WHO/ILO joint estimates of the work-related burden of disease and injury, 2000-2016: technical report with data sources and methods

WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury, 2000–2016 TECHNICAL REPORT WITH DATA SOURCES AND METHODS

WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury, 2000–2016 TECHNICAL REPORTI WITH DATA SOURCES AND METHODS WHO/ILO joint estimates of the work-related burden of disease and injury, 2000-2016: technical report with data sources and methods ISBN (WHO) 978-92-4-003492-1 (electronic version) ISBN (WHO) 978-92-4-003493-8 (print version) ISBN (ILO) 978-92-2-035429-2 (print) ISBN (ILO) 978-92-2-035430-8 (web PDF) © World Health Organization and International Labour Organization, 2021 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that the World Health Organization (WHO) or the International Labour Organization (ILO) endorse any specific organization, products or services. The unauthorized use of the WHO or ILO names or logos is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO) or the International Labour Organization (ILO). Neither WHO nor ILO are responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/rules). Suggested citation. WHO/ILO joint estimates of the work-related burden of disease and injury, 2000-2016: technical report with data sources and methods: Geneva: World Health Organization and the International Labour Organization, 2021. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/copyright. ILO publications and digital products can be obtained through major booksellers and digital distribution platforms, For more information, visit the website: www.ilo.org/publns. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third- party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO or ILO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO or ILO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO and ILO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO or ILO be liable for damages arising from its use. ACKNOWLEDGEMENTS v LIST OF ABBREVIATIONS vi EXECUTIVE SUMMARY vii 1. INTRODUCTION 1 1.1. Estimate production 1 1.2. Pairs of occupational risk factor and health outcome 2 1.3. Disaggregation by WHO region, sex and age 3 2. OCCUPATIONAL RISK FACTOR AND HEALTH OUTCOME PAIRS 4 2.1. Established pairs 4 2.2. Recently added pairs 4 3. DATA SOURCES 9 3.1. Established pairs 9 3.2. Recently added pairs 9 3.2.1. Exposure 10 3.2.2. Effect estimates 13 3.2.3. Total burden of disease envelopes 13 3.2.4. Other data sources 13 4. ESTIMATION METHODS 14 4.1. Established pairs 14 4.2. Recently added pairs 15 4.2.1. Exposure to long working hours 15 4.2.2. Burden of disease 19 4.2.3. Uncertainty range calculations 20 4.2.4. Sensitivity analyses 20 4.3. Inequalities in work-related burden of disease 20 CONTENTS 5. DISCUSSION 21 5.1. Strengths 21 5.2. Limitations 22 6. CONCLUSION 23 REFERENCES 24 ANNEXES 28 iv ACKNOWLEDGEMENTS This report presents the data sources and methods of the World Health Organization/International Labour Organization Joint Estimates of the Work-related Burden of Disease and Injury (WHO/ILO Joint Estimates). The WHO/ILO Joint Estimates were produced by Frank Pega (WHO), Natalie Momen (WHO), Kai Streicher (WHO) and Bálint Náfrádi (ILO). Frank Pega and Natalie Momen were the lead writers of this report; the drafting team also included Subas Neupane (WHO), Yuka Ujita (ILO), Bálint Náfrádi and Halim Hamzaoui (ILO). Bochen Cao (WHO), Ahmadreza Hosseinpoor (WHO), Kathleen Krupinski (WHO), Franklin Muchiri (ILO) and Annette Prüss- Üstün (WHO) also provided valuable technical inputs to this report. Frank Pega also coordinated the development and production of the WHO/ILO Joint Estimates and this report; the ILO focal point was Yuka Ujita and then Halim Hamzaoui. Maria Neira (WHO) and Vera Paquete-Perdigão (ILO) provided overall guidance. The National Institute of Occupational Health and Poison Control, Chinese Center for Disease Control and Prevention shared survey data on exposure to long working hours for the People’s Republic of China. Eurostat produced and shared the transition probabilities for exposure to long working hours for 27 countries in the European Region. Financial support for the preparation of this publication was provided by the United States Centers for Disease Control and Prevention National Institute for Occupational Safety and Health through its cooperative agreement with WHO (grant nos 1E11 OH0010676-02, 6NE11OH010461-02-01 and 5NE11OH010461-03-00); the German Federal Ministry of Health (BMG Germany) under the BMG–WHO Collaborative Programme 2020–2023 (WHO specified award reference 70672); and the Spanish Agency for International Cooperation (AECID) (WHO specified award reference 71208). The European Union also provided financial support to the ILO through the Vision Zero Fund (VZF) project on filling data and knowledge gaps on occupational safety and health in global supply chains, implemented within the framework of the ILO Flagship Programme “Safety + Health for All”. The contents of this publication are solely the responsibility of WHO and the ILO, and they do not necessarily represent the official views of any of the WHO or ILO donors mentioned above. v