Optimizing Provider Recruitment for Influenza Surveillance Networks

Abstract

The increasingly complex and rapid transmission dynamics of many infectious diseases necessitates the use of new, more advanced methods for surveillance, early detection, and decision-making. Here, we demonstrate that a new method for optimizing surveillance networks can improve the quality of epidemiological information produced by typical provider-based networks. Using past surveillance and Internet search data, it determines the precise locations where providers should be enrolled. When applied to redesigning the provider-based, influenza-like-illness surveillance network (ILINet) for the state of Texas, the method identifies networks that are expected to significantly outperform the existing network with far fewer providers. This optimized network avoids informational redundancies and is thereby more effective than networks designed by conventional methods and a recently published algorithm based on maximizing population coverage. We show further that Google Flu Trends data, when incorporated into a network as a virtual provider, can enhance but not replace traditional surveillance methods.

Data Blindspots: High-Tech Disease Surveillance Misses the Poor

“…Past work on infectious disease surveillance has focused at the statelevel (Polgreen et al 2009;Scarpino et al 2012) or assumed that risk was evenly spread across well-mixed populations (Pelat et al 2014). Surveillance studies focused on broader definitions of health and on chronic diseases have found similar disparities to the ones presented here (Liao et al 2004;Kandula et al 2007).…”

Improving incidence estimation in practice-based sentinel surveillance networks using spatial variation in general practitioner density

“…Here, we showed that bias due to variations in local GP density around participants could be reduced using weighted estimators, and supported this conclusion with empirical evidence. This approach could be used to introduce more flexibility for developing algorithms to identify the best placement of data providers [3–7]. …”

“…In this last respect, seemingly reasonable solutions may have undesirable features: we found here that selecting one GP in each of the 100 largest French towns would lead to large underestimation of incidence. Several approaches have been described to the optimal selection of data providers in surveillance, taking into account spatial coverage or representativeness of monitored population [3–7]. In our experience, the effective use of such approaches in GP based surveillance is difficult as it is not possible to have a stable roster of GPs who want to participate in surveillance [8].…”

“…Another possibility is to make the monitored population representative in terms of age, sex or socio-economic status [5, 6]. It may also be possible to select participants so that estimates from the surveillance network are as close as possible to a reference epidemiologic signal, for example influenza-like hospitalizations when monitoring flu in the community [7]. …”

Enhancing disease surveillance with novel data streams: challenges and opportunities

“…For example, NDS have facilitated expansion of dengue and influenza surveillance to countries without infrastructure capable of real time surveil-lance [5, 17, 21, 22]. This has also been done in the context of hospitalizations in Texas [23], mental illness, psychological manifestations of physical morbidities [24, 25], and search queries from clinical decision support sites, such as UpToDate [26]. In these cases, although NDS-based systems are being asked to estimate data that is actually being collected, those data are not available quickly enough for use in public health decision making.…”