Data Quality

In New Jersey, real-time emergency department (ED) data are currently received from EDs by Health Monitoring Systems Inc.’s (HMS) EpiCenter, which collects, manages and analyzes ED registration data for syndromic surveillance, and provides alerts to state and local health departments for surveillance anomalies.

EpiCenter receives pre-diagnostic chief complaint data from 78 of 80 acute care and satellite EDs. The need for more specific information raises the possibility that other data elements from EDs such as triage notes can be of utility in detecting outbreaks without a significant delay. This study evaluates the inclusion of triage notes in EpiCenter to detect a recent increased usage of synthetic cannabinoids. At the time of this evaluation, three New Jersey hospitals were providing triage notes in their EpiCenter data.

Objective

Describe the inclusion of triage notes into a syndromic surveillance system to enhance population health surveillance activities.

Illnesses related to synthetic marijuana use have been reported in many states, including Florida. Because these visits can present with a variety of symptoms, as well as be attributed to numerous diagnosis codes, it can be difficult to identify and quantify these visits. The Electronic Surveillance System for the Early Notification of Community-based Epidemics in Florida (ESSENCE-FL) receives chief complaint (CC) and discharge diagnosis (DD) data as free text allowing uncommon or new terms to be searched for within each patient visit. The main source of data for ESSENCE-FL is emergency department (ED) and urgent care center (UCC) data. There are currently 210 EDs and 33 UCCs throughout Florida that send their data to the ESSENCE-FL server. Using ESSENCE-FL, a free text query of patient CCs and DDs was used to identify visits related to synthetic marijuana use. This study is designed to analyze these identified visits for trends over time, geographical distribution and descriptive statistics and demographics.

Objective

One of the numerous functions of syndromic data has been the identification of visits of public health interest using customized free text queries. A specific query of syndromic data was created to search for and identify emergency department (ED) and urgent care center (UCC) visits possibly related to the use of synthetic marijuana to describe and quantify this public health issue in Florida.

Syndromic surveillance requires reliable, accurate, and complete healthcare encounter data to assess patterns of illness and respond to public health events. Illinois implemented syndromic surveillance statewide in response to Meaningful Use reporting objectives. To address the need for continuous, automated assessment following initial on-boarding of facility Emergency Department data, we developed an R script to assess the quality of data in the private BioSense locker database.

This script builds upon and adapts from scripts previously developed for syndromic surveillance and data quality assessment.

Objective

To describe an R script developed to assess and produce reports on data quality in the BioSense locker database.

The CMS EHR Incentive Programs include a measure for meaningful use of EHR systems for submitting syndromic surveillance messages to public health. The Stage 2 measure defines the standard for transmission to be HL7 v2.5.1 Admit/Discharge/Transfer messages according to the PHIN Messaging Guide for Syndromic Surveillance and Conformance Clarification for EHR Certification of Electronic Syndromic Surveillance, Addendum to PHIN Messaging Guide for Syndrome Surveillance. The National Institute of Standards and Technology (NIST) provides an online testing tool for validating messages. While some jurisdictions use the Biosense platform for receiving, managing, and analyzing syndromic surveillance data, there is no consistent tool that is available to jurisdictions to assess the quality and conformance of data submissions both at the time of on-boarding a new reporting facility and on an ongoing basis during production operations.

The New York City Citywide Immunization Registry (CIR), the immunization information system for NYC that has been operational since 1997, has as part of its software suite an Open Source, webbased data quality assurance (QA) tool used by its research scientists to qualify new sites for reporting data electronically via HL7 v2 messages, and for monitoring the ongoing quality of data submissions over time. A validation process evaluates incoming messages against the rules established by an implementation guide (IG) and stores the result of the evaluation in a CIR database table that is accessible by the QA Tool which displays the data to an administrative user. This project served as a proof-of-concept for implementing a similar process for syndromic surveillance.

Objective

To leverage an existing open source quality assurance software tool created for the immunization domain and modify it to serve as a quality assurance tool for syndromic surveillance messages.

The Public Health Agency of Canada (PHAC) established the Canadian Chronic Disease Surveillance System (CCDSS) in 2009 to facilitate national estimates of chronic disease prevalence, incidence, and health outcomes. The CCDSS uses population-based linked health administrative databases from all provinces/territories (P/Ts) and a distributed analytic protocol to produce standardized disease estimates.

Objective

To describe the process, benefits, and challenges of implementing a distributed model for chronic disease surveillance across thirteen Canadian jurisdictions.

