Presented January 26, 2017.
This presentation will describe the steps involved in machine learning and will include a demo an application to detect carbon monoxide poisoning in the Kansas syndromic surveillance data.
Air Quality
In general, data from public health surveillance can be used for short- and long-term planning and response through retrospective data analysis of trends over time or specific events. Combining health outcome data (e.g., hospitalizations or deaths) with environmental and socio-demographic information also provides a more complete picture of most vulnerable populations. Using syndromic surveillance systems for climate and health surveillance offers the unique opportunity to help quantify and track in near-real time the burden of disease from climate and weather impacts.
Description
Wildfires occur annually in Oregon, and the health risks of wildfire smoke are well documented1. Before implementing syndromic surveillance through Oregon ESSENCE, assessing the health effects of wildfires in real time was very challenging. Summer 2015 marked the first wildfire season with 60 of 60 eligible Oregon emergency departments (EDs) reporting to ESSENCE. The Oregon ESSENCE team developed a wildfire surveillance pilot project with two local public health authorities (LPHAs) to determine their surveillance needs and practices and developed a training program to increase capacity to conduct surveillance at the local level. Following the training, one of the LPHAs integrated syndromic surveillance into its routine surveillance practices. Oregon ESSENCE also integrated the evaluation findings into the summer 2016 statewide wildfire surveillance plan.
Objective
To build capacity to conduct syndromic surveillance at the local level by leveraging a health surveillance need.
Description
The impact of poor air quality (AQ) on human health is a global issue, with periods of poor AQ known to occur in multiple locations, across different countries at, or around the same time. The Public Health England (PHE) Emergency Department Syndromic Surveillance System (EDSSS) is a public health legacy of the London 2012 Olympic and Paralympic Games, monitoring anonymised daily attendance data in near real-time from a sentinel network of up to 38 EDs across England and Northern Ireland during 2014. The Organisation de la Surveillance COordonnée des URgences (OSCOUR®) is a similar ED system coordinated by Santé publique France and has been running in France since 2004, established following a major heatwave in 2003 to improve real-time public health surveillance capabilities. This truly national network included around 540 EDs in 2014.
Objective
To assess the impact on human health observed in association with periods of poor air quality which extended across international borders, affecting both London (UK) and Paris (France). In particular to quantify increased levels of emergency department (ED) attendances for asthma and wheeze/ difficulty breathing, and how different age groups were affected. Here, using ED syndromic surveillance from England and France, we aimed to identify and describe the acute impact of periods of particularly poor air quality during 2014 on human health in both London and Paris.
Description
It is estimated that in the United States (US), unintentional non-fire related CO poisoning causes an average of 439 deaths annually, and in 2007 confirmed CO poisoning cases resulted in 21,304 ED visits and 2,302 hospitalizations (71 per million and 8 per million population, respectively)1 . Despite the significant risk of morbidity and mortality associated with CO poisoning, existing surveillance systems in the United States are limited. This study is the first to focus specifically on CO poisoning trends within the VHA population.
Objective
To describe characteristics of Veterans Health Administration (VHA) patients with ICD 9/10 CM inpatient discharge and/or emergency department (ED)/urgent care outpatient encounter codes for carbon monoxide (CO) poisoning.