Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Estimated global mortality associated with the first 12 months of 2009 pandemic influenza A H1N1 virus circulation: a modelling study

Estimated global mortality associated with the first 12 months of 2009 pandemic influenza A H1N1... Articles Estimated global mortality associated with the fi rst 12 months of 2009 pandemic infl uenza A H1N1 virus circulation: a modelling study Fatimah S Dawood, A Danielle Iuliano, Carrie Reed, Martin I Meltzer, David K Shay, Po-Yung Cheng, Don Bandaranayake, Robert F Breiman, W Abdullah Brooks, Philippe Buchy, Daniel R Feikin, Karen B Fowler, Aubree Gordon, Nguyen Tran Hien, Peter Horby, Q Sue Huang, Mark A Katz, Anand Krishnan, Renu Lal, Joel M Montgomery, Kåre Mølbak, Richard Pebody, Anne M Presanis, Hugo Razuri, Anneke Steens, Yeny O Tinoco, Jacco Wallinga, Hongjie Yu, Sirenda Vong, Joseph Bresee, Marc-Alain Widdowson Summary Lancet Infect Dis 2012; Background 18 500 laboratory-confi rmed deaths caused by the 2009 pandemic infl uenza A H1N1 were reported 12: 687–95 worldwide for the period April, 2009, to August, 2010. This number is likely to be only a fraction of the true number Published Online of the deaths associated with 2009 pandemic infl uenza A H1N1. We aimed to estimate the global number of deaths June 26, 2012 during the fi rst 12 months of virus circulation in each country. http://dx.doi.org/10.1016/ S1473-3099(12)70121-4 Methods We calculated crude respiratory mortality rates associated with the 2009 pandemic infl uenza A H1N1 strain This online publication has by age (0–17 years, 18–64 years, and >64 years) using the cumulative (12 months) virus-associated symptomatic attack been corrected. The corrected version fi rst rates from 12 countries and symptomatic case fatality ratios (sCFR) from fi ve high-income countries. To adjust crude appeared at thelancet.com/ mortality rates for diff erences between countries in risk of death from infl uenza, we developed a respiratory mortality infection on June 27, 2012 multiplier equal to the ratio of the median lower respiratory tract infection mortality rate in each WHO region mortality See Comment page 651 stratum to the median in countries with very low mortality. We calculated cardiovascular disease mortality rates Infl uenza Division, Centers for associated with 2009 pandemic infl uenza A H1N1 infection with the ratio of excess deaths from cardiovascular and Disease Control and respiratory diseases during the pandemic in fi ve countries and multiplied these values by the crude respiratory disease Prevention, Atlanta, GA, USA (F S Dawood MD, mortality rate associated with the virus. Respiratory and cardiovascular mortality rates associated with 2009 pandemic A D Iuliano PhD, C Reed DSc, infl uenza A H1N1 were multiplied by age to calculate the number of associated deaths. D K Shay MD, P-Y Cheng PhD, R Lal PhD, J M Montgomery PhD, Findings We estimate that globally there were 201 200 respiratory deaths (range 105 700–395 600) with an additional J Bresee MD, M-A Widdowson VetMB); 83 300 cardiovascular deaths (46 000–179 900) associated with 2009 pandemic infl uenza A H1N1. 80% of the Scientifi c and Program Services respiratory and cardiovascular deaths were in people younger than 65 years and 51% occurred in southeast Asia and Branch, Division of Africa. Preparedness and Emerging Infections, Centers for Disease Control and Prevention, Interpretation Our estimate of respiratory and cardiovascular mortality associated with the 2009 pandemic infl uenza Atlanta, GA, USA A H1N1 was 15 times higher than reported laboratory-confi rmed deaths. Although no estimates of sCFRs were (M I Meltzer PhD); National available from Africa and southeast Asia, a disproportionate number of estimated pandemic deaths might have Centre for Biosecurity and occurred in these regions. Therefore, eff orts to prevent infl uenza need to eff ectively target these regions in future Infectious Disease, Environmental Science and pandemics. Research Institute, Upper Hutt, New Zealand Funding None. (D Bandaranayake MBBS, Q S Huang PhD); Global Disease Detection Division, Kenya Introduction virological surveillance without standardised case re- Medical Research Infl uenza pandemics are typically characterised by porting or population denominators needed to estimate Institute/Centers for Disease higher, but widely varying, number of deaths than are incidence. Second, the level and timing of the circulation Control and Prevention, 1 3–5 seasonal epidemics. The emergence of pandemic of the pandemic virus might vary by region and country. Nairobi, Kenya (R F Breiman MD, D R Feikin MD, infl uenza A H1N1 in April, 2009, led WHO to Third, the severity of infl uenza might vary by region and M A Katz MD); International request that countries report all laboratory-confi rmed country due to diff erences in access to and quality of health Centre for Diarrhoeal Disease deaths associated with it. For the period up to August, care, nutritional status, prevalence of underlying chronic Research, Dhaka, Bangladesh 2010, 18 500 deaths associated with laboratory-confi rmed disorders, age distribution of the populations, and the use (W A Brooks MD); Johns 2 6–10 Hopkins Bloomberg School of 2009 pandemic infl uenza A H1N1 have been reported. of infl uenza vaccines and antiviral drugs. Public Health, Baltimore, MD, This number is likely to be an underestimate because Infl uenza-associated mortality is often estimated USA (W A Brooks); National diagnostic specimens are not always obtained from indirectly, by use of statistical models, as the number of Infl uenza Center, Phnom Penh, 11–13 Cambodia (P Buchy MD); people who die with infl uenza and the viruses might no excess deaths during periods of circulation of the virus. University of Alabama, longer be detectable by the time of death in some people. Infl uenza-associated deaths that might be missed by Birmingham, AL, USA Estimation of the 2009-pandemic-associated mortality direct counts of only laboratory-confi rmed deaths (K B Fowler DrPH); University of presents several challenges. First, data for infl uenza are taken into account with the use of these indirect California, Berkeley, CA, USA (A Gordon PhD); John E Fogarty in many countries are sparse and obtained through approaches. However, indirect estimation might not be www.thelancet.com/infection Vol 12 September 2012 687 Articles International Center, National as easy to apply in settings where infl uenza viruses of estimates of each parameter. We then estimated the Institutes of Health, Bethesda, circulate perennially without a clearly defi ned season or number of respiratory deaths associated with MD, USA (A Gordon); National where data for circulation with other respiratory viruses 2009 pandemic infl uenza A H1N1 for each age and Institute for Hygiene and are not available. Additionally, methods to estimate country-risk group by multiplying sAR, sCFR, RMM, and Epidemiology, Hanoi, Vietnam (N T Hien MD); Wellcome Trust excess mortality require the availability of vital statistics age group population. Major Overseas Programme, data that are sparse in some regions of the world and not Because pandemic H1N1 introduction was delayed in Oxford University Clinical always available in aggregate until 2–3 years after deaths some countries by up to 8 months after the emergence of Research Unit, Hanoi, Vietnam (P Horby MBBS); Centre for are reported in countries that do have data. Thus, new the virus in the Americas, our approach implicitly Community Medicine, All India methods are needed to produce timely and representative estimates the mortality attributable to the pandemic Institute of Medical Sciences, estimates of worldwide infl uenza-associated mortality. virus during the fi rst 12 months of circulation with yearly New Delhi, India A global estimate of the mortality associated with the cumulative attack rates in each country and not a (A Krishnan MBBS); Naval Medical Research Unit–6, Lima, 2009 pandemic infl uenza A H1N1 is needed to document contemporaneous complete calendar year. Peru ( J M Montgomery, the eff ect of the pandemic on the world’s population and Estimates of the cumulative sAR for the fi rst 12 months H Razuri MD, Y O Tinoco DVM); to help guide allocation and delivery of prevention and after virus introduction in each country were obtained Department of Epidemiology, treatment measures during future pandemics. So far, from 17 sites in 13 countries—Bangladesh, Cambodia, Statens Serum Institute, Copenhagen, Denmark most re ported estimates of mortality associated with Denmark, Germany, India, Kenya, the Netherlands, New (K Mølbak MD); Medical 2009 pandemic infl uenza A H1N1 are from high-income, Zealand, Nicaragua, Peru, the UK (England), the USA, Research Council Biostatistics 14–17 temperate countries. We developed a new approach to and Vietnam. Data from the site in Cambodia were Unit, Cambridge, UK estimate global mortality and the number of years of life only gathered during 5 months and therefore were not (R Pebody MBBS, A M Presanis PhD); Centre for lost (YLL) associated with the fi rst year of circulation of the included in the fi nal model. For most sites, sARs Infectious Disease Control, 2009 pandemic infl uenza A H1N1 virus in each country. were calculated with raw data gathered with a standard National Institute for Public template. Most sAR estimates were calculated from rates Health and the Environment, Methods of outpatient visits for infl uenza-associated infl uenza-like Bilthoven, Netherlands (A Steens MSc, J Wallinga PhD); Estimation of respiratory mortality rate illness divided by the proportion of individuals seeking Chinese Centers for Disease To calculate the base respiratory mortality rate associated medical care according to surveys of the use of health- Control and Prevention, with the 2009 pandemic infl uenza A H1N1 (unadjusted for care patterns for respiratory illness during the pandemic; Beijing, China (H Yu MD); and diff erences between regions in risk of death) for each or from prospective, active community-based surveillance Institut Pasteur, Phnom Penh, Cambodia (S Vong MD) country, we estimated the 12 month cumulative of respiratory illness with respiratory specimens that Correspondence to: symptomatic attack rate (sAR) with data from high- were obtained and tested for infl uenza viruses (detailed Dr Fatimah S Dawood or income, middle-income, and low-income countries and description of data sources and calculation methods is Dr Marc-Alain Widdowson, 14,19–28 multiplied it by the estimated symptomatic case fatality provided in appendix p 1). Estimates of the sARs were Infl uenza Division, Centers for ratio (sCFR) from select high-income countries. sAR was 4–33% for the paediatric age group, 0–22% for people Disease Control and Prevention, 1600 Clifton Road MS A-32, defi ned as the percentage of the population who developed aged 18–64 years, and 0–4% for those older than 64 years Atlanta, GA 30333, USA a symp tomatic respiratory illness associated with (table 1). We assumed that the true sAR of the fdawood@cdc.gov or laboratory-confi rmed 2009 pandemic infl uenza A H1N1. 2009 pandemic infl uenza A H1N1 virus in all age groups mwiddowson@cdc.gov sCFR was defi ned as the percentage of individuals with was greater than zero; therefore, we set the lower bound See Online for appendix symptomatic respiratory illness associated with laboratory- of the sAR for people aged 18–64 years and older than confi rmed 2009 pandemic infl uenza A H1N1 who died. 64 years as 1% in our model. Because of the huge variation We stratifi ed the estimates of sAR and sCFR into three in sARs within and between countries, we defi ned the age groups: 0–17 years (paediatric age group), 18–64 years, probability of any particular sAR value in the model with and greater than 64 years. We did not further stratify the a uniform distribution (ie, an equal probability of any paediatric age group largely because estimates for sAR value in the range). smaller strata lack precision. In previous pandemics, risk We obtained all available age-stratifi ed estimates of of death due to infl uenza has varied between countries, the 2009 pandemic infl uenza A H1N1 sCFR from six possibly because of diff erences in the underlying chronic studies in Denmark, the Netherlands, New Zealand, the 14,23–26 diseases and co-infections and access to antibiotics and UK, and the USA (two estimates). Sites in the UK, intensive care. Therefore, to adjust sCFR estimates for the USA, and Denmark were each adjusted for under- diff erences in risk of death from respiratory disease in reporting of deaths according to their own methods, low-income versus high-income countries, we stratifi ed whereas sites in the Netherlands and New Zealand countries into three risk groups. We adjusted the estimated sCFR based on laboratory-confi rmed, reported estimated base mortality in each risk group using a risk- deaths only. sCFR estimates were 0·002–0·013% for the group-specifi c respiratory mortality multiplier (RMM). paediatric age group, 0·018–0·159% for people aged The RMM was calculated by use of WHO’s estimates of 18–64 years, and 0·090–0·308% for those older than mortality rate for lower respiratory tract infection. Thus, 64 years (table 1). Because the estimates of sCFR had for each of the nine age–risk strata, we built a Monte wide and overlapping 95% CIs, we fi rst built a triangular Carlo probability model using SAS (version 9.2) with distribu tion for each sCFR by setting the median of the probability distributions defi ned according to the range triangular distribution as being equal to the point 688 www.thelancet.com/infection Vol 12 September 2012 Articles estimate and the minimum and maximum values were 0–17 years 18–64 years >64 years set as the fi fth and 95th percentiles (tenth and Countries