Did Modeling Overestimate the Transmission Potential of Pandemic (H1N1-2009)? Sample Size Estimation for Post-Epidemic Seroepidemiological Studies

Abstract

BackgroundSeroepidemiological studies before and after the epidemic wave of H1N1-2009 are useful for estimating population attack rates with a potential to validate early estimates of the reproduction number, R, in modeling studies.Methodology/Principal FindingsSince the final epidemic size, the proportion of individuals in a population who become infected during an epidemic, is not the result of a binomial sampling process because infection events are not independent of each other, we propose the use of an asymptotic distribution of the final size to compute approximate 95% confidence intervals of the observed final size. This allows the comparison of the observed final sizes against predictions based on the modeling study (R = 1.15, 1.40 and 1.90), which also yields simple formulae for determining sample sizes for future seroepidemiological studies. We examine a total of eleven published seroepidemiological studies of H1N1-2009 that took place after observing the peak incidence in a number of countries. Observed seropositive proportions in six studies appear to be smaller than that predicted from R = 1.40; four of the six studies sampled serum less than one month after the reported peak incidence. The comparison of the observed final sizes against R = 1.15 and 1.90 reveals that all eleven studies appear not to be significantly deviating from the prediction with R = 1.15, but final sizes in nine studies indicate overestimation if the value R = 1.90 is used.ConclusionsSample sizes of published seroepidemiological studies were too small to assess the validity of model predictions except when R = 1.90 was used. We recommend the use of the proposed approach in determining the sample size of post-epidemic seroepidemiological studies, calculating the 95% confidence interval of observed final size, and conducting relevant hypothesis testing instead of the use of methods that rely on a binomial proportion.

Comparison of the pandemic H1N1 2009 experience in the Southern Hemisphere with pandemic expectations

“…Values in the range 1.2-1.4 are consistent with estimates of R obtained from seroprevalence surveys of pH1N1. 26,27 R has been estimated to be 2.0 (with a range of 1.4-2.8) for the 1918 pandemic, 1.6 for the pandemic of 1957 and 1.8 for the 1968 pandemic. R varies year-to-year for seasonal influenza with a mean around 1.3 and a range of 0.9-2.1.…”

Scalable Vaccine Distribution in Large Graphs given Uncertain Data

“…One source of uncertainty is public-health surveillance [39,16,34,30,12,5]. Generally there are three types of surveillance: population-based, health provider-based and lab-based.…”

“…SURVEILLANCE pi ∈ P Each node takes a probability from P (which is a finite set of probabilities like {0.1, 0.5, 0.9}). See Surveillance pyramid [39,34,30].…”

“…Handling Uncertainty. Many studies in epidemiology try to estimate the total infections using the surveillance pyramid [39,16,34,30,12,5]. More generally, missing data in networks is an important yet relatively poorly understood problem.…”

Hidden Hazards: Finding Missing Nodes in Large Graph Epidemics

“…There are many reasons why data may be missing in a cascade. In epidemiology for example, surveillance data on who is infected is limited and noisy [21] -the well-known 'surveillance-pyramid' demonstrates that detected infections are often only a fraction of the actual infections [17]. In Facebook, most users keep their activity and profiles private, while in Twitter only a percentage sample of Tweets are accessible by the pub-…”