iiinova — where your digital self keeps evolving

Reposted by Dan Zeltzer

Journal of Public Economics @jpube.bsky.social · 12/08/2025

Just published in @jpube.bsky.social: "Impacts of home-care subsidies: Evidence from quasi-random assignment" By Yuval Ofek-Shanny, @avnershlain.bsky.social, & @dzeltzer.bsky.social www.sciencedirect.com/science/arti... #econsky #publiceconomics

162

Dan Zeltzer @dzeltzer.bsky.social · 14/04/2025

I'm happy to share that I will be joining Berkeley School of Public Health as an associate professor this summer. Grateful for the opportunity and sad to leave the Berglas School of Economics at Tel Aviv University, my home for years as both student and faculty.

350

Reposted by Dan Zeltzer

Eric Topol @erictopol.bsky.social · 04/04/2025

When physicians with access to A.I. are compared with A.I. for virtual urgent care visits The A.I. performed better @annalsofim.bsky.social www.acpjournals.org/doi/10.7326/...

209615

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

This work is with a team of medical and technology experts I am lucky to collaborate with: Zehavi Kugler, Lior Hayat, Tamar Brufman, Ran Ilan Ber, Keren Leibovich, Tom Beer, Ilan Frank, Ran Shaul (KHealth) Caroline Goldzweig, and Joshua Pevnick (Cedars-Sinai Medical Center).

000

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

Natural follow-up questions: causal impacts of AI on decision quality, speed, and outcomes. Best ways to combine human and algorithmic expertise. Trust, over/under reliance, continued learning. Lots of more work to do!

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

Physicians had to scroll to see recommendations, and we don't know if they did. We didn't study AI impacts. Still, results show well-calibrated models can do very well - more than the average doctor - on common symptoms that account for the bulk of virtual urgent care cases.

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

AI excelled at guideline adherence (e.g., no antibiotics for viral URI) & incorporating EHR data (e.g., noticing recurrent UTI history that physicians missed). Physicians did better when patient narratives evolved or when visual findings were important.

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

Overall scores were equal in 68% of cases, better for AI in 21%, and better for physicians in 11% (these are actual physician decisions in urgent care cases with common symptoms - not medschool exams or vignettes)

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

AI recommendations were rated optimal in 77% of cases vs 67% for physicians, and potentially harmful in 2.8% vs 4.6% for physicians.

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

Experienced physician adjudicators viewed full case info and independently rated AI recommendations & physician decisions on a 4-point scale. If two independent adjudicators disagreed, a third was brought in, but we didn't force consensus and weighted final ratings equally.

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

We sampled 461 adult visits from one month in 2024 with respiratory, urinary, vaginal, eye, & dental symptoms - which in previous work showed had high AI-clinician diagnostic concordance; symptoms represent 2/3 of clinic volume. www.mcpdigitalhealth.org/article/S294...

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

The AI, commercially developed by KHealth, is an ensemble of discriminative models trained and tested on clinical EHR + med guidelines. To be reliable, the model withholds recommendations when confidence is low (about 1/5 of cases).

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

The setting is CS-Connect, a telemedicine clinic where AI first automates patient intake via structured chat. AI predicts the differential diagnoses and recommends Rx, labs & referrals. Physicians then conduct a video visit.

100

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

New @annalsofim.bsky.social paper: In 461 @cedarssinai.bsky.social virtual urgent care cases with common symptoms, AI recommendations were rated optimal more often than physicians' (77% vs 67%), & less often potentially harmful (2.8% vs 4.6%).

120

Dan Zeltzer @dzeltzer.bsky.social · 05/04/2025

Agree! While the focus was AI accuracy, this study is from a clinic where (calibrated, discriminative, ensemble) models already support physician work by expediting intake and providing (when confident) diagnostic predictions and care recommendations. And we plan to study collaboration impacts next!

030

Reposted by Dan Zeltzer

Avner Strulov-Shlain 🎗️ @avnershlain.bsky.social · 21/03/2025

⛽ For an academic study, we (@akraft.bsky.social and I) wish to interview US Gas Station owners to learn more about their business practices. Please share or send anyone you know to Andreas or me. Thanks!

051

Reposted by Dan Zeltzer

Todd Jones @toddrjones.bsky.social · 21/03/2025

Mississippi State Univ. is hiring a tenure-track Econ Assistant Prof w/ August start! Open field; pref. for applied micro & teaching PhD Micro I & II (Micro II this fall). Apply: MSU site & www.aeaweb.org/joe/listing.php?JOE_ID=2025-01_111475768 Please repost—it's an off-cycle search. Thanks!

04842

Dan Zeltzer @dzeltzer.bsky.social · 16/03/2025

AGI is coming in 2 years max

000

Dan Zeltzer @dzeltzer.bsky.social · 27/02/2025

Looks like MC>MB from just scaling... open.substack.com/pub/garymarc...

010

Dan Zeltzer @dzeltzer.bsky.social · 27/02/2025

ishaniganguli.bsky.social @davidmcutler.bsky.social Ateev Mehrotra et al.

000

Dan Zeltzer @dzeltzer.bsky.social · 27/02/2025

New DiD paper on US telemed adoption finds higher adoption assoc. w/ slightly more visits but lower spending (incl low value). Mirrors our findings from Israel, suggesting robustness across payment and delivery systems! US: doi.org/10.1001/jamainternmed.2024.8354 ISR: doi.org/10.1093/jeea/jvad035

120

Reposted by Dan Zeltzer

Senator Patty Murray @murray.senate.gov · 25/02/2025

Raphael is a disabled Army veteran who worked at the VA in Seattle and played a critical role in processing servicemembers' disability benefits. He's received nothing but stellar performance reviews. Elon Musk fired him with ZERO justification. Help share his story.

