With the epidemic shifting over time it is often difficult to properly allocate resources to areas until it is already too late 5, 6. Traditional methods to capture community characteristics, focused on economic, broad healthcare, and survey outcomes, only capture a fraction of what matters during the everyday lives of community members, especially as communities change from year to year 4. There is suspected to be large heterogeneity among risk factors per community across the US 3. One of the key challenges to the epidemic is that its underlying driving force seemingly changes across time and communities, for example, from prescription drug abuse to cheap and readily available synthetic opioids (e.g.
The United States has been attempting to tackle an opioid epidemic for over two decades, with age-adjusted opioid-related deaths increasing by 350% over 20 years from 1999 to 2020 1. A model built using linear auto-regression and traditional socioeconomic data gave 7% error (MAPE) or within 2.93 deaths per 100,000 people on average our proposed architecture was able to forecast yearly death rates with less than half that error: 3% MAPE and within 1.15 per 100,000 people. Trained over five years and evaluated over the next two years T rOP demonstrated state-of-the-art accuracy in predicting future county-specific opioid trends. TOP builds on recent advances in sequence modeling, namely transformer networks, to use changes in yearly language on Twitter and past mortality to project the following year’s mortality rates by county. Here, we develop and evaluate, T rOP ( Transformer for Opiod Prediction), a model for community-specific trend projection that uses community-specific social media language along with past opioid-related mortality data to predict future changes in opioid-related deaths. AI-based language analyses, having recently shown promise in cross-sectional (between-community) well-being assessments, may offer a way to more accurately longitudinally predict community-level overdose mortality. opioid epidemic is difficult due to our inability to accurately predict changes in opioid mortality across heterogeneous communities. Targeting of location-specific aid for the U.S.
0 kommentar(er)
