Professional Documents
Culture Documents
GlobalHealthEstimatesTechnicalPaperWHO/HIS/HSI/GHE/2013.3
Acknowledgments
ThisTechnicalReport waswritten byColin Mathers,GretchenStevens andDoris MaFat withinputs and assistancefromWahyuRetnoMahanani,JessicaHoandLiLiu.Estimatesofregionaldeathsbycausefor years 20002011 were primarily prepared by Colin Mathers, Gretchen Stevens, Jessica Ho, Doris Ma Fat and Wahyu Retno Mahanani, of the Mortality and Burden of Disease Unit in the WHO Department of Health Statistics and Information Systems, in the Health Systems and Innovation Cluster of the World Health Organization (WHO), Geneva, drawing heavily on advice and inputs from other WHO Departments, collaborating United Nations (UN) Agencies, and WHO expert advisory groups and academiccollaborators. Many of the inputs to these estimates result from collaborations with Interagency Groups, expert advisory groups and academic groups. The most important of these include the Interagency Group on Child Mortality Estimation (UNIGME), the UN Population Division, the Child Health Epidemiology Reference Group (CHERG), the Maternal Mortality Expert and Interagency Group (MMEIG), the International Agency for Research on Cancer, WHO QUIVER, and the Global Burden of Disease 2010 Study Collaborating Group. While it is not possible to name all those who provided advice, assistance or data, both inside and outside WHO, we would particularly like to note the assistance and inputs provided by Kirill Andreev, Diego Bassani, Bob Black, Ties Boerma, Phillipe Boucher, Freddie Bray, Tony Burton, Harry Campbell, Doris Chou, Richard Cibulskis, Simon Cousens, Jacques Ferlay, Marta Gacic Dobo, Richard Garfield, Alison Gemmill, Patrick Gerland, Peter Ghys, Philippe Glaziou, Danan Gu, Ken Hill,KacemIaych,MieInoue,RobertJakob,DeanJamison,PrabhatJha,HopeJohnson,JoyLawn,NanLi, Li Liu, Rafael Lozano, Chris Murray, Lori Newman, Mikkel Oestergaard, Max Parkin, Margie Peden, Francois Pelletier, Juergen Rehm, Igor Rudan, Lale Say, Emily Simons, Charalampos Sismanidis, Thomas Spoorenberg, Karen Stanecki, Peter Strebel, Emi Suzuki, Tamitza Toroyan, Theo Vos, Tessa Wardlaw, RichardWhite,JohnWilmothandDanzhenYou. Estimatesandanalysisareavailableat: http://www.who.int/gho/mortality_burden_disease/en/index.html Forfurtherinformationabouttheestimatesandmethods,pleasecontacthealthstat@who.int In this series 1. WHO methods and data sources for life tables 19902011 (Global Health Estimates Technical Paper WHO/HIS/HSI/GHE/2013.1) 2.CHERGWHO methodsand datasources forchild causesof death20002011 (GlobalHealth Estimates TechnicalPaperWHO/HIS/HSI/GHE/2013.2) 3. WHO methods and data sources for global causes of death 20002011 (Global Health Estimates TechnicalPaperWHO/HIS/HSI/GHE/2013.3)
Table of Contents
Acknowledgments ..........................................................................................................................................i TableofContents..........................................................................................................................................ii 1 Introduction.1 2 Populationandallcausemortalityestimatesforyears20002011........................................................3 2.1 Allcausemortalityandpopulationestimates................................................................................. 3 2.2 Estimationofneonatal,infantandunder5mortalityrates............................................................3 2.3 Allcausemortalitycomputedfromcivilregistrationdata..............................................................4 2.4 Allcausemortalityprojectedfromcivilregistrationdata...............................................................4 2.5 Countrieswithotherinformationonlevelsofadultmortality.......................................................5 2.6 Mortalityshocksepidemics,conflictsanddisasters..................................................................... 6 3 Countrieswithuseabledeathregistrationdata...................................................................................... 7 3.1 Dataandestimates.......................................................................................................................... 7 3.2 Inclusioncriteriaforcountrieswithhighqualitydeathregistrationdata.......................................7 3.3 Redistributionofunknownsex/ageandgarbagecodesandadjustmentforincompletedeath registration .....................................................................................................................................12 3.4 MappingtoGHEcauselists............................................................................................................ 12 3.5 Interpolationandextrapolationformissingcountryyears...........................................................14 ........................................................................................................ 14 3.6 Adjustmentofspecificcauses 3.7 Othernationallevelinformationoncausesofdeath................................................................... 14 4 Childmortalitybycause........................................................................................................................ 19 4.1 Causesofunder5deathincountrieswithgooddeathregistrationdata.....................................19 .....................................................19 4.2 Causesofneonataldeath(deathsatlessthan28daysofage) 4.3 Causesofchilddeathatages159monthslowmortalitycountries ...........................................20 4.4 Causesofchilddeathatages159monthshighmortalitycountries.........................................20 4.5 CausesofchilddeathforChinaandIndia..................................................................................... 21 4.6 InclusionofWHOCHERGestimatesinGlobalHealthEstimates20002011................................21 5 Methodsforspecificcauseswithadditionalinformation..................................................................... 22 5.1 Tuberculosis...................................................................................................................................22 5.2 HIV/AIDSandsexuallytransmitteddiseases................................................................................. 22 5.3 Malaria...........................................................................................................................................22 5.4 Whoopingcough............................................................................................................................ 23 5.5 Measles..........................................................................................................................................23 5.6 Schistosomiasis.............................................................................................................................. 24 ii
5.7 Maternalcausesofdeath.............................................................................................................. 24 5.8 Cancers...........................................................................................................................................24 5.9 Alcoholuseanddrugusedisorders............................................................................................... 25 5.10 Epilepsy..........................................................................................................................................25 5.11 Roadinjuries..................................................................................................................................25 5.11.1 5.11.2 5.11.3 5.11.4 Countrieswithdeathregistrationdata............................................................................. 26 Countrieswithothersourcesofinformationoncausesofdeath....................................26 Countrieswithpopulationslessthan150000.................................................................26 Countrieswithouteligibledeathregistrationdata..........................................................26
5.12 Conflictandnaturaldisasters........................................................................................................ 28 6 Othercausesofdeathforcountrieswithoutuseabledata................................................................... 30 7 Uncertaintyofestimates....................................................................................................................... 33 References.37 ........................................................................... 43 AnnexTableA GHEcausecategoriesandICD10codes AnnexTableB Firstlevelcategoriesforanalysisofchildcausesofdeath...............................................48 AnnexTableC ReassignmentofICD10codesforcertainneonataldeaths...........................................49 AnnexTableD Countrygroupingsusedforregionaltabulations.............................................................51 D.1 WHORegionsandMemberStates................................................................................................ 51 D.2 CountriesgroupedbyWHORegionandaverageincomepercapita*..........................................52 D.3 WorldBankincomegrouping*...................................................................................................... 53 D.4 WorldBankRegions....................................................................................................................... 54 D.5 MillenniumDevelopmentGoal(MDG)Regions............................................................................ 55 AnnexTableE MappingofIndiaMDScategoriestoGHEcauses.............................................................56 AnnexTableF Methodsusedforestimationofchildandadultmortalitylevels,andcausesofdeath,by country,20002011 ........................................................................................................... 58 AnnexTableG Methodsusedtoestimateroadtrafficdeathsfor182participatingcountries...............64
iii
Introduction
Global, regional, and country statistics on population and health indicators are important for assessing development and health progress and for guiding resource allocation. The demand is growing for timely datato monitor progress inhealth outcomes such as childmortality, maternal mortality, life expectancy and age and causespecific mortality rates. Much of the current focus is on monitoring progress towardsthetargetsofthe(healthrelated)MillenniumDevelopmentGoals(MDGs),includingtimeseries and countrylevel estimates that are regularly updated. But increasingly, the demand is for comprehensive estimates across the full spectrum, including noncommunicable diseases (NCDs) and injuries. WHO has previously published comprehensive estimates of deaths by region, cause, age and sex for years 2000 and 2002 (1), 2001 (2), 2004 (3) and 2008 (4). Beginning with the 2004 estimates, WHO has also released summary estimates of causes of death for its Member States (5). These successive single year estimates did not form a time series, as each revision involved revisions to data and methods for a range of inputs. To address the increasing demand for time series, for countrylevel estimates, and for comprehensive estimates across NCD and injury causes, as well as the more traditional priorities in infectious and parasitic diseases, updated Global Health Estimates (GHE) are being released, commencingwithregionallevelestimatesofdeathsbycause,ageandsexforyears20002011(6). This technical paper documents the data sources and methods used for preparation of these regional level cause of death estimates for years 20002011. Annex Table A lists the cause of death categories and their definitions in terms of the International Classification of Diseases, Tenth Revision (ICD10) (7). These estimates are available for years 2000 and 2011 for selected regional groupings of countries (6), definedinAnnexTablesD,athttp://www.who.int/healthinfo/global_health_estimates/en/. Comprehensive estimates of mortality, causes of death, DALYs for diseases, injuries and risk factors were released in December 2012 (810) by the Institute of Health Metrics and Evaluation (IHME) as part oftheGlobalBurdenofDisease2010study(GBD2010).WHOwasacollaboratorinthestudyfrom2007 to2011,butdidnotendorsethefinalresults,asitwasunabletoobtainfullaccesstotheresultspriorto publication or to evaluate them. In some areas, the results of the GBD 2010 differ substantially from existinganalysesdonebyWHOandotherUnitedNationsagenciesatglobal,regionalandcountrylevels. In many other areas, the GBD 2010 results are updates that are broadly similar to previous WHO analyses. Further work with IHME and expert groups is needed to examine the reasons for current differences. One of the six core functions of WHO is monitoring of the health situation, trends and determinants in theworld.OvertheyearsithascooperatedcloselywithotherUNpartneragencieslikeUNICEF,UNAIDS, UNFPA and the UN Population Division to collect and compile global health statistics. There are a number of established UN multiagency expert group mechanisms for cross cutting topics such as child mortality (the UNIGME including UNICEF/WHO/ UNPD/World Bank and the UNIGME Technical Advisory Group) and child causes of death (CHERG, WHO/UNICEF), specific diseases such as HIV/AIDS (UNAIDS Reference Group), maternal mortality (MMEIG including WHO/UNICEF/UNFPA/World Bank), tuberculosis (WHO STAG), malaria (Malaria Reference Group and Roll Back Malaria Malaria Monitoring andEvaluationReferenceGroup). These WHO Global Health Estimates provide a comprehensive and comparable set of cause of death estimatesfromyear2000onwards,consistentwithandincorporatingUNagency,interagencyandWHO estimatesforpopulation,births,allcausedeathsandspecificcausesofdeath,including: o mostrecentvitalregistration (VR)dataforall countrieswheretheVR dataqualityisassessed as useable; Page 1
o o o updated and additional information on levels and trends for child and adult mortality in many countrieswithoutgooddeathregistrationdata improvementsinmethodsusedfortheestimationofcausesofchilddeathsincountrieswithout gooddeathregistrationdata. Updated assessments of levels and trends for specific causes of death by WHO programs and interagencygroups.Theseinclude: o TuberculosisWHO HIVUNAIDSandWHO MalariaWHO VaccinepreventablechildcausesWHO OthermajorchildcausesWHOandCHERG MaternalmortalityMMEIG CancersIARC RoadtrafficaccidentsWHO Conflict and natural disasters WHO and the Collaborating Center for Research on the EpidemiologyofDisasters(CRED)
GBD 2010 study estimates for other causes in countries without useable VR data or other nationallyrepresentativesourcesofinformationoncausesofdeath.
Because these estimates draw on new data and on the result of the GBD 2010 study, and there have beensubstantialrevisionstomethodsformanycauses,theseestimatesfortheyears20002011arenot directly comparable with previous WHO estimates for 2008 and earlier years. These are provisional estimates and will be further revised in the process of extending the series to 2012 for release at country level in late 2013. WHO and collaborators will continue to include new data and improve methods,anditisanticipatedthatsomecauseswillbesubstantiallyupdatedinthenextrevision. TheseGlobalHealthEstimatesrepresentthebestestimatesofWHO,basedontheevidenceavailableto itupuntilMay2013,ratherthantheofficialestimatesofMemberStates,andhavenotnecessarilybeen endorsed by Member States. They have been computed using standard categories, definitions and methods to ensure crossnational comparability and may not be the same as official national estimates producedusingalternate,potentiallyequallyrigorousmethods.Thefollowingsectionsofthisdocument provideexplanatorynotesondatasourcesandmethodsforpreparingmortalityestimatesbycause.
Page 2
www.childmortality.org
Page 3
U5MR estimates produced by UNIGME, U5MR and NMR data points were rescaled for all years to matchtheUNIGMEestimates. For countries where child mortality is strongly affected by HIV, the NMR was estimated initially using neonatal and child mortality observations for nonAIDS deaths, calculated by subtracting from total death rates the estimated HIV death rates in the neonatal and 159 month periods respectively, and then AIDS neonatal deaths be added back on to the nonHIV neonatal deaths to compute the total estimatedneonataldeathrate. ThefollowingstatisticalmodelwasusedtoestimateNMR:
log(NMR/1000)=0+1*log(U5MR/1000)+2*([log(U5MR/1000)]2)
with additional random effect intercept parameters for both country and region. For countries with good vital registration data covering the period 19902011, random effects parameters for slope or trend parameters were also added. Based on predictive performance evaluation using tenfold cross validation, the statistical model fitted to data point for 1990 onwards were retained and only the most recentdatapointfromeachsurveywasincluded(16).
were also considered, but the implied low level of adult mortality could not be reconciled with intercensal survival between the 1979 Afghan census and 200305 Afghan household listing, or with population estimates from 200305 Household listing and more recent surveys in 20072008 and 2011, or with intercensal estimates of the trends in fertility, and international migration based on UNHCR statistics on the number of Afghan refugees. Additionally, they would imply that Afghan adult mortality levelsweresubstantiallylowerthanthoseinneighboringcountries. As a result, the life tables for Afghanistan are based on provisional analyses by UN Population Division using the West model of the CoaleDemeny Model Life Tables with three parameters: (1) estimates of infant mortality, (2) estimates of child mortality, and (3) adjusted estimates of adult mortality (45q15) derived from (a) recent household deaths data from the 1979 census; (b) implied relationship between child mortality and adult mortality based on the UN South Asian and West model of the CoaleDemeny Model Life Tables, and (c) levels of adult mortality based on sample registration data from neighboring countriesforrecentyears.
