Predictive performance of telenursing complaints in influenza surveillance: a prospective cohort study in Sweden

Abstract

Syndromic data sources have been sought to improve the timely detection of increased influenza transmission. This study set out to examine the prospective performance of telenursing chief complaints in predicting influenza activity. Data from two influenza seasons (2007/08 and 2008/09) were collected in a Swedish county (population 427,000) to retrospectively determine which grouping of telenursing chief complaints had the largest correlation with influenza case rates. This grouping was prospectively evaluated in the three subsequent seasons. The best performing telenursing complaint grouping in the retrospective algorithm calibration was fever (child, adult) and syncope (r=0.66; p<0.001). In the prospective evaluation, the performance of 14-day predictions was acceptable for the part of the evaluation period including the 2009 influenza pandemic (area under the curve (AUC)=0.84; positive predictive value (PPV)=0.58), while it was strong (AUC=0.89; PPV=0.93) for the remaining evaluation period including only influenza winter seasons. We recommend the use of telenursing complaints for predicting winter influenza seasons. The method requires adjustments when used during pandemics.

Epidemiological and statistical basis for detection and prediction of influenza epidemics

“…Two streams of data used for routine influenza surveillance in Östergötland County, Sweden (population 445,000 inhabitants) were used in this study: data on clinical influenza-diagnoses and data on syndromic chief complaints from a national telenursing service. The latter data source had previously been found to provide indications of increased influenza activity up to 2 weeks ahead of the former (Timpka et al 2014a(Timpka et al , 2014b.…”

“…Data from this learning set were used to determine the grouping of telenursing chief complaints with the largest correlation strength and best lead time between influenza-diagnosis data and telenursing data. The optimal combination of chief complaints was found to be fever (adult, child), and the most favorable lead time was 14 days (with telenursing data preceding influenza-diagnosis data) (Timpka et al 2014a(Timpka et al , 2014b. Using the predicted peak timing, the peak intensity prediction component was applied to influenza-diagnosis data from the corresponding epidemics to estimate the peak intensity on the predicted peak day.…”

“…The search of the meta-narrative review identified eight studies reporting prospective algorithm performance based on data from naturally occurring influenza epidemics: three studies (Closas et al 2012, Martínez-Beneito et al 2008, Cowling et al 2006) evaluated one or more detection algorithms and five (Timpka et al 2014a, Yuan et al 2013, Jiang et al 2009, Burkom et al 2007, Viboud et al 2003) evaluated prediction algorithms (Figure 4). …”

“…However, in that study, the predictions were made on day 13 and day 22 of a single ongoing epidemic. Another study applying a Shewhart-type model on telenursing data from a Swedish county to predict influenza epidemics over three seasons, covering two winter influenza seasons and the A (pH1N1) pandemic in 2009, showed outstanding performance for the winter influenza seasons on a daily basis and excellent performance on a weekly basis (Timpka et al 2014a). However, the performance was poor for the pandemic both on a daily basis and on a weekly basis.…”

“…In the past few years, a considerable amount of research has focused on the use of interactive health information technology-referred to as eHealth systems-to improve the effectiveness of infectious disease surveillance (Castillo-Salgado 2010). Researchers have focused on the use of Internet search engines (Kim et al 2013, Sharpe et al 2016, Pollett et al 2017, telenursing data (Timpka et al 2014a), mini-blogs (Nagel et al 2013, YomTov et al 2014, Sharpe et al 2016, and records of over-the-counter drug sales (Kirian & Weintraub 2010, Socan et al 2012. This implies that the area where the need for knowledge is most immediate is the detection and prediction of influenza activity at local levels (Shaman et al 2013).…”