Global mortality estimates for the 2009 Influenza Pandemic from the GLaMOR project: a modeling study.

Simonsen L, Spreeuwenberg P, Lustig R, Taylor RJ, Fleming DM, Kroneman M, Van Kerkhove MD, Mounts AW, Paget WJ, GLaMOR Collaborating Teams. · 2013

Read Global mortality estimates for the 2009 Influenza Pandemic from the GLaMOR project: a modeling study. by Simonsen L, Spreeuwenberg P, Lustig R, Taylor RJ, Fleming DM, Kroneman M, Van Kerkhove MD, Mounts AW, Paget WJ, GLaMOR Collaborating Teams. (2013) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc3841239.txt.

Introduction

Lone Simonsen leads a team, with named collaborating teams, in a modeling study in PLoS Medicine in 2013. They estimate deaths from the 2009 influenza pandemic. Lab-confirmed death counts were low, which led some to call the event mild. The team ask what the true toll was.

Their two-stage model starts from death data in countries with good records, then extends the pattern to the world. Lab counts miss deaths where no test was done, above all in poorer regions. A model-based toll gives a fuller base for judging the pandemic and for planning the next one. The study is part of a project built to standardize such counts across countries, so future pandemics can be sized in a shared way.

Catalog record

CREATORSimonsen L, Spreeuwenberg P, Lustig R, Taylor RJ, Fleming DM, Kroneman M, Van Kerkhove MD, Mounts AW, Paget WJ, GLaMOR Collaborating Teams.
YEAR2013
COLLECTIONPAPERS
FORMATTXT
FILE SIZE67 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC3841239