Page 5
China Life tables for years since 2000 have been revised to take into account a faster rate of decline for adult mortality than previously projected in the World Population Prospects 2010 revision. Unpublished analyses of the China 2010 census data on adult mortality by UN Population Division have adjusted for underreporting of deaths resulting in estimates of adult mortality rates for 2010 quite similar to those reportedbytheChinaDiseaseSurveillancePointsSystem(23). Egypt Life tables have been based on official estimates of life expectancy available through 2012, and in turn derived from death registration data for Egypt. The age pattern of mortality is based on official life tables for various years from 1960 to 2010 adjusted for infant and child mortality as estimated by UN IGME,andadultmortality. SaudiArabia The World Population Prospects 2010 revision based estimates of adult mortality for Saudi Arabia using model life tables with estimates of child mortality as input. Estimates of adult mortality have been provisionally updated using adjusted death rates by age and sex from the 1999 Demographic Survey, 2004 Census and 2007 Demographic Survey adjusted for infant and child mortality, and oldage mortality.Lifetablesbasedonannualdeathsfromthe2000DemographicSurvey,aswellason2005and 2009registereddeathswerealsoconsidered. SouthSudanandSudan The former Sudan became two countries, South Sudan and Sudan, on 9 July 2011. Previously published WHO and UN life tables refer to the former Sudan. Life tables for the two Member States of South Sudan and Sudan are based on provisional analyses of population and mortality rates for the territories correspondingtothecurrentSouthSudanandSudanovertheperiod1990to2011. Infant and child mortality for South Sudan and Sudan are derived from UNIGME estimates published in 2012 (14). Life tables are based on provisional unpublished analyses of the UN Population Division, deriving adult mortality rates from estimates of infant and child mortality by assuming that the age pattern of mortality conforms to the North model of the CoaleDemeny Model Life Tables. The demographicimpactsofAIDSandconflicthavealsobeenfactoredintothemortalityestimates.
3.2 Inclusion criteria for countries with high quality death registration data
WeappliedthefollowinginclusioncriteriatodataintheWHOmortalitydatabase: Atleastfiveyearsofdataareavailableduring1998present; Thedataareavailablefor5yearagegroupstoages85andover; Thedataareforacountrywhosepopulationin2008wasgreaterthan500,000; ThedataareforacountrythatiscurrentlyaWHOMemberState; Thedatafulfillqualitycriteriapertainingtogarbagecodesandcompleteness,asdescribed below.
For 131 Member States with vital registration systems who have provided summary data to WHO, demographic techniques (such as Brass GrowthBalance method, Generalized GrowthBalance method or Bennett Horiuchi method) were first applied to assess the level of completeness of recorded mortality data in the population above five years of age. We then calculated the proportion of deaths withunderlyingcausecodedtoashortlistofsocalledgarbagecodes: symptoms,signsandilldefinedconditions(ICD10codesR00R99), injuriesundeterminedwhetherintentionalorunintentional(ICD10Y10Y34,Y87.2), illdefinedcancers(C76,C80,andC97),and illdefinedcardiovasculardiseases(I47.2,I49.0,I46,I50,I51.4,I51.5,I51.6,I51.9andI70.9).
Page 7
Table3.1.Characteristicsofuseablecountryvitalregistrationdata (Only countries fulfilling the first four inclusion criteria listed above are included in this table. ICD10 codesincludedinthegarbagecategoryaregiveninthetextabove).
Country Firstyear Lastyear Average 1998+ available usability available 2000+ 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 2004 1998 1998 1998 2000 1998 1998 1998 2004 2010 2011 2011 2011 2007 2009 2009 2010 2011 2009 2009 2009 2011 2011 2010 2011 2011 2011 2010 2011 2009 2011 2011 55% 79% 66% 95% 90% 84% 88% 88% 76% 79% 94% 94% 89% 87% 87% 90% 73% 88% 87% 59% 61% 58% 94% 97% Rangeof completeness 67% 100% 66% 100% 100% 81% 99% 100% 87% 100% 100% 100% 93% 90% 98% 96% 90% 99% 100% 72% 99% 75% 100% 100% 71% 100% 81% 100% 100% 96% 100% 100% 91% 100% 100% 100% 96% 95% 100% 98% 91% 100% 100% 73% 100% 75% 100% 100% Rangeof garbage fraction 18% 20% 3% 5% 1% 2% 10% 12% 12% 16% 6% 6% 6% 4% 8% 1% 16% 10% 12% 16% 32% 18% 5% 2% Notes
Albania Argentina Armenia Australia Austria Azerbaijan Belarus Belgium Brazil Bulgaria Canada Chile Colombia CostaRica Croatia Cuba Cyprus CzechRepublic Denmark Ecuador Egypt ElSalvador Estonia Finland
20% Excludedduetolow usability 22% Excludedduetohigh proportiongarbage 6% Excludedduetolow usability 6% 14% 34% Excludedduetohigh proportiongarbage 13% Summarizedcauselist used 15% 21% 28% Excludedduetohigh proportiongarbage 8% 11% 8% 7% 17% 9% 24% 15% 14% 23% Excludedduetolow usability 41% Excludedduetolow usability 25% Excludedduetolow usability 8% 3%
Page 8
Country Firstyear Lastyear Average 1998+ available usability available 2000+ 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 2000 1998 1998 1998 1998 1998 1999 1998 2004 2009 2010 2011 2010 2009 2011 2009 2010 2010 2010 2011 2010 2011 2010 2010 2010 2011 2010 2009 2011 2009 2011 2009 2008 2011 2011 2009 85% 53% 87% 75% 73% 94% 94% 94% 90% 90% 89% 83% 87% 90% 92% 94% 90% 95% 70% 86% 97% 89% 80% 83% 74% 82% 74% Rangeof completeness 100% 78% 100% 100% 89% 99% 100% 100% 100% 100% 100% 84% 98% 91% 99% 99% 100% 100% 93% 100% 100% 100% 84% 91% 100% 100% 100% 100% 83% 100% 100% 90% 100% 100% 100% 100% 100% 100% 89% 98% 95% 100% 100% 100% 100% 93% 100% 100% 100% 91% 93% 100% 100% 100% Rangeof garbage fraction 14% 7% 11% 24% 12% 4% 5% 5% 8% 8% 9% 3% 9% 3% 5% 2% 8% 5% 23% 13% 3% 11% 8% 10% 25% 17% 22% Notes
France Georgia Germany Greece Guatemala Hungary Iceland Ireland Israel Italy Japan Kazakhstan Kuwait Kyrgyzstan Latvia Lithuania Mauritius Mexico Montenegro Netherlands NewZealand Norway Panama Philippines Poland Portugal Qatar
16% 69% Excludedduetolow usability 14% 27% Excludedduetohigh proportiongarbage 22% Excludedduetohigh proportiongarbage 7% 6% 8% 14% 12% 13% 11% Summarizedcauselist used 14% 8% 11% 6% 15% 6% 28% Excludedduetolow usability 15% 4% 12% 14% 13% 28% Excludedduetohigh proportiongarbage 22% 32% Excludedduetohigh proportiongarbage
Page 9
Country Firstyear Lastyear Average 1998+ available usability available 2000+ 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 2011 2011 2011 2010 2011 2011 2010 2010 2009 2011 2006 2010 2010 2010 2006 2008 2011 2010 2008 2009 2005 2009 85% 88% 92% 95% 72% 74% 94% 89% 68% 89% 55% 89% 89% 84% 48% 95% 96% 93% 93% 83% 83% 86% Rangeof completeness 90% 89% 99% 100% 84% 74% 100% 99% 81% 100% 74% 100% 100% 96% 78% 100% 100% 100% 100% 100% 85% 93% 100% 91% 100% 100% 89% 84% 100% 100% 88% 100% 74% 100% 100% 98% 88% 100% 100% 100% 100% 100% 87% 95% Rangeof garbage fraction 13% 2% 0% 4% 12% 2% 4% 9% 19% 9% 23% 10% 10% 9% 39% 2% 3% 6% 7% 16% 2% 7% Notes
Republicof Korea Republicof Moldova Romania Russian Federation Serbia Singapore Slovakia Slovenia SouthAfrica Spain SriLanka Sweden Switzerland TFYR Macedonia Thailand Trinidadand Tobago Ukraine United Kingdom UnitedStates ofAmerica Uruguay Uzbekistan Venezuela (Bolivarian Republicof)
21% 7% 8% 6% Summarizedcauselist used 18% 4% 11% 12% 32% Excludedduetolow usability 13% 32% Excludedduetolow usability 12% 13% 15% 54% Excludedduetolow usability 5% 6% Summarizedcauselist used 8% 10% 17% 6% Summarizedcauselist usedforsomeyears 9%
Asummaryusabilityscorewascalculatedasfollows: (PercentUsable)=Completeness(%)*(1ProportionGarbage) All countries with a mean percent usable below 70% during the period 2000 to latest available year wereexcluded(seeTable3.1). Thequalityofcauseofdeathcodingwasfurtherinvestigatedintheremainingcountries.Theproportion of deaths assigned to an expanded list of illdefined causes (Table 3.2) was calculated for each year in the period 20002011. For the period 20052011 countries had reported an average of 5 years of data. Data from a country were excluded if the average proportion of illdefined causes was above 25% for 20052011 (if available) or 20002004 (if more recent data were not available). Based on this analysis, data from Argentina, Azerbaijan, Bulgaria, Greece, Guatemala, Poland, and Qatar were excluded (Table 3.1). Table3.2.Expandedlistofgarbagecodes
ICD10code(s) Description A40A41 D65 E86 I10 I269 I46 I472 I490 I50 I514 I515 I516 I519 I709 I99 J81 J96 K72 N17 N18 N19 P285 Streptococcalandothersepticaemia Disseminatedintravascularcoagulation[defibrinationsyndrome] Volumedepletion Essential(primary)hypertension Pulmonaryembolismwithoutmentionofacutecorpulmonale Cardiacarrest Ventriculartachycardia Ventricularfibrillationandflutter Heartfailure Myocarditis,unspecified Myocardialdegeneration Cardiovasculardisease,unspecified Heartdisease,unspecified Generalizedandunspecifiedatherosclerosis Otherandunspecifieddisordersofcirculatorysystem Pulmonaryoedema Respiratoryfailure,notelsewhereclassified Hepaticfailure,notelsewhereclassified Acuterenalfailure Chronicrenalfailure Unspecifiedrenalfailure Respiratoryfailureofnewborn C76,C80,C97 Illdefinedcancersites
Y10Y34,Y872 Externalcauseofdeathnotspecifiedasaccidentallyorpurposelyinflicted
Page 11
3.3 Redistribution of unknown sex/age and garbage codes and adjustment for incomplete death registration
First, deaths of unknown sex prorata within causeage groups of known sexes were redistributed, and then deaths of unknown age prorata within causesex groups of known ages. Deaths coded to garbage codes were reassigned using previously published methods (25). We redistributed deaths coded to symptoms, signs and illdefined conditions prorata to all noninjury causes of death, and injuries with undetermined intent prorata to all injury causes of death. Cancers with unspecified site were redistributed prorata to all sites excluding liver, pancreas, ovary, and lung. Additionally, we redistributed cancer of uterus, part unspecified (C55) prorata to cervix uteri (C53) and corpus uteri (C54). Illdefined cardiovascular causes were redistributed to ischaemic heart disease and other cardiovascular causes of death. Finally, the total number of deaths was adjusted for incomplete recordingofdeathsusingthecompletenessestimatesdescribedinSection3.2.
Page 12
Table 3.3. Short cause list used for vital registration data coded using ICD9 or ICD10 abbreviated cause lists
GHE code 1 3 9 20 38 39 42 49 60 61 62 63 64 65 66 68 70 71 72 80 110 117 121 126 140 151 152 153 160 161 162 163 Shortlistcausecategory I. Communicable,maternal,perinatalandnutritionalconditions A1.Tuberculosis A3.HIV/AIDS
A.Infectiousandparasiticdiseases B.Respiratoryinfections B1.Lowerrespiratoryinfections C.Maternalconditions D.Neonatalconditions II.Noncommunicablediseases A.Malignantneoplasms A1.Mouthandoropharynxcancers A2.Oesophaguscancer A3.Stomachcancer A4.Colonandrectumcancers A5.Livercancer A7.Trachea,bronchusandlungcancers A9.Breastcancer A10.Cervixutericancer A13. Prostatecancer
C.Diabetesmellitus H.Cardiovasculardiseases I.Respiratorydiseases J.Digestivedisorders K.Genitourinarydiseases N.Congenitalanomalies III.Injuries A.Unintentionalinjuries A1.Roadinjury B1.Selfharm B2.Interpersonalviolence B3.Collectiveviolenceandlegalintervention B.Intentionalinjuries
82+94 E/F.Mentalandneurologicaldisorders
population covered by these two systems. The sample vital registration system data for years 1987 to 2010 was provided in separate tabulations for urban and rural sampled populations, with more urban than rural sampling. The urban and rural crude deaths rates by age, sex and cause were weighted for each year using the UN Population Divisions estimated urban and rural population fractions, and the resulting death rates reapplied to the UN total estimated population by age and sex. The DSP sample sites are considered to be nationally representative and the resulting total deaths by age, sex and cause were not reweighted. For both sets of data, annual data were rescaled so total deaths by age and sex matchedtheestimatedallcauseenvelopesforChina(seeSection2.5). Table 3.4. Total deaths and population covered by the Chinese vital registration system (VR) and the DiseaseSurveillancePointssystem(DSP)
Vitalregistration system Year 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
datanotavailable.
Numberofdeaths 711,946 626,392 295,906 310,826 379,057 475,289 471,219 505,021 558,915
Population
Both sets of data were assessed and compared for suitability in estimating 20002011 causespecific mortalityfor Chinaat thenational level.As seenin Figure3.1, bothsets ofdata gavequite similar cause distributions at major cause group level by age, across the period 20002010. Additionally, comparison for more detailed major causes of death did not give any clear indication that one data set was of systematicallyhigherqualitythantheother.Wethereforebasedtheupdateofcauseofdeathestimates forChinaonanaverageoftheestimatesfromthetwosystems. For all except the leading causes of death, there are considerable fluctuations across 5year age groups and year in numbers of deaths, due to stochastic variation and perhaps also variations in recording causeofdeathfromyeartoyearorsamplesitetosamplesite.Inordertosmooththesefluctuationsand to estimate underlying trends, cubic spline smoothing was used as follows. For the VR data, cubic spline curveswerefittedtoagesexcausespecificdeathsforyears19872010usinganegativebinomialmodel with population as offset and with knot points at years 1992, 1997, 2003, and 2007. For the DSP data, cubic spline curves were fitted to agesexcause specific deaths for years 19952010 using a negative binomial model with population as offset and with knot points at years 2004, 2007 and 2010. Final estimates for China were calculated as the average of the fitted spline estimates from VR and DSP for years20002011. World Health Organization Page 15
Figure 3.1. Sample vital registration data (VR) and Disease Surveillance Points data (DSP), China: comparisonofcausefractionsforthreemajorcausegroupsbyage,late1990s,2005and2010
The resulting causespecific estimates were further adjusted with information from WHO technical programmes and UNAIDS on specific causes (see Section 5) and from the GBD 2010 for certain specific subcause categories where deaths were either not recorded or recorded to only selected categories in the DSP and/or VR datasets. GBD 2010 analyses were used for GHE causes 59 (STDs), 20 (hepatitis C), 26 (leishmaniasis), 3436 (intestinal nematode infections), 115 (inflammatory heart diseases), and 119 (asthma). Additionally, DSP broad cause group totals were redistributed to detailed subcauses using GBD 2010 cause fractional distributions for the following categories: 82+94 (mental and behavioural disordersandneurologicalconditions),134(musculoskeletaldisorders)and147(oralconditions).Rabies deaths were revised using data on reported human rabies deaths from the Chinese Center for Disease ControlandPrevention(31). For estimates of causes of death under age 5, a separate analysis was undertaken based on an analysis of 206 Chinese communitybased longitudinal studies that reported multiple causes of child death (see Section 4.5 below. The CHERG conducted a systematic search of publically available Chinese databases in collaboration with researchers from Peking University. Information was obtained from the Chinese Ministry of Health and Bureau of Statistics websites, Chinese National Knowledge Infrastructure (CNKI) database and Chinese Health Statistics Yearbooks published between 19902008. A model was developed to assign the total number of child deaths to provinces, age groups and main causes of child death.
