Tech Talk
summary
Evaluates Oakland’s Shoo the Flu program that vaccinates up to 10,000 elementary students per year and links surveys (6,000+ students), 31 million student-days of absences, community hospitalization records (~3 million people) and Kaiser’s 57 million person-weeks using a difference-in-differences design. Builds a reproducible pipeline with GitHub, Docker and a HIPAA-compliant AWS cloud, adds unit tests and an internal replication step with two independent analysts (0.001 tolerance) and publishes code/data on Open Science Framework.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:07Shu the Flu
- 5:02reproducibility crisis
- 5:42cognitive fallacies
- 7:47traditional scientific process
- 10:04reproducible workflows
- 11:38team
- 13:04data
- 14:28HIPAA compliance
- 15:57Github repository
- 16:20Internal replication process
- 19:07Open Science Framework
- 19:50Teaching
- 21:40Grade Scope
- 21:53Questions Comments
related talks
15mFarbod Abolhassani, University of Toronto | KubeCon + CloudNativeCon Europe 2020 – Virtual
19mData science education as a public health intervention in E. Baltimore
33mCOVIDcast: An ecosystem for COVID-19 tracking and forecasting
45mGoogle Health: Connecting Mobile Patients
41mUsing Data Riches: A Tale of two projects
24mR Consortium initiatives in medicine
01h 15mSunday Keynotes
24mUsing R Bridge the Gap in Electronic Medical Record Reporting