Researcher & Data Scientist

Nathaniel Hendrix

I'm a researcher and data scientist with the American Board of Family Medicine and their Center for Professionalism and Value in Health Care. My work focuses on natural language processing of clinical notes, epidemiology, and artificial intelligence for decision support.

CV (PDF) · · Substack · Bluesky

Questions that motivate me

  • How can we reduce the heterogeneity of diagnoses by different clinicians?
  • What would a healthcare system designed around AI look like?
  • How can we model the diffusion of novel practice patterns among clinicians?
  • How can we accelerate the integration of new scientific knowledge into clinical decisions?
  • To what extent can we automate the process of learning from observational data?
  • How would our world change if we substantially lengthened our lifespans?
Open to consulting Cost-effectiveness analyses, observational studies, NLP in EHRs, making sense of weird unruly research areas. Rates dependent on how fun it is and who you are.

Recent activity

2025
Sep 30
Our paper on documentation of compounded GLP-1s in primary care has been published by Pharmacoepidemiology and Drug Safety. Self-archived version here.
Sep 28
Posted a preprint about how the COVID pandemic disrupted vaccination in primary care. This decentralization can make it hard to track vaccination rates and preventive care needs.
Aug 10
American Journal of Epidemiology published a paper I led, with coauthors from AAFP, ABFM, and Stanford, on how unreliable ICD-10 codes are for identifying COVID-19 patients and how NLP provides a way to improve cohort identification.
Jul 18
I'm on a new paper in the Journal of Primary Care and Community Health, led by Esther Velásquez at Stanford, on socioeconomic gradients of COVID antiviral access within race/ethnicity groups.
Jun 27
JABFM published a paper I'm on (led by Jeongyoung Park, AAFP / Robert Graham Center) on documentation of social determinants of health in primary care.
May 19
PLOS One published a paper of mine (with ABFM, Stanford, and CDC researchers) on primary care providers' variation in willingness to diagnose long COVID.
May 13
Posted a preprint estimating how many GLP-1 users in primary care use compounded versions of these drugs.
2024
Nov 5
New paper with collaborators at UCSF on how team efficiency and composition affects EHR-related burnout.
Oct 31
Published a policy brief on how few family physicians love their EHRs.
Sep 25
Accepted as a fellow in the third year of the NIH's AIM-AHEAD initiative, focused on AI and health equity.
Sep 11
First NIH award — a 2-year R03 from AHRQ on characterizing patterns of practice change in primary care following guideline updates.
Sep 10
New paper led by A Jay Holmgren, with UCSF and ONC, on EHR usability and burnout among family physicians.
Jun 17
Paper in JAMIA on how medical specialty boards can contribute to federal data collection on EHR policy.
Mar 26
Paper on physician satisfaction with EHRs and interoperability in JAMA Network Open.
Feb 8
Preprint: Heterogeneity of Diagnosis and Documentation of Post-COVID Conditions in Primary Care.
Jan 4
Preprint: Comparative Analysis of Three Surveys on Primary Care Providers' Experiences with Interoperability and EHRs.
2023
Dec 8
Preprint: Blueprinting the Future: Automatic Item Categorization using Hierarchical Zero-Shot and Few-Shot Classifiers. Led by Ting Wang.
Oct 31
Presented a workshop on NLP for clinical researchers at NAPCRG 2023, with linked Google Colab code.
Aug 7
Preprint with NCATS: Influence of Prior SARS-CoV-2 Infection on COVID-19 Severity.
Jun 14
How Do Family Physicians Document Patients' Social Needs in EHRs? published in JABFM.
Apr 7
NASEM posted video of a talk I gave on AI in medical education.
Mar 23
Variation in Family Physicians' Experiences Across Different EHR Platforms published in JGIM.