burden of injury along the development spectrum

Haagsma JA, et al. Inj Prev 2020;26:i12–i26. doi:10.1136/injuryprev-2019-043296i12

Original research

Burden of injury along the development spectrum: associations between the Socio- demographic Index and disability- adjusted life year estimates from the Global Burden of Disease Study 2017Juanita A Haagsma,1 Spencer L James ,2 Chris D Castle,2 Zachary V Dingels,2 Jack T Fox,2 Erin B Hamilton,2 Zichen Liu,2 Lydia R Lucchesi,2 Nicholas L S Roberts,2 Dillon O Sylte,2 Oladimeji M Adebayo,3 Alireza Ahmadi,4 Muktar Beshir Ahmed,5 Miloud Taki Eddine Aichour,6 Fares Alahdab,7 Suliman A Alghnam,8 Syed Mohamed Aljunid,9,10 Rajaa M Al- Raddadi,11 Ubai Alsharif,12 Khalid Altirkawi,13 Mina Anjomshoa,14 Carl Abelardo T Antonio,15,16 Seth Christopher Yaw Appiah,17,18 Olatunde Aremu,19 Amit Arora,20,21 Hamid Asayesh,22 Reza Assadi,23 Ashish Awasthi,24 Beatriz Paulina Ayala Quintanilla,25,26 Shivanthi Balalla,27 Amrit Banstola,28 Suzanne Lyn Barker- Collo,29 Till Winfried Bärnighausen,30,31 Shahrzad Bazargan- Hejazi,32,33 Neeraj Bedi,34 Masoud Behzadifar,35 Meysam Behzadifar,36 Corina Benjet,37 Derrick A Bennett,38 Isabela M Bensenor,39 Soumyadeep Bhaumik,40 Zulfiqar A Bhutta,41,42 Ali Bijani,43 Guilherme Borges,37 Rohan Borschmann,44,45 Dipan Bose,46 Soufiane Boufous,47 Alexandra Brazinova,48 Julio Cesar Campuzano Rincon,49,50 Rosario Cárdenas,51 Juan J Carrero,52 Félix Carvalho,53 Carlos A Castañeda- Orjuela,54,55 Ferrán Catalá-López,56,57 Jee- Young J Choi,58 Devasahayam J Christopher,59 Christopher Stephen Crowe,60 Koustuv Dalal,61,62 Ahmad Daryani,63 Dragos Virgil Davitoiu,64,65 Louisa Degenhardt,2,66 Diego De Leo,67 Jan- Walter De Neve,30 Kebede Deribe,68,69 Getenet Ayalew Dessie,70 Gabrielle Aline deVeber,71 Samath Dhamminda Dharmaratne,2,72 Linh Phuong Doan,73 Kate A Dolan,74 Tim Robert Driscoll,75 Manisha Dubey,76 Ziad El- Khatib,77,78 Christian Lycke Ellingsen,79 Maysaa El Sayed Zaki,80 Aman Yesuf Endries,81 Sharareh Eskandarieh,82 Andre Faro,83 Seyed- Mohammad Fereshtehnejad,84,85 Eduarda Fernandes,86 Irina Filip,87,88 Florian Fischer,89 Richard Charles Franklin,90 Takeshi Fukumoto,91,92 Kebede Embaye Gezae,93 Tiffany K Gill,94 Alessandra C Goulart,95,96 Ayman Grada,97 Yuming Guo,98,99 Rahul Gupta,100,101 Hassan Haghparast Bidgoli,102 Arvin Haj- Mirzaian,103,104 Arya Haj- Mirzaian,103,105 Randah R Hamadeh,106 Samer Hamidi,107 Josep Maria Haro,108,109 Hadi Hassankhani,110,111 Hamid Yimam Hassen,112,113 Rasmus Havmoeller,114 Delia Hendrie,115 Andualem Henok,112 Martha Híjar,116,117 Michael K Hole,118 Enayatollah Homaie Rad,119,120 Naznin Hossain,121,122 Sorin Hostiuc,123,124 Guoqing Hu,125 Ehimario U Igumbor,126,127 Olayinka Stephen Ilesanmi,128 Seyed Sina Naghibi Irvani,129 Sheikh Mohammed Shariful Islam,130,131 Rebecca Q Ivers,132 Kathryn H Jacobsen,133 Nader Jahanmehr,134,135 Mihajlo Jakovljevic,136 Achala Upendra Jayatilleke,137,138 Ravi Prakash Jha,139 Jost B Jonas,140,141 Zahra Jorjoran Shushtari,142 Jacek Jerzy Jozwiak,143 Mikk Jürisson,144 Ali Kabir,145 Rizwan Kalani,146 Amir Kasaeian,147,148 Abraham Getachew Kelbore,149 Andre Pascal Kengne,150,151 Yousef Saleh Khader,152 Morteza Abdullatif Khafaie,153 Nauman Khalid,154 Ejaz Ahmad Khan,155 Abdullah T Khoja,156,157 Aliasghar A Kiadaliri,158 Young- Eun Kim,159 Daniel Kim,160 Adnan Kisa,161 Ai Koyanagi,162,163 Barthelemy Kuate Defo,164,165

To cite: Haagsma JA, James SL, Castle CD, et al. Inj Prev 2020;26:i12–i26.

► Additional material is published online only. To view, please visit the journal online (http:// dx. doi. org/ 10. 1136/ injuryprev- 2019- 043296).

For numbered affiliations see end of article.

Correspondence toDr Spencer L James, Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA 98121, USA; spencj@ uw. edu

Received 3 May 2019Revised 8 August 2019Accepted 12 August 2019Published Online First 8 January 2020

© Author(s) (or their employer(s)) 2020. Re- use permitted under CC BY. Published by BMJ.

on October 18, 2021 by guest. P

rotected by copyright.http://injuryprevention.bm

j.com/

Inj Prev: first published as 10.1136/injuryprev-2019-043296 on 8 January 2020. D

ownloaded from

on O

ctober 18, 2021 by guest. Protected by copyright.

http://injuryprevention.bmj.com

/Inj P

rev: first published as 10.1136/injuryprev-2019-043296 on 8 January 2020. Dow

nloaded from



j.com/


ownloaded from

on O



/Inj P


nloaded from



j.com/


ownloaded from

on O



/Inj P


nloaded from



j.com/


ownloaded from

on O



/Inj P


nloaded from



j.com/


ownloaded from

on O



/Inj P


nloaded from



j.com/


ownloaded from

on O



/Inj P


nloaded from



j.com/


ownloaded from

on O



/Inj P


nloaded from



j.com/


ownloaded from

on O



/Inj P


nloaded from



j.com/


ownloaded from

on O



/Inj P


nloaded from



j.com/


ownloaded from

http://injuryprevention.bmj.com/

http://orcid.org/0000-0003-4653-2507

http://dx.doi.org/10.1136/injuryprev-2019-043296


http://crossmark.crossref.org/dialog/?doi=10.1136/injuryprev-2019-043296&domain=pdf&date_stamp=2020-08-31




















Haagsma JA, et al. Inj Prev 2020;26:i12–i26. doi:10.1136/injuryprev-2019-043296 i13

Original research

Burcu Kucuk Bicer,166,167 Manasi Kumar,168,169 Ratilal Lalloo,170 Hilton Lam,171 Faris Hasan Lami,172 Van C Lansingh,173,174 Janet L Leasher,175 Shanshan Li,98 Shai Linn,176 Raimundas Lunevicius,177,178 Flavia R Machado,179 Hassan Magdy Abd El Razek,180 Muhammed Magdy Abd El Razek,181 Narayan Bahadur Mahotra,182 Marek Majdan,183 Azeem Majeed,184 Reza Malekzadeh,185,186 Manzoor Ahmad Malik,187,188 Deborah Carvalho Malta,189 Ana- Laura Manda,190 Mohammad Ali Mansournia,191 Benjamin Ballard Massenburg,60 Pallab K Maulik,192,193 Hailemariam Abiy Alemu Meheretu,70,194 Man Mohan Mehndiratta,195,196 Addisu Melese,197 Walter Mendoza,198 Melkamu Merid Mengesha,199 Tuomo J Meretoja,200,201 Atte Meretoja,202,203 Tomislav Mestrovic,204,205 Tomasz Miazgowski,206 Ted R Miller,115,207 GK Mini,208,209 Erkin M Mirrakhimov,210,211 Babak Moazen,30,212 Naser Mohammad Gholi Mezerji,213 Roghayeh Mohammadibakhsh,214 Shafiu Mohammed,30,215 Mariam Molokhia,216 Lorenzo Monasta,217 Stefania Mondello,218,219 Pablo A Montero- Zamora,220,221 Yoshan Moodley,222 Mahmood Moosazadeh,223 Ghobad Moradi,224,225 Maziar Moradi- Lakeh,226 Lidia Morawska,227 Ilais Moreno Velásquez,228 Shane Douglas Morrison,229 Marilita M Moschos,230,231 Seyyed Meysam Mousavi,232,233 Srinivas Murthy,234 Kamarul Imran Musa,235 Gurudatta Naik,236 Farid Najafi,237 Vinay Nangia,238 Bruno Ramos Nascimento,239 Duduzile Edith Ndwandwe,240 Ionut Negoi,64,241 Trang Huyen Nguyen,242 Son Hoang Nguyen,242 Long Hoang Nguyen,242 Huong Lan Thi Nguyen,243 Dina Nur Anggraini Ningrum,244,245 Yirga Legesse Nirayo,246 Richard Ofori- Asenso,247,248 Felix Akpojene Ogbo,249 In- Hwan Oh,250 Olanrewaju Oladimeji,251,252 Andrew T Olagunju,253,254 Tinuke O Olagunju,255 Pedro R Olivares,256 Heather M Orpana,257,258 Stanislav S Otstavnov,259,260 Mahesh P A,261 Smita Pakhale,262 Eun- Kee Park,263 George C Patton,264,265 Konrad Pesudovs,266 Michael R Phillips,267,268 Suzanne Polinder,1 Swayam Prakash,269 Amir Radfar,270,271 Anwar Rafay,272 Alireza Rafiei,273,274 Siavash Rahimi,275 Vafa Rahimi- Movaghar,276 Muhammad Aziz Rahman,277,278 Rajesh Kumar Rai,279,280 Kiana Ramezanzadeh,281 Salman Rawaf,184,282 David Laith Rawaf,283,284 Andre M N Renzaho,249,285 Serge Resnikoff,286,287 Shahab Rezaeian,288 Leonardo Roever,289 Luca Ronfani,217 Gholamreza Roshandel,185,290 Yogesh Damodar Sabde,291 Basema Saddik,292 Payman Salamati,276 Yahya Salimi,237,293 Inbal Salz,294 Abdallah M Samy,295 Juan Sanabria,296,297 Lidia Sanchez Riera,298,299 Milena M Santric Milicevic,300,301 Maheswar Satpathy,302,303 Monika Sawhney,304 Susan M Sawyer,44,264 Sonia Saxena,305 Mete Saylan,306 Ione J C Schneider,307 David C Schwebel,308 Soraya Seedat,309 Sadaf G Sepanlou,185,186 Masood Ali Shaikh,310 Mehran Shams- Beyranvand,311,312 Morteza Shamsizadeh,313 Mahdi Sharif- Alhoseini,276 Aziz Sheikh,314,315 Jiabin Shen,316 Mika Shigematsu,317 Rahman Shiri,318 Ivy Shiue,319 João Pedro Silva,53 Jasvinder A Singh,320,321 Dhirendra Narain Sinha,322,323 Adauto Martins Soares Filho,324 Joan B Soriano,325,326 Sergey Soshnikov,327 Ireneous N Soyiri,328,329 Vladimir I Starodubov,330 Dan J Stein,323,331 Mark A Stokes,332 Mu’awiyyah Babale Sufiyan,333 Jacob E Sunshine,334 Bryan L Sykes,335 Rafael Tabarés- Seisdedos,336,337 Karen M Tabb,338 Arash Tehrani- Banihashemi,226,339 Gizachew Assefa Tessema,340,341 Jarnail Singh Thakur,342 Khanh Bao Tran,343,344 Bach Xuan Tran,345 Lorainne Tudor Car,346 Olalekan A Uthman,347 Benjamin S Chudi Uzochukwu,348 Pascual R Valdez,349,350 Elena Varavikova,351 Ana Maria Nogales Vasconcelos,352,353 Narayanaswamy Venketasubramanian,354,355 Francesco S Violante,356,357 Vasily Vlassov,358 Yasir Waheed,359 Yuan- Pang Wang,360 Tissa Wijeratne,361,362 Andrea Sylvia Winkler,363,364 Priyanka Yadav,365 Yuichiro Yano,366 Muluken Azage Yenesew,194 Paul Yip,367,368 Engida Yisma,369 Naohiro Yonemoto,370 Mustafa Z Younis,371,372 Chuanhua Yu,373,374 Shamsa Zafar,375 Zoubida Zaidi,376 Sojib Bin Zaman,377,378 Mohammad Zamani,379 Yong Zhao,380 Sanjay Zodpey,381 Simon I Hay,2,382 Alan D Lopez,2,383 Ali H Mokdad,2,382 Theo Vos2,382



j.com/


ownloaded from



Original research

AbsTrACTbackground The epidemiological transition of noncommunicable diseases replacing infectious diseases as the main contributors to disease burden has been well documented in global health literature. Less focus, however, has been given to the relationship between sociodemographic changes and injury. The aim of this study was to examine the association between disability- adjusted life years (DALYs) from injury for 195 countries and territories at different levels along the development spectrum between 1990 and 2017 based on the Global Burden of Disease (GBD) 2017 estimates.Methods Injury mortality was estimated using the GBD mortality database, corrections for garbage coding and CODEm—the cause of death ensemble modelling tool. Morbidity estimation was based on surveys and inpatient and outpatient data sets for 30 cause- of- injury with 47 nature- of- injury categories each. The Socio- demographic Index (SDI) is a composite indicator that includes lagged income per capita, average educational attainment over age 15 years and total fertility rate.results For many causes of injury, age- standardised DALY rates declined with increasing SDI, although road injury, interpersonal violence and self- harm did not follow this pattern. Particularly for self- harm opposing patterns were observed in regions with similar SDI levels. For road injuries, this effect was less pronounced.Conclusions The overall global pattern is that of declining injury burden with increasing SDI. However, not all injuries follow this pattern, which suggests multiple underlying mechanisms influencing injury DALYs. There is a need for a detailed understanding of these patterns to help to inform national and global efforts to address injury- related health outcomes across the development spectrum.

InTrOduCTIOnInjury is an important cause of morbidity and mortality in nations at any point of the development spectrum. Previous research has shown that in 2015, injuries accounted for 11% of the global burden of disease, expressed in disability- adjusted life years (DALYs), with an estimated 973 million people sustaining injuries warranting some type of healthcare and 4.7 million deaths.1 Globally, since 1990, focused injury burden research has documented a declining trend in the burden of injury of all the major causes of injury.2

The epidemiological transition of noncommunicable diseases (NCDs) replacing infectious diseases as the main contributors to disease burden has been well- documented.1 3 4 However, less focus has been given to the relationship between sociodemo-graphic changes and injury outcomes. Up till now, few studies have been performed that studied the relationship between sociodemographic changes and overall injury rates. There have been reports on the associations of gross domestic product and unemployment with suicides, homicides, road injury and unin-tentional injuries.5–12 However, these studies focused on one specific cause of injury and on one type of injury outcome, mostly mortality. The findings of these studies indicated that the relationship between economic development and injury burden is not straightforward and mediated by many factors. A better understanding of this relationship may be achieved by investi-gating all causes of injury as well as looking at both fatal and non- fatal injury outcome.

Insight into the epidemiological transitions with regard to injuries can be achieved by a systematic analysis of the relation-ship between development and trends in mortality, incidence and burden of disease using a standardised approach. A system-atic analysis may also reveal where health gains outpace or fall

behind changes in development and allow for the identification of determinants and mediating factors of injury burden. This information allows identification determinants of injury burden. This information serves as a crucial input for guiding health system investments and priority- setting at the global, regional, national and subnational levels.

The Global Burden of Disease (GBD) 2015 study introduced a measure of development, the Socio- demographic Index (SDI). SDI combines information on income per capita, education and fertility. Comparisons between DALYs and SDI showed that age- standardised DALY rates for many communicable diseases declined profoundly over time, whereas improvements in SDI correlated strongly with the increasing importance of NCDs.4

This paper aims to provide an overview of injury mortality, incidence and DALYs from the GBD 2017 study, with detailed information on a range of causes of injuries; to examine the asso-ciation between years of life lost (YLLs), YLDs and DALYs from injury and development, as measured by SDI, cause of injury, GBD region and over time; and to assess in which regions injury DALYs outpace or fall behind changes in development.

MeThOdsGbd 2017 studyThe GBD 2017 study methods and results have been described in extensive detail elsewhere, including description of the analyt-ical estimation framework used to measure deaths, YLLs, YLDs and DALYs.4 13 14 A summary overview of the GBD study is provided in online supplementary appendix 1. The methodolog-ical components specific to injuries estimation and SDI calcula-tion are summarised below.

Injury incidence and death are defined as ICD-9 codes E800–E999 and ICD-10 chapters V–Y, except for deaths and cases of drug overdoses and unintentional alcohol poisoning, which are classified under drug and alcohol use disorders. These external cause- of- injury codes or ‘E codes’ are designated as mutually exclusive and collectively exhaustive within the injuries estima-tion process. In terms of the nature- of- injury codes (eg, the lower extremity amputation that can occur with a road injury), inju-ries were categorised into 47 mutually exclusive and collectively exhaustive nature- of- injury categories using chapters S and T in International Classification of Disease (ICD) ICD-10 and codes 800–999 in ICD-9 to quantify the various disabling outcomes of each cause of injury. Some injuries are trivial and unlikely to account for an important number of DALYs; hence, we only included injuries in our morbidity analysis that warranted some form of healthcare.

Injury mortality and YLLsThe overall approach to estimate causes of death is provided in related publications.13 15 16 A summary is as follows. We first mapped data sources using different versions of ICD or alter-native classification systems to the GBD cause list. These data sources included vital registration, verbal autopsy, mortality surveillance, censuses, surveys, hospitals, police records and mortuary data. We then made adjustments for ill- defined causes of death such that they mapped to an underlying cause of death. Next, we conducted ensemble models using GBD cause of death ensemble modelling (CODEm) software to estimate cause- specific mortality by age, sex, country, year and cause. CODEm is described in more detail elsewhere but in summary explores a large variety of possible models to estimate trends in causes of death using an algorithm to select varying combinations of covariates that are run through several modelling classes. The



j.com/


ownloaded from

https://dx.doi.org/10.1136/injuryprev-2019-043296



Original research

method then creates an ensemble of best- performing models that are determined by evaluating out- of- sample predictive validity. Deaths are then rescaled for each cause so that the sum equals the number of deaths from all causes to ensure internal consis-tency. YLLs were calculated by multiplying deaths by the residual life expectancy at the age of death based on the GBD 2017 stan-dard model life table.12

Injury incidence, prevalence and years lived with disabilityOur method for estimating the incidence, prevalence and years lived with disability in non- fatal injury outcomes is provided in other GBD publications.2 14 A summary is as follows. We used DisMod- MR V.2.1 (a meta- regression tool for epidemio-logical modelling) to model injury incidence using data from emergency department and hospital records and survey data to produce cause- of- injury incidence by location, year, age and sex. Across every injury cause model, we used national income per capita as a covariate on excess mortality, which forces a negative relationship between income and mortality to take into account higher case fatality in lower- resource settings. After modelling incidence of each cause of injury, we used a severity hierarchy to identify the nature- of- injury category that would lead to the most long- term burden when an indi-vidual experiences multiple injuries. This hierarchy is based on pooled data sets of follow- up studies in which we translated each individual’s health status measure at 1 year after injury into a disability weight. This process is described in more detail in the GBD literature.12 14 17–22 Then, we generated matrices of the proportions of each cause of injury that are expected to lead to each nature of injury as determined in dual- coded (eg, both cause- of- injury and nature- of- injury coded) hospital and emergency department data sets and data from the Chinese National Injury Surveillance System.23 These data sets were used because the data were available in microdata format and they included dual- coded data in the format required for this specific part of the analysis. The resulting cause–nature matrices varied by injury warranting hospital admission versus injury warranting other healthcare, high- income/low- income countries, male/female and age group. In the next stage, we estimated short- term disability by cause and nature‐of‐injury category based on average duration for treated cases for each nature- of- injury category and for inpatient and outpatient injuries from the Dutch Injury Surveillance System.17 18 For 19 of the 47 nature- of- injury categories (eg, foreign body in ear, poisoning and fracture in ear), we supplemented these estimates with expert- driven estimates of short- term duration for nature- of- injury categories when the data set had insuffi-cient information. For untreated injuries, the average factor by which the duration of short- term injury outcomes is increased for a given nature- of- injury category when the injury goes untreated was estimated.

For longer- term injuries, we calculated the proportion of injuries that would result in disability lasting more than a year for each nature- of- injury category by admission status and age. This calculation was based on an assumption that disability from injury affects all cases in the short term with a proportion having persistent disability 1 year after the injury greater than the pre- injury health status. These probabilities of developing perma-nent health loss were based on a pooled data set of seven large follow- up studies from China, the Netherlands and the USA that used patient- reported outcome measures to assess health status.17–22 24 25 We used the GBD healthcare access and quality (HAQ) index to estimate the ratio of treated to untreated injuries

for each country–year grouping.26 The HAQ index is scaled from 0 to 100 and is based on 32 causes of death, covering a range of health service areas, which should not occur if effective care is present. Finally, we used DisMod- MR V.2.1 to compute the long- term prevalence (ie, 1 year or more) for each cause–nature combination from incidence, which also incorporated increased mortality risk of certain nature of injuries, such as traumatic brain injury based on meta- analyses of studies providing stan-dardised mortality ratios of these conditions. YLDs were calcu-lated as prevalence of a health state multiplied by a disability weight. These estimates were then corrected for comorbidity with other non- fatal diseases using methods described elsewhere in the GBD study.13

socio-demographic IndexSDI is a composite indicator that includes income per capita, average educational attainment over age 15 years and total fertility rate under age 25 years. The SDI has a value that ranges from 0 to 1. 0 represents the lowest income per capita, lowest educational attainmentand highest fertility under age 25 years observed across all GBD geographies from 1980 to 2017. 1 represents the highest income per capita, highest educational attainment and lowest fertility under 25 years observed across all GBD geographies from 1980 to 2017. The average relationship between YLLs, YLDs and YLDs divided by DALYs was calcu-lated with SDI using Gaussian process regression modelling. We used these estimates of expected DALY rates that were predicted based on the full range of SDI to determine whether observed health patterns deviated from trends associated with changes along the development spectrum.

GATher complianceThis study complies with the GATHER (Guidelines for Accurate and Transparent Health Estimates Reporting) recommendations (online supplementary appendix 2).

resuLTsMortality, incidence and burden of injury, 2017In 2017, worldwide 55.9 million (95% Uncertainty Interval (UI) 55.4 to 56.5 million) people died. Of these deaths, 4.5 million (95% UI 4.3 to 4.6 million), 8.0% (95% UI 7.7% to 8.2%), were due to injuries. Major causes of injury deaths were road injury (27.7%), self- harm (17.7%), falls (15.5%) and interpersonal violence (9.0%).