LIST OF ABBREVIATIONS CI confidence interval CRA Comparative Risk Assessment DALY disability-adjusted life year GATHER Guidelines for accurate and transparent health estimates reporting ILO International Labour Organization SDG Sustainable Development Goal UN United Nations UR uncertainty range WHO World Health Organization vi WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 EXECUTIVE SUMMARY The World Health Organization (WHO) and International Labour Organization (ILO) have produced their first WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury (WHO/ILO Joint Estimates), within the established methodological framework of the global Comparative Risk Assessment. These estimates include exposure to occupational risk factors and burden of disease attributable to exposure to occupational risk factors. Attributable burdens of disease have been estimated for 41 pairs of occupational risk factor and health outcome (i.e. disease or injury, referred to as “cause” elsewhere). For 39 established pairs, population attributable fractions were extracted from the published literature and applied to the disease burden envelopes provided by WHO Global Health Estimates. For burden of disease, the numbers of attributable deaths and disability-adjusted life years lost were estimated. Based on a series of systematic reviews and meta-analyses conducted for the WHO/ILO Joint Estimates, the production of estimates for several additional pairs of occupational risk factor and health outcome has been considered. The bodies of evidence on the occupational risk factor of long working hours and the health outcomes of ischaemic heart disease and stroke met pre- specified criteria for quality of evidence and strength of evidence, and WHO and the ILO have produced estimates for these pairs. Exposure to long working hours was estimated using a three- model approach, followed by calculation of the population attributable fractions and ultimately the attributable burden of disease. This Technical Report presents the analytical framework, data sources and methods of the WHO/ ILO Joint Estimates; the full set of these inter-agency estimates is reported in the separate Global Monitoring Report. All estimates of burden of disease were produced at the country, regional and global levels, and are disaggregated by sex and age group. The estimates were reported according to the Guidelines for accurate and transparent health estimates reporting (GATHER). It is anticipated that these estimates will improve understanding of the work-related burden of disease, and provide a base for policy and practice in occupational and workers’ health and safety, nationally, regionally and globally. vii

1. INTRODUCTION To achieve the Sustainable Development Goals (SDGs) and meet the aims of the 2030 United Nations (UN) Agenda (1), the World Health Organization (WHO) and the International Labour Organization (ILO) have developed the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury (WHO/ ILO Joint Estimates). These estimates are required to monitor progress towards the goals and agenda in relation to occupational and workers’ health and safety, and to plan and implement actions to prevent work-related burden of disease. For the first time, these two UN Specialized Agencies have together produced estimates of the exposure to occupational risk factors (as defined by Ezzati et al. (2) and WHO (3)) and the resulting burden of disease attributable to these factors, for the period 2000–2016. Consistent with the terminology and classification of the overarching analytical framework (Ezzati et al. (2)), the term “burden of disease” refers to the combined burdens of three types of health outcomes, namely communicable diseases, non-communicable diseases and injuries. This work builds on previous initiatives (4–15). In this Technical Report, we describe the analytical framework, data sources and methods used to produce the WHO/ILO Joint Estimates for both established and recently added pairs of occupational risk factor and health outcome. 1.1. Estimate production All WHO/ILO Joint Estimates are produced within the framework of the global Comparative Risk Assessment (CRA), and could provide an additional indicator for occupational and workers’ health and safety as well as monitoring progress towards the SDGs (16). The CRA conceptual framework is a web of hierarchically organized risk factors and health outcomes (referred to as “causes” by Ezzati et al. (2)) that contribute to health loss (e.g. loss of life or of years lived without disability) (2), enabling the quantification of exposure to defined risk factors and the burden of disease from a specific health outcome attributable to these risk factors (17). Combining information on prevalence of exposure to a defined risk factor with information about the increased risk of the incidence of or mortality from a defined health outcome among people exposed to the risk factor allows the calculation of the population attributable fraction for this pair of risk factor and health outcome (i.e. the proportional reduction in death or disease from this health outcome that would occur if exposure to the risk factor were removed or reduced to a counterfactual exposure