arXiv subjects

Marie-Laure Charpignon

Publications and source records attributed to Marie-Laure Charpignon.

6 recordsLinked to original sources

arXiv2026

Socio-Spatial Patterns of Suicide Mortality in the United States

Suicides cause over 49000 deaths yearly in the United States, 55% involving firearms. Suicide mortality exhibits substantial geographical and sociodemographic heterogeneity; yet the role of social networks remains underexplored. To assess how suicide risk and firearm restriction policies propagate through social ties, we integrate county-level suicide mortality data (2010-2022) with the Facebook Social Connectedness Index (SCI). We also examine Extreme Risk Protection Orders (ERPO), state-level policies restricting firearm access for individuals at risk of self-harm. In two-way fixed effects regressions, a one-standard-deviation increase in the SCI-weighted average suicide mortality rate of connected counties was associated with +2.78 deaths per 100,000 in a focal county, while a one-standard-deviation increase in ERPO social exposure was associated with -0.214 deaths per 100,000. These associations persisted when adjusting for geographic proximity and including state-by-year fixed effects, and confirm the effect of social networks on diffusion of both harmful exposures and protective interventions.

stat.AP↗

arXiv2025

Statistical vs. Deep Learning Models for Estimating Substance Overdose Excess Mortality in the US

Substance overdose mortality in the United States claimed over 80,000 lives in 2023, with the COVID-19 pandemic exacerbating existing trends through healthcare disruptions and behavioral changes. Estimating excess mortality, defined as deaths beyond expected levels based on pre-pandemic patterns, is essential for understanding pandemic impacts and informing intervention strategies. However, traditional statistical methods like SARIMA assume linearity, stationarity, and fixed seasonality, which may not hold under structural disruptions. We present a systematic comparison of SARIMA against three deep learning (DL) architectures (LSTM, Seq2Seq, and Transformer) for counterfactual mortality estimation using national CDC data (2015-2019 for training/validation, 2020-2023 for projection). We contribute empirical evidence that LSTM achieves superior point estimation (17.08% MAPE vs. 23.88% for SARIMA) and better-calibrated uncertainty (68.8% vs. 47.9% prediction interval coverage) when projecting under regime change. We also demonstrate that attention-based models (Seq2Seq, Transformer) underperform due to overfitting to historical means rather than capturing emergent trends. Ourreproducible pipeline incorporates conformal prediction intervals and convergence analysis across 60+ trials per configuration, and we provide an open-source framework deployable with 15 state health departments. Our findings establish that carefully validated DL models can provide more reliable counterfactual estimates than traditional methods for public health planning, while highlighting the need for calibration techniques when deploying neural forecasting in high-stakes domains.

cs.LG↗

arXiv2025

Reflections from Research Roundtables at the Conference on Health, Inference, and Learning (CHIL) 2025

The 6th Annual Conference on Health, Inference, and Learning (CHIL 2025), hosted by the Association for Health Learning and Inference (AHLI), was held in person on June 25-27, 2025, at the University of California, Berkeley, in Berkeley, California, USA. As part of this year's program, we hosted Research Roundtables to catalyze collaborative, small-group dialogue around critical, timely topics at the intersection of machine learning and healthcare. Each roundtable was moderated by a team of senior and junior chairs who fostered open exchange, intellectual curiosity, and inclusive engagement. The sessions emphasized rigorous discussion of key challenges, exploration of emerging opportunities, and collective ideation toward actionable directions in the field. In total, eight roundtables were held by 19 roundtable chairs on topics of "Explainability, Interpretability, and Transparency," "Uncertainty, Bias, and Fairness," "Causality," "Domain Adaptation," "Foundation Models," "Learning from Small Medical Data," "Multimodal Methods," and "Scalable, Translational Healthcare Solutions."

cs.LG↗

arXiv2025

Investigating Primary Care Indications to Improve Electronic Health Record in Dementia Target Trial Emulation

Missing data, inaccuracies in medication lists, and recording delays in electronic health records (EHR) are major limitations for target trial emulation (TTE), the process by which EHR data are used to retrospectively emulate a randomized control trial. EHR TTE relies on recorded data that proxy true drug exposures and outcomes. We investigate the under-utilized criterion that a patient has indications of primary care provider (PCP) encounters within the EHR. Such patients tend to have more records overall and a greater proportion of the types of encounters that materialize comprehensive and up-to-date records. We examine the impact of including a PCP feature in the TTE model or as an eligibility criterion for cohort selection, contrasted with ignoring it altogether. To that end, we compare the estimated effects of two first line antidiabetic drug classes on the onset of Alzheimer's Disease and Related Dementias (ADRD). We find that the estimated treatment effect is sensitive to the consideration of a PCP feature, particularly when used as an eligibility criterion. Our work suggests that this PCP feature should be further researched.

stat.AP↗

arXiv2024

What we should learn from pandemic publishing

Authors of COVID-19 papers produced during the pandemic were overwhelmingly not subject matter experts. Such a massive inflow of scholars from different expertise areas is both an asset and a potential problem. Domain-informed scientific collaboration is the key to preparing for future crises.

physics.soc-ph↗

arXiv2023

Study of the effectiveness of incentive measures on Covid-19 vaccination in the United States of America

With COVID-19 having emerged as the most widespread human pandemic disease in a century, the need to control its spread to avoid massive loss of life became more than necessary, and extremely fast. Several vaccines were developed and the task of policy makers was suddenly to convince the reluctant population to be vaccinated by various means. While some countries have chosen a policy of mandatory vaccination or punitive incentives, many states in the United States have adopted various incentives to try to increase vaccination coverage. A study we conducted in recent months quantified the effect of these measures on the proportion of the population vaccinated, using the synthetic control method, by simulating what would have happened without these measures. The aim now is to generalize this study to smaller scales, to improve the results of our previous study, to quantify their robustness and to provide a tool that can be used by policy makers to adapt their behavior in light of the results obtained.

stat.AP↗