborator to advance intervention design decision-making via computational framework.

defense.gov

Department of Defense Announces Fiscal Year 2024 University Research Funding Awards

DOD announced $221 million in awards for basic defense-related research projects as part of the Multidisciplinary University Research Initiative program.

021

Miguel Hernan @miguelhernan.org · 14/02/2024

jamanetwork.com/journals/jam... www.neurology.org/doi/10.1212/... www.sciencedirect.com/science/arti... ajph.aphapublications.org/doi/10.2105/... link.springer.com/article/10.1... More recently www.thelancet.com/journals/lan... www.bmj.com/content/371/... ajph.aphapublications.org/doi/10.2105/...

jamanetwork.com

Childhood Malnutrition and Postwar Reconstruction in Rural El Salvador

ContextThe 1992 peace settlement that ended the civil war in El Salvador included land redistribution and other provisions designed to improve the socioeconom

021

Miguel Hernan @miguelhernan.org · 14/02/2024

First, you're giving me too much credit. Second, I doubt descriptive studies are hard to publish. My colleagues and I have been publishing field work, surveys, etc. for >25 years. Doing this work I learned that good description (measurement) is the foundation of sound causal inference. Examples 👇

131

Miguel Hernan @miguelhernan.org · 12/02/2024

Don't let the causal inference buzz fool you: Description is the foundation of science. We've described the 3-year health impact of #COVID19 in Madrid, the EU region with the highest life expectancy. If we can't describe, no causal inference can follow. academic.oup.com/ofid/article...

academic.oup.com

Three Years of the Coronavirus Disease 2019 Pandemic in a European Region: A Population-Based Longitudinal Assessment in Madrid Between 2020 and 2022

Using population-based data from several linked administrative and clinical databases, we characterized the health burden of COVID-19 longitudinally in the near

03611

Miguel Hernan @miguelhernan.org · 05/01/2024

Have you ever been advised to state your hypothesis in your grant application? Sander Greenland and I argue that stating hypotheses is unnecessary. lnkd.in/eW4ccgae Who cares what you guess the qualitative answer is before doing the study? Just do the study. Your hypothesis is nobody's business.

0135

Miguel Hernan @miguelhernan.org · 22/11/2023

Hmm. I do. Causal DAGs help me identify and illustrate biases in the design and analysis of health studies. Having said that, I have supersmart colleagues that can reach the right conclusion without DAGs (congrats if you are one of them). For the rest of us mortals, causal DAGs are helpful tools.

130