BACKGROUND: Opportunities and challenges exist for local public health departments with various initiatives related to health information technology, health information exchanges, and the expanding need for information to address emerging health threats and monitor disease trends. This presentation will highlight the results from matching electronic lab reporting (ELR) information with syndromic surveillance data to more fully describe and characterize real disease outcomes and community impact. The County of San Diego is leveraging both ELR and syndromic surveillance systems to identify additional “meaningful” value of these Meaningful Use information sources, effectively describe influenza disease trends, and determine improvements which can be made to how disease surveillance may function in a more electronic era.
METHODS: A retrospective analysis was conducted in which 595 influenza-specific electronic lab reports dated from January 1 to December 31, 2016 were matched with a local hospital’s corresponding HL7 messages via patient medical record numbers using SAS statistical software. After utilizing a syndromic surveillance text classifier, frequencies and percentages were calculated for key flu-related chief complaint categories as well as symptom combinations consistent with the Council of State and Territorial Epidemiologists (CSTE) influenza-like illness (ILI) case definition. Proportions of patients with influenza-related and ILI chief complaints were also obtained.
RESULTS: Of 1,626 chief complaints identified, those most frequently reported among patients with lab-confirmed influenza included fever (24.35%), cough (20.66%), or other respiratory complaint (8.49%). Among the matched 595 patients, almost all (568; 95.46%) had at least one identified key flu-related chief complaint term while 27 (4.54%) had chief complaints unrelated to influenza. Of 433 chief complaints containing ILI symptoms, 51 (11.78%) specifically mentioned any variation of “flu-like” or “ILI” symptoms. Most (60.28%) were composed of a combination of fever and cough, while combinations of fever with sore throat, congestion, or rhinorrhea were less common (13.63%, 8.55%, and 5.77%, respectively). More than half (326; 54.79%) of patients either had a chief complaint that referred to flu-like/ILI symptoms or contained a combination of ILI symptoms.
CONCLUSIONS: Among our results, the majority of patients with laboratory-confirmed influenza had at least one influenza symptom chief complaint term and about half of the patients had chief complaints consistent with the more traditional ILI criteria. With the growing amount of electronic health data available, the value of linking records across systems is evident as opportunities exist to expand public health capabilities to improve the health of populations impacted by diseases like influenza.