with data inputs for symptomatic attack rate 90th percentiles for US estimates), respectively. We then Cambodia (Kapong Cham) *† 1% ·· ·· sampled each tri angular distribution an equal number of Bangladesh (Dhaka)*4% 0% 0% times to produce a sample of 10 000 points from which Denmark ‡§ 13% 4% 1% the fi nal model was sampled. As a sensitivity analysis, we Germany¶|| 32% 12% 2% re peated the same procedure using a uniform India (Ballabgarh)*|| 7% 4% 2% distribution for each sCFR estimate (results shown in Kenya (Kibera)*|| 11% 6% 0% appendix p 5). Kenya (Lwak)*|| 22% 22%** 0% Netherlands †† 13% 1% 0% RMM New Zealand †† 20% 7% 1% To adjust for diff erences in risk of death from respiratory Nicaragua (Managua) * 19% ·· ·· complications between countries, we developed an RMM 19,27 Peru (Cusco)*4% 1% 0% for all age groups with WHO’s 2008 country-specifi c 19,27 Peru (Lima) * 23% 7% 3% estimated mortality rates for lower respiratory tract 19,27 Peru (Madre de Dios) * 12% 2% 0% infection and WHO’s classifi cation of member states 19,27 into fi ve mortality strata—very low child and adult Peru (Tumbes) * 33% 5% 2% mortality; low child and adult mortality; low child and UK (England) ‡ 10% 3% 0% high adult mortality; high child and adult mortality; and USA ‡ 32% 12% 4% high child and very high adult mortality. We defi ned the Vietnam (Hanoi)*|| 5% 2% 0% RMM for countries in the lowest WHO mortality stratum Countries with data inputs for symptomatic case fatality ratio‡‡ as equal to 1. We then calculated mortality rate ratios for USA 0·005% (0·004–0·008) 0·027% (0·019–0·040) 0·092% (0·066–0·136) all other countries as the mortality rate associated with Denmark § 0·005% (0·002–0·013) 0·031–0·077%§§ ·· lower respiratory tract infection in a given country Netherlands 0·004% (0·003–0·006) 0·019–0·051%§§ ·· divided by the median mortality associated with lower New Zealand|| 0·002% (0·0003–0·005) 0·018% (0·012–0·024) 0·122% (0·024–0·220) respiratory tract infection from the lowest WHO 25 UK (England) 0·008% (0·004–0·013) 0·028% (0·012–0·047) 0·308% (0·057–1·069) mortality stratum (24 per 100 000 individuals). 24 USA (WI and NY) § 0·013% (0·005–0·033) 0·159% (0·066–0·333) 0·090% (0·008–1·471) Mortality rate ratios are shown by country in appendix References are for the general methods used for data gathering but might not include estimates for the period p 6. To minimise the eff ect of individual mortality rate reported in this table. *Based on population-based surveillance or cohort study of respiratory illness with laboratory ratios calculated with data of variable quality, we plotted confi rmation of 2009 pandemic infl uenza A H1N1 virus infection. †Data from Kampong Cham were only gathered the median mortality rate ratio for lower respiratory tract over 5 months; therefore, estimates of symptomatic attack rates from this site were not included in the fi nal model. ‡Based on statistical modelling by use of multiple data sources (see appendix p 1 for details). §Data were available for infection by WHO region and all-cause mortality stratum the following age groups: 0–14 years, 15–64 years, and greater than 64 years. ¶Derived from population-based (appendix p 11). We then assigned countries to one of three surveillance for medically attended respiratory illness adjusted with the estimated proportion of people who sought risk groups on the basis of the resulting distribution and care for respiratory illness. ||No available reference. **Outlier estimate excluded from the fi nal model. ††Based on estimate of seroprevalence of 2009 pandemic infl uenza A H1N1 virus infection adjusted with the estimated used the median mortality rate ratio from each group as proportion of people who were seropositive and had symptoms of respiratory illness. ‡‡90% CI reported for USA and the RMM for that group (with uncertainty defi ned by the 95% CI for all other estimates. §§Only 95% CI used. IQR for the mortality rate ratios). These risk groups were Table 1: Data inputs for estimation of mortality associated with 2009 pandemic infl uenza A H1N1 by all-African countries, non-African countries with high age group child and adult mortality, and all other countries (table 2). Non-African countries with high child and adult mortality included Afghanistan, Bangladesh, Bhutan, Bolivia, 10 000 estimates for each age–risk stratum to population Burma, Djibouti, Ecuador, Egypt, Guatemala, Haiti, India, estimates for each country in the RMM stratum to Iraq, Maldives, Morocco, Nepal, Nicaragua, North Korea, calculate the number of deaths by age in each country. Pakistan, Peru, Somalia, Sudan, Timor-Leste, and Yemen. The distribution of mortality-rate estimates for each age–risk stratum was highly right skewed with the lower Population estimates 75% of the estimates clustered within a narrow range of We used the UN’s projected population estimates for values and the upper 25% of estimates dispersed across 2010 when available, and the US Census Bureau mid- a wide range of values (appendix p 12 shows an example year population estimates for 2010 for all remaining of the distribution of estimates from one Monte Carlo countries. Kosovo, Niue, and Vatican City were excluded simula tion). We summed the median and 25% and 75% from this analysis because no available age-stratifi ed estimates for each country and age group to calculate a population estimates were available. global point estimate and range, respectively, by age group and WHO region. Estimated respiratory mortality We ran our Monte Carlo model 10 000 times to estimate Estimated cardiovascular mortality mortality for each of the nine age–risk strata (a total of Since infl uenza deaths can arise from respiratory or 90 000 iterations). We then applied each of the resulting cardio vascular complications, we also estimated www.thelancet.com/infection Vol 12 September 2012 689 Articles cardiovascular deaths due to 2009 pandemic H1N1 in and respiratory deaths attributed to the virus in fi ve people older than 17 years (infl uenza-associated high-income and middle-income countries stratifi ed by cardiovascular deaths are rare in children). First, we age 64 years and younger and older than 64 years. To calculated ratios of estimated cardiovascular to calculate the ratio of cardiovascular to respiratory deaths respiratory deaths associated with 2009 pandemic associated with the 2009 pandemic infl uenza A H1N1 infl uenza A H1N1 using estimates of excess circulatory virus, we used estimates of excess deaths attributable to this strain for Argentina, Brazil, Chile, Mexico, and the USA. The estimated ratios were 0·3–1·6 (median 0·7) in Minimum Maximum Median Distribution people aged 64 years and younger and 1·3–2·0 (1·8) in Symptomatic attack rate those older than 64 years (appendix p 14). 0–17 years 4% 33% 13% Uniform Second, we multiplied the median estimate of the 18–64 years 1% 12% 5% Uniform ratio of cardio vascular to respiratory deaths for each age >64 years 1% 4% 2% Uniform group by the base respiratory mortality rate in each Symptomatic case fatality ratio* country for people aged 18–64 years and older than 0–17 years 0·0003% 0·033% 0·005% Natural† 64 years to calculate the base cardiovascular mortality 18–64 years 0·012% 0·327% 0·029% Natural† rate associated with 2009 pandemic infl uenza A H1N1 >64 years 0·009% 1·449% 0·124% Natural† (unadjusted for diff erences between regions in risk of Mortality multiplier for lower respiratory tract infection cardiovascular death). We then assessed diff erences in African region countries, all ages 3 7 5 Uniform risk of death from cardiovascular disease between Non-African region countries with high child 1 4 3 Uniform countries by calcu lating mortality rate ratios from and adult mortality rates, all ages WHO’s estimates of cardiovascular disease mortality All other countries, all ages 1 1 1 Uniform rate based on the same method as for the RMM. Because the median mortality rate ratios for each region– *Minimum and maximum values for each age group represent those of the sample of 10 000 points from which the fi nal model was sampled. †Refers to the representative sample built from an equal number of samples selected from mortality stratum were 1–2 (appendix p 13), we did not each of the six sources for the estimates; fi rst, we built a triangular distribution setting the median of the triangular use a mortality multiplier for cardiovascular disease. distribution as equal to the point estimate and the minimum and maximum set to the fi fth and 95th percentiles, Thus, for each age group in each country, we estimated respectively; and second, we sampled each triangular distribution an equal number of times to produce a sample of the cardiovascular mortality rate associated with the 10 000 points from which the fi nal model sampled. 2009 pandemic infl uenza A H1N1 virus as the base Table 2: Parameters and probability distribution used in mortality model for 2009 pandemic respiratory mortality rate multiplied by the ratio of infl uenza A H1N1 cardiovascular to respiratory deaths and population. Reported* All ages 0–17 years (n [range†]) 18–64 years (n [range†]) >64 years (n [range†]) n (range†) Rate per 100 000 Respiratory deaths Africa 168 58 800 (30 800–112 200) 7·0 18 500 (9400–32 400) 35 900 (19 100–68 000) 4500 (2300–11 800) Americas ≥8533 17 500 (9400–34 500) 1·9 3000 (1500–5300) 11 400 (6200–20 900) 3000 (1700–8300) Eastern Mediterranean 1019 17 900 (9200–35 400) 3·0 4500 (2200–8500) 11 700 (6100–22 700) 1700 (800–4300) Europe ≥4879 16 400 (9000–33 400) 1·8 1800 (900–3100) 10 400 (5700–18 700) 4200 (2300–11 700) Southeast Asia 1992 59 500 (30 400–119 200) 3·3 12 400 (6100–23 500) 40 500 (21 000–78 500) 6700 (3300–17 200) Western Pacifi c 1858 31 100 (17 000–60 800) 1·7 4300 (2200–7400) 21 400 (11 800–38 700) 5300 (3000–14 800) Global‡ ≥18 449 201 200 (105 700–395 600) 2·9 44 500 (22 400–80 100) 131 300 (69 900–247 500) 25 400 (13 400–113 500) Respiratory and cardiovascular deaths§ Africa 168 65 600 (34 600–125 900) 7·8 18 500 (9400–32 400) 41 100 (22 000–77 300) 6100 (3200–16 200) Americas ≥8533 29 700 (16 200–61 500) 3·2 3000 (1500–5300) 18 600 (10 200–34 000) 8100 (4500–22 300) Eastern Mediterranean 1019 23 600 (12 300–47 100) 3·9 4500 (2200–8500) 16 000 (8500–30 400) 3100 (1600–8200) Europe ≥4879 31 300 (17 200–67 600) 3·5 1800 (900–3100) 17 600 (9700–31 800) 11 900 (6600–32 700) Southeast Asia 1992 78 600 (40 900–158 900) 4·4 12 400 (6100–23 500) 54 000 (28 400–103 000) 12 200 (6400–32 500) Western Pacifi c 1858 55 700 (30 600–114 500) 3·1 4300 (2200–7400) 36 400 (20 100–65 800) 15 000 (8300–41 300) Global‡ ≥18 449 284 400 (151 700–575 400) 4·1 44 500 (22 400–80 100) 183 700 (98 800–342 200) 56 400 (30 500–233 700) *Number of laboratory-confi rmed deaths due to 2009 pandemic infl uenza A H1N1 reported to WHO during April, 2009, to August 1, 2010. †The range was calculated by summing the 25th and 75th percentiles of estimates in each age group per country. ‡The total of the regional estimates is not always equal to the global estimate because of rounding. §Cardiovascular deaths were only calculated for people aged 18 years and older. Table 3: Reported and estimated respiratory and cardiovascular deaths associated with 2009 pandemic infl uenza A H1N1 by WHO region during the fi rst 12 months of virus circulation in each country and by age 690 www.thelancet.com/infection Vol 12 September 2012 Articles YLL for the period April, 2009, to August, 2010 (table 3). To further document the eff ect of the 2009 Summation of the fi fth and 95th percentile estimates in pan demic infl uenza A H1N1 that dispropor tionately each age group and country would have resulted in an aff ected young people compared with seasonal infl uenza estimated range of 39 000–1 315 800 respiratory deaths epidemics, we estimated the YLL from re spiratory and associated with 2009 pandemic infl uenza A H1N1. cardiovascular deaths associated with the virus as Results from the sensitivity analysis in which a uniform number of estimated deaths in each age group multiplied distribution was used instead of a triangular distribution by the mean number of additional years of life expected to sample from each sCFR range were broadly similar for people in each age group in each country. (appendix p 5). To demonstrate the eff ect of adjusting for The average numbers of years of life expected for diff erences in respiratory mortality, the calculation was people in each age group were obtained from WHO’s repeated without the RMM, which resulted in an estimate 2008 life tables. To show the diff erences in the mortality of 112 900 respiratory deaths (range 61 500–218 200) burden for age groups between the 2009 H1N1 pandemic associated with 2009 pandemic infl uenza A H1N1. and seasonal infl uenza epidemics, we estimated YLL that 29% of the 201 200 estimated respiratory deaths would have been lost if the age distribution of people associated with the 2009 pandemic infl uenza A H1N1 who died with 2009 pandemic infl uenza A H1N1 had occurred in the African region (table 3). The estimated been similar to the age distribution due to seasonal mortality rate in the African region was about two to four infl uenza. We assumed that during typical seasonal times that in countries elsewhere (table 3). The estimated infl uenza epidemics, 1% of deaths occurred in people range of respiratory deaths by country are shown in younger than 18 years, 9% in those aged 18–64 years, and appendix p 15. 