Page 16
India Analysis of causes of death for India was based on data over a period of 3years (20012003) recorded bytheMillionDeathStudy(32,33),acomprehensivestudybasedonverbalautopsythatassignedcauses to all deaths in areas of India covered by the Sample Registration System. The Sample Registration System monitors a representative sample population of 6.3 million people in over 1 million homes in India. The 1991 census was used to randomly select 6671 areas from approximately 1 million having about1000inhabitantsineach. In 2001 the Indian Registrar General Surveyor introduced an enhanced form of verbal autopsy for assessing the cause of death. Verbal autopsy is a method of ascertaining the cause of death by interviewing a family member or caretaker of the deceased to obtain information on the clinical signs, symptoms and general circumstances that preceded the death. Details of methods and validation have been reported elsewhere (33). Verbal autopsy reports were independently coded to ICD10 categories byatleast twoofa totalof130 physicianstrainedin ICD10coding.In caseofdisagreement onthe ICD 10 codes at the chapter level, reconciliation between reports was conducted, followed by a third senior physiciansadjudication. A total of 136,000 deaths were enumerated between January 2001 and December 2003. Verbal autopsies could not be conducted for 12% of the deaths for reasons such as family migration or change of residence. An additional 9% of the reports could not be coded because of data quality problems, resultingafinaldatasetof122,848codedrecords. Thecausespecificproportionofdeathsineachfiveyearagecategoryfrom0to79yearsandforpeople aged 80 years and over was weighted by the inverse probability of a household being selected within rural and urban subdivisions of each state to account for the sampling design. National estimates for deathsandmortalityrateswerebasedonUnitedNations2005estimatesforIndia,byage,sexandarea. Figure 3.2. Percentage of deaths by cause and age for India: comparison of final GHE estimates for year2002withnationallevelresultsfromtheMillionDeathStudy,20012003
GlobalHealthEstimates:India,2002
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Age(years) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 Age(years)
MillionDeathStudy:India,20012003
Suicide,homicideandconflict Otherunintentionalinjuries Roadinjury Othernoncommunicable Chronicrespiratorydiseases Cancers Cardiovasculardiseases Maternal,neonatal,nutritional Otherinfectiousdiseases Lowerrespiratoryinfections Diarrhoealdiseases HIV,TBandmalaria
Page 17
TheGHEanalysisisbasedontheresultingnationallevelcausespecificmortalityproportionsderivedfor GHE cause categories from the Million Death Study. The mapping of the MDS cause categories to GHE cause categories, and the use of GBD 2010 analyses to redistribute deaths to detailed subcause categories is summarized in Annex Table E. GHE cause categories 26 (leishmaniasis) and 124 (appendicitis)werealsoestimatedusingGBD2010results. The resulting causespecific estimates were further adjusted with information from WHO technical programmes and UNAIDS on specific causes (see Section 5) and adjusted to match WHO estimates of agesex specific allcause mortality for India in 2002. Causespecific trends for India estimated in the GBD2010study(26)wereusedtoprojectcausefractionsforwardsto2011andbackwardsto2000. Figure 3.2 provides a comparison of the final proportional distributional estimates of deaths by cause and age for India in the year 2002 with the original distributions in the Million Death Study for 2001 2003.
Page 18
Causespecific estimates of deaths for children under age 5 were estimated for 17 cause categories using methods described elsewhere by Liu et al. (34) and on the WHO website (35). These previously published estimates for years 20002010 were updated to take account of revisions in child mortality levels (14), as well as causespecific estimates for HIV, tuberculosis, measles and malaria deaths (as described in Section 5). Inputs to the multivariate cause composition models were also updated as describedbelow.
4.1 Causes of under 5 death in countries with good death registration data
Death registration data were used directly for estimating causes of neonatal and under 5 child deaths for countries with good quality vital registration (VR) data with population coverage of >80%. VR data were considered as of good quality if the following criteria were met: (a) reasonable distribution of deaths by cause were reported without excessive use of implausible codes or certain codes, and (b) sufficient details of the coding was provided so that deaths could be grouped into appropriate categories used in the analysis. For countries with adequate death registration, data on causes of child deaths were extracted from the WHO mortality database, adjusted for coverage incompleteness where needed, and grouped according to the standard International Classification of Diseases, 10th revision (ICD10). For earlier years when ICD9 codes were used, a mapping system was applied to convert them into ICD10 codes (34,webappendix). Certain neonatal codes were reassigned from illdefined codes to more plausible codes (see Annex Table C). Annual data for years 2000 to the latest available year were included with data closest to the estimating year used where possible. Where the latest year available was earlier than 2011, the cause distribution for the latest available year was assumed to apply for subsequentyear(s),whichwasthenappliedtotheagespecifictotalnumberofchilddeaths.
4.3 Causes of child death at ages 159 months low mortality countries
For 51 low mortality countries without VR data or with VR data not meeting quality criteria (see Section 4.1),the cause distribution was estimated using a multinomial model applied to death registration data. Thismultinomialmodelappliedtodeathregistrationdatawasgenerallyusedforcountrieswithaverage U5MR<35fortheperiod20002010. Fortheestimatesforyears20002011,thepreviousvitalregistrationbasedmulticausemodel(VRMCM) model was revised to include additional death registration data and to update time series for covariates and extend them to 2011. The choice of covariates included in the model was not revisited for this regionallevel update. The multinomial logistic regression model was estimated using death registration data from countries with >80% complete cause of death (CoD) certification for years 19902011 to estimatetheproportionofdeathsduetopneumonia,diarrhea,meningitis,injuries,perinatal,congenital anomalies,otherNCDsandothercauses. The current version of the model used death registration data for the years 1990 to 2011, including 1,123 data points, representing 63 countries. The model included the following covariates that were determinedapriori:U5MRs,GNIpercapita(PPP,$international),WHOEuropeanandAmericanregions. Adjustments for the scalingup of Hib vaccine occurred within the model. The proportional distribution ofcausesofdeathwasthenappliedtotheHIVfreeandmeaslesfreeenvelopeforchildren159months ofage.JackknifeandMonteCarlosimulationmethodswereusedtoestimateuncertainty.
4.4 Causes of child death at ages 159 months high mortality countries
For79highmortalitycountries(averageU5MR>35fortheperiod20002010),thecausedistributionwas estimatedusingamultinomialmodelappliedto(largely)verbalautopsy(VA)datafromresearchstudies (34,36,37).Themulticausemodelfordeathsatages159monthswasusedtoderivemortalityestimates for seven causes of postneonatal death, including pneumonia, diarrhea, malaria, meningitis, injuries, congenital malformations, causes arising in the perinatal period (prematurity, birth asphyxia and trauma, sepsis and other conditions of the newborn), and other causes, based on 113 data points from 74 studies of postneonatal deaths from 33 countries that met inclusion criteria2. Studies were predominantly from lower income high mortality countries. Malnutrition deaths were included in the other cause of death category. Deaths due to unknown causes were excluded from the analysis. Deaths duetomeaslesandHIV/AIDSwereestimatedseparately. The resulting causespecific inputs were adjusted countrybycountry to fit the estimated 159 month death envelopes (excluding HIV and measles deaths) for corresponding years and then estimates were furtheradjusted for intervention coverage(pneumonia and meningitis estimatesadjusted for use of Hib vaccine; malaria estimates adjusted for insecticide treated mosquito nets (ITNs)). This method was used for countries without useable death registration data and with U5MR>26 and gross national income per capitalessthan$7,510.
Studies conducted in year 1980 or later, a multiple of 12 months in study duration, cause of death available for more than a single cause, with at least 25 deaths in children <5 years of age, each death represented once, and less than 25% of deaths due to unknown causes were included. Studies conducted in sub-groups of the study population (e.g. intervention groups in clinical trials) and verbal autopsy studies conducted without use of a standardized questionnaire or the methods could not be confirmed were excluded from the analysis.
2
Page 20
Page 21
5.1 Tuberculosis
Forcountrieswithdeathregistrationdata,tuberculosismortalityestimatesweregenerallybasedonthe most recently available vital registration data. For other countries, total tuberculosis deaths were derived from latest published WHO estimates (41), together with more detailed unpublished age distributionsbasedontheVRdataandnotificationsdata.
5.3 Malaria
CountriesoutsidetheWHOAfricanRegionandlowtransmissioncountriesinAfrica3. Estimates of the number of cases were made by adjusting the number of reported malaria cases for completeness of reporting, the likelihood that cases are parasitepositive and the extent of health service use. The procedure, which is described in the World Malaria Report 2012 (42), combines data reported by National Malaria Control Programs (reported cases, reporting completeness, likelihood that cases are parasite positive) with those obtained from nationally representative household surveys on health service use. If data from more than one household survey was available for a country, estimates of health service use for intervening years were imputed by linear regression. If only one household survey was available then health service use was assumed to remain constant over time; analysis summarized in the World Malaria Report 2008 (43) indicated that the percentage of fever cases seeking treatment in public sector facilities varies little over time in countries with multiple surveys. Such a procedureresultsinanestimatewithwideuncertaintyintervalsaroundthepointestimate. The number of deaths was estimated by multiplying the estimated number of P. falciparum malaria cases by a fixed case fatality rate for each country as described in the World Malaria Report 2012 (42). This method is used for all countries outside the African Region and for countries within the African Region where estimates of case incidence were derived from routine reporting systems and where malaria causes less than 5% of all deaths in children under 5. A case fatality rate of 045% is applied to the estimated number of P. falciparum cases for countries in the African Region and a case fatality rate of 03% for P. falciparum cases in other Regions. In situations where the fraction of all deaths due to malaria is small, the use of a case fatality rate in conjunction with estimates of case incidence was considered to provide a better guide to the levels of malaria mortality than attempts to estimate the fractionofdeathsduetomalaria. Somalia,SudanandhightransmissioncountriesintheWHOAfricanRegion. Child malaria deaths were estimated using the VAMCM described in Section 4.4. The VAMCM derives mortality estimates for malaria, as well as 7 other causes (pneumonia, diarrhea, congenital
3
Botswana, Cape Verde, Eritrea, Madagascar, Namibia, Swaziland, South Africa, and Zimbabwe
Page 22
malformation, causes arising in the perinatal period, injury, meningitis, and other causes) using multinomial logistic regression methods to ensure that all 8 causes are estimated simultaneously with the total cause fraction summing to 1. Malaria deaths were retrospectively adjusted for coverage of insecticidetreated nets (ITNs) and use of Haemophilus influenzae type b vaccine (34). The bootstrap method was employed to estimate uncertainty intervals by resampling from the studylevel data to estimatethe distribution of thepredicted percent of deathsdue to each cause. The estimated malaria mortality rate in children under 5 years for a country was used to determine malaria transmission intensityandthecorrespondingmalariaspecificmortalityratesinolderagegroups(43).
5.5 Measles
To estimate deaths attributable to measles, a new model of measles mortality developed by WHO Department of Immunization, Vaccines and Biologicals (IVB) was used to first estimate countryand yearspecific cases using surveillance data (44). The improved statistical model firstly estimates measles cases by country and year using surveillance data and making explicit projections about dynamic transitionsovertimeaswellasoverallpatternsinincidence. The age distribution of measles cases are then estimated using a logistic regression function fitted to 172,191 measles cases with data on age at infection from 102 countries over 20052009 extracted from WHO's monthly measles casebased reporting system. Two explanatory variables were included in the regression:1)the5yearrollingaverageofestimatedMCV1coverage,categorizedin<60%,6084%,and 85100%;and2)geographicregionclassifiedinto7groups. Countryspecific measles casefatality ratios (CFRs) for children 14 years of age were taken from a comprehensive review of communitybased studies (45). This review included 102 field studies conducted in 29 countries during the period 19742007. The set of CFRs were revised for two countries (India and Nepal) where additional studies have been published subsequent to the review (46, 47). The sameCFRswereusedforinfantsandforchildrenaged14years.Fortheperiod20002011,weassumed thatagespecificCFRsarenotdecliningovertime. Agespecific deaths are aggregated to derive measles deaths for all children below five and for ages five and over. The new method takes into account herd immunity and produces results that are fairly consistent with previous ones (48). Uncertainty is estimated by bootstrap sampling from the distribution of incidence and age distribution estimates. Updated estimates of measles deaths by age and country for years 20002011 were prepared using the above methods at the end of 2012 and summary results published in the Weekly Epidemiological Record (49).These were used for this update ofGHEcausesofdeathforyears20002011. Forcountriesexperiencingmeaslesoutbreaks,themeaslesdeathsweresplitintooutbreakandendemic deaths, the latter of which were smoothed using local regression (50). For the ages of 159 months, the endemicmeaslesdeathsandAIDSdeathswereaddedtothemeaslesandAIDSfreeallcausedeathsfor whichtheVAMCMderivedcausefractionswereapplied.Themeaslesoutbreakdeathswereaddedback attheend.Inplaceswheretheoutbreakdeathsresultedinanincreaseintheallcausedeathsby10%or World Health Organization Page 23
more, the original survey data were screened to examine whether a real increase in child mortality was indicated for the outbreak year. If there were survey data available for the years around the outbreak but no evidence of an increased mortality, the measles outbreak deaths were truncated at 10% of the allcause deaths. This was only necessary in few countries, almost all of which are in Africa and all occurredintheearly2000swhenmoremeaslesdeathswereestimated.
5.6 Schistosomiasis
ForthelastWHOupdateofburdenofdiseaseforyear2004(3),theincidenceandprevalenceofcasesof schistosomiasis infection were separately estimated by country for S.mansoni, S.haematobium and S.japonicumplus S.mekongi. The GBD 2004 estimated that schistosomiasis was responsible for around 41000deathsglobally(excludingattributablecancerdeaths)and36000insubSaharanAfrica,although others have argued that the figure should be much higher (51). Van der Werf et al (52), using limited data from Africa, estimated that schistosomiasis caused 210000 deaths annually. For the GBD 2004 update (3), very limited available data was used to conservatively estimate annual case fatality rates for prevalent cases at 0.01% for S.mansoni, 0.02% for S.haematobium, and 0.03% for S.japonicum and S.mekongi.Therewereestimatedtobe261millionprevalentcasesofschistosomiasisinfectionin2004. The GBD 2010 study estimated that there 11,650 deaths due to schistosomias in 2010, of which 1,813 were in the Middle East and North Africa, and only 61 in subSaharan Africa in 2010. Divided by the numbers of prevalent cases estimated by the GBD 2010, the implied case fatality rates for the Middle East and North Africa, and for Latin America are 0.01% and 0.02% respectively. In comparison, the implied African case fatality rate is almost 400 times smaller. Implied case fatality rates for nonAfrican regionsintheGBD2010weregenerallyconsistentwiththosepreviouslyestimatedbyWHOfortheyear 2004.Revised case fatality ratesof 0.0075% for S.mansoni, 0.015%for S.haematobium were applied to the prevalence rates estimated by GBD 2010 (53) to revise the estimates of schistosomiasis deaths for GHE. This resulted in an estimate of 17,600 deaths in subSaharan Africa and 23,300 deaths globally in 2011.