distribution) (Fig. 1). TECHNICAL REPORT WITH DATA SOURCES AND METHODS In collaboration with partners, WHO developed the CRA methodology in the late 1990s (2) and has successfully used this methodology to estimate work-related burden of disease (4, 18). The methodology is established to the point that it is applied to produce several SDG indicators that are endorsed by the UN Statistical Commission (e.g. SDG indicators 3.9.1, 3.9.2 and 3.9.3, mortality rate attributed to: household and ambient air pollution; unsafe water, unsafe sanitation and lack of hygiene; and unintentional poisoning, respectively). In producing the WHO/ILO Joint Estimates, WHO and the ILO have implemented this method jointly to estimate the work-related burden of disease attributable to selected occupational risk factors; this approach has enabled comparability and consistency between methods and estimates of exposure to risk factors and of burden of disease, across risk factors and health outcomes, and over time. The WHO/ILO Joint Estimates were produced with theoretical minimum risk exposure, using counterfactual exposure distribution (the exposure distribution that results in minimum population risk) (17) as the basis of comparison. All WHO/ILO Joint Estimates are reported in adherence with the Guidelines for accurate and transparent health estimates reporting (GATHER) (19) (Annex 1). This ensures transparency in the reporting of data sources, methods and results, along dedicated and agreed technical reporting guidelines specifically for health estimates. WHO consulted its Member States on the estimates in March 2020 and July 2020. The estimates were produced in September 2020. 1.2. Pairs of occupational risk factor and health outcome The WHO/ILO Joint Estimates currently cover a total of 41 pairs of occupational risk factor and health outcome: 39 previously established pairs and another two recently added pairs, for which the exposure and the burden of disease had not previously been estimated. FIGURE 1 COMPARATIVE RISK ASSESSMENT METHOD FOR BURDEN OF DISEASE ESTIMATION Distribution of exposure to risk factor in the population Risk ratio of health outcome when exposed to risk factor Total burden of disease estimates per health outcome Burden of disease from health outcome attributable to risk factor Population attributable fraction 2 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 1.3. Disaggregation by WHO region, sex and age Geographically, the WHO/ILO Joint Estimates are reported at the levels of country and region, and globally. The six regions used are those classified by WHO (African Region, Region of the Americas, South-East Asia Region, European Region, Eastern Mediterranean Region and Western Pacific Region). As estimates are reported disaggregated by country, they can be combined to produce estimates for other geographic regions, such as the five ILO-classified regions, or non-geographically defined regions, such as the World Bank income groups. The estimates of exposure to occupational risk factors are reported for 194 countries, but burden of disease estimates only for the 183 of these with populations greater than 90 000 in 2015 (20). Burden of disease estimates, and hence attributable burdens of disease, could not be produced for 11: Andorra, Cook Islands, Dominica, Marshall Islands, Monaco, Nauru, Niue, Palau, Saint Kitts and Nevis, San Marino and Tuvalu. All estimates are produced fully disaggregated by two socioeconomic variables: sex and age group. For the variable sex, estimates were produced for the three categories of both sexes, females and males. For the variable age group, estimates were produced for 18 groups (≥ 15, 15–19, 20–24, ..., 90–94 and ≥ 95 years). These disaggregations of the estimates enable monitoring of inequalities in burden of disease by sex and age group, both between and within countries. This fulfils the call to “leave no one behind” in the SDGs, and in sustainable development more broadly (1). INTRODUCTION 3 2. OCCUPATIONAL RISK FACTOR AND HEALTH OUTCOME PAIRS TECHNICAL REPORT WITH DATA SOURCES AND METHODS 2.1. Established pairs Estimates for 39 established pairs of occupational risk factor and health outcome (Table 1) were produced by determining the burden of disease using existing data and established WHO and ILO methodologies (Section 3.1 and 4.1) (2, 21, 22). 2.2. Recently added pairs The feasibility of producing estimates for several additional pairs of occupational risk factor and health outcome, identified by WHO and the ILO in consultation with individual experts at the beginning of the interagency work on the WHO/ILO Joint Estimates (26), has been investigated and considered. Domains, and specific selection criteria within each domain, were developed to systematically prioritize pairs of occupational risk factor and health outcome (Table 2). Scoping reviews of the evidence base were conducted for these potential additional pairs to assess (i) the availability of global data on exposure to the occupational risk factor and (ii) the existing systematic review and meta-analytic evidence of the effect of exposure to the occupational risk factor on the health outcome. The application of these criteria identified 16 prioritized