COVID-19 Data Drift

Cross-source consensus on COVID-19 Data Drift from 1 sources and 4 claims.

1 sources · 4 claims

How it works

COVID-19-era structural changes altered administrative data patterns used by opioid prognostic models. — Impact of COVID-19-related data drift on machine-learning prognostic models predicting 30-day opioid-related emergency department visits, hospitalisation or mortality: a population-level administrative data study in Alberta, Canada
The study attributed drift primarily to healthcare system disruptions rather than changes in the population receiving opioids. — Impact of COVID-19-related data drift on machine-learning prognostic models predicting 30-day opioid-related emergency department visits, hospitalisation or mortality: a population-level administrative data study in Alberta, Canada
Feature importance shifted during the pandemic, with opioid dose becoming more important than age for the pre-pandemic model on pandemic data. — Impact of COVID-19-related data drift on machine-learning prognostic models predicting 30-day opioid-related emergency department visits, hospitalisation or mortality: a population-level administrative data study in Alberta, Canada
Unmeasured pandemic-related factors compounded measured drift in prescription-based models. — Impact of COVID-19-related data drift on machine-learning prognostic models predicting 30-day opioid-related emergency department visits, hospitalisation or mortality: a population-level administrative data study in Alberta, Canada