There were 521 million (95% UI 493 to 548 million) cases of non- fatal injuries in 2017, representing an increase of 167 million from the 354 million (95% UI 338 to 372 million) cases of non- fatal injuries in 1990. The global age- standardised injury death rate was 57.9 per 100 000 (95% UI 55.9 to 59.2), with highest death rates for road injury (15.8 deaths per 100 000 (95% UI 15.2 to 16.3)), self- harm (10.0 deaths per 100 000 (95% UI 9.4 to 10.3)) and falls (9.2 deaths per 100 000 (95% UI 8.5 to 9.8)) (see online supplementary appendix table 1). Injury death rates were over twice as high in men compared with women (80.9 per 100 000 (95% UI 77.7 to 83.0) and 35.5 per 100 000 (95% UI 33.9 to 36.5), respectively). The global age- standardised injury incidence rate was 6762.6 per 100 000 (95% UI 6412.0 to 7118.1)), with highest incidence rates for falls (2237.6 new cases per 100 000 (95% UI 1989.7 to 2532.3)) and mechanical forces (943.6 new cases per 100 000 (95% UI 808.7 to 1100.6)) (see online supplementary appendix table 1). Injury incidence rates were almost twice as high in men compared with women (7827.1 per 100 000



j.com/


ownloaded from








Original research

Figure 1 Age- standardised YLL and YLD rates for 17 cause- of- injury categories by level of Socio- demographic Index. YLL, years of life lost.

(95% UI 7435.3 to 8242.9) and 5654.5 per 100 000 (95% UI 5351.3 to 5962.1), respectively).

Injuries contributed 10.1% (9.7%–10.5%) to the global burden of disease in 2017 (3267.0 DALYs per 100 000 (95% UI 3058.2 to 3505.1)). YLLs were responsible for the majority of the injury DALYs (77%; 2548 YLLs per 100 000 (95% UI 2462 to 2610)). The main contributors to injury DALYs were road injuries (871.1 DALYs per 100 000 (95% UI 827.9 to 917.3); 26.7%), falls (459.5 DALYs per 100 000 (95% UI 387.1 to 547.5); 14.1%), self- harm (429.0 per 100 000 (95% UI 401.6 to 443.5); 13.1%), interpersonal violence (334.3 DALYs per 100 000 (95% UI 304.7 to 360.5); 10.2%) and drowning (230.0 DALYs per 100 000 (95% UI 219.1 to 241.2); 7.0%) (see online supplementary appendix table 2). The injury burden was highest in Syria (16 341.1 DALYs per 100 000 (95% UI 15 892.7 to 16 858.4), Central African Republic (11 012.7 DALYs per 100 000 (95% UI 8807.9 to 12 913.8)) and Lesotho (7951.3 DALYs per 100 000 (95% UI 6424.8 to 9407.4)) and lowest in Maldives (1282.4 DALYs per 100 000 (95% UI 1138.1 to 1572.9)), Bermuda (1432.2 DALYs per 100 000 (95% UI 1267.5 to 1606.7)) and Italy (1458.1 DALYs per 100 000 (95% UI 1237.2 to 1739.4)) (see online supplementary appendix table 3. SDI level for each country in 2017 is also provided).

Change over timeBetween 1990 and 2017, the age- standardised injury DALY rates have declined from 4946 (95% UI 4655 to 5233) to 3267 DALYs (95% UI 3058 to 3505) per 100 000, with largest absolute declines in drowning (from 635 (95% UI 571 to 689) to 230 (95% UI 219 to 241) DALYs per 100 000), road injuries (from 1259 (95% UI 1182 to 1330) to 871 (95% UI 828 to 917) DALYs per 100 000), self- harm (from 687 (95% UI 621 to 723) to 429 (95% UI 402 to 443) DALYs per 100 000), and fire, heat and hot substances

(from 197 (95% UI 157 to 228) to 111 (95% UI 93 to 129) DALYs per 100 000). Between 1990 and 2017, the age- standardised rates of YLDs and YLLs from injuries declined by 7.8% and 38.8%, respectively, while incidence of injuries only declined by 0.9%.

burden of injury by sdI levelThe contribution of cause- of- injury category DALY rates to the total injury DALY rates differed by year, age category, sex and SDI level. The largest disparity in DALY rate by SDI level was found in 0–6 days olds, ranging from a high of 52 374 DALYs per 100 000 in the lowest SDI quintile to a low of 6109 DALYs per 100 000 in the highest SDI quintile. In men aged 15–49 years, conflict and terrorism stands out because of the high difference between highest and lowest DALY rates by level of SDI (countries with low SDI 496 DALYs (95% UI 414 to 589) per 100 000; countries with high SDI 2 DALYs (95% UI 1 to 2) per 100 000).

YLL and YLD rates by SDI levelFor many causes of injury, age- standardised YLL and YLD rates declined strikingly with increasing SDI, with propor-tionally largest decreases in YLL rates for conflict and terrorism (low SDI level 163.4 YLLs per 100 000; high SDI level 0.06 YLLs per 100 000), animal contact (low SDI level 140.0 YLLs per 100 000; high SDI level 2 YLLs per 100 000) and other unintentional injuries (low SDI level 7993 YLLs per 100 000; high SDI level 8.4 YLLs per 100 000). Figure 1 shows the level of age- standardised YLLs and YLDs per 100 000 against SDI (all regions, all years 1990–2017) by cause- of- injury. Largest decreases in YLD rates were seen for cause- of- injury categories conflict and terrorism, exposure to forces of nature and adverse effects of medical treatment. Exceptions were road injuries, self- harm and interpersonal



j.com/


ownloaded from






Original research

Figure 2 Co- evolution of all injury age- standardised DALY rates with SDI for the world and 21 GBD regions for 1990–2017 with comparison with the values expected on the basis of SDI alone. DALY, disability- adjusted life year; GBD, Global Burden of Disease; SDI, Socio- demographic Index.

violence. The age- standardised YLL rate of road injuries was highest at the low- middle range SDI levels and lowest at higher SDI levels, whereas YLDs from road injuries increased at higher SDI. The age- standardised road injuries YLL rate increased from low SDI to low- middle SDI, but declined at higher levels of SDI. For falls, at higher levels of SDI, the composition of the disease burden shifted towards YLDs as the primary driver of DALYs. YLLs made up 63%, 61% and 20% of DALYs from falls in low, middle and high SDI quin-tiles, respectively. For road injuries, the proportion of YLLs dropped from 91% in countries with low SDI to 70% in countries with high SDI.

Expected based on SDI versus observed burden of injury by SDI level, 1990–2017Figure 2 shows the level of all injury age- standardised DALYs per 100 000 against SDI by GBD region from 1990 to 2017 in comparison with expected values (black line) based on SDI alone. The icons appearing above the black line for DALYs represent worse than expected injury DALYs and the icons appearing below represent better than expected injury DALYs. As SDI generally increases over time, successive markers represent years between 1990 and 2017. Regions where injury DALY rates were notably greater than expected based on SDI included Central and Southern Sub- Saharan Africa, Oceania, Eastern Europe, Central Europe and high- income North America. Regions where injury DALY rates were notably lower than expected based on SDI included Eastern and Western Sub- Saharan Africa, South Asia, South-east Asia and Western Europe.

Road injuryThe expected road injury DALY rate by SDI shows that most regions decreased in terms of road injury DALYs as

SDI increased over time (see figure 3). South Asia, East Asia, Southern Sub- Saharan Africa and Eastern Europe are excep-tions to this pattern, showing an initial increase and then a decline. In GBD 2017, the regions with worse than expected road injury DALYs based on SDI included North Africa and Middle East, Southern and Central Sub- Saharan Africa, Eastern Europe and Oceania, while regions with markedly better than expected rates included Eastern Sub- Saharan Africa, South Asia and Southern Latin America.

Interpersonal violenceIn 2017, in all regions except for Southern Sub- Saharan Africa, Central Latin America, Tropical Latin America, Eastern Europe, Caribbean, Oceania and high- income North America, the observed interpersonal violence DALY rates were better than expected based on SDI (see figure 4). Between 1990 and 2017, in most regions with higher than expected DALYs, the gap between observed and expected interpersonal violence DALY rates decreased, except for Caribbean and Tropical Latin America, where the gap increased.

Self-harmThe patterns of observed and expected self- harm DALYs based on SDI by GBD regions between 1990 and 2017 differed markedly from those of other injuries (see figure 5). In 1990, observed self- harm DALY rates in East Asia and Eastern Europe were worse than expected based on SDI but rapidly declined over time, with observed DALY rates lower than expected in 2017. Southern Sub- Saharan Africa had worse than expected DALY rates but the other regions of Sub- Saharan Africa had better than expected DALY rates. North Africa and Middle East, Western Europe, Southeast Asia, and Andean, Central and Tropical Latin America all had better than expected DALY rates.



j.com/


ownloaded from



Original research

Figure 3 Co- evolution of road injury age- standardised DALY rates with SDI for the world and 21 GBD regions for 1990–2017 with comparison with the values expected on the basis of SDI alone. DALY, disability- adjusted life year; GBD, Global Burden of Disease; SDI, Socio- demographic Index.

Figure 4 Co- evolution of interpersonal violence age- standardised DALY rates with SDI for the world and 21 GBD regions for 1990–2017 with comparison with the values expected on the basis of SDI alone. DALY, disability- adjusted life year; GBD, Global Burden of Disease; SDI, Socio- demographic Index.

DrowningDrowning DALY rates between 1990 and 2017 decreased in almost every GBD region regardless of their SDI value (figure 6), except for Oceania, Eastern Europe and Southern

Sub- Saharan Africa. Eastern and Western Sub- Saharan Africa, North Africa and Middle East, Andean, Tropical, Central and Southern Latin America, Western Europe and Austral-asia had better than expected DALY rates, while Oceania,



j.com/


ownloaded from



Original research

Figure 5 Co- evolution of self- harm age- standardised DALY rates with SDI for the world and 21 GBD regions for 1990–2017 with comparison with the values expected on the basis of SDI alone. DALY, disability- adjusted life year; GBD, Global Burden of Disease; SDI, Socio- demographic Index.

Figure 6 Co- evolution of drowning age- standardised DALY rates with SDI for the world and 21 GBD regions for 1990–2017 with comparison with the values expected on the basis of SDI alone. DALY, disability- adjusted life year; GBD, Global Burden of Disease; SDI, Socio- demographic Index.

East Asia and Eastern Europe had worse than expected DALY rates based on SDI.

FallsThe patterns in falls globally followed more dynamic trends across regions as SDI increased from 1990 to 2017 (see figure 7). The regions that performed worse than expected in terms of SDI were Central Europe, Eastern Europe, South

Asia, Central Asia and Australasia. Among these, Central Asia and Central Europe decreased and then increased, while Eastern Europe increased and then decreased. South Asia decreased steadily, while Australasia increased steadily until recent years. Among regions that performed better than expected, Oceania had increasing rates as SDI increased, while high- income North America dropped precipitously and then started increasing as SDI increased.



j.com/


ownloaded from



Original research

Figure 7 Co- evolution of falls age- standardised DALY rates with SDI for the world and 21 GBD regions for 1990–2017 with comparison with the values expected on the basis of SDI alone. DALY, disability- adjusted life year; GBD, Global Burden of Disease; SDI, Socio- demographic Index.

dIsCussIOnFor many causes of injury, age- standardised DALY rates declined strikingly with increasing SDI, although road injury, interper-sonal violence and self- harm did not strictly follow this pattern. Particularly for self- harm opposing patterns were observed in regions with similar SDI levels, for example, the trends in high- income Asia Pacific were opposite the trends in Western Europe, despite their proximity in terms of SDI. For road injuries, this effect was less pronounced; for nearly all regions, road injury DALY rates declined after 2005.

In Southern Sub- Saharan Africa, injury DALYs were worse than expected based on SDI in the overall injuries category as well as many of the specific injuries. In this region, road injury and interpersonal violence were important causes explaining the gap between observed and expected levels of overall injury DALYs. Many underlying and intertwining determinants of the high levels of interpersonal violence have been cited, including income inequality and poverty, high unemployment, rapid social change, corruption and poor rule of law, gender inequality, family breakdown, access to firearms, and alcohol and drug abuse.27 Despite these difficulties, however, and the worse- than- expected performance relative to SDI, our findings show that the DALY rates in Southern Sub- Saharan Africa have decreased from 2000 to 2017. This trend tallies with a reported declining number of injury deaths among young adults in South Africa.28

Of regions with a middle- high SDI, Eastern Europe stands out, because for most causes of injury, DALYs were much worse than expected based on SDI, particularly in the period 1990–2005. A compelling explanation for this finding may be the dissolution of the former Soviet Union and the resulting social and economic consequences on health and mortality.29 However, others have argued that causes of the health crisis are more complex and may result from a combination of historical and contemporary forces, including lifestyle habits, such as alcohol use, economic

impoverishment, widening social inequality and the breakdown of political institutions.30 31 It should be noted that our study did aim to assess determinants of the burden of injury and caution is needed in attempting to draw conclusions with regard to possible reasons for regional trends and differences.

Another notable finding from our study was that for falls, at the higher levels of SDI, the composition of the disease burden shifted towards YLDs, rather than YLLs, as a more prominent driver of DALYs compared with areas with lower SDI. The proportion of DALYs due to YLDs also increases with higher levels of SDI among other injuries. It is possible that this shift in distribution reflects decreased mortality among injuries when people in higher SDI locations have access to better healthcare services. The shift in road injuries, for example, could be brought about by injuryprevention measures reducing the severity of the injury sustained (eg, seat belts and helmets) or by improved access to better quality care after an injury (eg, trauma systems). It is also possible that in age- standardised analyses, the shift towards YLDs may be due to the ageing of the population of countries with high SDI with commensurate age- related increases in injury incidence. For example, the incidence of falls increases substan-tially with age and most of the burden from falls in high- SDI countries occurs in the very old.32

LimitationsOur analysis has several limitations. First, as SDI and time are correlated, we may be over interpreting SDI as a driver of change as it could well be driven largely by other factors changing over time, not necessarily linked to SDI, such as climate change.

Second, limited data are available to quantify burden of injuries in the world. Major limitations of the cause- of- death data are low or absent coverage of vital registration or verbal autopsy data in many parts of the world, incompleteness of



j.com/


ownloaded from



Original research

What is already known on the subject

► Morbidity and mortality from injuries are known to be affected by socioeconomic development.

What this study adds

► This study provides more recent estimates of global morbidity and mortality from injuries with a greater level of detail than has previously been reported and with an updated method for measuring sociodemographic development.

► This study found that many injuries decreased in terms of morbidity and mortality as sociodemographic development increased over time, but also identified important exceptions to this trend.

► The study adds to the body of discussion on how economic development and sociodemographic changes should be considered in preventing future injury burden.

death certification systems and differences in the proportion of injury deaths classified in ill- defined codes.33–36 Few data were available for non- fatal injuries, and if data were avail-able, injury was frequently recorded as a mix of cause and nature- of- injury codes and often a preponderance of nature- of- injury codes, while our analyses require attributing health outcomes to cause of injury. As a result, many non- fatal injury hospitals and emergency departments data sets could not be used. Furthermore, short- term duration of several nature- of- injury categories was based on expert- driven estimates because patient data was not available. Besides, gathering data on deaths and morbidity due to forces of nature (ie, disasters) and collective violence is complicated by the fact that their aftereffects may severely disrupt the infrastructure of vital and health registration systems.37 The statistical methods that we have used to assess mortality, incidence and prevalence can borrow strength over time and geography to ensure an esti-mate for all causes and all countries. Nevertheless, estimates for populations and time periods with few or absent data are inherently less precise.

Non- fatal injuries are reported by both cause of injury and nature of injury. Since our model requires a one- to- one rela-tionship between cause- of- injury and nature- of- injury cate-gory, we developed a nature- of- injuries severity hierarchy that selects the injury that was likely to be responsible for the largest burden in a person with more than one injury. This means that we ignore the other injuries sustained by such individuals and this may have led to some underesti-mation of the burden of non- fatal injury. We decided to use such a hierarchy after it proved difficult to use statistical methods on sparse data to parse estimates across co- occur-ring injuries.

A second methodological limitation is the assessment of the probability of permanent health loss, one of the main drivers of non- fatal burden of disease. The probability of long- term injury was based on patient- reported outcome data from follow- up studies in just three countries. Also, long- term patient- reported outcome data may be influenced by response shift bias. Response shift is a change of outcome due to a change of the measurement perspective of the respon-dent (‘internal measurement scale’), where the usual change is towards adaptation. In our study, response shift may have resulted in an underestimation of the severity of long- term consequences of injury and consequently, to an underestima-tion of the non- fatal burden of injury.

Third, even though a strong correlation between SDI and injury DALYs, YLLs and YLDs was found, this cannot be inter-preted as being causal in nature, because income per capita and education, two of the three components of SDI, were also used as covariates in all of the injury models except exposure to forces of nature and collective violence and legal intervention. In its original formulation, Murray et al suggested that SDI utility may be improved in the future through consideration of additional societal elements, such as inequality in each component.1

COnCLusIOnsThe overall pattern is that of declining injury burden with increasing development. Not all injuries follow this pattern, suggesting that there are multiple underlying mechanisms influ-encing injury outcomes. The detailed understanding of these patterns helps to inform countries how best to respond to changes in injury outcomes that occur with development and, in case of countries where health gains outpace development, may

help to identify which prevention and/or healthcare measures have been taken in these countries.

Author affiliations1Department of Public Health, Erasmus University Medical Center, Rotterdam, The Netherlands2Institute for Health Metrics and Evaluation, University of Washington, Seattle, Washington, USA3Department of Medicine, University College Hospital Ibadan, Ibadan, Nigeria4Department of Anesthesiology, Kermanshah University of Medical Sciences, Kermanshah, Iran5Department of Epidemiology, Jimma University, Jimma, Ethiopia6Higher National School of Veterinary Medicine, Algiers, Algeria7Evidence Based Practice Center, Mayo Clinic Foundation for Medical Education and Research, Rochester, Minnesota, USA8Department of Population Health Research, King Abdullah International Medical Research Center, Riyadh, Saudi Arabia9Department of Health Policy and Management, Kuwait University, Safat, Kuwait10International Centre for Casemix and Clinical Coding, National University of Malaysia, Bandar Tun Razak, Malaysia11Department of Family and Community Medicine, King Abdulaziz University, Jeddah, Saudi Arabia12Department of Oral and Maxillofacial Surgery, University Hospital Knappschaftskrankenhaus Bochum, Bochum, Germany13King Saud University, Riyadh, Saudi Arabia14Social Determinants of Health Research Center, Rafsanjan University of Medical Sciences, Rafsanjan, Iran15Department of Health Policy and Administration, University of the Philippines Manila, Manila, Philippines16Department of Applied Social Sciences, Hong Kong Polytechnic University, Hong Kong, China17Department of Sociology and Social Work, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana18Center for International Health, Ludwig Maximilians University, Munich, Germany19School of Health Sciences, Birmingham City University, Birmingham, UK20School of Science and Health, Western Sydney University, Sydney, New South Wales, Australia21Oral Health Services, Sydney Local Health District, Sydney, New South Wales, Australia22Qom University of Medical Sciences, Qom, Iran23Education Development Center, Mashhad University of Medical Sciences, Mashhad, Iran24Indian Institute of Public Health, Gandhinagar, India25The Judith Lumley Centre, La Trobe University, Melbourne, Victoria, Australia26General Office for Research and Technological Transfer, Peruvian National Institute of Health, Lima, Peru27School of Public Health, Auckland University of Technology, Auckland, New Zealand28Department of Research, Public Health Perspective Nepal, Pokhara- Lekhnath Metropolitan City, Nepal29School of Psychology, University of Auckland, Auckland, New Zealand



j.com/


ownloaded from



Original research

30Heidelberg Institute of Global Health (HIGH), Heidelberg University, Heidelberg, Germany31T.H. Chan School of Public Health, Harvard University, Boston, Massachusetts, USA32Department of Psychiatry, Charles R. Drew University of Medicine and Science, Los Angeles, California, USA33Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, California, USA34Department of Community Medicine, Gandhi Medical College Bhopal, Bhopal, India35Social Determinants of Health Research Center, Lorestan University of Medical Sciences, Khorramabad, Iran36Department of Epidemiology and Biostatistics, Lorestan University of Medical Sciences, Khorramabad, Iran37Department of Epidemiology and Psychosocial Research, Ramón de la Fuente Muñiz National Institute of Psychiatry, Mexico City, Mexico38Nuffield Department of Population Health, University of Oxford, Oxford, UK39Department of Internal Medicine, University of São Paulo, São Paulo, Brazil40The George Institute for Global Health, New Delhi, India41Centre for Global Child Health, The Hospital for Sick Children, Toronto, Ontario, Canada42Centre of Excellence in Women and Child Health, Aga Khan University, Karachi, Pakistan43Social Determinants of Health Research Center, Babol University of Medical Sciences, Babol, Iran44Centre for Adolescent Health, Murdoch Childrens Research Institute, Melbourne, Victoria, Australia45School of Population and Global Health, University of Melbourne, Melbourne, Victoria, Australia46Transport & Digital Development, World Bank, Washington, District of Columbia, USA47Transport and Road Safety (TARS) Research Department, University of New South Wales, Sydney, New South Wales, Australia48Institute of Epidemiology, Comenius University, Bratislava, Slovakia49National Institute of Public Health, Cuernavaca, Mexico50School of Medicine, University of the Valley of Cuernavaca, Cuernavaca, Mexico51Department of Population and Health, Metropolitan Autonomous University, Mexico City, Mexico52Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden53UCIBIO, University of Porto, Porto, Portugal54Colombian National Health Observatory, National Institute of Health, Bogota, Colombia55Epidemiology and Public Health Evaluation Group, National University of Colombia, Bogota, Colombia56National School of Public Health, Carlos III Health Institute, Madrid, Spain57Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada58Department of Biochemistry and Biomedical Science, Seoul National University Hospital, Seoul, South Korea59Department of Pulmonary Medicine, Christian Medical College and Hospital (CMC), Vellore, India60Division of Plastic Surgery, University of Washington, Seattle, Washington, USA61Institute of Public Health Kalyani, Kalyani, India62School of Health Science, Orebro University, Orebro, Sweden63Toxoplasmosis Research Center, Mazandaran University of Medical Sciences, Sari, Iran64Department of General Surgery, Carol Davila University of Medicine and Pharmacy, Bucharest, Romania65Department of Surgery, Clinical Emergency Hospital Sf. Pantelimon, Bucharest, Romania66National Drug and Alcohol Research Centre, University of New South Wales, Sydney, New South Wales, Australia67Australian Institute for Suicide Research and Prevention, Griffith University, Mount Gravatt, Queensland, Australia68Department of Global Health and Infection, Brighton and Sussex Medical School, Brighton, UK69School of Public Health, Addis Ababa University, Addis Ababa, Ethiopia70Department of Nursing, Debre Markos University, Debre Markos, Ethiopia71Centre for Global Child Health, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada72Department of Community Medicine, University of Peradeniya, Peradeniya, Sri Lanka73Center of Excellence in Health Service Management, Nguyen Tat Thanh University, Ho Chi Minh, Vietnam74University of New South Wales, Sydney, New South Wales, Australia75Sydney School of Public Health, University of Sydney, Sydney, New South Wales, Australia76United Nations World Food Programme, New Delhi, India77Department of Public Health Sciences, Karolinska Institutet, Stockholm, Sweden