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Global Mortality Estimates for the 2009 Influenza Pandemic from the GLaMOR Project: A Modeling Study By Lone Simonsen, Peter Spreeuwenberg, Roger Lustig, Robert J Taylor, Douglas M Fleming, Madelon Kroneman, Maria D Van Kerkhove, Anthony W Mounts, W John Paget, Simon I Hay Published in: PLoS Medicine DOI: 10.1371/journal.pmed.1001558 ABSTRACT Lone Simonsen and colleagues use a two-stage statistical modeling approach to estimate the global mortality burden of the 2009 influenza pandemic from mortality data obtained from multiple countries. Please see later in the article for the Editors' Summary INTRODUCTION Recurring seasonal influenza epidemics impose a moderate, if variable, mortality burden every year. But when a new human-transmissible influenza virus emerges, the ensuing pandemic can be catastrophic; the 1918 Spanish influenza pandemic, for example, killed approximately 1%–2% of the global population. Understanding the global mortality impact of pandemic influenza—who died, where, and when—is fundamental to understanding how pandemics emerge and evolve, and will help to guide responses to future pandemics. And because so few pandemics have occurred in the modern era, it is essential that each one be studied thoroughly—even if, as was the case with the 2009 influenza A H1N1 pandemic (H1N1pdm09), the catastrophe failed to appear. As of 31 August 2010 the World Health Organization (WHO) received reports of 18,449 laboratory-confirmed deaths from H1N1pdm09 infection [1]. This modest number has caused many to wonder what all the excitement was about, and some to question whether the pandemic response was excessive [2],[3]. But what is not widely appreciated is that the laboratory-confirmed total greatly underestimates the mortality burden, because only a minority of influenza-related deaths are ever definitively diagnosed as such. Additional influenza deaths result from secondary bacterial infections and exacerbation of preexisting chronic conditions, but are not recorded as being in any way related to influenza infection. Statistical methods are therefore used to separate the influenza-attributable fraction of deaths from the background [4],[5]. These methods involve modeling seasonal cyclical patterns in mortality time series compiled from vital statistics, often coupled with viral surveillance data to provide information on the timing of influenza circulation. Vital statistics data usually contain information about cause of death, allowing researchers to estimate influenza-attributable “excess” deaths in broad categories such as pneumonia, respiratory, or cardiorespiratory deaths during influenza periods. The influenza-related excess in respiratory deaths can be measured with higher precision than the less specific all-cause and cardiorespiratory categories. However, because some deaths triggered by influenza are recorded as having been caused by underlying non-respiratory causes such as heart attack, stroke, diabetes, or chronic kidney conditions, analysis of the broader cardiorespiratory and all-cause categories typically captures more completely the influenza-related burden. The majority of deaths from seasonal influenza occur among people aged 65 y or older, but in a pandemic the proportion of deaths among the young increases [4],[6]–[8]. Single-country studies of H1N1pdm09 mortality, using various cause-of-death outcomes and modeling techniques [9]–[14], have repeatedly documented such an age shift. For example, a comprehensive hospital-based sentinel surveillance study by Liang and colleagues found that in China only 4% of cases, 8% of hospitalizations, and 23% of pandemic deaths occurred in persons over 50 y of age [15]. On a global level, Van Kerkhove et al. reported a median age of 46 y among fatal laboratory-confirmed cases [16]. McCallum et al. reported that in the Western Pacific region during 2009, only 1% of laboratory-confirmed cases and 13% of laboratory-confirmed deaths were among persons 65 y of age or older [17]. The Global Pandemic Mortality (GLaMOR) project aimed to make a conservative estimate of the global H1N1pdm09 mortality burden in 2009 using statistical models applied to mortality, virology, and other available data. The project was funded by WHO, which requested global and regional estimates of H1N1pdm09 influenza deaths for the year 2009; thus, all mentions of pandemic flu mortality refer specifically to deaths that occurred in the last 9 mo of 2009. We invited global collaborators to contribute national mortality data detailed by week, age, and cause of death for 2005 through 2009, at minimum. Our novel method was inspired by a study that estimated the 1918–1920 global pandemic mortality burden using a two-stage statistical approach [18]. In Stage 1, we used detailed time series of national mortality and virology data from collaborating teams to estimate the 2009 pandemic respiratory mortality in each collaborating country/administrative region. Each participating GLaMOR team included influenza experts with whom the core GLaMOR team discussed Stage 1 mortality estimates in detail. The Stage 1 model and results were made available to each collaborating team to encourage individual country publications of national burden estimates. In Stage 2, we used a hierarchical multiple imputation model that used geographical, economic, and health country indicators to project the Stage 1 single-country estimates to all world countries, and summed to obtain regional and world estimates (Figure S2) . A previous study of global H1N1pdm09 mortality, published in 2012 by Dawood et al. [19], was conducted before 2009 mortality data became available. To overcome that problem, these authors implemented a probability model that took into account symptomatic attack rates and case fatality ratios measured in a set of wealthier countries, then used a “respiratory mortality multiplier” based on pre-2009 data to adjust for differences in respiratory disease fatality rates in different parts of the world. In contrast, we used national vital statistics data for 2009 to measure the actual pandemic mortality and its age patterns in countries representing ∼35% of the global population, then estimated the burden in the remaining countries using a novel projection method. As a result, while our global burden estimate is comparable to that of Dawood et al. [19], the regional pattern we found, with the Americas hit hard and Europe largely spared, corresponds more closely to what was reported as the pandemic unfolded. In addition, our method simultaneously generates pre-pandemic global seasonal influenza mortality estimates that we have presented here for comparison. METHODS Data Sources and Preparation We obtained weekly virology data from the WHO FluNet [20] to identify influenza active periods (some collaborators provided more detailed virology data). We created 3-wk moving averages of these data with a lag of 1 wk, to achieve the clinically observed lag of 1 wk from disease onset to death. We inspected the data for weeks in which the reported count dropped dramatically coincident with a major holiday; in three cases (Spain, Mexico, and Japan) we replaced the reported number with the average of positive counts in the two surrounding weeks. In most cases we encountered insufficient subtyping of influenza A viruses. Age-specific virology data were not available. We requested weekly national mortality time series based on the “underlying” cause-of-death determination from 1 January 2005 to 31 December 2009, stratified by at least two age groups (<65 and ≥65 y) and by four International Classification of Diseases–10 (ICD-10)–coded outcomes: all causes, cardiorespiratory (J and I codes), respiratory (J codes), and pneumonia and influenza (codes J10–J18) (Table 1). Contributing countries/administrative zones represented ∼35% of the global population. To prepare the time series mortality data for modeling, we created 3-wk moving averages after removing data in affected countries from summer weeks with a documented heat wave or armed conflict with significant mortality. We de-trended the time series using a

Read the complete text