5.8 Cancers
Causespecific estimates for cancer deaths were derived from Globocan 2008 (55) for countries without useable death registration data. Sitespecific deaths were projected back to year 2000 and forwards to year2011usingtrendestimatesfromtheGBD2010(26). World Health Organization Page 24
Karposi sarcoma was excluded from the Globocan estimates as this is almost entirely a manifestation of HIV/AIDS, already included in the estimates for HIV/AIDS deaths. Deaths due to nonmelanoma skin cancerswereincludedintheseestimatesalongwithmelanoma,unlikeinGlobocan2008.
5.10 Epilepsy
The Million Death Study for India (32,33) recorded relatively high proportions of epilepsy deaths, resulting in an initial GHE estimate of 73,600 epilepsy deaths in India in 2010 compared to an estimated 21,650bytheGBD2010.GBD2010estimatesofuntreatedidiopathicepilepsyprevalencewereusedto calculate implied regional case fatality rates (CFR) and the implied Indian CFR of 0.34 was substantially higher than those for South East Asia (0.09) or the Middle East and North Africa (0.05). Indian epilepsy deaths were adjusted downwards to give an implied case fatality rate of 0.17 (close to the global average),resultinginanestimated35,480epilepsydeathsforIndiain2010.
5.11.1 Countries with death registration data This group includes 87 countries with death registration data meeting one of the following completeness criteria, viz. completeness for the year estimated at 80% or more, or average completenessforthedecadeincludingthecountryyearwas80%ormore. Thesecountriesfellintothreecategories: 1. Forcountrieswithdeathregistrationdatafortheyear2010whichexceededthenumberofroad traffic deaths reported in the survey death registration data was used. There were 33 countriesinthiscategory. 2. ForcountrieswherethelatestdeathregistrationdatasubmittedtoWHOwasearlierthan 2010, but not earlier than 2005 deaths for 2010 were estimated based on a projection of the most recent death registration data using the trends obtained through the survey. There were 40 countriesinthiscategory. 3. For countries where the reported road traffic deaths for 2010 obtained through the survey exceeded the estimate based on death registration data: The reported road traffic deaths (adjusted to the 1 year definition) were used. There were 12 countries in this category. There wereanadditional2countrieswherereporteddataforearlieryearswereprojectedto2010and usedbecausetheyexceededthedeathregistrationnumbers. 5.11.2 Countries with other sources of information on causes of death For India, Iran, Thailand and Viet Nam, data on total deaths by cause were available for a single year or an earlier recent single year or group of years (33,5860). For these countries, the regression method described below was used to project forward from the most recent year for which an estimate of total roadtrafficdeathswereavailable. 5.11.3 Countries with populations less than 150 000 For 13 small countries with populations of less than 150 000 people the deaths reported in the survey wereuseddirectlywithoutadjustment. 5.11.4 Countries without eligible death registration data For 78 countries which did not fall into any of the above groups, a regression model was used to estimate total road traffic deaths. The regression model produced estimates of total road traffic deaths accordingtotheacceptedInternationalClassificationofDiseasedefinition,whichcountsalldeathsthat follow from a road traffic death, regardless of the time period in which they occur (unlike many official road traffic surveillance data sources, where road traffic death data is based on a 30day definition following a road traffic crash). Where total deaths reported by Member States surveillance systems weregreaterthanthedeathsestimatedfromtheregression,thesewereused. Threeclassesofmodelsweretestedandanegativebinomialcountsmodelwaschosen:
ln N C 1 X 1 2 X 2 .... n X n ln Pop
(1)
where N is the total road traffic deaths (for a countryyear), C is a constant term, Xi are a set of explanatory covariates, Pop is the population for the countryyear, included as an offset, and is the negative binomial error term. This model was estimated using death registration data for the period 19502010 that were 80% or more complete for a given year or where the average completeness for
Page 26
thelastdecadewasgreaterorequalto80%.Italsoincludednationallyrepresentativeverbalautopsyor sampledeathregistrationdataforIndia,ChinaandVietnam. Threemodels(A,BandC)werechosenthathadgoodinsampleandoutofsamplefit,andforwhichall the covariates (see Table 5.1) were statistically significant at the 95% level. The final estimates were basedontheaveragepredictionsofthesethreebestmodels. Age distributions for road injury deaths were based on regional age distributions estimated in the GBD 2010study(26). Table5.1.Covariatesusedinthemodelforroadinjurydeaths
Independent variables ln(GDP) Description WHOestimatesofGrossDomesticProduct (GDP)percapita(internationaldollarsor purchasingpowerparitydollars,2005base) Totalvehiclesper1000persons Totalroads(km)per1000hectares Themaximumnationalspeedlimitsonrural roads(km/h)fromWHOquestionnaire Themaximumnationalspeedlimitsonurban roads(km/h)fromWHOquestionnaire Healthsystemaccessvariable(principal componentscorebasedonasetofcoverage indicatorsforeachcountry) Litersofalcohol(recordedplusunrecorded) peradultaged15+ Proportionofpopulationaged1516years Percentoftotalvehiclesthataremotorbikes Controlofcorruptionindex(unitsrangefrom about2.5to+2.5withhighervalues correspondingtobettercontrolofcorruption Existenceofnationalpoliciesthatencourage walkingand/orcycling Totalpopulation(usedasoffsetinnegative binomialregression Sourceofinformation WHOdatabase Included inmodels A,B,C
GSRRSsurveysandWHOdatabase InternationalFuturesdatabase(63) GSRRSsurvey(57) GSRRSsurvey(57) InstituteforHealthMetricsand Evaluationdataset(61) WHOdatabase WorldPopulationProspects2010 revision(UNDESA) GSRRSsurvey(57) WorldBank(62),International Futuresdatabase(63) GSRRSsurvey(57) WorldPopulationProspects2010 revision(12)
A,B,C A,B,C B B
C A,B,C
Page 27
Table 5.2. Estimated total global injury deaths (thousands) due to conflict: comparison of various time seriesandWHOestimates
Year GBD1990(a) WHO20002008 UCDCPRIO WHO2013(i) (h) 95 85 30 18 34 28 138 122 95 69 84 57 IHMEGBD2010(j)
63 53 26 18
(a)EstimatesandprojectionsbyMurrayandLopez(74) (b)WorldHealthReport2001(87)andWorldreportonviolenceandhealth(88). (c)WorldHealthReport2002(1) (d)RevisionforDiseaseControlPrioritiesStudy(2) (e)Globalburdenofdisease:2004update(3) (f)WorldHealthStatistics2007(89) (g)WHOestimatesofcausesofdeathforyear2008(4) (h)Sumofmainestimatesofconflictdeathsforstatestate,statenonstateandonesidedconflicts(7880) (i)RevisedWHOestimatesforyears19902011asdocumentedhere. (j)IHMEGlobalBurdenofDiseaseStudy2010(26).
The revised WHO estimates for total conflict deaths (in the column WHO 2013) are considerably lower than the previous WHO estimates for years 20002008 which used the earlier higher adjustment factor for underreporting, which in turn are lower than the previous estimates and projections in the original GBD study (74). The recently estimates for conflict deaths published by IHME in the GBD 2010 study, shown in the rightmost column, are considerable lower than the revised WHO estimates. For the year 2010, the IHME estimates are also lower than the main estimate from the UCDCPRIO databases for the same year. The IHME methods were based on a regression analysis of available allcause mortality data for countryyears in which battle deaths were reported in various databases. Lozano et al (26) cite (90) formoredetaileddocumentationoftheirmethods. The revised WHO estimates for conflict deaths were taken into account in preparing final allcause mortality envelopes for Member States for years 19902011 as follows. For countryyears where death ratesfromconflictordisastersexceeded1per10,000population,theestimatedannualagesexspecific conflict deaths were added to the life table death rates for the relevant year. In cases of extended conflicts where death rates fluctuated above and below 1 per 10,000, only the death rate in excess of 1 per10,000wasaddedtorelevantyears.
Page 29
Previous WHO comprehensive estimates of causes of death have relied on causeofdeath modelling and available data on cause of death distributions within each analysis subregion to estimate causes of death for countries without useable data and where WHO causespecific analyses were not available (2, 3). The IHME developed covariate based estimation models for a large number of single causes as inputstoitsoverallestimationofnumbersofdeathsbycountry,cause,ageandsexforyears19902010 in the GBD 2010 study (810). Results from these models are used as inputs to WHO Global Health Estimates for causes of death not addressed by WHO and UN Interagency estimation processes and wherecountriesdidnothaveuseabledeathregistrationdata,asdescribedbelow. SixdifferentmodellingstrategieswereusedbyIHMEforcausesofdeathdependingontheavailabilityof data (26,webappendix). For all major causes of death except HIV/AIDS and measles, IHME used ensemble modelling to create a weighted average of many individual covariatebased models (ranging from hundreds to thousands in some cases) for each specific cause (26,91). IHME cause of death estimation methods are thus complex and highly computerintensive. The overall outofsample predictive validity of the ensemble is usually not much different to that of the topranked model, but uncertaintyrangesaregenerallymuchwiderandmoreplausiblethanforsinglemodels. IHME results for priority causes such as HIV, TB, malaria, cancers, maternal mortality, child mortality differtovaryingdegreesfromthoseofWHOandUNagencypartners.Inpart,thisreflectsdifferencesin modelling strategies, but also the inclusion by IHME of data from verbal autopsy (VA) studies which has been mapped to ICD categories using IHMEdeveloped computer algorithms. WHO aims to work with IHME and expert groups to further improve data and methods, which requires that all input data and detailed analysis methods and results are made available. Figure 6.1 provides a comparison of major cause group death rates for the GBD 2010 and WHO GHE results for year 2010 for seven broad regional groupings. To ensure that the results of all the singlecause models summed to the allcause mortality estimate for each agesexcountryyear group, IHME applied a final step called CoDCorrect to rescale the cause specific estimates. This was done using repeated random draws from the uncertainty distributions of each single cause and from the allcause envelope, and proportionately rescaling each single cause estimate so they collectively summed to the envelope estimate. The overall effect is to squeeze or expandcauseswithwideruncertaintyrangesmorethanthosewithnarroweruncertaintyranges. GBD2010results,postCoDCorrect,wereusedasinputstoestimatecausefractionsbycountry,age,sex andyearforcausesofdeathatagesfiveyearsandaboveforwhichdeathregistrationdataand/orWHO and UN Interagency analyses (described in Section 5) were not available. For this set of causes, GBD 2010 countrylevel estimates for death rates at ages 5 and over for years 1990, 1995, 2000, 2005 and 2010 were interpolated to death rates for all years in the range 20002011 using cubic spline interpolation of log(death rates. Cause fraction distributions were then computed for the set of causes excluding WHO/Interagency causespecific estimates. For countries where these cause fractions were used (see Annex Table G), they were applied to the countrylevel residual mortality envelopes by age and sex after the WHO/Interagency causespecific estimates were subtracted from the WHO allcause envelopes. Table 6.1 summarizes the overall percentage change in the GBD 2010 estimates for each of these residual causes resulting from the above process. This provides a rough metric of how much inconsistency there is between the GBD 2010 and the GHE 2010 estimates for ages five and over as a resultofdifferencesinallcauseenvelopesandWHO/Interagencyestimatesforspecificcauses. World Health Organization Page 30
Figure6.1.ComparisonofGHEandIHMEdeathratesper100,000population,majorcausegroups,2010
Page 31
Table 6.1. Ratio of GHE total deaths for residual causes to GBD 2010 total deaths for residual causes, low and middleincome countries without useable death registration data, by WHO Region and age group,2000and2011 514 3.70 1.75 1.48 0.95 1.29 1.38 2.47 Ratio2000 1549 5069 1.50 1.36 1.25 1.10 0.91 0.84 0.89 0.98 1.09 1.04 0.78 0.77 1.20 1.06 70+ 1.44 1.01 1.09 1.11 1.11 1.00 1.16 514 4.15 2.05 1.65 0.54 1.05 1.21 2.77 Ratio2011 1549 5069 1.57 1.23 1.28 1.14 0.81 0.88 0.62 0.83 0.83 1.02 0.58 0.68 1.12 1.02 70+ 1.36 1.07 1.12 1.08 1.10 0.95 1.15
Page 32
Uncertainty of estimates
Countrylevel estimates of mortality for 2004 and 2008 previously released on the WHO website included guidance to users on the data sources and methods used for each country, in terms of four levelsofevidence.ComprehensiveuncertaintyrangeshavenotyetbeenaddressedfortheGHEcauseof death estimates although uncertainty ranges are available for many of the component analyses for specific causes (refer to the detailed documentation of sources in Sections 4 and 5). General guidance on the quality and uncertainty of these cause of death estimates for years 20002011 is provided in termsof the quality ofdata inputs and methodsused. These are broadlysummarized for WHO Member StatesinAnnexTableFforgeneralmortalityandcauseofdeathmethods. WHOsadoption of healthestimates is affectedby a numberof factors, includinga country consultation process for countrylevel health estimates, existing multiagency and expert group collaborative mechanisms, and compliance with minimum standards around data transparency, data and methods sharing. More detailed information on quality of data sources and methods, as well as estimated uncertaintyintervals,isprovidedinreferencedsourcesforspecificcauses(Sections4and5). Calculated uncertainty ranges depend on the assumptions and methods used. In practice, estimating uncertainty in a consistent way across health indicators has had limited success (i.e., estimates with uncertainty typically reflect some, but not all, source of uncertainty). Most methods for estimation of uncertainty rely on statistical techniques to assess variations across observations and take into account sampling error but are less successful in dealing with unknown systematic bias in observations. In particular, there is not yet sufficient research or consensus on the interpretation and use of verbal autopsy studies to ensure that systematic bias in assigning underlying cause of death can be fully addressedorresultinguncertaintyfullyquantified. The type and complexity of models used for global health estimates varies widely by research/institutionalgroupandhealthestimate.Morecomplexmodelsarenecessarytogeneratemore accurate uncertainty intervals. As expected, these are more difficult to transfer across research groups and require greater researcher expertise and time and computational resources to run. Where data are available and of high quality, estimates from different institutions are generally in agreement. Discrepancies are more likely to arise for countries where data are poor and for conditions where data aresparseandpotentiallybiased.Thisisbestaddressedthroughimprovingtheprimarydata. Country health information systems, including vital registration, need to be strengthened as a matter of priority, in order to provide a more solid empirical basis for monitoring health situation and trends is essential.SuchdataarealsocrucialforMemberStatesmonitoringoflocaltrendsinordertorespondto thechangingneedsoftheirpopulations. To improve monitoring of mortality, morbidity and risk factors the improving health information systemsshouldfocusonstrengthening: Although the GHE estimates for years 20002011 have large uncertainty ranges for some causes and some regions, they provide useful information on broad relativities of disease burden, on the relative World Health Organization Page 33 Death registration through civil registration and vital statistics systems (CRVS), local health and demographicstudiesandothersources Causeofdeathdatacollectionthroughvitalregistrationandverbalautopsyincommunities Regularhouseholdhealthsurveysthatincludebiologicalandclinicaldatacollection Completefacilityrecordingandreportingwithregularqualitycontrol
importance of different causes of death, and on regional patterns and inequalities. The data gaps and limitations in highmortality regions reinforces the need for caution when interpreting global comparative cause of death assessments and the need for increased investment in population health measurementsystems.Theuseofverbalautopsymethodsinsampleregistrationsystems,demographic surveillance systems and household surveys provides some information on causes of death in populations without wellfunctioning death registration systems, but there remain considerable challengesinthevalidationandinterpretationofsuchdata. Figure7.1summarizestheproportionaldistributionsofdeathsbyage,sexandcauseforyears2000and 2011. More detailed regional tabulations of deaths by cause, age and sex for years 2000 and 2011 are available in the WHO Global Health Observatory (www.who.int/gho) and as downloadable Excel spreadsheetsathttp://www.who.int/healthinfo/global_health_estimates/en/. Figure7.1Percentageofdeathsbycauseforglobalagesexgroups,2000and2011.