33,34 90% in people older than 64 years. We then 65% of 2009 pandemic infl uenza A H1N1 deaths redistributed our pandemic deaths according to this worldwide were in individuals aged 18–64 years (60% of seasonal age distribution and estimated the YLL using global population), although the age distribution varied the same equation as above. by region. Overall, 13% of respiratory deaths associated with 2009 pandemic infl uenza A H1N1 were in people Role of the funding source older than 64 years (8% of global population). We received no external funding for the analysis. The An additional 83 300 cardiovascular deaths (range corresponding author had full access to all the data used 46 000–179 900) associated with the 2009 pandemic in the analysis and had fi nal responsibility for the infl uenza A H1N1 were estimated to have occurred in decision to submit for publication. people older than 17 years, resulting in a total of 284 400 respiratory and cardiovascular deaths (table 3). Results 20% of these deaths occurred in people older than The total of the median estimates of country-specifi c 64 years. With the inclusion of cardiovascular mortality, respiratory deaths associated with the 2009 there was a reduction in the disparity in mortality pan demic infl uenza A H1N1 was 201 200 (range associated with the pandemic by region, although the calculated by summing the 25th and 75th percentile mortality rate in Africa remained about two to three times estimates in each age group in each country), more than higher than elsewhere. Total numbers of respiratory and ten times the number of global deaths reported to WHO cardiovascular deaths by country are shown in appendix Deaths ≥1500 500–1499 200–499 50–199 <50 Data not available Figure 1: Global distribution of deaths associated with 2009 pandemic infl uenza A H1N1 during the fi rst year of virus circulation by country www.thelancet.com/infection Vol 12 September 2012 691 Articles Mortality per 100 000 8·1–9·5 6·1–8·0 4·1–6·0 2·1–4·0 0–2·0 Data not available Figure 2: Estimated age-adjusted respiratory and cardiovascular mortality rate associated with 2009 pandemic infl uenza A H1N1 per 100 000 individuals by country illness. Addition of cardiovascular deaths associated with Average life YLL due to pandemic infl uenza A YLL if age distribution expectancy at H1N1 respiratory mortality of deaths was similar 2009 pandemic infl uenza A H1N1 among people older birth (years) to seasonal infl uenza* than 17 years increased the mortality burden to Africa 53 2 278 800 (1 194 500–4 196 500) 556 100 151 700–575 400 deaths (table 3). Our global estimate was Americas 76 1 050 600 (567 600–1 965 100) 361 800 more than 15 times higher than the number of laboratory- Eastern Mediterranean 65 862 500 (446 100–1 645 900) 219 200 confi rmed deaths reported to WHO during the fi rst 16 months of the pandemic. A disproportionate number Europe 75 927 600 (506 800–1 756 400) 360 900 of total deaths from cardiovascular and respiratory Southeast Asia 65 2 725 300 (1 407 900–5 243 400) 738 700 diseases (51%) was estimated to have occurred in the Western Pacifi c 75 1 862 200 (1 015 800–3 445 500) 605 000 African and southeast Asian regions where 38% of the Global 68 9 707 000 (5 138 700–18 252 800) 2 841 700 world’s population live and where data for infl uenza Data are total of medians (range), unless otherwise indicated. YLL=years of life lost. *Number of 2009 pandemic 35 incidence are scarce. Additionally, most deaths were infl uenza A H1N1 deaths is redistributed across age groups to approximate the typical age distribution of seasonal reported in people aged 18–64 years, consistent with infl uenza deaths in developed countries where estimates were available (90% for people aged >64 years, 9% for age 18–64 years, and 1% for age 0–17 years). previous reports. These fi ndings are in contrast to those for seasonal infl uenza deaths. Roughly 80–90% of these Table 4: YLL because of deaths associated with 2009 pandemic infl uenza A H1N1 during the fi rst arise in people aged 65 years and older in Australia, 12 months of virus circulation and YLL that would be lost with a seasonal infl uenza age distribution of deaths Denmark, Singapore, and the USA, where age-stratifi ed 33,37–39 estimates are available. However, the age distribution p 19, and estimated H1N1-associated respiratory and of seasonal infl uenza-associated deaths might diff er cardiovascular deaths and mortality rates by country are between settings because of diff erences in the prevalence shown in fi gures 1 and 2. of underlying illnesses (such as HIV/AIDS and Estimated YLL were 9 707 000 during the fi rst 12 months tuberculosis) in younger adults in some low-income of the pandemic (table 4). Southeast Asia was the region countries. The shift in the age distribution of deaths with the greatest YLL (table 4). Total YLL attributable to during the pandemic resulted in substantially more YLL deaths associated with 2009 pandemic infl uenza A H1N1 than would have occurred if the age distribution of was 3·4 times higher than if the age distribution of deaths deaths had been similar to that of most seasonal had been similar to that during seasonal epidemics in infl uenza epidemics, consistent with obser vations from developed countries where age-specifi c estimates of previous pandemics (panel). seasonal infl uenza mortality are available (table 4). The Estimates from previous pandemics indicate that increase in YLL associated with 2009 pandemic infl uenza infl uenza mortality rates vary substantially between 9,47 A H1N1 was due to the increased rate of death in people coun tries. Our accounting for diff erential risk of aged 64 years and younger. infl uenza-associated death between countries is supported by the fi ndings of Cohen and colleagues that Discussion excess seasonal pneumonia and infl uenza mortality in During the fi rst year of circulation of the 2009 pandemic people aged 65 years and older is at least three times infl uenza A H1N1 virus in each country, an estimated higher in South Africa than in the USA. Further, data 105 700–395 600 people died of associated respiratory from the Americas and the western Pacifi c show that risk 692 www.thelancet.com/infection Vol 12 September 2012 Articles of death associated with the 2009 pandemic was up to six Panel: Research in context times higher in indigenous than in non-indigenous 8 6,8 populations. In both studies, the prevalence of Systematic review underlying disorders (including malnutrition in South We searched PubMed for reports of studies in any language from April, 2009, to March, Africa) and access to health care were postulated to 2012, with estimates of the number of deaths or mortality due to the 2009 pandemic contribute to the increased risk of infl uenza-associated infl uenza A H1N1. We identifi ed 12 studies from eight countries in Europe, the Americas, 14,17,23,25,39–46 42 death. Nair and colleagues also estimated that mortality western Pacifi c, and Asia. In two studies from Australia (New South Wales) and rates for infl uenza-associated acute lower respiratory the UK, no deaths were estimated based on the excess all-cause mortality during periods tract infection in children younger than 5 years are three of infl uenza virus circulation. Two other UK studies were reported—in the England-only times higher in low-income countries than in high- study, the estimated number of deaths was 390–490 and in the study of only England income countries. and Wales the estimate was 1556. In one of the two studies from Mexico, the estimated WHO has estimated that an average of 0·004–0·008% mortality was 35 per 100 000 people in San Luis Potosi, whereas in the other study the per year of the world’s population (250 000–500 000 peo- estimate was 6200 deaths for Mexico based on excess pneumonia and infl uenza ple) die as a result of seasonal infl uenza. Estimates of mortality or 26 500 deaths based on excess all-cause mortality. According to the pandemic infl uenza mortality ranged from 0·03% of the estimates from two studies that included deaths in 2009 only, there were 6000 deaths in world’s population during the 1968 pandemic to 1–3% of Bangladesh and 0–256 deaths in Hong Kong, equal to an age-standardised mortality of 7,9,47 the world’s population during the 1918 pandemic. We 2·9–14·8 per 100 000 individuals. According to the estimates from the remaining studies, 14 23,39 estimate that 0·001–0·007% of the world’s population there were 8868–18 306 deaths in the USA, 121 or 151–473 in Denmark, and (n=6 891 433 594) died of respiratory complications 266–958 in the Netherlands. associated with 2009 pandemic infl uenza A H1N1 during Interpretation the fi rst year of virus circulation or 0·001–0·011% when Our estimated ranges of deaths associated with 2009 pandemic infl uenza A H1N1 for cardiovascular deaths were included. However, our each country overlap with reported estimates from Denmark (109–438 vs 151–473 or estimates are not directly comparable to those of 121), the Netherlands (326–1289 vs 266–958), the USA (5834–22 697 vs 8868–18 306), mortality associated with seasonal and previous China (Hong Kong only; 1·7–6·2 per 100 000 vs 2·9–14·8 per 100 000), and Bangladesh pandemic infl uenza for at least two reasons. First, the (3899–15 135 vs 6000). However, our estimated ranges are higher than most estimates WHO estimate of global seasonal infl uenza mortality is from the UK (1237–4946 vs 0 for the UK or 390–490 for England or 1556 for England and an extrapolation of esti mates from high-income 48 Wales), and Australia (406–1589 vs mortality rate of 0 on the basis of excess all-cause countries without accounting for regional variation in deaths in the Hunter New England region) and lower than estimates from Mexico risk of death from outcomes that might be associated (1670–6105 vs 6200 excess pneumonia and infl uenza deaths or 26 000 excess all-cause with infl uenza, though details of the methods have not deaths). Diff erences in study methods must be taken into account when estimates from been provided. Second, reported estimates of global these studies are compared. The results of our study add to the understanding of the pandemic mortality include data from several years of global eff ect of the 2009 pandemic infl uenza A H1N1 because we have estimated the pandemic virus circulation, whereas our estimate associated deaths in African and southeast Asian countries for which there is only one includes data for only the fi rst year of 2009 pandemic reported estimate from Bangladesh so far. We show that half of all global infl uenza A H1N1 virus circulation in each country. H1N1-associated respiratory and cardiovascular deaths could have occurred in Africa and We were unable to identify any factor or group of southeast Asia and that mortality rates might have been two to three times higher in factors that allowed stratifi cation of countries into Africa than in other regions of the world. transmission risk groups, and therefore we assumed the same range of 2009 pandemic infl uenza A H1N1 sARs for all countries. This approach results in point estimates fi ndings that the 2009 pandemic infl uenza A H1N1 attack that underestimate or overestimate deaths in countries rates varied substantially both within and between coun- 49–54 that had very high or very low transmission, respectively, tries. Second, we relied on only six estimates of but should produce ranges for most country-specifi c sCFR from high-income countries because none were estimates that are likely to capture the true number of available from low-income or middle-income countries deaths in each country. For this reason, we present only including China and India, which account for 37% of the ranges for our country-specifi c estimates. The world’s population. Although the available estimates of assumption of a range of 2009 pandemic infl uenza A sCFR were each calculated with diff erent methods for H1N1 sARs for all countries is less likely to bias our ascertaining the numerator (number of deaths associated regional and global estimates. with 2009 pandemic infl uenza A H1N1) and denominator Our model was limited by the scarcity of globally (number of symptomatic 2009 pandemic infl uenza A representative estimates of sAR and sCFR. We relied on H1N1 virus infections), most of these estimates were only 16 direct estimates of 1 year sARs for 2009 infl uenza fairly similar. However, each sCFR estimate had a A H1N1 that varied greatly, and were likely to be aff ected substantial amount of inherent uncertainty, particularly by an unidentifi ed combination of factors such as for the older than 64 years age group in which the diff erences in climate, population density, and popu- denominator for the sCFR was lower than in other age lation age structure. Data from serology studies of both groups. The uncertainty of the individual sCFR estimates symptomatic and asymptomatic infections support our widened the confi dence intervals of our fi nal estimates. www.thelancet.com/infection Vol 12 September 2012 693 Articles The lack of infl uenza sCFR or mortality rate estimates continued eff orts to strengthen infl uenza surveillance worldwide, particularly for infl uenza-associated mor- for low-income and middle-income countries is an tality, are needed both to guide seasonal infl uenza important knowledge gap in the understanding of the prevention strategies and to build infl uenza surveillance epidemiology of both seasonal and pandemic infl uenza. systems to provide better and more timely and globally To overcome the lack of data to inform sCFR estimates for representative data for infl uenza-associated mortality these settings, we developed an RMM from mortality rates during future pandemics. for lower respiratory tract infection that are subject to several limitations. First, mortality rates consist of both Contributors FSD, ADI, M-AW, DKS, MIM, JB, and CR all actively contributed to the disease incidence and case fatality rates. Thus, the RMM is design of the analysis; CR, MIM, DKS, P-YC, DB, RFB, WAB, PB, DRF, aff ected by diff erences in the incidence of non-infl uenza KBF, AG, NTH, PH, QSH, MAK, AK, RL, JMM, KM, RP, AMP, HR, AS, respiratory infections, and the dearth of estimates of YOT, JW, HY, and