Development of a Machine Learning Model Using Multiple, Heterogeneous Data Sources to Estimate Weekly US Suicide Fatalities
- 1 December 2020
- journal article
- research article
- Published by American Medical Association (AMA) in JAMA Network Open
- Vol. 3 (12) , e2030932
Abstract
Suicide is a significant contributor to global mortality, resulting in nearly 800 000 deaths worldwide each year.1 Currently, suicide rates in the US are at their highest levels in more than 50 years after having experienced a relatively rapid, recent increase of 35% from 1999 to 2018 alone.2,3 Despite the urgency of this public health problem, real-time information on suicide fatality trends to guide prevention efforts is lacking. This is because national statistics on suicide rates are delayed by 1 to 2 years.4,5 Official statistics on suicide fatalities are produced by the Centers for Disease Control and Prevention using data from the National Vital Statistics System. The National Vital Statistics System collects information from death certificates that are submitted by states to the Centers for Disease Control and Prevention from more than 2000 medical examiner and coroner offices in the US.6 Reasons for the delay in suicide fatality data include the decentralized nature of mortality statistics, critical shortages of workers such as forensic pathologists, variation in local information technology infrastructure, and the time needed to investigate deaths due to suicide.4,7,8Keywords
This publication has 30 references indexed in Scilit:
- Tracking Suicide Risk Factors Through Twitter in the USCrisis, 2014
- Reassessing Google Flu Trends Data for Detection of Seasonal and Pandemic Influenza: A Comparative Epidemiological Study at Three Geographic ScalesPLOS Computational Biology, 2013
- Predicting National Suicide Numbers with Social Media DataPLOS ONE, 2013
- Influenza Forecasting with Google Flu TrendsPLOS ONE, 2013
- CRITICAL QUESTIONS FOR BIG DATAInformation, Communication & Society, 2012
- Internet monitoring of suicide risk in the populationJournal of Affective Disorders, 2010
- Social Media Use in the United States: Implications for Health CommunicationJournal of Medical Internet Research, 2009
- Super LearnerStatistical Applications in Genetics and Molecular Biology, 2007
- A tutorial on support vector regressionStatistics and Computing, 2004
- Time series segmentation: A sliding window approachInformation Sciences, 1995