Males,2000
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 Age(years)
Females,2000
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 Age(years)
Suicide,homicideandconflict Otherunintentionalinjuries Roadinjury Othernoncommunicable Chronicrespiratorydiseases Cancers Cardiovasculardiseases Maternal,neonatal,nutritional Otherinfectiousdiseases Lowerrespiratoryinfections Diarrhoealdiseases HIV,TBandmalaria
Males,2011
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 Age(years)
Females,2011
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 Age(years)
Suicide,homicideandconflict Otherunintentionalinjuries Roadinjury Othernoncommunicable Chronicrespiratorydiseases Cancers Cardiovasculardiseases Maternal,neonatal,nutritional Otherinfectiousdiseases Lowerrespiratoryinfections Diarrhoealdiseases HIV,TBandmalaria
Page 34
These estimates for years 20002011 supercede and replace all previous estimates for global and regional causes of death published by WHO. They are not directly comparable with previous WHO estimates for 2008 and earlier years and differences should not be interpreted as trends. Figures 7.2 and 7.3 provide summary comparisons of the GHE estimates for year 2008 with the previous WHO estimatesforyear2008publishedin2011(4,5).Thesefiguresillustratethattherehasbeenlittlechange in the relative ranking for the leading causes of death, although estimated numbers of deaths are somewhat lower for most causes. This partially reflects downwards revision of allcause envelopes in recent successive revisions by UNIGME and UN Population Division, but also reflects accelerating declinesinchildmortality,andtoalesserextent,adultmortalityinrecentyears. Theseareprovisionalestimatesandwillbefurtherrevisedintheprocessofupdatingto2012forrelease at country level in late 2013. WHO and collaborators will continue to include new data and improve methods,anditisanticipatedthatsomecauseswillbefurtherupdatedinthenextrevision. Figure7.2.Changein10leadingcausesofdeathatgloballevel,GHEestimatesfor2011comparedwith previousWHOcauseofdeath(COD)estimatesforyear2008(4,5)
COD08(a)
Diseaseorinjury Is cha emi chea rtdi s ea s e Cerebrovas cul a rdi s ea s e Lowerres pi ratoryi nfecti ons COPD Di a rrhoea l di s ea s es HIV/AIDS Lung ca ncer Di a betes mel l i tus Road i njury Hypertens i ve heartdi s eas e Total deaths (millions) Rank 7.25 6.15 3.46 3.28 2.46 1.78 1.39 1.26 1.21 1.15 1.00 1 2 3 4 5 6 7 8 9 10 13
GHE2011(b)
Total deaths Rank (millions) Diseaseorinjury 1 7.02 Is cha emi cheartdi s eas e 2 3 4 5 6 7 8 9 10 11 6.25 Cerebrova s cul a rdi s eas e 3.20 Lowerres pi ra toryi nfecti ons 2.97 COPD 1.89 Di a rrhoea l di s ea s es 1.59 HIV/AIDS 1.48 Lungca ncer 1.39 Di a betes mel l i tus 1.26 Roa d i njury 1.17 Preterm bi rth compl i ca ti ons 1.06 Hypertens i ve hea rtdi s ea s e
Page 35
1200
1000
Suicide,homicideandconflict Otherunintentionalinjuries
800
Roadinjury Othernoncommunicable
600
400
Otherinfectiousdiseases HIV,TBandmalaria
200
WorldAfricaAmericasEastern Mediterranean
1400
1200
1000
Suicide,homicideandconflict Otherunintentionalinjuries
800
Roadinjury Othernoncommunicable
600
400
Otherinfectiousdiseases HIV,TBandmalaria
200
HighincomeEuropeSouthEastAsiaWesternPacific
Page 36
References
(1) (2) (3) (4) (5) World Health Organization. World health report 2002. Reducing risks, promoting healthy life. Geneva,WorldHealthOrganization,2002. Lopez,A.D.,Mathers,C.D.,Ezzati,M.,Murray,C.J.L.,&Jamison,D.T.Globalburdenofdiseaseand riskfactors.NewYork,OxfordUniversityPress,2006. World Health Organization. The global burden of disease: 2004 update. Geneva, World Health Organization,2008. WorldHealthOrganization.Causesofdeath2008:datasourcesandmethods. http://www.who.int/healthinfo/global_burden_disease/cod_2008_sources_methods.pdf. WorldHealthOrganization.Countrylevelmortalityestimatesbycause,age,andsexfortheyear 2008.Geneva:WHO.Availableat http://www.who.int/healthinfo/global_burden_disease/estimates_country/en/index.html WorldHealthOrganization.Globalhealthestimatesfordeathsbycause,age,andsexforyears 20002011.Geneva:WHO.Availableat http://www.who.int/healthinfo/global_health_estimates/en/ InternationalClassificationofDiseases10thRevision.Geneva,WorldHealthOrganization,1990. Murray CJ, Ezzati M, Flaxman AD, et al. GBD 2010: a multiinvestigator collaboration for global comparativedescriptiveepidemiology.Lancet,2012,380(9859):20558.
(6)
(7) (8)
(9) Wang H, DwyerLindgren L, Lofgren KT, et al. Agespecific and sexspecific mortality in 187 countries, 19702010: a systematic analysis for the Global Burden of Disease Study 2010. The Lancet.2012Dec13;380:20712094. (10) InstituteforHealthMetricsandEvaluation2013.GBDCompare.Availableat http://viz.healthmetricsandevaluation.org/gbdcompare/(accessed24June2013). (11) WorldHealthOrganization.WHOmethodsanddatasourcesforlifetables19902011.Global HealthEstimatesTechnicalPaperWHO/HIS/HSI/GHE/2013.1) (12) UnitedNations,DepartmentofEconomicandSocialAffairs,PopulationDivision.WorldPopulation Prospectsthe2010revision.NewYork,UnitedNations,2011. (13) UnitedNations,DepartmentofEconomicandSocialAffairs,PopulationDivision.World PopulationProspectsthe2012revision.NewYork,UnitedNations,2013. (14) UNICEF, WHO, The World Bank and UN Population Division. Levels and Trends of Child Mortality Report 2012, Estimates developed by the UN Interagency Group for Child Mortality Estimation. UNICEF,NewYork,2012. (15) ThePLoSMedicineCollectiononChildMortalityEstimationMethods.PLoSMedicine,2012. Availableat:
http://www.ploscollections.org/article/browseIssue.action?issue=info:doi/10.1371/issue.pcol.v07.i19.
(16) OestergaardMZ,etal.NeonatalMortalityLevelsfor193Countriesin2009withTrendssince1990: ASystematicAnalysisofProgress,Projections,andPriorities.PLoSMedicine,2011,8(8): e1001080.doi:10.1371/journal.pmed.1001080 (17) Murray CJL, Ferguson BD, Lopez AD, Guillot M, Salomon JA, Ahmad O. Modified logit life table system:principles,empiricalvalidationandapplication.PopulationStudies,2003,57(2):118. World Health Organization Page 37
(18) UnitedNations,DepartmentofEconomicandSocialAffairs,PopulationDivision.File02Latest datasourcesusedtoderiveestimatesfortotalpopulation,fertility,mortalityandmigrationsby countriesorareasinWorldPopulationProspectsthe2010revisions.NewYork,UnitedNations PopulationDivision,2012.Availableat:http://esa.un.org/wpp/Excel Data/WPP2010_F02_METAINFO.xls (19) UNAIDS.2012UNAIDSReportontheGlobalAIDSEpidemic.Geneva,UNAIDS,2012. (20) UnitedNations,DepartmentofEconomicandSocialAffairs,PopulationDivision.WorldPopulation Prospects:The2012Revision,Provisionalresults(unpublished). (21) AfghanPublicHealthInstitute,MinistryofPublicHealth,CentralStatisticsOrganization,ICF Macro,IndianInstituteofHealthManagementResearch,andWorldHealthOrganization. AfghanistanMortalitySurvey2010.Calverton,Maryland,USA:APHI/MoPH,CSO,ICFMacro, IIHMRandWHO/EMRO,2011. (22) CentralStatisticsOrganisation(CSO)andUNICEF.AfghanistanMultipleIndicatorClusterSurvey 20102011:FinalReport.Kabul,CentralStatisticsOrganisationandUNICEF,2012. (23) ChineseCenterforDiseaseControlandPrevention.NationalDiseaseSurveillanceSystem monitoringcausesofdeath2010.Beijing,MilitaryMedicalSciencePress,2012. (24) WorldHealthOrganization.MortalityDatabase.Availableat: http://www.who.int/healthinfo/mortality_data/en/index.html (25) MathersCD,LopezAD,MurrayCJL,EzzatiM,JamisonDT.Theburdenofdiseaseandmortalityby condition:data,methodsandresultsfor2001.Globalburdenofdiseaseandriskfactors.NewYork, OxfordUniversityPress,2006.p.45240. (26) LozanoR,etal.Globalandregionalmortalityfrom235causesofdeathfor20agegroupsin1990 and2010:asystematicanalysisfortheGlobalBurdenofDiseaseStudy2010.Lancet,2012, 380(9859):2095128. (27) WHO,UNICEF,UNFPA,WorldBank.Trendsinmaternalmortality:1990to2010.Geneva:World HealthOrganization;2012. (28) ChinaMinistryofHealthUnpublishedtabulationsVitalRegistrationSystemcauseofdeathdata submittedannuallytoWHO. (29) [NationalDiseaseSurveillanceSystemmonitoringcausesof death2010].ChineseCenterforDiseaseControlandPrevention,Beijing,MilitaryMedicalScience Press,2012,ISBN9787802458277. (30) [CauseofdeathdatafromChineseDiseaseSurveillancePoints],China MinistryofHealth,20042009. (31) ChineseCenterforDiseaseControlandPrevention.CDC.Reportedhumanrabiesdeaths1950 2010.ChineseCDC:Beijing. (32) RegistrarGeneralofIndia.CausesofDeathinIndiain20012003.NewDelhi,RegistrarGeneralof India,GovernmentofIndia,2009. (33) JhaP,GajalakshmiV,GuptaPC,KumarR,MonyP,DhingraNetal.Prospectivestudyofone milliondeathsinIndia:rationale,design,andvalidationresults.PLoSMed2006February;3(2):e18.
Page 38
(34) Liu L, Johnson HL, Cousens S, Perin J, Scott S, Lawn JE, Rudan I, Campbell H, Cibulskis R, Li M, Mathers C, Black RE, for the Child Health Epidemiology Reference Group of WHO and UNICEF. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 withtimetrendssince2000.Lancet,2012,379:215161. (35) World Health Organization. Methodology for WHO mortality estimates. Available at: http://www.who.int/healthinfo/statistics/mortality/en/index2.html (36) BlackRE,CousensS,JohnsonHetal.Global,RegionalandNationalCausesofChildMortality, 2008.Lancet,2010,375(9730):196987. (37) Johnson H, Liu L, Walker CF, Black RE. Estimating the distribution of causes of child deaths in high mortalitycountrieswithincompletedeathcertification.IntJEpidemiol,2010,39(4):11031114. (38) Liu et al. National, regional and statelevel causes of child mortality in India in 20002010: a systematicsubnationalanalysis.Underpreparation,2013. (39) Bassani DG, Kumar R, Awasthi S et al. Causes of neonatal and child mortality in India: a nationally representativemortalitysurvey.Lancet,2010,376(9755):18531860. (40). Rudan I, Chan KY, Zhang JS et al. Causes of deaths in children younger than 5 years in China in 2008.Lancet,2010,375(9720):10831089. (41) WorldHealthOrganization.GlobalTuberculosisReport2012.Geneva,WHO,2012. (42) WorldHealthOrganization.WorldMalariaReport2012.Geneva,WHO,2012. (43) WorldHealthOrganization.WorldMalariaReport2008.Geneva,WHO,2008. (44) Chen S, Fricks J, Ferrari MJ. Tracking measles infection through nonlinear state space models. JournaloftheRoyalStatisticalSocietySeriesC,2011,61(1). (45) Wolfson LJ, Grais RF, Luquero FJ, Birmingham ME, Strebel PM. Estimates of measles case fatality ratios:acomprehensivereviewofcommunitybasedstudies.IntJEpidemiol,2009,38(1):192205. (46) Joshi AB, Luman ET, Nandy R, Subedi BK, Liyanage JBL, Wierzba TF. Measles deaths in Nepal: estimating the national casefatality ratio. Bulletin of the World Health Organization, 2009, 87(6):456465. (47) Sudfeld CR, Halsey NA. Measles case fatality ratio in India a review of community based studies. IndianPediatr,2009,46(11):9839. (48) Wolfson LJ, Strebel PM, GacicDobo M, Hoekstra EJ, McFarland JW, Hersh BS. Has the 2005 measles mortality reduction goal been achieved? A natural history modelling study. Lancet, 2007, 369(9557):191200. (49) World Health Organization. Progress in global control and regional elimination of measles, 2000 2011.Weeklyepidemiologicalrecord,2013,88(3):2936. (50) Cleveland WS, Loader CL. Smoothing by local regression: principles and methods. In: Haerdle W, Schimek MG, editors. Statistical theory and computational aspects of smoothing. New York, Springer,1996,1049. (51) Hotez PJ, Bundy DA, Beegle K, Brooker S, Drake L, de Silva NR et al. Helminth infections: soil transmitted helminth infections and schistosomiasis. In: Jamison DT, Breman JG, Measham AR, Alleyne G, Evans D, Claeson M et al., eds. Disease control priorities in developing countries, 2nd edit.NewYork,OxfordUniversityPress,2006:467482. World Health Organization Page 39
(52) Van der Werf MJ, de Vlas SJ. Morbidity and infection with schistosomes or soiltransmitted helminths.Rotterdam,ErasmusUniversity,2001. (53) Vos T, Flaxman AD, Naghavi M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries, 19902010: a systematic analysis for the Global Burden of Disease Study 2010.TheLancet.2012Dec13;380:21632196. (54) WilmothJR,MizoguchiN,OestergaardMZ,SayL,MathersCD,ZureickBrownS,InoueM,ChouD. A New Method for Deriving Global Estimates of Maternal Mortality. Statistics, Politics, and Policy. 3(2),DOI:10.1515/21517509.1038,July2012 (55) Ferlay J, Shin H, Bray F, Foreman D, Mathers CD, Parkin DM. Estimates of worldwide burden of cancerin2008:Globocan2008.InternationalJournalofCancer2010;127(12):2893917. (56) UnitedNationsOfficeonDrugsandCrime.WorldDrugReport2012.UNODC:Vienna.2012 (57) WorldHealthOrganization2013.Globalstatusreportonroadsafety2013:supportingadecadeof action.Geneva,WHO,2013. (58) Khosravi A, Taylor R, Naghavi N, Lopez AD. Mortality in the Islamic Republic of Iran, 19642004. BulletinoftheWorldHealthOrganization,2007,85:60714 (59) Porapakkham Y, Rao C, Pattaraarchachai J, Polprasert W, Vos T, Adair T et al. Estimated causes of deathinThailand,2005:implicationsforhealthpolicy.PopulationHealthMetrics,2010,8:14. (60) Ngo AD, Rao C, Hoa NP, Adair T, Chuc NTK. Mortality patterns in Vietnam, 2006: findings from a nationalverbalautopsysurvey.BMCResearchNotes,2010,3:78. (61) IHME.Healthsystemaccessref (62) Kaufmann D, Kraay A, Mastruzzi M. , Massimo. Governance Matters VIII : Aggregate and IndividualGovernanceIndicators19962008.WorldBank2009. (63) Hughes BB, et al. The International Futures (IFs) modeling system, version 6.54. Frederick S. PardeeCenterforInternationalFutures,JosefKorbelSchoolofInternationalStudies,Universityof Denver,www.ifs.du.edu. (64) CRED.EMDAT:TheCREDInternationalDisasterDatabase.Belgium,UniversitCatholiquede Louvain,2012. (65) HeH,OguchiT,ZhouR,ZhangJ,QiaoS.Damageandseismicintensityofthe1996Lijiang earthquake,Vhina:aGISanalysis.Technicalreport.Tokyo,CenterforSpatialInformationScience, UniversityofTokyo,2001.Availableat:http://www.csis.utokyo.ac.jp/english/dp/dp.html (accessed18January2008). (66) NaghiiMR.Publichealthimpactandmedicalconsequencesofearthquakes.PanAmericanJournal ofPublicHealth,2005,18:216221. (67) NishikioriN,AbeT,CostaDG,DharmaratneSD,KuniiO,MojiK.Whodiedasaresultofthe tsunami?RiskfactorsofmortalityamonginternallydisplacedpersonsinSriLanka:aretrospective cohortanalysis.BMCPublicHealth,2006,6:73. (68) DoocyS,RofiA,MoodieC,SpringE,BradleyS,BurnhamGetal.TsunamimortalityinAceh Province,Indonesia.BulletinoftheWorldHealthOrganization,2007,85:273278.
