![[ Visit Client Website ]](images/banner.png)
BACKGROUND: Influenza is five times more likely to cause severe illness (such as hospitalizations, mortality, pre-term labor and delivery) in pregnant women than women who are not pregnant. Hospitalizations for influenza are not currently reportable in Florida. Administrative data on hospitalized pregnant women with influenza infection are recorded but available yearly through the Florida Agency for Healthcare Administration (AHCA) hospital discharge dataset. This limits timely evidence-based decision making about a sub-population at risk of severe health outcomes from influenza infection. This investigation aimed to determine if emergency department (ED) visit data received daily or hourly and accessed through Florida’s syndromic surveillance system (ESSENCE-FL) could be used to timely estimate hospitalizations for influenza in pregnant women during the influenza season.
METHODS: The AHCA hospital discharge dataset for hospital admissions from 2008-2012 was queried for women aged 14-49 diagnosed with influenza (ICD9-CM Code: 487 or 488). Record review identified women, pregnant at time of admission (ICD9-CM Codes: 630-679 and V220-V222). ESSENCE-FL was used to query ED visit data from 2008-2012 for women aged 14-49 with “preg” as a chief complaint and with an influenza-like-illness (ILI) syndrome or influenza subsyndrome. Data were matched by facility name. Records where the facility was represented in both data sets were retained (91 facilities). Historical ED visit records for influenza and ILI in ESSENCE-FL were compared to historical AHCA records of pregnant women admitted to the hospital with diagnosed influenza.
RESULTS: The hospital discharge dataset contained records of 858 pregnant women from 91 facilities admitted to Florida hospitals. From the same 91 facilities, there were1,841 ED visits identified via ESSENCE-FL from women captured in the ILI syndrome, 1,217 ED visits from pregnant women were captured in the influenza subsyndrome. Trend analysis showed that ED visits for pregnant women categorized as having “influenza” as their subsyndrome more closely approximated the timing and volume of pregnant women diagnosed with influenza recorded in the AHCA hospital discharge data.
CONCLUSIONS: Although incomplete, ED chief complaint data captured in syndromic surveillance systems can be used as a proxy to timely estimate influenza hospitalizations of pregnant women during flu season. While rich and complete delays in AHCA hospital discharge data do not allow for timely informed decision-making. Pregnant women accounted for large numbers of ED visits and hospitalizations, emphasizing the need for this group to get vaccinated. Further analyses will compare the populations captured in the ILI syndrome and those in the influenza subsyndrome.