SV provided the data used in this analysis. FSD and incidence of lower respiratory tract infection for most ADI did the data analysis. FSD wrote the fi rst draft of the report, and all authors contributed to the interpretation of the results of the analysis countries prevented us from adjusting for diff erences in and to the revision and fi nal preparation of the report for submission. incidence. Second, we used one RMM for all age groups Confl icts of interest because the age-stratifi ed rate ratios in the 0–17 year group We declare that we have no confl icts of interest. were up to 58 times higher than the all-age ratios, probably Acknowledgments due to a larger diff erence in incidence of non-infl uenza We thank WHO’s Ad-Hoc Working Group on Pandemic Mortality respiratory infections in this age group. By contrast, the Burden for valuable insights; Patrick Glew at the Oakridge Institute of all-age rate ratios were similar to ratios in the older age Science and Education and the Centers for Disease Control and groups that had the most number of deaths associated Prevention for organisation of data used to calculate the mortality multiplier; the Dutch Ministry of Health, Welfare, and Sport for their with the 2009 pandemic infl uenza A H1N1. Because the support of the study that resulted in data from the Netherlands and relative risk of infl uenza-associated re spiratory death is M Van Boven of the Centre for Infectious Disease Control, National likely to vary between age groups, use of one all-age RMM Institute for Public Health and the Environment, Bilthoven, Netherlands is likely to result in an underestimation of virus-associated for his work on the study that resulted in data from the Netherlands; Angel Balmaseda, Guillermina Kuan, and Eva Harris for their work on mortality in children in low-income countries. obtaining data from Nicaragua; Ernesto Ortiz, Abel Estela, One additional potential methodological limitation of Maria Luisa Morales, Judith Patricia Breña, and Candice Romero for our analysis is that we summed the median and IQR their operational management and supervision at the surveillance sites estimates for each age group in each country to calculate in Peru, all from the Naval Medical Research Unit-6, Lima; and Udo Buchholz for his contribution of data used to estimate the a point estimate and range of the total number of deaths symptomatic attack rate in Germany. The fi ndings and conclusions in by country, region, and globally. The assumption with this report are those of the authors and do not necessarily represent the this approach is that symptomatic attack rates are views of the Centers for Disease Control and Prevention. positively correlated between age groups in each country References (ie, it is unlikely for one age group to have an attack rate 1 Kasowski EJ, Garten RJ, Bridges CB. Infl uenza pandemic epidemiologic and virologic diversity: reminding ourselves of the on the higher end of the global range of sARs and possibilities. Clin Infect Dis 2011; 52 (suppl 1): S44–49. another age group to have an attack rate on the lower 2 WHO. Pandemic (H1N1) 2009–update 112. www.who.int/csr/ end of the global range). Had we summed the result of don/2010_08_06/en/index.html (accessed Feb 24, 2011). each Monte Carlo iteration for each age group in each 3 Michalove J, Ndahwouh T, Diop O, et al, eds. Delayed pandemic H1N1 transmission in West Africa, May 2009–March 2010. Options for the country, then summed the results for each country, and Control of Infl uenza VII Conference; Hong Kong; Sept 3–7, 2010. taken the median and IQR of the distribution of the 4 Miller E, Hoschler K, Hardelid P, Stanford E, Andrews N, global estimates, our estimate would have been 25% Zambon M. Incidence of 2009 pandemic infl uenza A H1N1 infection in England: a cross-sectional serological study. Lancet higher. A fi nal potential limitation is that we assumed 2010; 375: 1100–08. that the life expectancy of people who died with 5 Ortiz JR, Lafond KE, Wong TA, Uyeki TM. Pandemic infl uenza in 2009 pandemic infl uenza A H1N1 was equal to that of Africa, lessons learned from 1968: a systematic review of the literature. Infl uenza Other Respi Viruses 2012; 6: 11–24. people of the same age in the general population, 6 Cohen C, Simonsen L, Kang JW, et al. Elevated infl uenza-related although people who die from infl uenza might be more excess mortality in South African elderly individuals, 1998–2005. likely to have underlying disorders that are associated Clin Infect Dis 2010; 51: 1362–69. with a reduced life expectancy. 7 Johnson NP, Mueller J. Updating the accounts: global mortality of the 1918–1920 “Spanish” infl uenza pandemic. Bull Hist Med 2002; We estimated that 151 700–575 400 respiratory and 76: 105–15. cardiovascular deaths associated with 2009 pan demic 8 La Ruche G, Tarantola A, Barboza P, Vaillant L, Gueguen J, infl uenza A H1N1 occurred during the fi rst year of virus Gastellu-Etchegorry M. The 2009 pandemic H1N1 infl uenza and indigenous populations of the Americas and the Pacifi c. circulation in every country in the world. Our fi ndings Euro Surveill 2009; 14: 19366. emphasise the need to improve the global response to 9 Murray CJ, Lopez AD, Chin B, Feehan D, Hill KH. Estimation of future infl uenza pandemics and expand production and potential global pandemic infl uenza mortality on the basis of vital registry data from the 1918–20 pandemic: a quantitative analysis. improve delivery of infl uenza vaccines to Africa and Lancet 2006; 368: 2211–18. southeast Asia because these countries might have 10 Nair H, Brooks WA, Katz M, et al. Global burden of respiratory borne a disproportionate burden of pandemic mortality infections due to seasonal infl uenza in young children: a systematic review and meta-analysis. Lancet 2011; 378: 1917–30. during the fi rst year of virus circulation. Additionally, 694 www.thelancet.com/infection Vol 12 September 2012 Articles 11 Thompson WW, Weintraub E, Dhankhar P, et al. Estimates of US 33 Thompson WW, Shay DK, Weintraub E, et al. Mortality associated infl uenza-associated deaths made using four diff erent methods. with infl uenza and respiratory syncytial virus in the United States. Infl uenza Other Respi Viruses 2009; 3: 37–49. JAMA 2003; 289: 179–86. 12 Serfl ing RE. Methods for current statistical analysis of excess 34 Estimates of deaths associated with seasonal infl uenza—United pneumonia-infl uenza deaths. Public Health Rep 1963; 78: 494–506. States, 1976–2007. MMWR Morb Mortal Wkly Rep 2010; 59: 1057–62. 13 Simonsen L, Clarke MJ, Stroup DF, Williamson GD, Arden NH, 35 Yazdanbakhsh M, Kremsner PG. Infl uenza in Africa. PLoS Med Cox NJ. A method for timely assessment of infl uenza-associated 2009; 6: e1000182. mortality in the United States. Epidemiology 1997; 8: 390–95. 36 Vaillant L, La Ruche G, Tarantola A, Barboza P. Epidemiology of 14 Shrestha SS, Swerdlow DL, Borse RH, et al. Estimating the fatal cases associated with pandemic H1N1 infl uenza 2009. burden of 2009 pandemic infl uenza A (H1N1) in the United States Euro Surveill 2009; 14: 19309. (April 2009–April 2010). Clin Infect Dis 2011; 52 (suppl 1): S75–82. 37 Chow A, Ma S, Ling AE, Chew SK. Infl uenza-associated deaths in 15 Viboud C, Miller M, Olson D, Osterholm M, Simonsen L. tropical Singapore. Emerg Infect Dis 2006; 12: 114–21. Preliminary estimates of mortality and years of life lost associated 38 Newall AT, Wood JG, Macintyre CR. Infl uenza-related with the 2009 A/H1N1 pandemic in the US and comparison with hospitalisation and death in Australians aged 50 years and older. past infl uenza seasons. PLoS Curr 2010: RRN1153. Vaccine 2008; 26: 2135–41. 16 Wielders CC, van Lier EA, van ‘t Klooster TM, et al. The burden of 39 Nielsen J, Mazick A, Glismann S, Molbak K. Excess mortality 2009 pandemic infl uenza A(H1N1) in the Netherlands. related to seasonal infl uenza and extreme temperatures in Eur J Public Health 2012; 22: 150–57. Denmark, 1994–2010. BMC Infect Dis 2011; 11: 350. 17 Charu V, Chowell G, Palacio Mejia LS, et al. Mortality burden of the 40 Comas-Garcia A, Garcia-Sepulveda CA, Mendez-de Lira JJ, A/H1N1 pandemic in Mexico: a comparison of deaths and years of Aranda-Romo S, Hernandez-Salinas AE, Noyola DE. Mortality life lost to seasonal infl uenza. Clin Infect Dis 2011; 53: 985–93. attributable to pandemic infl uenza A (H1N1) 2009 in San Luis 18 WHO. Disease and injury country estimates. 2011. http://www.who. Potosi, Mexico. Infl uenza Other Respi Viruses 2011; 5: 76–82. int/healthinfo/global_burden_disease/estimates_country/en/index. 41 Hardelid P, Andrews N, Pebody R. Excess mortality monitoring in html (accessed Feb 11, 2011). England and Wales during the infl uenza A(H1N1) 2009 pandemic. 19 Unit-6 NMR. Burden of illness and risk factors for transmission of Epidemiol Infect 2011; 139: 1431–39. seasonal infl uenza in four distinct regions of Peru. http:// 42 Muscatello DJ, Cretikos M, Macintyre CR. All-cause mortality proyectoinfl uenza.pe (accessed Aug 8, 2011). during fi rst wave of pandemic (H1N1) 2009, New South Wales, 20 Bandaranayake D, Huang QS, Bissielo A, et al. Risk factors and Australia, 2009. Emerg Infect Dis 2010; 16: 1396–402. immunity in a nationally representative population following the 43 Wijngaard CC, Asten L, Koopmans MP, et al. Comparing pandemic 2009 infl uenza A(H1N1) pandemic. PLoS One 2010; 5: e13211. to seasonal infl uenza mortality: moderate impact overall but high 21 Brooks WA, Goswami D, Rahman M, et al. Infl uenza is a major mortality in young children. PLoS One 2012; 7: e31197. contributor to childhood pneumonia in a tropical developing 44 Hardelid P, Pebody R, Andrews N. Mortality caused by infl uenza country. Pediatr Infect Dis J 2010; 29: 216–21. and respiratory syncytial virus by age group in England and Wales 22 Kuan G, Gordon A, Aviles W, et al. The Nicaraguan pediatric 1999–2010. Infl uenza Other Respi Viruses 2012; published online dengue cohort study: study design, methods, use of information March 9. DOI:10.1111/j.1750-2659.2012.00345.x. technology, and extension to other infectious diseases. 45 Homaira N, Luby SP, Alamgir A, et al. Infl uenza-associated Am J Epidemiol 2009; 170: 120–29. mortality in 2009 in four sentinel sites in Bangladesh. 23 Molbak K, Widgren K, Jensen KS, et al. Burden of illness of the Bull World Health Organ 2012; 90: 272–78. 2009 pandemic of infl uenza A (H1N1) in Denmark. Vaccine 2011; 46 Yang L, Chan KP, Cowling BJ, et al. Excess mortality associated with 29 (suppl 2): B63–69. the 2009 pandemic of infl uenza A(H1N1) in Hong Kong. 24 Presanis AM, De Angelis D, Hagy A, et al. The severity of pandemic Epidemiol Infect 2011. DOI:10.1017/S0950268811002238. H1N1 infl uenza in the United States, from April to July 2009: a 47 Mathews JD, Chesson JM, McCaw JM, McVernon J. Understanding Bayesian analysis. PLoS Med 2009; 6: e1000207. infl uenza transmission, immunity and pandemic threats. 25 Presanis AM, Pebody RG, Paterson BJ, et al. Changes in severity of Infl uenza Other Respi Viruses 2009; 3: 143–49. 2009 pandemic A/H1N1 infl uenza in England: a Bayesian evidence 48 WHO. Infl uenza (seasonal). 2009. http://www.who.int/ synthesis. BMJ 2011; 343: d5408. mediacentre/factsheets/fs211/en/ (accessed June 4, 2011). 26 Steens A, Waaijenborg S, Teunis PF, et al. Age-dependent patterns 49 Chan YJ, Lee CL, Hwang SJ, et al. Seroprevalence of antibodies to of infection and severity explaining the low impact of pandemic (H1N1) 2009 infl uenza virus among hospital staff in a 2009 infl uenza A (H1N1): evidence from serial serologic surveys in medical center in Taiwan. J Chin Med Assoc 2010; 73: 62–66. the Netherlands. Am J Epidemiol 2011; 174: 1307–15. 50 Chen MI, Lee VJ, Lim WY, et al. 2009 infl uenza A(H1N1) 27 Tinoco Y, Razuri H, Ortiz EJ, et al. Preliminary population-based seroconversion rates and risk factors among distinct adult cohorts epidemiological and clinical data on 2009 pandemic H1N1 infl uenza in Singapore. JAMA 2010; 303: 1383–91. A (pH1N1) from Lima, Peru. Infl uenza Other Respi Viruses 2009; 51 Moghadami M, Moattari A, Tabatabaee HR, et al. High titers of 3: 253–56. hemagglutination inhibition antibodies against 2009 H1N1 28 Vong S, Goyet S, Ly S, et al. Under-recognition and reporting of infl uenza virus in Southern Iran. Iran J Immunol 2010; 7: 39–48. dengue in Cambodia: a capture-recapture analysis of the National 52 Tsai TF, Pedotti P, Hilbert A, et al. Regional and age-specifi c Dengue Surveillance System. Epidemiol Infect 2011; 140: 491–99. patterns of pandemic H1N1 infl uenza virus seroprevalence inferred 29 WHO. List of member states by WHO region and mortality from vaccine clinical trials, August-October 2009. Euro Surveill 2010; stratum. http://www.who.int/whr/2003/en/member_ 15: 19624. states_182-184_en.pdf (accessed Oct 11, 2010). 53 Zimmer SM, Crevar CJ, Carter DM, et al. Seroprevalence following 30 UN. Department of Economic and Social Aff airs, Population the second wave of Pandemic 2009 H1N1 infl uenza in Pittsburgh, Division, Population Estimates and Projections Section. http://esa. PA, USA. PLoS One 2010; 5: e11601. un.org/unpd/wpp/unpp/panel_population.htm (accessed Feb 30, 54 Gilbert GL, Cretikos MA, Hueston L, Doukas G, O’Toole B, 2011). Dwyer DE. Infl uenza A (H1N1) 2009 antibodies in residents of New 31 US Census Bureau International Data Base, Population Division. South Wales, Australia, after the fi rst pandemic wave in the US Census Bureau. 2010. http://www.census.gov/ipc/www/idb/ 2009 southern hemisphere winter. PLoS One 2010; 5: e12562. region.php (Feb 11, 2011). 32 WHO. Life tables for WHO member states. 2011. http://www.who. int/healthinfo/statistics/mortality_life_tables/en/index.html (Feb 11, 2011). www.thelancet.com/infection Vol 12 September 2012 695 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Lancet Infectious Diseases Unpaywall

Loading next page...