Page 40
(69) HeidelbergInstituteonInternationalConflictResearch.Conflictbarometer.Departmentof PoliticalScience,UniversityofHeidelberg,2012.Availableat: http://www.hiik.de/en/konfliktbarometer/. (70) ProjectPloughshares.Armedconflictsreport.Waterloo,Canada,ProjectPloughshares,2005. Availableat:http://www.ploughshares.ca/. (71) MarshallMG,GurrTR.Peaceandconflict2005:aglobalsurveyofarmedconflicts,self determinationmovements,anddemocracy.UniversityofMaryland,CenterforInternational DevelopmentandConflictManagement,2005. (72) InternationalPeaceResearchInstitute.UCDP/PRIOArmedConflictDataset.Oslo,PRIO,2009. Availableat:http://www.prio.no/CSCW/Datasets/ArmedConflict/(accessed2November2009). (73) MurrayCJ,KingG,LopezAD,TomijimaN,KrugEG.Armedconflictasapublichealthproblem. BritishMedicalJournal,2002,324(7333):346349. (74) MurrayCJL,LopezAD.TheGlobalBurdenofDisease:acomprehensiveassessmentofmortality anddisabilityfromdiseases,injuriesandriskfactorsin1990andprojectedto2020.Cambridge, HarvardSchoolofPublicHealth,1996. (75) ObermeyerZ,MurrayCJL,GakidouE.FiftyyearsofviolentwardeathsfromVietnamtoBosnia: analysisofdatafromtheworldhealthsurveyprogramme.BritishMedicalJournal,2008, 336:14826. (76) LacinaB,GleditschNP.Monitoringtrendsinglobalcombat:anewdatasetofbattledeaths.EurJ Popul,2005,21:145166. (77) Garfield,R,BloreJ.DirectConflictDeaths.Unpublishedreportpreparedonbehalfofthe CollectiveViolenceExpertGroupfortheGlobalBurdenofDiseaseStudy,2009. (78) InternationalPeaceResearchInstitute2012.UCDP/PRIOBattleRelatedDeathsDatasetv.52012b, 19892011.Oslo,PRIO,2013.Availableat: http://www.pcr.uu.se/research/ucdp/datasets/ucdp_battlerelated_deaths_dataset/(accessed4 February2013). (79) InternationalPeaceResearchInstitute2012.UCDP/PRIONonStateConflictDatasetv.2.42012, 19892011.Oslo,UCDP,2013.Availableat: http://www.pcr.uu.se/research/ucdp/datasets/ucdp_nonstate_conflict_dataset_/(accessed4 February2013). (80) InternationalPeaceResearchInstitute2012.UCDP/PRIOOneSidedViolenceDatasetv.1.42012, 19892011.Oslo,PRIO,2013.Availableat: http://www.pcr.uu.se/research/ucdp/datasets/ucdp_onesided_violence_dataset/(accessed4 February2013). (81) IraqFamilyHealthSurveyStudyGroup.ViolenceRelatedMortalityinIraqfrom2002to2006.N EnglJMed,2008,NEJMsa0707782. (82) IraqBodyCount.Iraqideathsfromviolence20032011.Availableat: http://www.iraqbodycount.org/ (83) InternationalCampaigntoBanLandmines.Landminemonitor.Availableat:http://www.the monitor.org/
Page 41
(84) HoefflerA.DealingwiththeconsequencesofviolentconflictsinAfrica.BackgroundPaperforthe AfricanDevelopmentBank,2008.Availableat:http://users.ox.ac.uk/~ball0144/consequences.pdf (85) WorldHealthOrganization.EuropeanProgrammeforInterventionEpidemiologyTraining. Retrospectivemortalitysurveyamongtheinternallydisplacedpopulation,GreaterDarfur,Sudan, August2004.Geneva,WorldHealthOrganization,2004.Availableat: http://www.who.int/disasters/repo/14652.pdf (86) OfficefortheCoordinationofHumanitarianAffairsOccupiedPalestinianTerritory.Protectionof Civilians:CasualtyDatabase.Availableat: http://www.ochaopt.org/poc.aspx?id=1010002 (87) WorldHealthOrganization.Worldhealthreport2001.MentalHealth:NewUnderstanding,New Hope.Geneva,WorldHealthOrganization,2001. (88) KrugEG,etal.WorldReportonviolenceandhealth.Geneva:WorldHealthOrganization,2002. (89) WorldHealthOrganization.WorldHealthStatistics2007.Geneva:WorldHealthOrganization, 2007. (90) MurrayC,LopezAD,WangH.Mortalityestimationfornationalpopulations:methodsand applications.Seattle,UniversityofWashingtonPress,2012. (91) ForemanKJ,LozanoR,LopezAD,MurrayCJL.Modelingcausesofdeath:anintegratedapproach usingCODEm.PopulationHealthMetrics.2012;10:1.
Page 42
ICD-10 code
A00-B99, G00-G04, N70-N73, J00-J22, H65-H68, O00-O99, P00-P96, E00-E02, E40-E46, E50-E64, D50-D53, D64.9, U04 A00-B99, G00, G03-G04, N70-N73 A15-A19, B90 A50-A64, N70-N73 A50-A53 A55-A56 A54 A59 A57-A58, A60-A64, N70-N73 B20-B24 A00, A01, A03, A04, A06-A09 A33-A37, B05 A37 A36 B05 A33-A35 A39, G00, G03
b
5. Childhood-cluster diseases a. Whooping cough b. Diphtheria c. Measles d. Tetanus 6. Meningitis 7. Encephalitis 8. Hepatitis B 9. Hepatitis C 10. Parasitic and vector diseases a. Malaria b. Trypanosomiasis c. Chagas disease d. Schistosomiasis e. Leishmaniasis f. Lymphatic filariasis
A83-A86, B94.1, G04 B16-B19 (minus B17.1, B18.2) B17.1, B18.2 A30, A71, A82, A90-A91, B50-B57, B65, B73, B74.0-B74.2 B50-B54, P37.3, P37.4 B56 B57 B65 B55 B74.0-B74.2 B73 A30 A90-A91 A71 A82 B76-B77, B79 B77 B79 B76 A02, A05, A20-A28, A31, A32, A38, A40-A49, A65-A70, A74A79, A80-A81, A87-A89, A92-A99, B00-B04, B06-B15, B25-B49, B58-B60, B64, B66-B72, B74.3-B74.9, B75,B78, B80-B89, B91B99 (minus B94.1)
k. Rabies 11. Intestinal nematode infections a. Ascariasis b. Trichuriasis c. Hookworm disease 12. Other infectious diseases
Page 43
Code
38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 47 58 59 60
ICD-10 code
J00-J22, H65-H68,P23, U04 J09-J22, P23, U04 J00-J06 H65-H68 O00-O99 O44-O46, O67, O72 O85-O86 O10-O16 O64-O66 O00-O07 O20-O43, O47-O63, O68-O71, O73-O75, O87-O99 P00-P96 excl P37.3, P37.4
b b
1. Lower respiratory infections 2. Upper respiratory infections 3. Otitis media C. Maternal conditions 1. Maternal haemorrhage 2. Maternal sepsis 3. Hypertensive disorders of pregnancy 4. Obstructed labour 5. Abortion 6. Other maternal conditions D. Neonatal conditions 1. Preterm birth complications
P05, P07, P22, P27-P28 P03, P10-P15, P20-P21, P24-P26, P29 P35-P39 (excluding P37.3, P37.4) P00-P02, P04, P08, P50-P96 E00-E02, E40-E46, E50-E64, D50-D53, D64.9 E40-E46 E00-E02 E50 D50, D64.9 D51-D53, E51-E64 C00-C97, D00-D48, D55-D64 (minus D 64.9), D65-D89, E03E07, E10-E16, E20-E34, E65-E88, F01-F99, G06-G98, H00H61, H68-H93, I00-I99, J30-J98, K00-K92, N00-N64, N75-N98, b b L00-L98, M00-M99, Q00-Q99, X41-X42 , X45 C00-C97
d
2. Birth asphyxia and birth trauma 3. Neonatal sepsis and infections 4. Other neonatal conditions E. Nutritional deficiencies 1. Protein-energy malnutrition 2. Iodine deficiency 3. Vitamin A deficiency 4. Iron-deficiency anaemia 5. Other nutritional disorders II. Noncommunicable diseases
a
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
A. Malignant neoplasms 1. Mouth and oropharynx cancers 2. Oesophagus cancer 3. Stomach cancer
d d d
C00-C14 C15 C16 C18-C21 C22 C25 C33-C34 C43-C44 C50 C53
d
7. Trachea, bronchus and lung cancers 8. Melanoma and other skin cancers 9. Breast cancer
d d d
11. Corpus uteri cancer 12. Ovary cancer 13. Prostate cancer 14. Bladder cancer
d d d
C81-C90, C96 C91-C95 C17, C23, C24, C26-C32, C37-C41, C45-C49, C51, C52,C57C60, C62-C66, C68-C80, C97
Page 44
Code
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 I.
ICD-10 code
D00-D48 E10-E14 D55-D64 (minus D64.9), D65-D89, E03-E07, E15-E34, E65-E88 F04-F99, X41-X42 , X45 F32-F33, F34.1 F30-F31 F20-F29 F10, X45
c c c c
11. Other mental and behavioural disorders F. Neurological conditions 1. Alzheimers disease and other dementias 2. Parkinson disease 3. Epilepsy 4. Multiple sclerosis 5. Migraine 6. Non-migraine headache 7. Other neurological conditions G. Sense organ diseases 1. Glaucoma 2. Cataracts 3. Refractive errors 4. Macular degeneration 5. Other vision loss 6. Other hearing loss 7. Other sense organ disorders H. Cardiovascular diseases 1. Rheumatic heart disease 2. Hypertensive heart disease 3. Ischaemic heart disease 4. Stroke 5. Cardiomyopathy, myocarditis, endocarditis 6. Other cardiovascular diseases Respiratory diseases 1. Chronic obstructive pulmonary disease 2. Asthma 3. Other respiratory diseases J. Digestive diseases
e e
F04-F09, F17, F34-F39 (minus F34.1), F45-F48, F51-F69, F80F83, F88-F89, F93-F99 F01-F03, G06 -G98 F01-F03, G30-G31 G20-G21 G40-G41 G35 G43 G44 G06-G12, G23-G25, G36-G37, G45-G98 H00-H61, H69-H93 H40 H25-H26 H49-H52 H35.3 H30-H35 (minus H35.3), H53-H54 H90-H91 H00-H21, H27, H43-H47, H55-H61, H69-H83, H92-H93 I00-I99 I01-I09 I10-I15 I20-I25 I60-I69 I30-I33, I38, I40, I42 I00, I26-I28, I34-I37, I44-I51, I70-I99 J30-J98 J40-J44 J45-J46 J30-J39, J47-J98 K20-K92
Page 45
Code
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163
ICD-10 code
K25-K27 K70, K74 K35-K37 K20-K22, K28-K31, K38-K66, K71-K73, K75-K92 N00-N64, N75-N76, N80-N98 N00-N19 N40 N20-N23 N25-N39, N41-N45, N47-N51 N46, N97 N60-N64, N75-N76, N80-N96, N98 L00-L98 M00-M99 M05-M06 M15-M19 M10 M45-M48, M50-M54 M00, M02, M08, M11-M13, M20-M43, M60-M99 Q00-Q99 Q00, Q05 Q35-Q37 Q90 Q20-Q28 Q91-Q99 Q01-Q04, Q06-Q18, Q30-Q34, Q38-Q89 K00-K14 K00-K04, K06-K14 K05 V01-Y89 V01-X40, X43-X44, X46-59, Y40-Y86, Y88, Y89 V01-V04, V06, V09-V80, V87, V89, V99 X40, X43-X44, X46-X49 W00-W19 X00-X19 W65-W74 X30-X39 Rest of V, W20-W64, W75-W99, X20-X29, X50-X59, Y40-Y86, Y88, Y89 X60-Y09, Y35-Y36, Y870, Y871 X60-X84, Y870 X85-Y09, Y871 Y35-Y36
Page 46
, not available
Deaths coded to Symptoms, signs and ill-defined conditions (R00-R99) are distributed proportionately to all causes within Group I and Group II.
