1265. Application of the ALERT Influenza Trigger for Enhanced Prevention Activities
Session: Poster Abstract Session: Healthcare Epidemiology: Outbreaks
Friday, October 5, 2018
Room: S Poster Hall
Posters
  • ID WEEK 2018 - influenza trigger Poster Final.pdf (1.2 MB)
  • Background: Accurate prediction of the onset of increased influenza activity in a healthcare setting can allow for optimal use of enhanced prevention activities. The ALERT (Above Local Elevated Respiratory Illness Threshold) algorithm, described by Reich et al. (2015), utilizes historical weekly case counts of laboratory-confirmed influenza infections to set a trigger point of cases/week that anticipates elevated disease incidence. This can then be used in real-time, during subsequent influenza seasons, for initiation of enhanced prevention, including masking by non-vaccinated healthcare workers.

    Methods: Historical data collected from UnityPoint Health-Des Moines (3-hospital, 800-bed system), between 2011 and 2014, was analyzed using the ALERT method (Reich et al. 2015) to set a threshold number of diagnosed influenza cases per week to predict the start of the influenza season. Each following year the threshold was re-analyzed, adding the most recent year’s data to the historical data set. Our goal was to capture at least 80% of influenza cases within our “ALERT period,” without prolonging the duration of heightened prevention efforts.

    Results: For the initial year of real-time application (2015 – 2016), the threshold was defined as 7 cases. In subsequent years, the threshold was 5. Compared to the three years prior, use of the ALERT method resulted in more accurate and consistent identification of the influenza season, including anticipating the increase in cases and defining the total duration of the season.

    Influenza Season Metrics

    Threshold trigger

    Onset date of enhanced prevention

    Total cases per season

    Percent of cases captured during alert

    Duration of alert period (weeks)

    2012-13

    N/A

    11/15/12

    725

    99%

    22

    2013-14

    N/A

    12/30/13

    255

    77%

    9

    2014-15

    N/A

    12/11/14

    773

    73%

    19

    ALERT Trigger Initiated

    2015-16

    7

    2/22/16

    250

    87%

    13

    2016-17

    5

    1/3/17

    791

    95%

    17

    2017-18*

    5

    12/11/17

    1445

    98%

    20

    *As of April 28, 2018

    Conclusion: The ALERT method utilizes health system specific historical data to more precisely define the period of high influenza incidence allowing for focused utilization of enhanced measures to prevent transmission. This results in a safer environment, optimal use of resources and improved employee and patient satisfaction.

    Lisa Veach, MD, FACP, FSHEA1, Julie Gibbons, BSN, RN1, Katarina Smolka, BS2 and Veena Ramachandran, MD1, (1)UnityPoint Health, Des Moines, IA, (2)Regis University, Denver, CO

    Disclosures:

    L. Veach, None

    J. Gibbons, None

    K. Smolka, None

    V. Ramachandran, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.