additional pairs of risk factor and health outcome for further consideration through evidence review and synthesis. Supported by more than 220 individual experts in 35 countries, WHO and the ILO then conducted a series of systematic reviews and meta-analyses of the evidence for these prioritized pairs for burden of disease estimation (for an overview see Pega et al. (27)). To ensure that these systematic reviews and meta- analyses were tailored to fulfil the specific evidence and data needs of the WHO/ILO Joint Estimates, and that any estimates obtained were based on the latest and entire bodies of evidence currently available, all systematic reviews and meta-analyses followed peer-reviewed pre-published protocols (28–37) and were conducted and reported along WHO and ILO standards for such evidence syntheses (38–45). TABLE 1 ESTABLISHED PAIRS OF OCCUPATIONAL RISK FACTOR AND HEALTH OUTCOME Risk factora Health outcomeb 1 Occupational exposure to asbestos Trachea, bronchus and lung cancers 2 Occupational exposure to asbestos Ovary cancer 3 Occupational exposure to asbestos Larynx cancer 4 Occupational exposure to asbestos Mesothelioma 5 Occupational exposure to arsenic Trachea, bronchus and lung cancers 6 Occupational exposure to benzene Leukaemia 7 Occupational exposure to beryllium Trachea, bronchus and lung cancers 8 Occupational exposure to cadmium Trachea, bronchus and lung cancers 9 Occupational exposure to chromium Trachea, bronchus and lung cancers 10 Occupational exposure to diesel engine exhaust Trachea, bronchus and lung cancers 11 Occupational exposure to formaldehyde Nasopharynx cancer 12 Occupational exposure to formaldehyde Leukaemia 13 Occupational exposure to nickel Trachea, bronchus and lung cancers 14 Occupational exposure to polycyclic aromatic hydrocarbons Trachea, bronchus and lung cancers 15 Occupational exposure to silica Trachea, bronchus and lung cancers 16 Occupational exposure to sulphuric acid Larynx cancer 17 Occupational exposure to trichloroethylene Kidney cancer 18 Occupational asthmagens Asthma 19 Occupational particulate matter, gases and fumes Chronic obstructive pulmonary disease 20 Occupational noise Other hearing loss 21 Occupational injuriesc Pedestrian road injuries 22 Occupational injuriesc Cyclist road injuries 23 Occupational injuriesc Motorcyclist road injuries 24 Occupational injuriesc Motor vehicle road injuries 25 Occupational injuriesc Other road injuries 26 Occupational injuriesc Other transport injuries 27 Occupational injuriesc Poisoning by carbon monoxide 28 Occupational injuriesc Poisoning by other means 29 Occupational injuriesc Falls 30 Occupational injuriesc Fire, heat and hot substances 31 Occupational injuriesc Drowning 32 Occupational injuriesc Unintentional firearm injuries 33 Occupational injuriesc Other exposure to mechanical forces 34 Occupational injuriesc Pulmonary aspiration and foreign body in airway 35 Occupational injuriesc Foreign body in other body part 36 Occupational injuriesc Non-venomous animal contact 37 Occupational injuriesc Venomous animal contact 38 Occupational injuriesc Other unintentional injuries 39 Occupational ergonomic factors Back and neck pain a Defined as per the Global Burden of Disease Study classification (25). b Defined as per the burden of disease classification of the WHO Global Health Estimates (26) with the exception of injuries, which are defined as per Global Burden of Disease Study classification (25). c Throughout this report the term “Occupational injuries” is used as defined by Ezzati et al. (2, 3) to represent an occupational risk factor within the framework of the global Comparative Risk Assessment. This definition differs from that adopted by the 1982 Thirteenth International Conference of Labour Statisticians (27), and was revised by the 1998 Sixteenth International Conference of Labour Statisticians (28) to mean “any personal injury, disease or death resulting from an occupational accident”. OCCUPATIONAL RISK FACTOR AND HEALTH OUTCOME PAIRS 5 TABLE 2 DOMAINS AND CRITERIA FOR SELECTING ADDITIONAL PAIRS OF OCCUPATIONAL RISK FACTOR AND HEALTH OUTCOME FOR THE CONSIDERATION OF ESTIMATE PRODUCTION Domain Criterion Large burden of disease Prioritize pairs of occupational risk factor and health outcome that our crude estimates suggest accrue a larger burden of disease (threshold: ≥ 6000 deaths or ≥ 30 000 DALYs) compared with others Existing data on exposed population and effect estimate Prioritize occupational risk factors for which both a global database (ideally disaggregated by country, sex, age group, industry and level of exposure) of the exposed population and a relative effect estimate exist over those without such data Existing systematic review evidence Prioritize pairs of occupational risk factor and health outcome that our scoping review suggests have at least some prior systematic review evidence with strong relative effect estimates over pairs for which evidence has not yet been systematically reviewed; relative effect estimates (for example, risk ratios, odds ratios and hazard ratios) are considered strong if they are derived through meta-analysis from multiple high-quality studies, such as randomized controlled trials or prospective cohort studies