 
/lp/unpaywall/estimated-global-mortality-associated-with-the-first-12-months-of-2009-YMu8AA42iI

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
1473-3099
DOI
10.1016/s1473-3099(12)70121-4
Publisher site
See Article on Publisher Site

Abstract

Articles Estimated global mortality associated with the fi rst 12 months of 2009 pandemic infl uenza A H1N1 virus circulation: a modelling study Fatimah S Dawood, A Danielle Iuliano, Carrie Reed, Martin I Meltzer, David K Shay, Po-Yung Cheng, Don Bandaranayake, Robert F Breiman, W Abdullah Brooks, Philippe Buchy, Daniel R Feikin, Karen B Fowler, Aubree Gordon, Nguyen Tran Hien, Peter Horby, Q Sue Huang, Mark A Katz, Anand Krishnan, Renu Lal, Joel M Montgomery, Kåre Mølbak, Richard Pebody, Anne M Presanis, Hugo Razuri, Anneke Steens, Yeny O Tinoco, Jacco Wallinga, Hongjie Yu, Sirenda Vong, Joseph Bresee, Marc-Alain Widdowson Summary Lancet Infect Dis 2012; Background 18 500 laboratory-confi rmed deaths caused by the 2009 pandemic infl uenza A H1N1 were reported 12: 687–95 worldwide for the period April, 2009, to August, 2010. This number is likely to be only a fraction of the true number Published Online of the deaths associated with 2009 pandemic infl uenza A H1N1. We aimed to estimate the global number of deaths June 26, 2012 during the fi rst 12 months of virus circulation in each country. http://dx.doi.org/10.1016/ S1473-3099(12)70121-4 Methods We calculated crude respiratory mortality rates associated with the 2009 pandemic infl uenza A H1N1 strain This online publication has by age (0–17 years, 18–64 years, and >64 years) using the cumulative (12 months) virus-associated symptomatic attack been corrected. The corrected version fi rst rates from 12 countries and symptomatic case fatality ratios (sCFR) from fi ve high-income countries. To adjust crude appeared at thelancet.com/ mortality rates for diff erences between countries in risk of death from infl uenza, we developed a respiratory mortality infection on June 27, 2012 multiplier equal to the ratio of the median lower respiratory tract infection mortality rate in each WHO region mortality See Comment page 651 stratum to the median in countries with very low mortality. We calculated cardiovascular disease mortality rates Infl uenza Division, Centers for associated with 2009 pandemic infl uenza A H1N1 infection with the ratio of excess deaths from cardiovascular and Disease Control and respiratory diseases during the pandemic in fi ve countries and multiplied these values by the crude respiratory disease Prevention, Atlanta, GA, USA (F S Dawood MD, mortality rate associated with the virus. Respiratory and cardiovascular mortality rates associated with 2009 pandemic A D Iuliano PhD, C Reed DSc, infl uenza A H1N1 were multiplied by age to calculate the number of associated deaths. D K Shay MD, P-Y Cheng PhD, R Lal PhD, J M Montgomery PhD, Findings We estimate that globally there were 201 200 respiratory deaths (range 105 700–395 600) with an additional J Bresee MD, M-A Widdowson VetMB); 83 300 cardiovascular deaths (46 000–179 900) associated with 2009 pandemic infl uenza A H1N1. 80% of the Scientifi c and Program Services respiratory and cardiovascular deaths were in people younger than 65 years and 51% occurred in southeast Asia and Branch, Division of Africa. Preparedness and Emerging Infections, Centers for Disease Control and Prevention, Interpretation Our estimate of respiratory and cardiovascular mortality associated with the 2009 pandemic infl uenza Atlanta, GA, USA A H1N1 was 15 times higher than reported laboratory-confi rmed deaths. Although no estimates of sCFRs were (M I Meltzer PhD); National available from Africa and southeast Asia, a disproportionate number of estimated pandemic deaths might have Centre for Biosecurity and occurred in these regions. Therefore, eff orts to prevent infl uenza need to eff ectively target these regions in future Infectious Disease, Environmental Science and pandemics. Research Institute, Upper Hutt, New Zealand Funding None. (D Bandaranayake MBBS, Q S Huang PhD); Global Disease Detection Division, Kenya Introduction virological surveillance without standardised case re- Medical Research Infl uenza pandemics are typically characterised by porting or population denominators needed to estimate Institute/Centers for Disease higher, but widely varying, number of deaths than are incidence. Second, the level and timing of the circulation Control and Prevention, 1 3–5 seasonal epidemics. The emergence of pandemic of the pandemic virus might vary by region and country. Nairobi, Kenya (R F Breiman MD, D R Feikin MD, infl uenza A H1N1 in April, 2009, led WHO to Third, the severity of infl uenza might vary by region and M A Katz MD); International request that countries report all laboratory-confi rmed country due to diff erences in access to and quality of health Centre for Diarrhoeal Disease deaths associated with it. For the period up to August, care, nutritional status, prevalence of underlying chronic Research, Dhaka, Bangladesh 2010, 18 500 deaths associated with laboratory-confi rmed disorders, age distribution of the populations, and the use (W A Brooks MD); Johns 2 6–10 Hopkins Bloomberg School of 2009 pandemic infl uenza A H1N1 have been reported. of infl uenza vaccines and antiviral drugs. Public Health, Baltimore, MD, This number is likely to be an underestimate because Infl uenza-associated mortality is often estimated USA (W A Brooks); National diagnostic specimens are not always obtained from indirectly, by use of statistical models, as the number of Infl uenza Center, Phnom Penh, 11–13 Cambodia (P Buchy MD); people who die with infl uenza and the viruses might no excess deaths during periods of circulation of the virus. University of Alabama, longer be detectable by the time of death in some people. Infl uenza-associated deaths that might be missed by Birmingham, AL, USA Estimation of the 2009-pandemic-associated mortality direct counts of only laboratory-confi rmed deaths (K B Fowler DrPH); University of presents several challenges. First, data for infl uenza are taken into account with the use of these indirect California, Berkeley, CA, USA (A Gordon PhD); John E Fogarty in many countries are sparse and obtained through approaches. However, indirect estimation might not be www.thelancet.com/infection Vol 12 September 2012 687 Articles International Center, National as easy to apply in settings where infl uenza viruses of estimates of each parameter. We then estimated the Institutes of Health, Bethesda, circulate perennially without a clearly defi ned season or number of respiratory deaths associated with MD, USA (A Gordon); National where data for circulation with other respiratory viruses 2009 pandemic infl uenza A H1N1 for each age and Institute for Hygiene and are not available. Additionally, methods to estimate country-risk group by multiplying sAR, sCFR, RMM, and Epidemiology, Hanoi, Vietnam (N T Hien MD); Wellcome Trust excess mortality require the availability of vital statistics age group population. Major Overseas Programme, data that are sparse in some regions of the world and not Because pandemic H1N1 introduction was delayed in Oxford University Clinical always available in aggregate until 2–3 years after deaths some countries by up to 8 months after the emergence of Research Unit, Hanoi, Vietnam (P Horby MBBS); Centre for are reported in countries that do have data. Thus, new the virus in the Americas, our approach implicitly Community Medicine, All India methods are needed to produce timely and representative estimates the mortality attributable to the pandemic Institute of Medical Sciences, estimates of worldwide infl uenza-associated mortality. virus during the fi rst 12 months of circulation with yearly New Delhi, India A global estimate of the mortality associated with the cumulative attack rates in each country and not a (A Krishnan MBBS); Naval Medical Research Unit–6, Lima, 2009 pandemic infl uenza A H1N1 is needed to document contemporaneous complete calendar year. Peru ( J M Montgomery, the eff ect of the pandemic on the world’s population and Estimates of the cumulative sAR for the fi rst 12 months H Razuri MD, Y O Tinoco DVM); to help guide allocation and delivery of prevention and after virus introduction in each country were obtained Department of Epidemiology, treatment measures during future pandemics. So far, from 17 sites in 13 countries—Bangladesh, Cambodia, Statens Serum Institute, Copenhagen, Denmark most re ported estimates of mortality associated with Denmark, Germany, India, Kenya, the Netherlands, New (K Mølbak MD); Medical 2009 pandemic infl uenza A H1N1 are from high-income, Zealand, Nicaragua, Peru, the UK (England), the USA, Research Council Biostatistics 14–17 temperate countries. We developed a new approach to and Vietnam. Data from the site in Cambodia were Unit, Cambridge, UK estimate global mortality and the number of years of life only gathered during 5 months and therefore were not (R Pebody MBBS, A M Presanis PhD); Centre for lost (YLL) associated with the fi rst year of circulation of the included in the fi nal model. For most sites, sARs Infectious Disease Control, 2009 pandemic infl uenza A H1N1 virus in each country. were calculated with raw data gathered with a standard National Institute for Public template. Most sAR estimates were calculated from rates Health and the Environment, Methods of outpatient visits for infl uenza-associated infl uenza-like Bilthoven, Netherlands (A Steens MSc, J Wallinga PhD); Estimation of respiratory mortality rate illness divided by the proportion of individuals seeking Chinese Centers for Disease To calculate the base respiratory mortality rate associated medical care according to surveys of the use of health- Control and Prevention, with the 2009 pandemic infl uenza A H1N1 (unadjusted for care patterns for respiratory illness during the pandemic; Beijing, China (H Yu MD); and diff erences between regions in risk of death) for each or from prospective, active community-based surveillance Institut Pasteur, Phnom Penh, Cambodia (S Vong MD) country, we estimated the 12 month cumulative of respiratory illness with respiratory specimens that Correspondence to: symptomatic attack rate (sAR) with data from high- were obtained and tested for infl uenza viruses (detailed Dr Fatimah S Dawood or income, middle-income, and low-income countries and description of data sources and calculation methods is Dr Marc-Alain Widdowson, 14,19–28 multiplied it by the estimated symptomatic case fatality provided in appendix p 1). Estimates of the sARs were Infl uenza Division, Centers for ratio (sCFR) from select high-income countries. sAR was 4–33% for the paediatric age group, 0–22% for people Disease Control and Prevention, 1600 Clifton Road MS A-32, defi ned as the percentage of the population who developed aged 18–64 years, and 0–4% for those older than 64 years Atlanta, GA 30333, USA a symp tomatic respiratory illness associated with (table 1). We assumed that the true sAR of the fdawood@cdc.gov or laboratory-confi rmed 2009 pandemic infl uenza A H1N1. 2009 pandemic infl uenza A H1N1 virus in all age groups mwiddowson@cdc.gov sCFR was defi ned as the percentage of individuals with was greater than zero; therefore, we set the lower bound See Online for appendix symptomatic respiratory illness associated with laboratory- of the sAR for people aged 18–64 years and older than confi rmed 2009 pandemic infl uenza A H1N1 who died. 64 years as 1% in our model. Because of the huge variation We stratifi ed the estimates of sAR and sCFR into three in sARs within and between countries, we defi ned the age groups: 0–17 years (paediatric age group), 18–64 years, probability of any particular sAR value in the model with and greater than 64 years. We did not further stratify the a uniform distribution (ie, an equal probability of any paediatric age group largely because estimates for sAR value in the range). smaller strata lack precision. In previous pandemics, risk We obtained all available age-stratifi ed estimates of of death due to infl uenza has varied between countries, the 2009 pandemic infl uenza A H1N1 sCFR from six possibly because of diff erences in the underlying chronic studies in Denmark, the Netherlands, New Zealand, the 14,23–26 diseases and co-infections and access to antibiotics and UK, and the USA (two estimates). Sites in the UK, intensive care. Therefore, to adjust sCFR estimates for the USA, and Denmark were each adjusted for under- diff erences in risk of death from respiratory disease in reporting of deaths according to their own methods, low-income versus high-income countries, we stratifi ed whereas sites in the Netherlands and New Zealand countries into three risk groups. We adjusted the estimated sCFR based on laboratory-confi rmed, reported estimated base mortality in each risk group using a risk- deaths only. sCFR estimates were 0·002–0·013% for the group-specifi c respiratory mortality multiplier (RMM). paediatric age group, 0·018–0·159% for people aged The RMM was calculated by use of WHO’s estimates of 18–64 years, and 0·090–0·308% for those older than mortality rate for lower respiratory tract infection. Thus, 64 years (table 1). Because the estimates of sCFR had for each of the nine age–risk strata, we built a Monte wide and overlapping 95% CIs, we fi rst built a triangular Carlo probability model using SAS (version 9.2) with distribu tion for each sCFR by setting the median of the probability distributions defi ned according to the range triangular distribution as being equal to the point 688 www.thelancet.com/infection Vol 12 September 2012 Articles estimate and the minimum and maximum values were 0–17 years 18–64 years >64 years set as the fi fth and 95th percentiles (tenth and Countries with data inputs for symptomatic