78World Health Programme, Université du Québec en Abitibi- Témiscamingue, Rouyn- Noranda, Quebec, Canada79Department of Pathology, Stavanger University Hospital, Stavanger, Norway80Department of Clinical Pathology, Mansoura University, Mansoura, Egypt81Public Health Department, Saint Paul’s Hospital Millennium Medical College, Addis Ababa, Ethiopia82Multiple Sclerosis Research Center, Tehran University of Medical Sciences, Tehran, Iran83Department of Psychology, Federal University of Sergipe, Sao Cristovao, Brazil84Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden85Division of Neurology, University of Ottawa, Ottawa, Ontario, Canada86REQUIMTE/LAQV, University of Porto, Porto, Portugal87Psychiatry Department, Kaiser Permanente, Fontana, California, USA88School of Health Sciences, A.T. Still University, Mesa, Arizona, USA89Department of Population Medicine and Health Services Research, Bielefeld University, Bielefeld, Germany90College of Public Health, Medical and Veterinary Science, James Cook University, Douglas, Queensland, Australia91Gene Expression & Regulation Program, The Wistar Institute, Philadelphia, Pennsylvania, USA92Department of Dermatology, Kobe University, Kobe, Japan93Department of Biostatistics, Mekelle University, Mekelle, Ethiopia94Adelaide Medical School, University of Adelaide, Adelaide, South Australia, Australia95Center for Clinical and Epidemiological Research, University of São Paulo, Sao Paulo, Brazil96Internal Medicine Department, University Hospital, University of São Paulo, Sao Paulo, Brazil97School of Medicine, Boston University, Boston, Massachusetts, USA98School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia99Department of Epidemiology and Biostatistics, Zhengzhou University, Zhengzhou, China100March of Dimes, Arlington, Virginia, USA101School of Public Health, West Virginia University, Morgantown, West Virginia, USA102Institute for Global Health, University College London, London, UK103Department of Pharmacology, Tehran University of Medical Sciences, Tehran, Iran104Obesity Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran105Department of Radiology, Johns Hopkins University, Baltimore, Maryland, USA106Department of Family and Community Medicine, Arabian Gulf University, Manama, Bahrain107School of Health and Environmental Studies, Hamdan Bin Mohammed Smart University, Dubai, United Arab Emirates108Biomedical Research Networking Center for Mental Health Network (CiberSAM), Madrid, Spain109Research and Development Unit, San Juan de Dios Sanitary Park, Sant Boi de Llobregat, Spain110School of Nursing and Midwifery, Tabriz University of Medical Sciences, Tabriz, Iran111Independent Consultant, Tabriz, Iran112Department of Public Health, Mizan- Tepi University, Teppi, Ethiopia113Unit of Epidemiology and Social Medicine, University Hospital Antwerp, Wilrijk, Belgium114Clinical Sciences, Karolinska University Hospital, Stockholm, Sweden115School of Public Health, Curtin University, Perth, Western Australia, Australia116Research Coordination, AC Environments Foundation, Cuernavaca, Mexico117CISS, National Institute of Public Health, Cuernavaca, Mexico118Department of Pediatrics, Dell Medical School, University of Texas Austin, Austin, Texas, USA119Guilan Road Trauma Research Center, Guilan University of Medical Sciences, Rasht, Iran120Social Determinants of Health Research Center, Guilan University of Medical Sciences, Rasht, Iran121Department of Pharmacology and Therapeutics, Dhaka Medical College, Dhaka University, Dhaka, Bangladesh122Department of Pharmacology, Bangladesh Industrial Gases Limited, Tangail, Bangladesh123Faculty of Dentistry, Department of Legal Medicine and Bioethics, Carol Davila University of Medicine and Pharmacy, Bucharest, Romania124Clinical Legal Medicine Department, National Institute of Legal Medicine Mina Minovici, Bucharest, Romania125Department of Epidemiology and Health Statistics, Central South University, Changsha, China126School of Public Health, University of the Western Cape, Bellville, Cape Town, South Africa127Department of Public Health, Walter Sisulu University, Mthatha, South Africa128Department of Community Medicine, University of Ibadan, Ibadan, Nigeria



j.com/


ownloaded from



Original research

129Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran130Institute for Physical Activity and Nutrition, Deakin University, Burwood, Victoria, Australia131Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia132Injury Division, The George Institute for Global Health, Newtown, New South Wales, Australia133Department of Global and Community Health, George Mason University, Fairfax, Virginia, USA134School of Management and Medical Education, Shahid Beheshti University of Medical Sciences, Tehran, Iran135Safety Promotion and Injury Prevention Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran136Department for Health Care and Public Health, I.M. Sechenov First Moscow State Medical University, Moscow, Russia137Institute of Medicine, University of Colombo, Colombo, Sri Lanka138Faculty of Graduate Studies, University of Colombo, Colombo, Sri Lanka139Department of Community Medicine, Banaras Hindu University, Varanasi, India140Department of Ophthalmology, Heidelberg University, Mannheim, Germany141Beijing Institute of Ophthalmology, Beijing Tongren Hospital, Beijing, China142Social Determinants of Health Research Center, University of Social Welfare and Rehabilitation Sciences, Tehran, Iran143Department of Family Medicine and Public Health, University of Opole, Opole, Poland144Institute of Family Medicine and Public Health, University of Tartu, Tartu, Estonia145Minimally Invasive Surgery Research Center, Iran University of Medical Sciences, Tehran, Iran146Department of Neurology, University of Washington, Seattle, Washington, USA147Hematology- Oncology and Stem Cell Transplantation Research Center, Tehran University of Medical Sciences, Tehran, Iran148Pars Advanced and Minimally Invasive Medical Manners Research Center, Iran University of Medical Sciences, Tehran, Iran149Department of Dermatology, Wolaita Sodo University, Wolaita Sodo, Ethiopia150Non- communicable Diseases Research Unit, Medical Research Council South Africa, Cape Town, South Africa151Department of Medicine, University of Cape Town, Cape Town, South Africa152Department of Public Health and Community Medicine, Jordan University of Science and Technology, Ramtha, Jordan153Social Determinants of Health Research Center, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran154School of Food and Agricultural Sciences, University of Management and Technology, Lahore, Pakistan155Epidemiology and Biostatistics Department, Health Services Academy, Islamabad, Pakistan156Department of Public Health, Imam Muhammad Ibn Saud Islamic University, Riyadh, Saudi Arabia157Department of Health Policy and Management, Johns Hopkins University, Baltimore, Maryland, USA158Clinical Epidemiology Unit, Lund University, Lund, Sweden159Department of Preventive Medicine, Korea University, Seoul, South Korea160Department of Health Sciences, Northeastern University, Boston, Massachusetts, USA161School of Health Sciences, Kristiania University College, Oslo, Norway162CIBERSAM, San Juan de Dios Sanitary Park, Sant Boi de Llobregat, Spain163Catalan Institution for Research and Advanced Studies (ICREA), Barcelona, Spain164Department of Demography, University of Montreal, Montreal, Quebec, Canada165Department of Social and Preventive Medicine, University of Montreal, Montreal, Quebec, Canada166Department of Public Health, Yuksek Ihtisas University, Ankara, Turkey167Department of Public Health, Hacettepe University, Ankara, Turkey168Department of Psychiatry, University of Nairobi, Nairobi, Kenya169Division of Psychology and Language Sciences, University College London, London, UK170School of Dentistry, The University of Queensland, Brisbane, Queensland, Australia171Institute of Health Policy and Development Studies, National Institutes of Health, Manila, Philippines172Department of Community and Family Medicine, University of Baghdad, Baghdad, Iraq173HelpMeSee, New York City, New York, USA174International Relations Department, Mexican Institute of Ophthalmology, Queretaro, Mexico175College of Optometry, Nova Southeastern University, Fort Lauderdale, Florida, USA176School of Public Health, University of Haifa, Haifa, Israel177Department of General Surgery, Aintree University Hospital National Health Service (NHS) Foundation Trust, Liverpool, UK178Department of Surgery, University of Liverpool, Liverpool, UK179Anesthesiology, Pain and Intensive Care Department, Federal University of São Paulo, Sao Paulo, Brazil

180Radiology Department, Mansoura Faculty of Medicine, Mansoura, Egypt181Ophthalmology Department, Aswan Faculty of Medicine, Aswan, Egypt182Institute of Medicine, Tribhuvan University, Kathmandu, Nepal183Department of Public Health, Trnava University, Trnava, Slovakia184Department of Primary Care and Public Health, Imperial College London, London, UK185Digestive Diseases Research Institute, Tehran University of Medical Sciences, Tehran, Iran186Non- communicable Diseases Research Center, Shiraz University of Medical Sciences, Shiraz, Iran187Department of Humanities and Social Sciences, Indian Institute of Technology, Roorkee, Haridwar, India188Department of Development Studies, International Institute for Population Sciences, Mumbai, India189Department of Maternal and Child Nursing and Public Health, Federal University of Minas Gerais, Belo Horizonte, Brazil190Surgery Department, Emergency University Hospital Bucharest, Bucharest, Romania191Department of Epidemiology and Biostatistics, Tehran University of Medical Sciences, Tehran, Iran192Research Department, The George Institute for Global Health, New Delhi, India193School of Medicine, University of New South Wales, Sydney, New South Wales, Australia194School of Public Health, Bahir Dar University, Bahir Dar, Ethiopia195Neurology Department, Janakpuri Super Specialty Hospital Society, New Delhi, India196Neurology Department, Govind Ballabh Institute of Medical Education and Research, New Delhi, India197Department of Medical Laboratory Sciences, Bahir Dar University, Bahir Dar, Ethiopia198Peru Country Office, United Nations Population Fund (UNFPA), Lima, Peru199Department of Epidemiology and Biostatistics, Haramaya University, Harar, Ethiopia200Breast Surgery Unit, Helsinki University Hospital, Helsinki, Finland201University of Helsinki, Helsinki, Finland202Neurocenter, Helsinki University Hospital, Helsinki, Finland203School of Health Sciences, University of Melbourne, Melbourne, Victoria, Australia204Clinical Microbiology and Parasitology Unit, Zora Profozic Polyclinic, Zagreb, Croatia205University Centre Varazdin, University North, Varazdin, Croatia206Department of Propedeutics of Internal Diseases & Arterial Hypertension, Pomeranian Medical University, Szczecin, Poland207Pacific Institute for Research & Evaluation, Calverton, Maryland, USA208Achutha Menon Centre for Health Science Studies, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum, India209Global Institute of Public Health (GIPH), Ananthapuri Hospitals and Research Centre, Trivandrum, India210Faculty of Internal Medicine, Kyrgyz State Medical Academy, Bishkek, Kyrgyzstan211Department of Atherosclerosis and Coronary Heart Disease, National Center of Cardiology and Internal Disease, Bishkek, Kyrgyzstan212Institute of Addiction Research (ISFF), Frankfurt University of Applied Sciences, Frankfurt, Germany213Department of Biostatistics, Hamadan University of Medical Sciences, Hamadan, Iran214Hamadan University of Medical Sciences, Hamadan, Iran215Health Systems and Policy Research Unit, Ahmadu Bello University, Zaria, Nigeria216Faculty of Life Sciences and Medicine, King’s College London, London, UK217Clinical Epidemiology and Public Health Research Unit, Burlo Garofolo Institute for Maternal and Child Health, Trieste, Italy218Department of Biomedical and Dental Sciences and Morphofunctional Imaging, University of Messina, Messina, Italy219Department of Neurology, Oasi Research Institute, Troina, Italy220Department of Public Health Sciences, University of Miami, Miami, Florida, USA221Center for Health Systems Research, National Institute of Public Health, Cuernavaca, Mexico222Department of Public Health Medicine, University of KwaZulu- Natal, Durban, South Africa223Health Sciences Research Center, Mazandaran University of Medical Sciences, Sari, Iran224Social Determinants of Health Research Center, Kurdistan University of Medical Sciences, Sanandaj, Iran225Department of Epidemiology and Biostatistics, Kurdistan University of Medical Sciences, Sanandaj, Iran226Preventive Medicine and Public Health Research Center, Iran University of Medical Sciences, Tehran, Iran227International Laboratory for Air Quality and Health, Queensland University of Technology, Brisbane, Queensland, Australia228Gorgas Memorial Institute for Health Studies, Panama City, Panama



j.com/


ownloaded from



Original research

229Department of Surgery, University of Washington, Seattle, Washington, USA2301st Department of Ophthalmology, University of Athens, Athens, Greece231Biomedical Research Foundation, Academy of Athens, Athens, Greece232Health Management Reserach Center, Baqiyatallah University of Medical Sciences, Tehran, Iran233Department of Health Management and Economics, Tehran University of Medical Sciences, Tehran, Iran234Department of Pediatrics, University of British Columbia, Vancouver, British Columbia, Canada235School of Medical Sciences, Science University of Malaysia, Kubang Kerian, Malaysia236Department of Epidemiology, University of Alabama at Birmingham, Birmingham, Alabama, USA237Department of Epidemiology & Biostatistics, Kermanshah University of Medical Sciences, Kermanshah, Iran238Suraj Eye Institute, Nagpur, India239Hospital of the Federal University of Minas Gerais, Federal University of Minas Gerais, Belo Horizonte, Brazil240Cochrane South Africa, South African Medical Research Council, Cape Town, South Africa241General Surgery Department, Emergency Hospital of Bucharest, Bucharest, Romania242Center of Excellence in Behavioral Medicine, Nguyen Tat Thanh University, Ho Chi Minh, Vietnam243Institute for Global Health Innovations, Duy Tan University, Hanoi, Vietnam244Public Health Department, Universitas Negeri Semarang, Kota Semarang, Indonesia245Graduate Institute of Biomedical Informatics, Taipei Medical University, Taipei City, Taiwan246Clinical Pharmacy Unit, Mekelle University, Mekelle, Ethiopia247Centre of Cardiovascular Research and Education in Therapeutics, Monash University, Melbourne, Victoria, Australia248Independent Consultant, Accra, Ghana249Translational Health Research Institute, Western Sydney University, Penrith, New South Wales, Australia250Department of Preventive Medicine, Kyung Hee University, Dongdaemun- gu, South Korea251HAST, Human Sciences Research Council, Durban, South Africa252School of Public Health, Faculty of Health Sciences, University of Namibia, Osakhati, Namibia253Department of Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, ON, Canada254Department of Psychiatry, University of Lagos, Lagos, Nigeria255Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada256Institute of Physical Activity and Health, Autonomous University of Chile, Talca, Chile257Applied Research Division, Public Health Agency of Canada, Ottawa, Ontario, Canada258School of Psychology, University of Ottawa, Ottawa, Ontario, Canada259Analytical Center, Moscow Institute of Physics and Technology, Dolgoprudny, Russia260Committee for the Comprehensive Assessment of Medical Devices and Information Technology, Health Technology Assessment Association, Moscow, Russia261Department of Respiratory Medicine, Jagadguru Sri Shivarathreeswara Academy of Health Education and Research, Mysore, India262Department of Medicine, Ottawa Hospital Research Institute, University of Ottawa, Ottawa, Ontario, Canada263Department of Medical Humanities and Social Medicine, Kosin University, Busan, South Korea264Department of Paediatrics, University of Melbourne, Melbourne, Victoria, Australia265Population Health Department, Murdoch Childrens Research Institute, Melbourne, Victoria, Australia266School of Optometry and Vision Science, University of New South Wales, Sydney, New South Wales, Australia267Shanghai Mental Health Center, Shanghai Jiao Tong University, Shanghai, China268Department of Psychiatry, Department of Epidemiology, Columbia University, New York City, New York, USA269Department of Nephrology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India270College of Medicine, University of Central Florida, Orlando, Florida, USA271College of Graduate Health Sciences, A.T. Still University, Mesa, Arizona, USA272Department of Epidemiology & Biostatistics, Contech School of Public Health, Lahore, Pakistan273Department of Immunology, Mazandaran University of Medical Sciences, Sari, Iran274Molecular and Cell Biology Research Center, Mazandaran University of Medical Sciences, Sari, Iran275Faculty of Medicine, Mazandaran University of Medical Sciences, Sari, Iran

276Sina Trauma and Surgery Research Center, Tehran University of Medical Sciences, Tehran, Iran277School of Nursing and Healthcare Professions, Federation University, Heidelberg, Victoria, Australia278National Centre for Farmer Health, Deakin University, Waurn Ponds, Victoria, Australia279Society for Health and Demographic Surveillance, Suri, India280Department of Economics, University of Göttingen, Göttingen, Germany281Department of Pharmacology, Shahid Beheshti University of Medical Sciences, Tehran, Iran282Academic Public Health Department, Public Health England, London, UK283WHO Collaborating Centre for Public Health Education and Training, Imperial College London, London, UK284University College London Hospitals, London, UK285School of Social Sciences and Psychology, Western Sydney University, Penrith, New South Wales, Australia286Brien Holden Vision Institute, Sydney, New South Wales, Australia287Organization for the Prevention of Blindness, Paris, France288Kermanshah University of Medical Sciences, Kermanshah, Iran289Department of Clinical Research, Federal University of Uberlândia, Uberlândia, Brazil290Golestan Research Center of Gastroenterology and Hepatology, Golestan University of Medical Sciences, Gorgan, Iran291National Institute for Research in Environmental Health, Indian Council of Medical Research, Bhopal, India292College of Medicine, University of Sharjah, Sharjah, United Arab Emirates293Social Development & Health Promotion Research Center, Kermanshah University of Medical Sciences, Kermanshah, Iran294Health and Disability Intelligence Group, Ministry of Health, Wellington, New Zealand295Department of Entomology, Ain Shams University, Cairo, Egypt296Department of Surgery, Marshall University, Huntington, West Virginia, USA297Department of Nutrition and Preventive Medicine, Case Western Reserve University, Cleveland, Ohio, USA298Rheumatology Department, University Hospitals Bristol NHS Foundation Trust, Bristol, UK299Institute of Bone and Joint Research, University of Sydney, Syndey, New South Wales, Australia300Institute of Social Medicine, University of Belgrade, Belgrade, Serbia301Centre- School of Public Health and Health Management, University of Belgrade, Belgrade, Serbia302UGC Centre of Advanced Study in Psychology, Utkal University, Bhubaneswar, India303Udyam- Global Association for Sustainable Development, Bhubaneswar, India304Department of Public Health Sciences, University of North Carolina at Charlotte, Charlotte, North Carolina, USA305School of Public Health, Imperial College London, London, UK306Market Access Department, Bayer, Istanbul, Turkey307School of Health Sciences, Federal University of Santa Catarina, Ararangua, Brazil308Department of Psychology, University of Alabama at Birmingham, Birmingham, Alabama, USA309Department of Psychiatry, Stellenbosch University, Cape Town, South Africa310Independent Consultant, Karachi, Pakistan311School of Medicine, Dezful University of Medical Sciences, Dezful, Iran312School of Medicine, Alborz University of Medical Sciences, Karaj, Iran313Chronic Diseases (Home Care) Research Center, Hamadan University of Medical Sciences, Hamadan, Iran314Centre for Medical Informatics, University of Edinburgh, Edinburgh, UK315Division of General Internal Medicine and Primary Care, Harvard University, Boston, Massachusetts, USA316Center for Pediatric Trauma Research, Research Institute at Nationwide Children’s Hospital, Columbus, Ohio, USA317National Institute of Infectious Diseases, Tokyo, Japan318Finnish Institute of Occupational Health, Helsinki, Finland319Institute of Medical Epidemiology, Martin Luther University Halle- Wittenberg, Halle, Germany320Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA321Medicine Service, US Department of Veteran Affairs, Birmingham, Alabama, USA322Department of Epidemiology, School of Preventive Oncology, Patna, India323Department of Epidemiology, Healis Sekhsaria Institute for Public Health, Mumbai, India324Department of Diseases and Noncommunicable Diseases and Health Promotion, Federal Ministry of Health, Brasilia, Brazil325Hospital Universitario de la Princesa, Autonomous University of Madrid, Madrid, Spain326Centro de Investigación Biomédica en Red Enfermedades Respiratorias (CIBERES), Madrid, Spain



j.com/


ownloaded from



Original research

327Department of Research Development, Federal Research Institute for Health Organization and Informatics of the Ministry of Health (FRIHOI), Moscow, Russia328Hull York Medical School, University of Hull, Hull City, UK329Usher Institute of Population Health Sciences and Informatics, University of Edinburgh, Edinburgh, UK330Federal Research Institute for Health Organization and Informatics of the Ministry of Health (FRIHOI), Moscow, Russia331Department of Psychiatry and Mental Health, University of Cape Town, Cape Town, South Africa332Department of Psychology, Deakin University, Burwood, Victoria, Australia333Department of Community Medicine, Ahmadu Bello University, Zaria, Nigeria334Department of Anesthesiology & Pain Medicine, University of Washington, Seattle, Washington, USA335Department of Criminology, Law and Society, University of California Irvine, Irvine, California, USA336Department of Medicine, University of Valencia, Valencia, Spain337Carlos III Health Institute, Biomedical Research Networking Center for Mental Health Network (CiberSAM), Madrid, Spain338School of Social Work, University of Illinois, Urbana, Illinois, USA339Department of Community Medicine, Iran University of Medical Sciences, Tehran, Iran340Institute of Public Health, University of Gondar, Gondar, Ethiopia341School of Public Health, University of Adelaide, Adelaide, South Australia, Australia342School of Public Health, Post Graduate Institute of Medical Education and Research, Chandigarh, India343Molecular Medicine and Pathology Department, University of Auckland, Auckland, New Zealand344Clinical Hematology and Toxicology, Military Medical University, Hanoi, Vietnam345Department of Health Economics, Hanoi Medical University, Hanoi, Vietnam346Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore, Singapore347Division of Health Sciences, University of Warwick, Coventry, UK348Department of Community Medicine, University of Nigeria Nsukka, Enugu, Nigeria349Argentine Society of Medicine, Buenos Aires, Argentina350Velez Sarsfield Hospital, Buenos Aires, Argentina351Central Research Institute of Cytology and Genetics, Federal Research Institute for Health Organization and Informatics of the Ministry of Health (FRIHOI), Moscow, Russia352Department of Statistics, University of Brasília, Brasília, Brazil353Directorate of Social Studies and Policies, Federal District Planning Company, Brasília, Brazil354Raffles Neuroscience Centre, Raffles Hospital, Singapore, Singapore355Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore356Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy357Occupational Health Unit, Sant’Orsola Malpighi Hospital, Bologna, Italy358Department of Health Care Administration and Economics, National Research University Higher School of Economics, Moscow, Russia359Foundation University Medical College, Foundation University, Islamabad, Pakistan360Department of Psychiatry, University of São Paulo, São Paulo, Brazil361Department of Psychology and Counselling, University of Melbourne, Melbourne, Victoria, Australia362Department of Medicine, University of Melbourne, St Albans, Victoria, Australia363Institute of Health and Society, University of Oslo, Oslo, Norway364Department of Neurology, Technical University of Munich, Munich, Germany365Centre for the Study of Regional Development, Jawahar Lal Nehru University, New Delhi, India366Department of Preventive Medicine, Northwestern University, Chicago, Illinois, USA367Centre for Suicide Research and Prevention, University of Hong Kong, Hong Kong, China368Department of Social Work and Social Administration, University of Hong Kong, Hong Kong, China369School of Allied Health Sciences, Addis Ababa University, Addis Ababa, Ethiopia370Department of Psychopharmacology, National Center of Neurology and Psychiatry, Tokyo, Japan371Health Economics & Finance, Jackson State University, Jackson, Mississippi, USA372School of Medicine, Tsinghua University, Beijing, China373Department of Epidemiology and Biostatistics, Wuhan University, Wuhan, China374Global Health Institute, Wuhan University, Wuhan, China375Department of Obstetrics & Gynaecology, A.C.S. Medical College and Hospital, Islamabad, Pakistan376Department of Epidemiology, University Hospital of Setif, Setif, Algeria377Maternal and Child Health Division, International Centre for Diarrhoeal Disease Research, Dhaka, Bangladesh378Department of Medicine, Monash University, Melbourne, Victoria, Australia379Student Research Committee, Babol University of Medical Sciences, Babol, Iran