b
As from 2006, deaths from causes F10-F19 with fourth character .0 (Acute intoxication) are coded to the category of accidental poisoning according to the updated ICD-10 instructions. Cancer deaths coded to ICD categories for malignant neoplasms of other and unspecified sites including those whose point of origin cannot be determined, and secondary and unspecified neoplasms (ICD-10 C76, C80, C97) were redistributed pro-rata across the footnoted malignant neoplasm categories within each agesex group, so that the category Other malignant neoplasms includes only malignant neoplasms of other specified sites (Ref Mathers et al 2006 DCP chapter). Ischaemic heart disease deaths may be miscoded to a number of so-called cardiovascular garbage codes. These include heart failure, ventricular dysrhythmias, generalized atherosclerosis and ill-defined descriptions and complications of heart disease. Proportions of deaths coded to these causes were redistributed to ischaemic heart disease as described in (GPE discussion paper). Relevant ICD-10 codes are I47.2, I49.0, I46, I50, I51.4, I51.5, I51.6, I51.9 and I70.9. Injury deaths where the intent is not determined (Y10-Y34, Y872) are distributed proportionately to all causes below the group level for injuries. For countries with 3-digit ICD10 data, for Road injury use: V01-V04, V06, V09-V80, V87, V89 and V99. For countries with 4-digit ICD10 data, for Road injury use: V01.1-V01.9, V02.1-V02.9, V03.1-V03.9, V04.1-V04.9, V06.1-V06.9, V09.2, V09.3, V10.3-V10.9, V11.3-V11.9, V12.3-V12.9, V13.3V13.9, V14.3-V14.9, V15.4-V15.9, V16.4-V16.9, V17.4-V17.9, V18.4-V18.9, V19.4-V19.9, V20.3-V20.9, V21.3-V21.9, V22.3-V22.9, V23.3-V23.9, V24.3-V24.9, V25.3-V25.9, V26.3-V26.9, V27.3-V27.9, V28.3-V28.9, V29.4-V29.9, V30.4.V30.9, V31.4-V31.9, V32.4V32.9, V33.4-V33.9, V34.4-V34.9, V35.4-V35.9, V36.4-V36.9, V37.4-V37.9, V38.4-V38.9, V39.4-V39.9, V40.4-V40.9, V41.4-V41.9, V42.4-V42.9, V43.4-V43.9, V44.4-V44.9, V45.4-V45.9, V46.4-V46.9, V47.4-V47.9, V48.4-V48.9, V49.4-V49.9, V50.4-V50.9, V51.4V51.9, V52.4-V52.9, V53.4-V53.9, V54.4-V54.9, V55.4-V55.9, V56.4-V56.9, V57.4-V57.9, V58.4-V58.9, V59.4-V59.9, V60.4-V60.9, V61.4-V61.9, V62.4-V62.9, V63.4-V63.9, V64.4-V64.9, V65.4-V65.9, V66.4-V66.9, V67.4-V67.9, V68.4-V68.9, V69.4-V69.9, V70.4V70.9, V71.4-V71.9, V72.4-V72.9, V73.4-V73.9, V74.4-V74.9, V75.4-V75.9, V76.4-V76.9, V77.4-V77.9, V78.4-V78.9, V79.4-V79.9, V80.3-V80.5, V81.1, V82.1, V82.8-V82.9, V83.0-V83.3, V84.0-V84.3, V85.0-V85.3, V86.0-V86.3, V87.0-V87.9, V89.2-V89.3, V89.9, V99 and Y850.
g f e d
Page 47
ICD-10 code
A00-Y89 A00-B99, D50-D53, D64.9, E00-E02, E40-E64, G00, G03-G04, H65-H66, J00J22, J85, N30, N34, N390, N70-N73, O00-P96, U04
B20-B24 A00-A09 A37 A33-A35 B05 A39, A83-A87, G00, G03, G04 B50-B54, P37.3, P37.4 H65-H66, J00-J22, J85, P23, U04 P01.0, P01.1, P07, P22, P25-P28, P61.2, P77 P01.7-P02.1, P02.4-P02.6, P03, P10-P15, P20-P21, P24, P50, P90-P91 P35-P39 (exclude P37.3, P37.4)
Remainder C00-C97, D00-D48, D55-D64 (exclude D64.9), D65-D89, E03-E34, E65-E88, F01F99, G06-G98, H00-H61, H68-H93, I00-I99, J30-J84, J86-J98, K00-K92, L00-L98, M00-M99, N00-N28, N31-N32, N35-N64 (exclude N39.0), N75-N98, Q00-Q99 Q00-Q99 Remainder V01-Y89
Deaths coded to Symptoms, signs and ill-defined conditions (780-799 in ICD-9 and R00-R99 in ICD-10) are distributed proportionately to all causes within Group I and Group II.
Page 48
Cause A153 A162 A165 A169 A170 A180 A320 A321 A327 A328 A329 A35 A40 A401 A402 A403 A408 A409 A41 A410 A412 A413 A415 A418 A419 B00 B000 B004 B007 B008 B009 B01 B010 B011 B012 B018 B019 B059 B060 B068 continued
Recode P370 P370 P370 P370 P370 P370 P372 P372 P372 P372 P372 A33 P36 P360 P361 P361 P361 P361 P36 P362 P363 P368 P368 P368 P369 P35 P352 P352 P352 P352 P352 P35 P358 P358 P358 P358 P358 P358 P350 P350
Cause D649 D65 D696 D699 E101 E102 E110 E112 E116 E117 E140 E144 E145 E147 E149 E343 E86 E87 E870 E871 E872 E874 E875 E876 E877 E878 F322 F328 F329 F439 G91 G911 G912 G913 G919 G930 G931 G936 G952 I050
Recode P614 P60 D694 P549 P702 P702 P702 P702 P702 P702 P702 P702 P702 P702 P702 P051 P74 P74 P742 P742 P740 P748 P743 P743 P744 P744 P914 P914 P914 P209 Q03 Q039 Q039 Q039 Q039 Q046 P219 P524 P025 Q232
Cause I471 I472 I479 I48 I490 I494 I498 I499 I50 I500 I501 I509 I517 I518 I519 I60 I603 I607 I608 I609 I61 I610 I612 I615 I616 I618 I619 I620 I629 I632 I633 I634 I635 I638 I639 I64 I671 J12 J120 J121
Recode P291 P291 P291 P29 P291 P291 P291 P291 P29 P290 P290 P290 Q248 Q248 Q249 P52 P525 P525 P525 P525 P52 P524 P524 P524 P524 P524 P524 P528 P529 P529 P529 P529 P529 P529 P529 P52 I607 P23 P230 P230
Cause J698 J70 J709 J80 J840 J841 J848 J849 J85 J850 J851 J852 J860 J869 J90 J930 J931 J938 J939 J940 J941 J942 J948 J96 J960 J961 J969 J980 J981 J982 J984 J985 J986 J988 J989 K220 K311 K44 K440 K441
Recode P249 P24 P249 P22 P258 P258 P258 P258 P28 P288 P288 P288 P288 P288 P28 P251 P251 P251 P251 P288 P288 P548 P288 P28 P285 P285 P285 P288 P281 P250 P288 P288 P288 P288 P289 Q395 Q400 Q79 Q790 Q790
Cause K760 K761 K762 K763 K767 K768 K769 K819 K82 K828 K830 K831 K838 K839 K85 K868 K869 K904 K909 K920 K922 K928 K929 N133 N139 N17 N170 N171 N172 N179 N180 N188 N189 N19 N359 N390 N433 N883 R001 R011
Recode P788 P788 P788 P788 P788 P788 P788 P788 P78 P788 P788 P788 P788 P788 P78 P788 P788 P788 P788 P540 P543 P788 P789 Q620 Q623 P96 P960 P960 P960 P960 P960 P960 P960 P96 Q643 P393 P835 P010 P209 P298
Page 49
Annex Table C (continued): Re-assignment of ICD-10 codes for certain neonatal deaths.
Cause B069 B09 B25 B250 B251 B258 B259 B270 B370 B371 B372 B373 B374 B375 B376 B377 B378 B379 B509 B54 B582 B589 D500 D609 D62
Recode P350 P35 P35 P351 P351 P351 P351 P358 P375 P375 P375 P375 P375 P375 P375 P375 P375 P375 P373 P37 P371 P371 P549 D610 P61
Cause I059 I071 I080 I340 I348 I35 I350 I351 I352 I359 I370 I379 I38 I42 I420 I421 I422 I429 I442 I443 I455 I458 I459 I460 I469
Recode Q238 Q228 Q238 Q233 Q238 Q23 Q230 Q231 Q238 Q238 Q221 Q223 I42 I42 I424 Q248 I424 I424 Q246 Q246 Q246 Q246 Q246 P291 P291
Cause J128 J129 J13 J14 J15 J150 J151 J152 J153 J154 J155 J156 J157 J158 J159 J16 J18 J180 J181 J188 J189 J386 J439 J69 J690
Recode P230 P230 P23 P23 P23 P236 P235 P232 P233 P236 P234 P236 P236 P236 P236 P23 P23 P239 P239 P239 P239 Q318 P250 P24 P249
Cause K449 K561 K562 K565 K566 K57 K593 K625 K631 K633 K65 K650 K659 K660 K661 K720 K729 K732 K745 K746 K750 K752 K758 K759
Recode Q790 Q438 Q438 Q433 P769 Q43 Q431 P542 P780 P788 P78 P781 P781 Q433 P548 P788 P788 P788 P788 P788 P788 P788 P788 P788
Cause R030 R040 R042 R048 R049 R05 R060 R064 R068 R090 R092 R160 R162 R230 R509 R568 R571 R58 R601 R628 R629 R630 R638 R75
Recode P292 P548 P269 P548 P548 P28 P228 P228 P228 P219 P285 Q447 Q447 Q249 P819 P90 P741 P54 P833 P059 P059 P929 P929 B24
Page 50
Page 51
D.2 Countries grouped by WHO Region and average income per capita*
High income Andorra, Australia, Austria, Bahamas, Bahrain, Barbados, Belgium, Brunei Darussalam Canada, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Kuwait, Luxembourg, Malta, Monaco, Netherlands, ,New Zealand, Norway, Oman, Poland, Portugal, Qatar, Republic of Korea, Saint Kitts and Nevis, San Marino, Saudi Arabia, Singapore, Slovakia, Slovenia, Spain, Sweden, Switzerland, Trinidad and Tobago, United Arab Emirates, United Kingdom,UnitedStatesofAmerica Low and middle income WHO African Region Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Cape Verde, Central African Republic, Chad, Comoros, Congo, Cte d'Ivoire, Democratic Republic of the Congo, Equatorial Guinea**, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Guinea, GuineaBissau, Kenya, Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Seychelles, Sierra Leone, South Africa, Swaziland, Togo, Uganda, United Republic of Tanzania,Zambia,Zimbabwe WHO Region of the Americas Antigua and Barbuda, Argentina, Belize, Bolivia (Plurinational State of), Brazil, Chile, Colombia, Costa Rica, Cuba, Dominica, Dominican Republic, Ecuador, El Salvador, Grenada Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Saint Lucia, Saint Vincent and the Grenadines,Suriname,Uruguay,Venezuela(BolivarianRepublicof) WHO South-East Asia Region Bangladesh, Bhutan, Democratic People's Republic of Korea, India, Indonesia, Maldives, Myanmar, Nepal,SriLanka,Thailand,TimorLeste WHO European Region Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Montenegro, Republic of Moldova, Romania, Russian Federation, Serbia, Tajikistan,TheformerYugoslavRepublicofMacedonia,Turkey,Turkmenistan,Ukraine,Uzbekistan WHO Eastern Mediterranean Region Afghanistan, Djibouti, Egypt, Iran (Islamic Republic of), Iraq, Jordan, Lebanon, Libya, Morocco, Pakistan, Somalia,SouthSudan,Sudan,SyrianArabRepublic,Tunisia,Yemen WHO Western Pacific Region Cambodia, China, Cook Islands, Fiji, Kiribati, Lao People's Democratic Republic, Malaysia, Marshall Islands,Micronesia(FederatedStatesof),Mongolia,Nauru,Niue,Palau,PapuaNewGuinea,Philippines, Samoa,SolomonIslands,Tonga,Tuvalu,Vanuatu,VietNam
* This regional grouping classifies WHO Member States according to the World Bank income categories for the year 2011 (WorldBanklistofeconomies,July2012)andtheWHOregion. **EquatorialGuineaisclassifiedbytheWorldBankashighincome,itiskeptherewithuppermiddleincometoavoida regionalgroupingcontainingonlyonecountryandbecauseitsmortalityprofileisnotdissimilartoneighbouringcountries.
Page 52
Page 53
Page 54
Communicable,maternal,perinatalandnutritionalconditions
Page 56
2D01 2D02 Epilepsy Otherneuropsychiatricdisorders 97 8393, 95, 96, 98101 133 135139 103109 150 111 113 112 114 ApportionedtoaccordingtoGBD2010causefractions ApportionedtoaccordingtoGBD2010causefractions ApportionedtoaccordingtoGBD2010causefractions
2F01 2F02 2F03 2F04 2G01 2G02 2G03 2G04 2G05 2G06 2H01 2H02 2J01 2J02
Skindiseases Musculoskeletaldisorders Senseorgandisorders Oralconditions Rheumaticheartdisease Ischaemicheartdiseases Hypertensiveheartdiseases Cerebrovasculardisease Heartfailure Othercardiovasculardiseases Asthma and chronic pulmonarydisease obstructive
115,116 118,119 120 122 86, 123, 125,154 124,125 127 128132 141146
Apportioned to alcohol use disorders, liver cirrhosis, other gastrointestinal,andaccidentalpoisoningaccordingtoGBD2010cause fractions ApportionedtoaccordingtoGBD2010causefractions
2J03 2K01 2K02 2L01 Injuries 3A01 3A02 3A03 3A04 3A05 3A06 3A07 3B01 3B02 3C01
ApportionedtoaccordingtoGBD2010causefractions ApportionedtoaccordingtoGBD2010causefractions
Transportaccidents Poisonings Falls Fires Drownings Venomoussnakes,animalsandplants Otherunintentionalinjuries Selfinflictedinjuries(suicide) War, violence and other intentional injuries UndeterminedIntent
NonroadtransportinjuryestimatedusingGBD2010analysis
Redistributedprorataacrossintentional&unintentionalinjurycauses.
Symptoms,signsandIlldefinedconditions 4A01 4A02 4A03 Senility Other Illdefined and abnormal findings Unspecifieddeaths Redistributedprorataacrossnoninjurycausecategories. Redistributedprorataacrossnoninjurycausecategories. Redistributedprorataacrossallcausecategories.
