Health and Economic impacts of an Early Labor Induction Policy for High-BMI Mothers (2026) with Maria K. Gregersen
Health Economics
We expand the literature on marginal returns to birth interventions by studying a common intervention: early induction of labor for a growing share of pregnancies: high-BMI women. We exploit Danish guidelines which recommend routine induction at 7 days after the expected due date instead of 10-13 days after for mothers with a pre-pregnancy BMI of at least 35. Our results show that early labor induction improves immediate maternal and neonatal health, reduces universal nurse visits during the child’s first year of life, as well as maternal postpartum depression risks (suggestive).
Digital Microsteps as Scalable Adjuncts for Adults Using GLP-1 Receptor Agonists: A Randomized Clinical Trial (2026)
with Maya Adam, Till Bärnighausen, Fatima Rodriguez, Doron Amsalem, Eleni Linos
JAMA Network Open
Among adults using glucagon-like peptide-1 receptor agonists (GLP-1 RAs), can a brief digital intervention, consisting of written behavior change prompts accompanied by short videos, boost expectations to adopt health lifestyle behaviors? In this randomized clinical trial of 5054 global adults using GLP-1 RAs, single exposure to behavior change prompts with either storytelling or didactic video boosters significantly improved behavioral expectation to adopt health behaviors immediately after exposure, with effects still evident 2 weeks later. The storytelling video showed stronger effects across most domains.
Contribution: Co-first author
AI-assisted teams outperform AI-led teams but not human-only teams in assessing research reproducibility in quantitative social science with Abel Brodeur et al. (2026)
Proceedings of the National Academy of Science
Large Language Models (LLMs) such as ChatGPT are transforming how scientists conduct and validate research, offering promise as tools to improve scientific reproducibility. However, computational reproducibility and error detection remain expensive and labor-intensive. We experimentally test how collaboration between researchers and LLM assistants influences the reproduction of quantitative social science findings across different levels of AI autonomy. We randomly assigned 288 researchers to 103 teams working under three conditions: human-only, AI-assisted (using ChatGPT as a collaborative tool), or AI-led (ChatGPT operating with minimal human oversight). Teams reproduced published results from leading social science journals, detected coding errors, and proposed robustness checks. Human-only and AI-assisted teams achieved comparable reproduction rates (94% vs. 91%) and performed similarly on most outcomes, except human-only teams identified significantly more major coding errors. Both substantially outperformed AI-led teams, which achieved only a 37% reproduction rate, detected fewer errors across all categories, proposed weaker robustness checks, and required more time. This autonomous approach, however, likely represents only a lower bound of AI capabilities. Despite rapid model advances, expert human judgment currently remains indispensable for reliable empirical verification. While AI assistance did not degrade most outcomes, it provided no measurable advantages and was associated with reduced detection of major errors. However, the 37% autonomous reproduction rate indicates that AI could provide value in settings where scale or cost constraints preclude human review of papers, even though general-purpose LLMs offer no immediate advantages for human-supervised verification.
Contribution: minor (data collection, limited writing)