380School of Public Health and Management, Chongqing Medical University, Chongqing, China381Indian Institute of Public Health, Public Health Foundation of India, Gurugram, India382Department of Health Metrics Sciences, School of Medicine, University of Washington, Seattle, Washington, USA383University of Melbourne, Melbourne, Queensland, Australia

Acknowledgements Mihajlo Jakovljevic Serbia acknowledges support through the Grant OI 175 014 of the Ministry of Education Science and Technological Development of the Republic of Serbia. Shahrzad Bazargan- Hejai acknowledges support through the NIH National Center for Advancing Translational Science (NCATS) UCLA CTSI Grant Number UL1TR001881". Ashish Awasthi acknowledges support from the Department of Science and Technology, Government of India, New Delhi through INSPIRE Faculty program. Rafael Tabarés- Seisdedos acknowledges support in part by grant number PROMETEOII/2015/021 from Generalitat Valenciana and the national grant PI17/00719 from ISCIII- FEDER. Abdallah M Samy acknowledges support from a fellowship from the Egyptian Fulbright Mission Program. Eduarda Fernandes acknowledges support ID/MULTI/04378/2019 and UID/QUI/50006/2019 with FCT/MCTES support through Portuguese national funds. Félix Carvalho acknowledges support ID/MULTI/04378/2019 and UID/QUI/50006/2019 with FCT/MCTES support through Portuguese national funds. Ilais Moreno Velásquezis acknowledges support from the Sistema Nacional de Investigacion, SENACYT (Panama). Louisa Degenhardt acknowledges support by an NHMRC research fellowship (#1135991) and by NIH grant NIDA R01DA1104470; The National Drug and Alcohol Research Centre is supported by funding from the Australian Government Department of Health under the Drug and Alcohol Program. Milena Santric Milicevic acknowledges the support from the Ministry of Education, Science and Technological Development, Republic of Serbia (Contract No. 175087). Kebede Deribe KD is supported by a grant from the Wellcome Trust [grant number 201900] as part of his International Intermediate Fellowship. Syed Aljunid acknowledges support from the International Centre for Casemix and Clinical Coding, Faculty of Medicine, National University of Malaysia and Department of Health Policy and Management, Faculty of Public Health, Kuwait University for the approval and support to participate in this research project. Jan- Walter De Neve was supported by the Alexander von Humboldt Foundation. Michael R Phillips acknowledges support from the Chinese National Natural Science Foundation of China (NSFC, No. 81371502). Sheikh Mohammed Shariful Islam acknowledges support from the National Heart Foundation of Australia and from a senior research fellowship from Deakin University. Duduzile Edith Ndwandwe acknowledges support from Cochrane South Africa, South African Medical Research Council.Tissa Wijeratne acknowledges the Department of Medicine, Faculty of Medicine, University of Rajarata, Saliyapura, Anuradhapura, Sri Lanka for their support.

Funding Funding for GBD 2017 was provided by the Bill and Melinda Gates Foundation.

Competing interests Dr. Carl Abelardo T Antonio reports personal fees from Johnson & Johnson (Philippines), Inc., outside the submitted work. Dr. Jasvinder Singh reports personal fees from Crealta/Horizon, Medisys, Fidia, UBM LLC, Medscape, WebMD, Clinical Care options, Clearview healthcare partners, Putnam associates, Spherix, the National Institutes of Health and the American College of Rheumatology, stock options in Amarin pharmaceuticals and Viking pharmaceuticals, participating in the steering committee of OMERACT, an international organization that develops measures for clinical trials and receives arm’s length funding from 12 pharmaceutical companies, including Amgen, Janssen, Novartis, Roche, UCB Group, Ardea/Astra Zeneca, Bristol Myers Squibb, Celgene, EliLilly, Horizon Pharma, Pfizer, and Centrexion. Dr. Josep Maria Haro reports personal fees from Roche and Lundbeck, and that the institute for which they work provides services to Eli Lilly and Co., outside the submitted work. Dr. Mete Saylan is an employee of Bayer AG, outside the submitted work. Dr Sheikh Mohammed Shariful Islam is funded by National Heart Foundation of Australia and supported by a senior research fellowship from Deakin University, outside the submitted work. Dr. Spencer James reports grants from Sanofi Pasteur, outside the submitted work.

Patient consent for publication Not required.

Provenance and peer review Not commissioned; externally peer reviewed.

data availability statement Data are available in a public, open access repository ( ghdx. healthdata. org). Select data are available on reasonable request. Select input data may be obtained from a third party and are not publicly available.

Open access This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given,



j.com/


ownloaded from



Original research

and indication of whether changes were made. See: https:// creativecommons. org/ licenses/ by/ 4. 0/.

OrCId idSpencer L James http:// orcid. org/ 0000- 0003- 4653- 2507

RefeRences 1 Murray CJL, Barber RM, Foreman KJ, et al. Global, regional, and national disability-

adjusted life years (DALYs) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990–2013: quantifying the epidemiological transition. The Lancet 2015;386:2145–91.

2 Haagsma JA, Graetz N, Bolliger I, et al. The global burden of injury: incidence, mortality, disability- adjusted life years and time trends from the global burden of disease study 2013. Inj Prev 2016;22:3–18.

3 Omran AR. The epidemiologic transition. A theory of the epidemiology of population change. Milbank Mem Fund Q 1971;49:509–38.

4 Kassebaum NJ, Arora M, Barber RM, et al. Global, regional, and national disability- adjusted life- years (DALYs) for 315 diseases and injuries and healthy life expectancy (HALE), 1990–2015: a systematic analysis for the global burden of disease study 2015. The Lancet 2016;388:1603–58.

5 Ahmed N, Andersson R. Unintentional injury mortality and socioeconomic development among 15-44- year- olds: in a health transition perspective. Public Health 2000;114:416–22.

6 Moniruzzaman S, Andersson R. Relationship between economic development and risk of injuries in older adults and the elderly. A global analysis of unintentional injury mortality in an epidemiologic transition perspective. Eur J Public Health 2005;15:454–8.

7 Moniruzzaman S, Andersson R. Economic development as a determinant of injury mortality - a longitudinal approach. Soc Sci Med 2008;66:1699–708.

8 Muazzam S, Nasrullah M. Macro determinants of cause- specific injury mortality in the OECD countries: an exploration of the importance of GDP and unemployment. J Community Health 2011;36:574–82.

9 Fountoulakis KN. Suicide and the economic situation in Europe: are we experiencing the development of a ’reverse stigma’? Br J Psychiatry 2016;209:273–4.

10 Vallin J. Socio- Economic determinants of mortality in industrialized countries. Popul Bull UN 1980;13:26–41.

11 Bishai D, Quresh A, James P, et al. National road casualties and economic development. Health Econ 2006;15:65–81.

12 Khan UR, Sengoelge M, Zia N, et al. Country level economic disparities in child injury mortality. Arch Dis Child 2015;100(Suppl 1):S29–S33.

13 Wang H, Naghavi M, Allen C, et al. Global, regional, and national life expectancy, all- cause mortality, and cause- specific mortality for 249 causes of death, 1980–2015: a systematic analysis for the global burden of disease study 2015. The Lancet 2016;388:1459–544.

14 GBD 2015 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990-2015: a systematic analysis for the global burden of disease study 2015. Lancet 2016;388:1545–602.

15 Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the global burden of disease study 2010. The Lancet 2012;380:2095–128.

16 GBD 2013 Mortality and Causes of Death Collaborators. Global, regional, and national age- sex specific all- cause and cause- specific mortality for 240 causes of

death, 1990-2013: a systematic analysis for the global burden of disease study 2013. Lancet 2014.

17 A Haagsma J, Van Beeck EF, Toet H. Posttraumatic stress disorder following injury: trajectories and impact on health- related quality of life. J Depress Anxiety 2013;S4.

18 Polinder S, van Beeck EF, Essink- Bot ML, et al. Functional outcome at 2.5, 5, 9, and 24 months after injury in the Netherlands. J Trauma 2007;62:133–41.

19 Ringburg AN, Polinder S, van Ierland MCP, et al. Prevalence and prognostic factors of disability after major trauma. J Trauma 2011;70:916–22.

20 Ringburg AN, Polinder S, Meulman TJ, et al. Cost- effectiveness and quality- of- life analysis of physician- staffed helicopter emergency medical services. Br J Surg 2009;96:1365–70.

21 United States National Study on the Costs and Outcomes of Trauma Care 2001-2003. Johns Hopkins Bloomberg School of Public Health, University of Washington, Westat, Inc.

22 van Loey NE, van Beeck EF, Faber BW, et al. Health- related quality of life after burns: a prospective multicenter cohort study with 18 months follow- up. J Trauma Acute Care Surg 2011.

23 Duan L, Deng X, Wang Y, et al. The National injury surveillance system in China: a six- year review. Injury 2015;46:572–9.

24 Agency for Healthcare Research and Quality (AHRQ). Medical Expenditure Panel Survey 1996-2012.

25 Mackenzie EJ, Rivara FP, Jurkovich GJ, et al. The national study on costs and outcomes of trauma. J Trauma 2007;63(6 Suppl):S54–S67.

26 Fullman N, Yearwood J, Abay SM, et al. Measuring performance on the healthcare access and quality index for 195 countries and territories and selected subnational locations: a systematic analysis from the global burden of disease study 2016. The Lancet 2018;391:2236–71.

27 Norman R, Matzopoulos R, Groenewald P, et al. The high burden of injuries in South Africa. Bull World Health Organ 2007;85:695–702.

28 Bah S. Unnoticed decline in the number of unnatural deaths in South Africa. S Afr Med J 2004;94:442–3.

29 Grigoriev P, Shkolnikov V, Andreev E, et al. Mortality in Belarus, Lithuania, and Russia: divergence in recent trends and possible explanations. Eur J Population 2010;26:245–74.

30 Chen LC, Wittgenstein F, McKeon E. The upsurge of mortality in Russia: causes and policy implications. Popul Dev Rev 1996;22:517–30.

31 Men T, Brennan P, Boffetta P, et al. Russian mortality trends for 1991-2001: analysis by cause and region. BMJ 2003;327.

32 Hartholt KA, Stevens JA, Polinder S, et al. Increase in fall- related hospitalizations in the United States, 2001-2008. J Trauma 2011;71:255–8.

33 Mahapatra P, Shibuya K, Lopez AD, et al. Civil registration systems and vital statistics: successes and missed opportunities. The Lancet 2007;370:1653–63.

34 Joubert J, Rao C, Bradshaw D, et al. Characteristics, availability and uses of vital registration and other mortality data sources in post- democracy South Africa. Glob Health Action 2012;5:19263–19.

35 Obermeyer Z, Rajaratnam JK, Park CH, et al. Measuring adult mortality using sibling survival: a new analytical method and new results for 44 countries, 1974-2006. PLoS Med 2010;7:e1000260.

36 Setel PW, Macfarlane SB, Szreter S, et al. A scandal of invisibility: making everyone count by counting everyone. The Lancet 2007;370:1569–77.

37 Obermeyer Z, Murray CJL, Gakidou E. Fifty years of violent war deaths from Vietnam to Bosnia: analysis of data from the world health survey programme. BMJ 2008;336:1482–6.



j.com/


ownloaded from

https://creativecommons.org/licenses/by/4.0/

https://creativecommons.org/licenses/by/4.0/

http://orcid.org/0000-0003-4653-2507

http://dx.doi.org/10.1016/S0140-6736(15)61340-X

http://dx.doi.org/10.1016/S0140-6736(15)61340-X


http://dx.doi.org/10.2307/3349375

http://dx.doi.org/10.1016/S0140-6736(16)31460-X

http://dx.doi.org/10.1038/sj.ph.1900646

http://dx.doi.org/10.1093/eurpub/cki014

http://dx.doi.org/10.1016/j.socscimed.2007.12.020

http://dx.doi.org/10.1007/s10900-010-9343-5

http://dx.doi.org/10.1007/s10900-010-9343-5

http://dx.doi.org/10.1192/bjp.bp.115.171785

http://dx.doi.org/10.1002/hec.1020

http://dx.doi.org/10.1136/archdischild-2013-305847

http://dx.doi.org/10.1016/S0140-6736(16)31012-1

http://dx.doi.org/10.1016/S0140-6736(16)31678-6

http://dx.doi.org/10.1016/S0140-6736(12)61728-0

http://dx.doi.org/10.4172/2167-1044.S4-002

http://dx.doi.org/10.1097/TA.0b013e31802b71c9

http://dx.doi.org/10.1097/TA.0b013e3181f6bce8

http://dx.doi.org/10.1002/bjs.6720

http://www.ncbi.nlm.nih.gov/pubmed/22439227


http://dx.doi.org/10.1016/j.injury.2014.12.013

http://dx.doi.org/10.1097/TA.0b013e31815acb09

http://dx.doi.org/10.1016/S0140-6736(18)30994-2

http://dx.doi.org/10.1016/S0140-6736(18)30994-2

http://dx.doi.org/10.2471/BLT.06.037184



http://dx.doi.org/10.1007/s10680-010-9210-1

http://dx.doi.org/10.2307/2137719

http://dx.doi.org/10.1136/bmj.327.7421.964

http://dx.doi.org/10.1097/TA.0b013e31821c36e7

http://dx.doi.org/10.1016/S0140-6736(07)61308-7

http://dx.doi.org/10.3402/gha.v5i0.19263

http://dx.doi.org/10.3402/gha.v5i0.19263

http://dx.doi.org/10.1371/journal.pmed.1000260

http://dx.doi.org/10.1371/journal.pmed.1000260

http://dx.doi.org/10.1016/S0140-6736(07)61307-5

http://dx.doi.org/10.1136/bmj.a137


Appendix 1

Summary of General Global Burden of Disease Study Methods

The Institute for Health Metrics and Evaluation with a growing collaboration of scientists

produces annual updates of the Global Burden of Disease study. Estimates span the period

from 1990 to the most recent completed year (2017). By the time of the release of GBD 2017 in

November 2018, there were 3,676 collaborators in 144 countries and 2 territories who

contributed to this global public good. Annual updates allow incorporation of new data and

method improvements to ensure that the most up-to-date information is available to policy

makers in a timely fashion to help make resource allocation decisions.

The guiding principle of GBD is to assess health loss due to mortality and disability

comprehensively, where we define disability as any departure from full health. In GBD 2017,

estimates were made for 195 countries and territories, and 579 subnational locations, for 28

years starting from 1990, for 23 age groups and both sexes. Deaths were estimated for 282

diseases and injuries, while prevalence and incidence were estimated for 355 diseases and

injuries. In order to allow meaningful comparisons between deaths and non-fatal disease

outcomes as well as between diseases, the data on deaths and prevalence are summarised in a

single indicator, the disability-adjusted life-year (DALY). DALYs are the sum of years of life lost

(YLLs) and years lived with disability (YLDs). YLLs are estimated as the multiplication of counts of

death and a standard, “ideal”, remaining life expectancy at the age of death. The standard life

expectancy is derived from the lowest observed mortality rates in any population in the world

greater than 5 million. YLDs are estimated as the product of prevalence of individual

consequences of disease (or “sequelae”) times a disability weight that quantifies the relative

severity of a sequela as a number between zero (representing “full health”) and 1 (representing

death). Disability weights have been estimated in nine population surveys and an open-access

internet survey in which respondents are asked to choose the “healthier” between random

pairs of health states that are presented with a short description of the main features.

All-cause mortality rates are estimated from vital registration data in countries with complete

coverage1. For other countries, the probabilities of death before age 5 and between ages 15

and 60 are estimated from censuses and surveys asking mothers to provide a history of children

ever born and those still alive, and surveys asking adults about siblings who are alive or have

passed away. Using model life tables, these probabilities of death are transformed into age-

specific death rates by location, year, and sex.

For cause of death estimation, GBD has collated a large database of cause of death data from

vital registrations and verbal autopsy surveys in which relatives are asked a standard set of

questions to ascertain the likely cause of death, supplemented with police and mortuary data

for injury deaths in countries with no other data2. For countries with vital registration data, the

completeness is assessed with demographic methods based on comparing recorded deaths

with population counts between two successive censuses. The cause of death information is

provided in a large number of different classification systems based on versions of the

Supplementary material Inj Prev

doi: 10.1136/injuryprev-2019-043296–15.:10 2020;Inj Prev, et al. Haagsma JA

International Classification of Diseases or bespoke classifications in some countries. All data are

mapped into the disease and injury categories of GBD. All classification systems contain codes

that are less informative because they lack a specific diagnosis (eg, unspecified cancer) or refer

to codes that cannot be underlying cause of death (eg, low back pain or senility) or are

intermediate causes (eg, heart failure or sepsis). Such deaths are redistributed to more precise

underlying causes of death. After these redistributions and corrections for under-registration,

the data are analysed in CODEm (cause of death ensemble model), a highly systematised tool

that runs many different models on the same data and chooses an ensemble of models that

best reflects all the available input data. Models are chosen with variations in the statistical

approach (“mixed effects” of spatiotemporal Gaussian Process Regression), in the unit of

analysis (rates or cause fractions), and the choice of predictive covariates. The statistical

performance of all models is tested by holding out 30% of the data and checking how well a

model covers the data that were held out. To enforce consistency from CODEm, the sum of all

cause-specific mortality rates is scaled to that of the all-cause mortality rates in each age, sex,

location, and year category.

Non-fatal estimates are based on systematic reviews of published papers and unpublished

documents, survey microdata, administrative records of health encounters, registries, and

disease surveillance systems3. Our Global Health Data Exchange (GHDx,

http://ghdx.healthdata.org/) is the largest repository of health data globally. We first set a

reference case definition and/or study method that best quantifies each disease or injury or

consequence thereof. If there is evidence of a systematic bias in data that used different case

definitions or methods compared to reference data we adjust those data points to reflect what

its value would have been if measured as the reference. This is a necessary step if one wants to

use all data pertaining to a particular quantity of interest rather than choosing a small subset of

data of the highest quality only. DisMod-MR 2.1, a Bayesian meta-regression tool, is our main

method of analyzing non-fatal data. It is designed as a geographical cascade where a first model

is run on all the world’s data, which produces an initial global fit and estimates coefficients for

predictor variables and the adjustments for alternative study characteristics. The global fit

adjusted by the values of random effects for each of seven GBD super-regions, the coefficients

on sex and country predictors, are passed down as data to a model for each super-region

together with the input data for that geography. The same steps are repeated going from

super-region to 21 region fits and then to 195 fits by country and where applicable a further

level down to subnational units. Below the global fit, all models are run separately by sex and

for six time periods: 1990, 1995, 2000, 2005, 2010, and 2017. During each fit all data on

prevalence, incidence, remission, and mortality are forced to be internally consistent. For most

diseases, the bulk of data on prevalence or incidence is at the disease level with fewer studies

providing data on the proportions of cases of disease in each of the sequelae defined for the

disease. The proportions in each sequela are pooled using DisMod-MR 2.1 or meta-analysis, or

derived from analyses of patient-level datasets. The multiplication of prevalent cases for each

disease sequela and the appropriate disability weight produces YLD estimates that do not yet

take into account comorbidity. To correct for comorbidity, these data are used in a simulation

to create hypothetical individuals in each age, sex, location, and year combination who

experience no, one, or multiple sequelae simultaneously. We assume that disability weights are



multiplicative rather than additive as this avoids assigning a combined disability weight value in

any individual to exceed 1, ie, be worse than a “year lost due to death”. This comorbidity

adjustment leads to an average scaling down of disease-specific YLDs ranging from about 2% in

young children up to 17% in oldest ages.

All our estimates of causes of death are categorical: each death is assigned to a single

underlying cause. This has the attractive property that all estimates add to 100%. For risks, we

use a different, “counterfactual” approach, ie, answering the question: “what would the burden

have been if the population had been exposed to a theoretical minimum level of exposure to a

risk”. Thus, we need to define what level of exposure to a risk factor leads to the lowest amount

of disease. We then analyse data on the prevalence of exposure to a risk and derive relative

risks for any risk-outcome pair for which we find sufficient evidence of a causal relationship.

Prevalence of exposure is estimated in DisMod-MR 2.1, using spatiotemporal Gaussian Process

Regression, or from satellite imagery in the case of ambient air pollution. Relative risk data are

pooled using meta-analysis of cohort, case-control and/or intervention studies. For each risk

and outcome pair, we evaluate the evidence and judge if the evidence falls into the categories

of “convincing” or “probable” as defined by the World Cancer Research Fund4.

From the prevalence and relative risk results, population attributable fractions are estimated

relative to the theoretical minimum risk exposure level (TMREL). When we aggregate estimates

for clusters of risks, eg, metabolic or behavioural risks, we use a multiplicative function rather

than simple addition and take into account how much of each risk is mediated through another

risk. For instance, some of the risk of high body mass index is directly onto stroke as an

outcome but much of its impact is mediated through high blood pressure, high cholesterol, or

high fasting plasma glucose, and we would not want to double count the mediated effects

when we estimate aggregates across risk factors5.

Uncertainty is propagated throughout all these calculations by creating 1,000 values for each

prevalence, death, YLL, YLD, or DALY estimate and performing aggregations across causes and

locations at the level of each of the 1,000 values for all intermediate steps in the calculation.

The lower and upper bounds of the 95% uncertainty interval are the 25th and 975th values of

the ordered 1,000 values. For all age-standardised rates, GBD uses a standard population

estimated elsewhere in the GBD analytical process.