37699399

Reposted by Dan Zeltzer

Institute for Replication @i4replication.bsky.social · 24/02/2025

After being alerted about possible misconduct, the I4R are reproducing published papers that use data from a specific NGO (GDRI). This thread releases the first 2 reports and provides more information about the work and responses/statements from authors journals and journals. 🧵

313759

Reposted by Dan Zeltzer

Ben Collins @bencollins.bsky.social · 03/02/2025

Brief break from hawking The Onion to tell everyone I just subscribed to Wired. Please support publications who do brave work right now. www.wired.com/v2/offers/wi...

wired.com

WIRED - The Latest in Technology, Science, Culture and Business | WIRED

Unlimited Access| Best-in-class reporting that's too important to ignore - plus gear reviews, podcasts, and more. | Plus, FREE Stickers!

262105351618

Reposted by Dan Zeltzer

Ben Golub @bengolub.bsky.social · 17/01/2025

PLEASE RT :) Submit to the annual Network Science and Economics Conference, the largest US event for network theory and empirical research in and adjacent to economics Deadline Jan 30 (soon!) Conference Apr 11-13 at Stanford

48076

Dan Zeltzer @dzeltzer.bsky.social · 16/01/2025

Sad about David Lynch. Beyond his famously great stuff, I remember really enjoying "The Intervew Project", an experimental documentary where his team traveled for weeks all over America and filmed random people they met along the way.

070

Reposted by Dan Zeltzer

Joshua Goodman @joshua-goodman.com · 14/01/2025

For those who didn't see Janet Currie's AEA Presidential Address live, here's the recording of her talk titled: "Investing in Children to Address the Youth Mental Health Crisis" www.aeaweb.org/webcasts/202...

aeaweb.org

American Economic Association: AEA Presidential Address

2277

Dan Zeltzer @dzeltzer.bsky.social · 11/01/2025

Mostly raw (btw asking it to comment code and make it nice is also something I do, separately)

030

Dan Zeltzer @dzeltzer.bsky.social · 11/01/2025

Gave LLM my research project code and a finished paper as a template, and asked it to draft the sample construction section based on the code. Worked really well! (A bit too detailed but I feel deleting irrelevant stuff is much easier).

120

Reposted by Dan Zeltzer

Pedro Sant’Anna @pedrosantanna.bsky.social · 03/01/2025

I've decided to collect my DiD materials in a single place. psantanna.com/did-resources There, you will find - 14 lectures of my comprehensive DiD course - Shorter lectures/talks I have given on DiD - My DiD R/Stata/Python packages - Some DiD checklists - DiD materials from my friends Enjoy!

psantanna.com

Pedro H. C. Sant’Anna

23461145

Dan Zeltzer @dzeltzer.bsky.social · 21/12/2024

This progress on generalization seems like a big deal!

150

Dan Zeltzer @dzeltzer.bsky.social · 08/12/2024

My rules of thumb are keeping light exposure, food and drink, and physical activity aligned with destination time. Was it helpful in implementing just that or beyond?

140

Reposted by Dan Zeltzer

Alex Imas @aleximas.bsky.social · 01/12/2024

Why do I think these instances are interesting? To me, these are not just random instances that LLMs are sometimes wrong (like people), these are *diagnostic* that LLMs do not have a world model and are not “reasoning”, it exposes the basic architecture. Why? 1/n

510833

Reposted by Dan Zeltzer

Ethan Mollick @emollick.bsky.social · 23/11/2024

This may sound odd, but game-based benchmarks are some of the most useful for AI, since we have human scores and they require reasoning, planning & vision The hardest of all is Nethack. No AI is close, and I suspect that an AI that can fairly win/ascend would need to be AGI-ish. Paper: balrogai.com

718520

Dan Zeltzer @dzeltzer.bsky.social · 06/05/2024

I will continue in my visitor position at Stanford Econ for (most of) the next academic year. Very grateful for this incredible opportunity!

0110

Reposted by Dan Zeltzer

Avner Strulov-Shlain 🎗️ @avnershlain.bsky.social · 25/10/2023

🌼New WP 📉📈 How does receiving subsidized in-home care affects the health of sick elders? How does it affect the labor supply and income of their adult children? 🧵 with @dzeltzer.bsky.social and Yuval Ofek-Shanny papers.ssrn.com/sol3/papers....

Impacts of Home-Care Subsidies: Evidence from Quasi-Random Assignment
We study the impact of subsidizing home-based long-term care on recipients’ health and the labor supply of their working-age children. We use administrative data from Israel on the universe of welfare benefit applications linked with tax records of applicants and their adult children. To address the endogeneity of benefit recipients’ health status, we instrument for benefit receipt using the leniency of randomly assigned evaluators who assess the applicant’s functional status and determine benefit eligibility. We find that for compliers — applicants who receive subsidies only from more lenient evaluators — subsidizing home-based care has large adverse effects on recipient health but no detectable effects on the labor market outcomes of their children. The results are consistent with the crowd-out of self-care for the marginal recipient,
highlighting the need to assess the heterogeneous effects of home-care subsidies

1166