Scientific consensus on causality Prioritize pairs of occupational risk factor and health outcome that have a relatively high-quality body of supporting theoretical and empirical evidence on causality over pairs with a relatively low-quality body of such evidence; crucial supporting evidence includes evidence of a causal effect of the risk factor on the outcome, as well as on the causal pathways (or mechanisms) through which the risk factor affects the outcome, including key mediating factors (especially for more distal risk factors) Included in existing WHO or ILO methodology Prioritize pairs of occupational risk factor and health outcome that have already been included in the WHO/ILO methodologies over those that have not Large public interest or knock-on effect Prioritize pairs of occupational risk factor and health outcome that are of greater current public interest or have larger knock-on effects (e.g. effects on health workers that threaten health care provision) over those of lesser public interest or without any knock-on effects Preventable occupational risk factor Prioritize more easily preventable pairs of occupational risk factor and health outcome over those that are less easily preventable DALYs, disability-adjusted life years; ILO, International Labour Organization; WHO, World Health Organization. In order to harmonize the systematic reviews and to ensure consistency, WHO and the ILO: convened 48 coordination meetings with all lead reviewers; convened two face-to-face meetings with individual experts; provided several training workshops to build capacity of participating individual experts in conducting the specific systematic reviews along the agreed standards; established a Working Group of systematic review methodologists to strengthen global capacity for evidence synthesis for work-related burden of disease estimation; and, supported by individual experts, developed novel systematic review tools (46). An overview of all systematic reviews and a description of all innovations developed for this series is available elsewhere (27). The occupational risk factor of interest in four systematic reviews and meta-analyses was exposure to long working hours (here defined as ≥ 55 hours/week) (Table 3). Following the pre-published protocols (28–31), studies of the effect of exposure to long working hours on the risk of stroke, ischaemic heart disease, depressive disorder and alcohol use disorder were subject to systematic review and their reported estimates were included in meta-analyses (38–41) (Table 4). WHO and the ILO selected the pairs of occupational risk factor and health outcome with an evidence base (as presented in the systematic reviews and meta-analyses) that the organizations judged sufficient for the production of official burden of disease estimates (2, 48). Using Navigation Guide ratings (49), the body of evidence had to have been judged to be either of “high quality” or “moderate 6 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 TABLE 3 DEFINITION OF RISK FACTOR, RISK FACTOR LEVELS AND THEORETICAL MINIMUM RISK EXPOSURE LEVEL FOR EXPOSURE TO LONG WORKING HOURS Occupational risk factor Definition Levels (hours per week) Theoretical minimum risk exposure level Exposure to long working hoursa Working > 40 hours per week, that is, exceeding the standard working hours (35–40 hours per week) (i) 35–40 (ii) 41–48 (iii) 49–54 (iv) ≥ 55 Standard working hours defined as 35–40 hours per week a The ILO defines long working hours as > 48 hours per week, based on the Hours of Work (Industry) Convention, 1919 (No. 1) and Hours of Work (Commerce and Offices) Convention, 1930 (No. 30), which set the general standard of 48 hours of work per week. TABLE 4 SYSTEMATIC REVIEWS AND META-ANALYSES ON THE EFFECT OF EXPOSURE TO LONG WORKING HOURS ON VARIOUS HEALTH OUTCOMES Exposure to long working hours (hours per week) No. studies in meta-analysis (no. participants) Risk ratioa (95% CI) Quality of evidence (see Table 5) Strength of evidence of human data (see Table 6) Evidence sufficient to proceed to estimation (47) Ischaemic heart disease (39) 41–48 20 (312 209) 0.98 (0.91–1.07) Low Inadequate evidence of harmfulness No 49–54 18 (308 405) 1.05 (0.94–1.17) Low Inadequate evidence of harmfulness No ≥ 55 22 (339 680) 1.17 (1.05–1.31) Moderate Sufficient evidence of harmfulness Yes Stroke (38) 41–48 12 (265 937) 1.01 (0.91–1.12) Low Inadequate evidence of harmfulness No 49–54 17 (275 181) 1.13 (1.00–1.28) Moderate Limited evidence of harmfulness No ≥ 55 7 (162 644) 1.35 (1.13–1.61) Moderate Sufficient evidence of harmfulness Yes Depressive disorder (31) 41–48 8 (49 392) 1.03 (0.90–1.17) Low Inadequate evidence of harmfulness No 49–54 8 (49 392) 1.04 (0.95–1.13) Low Inadequate evidence of harmfulness No ≥ 55 17 (91 142) 1.05 (0.96–1.14) Low Inadequate evidence of harmfulness No Alcohol use disorder (40) 41–48 0 (0) – Low Inadequate evidence of harmfulness No 49–54 0 (0) – Low Inadequate evidence of harmfulness No ≥ 55 0 (0) – Low Inadequate evidence of harmfulness No CI, confidence interval. a Calculated for the effect of the category of exposure to long working hours on the health outcome, compared with the minimum risk exposure level, defined as working 35–40 