GBD uses a composite indicator or sociodemographic development, SDI, which reflects the

geometric mean of normalised values of a location’s income per capita, the average years of

schooling in the population 15 and over, and the total fertility rate under age 25. Countries and

territories are grouped into five quintiles of high, high-middle, middle, low-middle, and low SDI

based on their 2017 values.

1 GBD 2017 Collaborators. Global, regional, and national age- and sex-specific mortality and life

expectancy for 195 countries and territories, 1950–2017: a systematic analysis for the Global

Burden of Disease Study 2017. The Lancet 2018.



2 GBD 2017 Collaborators. Global, regional, and national age-sex-specific mortality for 282

causes of death for 195 countries and territories, 1980–2017: a systematic analysis for the

Global Burden of Disease Study 2017. The Lancet 2018.

3 GBD 2017 Collaborators. Global, regional, and national incidence, prevalence, and YLDs for

328 acute and chronic diseases and injuries for 195 countries, 1990-2017: a systematic

analysis for the Global Burden of Disease Study 2017. The Lancet 2018.

4 Food, nutrition, physical activity, and the prevention of cancer: a global perspective. 2007.

http://www.aicr.org/assets/docs/pdf/reports/Second_Expert_Report.pdf.

5 GBD 2017 Collaborators. Global, regional, and national comparative risk assessment of 84

behavioural, environmental and occupational, and metabolic risks or clusters of risks for 195

countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease

Study 2017. The Lancet 2018.



Appendix 2

GATHER checklist of information that should be included in reports of global health estimates, with

description of compliance and location of information for GBD 2017.

# GATHER checklist item Description of

compliance

Reference

Objectives and funding

1 Define the indicators, populations, and time periods for

which estimates were made.

Narrative provided in

paper and

appendix describing

indicators, definitions,

and populations

Main text (Methods)

and appendix

2 List the funding sources for the work. Funding sources listed in

paper

Summary (Funding)

Data Inputs

For all data inputs from multiple sources that are synthesised as part of the study:

3 Describe how the data were identified and how the data

were accessed.

Narrative description of

data seeking methods

provided

Main text (Methods) and

appendix

4 Specify the inclusion and exclusion criteria. Identify all ad-hoc

exclusions.

Narrative about

inclusion and exclusion

criteria by data type

provided; ad hoc

exclusions in cause-

specific write-ups

Main text (Methods) and

appendix

5 Provide information on all included data sources and their

main characteristics. For each data source used, report

reference information or contact name/institution,

population represented, data collection method, year(s) of

data collection, sex and age range, diagnostic criteria or

measurement method, and sample size, as relevant.

An interactive, online

data source tool that

provides metadata for

data sources by

component, geography,

cause, risk, or

impairment has been

developed

Online data citation

tools:

http://ghdx.healthdata.o

rg/gbd-2017

6 Identify and describe any categories of input data that have

potentially important biases (e.g., based on characteristics

listed in item 5).

Summary of known

biases by cause included

in appendix

Appendix

For data inputs that contribute to the analysis but were not synthesised as part of the study:

7 Describe and give sources for any other data inputs. Included in online data

source tool

http://ghdx.healthdata.o

rg/gbd-2017

For all data inputs:

8 Provide all data inputs in a file format from which data can be

efficiently extracted (e.g., a spreadsheet as opposed to a

PDF), including all relevant meta-data listed in item 5. For any

data inputs that cannot be shared due to ethical or legal

reasons, such as third-party ownership, provide a contact

name or the name of the institution that retains the right to

the data.

Downloads of input data

available through online

tools, including data

visualisation tools and

data query tools; input

data not available in

tools will be made

available upon request

Online data

visualisation tools,

data query tools, and

the Global Health Data

Exchange

Data analysis



9 Provide a conceptual overview of the data analysis method. A

diagram may be helpful.

Flow diagrams of the

overall methodological

processes, as well as

cause-specific modelling

processes, have been

provided

Main text (Methods)

and appendix

10 Provide a detailed description of all steps of the analysis,

including mathematical formulae. This description should

cover, as relevant, data cleaning, data pre-processing, data

adjustments and weighting of data sources, and

mathematical or statistical model(s).

Flow diagrams and

corresponding

methodological write-

ups for each cause, as

well as the databases

and modelling

processes, have been

provided

Main text (Methods)

and

appendix

11 Describe how candidate models were evaluated and how the

final model(s) were selected.

Provided in the


ups

Appendix

12 Provide the results of an evaluation of model performance, if

done, as well as the results of any relevant sensitivity

analysis.

Provided in the


ups

Appendix

13 Describe methods for calculating uncertainty of the

estimates. State which sources of uncertainty were, and were

not, accounted for in the uncertainty analysis.

Appendix Appendix

14 State how analytic or statistical source code used to generate

estimates can be accessed.

Appendix http://ghdx.healthdata.o

rg/gbd-2017/code

Results and Discussion

15 Provide published estimates in a file format from which data

can be efficiently extracted.

GBD 2017 results are

available through online

data visualisation tools,


Exchange, and the

online data query tool

Main text,

and online data tools

(data visualisation tools,

data query tools, and


Exchange)

16 Report a quantitative measure of the uncertainty of the

estimates (e.g. uncertainty intervals).

Uncertainty intervals are

provided with all results

Main text, appendix, and

online data tools (data

visualisation tools, data

query tools, and the

Global Health Data

Exchange)

17 Interpret results in light of existing evidence. If updating a

previous set of estimates, describe the reasons for changes in

estimates.

Discussion of

methodological changes

between GBD rounds

provided in the narrative

of the manuscript and

appendix

Main text (Methods and

Discussion) and

appendix

18 Discuss limitations of the estimates. Include a discussion of

any modelling assumptions or data limitations that affect

interpretation of the estimates.

Discussion of limitations

provided in the narrative

of the main paper, as

well as in the


ups

in the appendix

Main text (Limitations)

and appendix



2017 age-standardised rates per

100,000

Percentage change in age-

standardised rates between

1990 and 2017


100,000



1990 and 2017

All injuries57.9

(55.9 to 59.2)

-31.6%

(-33.6% to -29.1%)

6 763

(6 412 to 7 118)

-0.9%

(-2.3% to 0.6%)

Transport injuries17.0

(16.4 to 17.4)

-28.6%

(-33.2% to -24.8%)

823

(732 to 923)

8.3%

(4.3% to 12.2%)

Road injuries15.8

(15.2 to 16.3)

-29.0%

(-33.6% to -25.0%)

692

(605 to 786)

11.3%

(6.4% to 15.8%)

Pedestrian road injuries6.2

(5.9 to 6.8)

-37.1%

(-44.2% to -29.9%)

141

(116 to 168)

3.0%

(-3.4% to 9.5%)

Cyclist road injuries0.9

(0.7 to 1.0)

7.7%

(-12.6% to 31.4%)

152

(124 to 187)

24.1%

(17.0% to 31.6%)

Motorcyclist road injuries2.9

(2.5 to 3.0)

-11.1%

(-29.5% to 1.3%)

129

(105 to 157)

30.0%

(22.2% to 37.4%)

Motor vehicle road injuries5.8

(5.4 to 6.0)

-30.0%

(-36.2% to -23.8%)

207

(173 to 247)

-4.9%

(-10.5% to 0.9%)

Other road injuries0.1

(0.1 to 0.2)

-27.9%

(-40.5% to 13.3%)

63

(49 to 80)

38.0%

(30.5% to 45.4%)

Other transport injuries1.2

(1.1 to 1.4)

-22.1%

(-32.0% to -9.9%)

131

(109 to 158)

-5.1%

(-8.8% to -1.1%)

Unintentional injuries23.8

(22.4 to 24.7)

-35.5%

(-38.1% to -31.3%)

5 400

(5 075 to 5 739)

-1.4%

(-3.2% to 0.3%)

Falls9.2

(8.5 to 9.8)

-5.9%

(-13.7% to 3.5%)

2 238

(1 990 to 2 532)

-3.7%

(-7.4% to -0.3%)

Drowning4.0

(3.8 to 4.1)

-57.4%

(-60.1% to -53.3%)

5

(4 to 5)

-27.7%

(-32.0% to -22.5%)

Fire, heat, and hot substances1.6

(1.3 to 1.7)

-46.6%

(-49.7% to -38.8%)

119

(99 to 142)

-5.4%

(-11.1% to 0.3%)

Poisonings0.9

(0.7 to 1.0)

-37.4%

(-56.9% to -20.3%)

55

(44 to 68)

6.0%

(0.5% to 11.4%)

Poisoning by carbon monoxide0.5

(0.3 to 0.5)

-41.2%

(-60.8% to -28.9%)

14

(10 to 19)

6.3%

(-2.4% to 14.8%)

Poisoning by other means0.5

(0.4 to 0.5)

-33.5%

(-53.2% to -10.5%)

41

(32 to 52)

5.9%

(-0.2% to 11.9%)

Exposure to mechanical forces1.8

(1.5 to 1.8)

-36.2%

(-47.8% to -30.5%)

944

(809 to 1 101)

-1.7%

(-5.5% to 2.0%)

Unintentional firearm injuries0.3

(0.3 to 0.3)

-46.3%

(-51.7% to -37.7%)

25

(18 to 34)

-3.6%

(-10.2% to 2.9%)

Other exposure to mechanical forces1.5

(1.2 to 1.6)

-33.7%

(-48.3% to -27.0%)

918

(788 to 1 073)

-1.7%

(-5.5% to 2.1%)

Adverse effects of medical treatment1.6

(1.4 to 1.8)

-25.4%

(-30.3% to -16.3%)

437

(376 to 502)

41.9%

(34.8% to 49.3%)

Animal contact1.1

(0.6 to 1.2)

-35.1%

(-43.8% to -19.3%)

574

(492 to 671)

-7.1%

(-9.7% to -4.3%)

Venomous animal contact0.9

(0.5 to 1.1)

-34.0%

(-44.5% to -16.6%)

275

(225 to 334)

-0.6%

(-4.4% to 3.4%)

Non-venomous animal contact0.1

(0.1 to 0.2)

-41.5%

(-55.6% to -15.2%)

299

(244 to 363)

-12.3%

(-14.8% to -9.8%)

Foreign body1.7

(1.6 to 1.8)

-34.8%

(-39.4% to -30.8%)

303

(262 to 348)

-0.4%

(-2.2% to 1.5%)

Pulmonary aspiration and foreign body in

airway

1.6

(1.5 to 1.7)

-32.3%

(-36.7% to -28.3%)

18

(15 to 22)

9.7%

(4.4% to 14.9%)

Foreign body in eyes -- --230

(191 to 273)

-1.8%

(-3.9% to 0.4%)

Foreign body in other body part0.1

(0.1 to 0.1)

-56.9%

(-64.3% to -31.6%)

55

(45 to 67)

2.8%

(0.3% to 5.4%)

Environmental heat and cold exposure0.7

(0.5 to 0.8)

-47.3%

(-52.9% to -44.1%)

115

(96 to 137)

-1.5%

(-5.3% to 1.7%)

Exposure to forces of nature0.1

(0.1 to 0.1)

-86.6%

(-88.7% to -84.0%)

9

(8 to 11)

-83.1%

(-85.0% to -80.7%)

Other unintentional injuries1.2

(1.2 to 1.3)

-39.8%

(-43.3% to -35.0%)

602

(512 to 699)

-1.2%

(-5.2% to 2.6%)

Self-harm and interpersonal violence17.1

(16.3 to 17.5)

-28.8%

(-32.2% to -24.9%)

540

(487 to 596)

-8.0%

(-10.9% to -4.4%)

Self-harm10.0

(9.4 to 10.3)

-35.4%

(-39.5% to -30.4%)

50

(42 to 59)

-16.3%

(-21.4% to -11.5%)

Self-harm by firearm0.8

(0.7 to 1.0)

-34.6%

(-38.2% to -30.4%)

1

(0 to 1)

-29.3%

(-37.0% to -21.6%)

Self-harm by other specified means9.2

(8.5 to 9.5)

-35.4%

(-39.8% to -30.2%)

50

(42 to 58)

-16.2%

(-21.2% to -11.3%)

Interpersonal violence5.2

(4.7 to 5.5)

-22.0%

(-24.9% to -18.9%)

296

(253 to 345)

-6.5%

(-10.6% to -2.4%)

Assault by firearm2.2

(1.9 to 2.4)

-0.4%

(-5.6% to 5.3%)

7

(5 to 9)

7.2%

(-0.2% to 15.0%)

Assault by sharp object1.2

(0.9 to 1.4)

-34.1%

(-38.6% to -25.8%)

55

(43 to 70)

-8.5%

(-14.3% to -2.7%)

Assault by other means1.8

(1.6 to 2.1)

-32.3%

(-39.2% to -25.2%)

234

(200 to 274)

-6.4%

(-10.5% to -1.9%)

Conflict and terrorism1.7

(1.6 to 1.9)

-3.4%

(-12.0% to 6.6%)

167

(143 to 202)

-14.1%

(-19.3% to -5.3%)

Table 1: Age-standardized mortality and incidence rates in 2017 and percentage change from 1990 to 2017 by cause of injury

Deaths (95% UI) Incidence (95% UI)

Cause




100,000



1990 and 2017


100,000



1990 and 2017

Deaths (95% UI) Incidence (95% UI)

Cause

Executions and police conflict0.2

(0.2 to 0.2)

64.6%

(48.2% to 208.1%)

26

(22 to 32)

72.6%

(50.4% to 247.9%)




100,000



1990 and 2017


100,000



1990 and 2017


100,000



1990 and 2017

All injuries2 548

(2 462 to 2 610)

-38.8%

(-40.9% to -35.9%)

719

(529 to 948)

-7.8%

(-9.3% to -6.3%)

3 267

(3 058 to 3 505)

-34.0%

(-36.4% to -31.0%)

Transport injuries800

(776 to 823)

-34.0%

(-37.9% to -30.1%)

167

(120 to 223)

-2.9%

(-4.8% to -0.9%)

968

(914 to 1 024)

-30.1%

(-34.1% to -26.4%)

Road injuries745

(718 to 767)

-34.4%

(-38.5% to -30.4%)

126

(90 to 169)

2.2%

(0.3% to 4.0%)

871

(828 to 917)

-30.8%

(-35.0% to -26.9%)

Pedestrian road injuries270

(254 to 301)

-44.8%

(-51.3% to -37.2%)

34

(24 to 46)

-5.4%

(-8.0% to -2.9%)

304

(284 to 332)

-42.2%

(-48.6% to -35.0%)

Cyclist road injuries36

(32 to 41)

-1.8%

(-23.2% to 22.7%)

21

(15 to 29)

12.0%

(8.7% to 15.1%)

57

(49 to 66)

2.8%

(-12.4% to 18.7%)

Motorcyclist road injuries146

(128 to 155)

-16.1%

(-32.1% to -4.4%)

32

(22 to 43)

17.9%

(15.1% to 20.8%)

178

(157 to 192)

-11.5%

(-26.9% to -1.1%)

Motor vehicle road injuries285

(268 to 300)

-32.8%

(-37.7% to -24.7%)

32

(23 to 43)

-11.9%

(-13.4% to -10.4%)

317

(298 to 335)

-31.1%

(-35.8% to -23.6%)

Other road injuries7

(6 to 8)

-32.7%

(-45.6% to 10.7%)

8

(5 to 11)

31.2%

(27.7% to 34.2%)

14

(12 to 17)

-9.0%

(-23.0% to 19.3%)

Other transport injuries55

(51 to 65)

-27.6%

(-36.9% to -14.8%)

41

(29 to 55)

-15.7%

(-17.7% to -13.4%)

96

(83 to 112)

-23.0%

(-29.4% to -14.9%)

Unintentional injuries929

(866 to 969)

-48.4%

(-51.4% to -43.3%)

460

(333 to 618)

-9.9%

(-11.5% to -8.3%)

1 389

(1 241 to 1 560)

-39.9%

(-43.5% to -35.1%)

Falls217

(196 to 229)

-18.5%

(-31.7% to -6.2%)

243

(173 to 330)

-9.3%

(-10.7% to -7.9%)

459

(387 to 547)

-13.9%

(-21.3% to -8.0%)

Drowning228

(217 to 240)

-63.9%

(-66.6% to -59.8%)

2

(1 to 2)

-35.7%

(-39.0% to -32.3%)

230

(219 to 241)

-63.8%

(-66.5% to -59.7%)

Fire, heat, and hot substances71

(58 to 79)

-50.8%

(-55.8% to -39.3%)

40

(28 to 55)

-24.4%

(-29.4% to -19.3%)

111

(93 to 129)

-43.7%

(-49.3% to -34.1%)

Poisonings44

(33 to 49)

-44.4%

(-61.5% to -26.3%)

6

(4 to 8)

-8.1%

(-9.8% to -6.1%)

50

(39 to 56)

-41.6%

(-57.9% to -24.5%)

Poisoning by carbon monoxide19

(14 to 21)

-48.0%

(-65.2% to -36.0%)

1

(1 to 1)

-0.3%

(-3.4% to 2.9%)

20

(15 to 22)

-46.8%

(-63.8% to -35.1%)

Poisoning by other means25

(19 to 28)

-41.3%

(-58.5% to -16.5%)

5

(3 to 7)

-9.4%

(-11.6% to -7.2%)

30

(24 to 34)

-37.7%

(-53.7% to -15.3%)

Exposure to mechanical forces84

(72 to 88)

-40.9%

(-51.5% to -35.5%)

62

(43 to 87)

-9.8%

(-11.5% to -8.1%)

146

(124 to 173)

-30.8%

(-38.4% to -25.8%)

Unintentional firearm injuries14

(13 to 17)

-48.5%

(-53.7% to -39.0%)

4

(3 to 5)

-9.3%

(-10.7% to -8.0%)

18

(17 to 21)

-43.1%

(-48.2% to -34.5%)

Other exposure to mechanical forces70

(58 to 74)

-39.1%

(-52.5% to -32.4%)

58

(40 to 82)

-9.8%

(-11.6% to -8.1%)

128

(107 to 152)

-28.5%

(-38.0% to -23.1%)

Adverse effects of medical treatment58

(48 to 71)

-28.7%

(-37.2% to -14.6%)

4

(3 to 7)

41.8%

(34.7% to 49.2%)

63

(52 to 75)

-26.1%

(-35.1% to -11.7%)

Animal contact52

(29 to 62)

-39.0%

(-50.2% to -20.5%)

14

(9 to 19)

-16.1%

(-17.8% to -14.7%)

66

(42 to 77)

-35.3%

(-45.5% to -19.8%)

Venomous animal contact46

(23 to 55)

-38.0%

(-51.0% to -18.1%)

9

(6 to 13)

-11.3%

(-13.0% to -9.5%)

55

(33 to 65)

-34.7%

(-46.0% to -17.4%)

Non-venomous animal contact7

(5 to 10)

-44.7%

(-61.9% to -13.7%)

5

(3 to 7)

-24.4%

(-26.8% to -22.4%)

11

(8 to 15)

-38.1%

(-54.7% to -18.5%)

Foreign body83

(78 to 89)

-46.7%

(-51.4% to -41.9%)

12

(9 to 16)

-10.0%

(-12.6% to -7.2%)

95

(88 to 102)

-43.8%

(-48.3% to -39.0%)

Pulmonary aspiration and foreign body in

airway

78

(73 to 84)

-44.2%

(-48.7% to -39.0%)

2

(1 to 3)

-9.1%

(-14.6% to -3.1%)

80

(75 to 85)

-43.6%

(-48.2% to -38.5%)

Foreign body in eyes -- --3

(1 to 4)

-7.0%

(-10.6% to -4.5%)

3

(1 to 4)

-7.0%

(-10.6% to -4.5%)

Foreign body in other body part5

(4 to 7)

-68.6%

(-75.3% to -46.8%)

8

(5 to 10)

-11.2%

(-13.7% to -8.5%)

13

(10 to 15)

-49.0%

(-57.2% to -29.6%)

Environmental heat and cold exposure24

(16 to 27)

-51.8%

(-55.4% to -47.9%)

19

(14 to 26)

-12.5%

(-15.3% to -9.7%)

43

(32 to 51)

-39.7%

(-43.7% to -35.7%)

Exposure to forces of nature6

(6 to 7)

-88.1%

(-90.1% to -85.6%)

9

(7 to 12)

176.3%

(155.6% to 195.3%)

16

(13 to 19)

-72.6%

(-78.1% to -65.5%)

Other unintentional injuries61

(59 to 64)

-44.4%

(-48.4% to -39.0%)

49

(34 to 69)

-9.0%

(-10.6% to -7.5%)

110

(95 to 130)

-32.8%

(-37.4% to -27.9%)

Self-harm and interpersonal violence819

(782 to 843)

-29.1%

(-32.3% to -25.5%)

91

(71 to 114)

-5.5%

(-8.1% to -2.6%)

910

(872 to 944)

-27.3%

(-30.3% to -23.7%)

Self-harm424

(397 to 438)

-37.7%

(-41.8% to -32.9%)

5

(4 to 7)

-26.2%

(-28.6% to -23.7%)

429

(402 to 443)

-37.6%

(-41.7% to -32.8%)

Self-harm by firearm33

(28 to 42)

-35.0%

(-38.6% to -29.8%)

0

(0 to 0)

-31.5%

(-33.8% to -28.9%)

34

(28 to 42)

-35.0%

(-38.6% to -29.8%)

Self-harm by other specified means390

(364 to 405)

-37.9%

(-42.3% to -32.8%)

5

(4 to 7)

-26.1%

(-28.6% to -23.6%)

395

(369 to 410)

-37.8%

(-42.1% to -32.8%)

Interpersonal violence277

(248 to 294)

-22.6%

(-25.7% to -19.0%)

57

(44 to 73)

-8.1%

(-9.8% to -6.3%)

334

(305 to 361)

-20.5%

(-23.3% to -17.2%)

Assault by firearm123

(104 to 132)

1.0%

(-5.0% to 6.5%)

1

(1 to 2)

1.7%

(0.2% to 3.0%)

124

(106 to 134)

1.0%

(-5.0% to 6.4%)

Assault by sharp object59

(48 to 72)

-35.4%

(-40.0% to -26.6%)

6

(4 to 8)

-15.0%

(-17.1% to -12.7%)

65

(53 to 79)

-34.0%

(-38.5% to -25.8%)

Sexual violence -- --27

(18 to 40)

-1.4%

(-3.1% to 0.3%)

27

(18 to 40)

-1.4%

(-3.1% to 0.3%)

Assault by other means95

(83 to 112)

-34.4%

(-40.2% to -27.5%)

23

(16 to 30)

-14.0%

(-16.1% to -12.0%)

117

(106 to 134)

-31.2%

(-36.4% to -24.9%)

Conflict and terrorism107

(98 to 119)

-3.2%

(-12.1% to 7.6%)

27

(18 to 40)

4.8%

(-3.4% to 12.7%)

134

(119 to 152)

-1.7%

(-9.0% to 7.0%)

Executions and police conflict11

(11 to 12)

69.2%

(51.4% to 222.2%)

2

(1 to 2)

40.1%

(26.1% to 69.0%)

13

(13 to 14)