attack rate 90th percentiles for US estimates), respectively. We then Cambodia (Kapong Cham) *† 1% ·· ·· sampled each tri angular distribution an equal number of Bangladesh (Dhaka)*4% 0% 0% times to produce a sample of 10 000 points from which Denmark ‡§ 13% 4% 1% the fi nal model was sampled. As a sensitivity analysis, we Germany¶|| 32% 12% 2% re peated the same procedure using a uniform India (Ballabgarh)*|| 7% 4% 2% distribution for each sCFR estimate (results shown in Kenya (Kibera)*|| 11% 6% 0% appendix p 5). Kenya (Lwak)*|| 22% 22%** 0% Netherlands †† 13% 1% 0% RMM New Zealand †† 20% 7% 1% To adjust for diff erences in risk of death from respiratory Nicaragua (Managua) * 19% ·· ·· complications between countries, we developed an RMM 19,27 Peru (Cusco)*4% 1% 0% for all age groups with WHO’s 2008 country-specifi c 19,27 Peru (Lima) * 23% 7% 3% estimated mortality rates for lower respiratory tract 19,27 Peru (Madre de Dios) * 12% 2% 0% infection and WHO’s classifi cation of member states 19,27 into fi ve mortality strata—very low child and adult Peru (Tumbes) * 33% 5% 2% mortality; low child and adult mortality; low child and UK (England) ‡ 10% 3% 0% high adult mortality; high child and adult mortality; and USA ‡ 32% 12% 4% high child and very high adult mortality. We defi ned the Vietnam (Hanoi)*|| 5% 2% 0% RMM for countries in the lowest WHO mortality stratum Countries with data inputs for symptomatic case fatality ratio‡‡ as equal to 1. We then calculated mortality rate ratios for USA 0·005% (0·004–0·008) 0·027% (0·019–0·040) 0·092% (0·066–0·136) all other countries as the mortality rate associated with Denmark § 0·005% (0·002–0·013) 0·031–0·077%§§ ·· lower respiratory tract infection in a given country Netherlands 0·004% (0·003–0·006) 0·019–0·051%§§ ·· divided by the median mortality associated with lower New Zealand|| 0·002% (0·0003–0·005) 0·018% (0·012–0·024) 0·122% (0·024–0·220) respiratory tract infection from the lowest WHO 25 UK (England) 0·008% (0·004–0·013) 0·028% (0·012–0·047) 0·308% (0·057–1·069) mortality stratum (24 per 100 000 individuals). 24 USA (WI and NY) § 0·013% (0·005–0·033) 0·159% (0·066–0·333) 0·090% (0·008–1·471) Mortality rate ratios are shown by country in appendix References are for the general methods used for data gathering but might not include estimates for the period p 6. To minimise the eff ect of individual mortality rate reported in this table. *Based on population-based surveillance or cohort study of respiratory illness with laboratory ratios calculated with data of variable quality, we plotted confi rmation of 2009 pandemic infl uenza A H1N1 virus infection. †Data from Kampong Cham were only gathered the median mortality rate ratio for lower respiratory tract over 5 months; therefore, estimates of symptomatic attack rates from this site were not included in the fi nal model. ‡Based on statistical modelling by use of multiple data sources (see appendix p 1 for details). §Data were available for infection by WHO region and all-cause mortality stratum the following age groups: 0–14 years, 15–64 years, and greater than 64 years. ¶Derived from population-based (appendix p 11). We then assigned countries to one of three surveillance for medically attended respiratory illness adjusted with the estimated proportion of people who sought risk groups on the basis of the resulting distribution and care for respiratory illness. ||No available reference. **Outlier estimate excluded from the fi nal model. ††Based on estimate of seroprevalence of 2009 pandemic infl uenza A H1N1 virus infection adjusted with the estimated used the median mortality rate ratio from each group as proportion of people who were seropositive and had symptoms of respiratory illness. ‡‡90% CI reported for USA and the RMM for that group (with uncertainty defi ned by the 95% CI for all other estimates. §§Only 95% CI used. IQR for the mortality rate ratios). These risk groups were Table 1: Data inputs for estimation of mortality associated with 2009 pandemic infl uenza A H1N1 by all-African countries, non-African countries with high age group child and adult mortality, and all other countries (table 2). Non-African countries with high child and adult mortality included Afghanistan, Bangladesh, Bhutan, Bolivia, 10 000 estimates for each age–risk stratum to population Burma, Djibouti, Ecuador, Egypt, Guatemala, Haiti, India, estimates for each country in the RMM stratum to Iraq, Maldives, Morocco, Nepal, Nicaragua, North Korea, calculate the number of deaths by age in each country. Pakistan, Peru, Somalia, Sudan, Timor-Leste, and Yemen. The distribution of mortality-rate estimates for each age–risk stratum was highly right skewed with the lower Population estimates 75% of the estimates clustered within a narrow range of We used the UN’s projected population estimates for values and the upper 25% of estimates dispersed across 2010 when available, and the US Census Bureau mid- a wide range of values (appendix p 12 shows an example year population estimates for 2010 for all remaining of the distribution of estimates from one Monte Carlo countries. Kosovo, Niue, and Vatican City were excluded simula tion). We summed the median and 25% and 75% from this analysis because no available age-stratifi ed estimates for each country and age group to calculate a population estimates were available. global point estimate and range, respectively, by age group and WHO region. Estimated respiratory mortality We ran our Monte Carlo model 10 000 times to estimate Estimated cardiovascular mortality mortality for each of the nine age–risk strata (a total of Since infl uenza deaths can arise from respiratory or 90 000 iterations). We then applied each of the resulting cardio vascular complications, we also estimated www.thelancet.com/infection Vol 12 September 2012 689 Articles cardiovascular deaths due to 2009 pandemic H1N1 in and respiratory deaths attributed to the virus in fi ve people older than 17 years (infl uenza-associated high-income and middle-income countries stratifi ed by cardiovascular deaths are rare in children). First, we age 64 years and younger and older than 64 years. To calculated ratios of estimated cardiovascular to calculate the ratio of cardiovascular to respiratory deaths respiratory deaths associated with 2009 pandemic associated with the 2009 pandemic infl uenza A H1N1 infl uenza A H1N1 using estimates of excess circulatory virus, we used estimates of excess deaths attributable to this strain for Argentina, Brazil, Chile, Mexico, and the USA. The estimated ratios were 0·3–1·6 (median 0·7) in Minimum Maximum Median Distribution people aged 64 years and younger and 1·3–2·0 (1·8) in Symptomatic attack rate those older than 64 years (appendix p 14). 0–17 years 4% 33% 13% Uniform Second, we multiplied the median estimate of the 18–64 years 1% 12% 5% Uniform ratio of cardio vascular to respiratory deaths for each age >64 years 1% 4% 2% Uniform group by the base respiratory mortality rate in each Symptomatic case fatality ratio* country for people aged 18–64 years and older than 0–17 years 0·0003% 0·033% 0·005% Natural† 64 years to calculate the base cardiovascular mortality 18–64 years 0·012% 0·327% 0·029% Natural† rate associated with 2009 pandemic infl uenza A H1N1 >64 years 0·009% 1·449% 0·124% Natural† (unadjusted for diff erences between regions in risk of Mortality multiplier for lower respiratory tract infection cardiovascular death). We then assessed diff erences in African region countries, all ages 3 7 5 Uniform risk of death from cardiovascular disease between Non-African region countries with high child 1 4 3 Uniform countries by calcu lating mortality rate ratios from and adult mortality rates, all ages WHO’s estimates of cardiovascular disease mortality All other countries, all ages 1 1 1 Uniform rate based on the same method as for the RMM. Because the median mortality rate ratios for each region– *Minimum and maximum values for each age group represent those of the sample of 10 000 points from which the fi nal model was sampled. †Refers to the representative sample built from an equal number of samples selected from mortality stratum were 1–2 (appendix p 13), we did not each of the six sources for the estimates; fi rst, we built a triangular distribution setting the median of the triangular use a mortality multiplier for cardiovascular disease. distribution as equal to the point estimate and the minimum and maximum set to the fi fth and 95th percentiles, Thus, for each age group in each country, we estimated respectively; and second, we sampled each triangular distribution an equal number of times to produce a sample of the cardiovascular mortality rate associated with the 10 000 points from which the fi nal model sampled. 2009 pandemic infl uenza A H1N1 virus as the base Table 2: Parameters and probability distribution used in mortality model for 2009 pandemic respiratory mortality rate multiplied by the ratio of infl uenza A H1N1 cardiovascular to respiratory deaths and population. Reported* All ages 0–17 years (n [range†]) 18–64 years (n [range†]) >64 years (n [range†]) n (range†) Rate per 100 000 Respiratory deaths Africa 168 58 800 (30 800–112 200) 7·0 18 500 (9400–32 400) 35 900 (19 100–68 000) 4500 (2300–11 800) Americas ≥8533 17 500 (9400–34 500) 1·9 3000 (1500–5300) 11 400 (6200–20 900) 3000 (1700–8300) Eastern Mediterranean 1019 17 900 (9200–35 400) 3·0 4500 (2200–8500) 11 700 (6100–22 700) 1700 (800–4300) Europe ≥4879 16 400 (9000–33 400) 1·8 1800 (900–3100) 10 400 (5700–18 700) 4200 (2300–11 700) Southeast Asia 1992 59 500 (30 400–119 200) 3·3 12 400 (6100–23 500) 40 500 (21 000–78 500) 6700 (3300–17 200) Western Pacifi c 1858 31 100 (17 000–60 800) 1·7 4300 (2200–7400) 21 400 (11 800–38 700) 5300 (3000–14 800) Global‡ ≥18 449 201 200 (105 700–395 600) 2·9 44 500 (22 400–80 100) 131 300 (69 900–247 500) 25 400 (13 400–113 500) Respiratory and cardiovascular deaths§ Africa 168 65 600 (34 600–125 900) 7·8 18 500 (9400–32 400) 41 100 (22 000–77 300) 6100 (3200–16 200) Americas ≥8533 29 700 (16 200–61 500) 3·2 3000 (1500–5300) 18 600 (10 200–34 000) 8100 (4500–22 300) Eastern Mediterranean 1019 23 600 (12 300–47 100) 3·9 4500 (2200–8500) 16 000 (8500–30 400) 3100 (1600–8200) Europe ≥4879 31 300 (17 200–67 600) 3·5 1800 (900–3100) 17 600 (9700–31 800) 11 900 (6600–32 700) Southeast Asia 1992 78 600 (40 900–158 900) 4·4 12 400 (6100–23 500) 54 000 (28 400–103 000) 12 200 (6400–32 500) Western Pacifi c 1858 55 700 (30 600–114 500) 3·1 4300 (2200–7400) 36 400 (20 100–65 800) 15 000 (8300–41 300) Global‡ ≥18 449 284 400 (151 700–575 400) 4·1 44 500 (22 400–80 100) 183 700 (98 800–342 200) 56 400 (30 500–233 700) *Number of laboratory-confi rmed deaths due to 2009 pandemic infl uenza A H1N1 reported to WHO during April, 2009, to August 1, 2010. †The range was calculated by summing the 25th and 75th percentiles of estimates in each age group per country. ‡The total of the regional estimates is not always equal to the global estimate because of rounding. §Cardiovascular deaths were only calculated for people aged 18 years and older. Table 3: Reported and estimated respiratory and cardiovascular deaths associated with 2009 pandemic infl uenza A H1N1 by WHO region during the fi rst 12 months of virus circulation in each country and by age 690 www.thelancet.com/infection Vol 12 September 2012 Articles YLL for the period April, 2009, to August, 2010 (table 3). To further document the eff ect of the 2009 Summation of the fi fth and 95th percentile estimates in pan demic infl uenza A H1N1 that dispropor tionately each age group and country would have resulted in an aff ected young people compared with seasonal infl uenza estimated range of 39 000–1 315 800 respiratory deaths epidemics, we estimated the YLL from re spiratory and associated with 2009 pandemic infl uenza A H1N1. cardiovascular deaths associated with the virus as Results from the sensitivity analysis in which a uniform number of estimated deaths in each age group multiplied distribution was used instead of a triangular distribution by the mean number of additional years of life expected to sample from each sCFR range were broadly similar for people in each age group in each country. (appendix p 5). To demonstrate the eff ect of adjusting for The average numbers of years of life expected for diff erences in respiratory mortality, the calculation was people in each age group were obtained from WHO’s repeated without the RMM, which resulted in an estimate 2008 life tables. To show the diff erences in the mortality of 112 900 respiratory deaths (range 61 500–218 200) burden for age groups between the 2009 H1N1 pandemic associated with 2009 pandemic infl uenza A H1N1. and seasonal infl uenza epidemics, we estimated YLL that 29% of the 201 200 estimated respiratory deaths would have been lost if the age distribution of people associated with the 2009 pandemic infl uenza A H1N1 who died with 2009 pandemic infl uenza A H1N1 had occurred in the African region (table 3). The estimated been similar to the age distribution due to seasonal mortality rate in the African region was about two to four infl uenza. We assumed that during typical seasonal times that in countries elsewhere (table 3). The estimated infl uenza epidemics, 1% of deaths occurred in people range of respiratory deaths by country are shown in younger than 18 years, 9% in those aged 18–64 years, and appendix p 15. 