hours per week (standard working hours). quality” (Table 5) and the strength of the evidence had to have been rated as “sufficient evidence for harmfulness” (Table 6) (48). The organizations then selected the “best” effect estimate (risk ratios for morbidity versus mortality) based on strength of evidence ratings (48). If there was any evidence for fatal or non-fatal events of the health outcome rated as “sufficient evidence for harmfulness”, this was selected as the “best” estimate. In the event that both fatal and non-fatal events had the same rating, estimates for fatal events were prioritized. OCCUPATIONAL RISK FACTOR AND HEALTH OUTCOME PAIRS 7 According to the above-described criteria for proceeding to estimation, WHO and the ILO have been able to calculate the burdens of disease for the category of exposure to long working hours of ≥ 55 hours per week and the health outcomes of stroke and ischaemic heart disease (see last column in Table 4). Estimates for these two pairs are also provided in the accompanying Global Monitoring Report (26) as well as a dedicated scientific journal article (48). TABLE 6 NAVIGATION GUIDE STRENGTH OF EVIDENCE RATINGS (49) Strength of evidence rating Definition Sufficient evidence of harmfulness The available evidence usually includes consistent results from well designed, well conducted studies, and the conclusion is unlikely to be strongly affected by the results of future studies; for human evidence a positive relationship is observed between exposure and outcome where chance, bias and confounding can be ruled out with reasonable confidence. Limited evidence of harmfulness The available evidence is sufficient to determine the effects of the exposure, but confidence in the estimate is constrained by factors such as: the number, size or quality of individual studies; confidence in the effect; or inconsistency of findings across individual studies. As more information becomes available, the observed effect could change and this change may be large enough to alter the conclusion. For human evidence, a positive relationship is observed between exposure and outcome where chance, bias and confounding cannot be ruled out with reasonable confidence. Inadequate evidence of harmfulness Studies permit no conclusion about a toxic effect. The available evidence is insufficient to assess effects of the exposure because of the limited number or size of studies, the low quality of individual studies or inconsistency of findings across individual studies. More information may allow an estimation of effects. Evidence of lack of harmfulness The available evidence includes consistent results from well designed, well conducted studies, and the conclusion is unlikely to be strongly affected by the results of future studies. For human evidence, more than one study showed no effect on the outcome of interest at the full range of exposure levels that humans are known to encounter, where bias and confounding can be ruled out with reasonable confidence. The conclusion is limited to the age at exposure and/or other conditions and levels of exposure studied. TABLE 5 NAVIGATION GUIDE QUALITY OF EVIDENCE RATINGS (49) Quality of evidence rating Definition High Further research is very unlikely to change our confidence in the estimate of effect Moderate Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate 8 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 3. DATA SOURCES In this section, we report the various input data sources used to produce the WHO/ILO Joint Estimates for the established pairs, as well as the two recently added pairs (ischaemic heart disease and stroke attributable to exposure to long working hours). The input databases and sources are summarized in Table 7. 3.1. Established pairs For the 39 established pairs of occupational risk factor and health outcome included in this estimation (Table 1), WHO and the ILO derived the population attributable fractions (disaggregated by country, sex and age group) from the Global Burden of Disease Study (23) and total disease burden envelopes from the WHO Global Health Estimates (3). Each established pair’s attributable burden of disease was then estimated using the CRA framework (2). 3.2. Recently added pairs The WHO/ILO Joint Estimates on exposure to long working hours are produced primarily from global databases. The sources, availability, and coverage by country, area or territory and population of the data in these exposure databases on long working hours are described in detail below and depicted in Fig. 2. TABLE 7 SUMMARY OF THE MAIN DATA SOURCES USED AS INPUTS Database/estimates Content Source WHO/ILO global cross-sectional and longitudinal databases of working hours Prevalence of exposure to long working hours by exposure level (41–48, 49–54 and ≥ 55 hours per week), country/area/territory, sex and age group Labour Force Surveys, Gallup surveys UN population prospects No. people by country, sex and age group UN estimates WHO life tables Probability of dying by country, sex and age group WHO estimates WHO total disease burden envelope by health outcome No. DALYs and deaths by country, health outcome, sex and age group WHO estimates DALYs, disability-adjusted life years; ILO, International