64.7%

(48.8% to 191.5%)

Table 2: Age-standardized YLL, YLD, and DALY rates in 2017 and percentage change from 1990 to 2017 by cause of injury

YLLs (95% UI) YLDs (95% UI) DALYs (95% UI)

Cause



2017 age-standardised rates

per 100,000



1990 and 2017


per 100,000



1990 and 2017


per 100,000



1990 and 2017


per 100,000



1990 and 2017

High SDI Andorra23

(20 to 26)

-35.2%

(-46.5% to -21.8%)

854

(738 to 980)

-43.5%

(-54.6% to -30.1%)

911

(650 to 1 243)

2.5%

(1.1% to 3.9%)

1 765

(1 473 to 2 109)

-26.4%

(-35.3% to -16.5%)

High SDI Australia30

(27 to 33)

-32.1%

(-38.7% to -25.5%)

1 156

(1 033 to 1 284)

-43.2%

(-49.4% to -37.1%)

1 514

(1 077 to 2 057)

11.6%

(10.0% to 13.2%)

2 670

(2 207 to 3 237)

-21.3%

(-27.8% to -15.6%)

High SDI Austria29

(27 to 31)

-50.8%

(-53.9% to -47.8%)

1 009

(946 to 1 079)

-57.8%

(-60.5% to -54.9%)

892

(633 to 1 220)

-9.0%

(-10.3% to -7.7%)

1 901

(1 621 to 2 219)

-43.6%

(-47.5% to -39.8%)

High SDI Belgium38

(36 to 40)

-33.9%

(-37.6% to -30.0%)

1 348

(1 267 to 1 431)

-44.6%

(-48.2% to -41.2%)

968

(688 to 1 322)

6.6%

(4.9% to 8.3%)

2 316

(2 028 to 2 695)

-30.7%

(-34.7% to -26.3%)

High SDI Brunei44

(41 to 47)

-36.0%

(-41.7% to -29.6%)

1 839

(1 704 to 2 003)

-36.6%

(-42.1% to -30.0%)

1 031

(725 to 1 411)

-11.9%

(-15.0% to -9.0%)

2 871

(2 523 to 3 283)

-29.5%

(-34.2% to -24.4%)

High SDI Canada33

(31 to 35)

-29.3%

(-33.1% to -25.2%)

1 301

(1 225 to 1 375)

-38.3%

(-41.9% to -34.4%)

833

(593 to 1 132)

0.2%

(-1.1% to 1.5%)

2 134

(1 890 to 2 437)

-27.4%

(-31.2% to -24.0%)

High SDI Croatia38

(36 to 41)

-49.4%

(-52.2% to -46.3%)

1 223

(1 149 to 1 295)

-59.4%

(-61.9% to -56.8%)

1 389

(996 to 1 872)

-10.7%

(-12.7% to -8.2%)

2 612

(2 219 to 3 098)

-42.8%

(-46.3% to -39.3%)

High SDI Cyprus28

(25 to 31)

-48.6%

(-53.9% to -42.9%)

1 061

(963 to 1 170)

-49.6%

(-55.1% to -44.3%)

893

(633 to 1 218)

-8.8%

(-10.9% to -6.7%)

1 954

(1 673 to 2 298)

-36.7%

(-41.7% to -32.0%)

High SDI Czech Republic35

(33 to 37)

-55.9%

(-58.5% to -53.1%)

1 280

(1 203 to 1 361)

-54.6%

(-57.3% to -51.4%)

2 009

(1 433 to 2 748)

3.3%

(1.3% to 5.4%)

3 289

(2 702 to 4 037)

-30.9%

(-36.0% to -25.9%)

High SDI Denmark25

(23 to 26)

-60.3%

(-62.7% to -57.6%)

799

(745 to 855)

-64.2%

(-66.7% to -61.6%)

866

(617 to 1 183)

-1.8%

(-3.5% to -0.2%)

1 665

(1 404 to 1 995)

-46.5%

(-51.2% to -42.1%)

High SDI Estonia42

(37 to 48)

-65.9%

(-70.3% to -61.1%)

1 771

(1 548 to 2 018)

-69.8%

(-73.6% to -65.5%)

1 400

(996 to 1 904)

-18.3%

(-20.0% to -16.7%)

3 171

(2 703 to 3 730)

-58.2%

(-62.4% to -53.7%)

High SDI Finland35

(33 to 38)

-52.4%

(-55.3% to -49.0%)

1 282

(1 203 to 1 380)

-58.8%

(-61.5% to -55.7%)

1 014

(720 to 1 384)

4.4%

(2.9% to 5.9%)

2 296

(1 993 to 2 681)

-43.8%

(-47.9% to -39.3%)

High SDI France35

(33 to 37)

-51.9%

(-54.8% to -48.7%)

1 202

(1 128 to 1 277)

-55.7%

(-58.6% to -52.8%)

919

(654 to 1 258)

-5.1%

(-6.5% to -3.8%)

2 121

(1 842 to 2 452)

-42.4%

(-46.5% to -38.6%)

High SDI Georgia52

(49 to 54)

-12.8%

(-18.1% to -7.3%)

2 300

(2 187 to 2 407)

-21.0%

(-26.3% to -15.6%)

1 034

(741 to 1 398)

-8.1%

(-10.4% to -5.7%)

3 334

(3 024 to 3 705)

-17.4%

(-21.6% to -13.2%)

High SDI Germany26

(24 to 29)

-43.1%

(-48.3% to -37.2%)

919

(826 to 1 022)

-52.4%

(-57.1% to -47.1%)

869

(615 to 1 192)

2.4%

(0.9% to 3.8%)

1 787

(1 521 to 2 109)

-35.6%

(-40.7% to -30.5%)

High SDI Greece24

(23 to 25)

-39.7%

(-43.1% to -36.2%)

998

(938 to 1 062)

-43.2%

(-46.9% to -39.5%)

861

(611 to 1 177)

-2.2%

(-3.5% to -0.8%)

1 859

(1 594 to 2 175)

-29.6%

(-33.7% to -25.4%)

High SDI Iceland26

(25 to 27)

-40.9%

(-44.0% to -37.3%)

949

(906 to 997)

-48.3%

(-51.2% to -45.1%)

881

(627 to 1 204)

3.0%

(1.6% to 4.5%)

1 830

(1 565 to 2 160)

-32.0%

(-36.5% to -27.5%)

High SDI Ireland21

(19 to 22)

-48.6%

(-52.3% to -44.8%)

814

(754 to 876)

-52.6%

(-56.1% to -49.0%)

860

(610 to 1 169)

4.6%

(3.1% to 6.4%)

1 674

(1 417 to 1 979)

-34.1%

(-39.1% to -29.6%)

High SDI Italy20

(19 to 21)

-52.5%

(-55.5% to -49.6%)

686

(640 to 733)

-58.3%

(-61.0% to -55.4%)

772

(549 to 1 052)

-11.7%

(-13.0% to -10.3%)

1 458

(1 237 to 1 739)

-42.1%

(-46.1% to -38.1%)

High SDI Japan30

(29 to 31)

-25.5%

(-27.6% to -22.8%)

1 136

(1 101 to 1 178)

-30.4%

(-32.4% to -27.7%)

1 052

(754 to 1 430)

16.0%

(14.5% to 17.7%)

2 188

(1 894 to 2 550)

-13.8%

(-17.4% to -10.2%)

High SDI Latvia61

(54 to 68)

-52.1%

(-57.5% to -46.4%)

2 572

(2 282 to 2 884)

-56.7%

(-61.6% to -51.5%)

1 402

(996 to 1 910)

-21.5%

(-23.4% to -19.6%)

3 974

(3 443 to 4 553)

-48.6%

(-53.3% to -44.1%)

High SDI Lithuania75

(71 to 80)

-33.1%

(-37.3% to -28.5%)

3 140

(2 945 to 3 369)

-40.2%

(-44.2% to -35.9%)

1 482

(1 053 to 2 016)

-9.3%

(-11.2% to -7.4%)

4 622

(4 143 to 5 156)

-32.9%

(-36.4% to -29.1%)

High SDI Luxembourg28

(26 to 31)

-51.4%

(-56.1% to -46.6%)

938

(847 to 1 034)

-62.0%

(-65.8% to -57.9%)

931

(662 to 1 264)

-10.3%

(-11.6% to -8.7%)

1 869

(1 587 to 2 203)

-46.6%

(-51.0% to -42.2%)

High SDI Malta21

(20 to 22)

-29.1%

(-33.1% to -25.0%)

739

(701 to 780)

-32.5%

(-36.7% to -27.5%)

919

(654 to 1 253)

8.4%

(6.7% to 10.3%)

1 657

(1 384 to 1 993)

-14.6%

(-19.3% to -10.4%)

High SDI Netherlands26

(25 to 28)

-25.8%

(-30.0% to -21.3%)

803

(759 to 855)

-42.1%

(-45.3% to -38.1%)

727

(516 to 991)

1.2%

(-0.4% to 3.0%)

1 530

(1 306 to 1 796)

-27.3%

(-31.4% to -23.5%)

High SDI New Zealand32

(31 to 34)

-39.7%

(-42.8% to -36.2%)

1 410

(1 339 to 1 491)

-47.0%

(-49.8% to -44.1%)

1 860

(1 321 to 2 527)

9.9%

(7.7% to 12.3%)

3 270

(2 724 to 3 968)

-24.9%

(-29.9% to -20.3%)

High SDI Norway29

(28 to 29)

-46.7%

(-48.5% to -44.7%)

970

(942 to 1 007)

-55.9%

(-57.4% to -54.1%)

1 031

(746 to 1 383)

2.0%

(0.8% to 3.3%)

2 001

(1 720 to 2 353)

-37.6%

(-41.6% to -33.7%)

High SDI Poland41

(39 to 44)

-40.9%

(-44.2% to -37.2%)

1 657

(1 563 to 1 760)

-46.0%

(-49.3% to -42.5%)

1 708

(1 217 to 2 332)

-8.3%

(-11.4% to -5.3%)

3 365

(2 865 to 3 992)

-31.8%

(-35.8% to -28.0%)

High SDI Singapore15

(14 to 16)

-56.8%

(-59.7% to -54.0%)

611

(571 to 650)

-59.1%

(-61.8% to -56.5%)

966

(683 to 1 323)

2.9%

(0.8% to 4.8%)

1 577

(1 296 to 1 925)

-35.2%

(-40.7% to -30.0%)

High SDI Slovakia38

(35 to 41)

-48.8%

(-52.3% to -44.2%)

1 468

(1 372 to 1 583)

-51.6%

(-54.9% to -47.3%)

1 716

(1 227 to 2 349)

-10.5%

(-12.4% to -8.6%)

3 184

(2 671 to 3 831)

-35.7%

(-39.4% to -31.4%)

High SDI Slovenia40

(37 to 42)

-52.3%

(-55.5% to -48.9%)

1 318

(1 222 to 1 409)

-59.5%

(-62.6% to -56.5%)

2 034

(1 459 to 2 781)

1.4%

(-0.0% to 2.8%)

3 353

(2 772 to 4 094)

-36.3%

(-41.6% to -31.6%)

High SDI South Korea43

(39 to 46)

-48.1%

(-52.1% to -44.1%)

1 463

(1 357 to 1 569)

-61.1%

(-64.3% to -57.8%)

912

(645 to 1 249)

-24.0%

(-26.7% to -20.9%)

2 374

(2 098 to 2 719)

-52.1%

(-55.4% to -48.6%)

High SDI Spain19

(18 to 20)

-57.5%

(-60.0% to -54.7%)

668

(625 to 709)

-67.2%

(-69.3% to -65.0%)

836

(595 to 1 140)

-3.6%

(-5.0% to -2.1%)

1 504

(1 253 to 1 806)

-48.2%

(-52.8% to -43.8%)

High SDI Sweden27

(26 to 29)

-38.0%

(-41.1% to -34.9%)

950

(899 to 1 005)

-47.0%

(-50.0% to -43.9%)

958

(693 to 1 292)

7.3%

(5.7% to 8.9%)

1 907

(1 633 to 2 237)

-28.9%

(-33.3% to -24.7%)

High SDI Switzerland25

(24 to 27)

-58.2%

(-61.0% to -55.2%)

804

(751 to 860)

-66.4%

(-68.7% to -63.8%)

879

(629 to 1 199)

-21.1%

(-22.5% to -19.7%)

1 683

(1 421 to 2 015)

-52.0%

(-55.7% to -48.5%)

High SDI Taiwan (Province of China)40

(39 to 43)

-54.2%

(-56.5% to -51.8%)

1 580

(1 497 to 1 667)

-59.8%

(-61.9% to -57.5%)

508

(373 to 669)

-1.9%

(-5.0% to 1.1%)

2 088

(1 936 to 2 260)

-53.1%

(-55.6% to -50.4%)

High SDI USA46

(45 to 47)

-17.2%

(-19.6% to -14.9%)

2 024

(1 963 to 2 081)

-26.0%

(-28.3% to -23.8%)

811

(589 to 1 093)

-31.2%

(-33.2% to -29.5%)

2 836

(2 605 to 3 129)

-27.6%

(-29.4% to -25.7%)

High SDI United Kingdom21

(21 to 21)

-33.0%

(-34.2% to -31.9%)

774

(762 to 787)

-44.3%

(-45.4% to -43.3%)

850

(603 to 1 160)

9.8%

(8.9% to 10.7%)

1 623

(1 377 to 1 938)

-25.0%

(-29.1% to -20.8%)

High-middle SDI American Samoa52

(48 to 56)

-17.3%

(-25.7% to -8.7%)

2 201

(2 029 to 2 403)

-19.5%

(-27.8% to -10.6%)

553

(413 to 720)

18.0%

(13.8% to 22.7%)

2 755

(2 516 to 3 007)

-14.0%

(-21.6% to -6.1%)

High-middle SDI Antigua and Barbuda37

(34 to 40)

-20.2%

(-27.1% to -13.2%)

1 698

(1 556 to 1 847)

-15.8%

(-24.0% to -7.3%)

406

(294 to 545)

6.0%

(1.7% to 10.5%)

2 104

(1 929 to 2 315)

-12.3%

(-19.3% to -5.1%)

High-middle SDI Argentina47

(43 to 53)

-19.0%

(-27.2% to -9.4%)

2 140

(1 917 to 2 391)

-19.2%

(-27.6% to -9.2%)

881

(628 to 1 194)

-1.5%

(-5.8% to 2.8%)

3 021

(2 676 to 3 424)

-14.7%

(-21.4% to -6.9%)

High-middle SDI Armenia32

(30 to 33)

-56.1%

(-58.3% to -53.8%)

1 311

(1 261 to 1 363)

-62.0%

(-64.3% to -59.8%)

1 027

(742 to 1 374)

-30.1%

(-33.9% to -26.3%)

2 338

(2 049 to 2 692)

-52.5%

(-55.3% to -49.6%)

High-middle SDI Azerbaijan31

(28 to 35)

-46.7%

(-53.0% to -37.9%)

1 471

(1 311 to 1 686)

-51.6%

(-57.7% to -44.0%)

1 049

(756 to 1 410)

-9.0%

(-13.2% to -4.4%)

2 520

(2 177 to 2 923)

-39.9%

(-45.2% to -33.6%)

High-middle SDI Bahrain23

(21 to 26)

-50.6%

(-56.2% to -44.3%)

1 014

(922 to 1 126)

-50.4%

(-55.5% to -44.4%)

575

(411 to 773)

-13.8%

(-19.1% to -8.3%)

1 589

(1 406 to 1 824)

-41.4%

(-46.2% to -35.9%)

High-middle SDI Barbados37

(34 to 41)

-19.7%

(-26.4% to -12.6%)

1 649

(1 512 to 1 795)

-20.9%

(-28.0% to -13.4%)

376

(273 to 504)

14.1%

(9.5% to 18.5%)

2 025

(1 851 to 2 220)

-16.1%

(-22.3% to -9.6%)

High-middle SDI Belarus65

(61 to 69)

-29.2%

(-33.6% to -24.0%)

2 753

(2 570 to 2 949)

-38.7%

(-43.0% to -33.8%)

1 401

(999 to 1 912)

-6.4%

(-9.1% to -3.8%)

4 154

(3 705 to 4 682)

-30.7%

(-34.5% to -26.4%)

High-middle SDI Bermuda24

(22 to 27)

-50.5%

(-55.1% to -45.5%)

1 004

(899 to 1 116)

-54.4%

(-59.6% to -48.5%)

428

(310 to 568)

6.7%

(0.9% to 11.6%)

1 432

(1 267 to 1 607)

-45.0%

(-50.2% to -39.5%)

High-middle SDI Bosnia and Herzegovina25

(23 to 29)

-35.9%

(-44.2% to -18.9%)

1 026

(951 to 1 116)

-38.6%

(-45.7% to -25.2%)

1 780

(1 293 to 2 393)

20.5%

(13.7% to 30.0%)

2 806

(2 313 to 3 417)

-10.8%

(-18.7% to -0.7%)

High-middle SDI Bulgaria34

(32 to 37)

-41.4%

(-44.9% to -37.8%)

1 488

(1 393 to 1 595)

-45.3%

(-48.8% to -41.6%)

1 573

(1 111 to 2 152)

-9.6%

(-11.8% to -7.4%)

3 061

(2 595 to 3 640)

-31.3%

(-35.3% to -27.7%)

High-middle SDI Chile40

(36 to 45)

-48.8%

(-54.4% to -42.4%)

1 688

(1 494 to 1 903)

-53.3%

(-58.7% to -47.0%)

740

(524 to 1 006)

-16.3%

(-20.2% to -12.0%)

2 428

(2 125 to 2 778)

-46.1%

(-51.1% to -40.9%)

High-middle SDI China46

(43 to 48)

-44.3%

(-48.9% to -41.1%)

1 893

(1 798 to 1 963)

-55.8%

(-58.8% to -53.4%)

580

(430 to 758)

21.2%

(14.9% to 27.9%)

2 473

(2 298 to 2 679)

-48.1%

(-51.8% to -44.6%)

High-middle SDI Georgia52

(49 to 54)

-12.8%

(-18.1% to -7.3%)

2 300

(2 187 to 2 407)

-21.0%

(-26.3% to -15.6%)

1 034

(741 to 1 398)

-8.1%

(-10.4% to -5.7%)

3 334

(3 024 to 3 705)

-17.4%

(-21.6% to -13.2%)

High-middle SDI Greenland104

(98 to 112)

-51.2%

(-55.0% to -46.6%)

4 474

(4 153 to 4 799)

-58.4%

(-61.9% to -54.5%)

948

(682 to 1 281)

-39.8%

(-41.8% to -38.0%)

5 422

(5 006 to 5 864)

-56.0%

(-59.2% to -52.6%)

High-middle SDI Guam57

(53 to 62)

1.5%

(-8.5% to 12.5%)

2 655

(2 441 to 2 881)

4.0%

(-6.0% to 15.2%)

548

(400 to 726)

30.5%

(26.7% to 34.1%)

3 203

(2 938 to 3 479)

7.8%

(-1.2% to 17.7%)

High-middle SDI Hungary39

(37 to 41)

-63.0%

(-65.0% to -60.9%)

1 334

(1 260 to 1 414)

-64.4%

(-66.4% to -62.2%)

1 636

(1 166 to 2 240)

-15.0%

(-17.5% to -12.5%)

2 970

(2 484 to 3 567)

-47.6%

(-51.7% to -43.5%)

High-middle SDI Iran53

(52 to 54)

-70.1%

(-71.9% to -68.7%)

2 512

(2 464 to 2 611)

-72.5%

(-73.4% to -71.3%)

732

(543 to 962)

-36.4%

(-42.1% to -30.7%)

3 244

(3 051 to 3 499)

-68.4%

(-69.9% to -66.8%)

High-middle SDI Israel23

(21 to 24)

-44.7%

(-48.3% to -37.4%)

831

(782 to 887)

-49.3%

(-52.7% to -44.3%)

882

(630 to 1 194)

6.9%

(4.2% to 11.5%)

1 713

(1 454 to 2 022)

-30.5%

(-35.2% to -25.3%)

High-middle SDI Kazakhstan74

(70 to 79)

-26.5%

(-30.4% to -22.2%)

3 501

(3 295 to 3 757)

-31.4%

(-35.4% to -26.3%)

1 074

(766 to 1 453)

-13.8%

(-17.4% to -10.5%)

4 574

(4 192 to 5 017)

-27.9%

(-31.6% to -23.8%)

High-middle SDI Kuwait25

(23 to 26)

-78.7%

(-80.1% to -77.2%)

1 073

(1 004 to 1 146)

-82.5%

(-83.6% to -81.3%)

626

(449 to 843)

-16.7%

(-20.7% to -13.0%)

1 700

(1 517 to 1 939)

-75.3%

(-77.4% to -72.7%)

High-middle SDI Lebanon33

(29 to 37)

-67.0%

(-71.8% to -62.0%)

1 447

(1 282 to 1 603)

-71.2%

(-75.1% to -67.6%)

803

(585 to 1 061)

-39.5%

(-47.3% to -31.6%)

2 249

(1 985 to 2 553)

-64.6%

(-68.2% to -61.2%)

High-middle SDI Libya85

(71 to 99)

24.8%

(3.1% to 48.4%)

4 132

(3 534 to 4 734)

28.3%

(7.4% to 51.4%)

855

(629 to 1 127)

16.3%

(2.8% to 40.1%)

4 987

(4 343 to 5 644)

26.1%

(8.9% to 43.7%)

High-middle SDI Macedonia24

(22 to 26)

-38.9%

(-44.2% to -33.7%)

997

(932 to 1 070)

-43.4%

(-48.3% to -38.6%)

1 529

(1 088 to 2 101)

0.9%

(-2.7% to 4.4%)

2 526

(2 065 to 3 095)

-22.9%

(-28.4% to -17.9%)

High-middle SDI Malaysia52

(47 to 57)

-25.8%

(-36.6% to -16.9%)

2 015

(1 844 to 2 215)

-27.8%

(-36.6% to -19.4%)

394

(284 to 524)

4.1%

(-1.9% to 10.4%)

2 409

(2 200 to 2 644)

-24.0%

(-32.2% to -16.5%)

High-middle SDI Mauritius38

(36 to 41)

-23.4%

(-29.1% to -17.8%)

1 746

(1 617 to 1 868)

-23.1%

(-28.9% to -17.3%)

316

(230 to 417)

17.7%

(12.9% to 22.4%)

2 062

(1 915 to 2 214)

-18.8%

(-24.1% to -13.4%)

Table 3: Age-standardised mortality, YLL, YLD, and DALY rates in 2017 and percentage change from 1990 to 2017 for all injuries by country

SDI Quintile Location

Deaths (95% UI) YLLs (95% UI) YLDs (95% UI) DALYs (95% UI)




per 100,000



1990 and 2017


per 100,000



1990 and 2017


per 100,000



1990 and 2017


per 100,000



1990 and 2017



High-middle SDI Montenegro33

(30 to 36)

-31.3%

(-37.5% to -24.4%)

1 341

(1 221 to 1 468)

-40.0%

(-45.7% to -33.6%)