33,34 90% in people older than 64 years. We then 65% of 2009 pandemic infl uenza A H1N1 deaths redistributed our pandemic deaths according to this worldwide were in individuals aged 18–64 years (60% of seasonal age distribution and estimated the YLL using global population), although the age distribution varied the same equation as above. by region. Overall, 13% of respiratory deaths associated with 2009 pandemic infl uenza A H1N1 were in people Role of the funding source older than 64 years (8% of global population). We received no external funding for the analysis. The An additional 83 300 cardiovascular deaths (range corresponding author had full access to all the data used 46 000–179 900) associated with the 2009 pandemic in the analysis and had fi nal responsibility for the infl uenza A H1N1 were estimated to have occurred in decision to submit for publication. people older than 17 years, resulting in a total of 284 400 respiratory and cardiovascular deaths (table 3). Results 20% of these deaths occurred in people older than The total of the median estimates of country-specifi c 64 years. With the inclusion of cardiovascular mortality, respiratory deaths associated with the 2009 there was a reduction in the disparity in mortality pan demic infl uenza A H1N1 was 201 200 (range associated with the pandemic by region, although the calculated by summing the 25th and 75th percentile mortality rate in Africa remained about two to three times estimates in each age group in each country), more than higher than elsewhere. Total numbers of respiratory and ten times the number of global deaths reported to WHO cardiovascular deaths by country are shown in appendix Deaths ≥1500 500–1499 200–499 50–199 <50 Data not available Figure 1: Global distribution of deaths associated with 2009 pandemic infl uenza A H1N1 during the fi rst year of virus circulation by country www.thelancet.com/infection Vol 12 September 2012 691 Articles Mortality per 100 000 8·1–9·5 6·1–8·0 4·1–6·0 2·1–4·0 0–2·0 Data not available Figure 2: Estimated age-adjusted respiratory and cardiovascular mortality rate associated with 2009 pandemic infl uenza A H1N1 per 100 000 individuals by country illness. Addition of cardiovascular deaths associated with Average life YLL due to pandemic infl uenza A YLL if age distribution expectancy at H1N1 respiratory mortality of deaths was similar 2009 pandemic infl uenza A H1N1 among people older birth (years) to seasonal infl uenza* than 17 years increased the mortality burden to Africa 53 2 278 800 (1 194 500–4 196 500) 556 100 151 700–575 400 deaths (table 3). Our global estimate was Americas 76 1 050 600 (567 600–1 965 100) 361 800 more than 15 times higher than the number of laboratory- Eastern Mediterranean 65 862 500 (446 100–1 645 900) 219 200 confi rmed deaths reported to WHO during the fi rst 16 months of the pandemic. A disproportionate number Europe 75 927 600 (506 800–1 756 400) 360 900 of total deaths from cardiovascular and respiratory Southeast Asia 65 2 725 300 (1 407 900–5 243 400) 738 700 diseases (51%) was estimated to have occurred in the Western Pacifi c 75 1 862 200 (1 015 800–3 445 500) 605 000 African and southeast Asian regions where 38% of the Global 68 9 707 000 (5 138 700–18 252 800) 2 841 700 world’s population live and where data for infl uenza Data are total of medians (range), unless otherwise indicated. YLL=years of life lost. *Number of 2009 pandemic 35 incidence are scarce. Additionally, most deaths were infl uenza A H1N1 deaths is redistributed across age groups to approximate the typical age distribution of seasonal reported in people aged 18–64 years, consistent with infl uenza deaths in developed countries where estimates were available (90% for people aged >64 years, 9% for age 18–64 years, and 1% for age 0–17 years). previous reports. These fi ndings are in contrast to those for seasonal infl uenza deaths. Roughly 80–90% of these Table 4: YLL because of deaths associated with 2009 pandemic infl uenza A H1N1 during the fi rst arise in people aged 65 years and older in Australia, 12 months of virus circulation and YLL that would be lost with a seasonal infl uenza age distribution of deaths Denmark, Singapore, and the USA, where age-stratifi ed 33,37–39 estimates are available. However, the age distribution p 19, and estimated H1N1-associated respiratory and of seasonal infl uenza-associated deaths might diff er cardiovascular deaths and mortality rates by country are between settings because of diff erences in the prevalence shown in fi gures 1 and 2. of underlying illnesses (such as HIV/AIDS and Estimated YLL were 9 707 000 during the fi rst 12 months tuberculosis) in younger adults in some low-income of the pandemic (table 4). Southeast Asia was the region countries. The shift in the age distribution of deaths with the greatest YLL (table 4). Total YLL attributable to during the pandemic resulted in substantially more YLL deaths associated with 2009 pandemic infl uenza A H1N1 than would have occurred if the age distribution of was 3·4 times higher than if the age distribution of deaths deaths had been similar to that of most seasonal had been similar to that during seasonal epidemics in infl uenza epidemics, consistent with obser vations from developed countries where age-specifi c estimates of previous pandemics (panel). seasonal infl uenza mortality are available (table 4). The Estimates from previous pandemics indicate that increase in YLL associated with 2009 pandemic infl uenza infl uenza mortality rates vary substantially between 9,47 A H1N1 was due to the increased rate of death in people coun tries. Our accounting for diff erential risk of aged 64 years and younger. infl uenza-associated death between countries is supported by the fi ndings of Cohen and colleagues that Discussion excess seasonal pneumonia and infl uenza mortality in During the fi rst year of circulation of the 2009 pandemic people aged 65 years and older is at least three times infl uenza A H1N1 virus in each country, an estimated higher in South Africa than in the USA. Further, data 105 700–395 600 people died of associated respiratory from the Americas and the western Pacifi c show that risk 692 www.thelancet.com/infection Vol 12 September 2012 Articles of death associated with the 2009 pandemic was up to six Panel: Research in context times higher in indigenous than in non-indigenous 8 6,8 populations. In both studies, the prevalence of Systematic review underlying disorders (including malnutrition in South We searched PubMed for reports of studies in any language from April, 2009, to March, Africa) and access to health care were postulated to 2012, with estimates of the number of deaths or mortality due to the 2009 pandemic contribute to the increased risk of infl uenza-associated infl uenza A H1N1. We identifi ed 12 studies from eight countries in Europe, the Americas, 14,17,23,25,39–46 42 death. Nair and colleagues also estimated that mortality western Pacifi c, and Asia. In two studies from Australia (New South Wales) and rates for infl uenza-associated acute lower respiratory the UK, no deaths were estimated based on the excess all-cause mortality during periods tract infection in children younger than 5 years are three of infl uenza virus circulation. Two other UK studies were reported—in the England-only times higher in low-income countries than in high- study, the estimated number of deaths was 390–490 and in the study of only England income countries. and Wales the estimate was 1556. In one of the two studies from Mexico, the estimated WHO has estimated that an average of 0·004–0·008% mortality was 35 per 100 000 people in San Luis Potosi, whereas in the other study the per year of the world’s population (250 000–500 000 peo- estimate was 6200 deaths for Mexico based on excess pneumonia and infl uenza ple) die as a result of seasonal infl uenza. Estimates of mortality or 26 500 deaths based on excess all-cause mortality. According to the pandemic infl uenza mortality ranged from 0·03% of the estimates from two studies that included deaths in 2009 only, there were 6000 deaths in world’s population during the 1968 pandemic to 1–3% of Bangladesh and 0–256 deaths in Hong Kong, equal to an age-standardised mortality of 7,9,47 the world’s population during the 1918 pandemic. We 2·9–14·8 per 100 000 individuals. According to the estimates from the remaining studies, 14 23,39 estimate that 0·001–0·007% of the world’s population there were 8868–18 306 deaths in the USA, 121 or 151–473 in Denmark, and (n=6 891 433 594) died of respiratory complications 266–958 in the Netherlands. associated with 2009 pandemic infl uenza A H1N1 during Interpretation the fi rst year of virus circulation or 0·001–0·011% when Our estimated ranges of deaths associated with 2009 pandemic infl uenza A H1N1 for cardiovascular deaths were included. However, our each country overlap with reported estimates from Denmark (109–438 vs 151–473 or estimates are not directly comparable to those of 121), the Netherlands (326–1289 vs 266–958), the USA (5834–22 697 vs 8868–18 306), mortality associated with seasonal and previous China (Hong Kong only; 1·7–6·2 per 100 000 vs 2·9–14·8 per 100 000), and Bangladesh pandemic infl uenza for at least two reasons. First, the (3899–15 135 vs 6000). However, our estimated ranges are higher than most estimates WHO estimate of global seasonal infl uenza mortality is from the UK (1237–4946 vs 0 for the UK or 390–490 for England or 1556 for England and an extrapolation of esti mates from high-income 48 Wales), and Australia (406–1589 vs mortality rate of 0 on the basis of excess all-cause countries without accounting for regional variation in deaths in the Hunter New England region) and lower than estimates from Mexico risk of death from outcomes that might be associated (1670–6105 vs 6200 excess pneumonia and infl uenza deaths or 26 000 excess all-cause with infl uenza, though details of the methods have not deaths). Diff erences in study methods must be taken into account when estimates from been provided. Second, reported estimates of global these studies are compared. The results of our study add to the understanding of the pandemic mortality include data from several years of global eff ect of the 2009 pandemic infl uenza A H1N1 because we have estimated the pandemic virus circulation, whereas our estimate associated deaths in African and southeast Asian countries for which there is only one includes data for only the fi rst year of 2009 pandemic reported estimate from Bangladesh so far. We show that half of all global infl uenza A H1N1 virus circulation in each country. H1N1-associated respiratory and cardiovascular deaths could have occurred in Africa and We were unable to identify any factor or group of southeast Asia and that mortality rates might have been two to three times higher in factors that allowed stratifi cation of countries into Africa than in other regions of the world. transmission risk groups, and therefore we assumed the same range of 2009 pandemic infl uenza A H1N1 sARs for all countries. This approach results in point estimates fi ndings that the 2009 pandemic infl uenza A H1N1 attack that underestimate or overestimate deaths in countries rates varied substantially both within and between coun- 49–54 that had very high or very low transmission, respectively, tries. Second, we relied on only six estimates of but should produce ranges for most country-specifi c sCFR from high-income countries because none were estimates that are likely to capture the true number of available from low-income or middle-income countries deaths in each country. For this reason, we present only including China and India, which account for 37% of the ranges for our country-specifi c estimates. The world’s population. Although the available estimates of assumption of a range of 2009 pandemic infl uenza A sCFR were each calculated with diff erent methods for H1N1 sARs for all countries is less likely to bias our ascertaining the numerator (number of deaths associated regional and global estimates. with 2009 pandemic infl uenza A H1N1) and denominator Our model was limited by the scarcity of globally (number of symptomatic 2009 pandemic infl uenza A representative estimates of sAR and sCFR. We relied on H1N1 virus infections), most of these estimates were only 16 direct estimates of 1 year sARs for 2009 infl uenza fairly similar. However, each sCFR estimate had a A H1N1 that varied greatly, and were likely to be aff ected substantial amount of inherent uncertainty, particularly by an unidentifi ed combination of factors such as for the older than 64 years age group in which the diff erences in climate, population density, and popu- denominator for the sCFR was lower than in other age lation age structure. Data from serology studies of both groups. The uncertainty of the individual sCFR estimates symptomatic and asymptomatic infections support our widened the confi dence intervals of our fi nal estimates. www.thelancet.com/infection Vol 12 September 2012 693 Articles The lack of infl uenza sCFR or mortality rate estimates continued eff orts to strengthen infl uenza surveillance worldwide, particularly for infl uenza-associated mor- for low-income and middle-income countries is an tality, are needed both to guide seasonal infl uenza important knowledge gap in the understanding of the prevention strategies and to build infl uenza surveillance epidemiology of both seasonal and pandemic infl uenza. systems to provide better and more timely and globally To overcome the lack of data to inform sCFR estimates for representative data for infl uenza-associated mortality these settings, we developed an RMM from mortality rates during future pandemics. for lower respiratory tract infection that are subject to several limitations. First, mortality rates consist of both Contributors FSD, ADI, M-AW, DKS, MIM, JB, and CR all actively contributed to the disease incidence and case fatality rates. Thus, the RMM is design of the analysis; CR, MIM, DKS, P-YC, DB, RFB, WAB, PB, DRF, aff ected by diff erences in the incidence of non-infl uenza KBF, AG, NTH, PH, QSH, MAK, AK, RL, JMM, KM, RP, AMP, HR, AS, respiratory infections, and the dearth of estimates of YOT, JW, HY, and SV provided