An operational epidemiological model for calibrating agent-based simulations of pandemic influenza outbreaks

Accuracy of State-Level Surveillance during Emerging Outbreaks of Respiratory Viruses: A Model-Based Assessment

It is long perceived that the more data collection, the more knowledge emerges about the real disease progression. During emergencies like the H1N1 and the severe acute respiratory syndrome coronavirus 2 pandemics, public health surveillance requested increased testing to address the exacerbated demand. However, it is currently unknown how accurately surveillance portrays disease progression through incidence and confirmed case trends. State surveillance, unlike commercial testing, can process specimens based on the upcoming demand (e.g., with testing restrictions). Hence, proper assessment of accuracy may lead to improvements for a robust infrastructure. Using the H1N1 pandemic experience, we developed a simulation that models the true unobserved influenza incidence trend in the State of Michigan, as well as trends observed at different data collection points of the surveillance system. We calculated the growth rate, or speed at which each trend increases during the pandemic growth phase, and we performed statistical experiments to assess the biases (or differences) between growth rates of unobserved and observed trends. We highlight the following results: 1) emergency-driven high-risk perception increases reporting, which leads to reduction of biases in the growth rates; 2) the best predicted growth rates are those estimated from the trend of specimens submitted to the surveillance point that receives reports from a variety of health care providers; and 3) under several criteria to queue specimens for viral subtyping with limited capacity, the best-performing criterion was to queue first-come, first-serve restricted to specimens with higher hospitalization risk. Under this criterion, the lab released capacity to subtype specimens for each day in the trend, which reduced the growth rate bias the most compared to other queuing criteria. Future research should investigate additional restrictions to the queue.

Estimating disease burden of a potential A(H7N9) pandemic influenza outbreak in the United States

BACKGROUND

Since spring 2013, periodic emergence of avian influenza A(H7N9) virus in China has heightened the concern for a possible pandemic outbreak among humans, though it is believed that the virus is not yet human-to-human transmittable. Till June 2017, A(H7N9) has resulted in 1533 laboratory-confirmed cases of human infections causing 592 deaths. The aim of this paper is to present disease burden estimates (measured by infection attack rates (IAR) and number of deaths) in the event of a possible pandemic outbreak caused by human-to-human transmission capability acquired by A(H7N9) virus. Even though such a pandemic will likely spread worldwide, our focus in this paper is to estimate the impact on the United States alone.

METHOD

The method first uses a data clustering technique to divide 50 states in the U.S. into a small number of clusters. Thereafter, for a few selected states in each cluster, the method employs an agent-based (AB) model to simulate human A(H7N9) influenza pandemic outbreaks. The model uses demographic and epidemiological data. A few selected non-pharmaceutical intervention (NPI) measures are applied to mitigate the outbreaks. Disease burden for the U.S. is estimated by combining results from the clusters applying a method used in stratified sampling.

RESULTS

Two possible pandemic scenarios with R 0 = 1.5 and 1.8 are examined. Infection attack rates with 95% C.I. (Confidence Interval) for R 0 = 1.5 and 1.8 are estimated to be 18.78% (17.3-20.27) and 25.05% (23.11-26.99), respectively. The corresponding number of deaths (95% C.I.), per 100,000, are 7252.3 (6598.45-7907.33) and 9670.99 (8953.66-10,389.95).

CONCLUSIONS

The results reflect a possible worst-case scenario where the outbreak extends over all states of the U.S. and antivirals and vaccines are not administered. Our disease burden estimations are also likely to be somewhat high due to the fact that only dense urban regions covering approximately 3% of the geographic area and 81% of the population are used for simulating sample outbreaks. Outcomes from these simulations are extrapolated over the remaining 19% of the population spread sparsely over 97% of the area. Furthermore, the full extent of possible NPIs, if deployed, could also have lowered the disease burden estimates.

Lessons from a decade of individual-based models for infectious disease transmission: a systematic review (2006-2015)

Background

Individual-based models (IBMs) are useful to simulate events subject to stochasticity and/or heterogeneity, and have become well established to model the potential (re)emergence of pathogens (e.g., pandemic influenza, bioterrorism). Individual heterogeneity at the host and pathogen level is increasingly documented to influence transmission of endemic diseases and it is well understood that the final stages of elimination strategies for vaccine-preventable childhood diseases (e.g., polio, measles) are subject to stochasticity. Even so it appears IBMs for both these phenomena are not well established. We review a decade of IBM publications aiming to obtain insights in their advantages, pitfalls and rationale for use and to make recommendations facilitating knowledge transfer within and across disciplines.

Methods

We systematically identified publications in Web of Science and PubMed from 2006-2015 based on title/abstract/keywords screening (and full-text if necessary) to retrieve topics, modeling purposes and general specifications. We extracted detailed modeling features from papers on established vaccine-preventable childhood diseases based on full-text screening.

Results

We identified 698 papers, which applied an IBM for infectious disease transmission, and listed these in a reference database, describing their general characteristics. The diversity of disease-topics and overall publication frequency have increased over time (38 to 115 annual publications from 2006 to 2015). The inclusion of intervention strategies (8 to 52) and economic consequences (1 to 20) are increasing, to the detriment of purely theoretical explorations. Unfortunately, terminology used to describe IBMs is inconsistent and ambiguous. We retrieved 24 studies on a vaccine-preventable childhood disease (covering 7 different diseases), with publication frequency increasing from the first such study published in 2008. IBMs have been useful to explore heterogeneous between- and within-host interactions, but combined applications are still sparse. The amount of missing information on model characteristics and study design is remarkable.

Conclusions

IBMs are suited to combine heterogeneous within- and between-host interactions, which offers many opportunities, especially to analyze targeted interventions for endemic infections. We advocate the exchange of (open-source) platforms and stress the need for consistent “branding”. Using (existing) conventions and reporting protocols would stimulate cross-fertilization between research groups and fields, and ultimately policy making in decades to come.

Electronic supplementary material

The online version of this article (doi:10.1186/s12879-017-2699-8) contains supplementary material, which is available to authorized users.