1 593

(1 133 to 2 174)

2.0%

(0.2% to 3.7%)

2 934

(2 443 to 3 547)

-22.7%

(-28.0% to -17.9%)

High-middle SDI Northern Mariana Islands51

(45 to 56)

-20.3%

(-32.3% to -6.1%)

2 110

(1 852 to 2 355)

-19.2%

(-32.7% to -2.9%)

493

(359 to 657)

4.9%

(1.6% to 8.4%)

2 603

(2 335 to 2 900)

-15.6%

(-26.9% to -1.7%)

High-middle SDI Oman61

(50 to 72)

-50.4%

(-60.5% to -37.1%)

2 601

(2 165 to 3 065)

-54.3%

(-63.5% to -41.9%)

611

(431 to 830)

-20.3%

(-25.2% to -15.3%)

3 212

(2 733 to 3 736)

-50.3%

(-58.8% to -38.8%)

High-middle SDI Portugal25

(24 to 27)

-64.0%

(-66.4% to -61.6%)

930

(863 to 997)

-71.7%

(-73.8% to -69.5%)

729

(517 to 998)

-27.8%

(-29.7% to -26.0%)

1 659

(1 436 to 1 910)

-61.4%

(-64.6% to -58.4%)

High-middle SDI Puerto Rico105

(102 to 109)

50.1%

(44.8% to 55.7%)

5 175

(5 011 to 5 336)

57.8%

(52.0% to 63.9%)

478

(341 to 640)

31.4%

(27.3% to 35.9%)

5 653

(5 442 to 5 868)

55.1%

(49.9% to 60.6%)

High-middle SDI Qatar43

(35 to 51)

-37.4%

(-50.1% to -20.5%)

1 785

(1 469 to 2 136)

-42.1%

(-53.9% to -26.8%)

670

(475 to 909)

-16.9%

(-20.5% to -13.0%)

2 454

(2 056 to 2 883)

-36.9%

(-46.9% to -24.7%)

High-middle SDI Romania40

(38 to 42)

-43.7%

(-46.7% to -40.7%)

1 721

(1 630 to 1 823)

-50.9%

(-53.6% to -48.1%)

1 596

(1 131 to 2 193)

-23.0%

(-26.1% to -20.0%)

3 317

(2 838 to 3 909)

-40.6%

(-43.8% to -37.5%)

High-middle SDI Russian Federation87

(86 to 89)

-24.0%

(-25.1% to -23.1%)

3 987

(3 925 to 4 056)

-28.2%

(-29.1% to -27.4%)

1 444

(1 026 to 1 962)

-9.2%

(-11.9% to -6.7%)

5 431

(5 008 to 5 946)

-24.0%

(-25.6% to -22.3%)

High-middle SDI Saudi Arabia72

(61 to 83)

-30.3%

(-46.7% to -12.7%)

2 989

(2 516 to 3 461)

-35.4%

(-49.0% to -19.8%)

675

(481 to 915)

-24.9%

(-29.6% to -20.0%)

3 664

(3 136 to 4 208)

-33.7%

(-45.5% to -20.8%)

High-middle SDI Serbia33

(31 to 35)

-43.1%

(-47.2% to -38.7%)

1 213

(1 132 to 1 309)

-54.5%

(-58.0% to -50.2%)

1 566

(1 118 to 2 133)

4.5%

(1.8% to 7.6%)

2 779

(2 324 to 3 356)

-33.2%

(-38.4% to -27.6%)

High-middle SDI The Bahamas71

(65 to 77)

-6.0%

(-14.2% to 3.0%)

3 382

(3 085 to 3 681)

-2.5%

(-11.7% to 7.2%)

407

(296 to 543)

4.1%

(0.2% to 7.6%)

3 789

(3 469 to 4 130)

-1.9%

(-10.2% to 7.0%)

High-middle SDI Turkey31

(28 to 34)

-38.0%

(-45.7% to -30.5%)

1 432

(1 318 to 1 556)

-43.5%

(-50.0% to -35.8%)

552

(395 to 740)

-24.3%

(-30.1% to -17.9%)

1 984

(1 785 to 2 197)

-39.2%

(-45.1% to -32.6%)

High-middle SDI Ukraine80

(76 to 85)

-11.6%

(-16.7% to -6.0%)

3 888

(3 681 to 4 115)

-12.7%

(-17.9% to -6.9%)

1 371

(971 to 1 872)

-7.5%

(-10.1% to -4.8%)

5 259

(4 810 to 5 777)

-11.4%

(-15.3% to -7.0%)

High-middle SDI United Arab Emirates78

(63 to 95)

-18.7%

(-40.4% to 8.2%)

3 005

(2 432 to 3 638)

-21.7%

(-40.9% to 0.8%)

707

(510 to 956)

-25.6%

(-29.6% to -21.2%)

3 712

(3 095 to 4 399)

-22.5%

(-38.0% to -5.2%)

High-middle SDI Uruguay59

(53 to 66)

-6.7%

(-16.7% to 3.1%)

2 553

(2 276 to 2 833)

-12.6%

(-22.3% to -2.5%)

877

(624 to 1 192)

-1.6%

(-5.2% to 1.8%)

3 430

(3 064 to 3 855)

-10.0%

(-17.7% to -2.3%)

High-middle SDI Virgin Islands69

(63 to 77)

-4.6%

(-14.8% to 7.4%)

2 874

(2 630 to 3 205)

-9.4%

(-18.5% to 2.6%)

413

(295 to 557)

7.7%

(2.5% to 13.6%)

3 287

(3 018 to 3 667)

-7.6%

(-15.9% to 3.2%)

Middle SDI Albania26

(22 to 31)

-37.1%

(-47.5% to -25.8%)

1 236

(1 037 to 1 454)

-41.7%

(-51.5% to -31.2%)

1 547

(1 101 to 2 124)

-4.6%

(-8.8% to -0.4%)

2 783

(2 272 to 3 380)

-25.6%

(-33.0% to -18.7%)

Middle SDI Algeria36

(31 to 47)

-44.2%

(-49.2% to -38.7%)

1 731

(1 483 to 2 172)

-48.2%

(-53.4% to -42.2%)

624

(455 to 820)

-18.9%

(-23.4% to -13.7%)

2 355

(2 039 to 2 823)

-42.7%

(-47.4% to -37.7%)

Middle SDI Botswana51

(46 to 58)

-29.0%

(-42.6% to -13.5%)

2 185

(1 929 to 2 487)

-30.6%

(-45.0% to -13.3%)

564

(417 to 741)

-10.9%

(-14.5% to -7.2%)

2 749

(2 449 to 3 106)

-27.3%

(-40.3% to -12.5%)

Middle SDI Brazil74

(73 to 75)

-24.0%

(-25.6% to -22.2%)

3 573

(3 516 to 3 625)

-25.4%

(-27.1% to -23.0%)

512

(375 to 675)

16.1%

(13.4% to 18.8%)

4 085

(3 936 to 4 247)

-21.9%

(-23.9% to -19.5%)

Middle SDI Colombia61

(53 to 68)

-55.8%

(-60.9% to -50.2%)

3 114

(2 753 to 3 501)

-56.0%

(-61.1% to -50.4%)

372

(277 to 486)

-29.7%

(-34.0% to -25.1%)

3 486

(3 101 to 3 876)

-54.1%

(-59.2% to -49.0%)

Middle SDI Costa Rica51

(47 to 54)

-14.5%

(-20.4% to -8.1%)

2 130

(1 987 to 2 280)

-8.5%

(-14.9% to -1.4%)

337

(242 to 451)

4.5%

(-0.1% to 9.1%)

2 468

(2 299 to 2 660)

-6.9%

(-12.6% to -0.6%)

Middle SDI Cuba48

(43 to 53)

-39.6%

(-45.3% to -32.9%)

1 641

(1 473 to 1 846)

-50.3%

(-55.3% to -44.2%)

409

(292 to 553)

8.5%

(4.5% to 12.7%)

2 051

(1 844 to 2 267)

-44.2%

(-49.1% to -38.7%)

Middle SDI Dominica56

(52 to 60)

-1.7%

(-9.2% to 6.6%)

2 732

(2 516 to 2 987)

4.1%

(-5.1% to 14.5%)

421

(308 to 558)

20.4%

(15.5% to 25.1%)

3 152

(2 913 to 3 432)

6.0%

(-2.3% to 15.1%)

Middle SDI Ecuador73

(66 to 80)

-14.1%

(-22.0% to -4.9%)

3 393

(3 070 to 3 759)

-13.5%

(-21.9% to -3.7%)

477

(353 to 628)

-16.0%

(-22.0% to -9.8%)

3 869

(3 532 to 4 258)

-13.8%

(-21.3% to -5.3%)

Middle SDI Equatorial Guinea56

(40 to 78)

-61.9%

(-73.0% to -47.3%)

2 312

(1 604 to 3 229)

-68.7%

(-78.3% to -54.6%)

615

(459 to 801)

-18.7%

(-23.4% to -13.3%)

2 927

(2 201 to 3 904)

-64.1%

(-73.4% to -50.8%)

Middle SDI Fiji49

(44 to 55)

-8.0%

(-21.9% to 7.7%)

2 336

(2 068 to 2 648)

-6.6%

(-20.8% to 10.1%)

526

(389 to 679)

27.0%

(23.1% to 30.8%)

2 862

(2 562 to 3 210)

-1.8%

(-15.1% to 13.2%)

Middle SDI Gabon72

(61 to 84)

-30.0%

(-41.7% to -15.7%)

3 018

(2 442 to 3 602)

-34.9%

(-47.5% to -20.3%)

730

(547 to 947)

-19.0%

(-21.1% to -16.8%)

3 748

(3 161 to 4 374)

-32.3%

(-43.0% to -20.3%)

Middle SDI Grenada45

(42 to 48)

-32.5%

(-37.0% to -27.3%)

1 902

(1 774 to 2 044)

-35.0%

(-40.2% to -29.0%)

439

(321 to 582)

7.7%

(3.2% to 12.0%)

2 341

(2 160 to 2 535)

-29.8%

(-35.0% to -24.3%)

Middle SDI Indonesia39

(37 to 42)

-49.5%

(-54.1% to -43.9%)

1 560

(1 473 to 1 648)

-56.1%

(-59.4% to -51.2%)

359

(271 to 458)

-3.8%

(-6.3% to -1.1%)

1 919

(1 787 to 2 050)

-51.1%

(-54.5% to -46.2%)

Middle SDI Jamaica56

(47 to 65)

89.2%

(58.4% to 123.0%)

2 616

(2 209 to 3 055)

94.5%

(62.2% to 130.8%)

449

(324 to 601)

21.9%

(16.7% to 26.7%)

3 065

(2 630 to 3 529)

78.9%

(53.7% to 107.7%)

Middle SDI Jordan29

(26 to 32)

-51.5%

(-57.9% to -43.4%)

1 387

(1 247 to 1 542)

-54.3%

(-60.8% to -46.2%)

494

(352 to 662)

-25.4%

(-29.5% to -20.9%)

1 881

(1 665 to 2 112)

-49.1%

(-54.9% to -41.8%)

Middle SDI Maldives26

(24 to 33)

-65.9%

(-69.7% to -58.0%)

958

(855 to 1 225)

-72.5%

(-75.9% to -65.0%)

324

(237 to 426)

-8.2%

(-14.6% to -1.3%)

1 282

(1 138 to 1 573)

-66.6%

(-70.6% to -58.9%)

Middle SDI Mexico74

(73 to 75)

-19.4%

(-21.1% to -18.1%)

3 535

(3 486 to 3 588)

-16.1%

(-17.8% to -14.3%)

437

(321 to 579)

-8.7%

(-11.2% to -6.6%)

3 971

(3 842 to 4 120)

-15.4%

(-16.9% to -13.7%)

Middle SDI Moldova59

(56 to 61)

-42.4%

(-45.1% to -39.6%)

2 598

(2 475 to 2 735)

-49.6%

(-52.7% to -46.0%)

1 291

(927 to 1 754)

-19.1%

(-21.8% to -16.5%)

3 889

(3 518 to 4 320)

-42.4%

(-45.3% to -39.3%)

Middle SDI Mongolia72

(65 to 81)

-21.7%

(-31.9% to -11.1%)

3 610

(3 229 to 4 037)

-21.0%

(-30.7% to -10.4%)

1 293

(938 to 1 723)

-2.3%

(-6.4% to 2.3%)

4 903

(4 355 to 5 520)

-16.8%

(-25.0% to -8.5%)

Middle SDI Namibia72

(59 to 87)

-37.4%

(-49.0% to -23.8%)

3 205

(2 551 to 3 994)

-38.0%

(-50.6% to -22.8%)

752

(563 to 963)

-38.7%

(-47.2% to -31.3%)

3 957

(3 288 to 4 766)

-38.1%

(-48.4% to -25.9%)

Middle SDI Panama47

(44 to 51)

-25.7%

(-31.4% to -19.8%)

2 319

(2 119 to 2 519)

-23.3%

(-29.8% to -16.6%)

343

(250 to 455)

-3.7%

(-9.1% to 2.0%)

2 663

(2 451 to 2 883)

-21.2%

(-27.2% to -15.1%)

Middle SDI Paraguay58

(48 to 69)

12.5%

(-7.9% to 37.3%)

2 572

(2 138 to 3 096)

5.8%

(-14.2% to 29.7%)

460

(334 to 612)

1.2%

(-3.2% to 5.5%)

3 032

(2 592 to 3 579)

5.1%

(-11.8% to 25.2%)

Middle SDI Peru41

(35 to 47)

-55.5%

(-62.2% to -48.0%)

1 913

(1 627 to 2 215)

-64.3%

(-70.3% to -57.7%)

442

(325 to 581)

-19.9%

(-26.3% to -13.1%)

2 354

(2 038 to 2 689)

-60.1%

(-65.7% to -53.8%)

Middle SDI Philippines56

(49 to 64)

-20.5%

(-30.4% to -8.9%)

2 536

(2 238 to 2 891)

-23.5%

(-33.4% to -12.4%)

359

(268 to 463)

17.6%

(13.3% to 22.0%)

2 895

(2 575 to 3 260)

-20.1%

(-29.2% to -9.8%)

Middle SDI Saint Lucia57

(53 to 61)

-20.3%

(-26.0% to -14.1%)

2 591

(2 409 to 2 782)

-17.9%

(-24.1% to -10.9%)

410

(299 to 546)

4.7%

(-0.2% to 9.4%)

3 001

(2 798 to 3 222)

-15.4%

(-21.2% to -8.9%)

Middle SDISaint Vincent and the

Grenadines

63

(59 to 67)

4.3%

(-3.1% to 12.0%)

2 917

(2 731 to 3 107)

7.0%

(-1.5% to 15.7%)

455

(334 to 601)

26.9%

(22.4% to 31.0%)

3 372

(3 156 to 3 618)

9.3%

(1.9% to 17.0%)

Middle SDI Seychelles55

(52 to 59)

-19.9%

(-25.7% to -13.1%)

2 332

(2 187 to 2 479)

-18.7%

(-25.3% to -12.3%)

362

(264 to 481)

14.4%

(8.6% to 20.5%)

2 694

(2 515 to 2 883)

-15.4%

(-21.4% to -9.5%)

Middle SDI South Africa84

(79 to 91)

-45.5%

(-50.3% to -41.0%)

3 965

(3 648 to 4 299)

-50.6%

(-54.4% to -46.5%)

612

(452 to 808)

-27.7%

(-30.1% to -25.1%)

4 577

(4 219 to 4 978)

-48.4%

(-52.1% to -44.6%)

Middle SDI Sri Lanka62

(52 to 72)

-55.2%

(-62.1% to -47.5%)

2 302

(1 900 to 2 728)

-62.4%

(-68.9% to -55.6%)

548

(409 to 715)

25.4%

(14.2% to 41.8%)

2 850

(2 430 to 3 327)

-56.5%

(-62.6% to -49.9%)

Middle SDI Suriname79

(71 to 88)

-17.4%

(-27.2% to -6.9%)

3 486

(3 107 to 3 886)

-22.1%

(-31.3% to -11.9%)

468

(344 to 617)

2.6%

(-2.0% to 7.5%)

3 955

(3 581 to 4 377)

-19.8%

(-28.4% to -10.3%)

Middle SDI Syria271

(267 to 276)

487.5%

(421.2% to 563.2%)

14 915

(14 744 to 15 112)

602.2%

(530.0% to 696.4%)

1 426

(1 029 to 1 949)

122.6%

(87.8% to 182.6%)

16 341

(15 893 to 16 858)

491.0%

(437.0% to 556.7%)

Middle SDI Thailand61

(55 to 67)

-38.5%

(-46.5% to -30.3%)

2 973

(2 675 to 3 267)

-38.8%

(-46.7% to -31.0%)

398

(289 to 527)

-4.5%

(-9.7% to 1.2%)

3 371

(3 026 to 3 703)

-36.1%

(-43.5% to -28.7%)

Middle SDI Tonga48

(42 to 54)

-16.1%

(-29.1% to -2.0%)

2 093

(1 833 to 2 370)

-14.5%

(-28.2% to 0.2%)

528

(392 to 682)

24.8%

(20.1% to 29.8%)

2 621

(2 330 to 2 940)

-8.7%

(-20.7% to 4.5%)

Middle SDI Trinidad and Tobago63

(51 to 76)

-7.0%

(-23.8% to 12.2%)

3 073

(2 504 to 3 725)

-3.6%

(-21.3% to 16.4%)

421

(306 to 561)

11.3%

(6.4% to 16.1%)

3 494

(2 904 to 4 148)

-2.0%

(-18.0% to 15.7%)

Middle SDI Tunisia48

(40 to 59)

-39.4%

(-51.5% to -25.1%)

1 983

(1 608 to 2 380)

-47.3%

(-57.3% to -36.0%)

561

(402 to 755)

-22.0%

(-26.9% to -16.8%)

2 544

(2 121 to 2 989)

-43.2%

(-51.9% to -33.6%)

Middle SDI Turkmenistan36

(33 to 39)

-44.8%

(-50.0% to -39.0%)

1 811

(1 650 to 2 006)

-49.4%

(-54.9% to -42.8%)

1 019

(735 to 1 371)

-13.4%

(-17.7% to -8.8%)

2 829

(2 497 to 3 229)

-40.5%

(-45.6% to -35.1%)

Middle SDI Uzbekistan42

(37 to 46)

-33.0%

(-40.5% to -25.0%)

2 032

(1 811 to 2 267)

-38.4%

(-45.5% to -31.0%)

970

(696 to 1 308)

-9.1%

(-12.6% to -5.4%)

3 002

(2 641 to 3 406)

-31.2%

(-37.4% to -25.5%)

Middle SDI Venezuela95

(81 to 111)

12.2%

(-5.0% to 31.3%)

5 025

(4 242 to 5 890)

20.7%

(1.8% to 41.1%)

440

(323 to 581)

2.9%

(-2.3% to 9.0%)

5 466

(4 708 to 6 319)

19.0%

(2.1% to 37.3%)

Middle SDI Vietnam68

(60 to 75)

-25.1%

(-36.0% to -13.0%)

2 559

(2 220 to 2 820)

-34.6%

(-44.9% to -23.1%)

374

(275 to 494)

11.6%

(5.0% to 17.9%)

2 933

(2 578 to 3 225)

-30.9%

(-41.0% to -20.3%)

Low-middle SDI Angola75

(65 to 87)

-55.6%

(-63.9% to -38.1%)

3 278

(2 803 to 3 849)

-63.6%

(-71.2% to -45.6%)

955

(718 to 1 212)

-26.2%

(-30.6% to -21.8%)

4 234

(3 689 to 4 901)

-58.9%

(-66.3% to -42.6%)

Low-middle SDI Belize83

(79 to 87)

7.6%

(-0.2% to 15.4%)

3 910

(3 710 to 4 119)

6.5%

(-1.6% to 15.9%)

494

(367 to 652)

25.3%

(20.8% to 29.9%)

4 404

(4 159 to 4 646)

8.3%

(0.8% to 16.7%)

Low-middle SDI Bhutan50

(40 to 58)

-46.0%

(-57.1% to -34.4%)

1 761

(1 378 to 2 086)

-55.5%

(-66.4% to -44.0%)

659

(496 to 859)

-12.4%

(-16.6% to -7.7%)

2 420

(1 958 to 2 811)

-48.6%

(-58.9% to -37.9%)

Low-middle SDI Bolivia60

(47 to 72)

-51.1%

(-62.7% to -39.4%)

2 580

(1 862 to 3 073)

-58.7%

(-70.9% to -47.2%)

444

(326 to 584)

-14.5%

(-18.8% to -9.8%)

3 024

(2 301 to 3 564)

-55.3%

(-66.8% to -44.2%)

Low-middle SDI Cambodia98

(81 to 114)

-40.6%

(-48.3% to -31.3%)

3 701

(3 137 to 4 333)

-49.6%

(-57.7% to -40.0%)

628

(469 to 812)

-32.9%

(-42.1% to -23.9%)

4 329

(3 741 to 5 012)

-47.7%

(-54.9% to -39.0%)

Low-middle SDI Cameroon79

(65 to 93)

-20.3%

(-33.6% to -5.5%)

2 850

(2 325 to 3 403)

-25.4%

(-39.3% to -10.2%)

614

(457 to 791)

-15.0%

(-17.5% to -12.6%)

3 464

(2 885 to 4 059)

-23.8%

(-35.8% to -11.0%)

Low-middle SDI Cape Verde58

(53 to 63)

-0.2%

(-12.8% to 12.4%)

2 542

(2 293 to 2 810)

-7.3%

(-19.7% to 5.5%)

522

(385 to 689)

-9.6%

(-13.5% to -5.4%)

3 064

(2 753 to 3 383)

-7.7%

(-18.0% to 2.8%)

Low-middle SDI Congo (Brazzaville)77

(65 to 92)

-37.4%

(-49.4% to -22.5%)

3 295

(2 672 to 3 994)

-43.4%

(-55.4% to -27.0%)

850

(639 to 1 089)

10.7%

(3.2% to 21.4%)

4 146

(3 471 to 4 893)

-37.1%

(-48.4% to -22.4%)

Low-middle SDI Djibouti69

(52 to 94)

-32.9%

(-50.1% to -9.5%)

2 416

(1 758 to 3 416)

-42.5%

(-58.0% to -15.8%)

857

(642 to 1 103)

-14.6%

(-17.3% to -11.5%)

3 273

(2 583 to 4 275)

-37.1%

(-50.4% to -15.3%)




per 100,000



1990 and 2017


per 100,000



1990 and 2017


per 100,000



1990 and 2017


per 100,000



1990 and 2017



Low-middle SDI Dominican Republic80

(69 to 92)

19.8%

(-2.8% to 42.5%)

3 744

(3 221 to 4 283)

10.1%

(-10.5% to 30.9%)

464

(337 to 622)

8.9%

(1.5% to 17.1%)

4 208

(3 681 to 4 764)

10.0%

(-8.7% to 28.2%)

Low-middle SDI Egypt54

(44 to 64)

-32.0%

(-43.2% to -21.1%)

2 473

(2 109 to 2 828)