Page 57
Annex Table F Methods used for estimation of child and adult mortality levels, and causes of death, by country, 20002011
Mortality method groups: A: B: Lifetablesbasedondeathratescomputedfromvitalregistrationdata. Projection of life table parameters l5 and l60 from adjusted vital registration data, smoothed with moving average, projected using modified logit system with latest available year's lx as standard; childmortalityfromtheUNIGME. Lifetablesbasedondeathratescomputedfromneighbouringregionalvitalregistrationdata. Life tables based on UNPDs World Population Prospects the 2010 revision, and child mortality estimatesfromtheUNIGME. Life tables based on UNPDs World Population Prospects the 2010 revision, updated with the latestHIV/AIDSmortalityfromUNAIDSandchildmortalityestimatesfromtheUNIGME. Life tables using method E together with unpublished draft updates provided by UN Population Division(seetext).
C: D: E: F:
Note (a): WHO and UN Interagency causespecific estimates for all Member States as documented in SectionXabove.
Country Afghanistan Albania Algeria Andorra Angola Antigua and Barbuda Argentina Armenia Australia Austria Azerbaijan Bahamas Bahrain Bangladesh Barbados Belarus
Under 5 child cause of death method VA multicause models VR multicause models VA multicause models VR multicause models VA multicause models VR data VR data VR multicause models VR data VR data VA multicause models VR data VR data VA multicause models VR data VR multicause models
Cause of death methods for ages 5+ GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data
2004
55%
2009
88%
Page 58
All-cause mortality method B B E D D B E A A B E E D E B A E E B Latest available VR data Average useability 2000+
Country Belgium Belize Benin Bhutan Bolivia Bosnia and Herzegovina Botswana Brazil Brunei Darussalam Bulgaria Burkina Faso Burundi Cambodia Cameroon Canada Cape Verde Central African Republic Chad Chile
Under 5 child cause of death method VR data VR data VA multicause models VA multicause models VA multicause models VR multicause models VA multicause models VR data VR multicause models VR data VA multicause models VA multicause models VA multicause models VA multicause models VR data VR multicause models VA multicause models VA multicause models VR data National VA model based on subnational Chinese studies only VR data VA multicause models VA multicause models VR multicause models VR data VA multicause models VR data VR data VR multicause models VR data VA multicause models VA multicause models VR data VA multicause models VR data VA multicause models VR multicause models VR multicause models
Cause of death methods for ages 5+ VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data
2009
88%
2010 2011
76% 79%
China Colombia Comoros Congo Cook Islands Costa Rica Cote d'Ivoire Croatia Cuba Cyprus Czech Republic Democratic People's Republic of Korea Democratic Republic of the Congo Denmark Djibouti Dominica Dominican Republic Ecuador Egypt
F A D E B A E B B B B D E B E B A A B
GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) VR data VR data VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a)
2011
87%
2010 2011
59% 61%
Page 59
All-cause mortality method A E E B E D B B E E A B E B B A E E A E D B B A D D D B B B A B D A E A B A D Latest available VR data Average useability 2000+
Country El Salvador Equatorial Guinea Eritrea Estonia Ethiopia Fiji Finland France Gabon Gambia Georgia Germany Ghana Greece Grenada Guatemala Guinea Guinea-Bissau Guyana Haiti Honduras Hungary Iceland India Indonesia Iran (Islamic Republic of) Iraq Ireland Israel Italy Jamaica Japan Jordan Kazakhstan Kenya Kiribati Kuwait Kyrgyzstan Lao People's Democratic Republic
Under 5 child cause of death method VR multicause models VA multicause models VA multicause models VR data VA multicause models VR multicause models VR data VR data VA multicause models VA multicause models VR multicause models VR data VA multicause models VR data VR data VA multicause models VA multicause models VA multicause models VR data VA multicause models VR multicause models VR data VR data State-level Indian-specific VA model VA multicause models VA multicause models VA multicause models VR data VR data VR data VR multicause models VR data VR multicause models VA multicause models VA multicause models VA multicause models VR data VA multicause models VA multicause models
Cause of death methods for ages 5+ GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data VR data GBD 2010 plus (a) VR data GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data GBD 2010 plus (a)
2009
58%
Page 60
All-cause mortality method B D E E D B B D E A A E B A D B B D C B B D E D D D B B A D E A B D D B A D A A Latest available VR data Average useability 2000+
Country Latvia Lebanon Lesotho Liberia Libyan Arab Jamahiriya Lithuania Luxembourg Madagascar Malawi Malaysia Maldives Mali Malta Marshall Islands Mauritania Mauritius Mexico Micronesia (Federated States of) Monaco Mongolia Montenegro Morocco Mozambique Myanmar Namibia Nauru Nepal Netherlands New Zealand Nicaragua Niger Nigeria Niue Norway Oman Pakistan Palau Panama Papua New Guinea Paraguay
Under 5 child cause of death method VR data VR multicause models VA multicause models VA multicause models VR multicause models VR data VR data VA multicause models VA multicause models VR multicause models VR multicause models VA multicause models VR data VA multicause models VA multicause models VR data VR data VA multicause models VR multicause models VA multicause models VR data VA multicause models VA multicause models VA multicause models VA multicause models VA multicause models VA multicause models VR data VR data VR multicause models VA multicause models VA multicause models VR multicause models VR data VR multicause models VA multicause models VR multicause models VR data VA multicause models VR multicause models
Cause of death methods for ages 5+ VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a)
2010
92%
2010
94%
Page 61
All-cause mortality method A B B B B B B B E B B B D B D D D B B E B B B D D A #N/A B A D B E B B D A A B D Latest available VR data Average useability 2000+
Country Peru Philippines Poland Portugal Qatar Republic of Korea Republic of Moldova Romania Russian Federation Rwanda Saint Kitts and Nevis Saint Lucia Saint Vincent and the Grenadines Samoa San Marino Sao Tome and Principe Saudi Arabia Senegal Serbia Seychelles Sierra Leone Singapore Slovakia Slovenia Solomon Islands Somalia South Africa Spain Sri Lanka Sudan Suriname Swaziland Sweden Switzerland Syrian Arab Republic Tajikistan Thailand The former Yugoslav Republic of Macedonia Timor-Leste
Under 5 child cause of death method VR multicause models VA multicause models VR data VR data VR multicause models VR data VR data VR data VR multicause models VA multicause models VR data VR data VR data VR multicause models VR data VA multicause models VR multicause models VA multicause models VR data VR multicause models VA multicause models VR data VR data VR data VA multicause models VA multicause models VA multicause models VR data VR multicause models VA multicause models VR data VA multicause models VR data VR data VR multicause models VA multicause models VR multicause models VR data VA multicause models
Cause of death methods for ages 5+ GBD 2010 plus (a) VR data GBD 2010 plus (a) VR data GBD 2010 plus (a) VR data VR data VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a)
2006 2010
48% 84%
Page 62
All-cause mortality method E A B A D A A E B D B E B B A D A D D E E Latest available VR data Average useability 2000+
Country Togo Tonga Trinidad and Tobago Tunisia Turkey Turkmenistan Tuvalu Uganda Ukraine United Arab Emirates United Kingdom United Republic of Tanzania United States Uruguay Uzbekistan Vanuatu Venezuela Viet Nam Yemen Zambia Zimbabwe
Under 5 child cause of death method VA multicause models VR multicause models VR data VR multicause models VR multicause models VA multicause models VR multicause models VA multicause models VR multicause models VR multicause models VR data VA multicause models VR data VR data VA multicause models VR multicause models VR data VR multicause models VA multicause models VA multicause models VA multicause models
Cause of death methods for ages 5+ GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) VR data GBD 2010 plus (a) VR data GBD 2010 plus (a) VR data VR data VR data GBD 2010 plus (a) VR data GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a) GBD 2010 plus (a)
Page 63
Annex Table G Methods used to estimate road traffic deaths for 182 participating countries
Country
Afghanistan Albania Andorra Angola Argentina Armenia Australia Austria Azerbaijan Bahamas Bahrain Bangladesh Barbados Belarus Belgium Belize Benin Bhutan Bolivia (Plurinational State of) Bosnia and Herzegovina Botswana Brazil Brunei Darussalam Bulgaria Burkina Faso Burundi Cambodia Cameroon Canada Cape Verde Central African Republic Chad Chile China Colombia
Group
4 4 3 4 1 4 1 1 1 1 1 4 1 1 1 1 4 4 4 4 4 1 1 1 4 4 4 4 1 4 4 4 1 1 1
Method
Regression estimate Regression estimate Reported deaths (small population) Regression estimate Projected death registration data Regression estimate Projected death registration data Death registration data Reported deaths (replacing death registration estimate) Projected death registration data Projected death registration data Regression estimate Reported deaths (replacing death registration estimate) Projected death registration data Projected death registration data Projected death registration data Regression estimate Regression estimate Regression estimate Regression estimate Regression estimate Projected death registration data Projected death registration data Reported deaths (replacing death registration estimate) Regression estimate Regression estimate Regression estimate Regression estimate Death registration data Regression estimate Regression estimate Reported deaths (replacing regression estimate) Death registration data Death registration data (refer to section 3.5) Projected death registration data
Latest VR data
2009
2010
Page 64
Country
Comoros Congo Cook Islands Costa Rica Cte d'Ivoire Croatia Cuba Cyprus Czech Republic Democratic Korea People's Republic of
Group
4 4 3 1 4 1 1 1 1 4 4 1 3 4 1 1 1 4 1 4 1 1 1 4 4 1 1 4 1 1 4 4 1 4 1 1 2 4
Method
Regression estimate Regression estimate Reported deaths (small population) Death registration data Regression estimate Death registration data Projected death registration data Death registration data Death registration data Regression estimate Regression estimate Projected death registration data Reported deaths (small population) Regression estimate Projected death registration data Death registration data Projected death registration data Regression estimate Death registration data Regression estimate Death registration data Reported deaths (replacing death registration estimate) Reported deaths (replacing death registration estimate) Reported deaths (replacing regression estimate) Regression estimate Projected death registration data Death registration data Regression estimate Projected death registration data Projected death registration data Regression estimate Regression estimate Projected death registration data Regression estimate Death registration data Projected death registration data Regression estimate projected from 2001-2003 data (32, 33) Regression estimate
Latest VR data
2010
Democratic Republic of the Congo Denmark Dominica Dominican Republic Ecuador Egypt El Salvador Equatorial Guinea Estonia Ethiopia Fiji Finland France Gabon Gambia Georgia Germany Ghana Greece Guatemala Guinea Guinea-Bissau Guyana Honduras Hungary Iceland India Indonesia
2006 2010
2010
2009 2010
2009 2009
2008
Page 65
Country
Iran (Islamic Republic of) Iraq Ireland Israel Italy Jamaica Japan Jordan Kazakhstan Kenya Kiribati Kuwait Kyrgyzstan Lao People's Democratic Republic Latvia Lebanon Lesotho Liberia Lithuania Luxembourg Madagascar Malawi Malaysia Maldives Mali Malta Marshall Islands Mauritania Mauritius Mexico Micronesia (Federated States of) Mongolia Montenegro Morocco Mozambique Myanmar Namibia Nepal Netherlands
Group
2 4 1 1 1 1 1 4 1 4 3 1 1 4 1 4 4 4 1 1 4 4 4 1 4 1 3 4 1 1 3 4 1 4 4 4 4 4 1
Method
Projected death registration data Regression estimate Death registration data Projected death registration data Projected death registration data Projected death registration data Projected reported deaths (replacing death registration estimate) Regression estimate Death registration data Regression estimate Reported deaths (small population) Projected death registration data Projected death registration data Regression estimate Death registration data Regression estimate Regression estimate Regression estimate Death registration data Projected death registration data Regression estimate Regression estimate Reported deaths (replacing regression estimate) Reported deaths(replacing death registration estimate) Regression estimate Death registration data Reported deaths (small population) Regression estimate Death registration data Projected reported deaths (replacing death registration estimate) Reported deaths (small population) Reported deaths (replacing regression estimate) Projected death registration data Regression estimate Regression estimate Regression estimate Reported deaths (replacing regression estimate) Regression estimate Death registration data
Latest VR data
2006
2010
2009 2009
2010
2010 2009
2008
2010
2010 2010
2009
2010
Page 66
Country
New Zealand Nicaragua Niger Nigeria Niue Norway Oman Pakistan Palau Panama Papua New Guinea Paraguay Peru Philippines Poland Portugal Qatar Republic of Korea Republic of Moldova Romania Russian Federation Rwanda Saint Kitts and Nevis Saint Lucia Saint Vincent and the Grenadines Samoa San Marino Sao Tome and Principe Saudi Arabia Senegal Serbia Seychelles Sierra Leone Singapore Slovakia Slovenia Solomon Islands South Africa Spain
Group
1 4 4 4 3 1 1 4 3 1 4 1 4 1 1 1 1 1 1 1 1 4 3 1 3 4 3 4 4 4 1 3 4 1 1 1 4 1 1
Method
Projected death registration data Regression estimate Regression estimate Regression estimate Reported deaths (small population) Reported deaths (replacing death registration estimate) Death registration data Regression estimate Reported deaths (small population) Projected death registration data Regression estimate Projected death registration data Regression estimate Projected death registration data Death registration data Death registration data Death registration data Projected death registration data Death registration data Reported deaths (replacing death registration estimate) Reported deaths (replacing death registration estimate) Regression estimate Reported deaths (small population) Projected death registration data Reported deaths (small population) Regression estimate Reported deaths (small population) Reported deaths (replacing regression estimate) Reported deaths (replacing regression estimate) Regression estimate Death registration data Reported deaths (small population) Regression estimate Death registration data Death registration data Death registration data Regression estimate Projected death registration data Reported deaths (replacing death registration est.)
Latest VR data
2008
2010 2010
2009
2009
2010 2009
2009 2009
Page 67
Country
Sri Lanka Sudan Suriname Swaziland Sweden Switzerland Syrian Arab Republic Tajikistan Thailand The Former Yugoslav Republic of Macedonia Timor-Leste Togo Tonga Trinidad and Tobago Tunisia Turkey Uganda Ukraine United Arab Emirates United Kingdom United Republic of Tanzania United States of America Uruguay Uzbekistan Vanuatu Venezuela (Bolivarian Republic of) Viet Nam West Bank and Gaza Strip Yemen Zambia Zimbabwe
Group
4 4 1 4 1 1 4 4 2 1 4 4 3 1 4 4 4 1 4 1 4 1 1 1 4 1 2 1 4 4 1
Method
Regression estimate Regression estimate Projected death registration data Regression estimate Death registration data Projected death registration data Regression estimate Regression estimate Projected death registration data Death registration data Regression estimate Regression estimate Reported deaths (small population) Projected death registration data Regression estimate Regression estimate Regression estimate Death registration data Regression estimate Death registration data Regression estimate Projected death registration data Projected death registration data Projected death registration data Regression estimate Projected death registration data Projected national verbal autopsy survey data Reported deaths (replacing regression estimate) Regression estimate Regression estimate Reported deaths (replacing regression estimate)
Latest VR data
2009
2010 2007
2008 2010
2007
2010
2010
Page 68