As part of this surveillance study for Avian Influenza both active and passive surveillance samples were tested using PCR and also utilized to validate the LAMP method. Active surveillance samples include pathological material and tracheal and cloacal swabs from ill poultry, which were subsequently assessed for avian influenza during diagnosis, and birds collected by hunters. Passive surveillance included environmental samples such as sand and bird faeces. Active surveillance samples were taken mostly from poultry farms across Ukraine, where infected birds are required to be diagnosed by State Scientific Research Institute of Laboratory Diagnostics and Veterinary Sanitary Expertise (SSRILDVSE) by Ukraine Law. Passive surveillance samples were taken primarily during the annual bird migration season. Development of simple, sensitive, and cheap methods for diagnostics of avian influenza is a very important task for practical veterinary medicine. LAMP is one of such methods. The technique is based on isothermal amplification of nucleic acids. It does not require special conditions and equipment (PCR cyclers), therefore it is cheaper in comparison with PCR. Accurate diagnosis is necessary for determining the risk associated with avian influenza in Ukraine and along the Dnipro River during the migratory season.

Objective

The performance of comparative analysis of sensitivity and results of detection of avian influenza virus by real time polymerase chain reaction (PCR-RT) and loop-mediated isothermal amplification of the nucleic acids (LAMP) was the main goal of the study.

To immediately monitor disease outbreaks, the application of laboratory-based surveillance is more popular in recent years. Taiwan Centers for Disease Control (TCDC) has developed LARS to collect the laboratory-confirmed cases caused by any of 20 pathogens daily via automated submitting of reports from hospital laboratory information system (LIS) to LARS since 2014 [1]. LOINC is used as standardized format for messaging inspection data [1, 2]. There are 37 hospitals have joined LARS, coverage rate about 59% of all hospitals in Taiwan. Recently, more than 10,000 of data are collected weekly and used in monitoring pathogen activity [3]. Therefore, it is important to ensure data quality that the data will lead to valuable information for public health surveillance.

Objective

To improve data quality and sustain a good quality data collected by Laboratory Automated Reporting System (LARS), we use a Threestage Data Quality Correction (3DQC) strategy to ensure data accuracy.

The Ministry of Health of Georgia accepted the Electronic Integrated Disease Surveillance System (EIDSS) as an official disease reporting system in 2012. The Georgian government adopted electronic reporting for both veterinary and human diseases in 2015. We conducted a comparative assessment of progress in the implementation of electronic reporting.

Objective

The objective of this study was to evaluate several aspects of the electronic disease reporting system and its abilities used in Georgia. Also, the study assessed if the system abilities are tailored to the national surveillance requirements. User attitudes (system strength and weaknesses) toward the system was also surveyed.

Data sets from disparate sources widely vary in the number and type of factors which most hamper integrity and timeliness of the data. To maintain high quality data, data sets must be regularly assessed, particularly for those vulnerabilities that each is especially prone to due to the methods involved in collecting the data. For surveillance practitioners charged with monitoring data from multiple data sources, keeping track of the issues that each data set is susceptible to, and quickly identifying any inconsistencies or deviations from normal trends, may be a challenge. An application that can track all those issues, and trigger alerts when patterns diverge from what is expected, could help to enhance the efficiency and effectiveness of the surveillance efforts.

Objective

An interactive, point-and-click application was developed to facilitate the routine assessment of known data quality factors that compromise the integrity and timeliness of data sets used at the Marion County Public Health Department (MCPHD). The code (and associated documentation) for this application is being made available for other surveillance practitioners to adopt.

NYS (excluding NYC) has a very robust Communicable Disease Electronic Surveillance System (CDESS). This system provides disease specific modules, as well as a tracking system for contacts, and a perinatal infant tracking system. This system provides an easy way for users to quickly download a file with all of their data.

NYS (excluding NYC) tracks, on average, 300 infants of hepatitis B surface antigen (HBsAg) positive mothers annually. CDESS provides an infant tracking module for local health departments (LHDs) to enter and monitor vaccine information, add multiple infants per mother, and track patient movement and loss to follow-up. The tool allows LHDs to analyze infants’ data by birth year cohort, with all of their current vaccination and serology information available in one record.

In 2013 and 2014, more than 13,000 cases of gonorrhea were reported to CDESS in NYS (excluding NYC). From November 2013 through May 2014, only 61% of cases were adequately treated with a regimen recommended by the Centers for Disease Control and Prevention (CDC) STD Treatment Guidelines for Gonorrhea , and 29% were missing treatment information. The CDESS system allows the LHDs to track patients who have inadequate and/or missing treatment information.

Objective

Improved methods for user analysis of communicable disease surveillance data in New York State (NYS), excluding New York City (NYC).