-39.9%

(-47.7% to -29.0%)

572

(416 to 763)

-17.7%

(-23.2% to -11.7%)

3 045

(2 660 to 3 451)

-36.7%

(-43.5% to -27.3%)

Low-middle SDI El Salvador106

(87 to 125)

-31.7%

(-43.7% to -19.0%)

4 924

(4 032 to 5 883)

-35.7%

(-47.7% to -23.5%)

560

(415 to 726)

-48.7%

(-55.7% to -41.1%)

5 484

(4 572 to 6 440)

-37.4%

(-47.9% to -26.3%)

Low-middle SDIFederated States of

Micronesia

73

(52 to 89)

-17.7%

(-41.9% to 4.8%)

3 282

(2 157 to 4 160)

-20.4%

(-48.0% to 3.3%)

574

(426 to 740)

24.5%

(19.8% to 29.1%)

3 856

(2 759 to 4 742)

-15.9%

(-41.0% to 6.1%)

Low-middle SDI Ghana76

(67 to 85)

-3.0%

(-20.1% to 15.0%)

2 646

(2 264 to 3 056)

-11.8%

(-26.7% to 5.8%)

606

(453 to 785)

-4.0%

(-7.3% to -0.6%)

3 252

(2 843 to 3 725)

-10.4%

(-23.1% to 4.1%)

Low-middle SDI Guatemala92

(82 to 102)

-27.9%

(-35.3% to -19.3%)

4 351

(3 894 to 4 872)

-29.5%

(-36.7% to -20.8%)

477

(355 to 616)

-35.7%

(-43.4% to -28.2%)

4 828

(4 351 to 5 366)

-30.2%

(-37.0% to -22.1%)

Low-middle SDI Guyana95

(84 to 107)

-2.2%

(-13.5% to 10.3%)

4 246

(3 738 to 4 780)

-1.8%

(-13.7% to 11.4%)

470

(345 to 620)

11.3%

(6.8% to 15.9%)

4 716

(4 231 to 5 247)

-0.7%

(-11.5% to 11.6%)

Low-middle SDI Honduras84

(68 to 103)

-29.0%

(-44.3% to -10.2%)

3 864

(3 095 to 4 647)

-36.1%

(-50.5% to -19.2%)

478

(361 to 612)

14.8%

(7.3% to 24.1%)

4 342

(3 549 to 5 175)

-32.8%

(-46.3% to -16.9%)

Low-middle SDI India85

(78 to 90)

-23.3%

(-29.6% to -15.8%)

3 070

(2 794 to 3 231)

-34.0%

(-38.2% to -28.5%)

657

(488 to 855)

2.8%

(-0.2% to 6.0%)

3 726

(3 396 to 3 990)

-29.5%

(-33.7% to -24.2%)

Low-middle SDI Iraq115

(112 to 118)

14.0%

(0.4% to 36.2%)

6 311

(6 151 to 6 472)

22.7%

(8.8% to 48.3%)

1 552

(1 125 to 2 064)

-30.1%

(-34.0% to -26.0%)

7 863

(7 434 to 8 416)

6.8%

(-2.8% to 21.6%)

Low-middle SDI Kenya76

(69 to 84)

-11.4%

(-26.3% to -1.2%)

2 568

(2 332 to 2 948)

-11.9%

(-23.0% to -2.3%)

788

(588 to 1 033)

6.6%

(5.1% to 8.3%)

3 356

(3 042 to 3 800)

-8.1%

(-18.1% to -0.2%)

Low-middle SDI Kyrgyzstan47

(45 to 49)

-52.7%

(-55.6% to -49.3%)

2 199

(2 094 to 2 325)

-56.0%

(-59.0% to -52.5%)

978

(708 to 1 311)

-20.2%

(-22.4% to -17.9%)

3 177

(2 895 to 3 530)

-48.9%

(-52.1% to -45.6%)

Low-middle SDI Laos69

(54 to 80)

-49.2%

(-60.3% to -37.7%)

3 203

(2 427 to 3 739)

-54.8%

(-65.8% to -44.3%)

421

(313 to 544)

-0.3%

(-4.2% to 3.8%)

3 624

(2 830 to 4 185)

-51.7%

(-62.3% to -41.1%)

Low-middle SDI Lesotho156

(126 to 184)

25.8%

(-1.5% to 56.5%)

7 282

(5 784 to 8 713)

30.3%

(0.9% to 63.8%)

669

(498 to 864)

6.7%

(3.7% to 10.0%)

7 951

(6 425 to 9 407)

28.0%

(1.4% to 57.4%)

Low-middle SDI Marshall Islands94

(80 to 109)

-12.2%

(-24.7% to 2.1%)

4 279

(3 618 to 5 057)

-13.9%

(-27.0% to 1.0%)

559

(415 to 724)

29.1%

(24.8% to 33.1%)

4 838

(4 141 to 5 637)

-10.5%

(-22.9% to 3.2%)

Low-middle SDI Mauritania62

(54 to 72)

-35.8%

(-45.6% to -22.9%)

2 276

(1 949 to 2 680)

-41.2%

(-50.9% to -25.8%)

560

(415 to 725)

-18.8%

(-21.5% to -15.9%)

2 836

(2 465 to 3 307)

-37.8%

(-46.3% to -24.6%)

Low-middle SDI Morocco46

(37 to 60)

-36.3%

(-47.8% to -24.0%)

2 144

(1 729 to 2 782)

-42.8%

(-53.5% to -31.3%)

566

(415 to 753)

-17.3%

(-21.8% to -12.4%)

2 710

(2 246 to 3 374)

-38.9%

(-48.1% to -29.1%)

Low-middle SDI Myanmar77

(68 to 86)

-32.7%

(-43.2% to -20.9%)

3 393

(3 002 to 3 837)

-39.3%

(-49.8% to -27.3%)

524

(398 to 668)

29.2%

(18.6% to 40.8%)

3 917

(3 487 to 4 375)

-34.7%

(-45.2% to -22.8%)

Low-middle SDI Nicaragua38

(34 to 45)

-46.7%

(-53.7% to -37.4%)

1 701

(1 496 to 1 993)

-52.9%

(-59.0% to -44.7%)

537

(395 to 701)

-44.2%

(-51.1% to -36.9%)

2 238

(1 986 to 2 550)

-51.1%

(-56.4% to -44.4%)

Low-middle SDI Nigeria55

(45 to 71)

-20.4%

(-35.8% to 2.1%)

2 229

(1 801 to 2 795)

-26.1%

(-38.9% to -6.1%)

565

(422 to 733)

-10.3%

(-13.4% to -7.0%)

2 794

(2 342 to 3 383)

-23.4%

(-34.2% to -6.6%)

Low-middle SDI North Korea61

(50 to 73)

14.4%

(-10.6% to 40.9%)

3 036

(2 462 to 3 645)

14.9%

(-11.2% to 44.0%)

525

(395 to 681)

36.3%

(31.5% to 40.9%)

3 560

(2 950 to 4 197)

17.6%

(-5.7% to 42.7%)

Low-middle SDI Pakistan67

(53 to 80)

0.3%

(-16.8% to 22.8%)

2 940

(2 324 to 3 513)

-9.5%

(-25.5% to 12.2%)

716

(538 to 919)

28.6%

(24.3% to 33.1%)

3 655

(3 034 to 4 239)

-3.9%

(-18.0% to 15.0%)

Low-middle SDI Palestine31

(27 to 34)

-55.3%

(-61.8% to -49.6%)

1 455

(1 242 to 1 607)

-60.6%

(-66.5% to -54.8%)

1 254

(916 to 1 672)

2.5%

(-3.6% to 8.7%)

2 709

(2 321 to 3 152)

-44.9%

(-51.0% to -38.9%)

Low-middle SDI Samoa47

(39 to 57)

-25.8%

(-39.8% to -10.5%)

1 949

(1 539 to 2 447)

-33.3%

(-48.0% to -16.7%)

616

(464 to 790)

40.5%

(35.3% to 45.9%)

2 565

(2 099 to 3 063)

-23.7%

(-37.9% to -8.9%)

Low-middle SDI Sao Tome and Principe52

(43 to 61)

-3.2%

(-20.7% to 15.6%)

2 035

(1 754 to 2 352)

-21.2%

(-34.6% to -5.2%)

627

(468 to 812)

-7.8%

(-11.1% to -4.7%)

2 661

(2 321 to 3 031)

-18.4%

(-29.6% to -5.5%)

Low-middle SDI Sudan62

(49 to 81)

-56.0%

(-63.9% to -42.9%)

3 105

(2 514 to 3 892)

-60.5%

(-68.6% to -46.1%)

744

(556 to 966)

-3.6%

(-7.0% to 1.2%)

3 849

(3 247 to 4 653)

-55.4%

(-63.7% to -40.7%)

Low-middle SDI Swaziland117

(94 to 142)

-6.4%

(-29.0% to 18.4%)

5 532

(4 443 to 6 744)

-4.1%

(-27.3% to 21.1%)

644

(474 to 837)

-10.5%

(-13.5% to -7.3%)

6 176

(5 081 to 7 375)

-4.8%

(-25.6% to 17.5%)

Low-middle SDI Tajikistan36

(33 to 40)

-37.4%

(-43.4% to -28.9%)

1 878

(1 715 to 2 095)

-38.6%

(-44.4% to -31.7%)

1 255

(934 to 1 647)

-1.8%

(-6.8% to 5.2%)

3 134

(2 764 to 3 542)

-27.7%

(-32.8% to -22.0%)

Low-middle SDI Timor-Leste44

(30 to 54)

-55.0%

(-69.5% to -45.8%)

1 894

(1 044 to 2 326)

-63.5%

(-79.5% to -54.6%)

907

(655 to 1 223)

-8.2%

(-15.2% to -0.7%)

2 801

(1 998 to 3 355)

-54.7%

(-67.6% to -47.0%)

Low-middle SDI Vanuatu85

(64 to 109)

-5.4%

(-28.1% to 26.3%)

3 855

(2 813 to 5 085)

-6.7%

(-31.6% to 27.3%)

699

(529 to 890)

39.3%

(35.6% to 42.8%)

4 555

(3 480 to 5 776)

-1.7%

(-24.3% to 29.6%)

Low-middle SDI Zambia75

(66 to 84)

-35.7%

(-46.4% to -16.2%)

2 628

(2 289 to 3 019)

-41.9%

(-53.8% to -11.2%)

761

(566 to 975)

-8.7%

(-10.5% to -6.9%)

3 389

(2 980 to 3 830)

-36.8%

(-47.7% to -10.5%)

Low-middle SDI Zimbabwe103

(89 to 118)

19.6%

(-3.9% to 42.8%)

4 144

(3 539 to 4 732)

24.8%

(-0.8% to 51.1%)

668

(503 to 852)

18.4%

(15.5% to 21.7%)

4 813

(4 167 to 5 449)

23.8%

(1.9% to 46.0%)

Low SDI Afghanistan103

(86 to 122)

-35.9%

(-51.9% to 37.6%)

4 913

(4 067 to 5 876)

-40.9%

(-56.9% to 44.4%)

1 995

(1 442 to 2 642)

-42.3%

(-48.5% to -34.9%)

6 909

(5 920 to 8 048)

-41.3%

(-53.2% to 3.5%)

Low SDI Bangladesh42

(37 to 47)

-48.9%

(-56.4% to -38.5%)

1 878

(1 628 to 2 153)

-58.3%

(-65.4% to -46.2%)

610

(457 to 791)

6.4%

(-0.5% to 14.5%)

2 489

(2 209 to 2 810)

-50.9%

(-58.3% to -39.0%)

Low SDI Benin95

(76 to 118)

-24.5%

(-36.8% to -8.9%)

3 565

(2 814 to 4 471)

-33.0%

(-45.2% to -18.5%)

659

(488 to 850)

-11.6%

(-14.1% to -9.0%)

4 224

(3 450 to 5 160)

-30.4%

(-40.9% to -17.6%)

Low SDI Burkina Faso94

(83 to 106)

-14.9%

(-26.5% to 1.2%)

3 585

(3 043 to 4 244)

-18.2%

(-30.6% to -1.9%)

595

(442 to 771)

-5.8%

(-8.8% to -3.0%)

4 181

(3 636 to 4 849)

-16.7%

(-27.8% to -2.4%)

Low SDI Burundi88

(74 to 105)

-34.3%

(-45.2% to -19.4%)

3 099

(2 613 to 3 757)

-39.1%

(-50.1% to -22.7%)

1 790

(1 298 to 2 346)

87.6%

(56.0% to 136.2%)

4 890

(4 156 to 5 817)

-19.1%

(-31.5% to -2.1%)

Low SDI Central African Republic201

(161 to 235)

27.0%

(5.1% to 79.0%)

10 134

(7 937 to 12 026)

31.9%

(6.8% to 113.4%)

879

(652 to 1 129)

33.4%

(23.1% to 48.2%)

11 013

(8 808 to 12 914)

32.0%

(8.6% to 105.1%)

Low SDI Chad89

(75 to 108)

-3.9%

(-17.2% to 10.7%)

3 577

(3 028 to 4 199)

-12.8%

(-27.0% to 3.1%)

776

(584 to 983)

-8.1%

(-13.7% to -3.1%)

4 353

(3 770 to 5 057)

-12.0%

(-23.8% to 1.0%)

Low SDI Comoros67

(56 to 82)

-35.5%

(-46.1% to -22.6%)

2 334

(1 905 to 2 948)

-42.6%

(-52.8% to -28.9%)

787

(582 to 1 014)

-22.0%

(-24.2% to -19.9%)

3 121

(2 631 to 3 805)

-38.5%

(-47.3% to -27.2%)

Low SDI Cote d'Ivoire89

(77 to 102)

-14.1%

(-27.2% to 1.9%)

3 329

(2 849 to 3 867)

-17.7%

(-31.0% to -1.2%)

645

(478 to 832)

-4.8%

(-6.9% to -2.8%)

3 974

(3 456 to 4 557)

-15.8%

(-27.4% to -1.9%)

Low SDI DR Congo78

(67 to 91)

-15.4%

(-28.5% to 6.5%)

3 624

(3 035 to 4 291)

-21.0%

(-35.1% to 6.2%)

770

(585 to 975)

12.4%

(6.9% to 20.3%)

4 394

(3 787 to 5 071)

-16.7%

(-29.6% to 7.2%)

Low SDI Eritrea105

(87 to 125)

-91.6%

(-93.0% to -89.9%)

4 025

(3 255 to 4 929)

-94.1%

(-95.2% to -92.7%)

1 857

(1 367 to 2 448)

-52.6%

(-57.5% to -46.3%)

5 882

(4 940 to 6 952)

-91.9%

(-93.1% to -90.4%)

Low SDI Ethiopia69

(63 to 77)

-61.0%

(-66.7% to -52.6%)

2 378

(2 142 to 2 650)

-68.6%

(-73.9% to -60.3%)

951

(709 to 1 211)

-14.4%

(-17.5% to -11.0%)

3 328

(2 982 to 3 708)

-61.7%

(-67.2% to -53.5%)

Low SDI Guinea85

(72 to 100)

-13.4%

(-26.8% to 1.0%)

3 238

(2 713 to 3 779)

-28.9%

(-41.8% to -12.5%)

647

(481 to 831)

-10.5%

(-13.0% to -7.9%)

3 885

(3 331 to 4 489)

-26.4%

(-37.7% to -12.0%)

Low SDI Guinea-Bissau99

(85 to 116)

-31.7%

(-44.2% to -16.9%)

3 796

(3 132 to 4 602)

-37.5%

(-50.1% to -20.5%)

632

(471 to 810)

-14.0%

(-16.5% to -11.4%)

4 427

(3 753 to 5 218)

-34.9%

(-46.7% to -19.9%)

Low SDI Haiti108

(90 to 132)

-38.4%

(-48.9% to -24.6%)

5 052

(4 206 to 6 110)

-45.1%

(-55.0% to -30.4%)

1 433

(1 052 to 1 894)

164.0%

(116.7% to 225.8%)

6 485

(5 560 to 7 616)

-33.5%

(-44.5% to -18.2%)

Low SDI Kiribati63

(52 to 74)

-7.9%

(-24.4% to 10.1%)

3 101

(2 532 to 3 653)

-9.5%

(-26.7% to 9.4%)

543

(408 to 689)

44.7%

(39.8% to 49.5%)

3 645

(3 080 to 4 220)

-4.1%

(-20.3% to 13.0%)

Low SDI Liberia61

(52 to 73)

-79.6%

(-82.6% to -75.8%)

2 173

(1 789 to 2 721)

-86.4%

(-88.8% to -82.8%)

825

(620 to 1 053)

-6.9%

(-16.5% to 4.6%)

2 997

(2 571 to 3 578)

-82.2%

(-84.7% to -78.5%)

Low SDI Madagascar65

(54 to 76)

-31.4%

(-42.1% to -19.9%)

2 356

(1 954 to 2 831)

-39.9%

(-50.4% to -26.2%)

792

(588 to 1 018)

-16.3%

(-18.4% to -14.3%)

3 148

(2 686 to 3 677)

-35.3%

(-44.1% to -24.2%)

Low SDI Malawi67

(59 to 76)

-31.7%

(-47.6% to 20.7%)

2 328

(1 998 to 2 712)

-38.6%

(-56.3% to 46.1%)

683

(511 to 878)

-7.1%

(-9.6% to -4.6%)

3 011

(2 616 to 3 431)

-33.5%

(-50.1% to 27.9%)

Low SDI Mali77

(64 to 97)

-31.1%

(-41.8% to -15.6%)

3 394

(2 845 to 4 117)

-35.7%

(-46.6% to -18.6%)

702

(529 to 891)

-2.8%

(-8.2% to 4.1%)

4 097

(3 518 to 4 868)

-31.7%

(-41.9% to -16.8%)

Low SDI Mozambique95

(83 to 109)

-25.7%

(-36.8% to -13.1%)

3 477

(2 998 to 4 024)

-35.8%

(-45.8% to -24.0%)

893

(673 to 1 134)

-40.4%

(-49.4% to -31.6%)

4 370

(3 863 to 4 951)

-36.8%

(-44.8% to -27.1%)

Low SDI Nepal70

(57 to 85)

-21.0%

(-37.7% to 0.4%)

2 499

(1 900 to 3 120)

-36.1%

(-51.9% to -10.9%)

675

(510 to 857)

10.5%

(6.0% to 15.3%)

3 173

(2 564 to 3 850)

-29.8%

(-44.0% to -8.0%)

Low SDI Niger71

(58 to 92)

-32.1%

(-43.9% to -15.7%)

2 858

(2 295 to 3 583)

-45.2%

(-56.5% to -26.3%)

625

(461 to 805)

-13.9%

(-16.4% to -11.5%)

3 483

(2 886 to 4 227)

-41.4%

(-52.2% to -24.6%)

Low SDI Papua New Guinea126

(103 to 150)

-19.4%

(-34.2% to -3.1%)

6 181

(5 054 to 7 436)

-21.5%

(-36.6% to -5.1%)

623

(471 to 793)

29.7%

(25.8% to 33.7%)

6 803

(5 652 to 8 041)

-18.5%

(-33.0% to -2.8%)

Low SDI Rwanda79

(66 to 96)

-57.9%

(-64.6% to -48.8%)

2 694

(2 228 to 3 358)

-65.6%

(-71.5% to -56.1%)

1 649

(1 195 to 2 229)

60.5%

(33.8% to 102.5%)

4 342

(3 642 to 5 196)

-51.0%

(-58.7% to -40.5%)

Low SDI Senegal67

(58 to 79)

-18.7%

(-30.0% to -5.8%)

2 342

(2 003 to 2 900)

-30.5%

(-42.0% to -13.2%)

618

(460 to 798)

-10.2%

(-12.4% to -8.0%)

2 960

(2 574 to 3 571)

-27.1%

(-36.8% to -12.6%)

Low SDI Sierra Leone76

(65 to 89)

-16.0%

(-30.1% to 4.7%)

2 978

(2 505 to 3 503)

-30.4%

(-43.7% to -9.8%)

781

(586 to 996)

16.2%

(7.0% to 30.2%)

3 759

(3 263 to 4 335)

-24.1%

(-37.0% to -5.9%)

Low SDI Solomon Islands105

(89 to 123)

-17.1%

(-33.3% to 2.0%)

4 394

(3 637 to 5 212)

-19.8%

(-37.2% to 0.2%)

696

(522 to 889)

29.1%

(26.1% to 31.8%)

5 090

(4 279 to 5 893)

-15.4%

(-32.0% to 2.9%)

Low SDI Somalia150

(120 to 189)

-7.0%

(-30.0% to 39.8%)

6 224

(5 056 to 7 791)

-11.8%

(-35.8% to 44.7%)

1 141

(848 to 1 462)

5.7%

(2.6% to 8.8%)

7 365

(6 143 to 8 961)

-9.5%

(-31.0% to 36.7%)

Low SDI South Sudan141

(121 to 168)

15.5%

(-12.1% to 64.5%)

6 224

(5 439 to 7 291)

25.0%

(-6.7% to 94.3%)

1 167

(867 to 1 495)

29.1%

(21.1% to 41.3%)

7 392

(6 536 to 8 440)

25.7%

(-2.2% to 80.5%)

Low SDI Tanzania61

(54 to 69)

-23.7%

(-36.8% to 8.7%)

2 258

(1 958 to 2 610)

-25.6%

(-41.1% to 29.0%)

739

(554 to 947)

-4.2%

(-6.6% to -1.9%)

2 997

(2 639 to 3 382)

-21.3%

(-34.5% to 18.8%)

Low SDI The Gambia72

(61 to 86)

-7.9%

(-22.9% to 10.0%)

2 416

(1 899 to 3 051)

-18.0%

(-33.4% to 1.5%)

592

(438 to 764)

-11.2%

(-14.6% to -8.2%)

3 008

(2 469 to 3 663)

-16.7%

(-29.8% to -1.1%)

Low SDI Togo73

(62 to 85)

-14.9%

(-29.6% to 2.0%)

2 658

(2 205 to 3 193)

-23.6%

(-37.7% to -5.0%)

588

(435 to 756)

-11.0%

(-13.6% to -8.1%)

3 246

(2 757 to 3 783)

-21.6%

(-34.0% to -6.0%)




per 100,000



1990 and 2017


per 100,000



1990 and 2017


per 100,000



1990 and 2017


per 100,000



1990 and 2017



Low SDI Uganda71

(61 to 84)

-18.4%

(-33.4% to 0.0%)

2 468

(2 095 to 2 997)

-24.8%

(-38.5% to -7.4%)

898

(680 to 1 134)

-19.3%

(-29.2% to -11.2%)

3 366

(2 918 to 3 962)

-23.4%

(-34.5% to -10.0%)

Low SDI Yemen127

(111 to 150)

5.3%

(-19.2% to 103.1%)

6 511

(5 792 to 7 626)

3.4%

(-21.8% to 125.9%)

787

(587 to 1 011)

-10.3%

(-14.6% to -5.6%)

7 298

(6 526 to 8 438)

1.8%

(-20.7% to 93.8%)



burden of injury along the development spectrum

Documents