started looking at $8.6 billion in inpatient healt...
# show-and-tell
j
started looking at $8.6 billion in inpatient health care costs for 1.2 million people yesterday: https://onefact.github.io/synthetic-healthcare-data/ only possible thanks to observable framework -- thank you! very smooth workflow to go from dbt / duckdb sql models to these visualizations. will be looking at LLMs on this data next to understand algorithmic fairness metrics and health equity, linked to census data (https://jaanli.github.io/american-community-survey/new-york-area/income-by-race)
🎉 5
❤️ 2
c
Hey, this looks very cool and I just cloned this repo today to start digging in to the details. I’m still getting more familiar with observable framework, but I’m in the healthcare world professionally and love to play with duckdb & dbt projects when I can. Can I confirm that I’ll need to submit the data use agreements to AHRQ in order to get access to the csv files you’re referencing?
j
Yes exactly! Let me know if you need a hand with it chris, feel free to email me if easier :) and appreciate the kind words! Our next step for this work is adding LLM prediction to assess publicly available health care AI (like what I built in grad school: https://arxiv.org/abs/1904.05342 and have since helped hospitals deploy) for compliance with the federal affordable care act non-discrimination clauses (https://www.hhs.gov/civil-rights/for-individuals/section-1557/index.html).
🙌 1