BACKGROUND: The National Respiratory and Enteric Viral Surveillance System (NREVSS) monitors the detection of various respiratory viruses, including influenza. While useful nationally, NREVSS data have limitations. They are typically hand-entered at the laboratory, thus requiring additional staff time or delaying reports, and lack patient demographics. Currently, only data from the Florida Bureau of Public Health Laboratories (BPHL) and NREVSS are used to inform influenza laboratory surveillance in Florida. Some laboratories in Florida have voluntarily submitted positive and negative influenza test results to the Florida Department of Health (DOH) through electronic laboratory reporting (ELR) since 2013. Mandatory submission for all laboratories participating in ELR began in June 2014 to provide more complete and timely data receipt. Because influenza laboratory results have never been reportable via paper, DOH does not have historical paper data available to validate newly received ELR data and requesting paper laboratory results directly from the laboratory for validation would be tedious and inefficient. Instead, this analysis uses existing influenza test results submitted by NREVSS participating Florida laboratories to validate newly received ELR data.
METHODS: The DOH receives ELR data daily. Influenza ELR data from 2013-2015 were correlated with NREVSS test results by Morbidity and Mortality Weekly Report (MMWR) week based on three parameters: total count of influenza tests, total count of influenza positive results, and the percent of positive influenza laboratory results. A significance level of p ≤0.05 between two of the three parameters when using a Spearman’s rank t-test to correlate NREVSS and ELR data determined correlation and therefore validity of the ELR data.
RESULTS: Of 69 laboratories regularly submitting ELR influenza test results, 37 laboratories also submitted NREVSS data as of November 14, 2015. Of these 37 laboratories, 24 were validated using NREVSS data. Two of the three parameters for all 24 laboratories were significantly correlated with the corresponding parameters in NREVSS with t-test significance levels ranging from p=0.046 to p<0.0001. DOH is working with the remaining 14 laboratories to correct data submission errors for future validation.
CONCLUSIONS: Validated ELR data have replaced NREVSS data in current Florida influenza laboratory surveillance. Substituting NREVSS data with ELR data improves the laboratory surveillance systems by providing timely, standardized, and complete data with demographics and builds on existing ELR processes. The methodology described here can be used to validate ELR data instead of using paper reporting data, which is easier and more efficient than paper-based validation.