the data used in this analysis. FSD and incidence of lower respiratory tract infection for most ADI did the data analysis. FSD wrote the fi rst draft of the report, and all authors contributed to the interpretation of the results of the analysis countries prevented us from adjusting for diff erences in and to the revision and fi nal preparation of the report for submission. incidence. Second, we used one RMM for all age groups Confl icts of interest because the age-stratifi ed rate ratios in the 0–17 year group We declare that we have no confl icts of interest. were up to 58 times higher than the all-age ratios, probably Acknowledgments due to a larger diff erence in incidence of non-infl uenza We thank WHO’s Ad-Hoc Working Group on Pandemic Mortality respiratory infections in this age group. By contrast, the Burden for valuable insights; Patrick Glew at the Oakridge Institute of all-age rate ratios were similar to ratios in the older age Science and Education and the Centers for Disease Control and groups that had the most number of deaths associated Prevention for organisation of data used to calculate the mortality multiplier; the Dutch Ministry of Health, Welfare, and Sport for their with the 2009 pandemic infl uenza A H1N1. Because the support of the study that resulted in data from the Netherlands and relative risk of infl uenza-associated re spiratory death is M Van Boven of the Centre for Infectious Disease Control, National likely to vary between age groups, use of one all-age RMM Institute for Public Health and the Environment, Bilthoven, Netherlands is likely to result in an underestimation of virus-associated for his work on the study that resulted in data from the Netherlands; Angel Balmaseda, Guillermina Kuan, and Eva Harris for their work on mortality in children in low-income countries. obtaining data from Nicaragua; Ernesto Ortiz, Abel Estela, One additional potential methodological limitation of Maria Luisa Morales, Judith Patricia Breña, and Candice Romero for our analysis is that we summed the median and IQR their operational management and supervision at the surveillance sites estimates for each age group in each country to calculate in Peru, all from the Naval Medical Research Unit-6, Lima; and Udo Buchholz for his contribution of data used to estimate the a point estimate and range of the total number of deaths symptomatic attack rate in Germany. The fi ndings and conclusions in by country, region, and globally. The assumption with this report are those of the authors and do not necessarily represent the this approach is that symptomatic attack rates are views of the Centers for Disease Control and Prevention. positively correlated between age groups in each country References (ie, it is unlikely for one age group to have an attack rate 1 Kasowski EJ, Garten RJ, Bridges CB. Infl uenza pandemic epidemiologic and virologic diversity: reminding ourselves of the on the higher end of the global range of sARs and possibilities. Clin Infect Dis 2011; 52 (suppl 1): S44–49. another age group to have an attack rate on the lower 2 WHO. Pandemic (H1N1) 2009–update 112. www.who.int/csr/ end of the global range). Had we summed the result of don/2010_08_06/en/index.html (accessed Feb 24, 2011). each Monte Carlo iteration for each age group in each 3 Michalove J, Ndahwouh T, Diop O, et al, eds. Delayed pandemic H1N1 transmission in West Africa, May 2009–March 2010. Options for the country, then summed the results for each country, and Control of Infl uenza VII Conference; Hong Kong; Sept 3–7, 2010. taken the median and IQR of the distribution of the 4 Miller E, Hoschler K, Hardelid P, Stanford E, Andrews N, global estimates, our estimate would have been 25% Zambon M. Incidence of 2009 pandemic infl uenza A H1N1 infection in England: a cross-sectional serological study. Lancet higher. A fi nal potential limitation is that we assumed 2010; 375: 1100–08. that the life expectancy of people who died with 5 Ortiz JR, Lafond KE, Wong TA, Uyeki TM. Pandemic infl uenza in 2009 pandemic infl uenza A H1N1 was equal to that of Africa, lessons learned from 1968: a systematic review of the literature. Infl uenza Other Respi Viruses 2012; 6: 11–24. people of the same age in the general population, 6 Cohen C, Simonsen L, Kang JW, et al. Elevated infl uenza-related although people who die from infl uenza might be more excess mortality in South African elderly individuals, 1998–2005. likely to have underlying disorders that are associated Clin Infect Dis 2010; 51: 1362–69. with a reduced life expectancy. 7 Johnson NP, Mueller J. Updating the accounts: global mortality of the 1918–1920 “Spanish” infl uenza pandemic. Bull Hist Med 2002; We estimated that 151 700–575 400 respiratory and 76: 105–15. cardiovascular deaths associated with 2009 pan demic 8 La Ruche G, Tarantola A, Barboza P, Vaillant L, Gueguen J, infl uenza A H1N1 occurred during the fi rst year of virus Gastellu-Etchegorry M. The 2009 pandemic H1N1 infl uenza and indigenous populations of the Americas and the Pacifi c. circulation in every country in the world. Our fi ndings Euro Surveill 2009; 14: 19366. emphasise the need to improve the global response to 9 Murray CJ, Lopez AD, Chin B, Feehan D, Hill KH. Estimation of future infl uenza pandemics and expand production and potential global pandemic infl uenza mortality on the basis of vital registry data from the 1918–20 pandemic: a quantitative analysis. improve delivery of infl uenza vaccines to Africa and Lancet 2006; 368: 2211–18. southeast Asia because these countries might have 10 Nair H, Brooks WA, Katz M, et al. Global burden of respiratory borne a disproportionate burden of pandemic mortality infections due to seasonal infl uenza in young children: a systematic review and meta-analysis. Lancet 2011; 378: 1917–30. during the fi rst year of virus circulation. Additionally, 694 www.thelancet.com/infection Vol 12 September 2012 Articles 11 Thompson WW, Weintraub E, Dhankhar P, et al. Estimates of US 33 Thompson WW, Shay DK, Weintraub E, et al. Mortality associated infl uenza-associated deaths made using four diff erent methods. with infl uenza and respiratory syncytial virus in the United States. Infl uenza Other Respi Viruses 2009; 3: 37–49. JAMA 2003; 289: 179–86. 12 Serfl ing RE. Methods for current statistical analysis of excess 34 Estimates of deaths associated with seasonal infl uenza—United pneumonia-infl uenza deaths. Public Health Rep 1963; 78: 494–506. States, 1976–2007. MMWR Morb Mortal Wkly Rep 2010; 59: 1057–62. 13 Simonsen L, Clarke MJ, Stroup DF, Williamson GD, Arden NH, 35 Yazdanbakhsh M, Kremsner PG. Infl uenza in Africa. PLoS Med Cox NJ. A method for timely assessment of infl uenza-associated 2009; 6: e1000182. mortality in the United States. Epidemiology 1997; 8: 390–95. 36 Vaillant L, La Ruche G, Tarantola A, Barboza P. Epidemiology of 14 Shrestha SS, Swerdlow DL, Borse RH, et al. Estimating the fatal cases associated with pandemic H1N1 infl uenza 2009. burden of 2009 pandemic infl uenza A (H1N1) in the United States Euro Surveill 2009; 14: 19309. (April 2009–April 2010). Clin Infect Dis 2011; 52 (suppl 1): S75–82. 37 Chow A, Ma S, Ling AE, Chew SK. Infl uenza-associated deaths in 15 Viboud C, Miller M, Olson D, Osterholm M, Simonsen L. tropical Singapore. Emerg Infect Dis 2006; 12: 114–21. Preliminary estimates of mortality and years of life lost associated 38 Newall AT, Wood JG, Macintyre CR. Infl uenza-related with the 2009 A/H1N1 pandemic in the US and comparison with hospitalisation and death in Australians aged 50 years and older. past infl uenza seasons. PLoS Curr 2010: RRN1153. Vaccine 2008; 26: 2135–41. 16 Wielders CC, van Lier EA, van ‘t Klooster TM, et al. The burden of 39 Nielsen J, Mazick A, Glismann S, Molbak K. Excess mortality 2009 pandemic infl uenza A(H1N1) in the Netherlands. related to seasonal infl uenza and extreme temperatures in Eur J Public Health 2012; 22: 150–57. Denmark, 1994–2010. BMC Infect Dis 2011; 11: 350. 17 Charu V, Chowell G, Palacio Mejia LS, et al. Mortality burden of the 40 Comas-Garcia A, Garcia-Sepulveda CA, Mendez-de Lira JJ, A/H1N1 pandemic in Mexico: a comparison of deaths and years of Aranda-Romo S, Hernandez-Salinas AE, Noyola DE. Mortality life lost to seasonal infl uenza. Clin Infect Dis 2011; 53: 985–93. attributable to pandemic infl uenza A (H1N1) 2009 in San Luis 18 WHO. Disease and injury country estimates. 2011. http://www.who. Potosi, Mexico. Infl uenza Other Respi Viruses 2011; 5: 76–82. int/healthinfo/global_burden_disease/estimates_country/en/index. 41 Hardelid P, Andrews N, Pebody R. Excess mortality monitoring in html (accessed Feb 11, 2011). England and Wales during the infl uenza A(H1N1) 2009 pandemic. 19 Unit-6 NMR. Burden of illness and risk factors for transmission of Epidemiol Infect 2011; 139: 1431–39. seasonal infl uenza in four distinct regions of Peru. http:// 42 Muscatello DJ, Cretikos M, Macintyre CR. All-cause mortality proyectoinfl uenza.pe (accessed Aug 8, 2011). during fi rst wave of pandemic (H1N1) 2009, New South Wales, 20 Bandaranayake D, Huang QS, Bissielo A, et al. Risk factors and Australia, 2009. Emerg Infect Dis 2010; 16: 1396–402. immunity in a nationally representative population following the 43 Wijngaard CC, Asten L, Koopmans MP, et al. Comparing pandemic 2009 infl uenza A(H1N1) pandemic. PLoS One 2010; 5: e13211. to seasonal infl uenza mortality: moderate impact overall but high 21 Brooks WA, Goswami D, Rahman M, et al. Infl uenza is a major mortality in young children. PLoS One 2012; 7: e31197. contributor to childhood pneumonia in a tropical developing 44 Hardelid P, Pebody R, Andrews N. Mortality caused by infl uenza country. Pediatr Infect Dis J 2010; 29: 216–21. and respiratory syncytial virus by age group in England and Wales 22 Kuan G, Gordon A, Aviles W, et al. The Nicaraguan pediatric 1999–2010. Infl uenza Other Respi Viruses 2012; published online dengue cohort study: study design, methods, use of information March 9. DOI:10.1111/j.1750-2659.2012.00345.x. technology, and extension to other infectious diseases. 45 Homaira N, Luby SP, Alamgir A, et al. Infl uenza-associated Am J Epidemiol 2009; 170: 120–29. mortality in 2009 in four sentinel sites in Bangladesh. 23 Molbak K, Widgren K, Jensen KS, et al. Burden of illness of the Bull World Health Organ 2012; 90: 272–78. 2009 pandemic of infl uenza A (H1N1) in Denmark. Vaccine 2011; 46 Yang L, Chan KP, Cowling BJ, et al. Excess mortality associated with 29 (suppl 2): B63–69. the 2009 pandemic of infl uenza A(H1N1) in Hong Kong. 24 Presanis AM, De Angelis D, Hagy A, et al. The severity of pandemic Epidemiol Infect 2011. DOI:10.1017/S0950268811002238. H1N1 infl uenza in the United States, from April to July 2009: a 47 Mathews JD, Chesson JM, McCaw JM, McVernon J. Understanding Bayesian analysis. PLoS Med 2009; 6: e1000207. infl uenza transmission, immunity and pandemic threats. 25 Presanis AM, Pebody RG, Paterson BJ, et al. Changes in severity of Infl uenza Other Respi Viruses 2009; 3: 143–49. 2009 pandemic A/H1N1 infl uenza in England: a Bayesian evidence 48 WHO. Infl uenza (seasonal). 2009. http://www.who.int/ synthesis. BMJ 2011; 343: d5408. mediacentre/factsheets/fs211/en/ (accessed June 4, 2011). 26 Steens A, Waaijenborg S, Teunis PF, et al. Age-dependent patterns 49 Chan YJ, Lee CL, Hwang SJ, et al. Seroprevalence of antibodies to of infection and severity explaining the low impact of pandemic (H1N1) 2009 infl uenza virus among hospital staff in a 2009 infl uenza A (H1N1): evidence from serial serologic surveys in medical center in Taiwan. J Chin Med Assoc 2010; 73: 62–66. the Netherlands. Am J Epidemiol 2011; 174: 1307–15. 50 Chen MI, Lee VJ, Lim WY, et al. 2009 infl uenza A(H1N1) 27 Tinoco Y, Razuri H, Ortiz EJ, et al. Preliminary population-based seroconversion rates and risk factors among distinct adult cohorts epidemiological and clinical data on 2009 pandemic H1N1 infl uenza in Singapore. JAMA 2010; 303: 1383–91. A (pH1N1) from Lima, Peru. Infl uenza Other Respi Viruses 2009; 51 Moghadami M, Moattari A, Tabatabaee HR, et al. High titers of 3: 253–56. hemagglutination inhibition antibodies against 2009 H1N1 28 Vong S, Goyet S, Ly S, et al. Under-recognition and reporting of infl uenza virus in Southern Iran. Iran J Immunol 2010; 7: 39–48. dengue in Cambodia: a capture-recapture analysis of the National 52 Tsai TF, Pedotti P, Hilbert A, et al. Regional and age-specifi c Dengue Surveillance System. Epidemiol Infect 2011; 140: 491–99. patterns of pandemic H1N1 infl uenza virus seroprevalence inferred 29 WHO. List of member states by WHO region and mortality from vaccine clinical trials, August-October 2009. Euro Surveill 2010; stratum. http://www.who.int/whr/2003/en/member_ 15: 19624. states_182-184_en.pdf (accessed Oct 11, 2010). 53 Zimmer SM, Crevar CJ, Carter DM, et al. Seroprevalence following 30 UN. Department of Economic and Social Aff airs, Population the second wave of Pandemic 2009 H1N1 infl uenza in Pittsburgh, Division, Population Estimates and Projections Section. http://esa. PA, USA. PLoS One 2010; 5: e11601. un.org/unpd/wpp/unpp/panel_population.htm (accessed Feb 30, 54 Gilbert GL, Cretikos MA, Hueston L, Doukas G, O’Toole B, 2011). Dwyer DE. Infl uenza A (H1N1) 2009 antibodies in residents of New 31 US Census Bureau International Data Base, Population Division. South Wales, Australia, after the fi rst pandemic wave in the US Census Bureau. 2010. http://www.census.gov/ipc/www/idb/ 2009 southern hemisphere winter. PLoS One 2010; 5: e12562. region.php (Feb 11, 2011). 32 WHO. Life tables for WHO member states. 2011. http://www.who. int/healthinfo/statistics/mortality_life_tables/en/index.html (Feb 11, 2011). www.thelancet.com/infection Vol 12 September 2012 695

Journal

The Lancet Infectious DiseasesUnpaywall

Published: Sep 1, 2012

There are no references for this article.