Labour Organization; UN, United Nations; WHO, World Health Organization. TECHNICAL REPORT WITH DATA SOURCES AND METHODS FIGURE 2 FLOW CHART SHOWING HOW DATA SOURCES, INPUT DATA AND MODELS WERE COMBINED TO PRODUCE MODEL OUTPUTS AND BURDEN ESTIMATES Data sources Input variables Proportion of people in exposure category by sex and age at first year of time window Proportion of people in exposure category by sex and age over the time window Estimates of prevalence of exposure category by sex and age at first year of time window Average annual transition probabilities between categories of exposure Number of exposed population over time window Estimates of relative risks Number of deaths by sex and age at each time unit during time window Number of persons by sex and age at first year of time window Model 1 Model 2 Model 3 Model 4 Prevalence of exposure category by sex and age at first year of time window Average annual transition probabilities between categories of exposure Number of exposed population over time window Estimated burden of disease Total number of deaths/DALYs by sex and age at estimation year Models Outputs Input data 1: Cross-sectional LFS Input data 2: Longitudinal LFS Input data 4: WHO life tables Input data 3: UN population prospect Input data 6: WHO and ILO systematic reviews Input data 5: WHO Global Health Estimates 3.2.1. Exposure (a) Cross-sectional database The WHO/ILO Global Cross-Sectional Working Hours Database includes 467 million observations from 2324 surveys conducted in 154 countries, areas and territories between 1 January 1976 and 31 December 2018 (Table 8), the majority of which are official household surveys (mainly Labour Force Surveys). Data from at least one survey are available for 77.4% (154/199) of the countries, areas and territories. Data are available for at least half of the countries, areas and territories within each region (Table 8). The number of surveys in the database by country, area and territory is presented in Fig. 3 and the number of surveys by country, area and territory grouped within region over time for the period 1976–2018 is shown in Fig. 4. Aggregate data are openly available to browse via the ILOSTAT web portal (https://ilostat.ilo.org), and fully disaggregated data for a country are available to the responsible national statistical office and other government agencies upon request. A more detailed description of the database, including all source surveys, is provided elsewhere (see Pega et al. (48)). 10 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 Source: Pega et al. (48). (b) Longitudinal database The WHO/ILO Global Longitudinal Working Hours Database was also established specifically for the WHO/ILO Joint Estimates. The database comprises 143 million observations from 739 quarterly datasets of Labour Force Surveys conducted in 15 countries between 1 January 2000 and 31 December 2018. These quarterly survey datasets use sample rotation to ensure sample overlaps, with measures taken repeatedly from the same survey participants over consecutive years. FIGURE 3 MAP OF COUNTRIES, AREAS AND TERRITORIES WITH DATA (COLOURS) AND WITHOUT DATA (GREY) IN THE WHO/ILO GLOBAL CROSS-SECTIONAL WORKING HOURS DATABASE TABLE 8 NUMBER OF SURVEYS AND NUMBER AND PERCENTAGE OF COUNTRIES, AREAS AND TERRITORIES COVERED IN THE WHO/ ILO GLOBAL CROSS-SECTIONAL WORKING HOURS DATABASE BY REGION AND GLOBALLY WHO regiona Global African Region Region of the Americas South-East Asia Region European Region Eastern Mediterranean Region Western Pacific Region No. countries, areas and territories 47 36 11 53 22 30 199 No. surveys 135 437 96 1435 66 155 2324 No. countries, areas and territories with ≥ 1 survey (% of countries, areas and territories) 37 (78.7%) 24 (66.7%) 10 (90.9%) 45 (84.9%) 11 (50.0%) 27 (90.0%) 154 (77.4%) a See Annex 2 for listing of countries, areas and territories within regions. 0 1–5 6–10 11–20 21–30 31–40 41–50 >50 Not applicable © WHO 2021. All rights reserved. Data Source: WHO/ILO Joint Estimates of the Work- related Burden of Disease and Injury Map Production: WHO GIS Centre for Health, DNA/DDI The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. DATA SOURCES 11 FIGURE 4 TIME AND COUNTRY, AREA AND TERRITORY DISTRIBUTION OF THE NUMBER OF SURVEYS IN THE WHO/ILO GLOBAL CROSS-SECTIONAL WORKING HOURS DATABASE GROUPED BY REGION African Region Angola Benin Botswana Burkina Faso Burundi Cabo Verde Cameroon Comoros Congo Côte d’Ivoire Democratic Republic of the Congo Eswatini Ethiopia Gambia Ghana Guinea Kenya Liberia Madagascar Malawi Mali Mauritania Mauritius Mozambique Namibia Niger Nigeria Rwanda Senegal Seychelles Sierra Leone South Africa Togo Uganda United Republic of Tanzania Zambia Zimbabwe 1980 1990 2000 2010 2020 Year Region of the Americas Argentina 000000000000000000000000000111111122222111110 Belize 000000000000000000000000000000000000021111110 Bolivia (Plurinational State of) 000000000000000000000000000001112101111222100 Brazil 000000000000000000011111012222222212433321110 Canada 111111111111111111111111111111111111111111100 Chile 000000000000001010101010100000100112221212110 Colombia 000000000000000000000000000000011111221111110 Costa Rica 000000000000000000000000000001001121222222210 