Mathes Robert

An interdisciplinary team convened by ISDS to translate public health use-case needs into well-defined technical problems recently identified the need for new pre-syndromic surveillance methods that do not rely on existing syndromes or pre-defined illness categories1. Our group has recently developed Multidimensional Semantic Scan (MUSES), a pre-syndromic surveillance approach that (1) uses topic modeling to identify newly emerging syndromes that correspond to rare or novel diseases; and (2) uses multidimensional scan statistics to identify emerging outbreaks that correspond to these syndromes and are localized to a particular geography and/or subpopulation2,3. Through a blinded evaluation on retrospective free-text ED chief complaint data from NYC DOHMH, we demonstrate that MUSES has great potential to serve as a safety net for public health surveillance, facilitating a rapid, targeted, and effective response to emerging novel disease outbreaks and other events of relevance to public health that do not fit existing syndromes and might otherwise go undetected.

Objective: We present a new approach for pre-syndromic disease surveillance from free-text emergency department (ED) chief complaints, and evaluate the method using historical ED data from New York City's Department of Health and Mental Hygiene (NYC DOHMH).

In 2012, an outbreak of Mycobacterium chelonae infections in tattoo recipients in Rochester, NY was found to be associated with premixed tattoo ink contaminated before distribution.1 In May 2012, a case of M. chelonae was reported in a New York City (NYC) resident who received a tattoo with ink alleged to have been diluted with tap water. When a second case of M. chelonae in a tattoo recipient was reported in March 2013, an investigation was initiated. M. chelonae is not reportable in NYC other than in clusters reported by providers or laboratories. To determine if there were additional tattoo-associated M. chelonae infections, we searched for cases using NYC ED syndromic surveillance.

Objective

To investigate tattoo-associated skin infections due to Mycobacterium chelonae using Emergency Department (ED) syndromic surveillance.

The role of public health in preparing for, responding to, and recovering from emergencies has expanded as a result of the massive impact recent disasters have had on affected populations. Nearly every large-scale disaster carries substantial public health risk and requires a response that addresses immediate effects of the disaster on a population (e.g., mass casualties and severe injuries, lack of shelter in severe weather), as well as subsequent secondary physical effects (e.g., carbon monoxide poisoning due to improper operation or location of carbon monoxide-producing devices such as generators) and emotional effects (e.g., grief, anxiety, and post-traumatic stress disorder) caused by the disaster. Disaster epidemiology has been identified as an evolving field that integrates a variety of data sources and technological and geospatial resources to expedite reporting and to increase the accuracy of information collected and used by emergency planners and incident managers. As the national organization that supports the activities of applied epidemiologists in state, tribal, local, territorial, and federal public health agencies, the Council of State and Territorial Epidemiologists (CSTE) assembled a Disaster Epidemiology Subcommittee of public health experts and practitioners from diverse fields of applied epidemiology to discuss the use of epidemiologic methods in all phases of the disaster management cycle. In 2012, the Subcommittee assessed state-level disaster epidemiology capacity with a focus on surveillance. 

Objective

The panel will discuss the current status of disaster surveillance capabilities at local and state health departments in the United States and will provide an overview of current resources available to epidemiologists for surveillance.

The NYC syndromic surveillance system has been monitoring syndromes from NYC emergency department (ED) visits for over a decade. We applied several aberration detection methodologies to a time series of ED visits in NYC spiked with synthetic outbreaks. This effort is part of a larger evaluation of the NYC syndromic system, funded by a grant from the Alfred P. Sloan Foundation.

Objective

To critically evaluate temporal aberration detection methodologies using New York City (NYC) syndromic surveillance data.

ARIMA models use past values (autoregressive terms) and past forecasting errors (moving average terms) to generate future forecasts, making it a potential candidate method for modeling citywide time series of syndromic data [1]. While past research supports the use of ARIMA modeling as a detection algorithm in syndromic surveillance [2], there has been little evaluation of an ARIMA model's prospective outbreak detection capabilities. We built an ARIMA model to prospectively detect simulated outbreaks in ED syndromic data. This method is one of eight being formally evaluated as part of a grant from the Alfred P. Sloan Foundation.

Objective

To evaluate seasonal autoregressive integrated moving average (ARIMA) models for prospective analysis of New York City (NYC) emergency department (ED) syndromic data.

The CC text field is a rich source of information, but its current use for syndromic surveillance is limited to a fixed set of syndromes that are routine, suspected, expected, or discovered by chance. In addition to syndromes that are routinely monitored by the NYC Department of Health and Mental Hygiene (e.g., diarrhea, respiratory), additional syndromes are occasionally monitored when requested by outside sources or when expected to increase during emergencies. During Hurricane Sandy, we discovered by manual inspection of data for a few EDs an increase in certain words in the CC field (e.g., 'METHADONE', 'DIALYSIS', and 'OXYGEN') that led to the creation of a 'needs medication' syndrome. Current syndromic surveillance systems cannot detect unanticipated events that are not defined a priori by keywords. We describe a simple data-driven method that routinely scans the CC field for increases in word frequency that might trigger further investigation and/or temporary monitoring.

Objective

To detect sudden increases in word frequency in the Emergency Department (ED) syndromic chief complaint (CC) text field.

As technology advances, the implementation of statistically and computationally intensive methods to detect unusual clusters of illness becomes increasingly feasible at the state and local level [2]. Bayesian methods allow for the incorporation of prior knowledge directly into the model, which could potentially improve estimation of expected counts and enhance outbreak detection. This method is one of eight being formally evaluated as part of a grant from the Alfred P. Sloan Foundation.

Objective

To adapt a previously described Bayesian model-based surveillance technique for cluster detection [1] to NYC Emergency Department (ED) visits.