Dominican Republic 000000000000000000000000111111111111111321100 Ecuador 000000000000000000000000110111111111211111110 El Salvador 000000000000000000000000000000000011212211100 Guatemala 000000000000000000000000001110100012112111100 Guyana 000000000000000000000000000000000000000001000 Haiti 000000000000000000000000000000000000100000000 Honduras 000000000000000000000000000001111111111111100 Jamaica 000000000000000000000000000000000000011110000 Mexico 000000000000000000011111101111111111111111110 Nicaragua 000000000000000000000000000000000000101000000 Panama 000000000000000000000000000000000021212111100 Paraguay 000000000000000000000000001111111023222222110 Peru 000000000000000000000000012222222222232222110 United States of America 000000000000000000111111111111111111111111110 Uruguay 000000000000000000000011111111111121111111100 Venezuela (Bolivarian Republic of) 000000000000000000000000000001111111100000000 1980 1990 2000 2010 2020 Year Eastern Mediterranean Region Afghanistan 000000000000000000000000000000001001210001000 Djibouti 000000000000000000000000000000000000000001000 Egypt 000000000000000000000000000000001111211111000 Iraq 000000000000000000000000000000010000100000000 Jordan 000000000000000000000000000000000000100100000 Lebanon 000000000000000000000000000000000000001000000 Occupied Palestinian territory, including east Jerusalem 000000000000000000000000111111111111121211100 Pakistan 000000000000000000000000000000111112222101100 Tunisia 000000000000000000000000000000000010010000000 United Arab Emirates 000000000000000000000000000000000000000001100 Yemen 000000000000000000000000000000000010001000000 1980 1990 2000 2010 2020 Year European Region Albania 000000000000000000000000000000011121110000000 Armenia 000000000000000000000000000000012111213322000 Austria 000000000000000000011111111133333333333332100 Belgium 000000011111111111111111111122222222222221000 Bosnia and Herzegovina 000000000000000000000000000000111111111111110 Bulgaria 000000000000000000000000111111122222222221000 Croatia 000000000000000000000000001111111122222221000 Cyprus 000000000000000000000000111112222222222221000 Czechia 000000000000000000000111222223333333333332100 Denmark 000000011111111111111111111122222222222221000 Estonia 000000000000000000000111111122222222222221000 Finland 000000000000000000011111111122222222222221000 France 000000011111111111111111111122222222222221000 Georgia 000000000000000000000000000000000000000001100 Greece 000000011111111111111122222233333333333332100 Hungary 000000000000000000001111111112222222222221000 Iceland 000000000000000000011111111122222222222211000 Ireland 000000011111111111111111111122222222222211000 Israel 000000000000000000000000000000000000111110000 Italy 000000011111111111111111111122222222223322100 Kyrgyzstan 000000000000000000000000000000000000011000000 Latvia 000000000000000000000011111112222222222221000 Lithuania 000000000000000000000011111112222222222221000 Luxembourg 000000011111111111111111111122222222222211000 Malta 000000000000000000000000000000011222222211000 Montenegro 000000000000000000000000000000000001111211100 Netherlands 000000010101111111111111111112222222222221000 North Macedonia 000000000000000000000000000000111111212111100 Norway 000000000000000000011111111122222222222221000 Poland 000000000000000000000111111112222222222221000 Portugal 000000000011111111111122222233333333333332000 Republic of Moldova 000000000000000000000000000000000111111211220 Romania 000000000000000000000111222222233333333332100 Russian Federation 000000000000000000000000000000000011211211100 Serbia 000000000000000000000000000000001111222321110 Slovakia 000000000000000000000011111112222233333331000 Slovenia 000000000000000000001111111112222222222221000 Spain 000000000011111111111112222233333333333332100 Sweden 000000000000000000011111111122222222222221000 Switzerland 000000000000000111112222222222233333333322100 Tajikistan 000000000000000000000000000000000100000000000 Turkey 000000000000000000000000000011111111111111000 Ukraine 000000000000000000000000000000000000010100000 United Kingdom 000000011111111122222222222223333333333332110 1980 1990 2000 2010 2020 Year South-East Asian Region Bangladesh 000000000000000000000000000000100011141011000 Bhutan 000000000000000000000000000000000000011000000 India 000000000000000000000000000000000001110000000 Indonesia 000000000000001111101111111111111212222111100 Maldives 000000000000000000000000000000000000000010000 Myanmar 000000000000000000000000000000000000110101100 Nepal 000000000000000000000000000000001001121001000 Sri Lanka 000000000000000000000000000000000112122110000 Thailand 000000000000000000000000000000000011123333300 Timor-Leste 000000000000000000000000000000000010010010000 1980 1990 2000 2010 2020 Year Western Pacific Region Australia 000000000000000000000000000000000001011000000 Brunei Darussalam 000000000000000000000000000000000000001001100 Cambodia 000000000000000000000000110000011012423110000 China 000000000000000000000000000000001001120000000 China, Hong Kong Special Administrative Region 000000000000000000000000000000000001101000000 Taiwan, China 000000000000000000000000000000000001111000000 Cook Islands 000000000000000000000000000000000000000010000 Fiji 000000000000000000000000000001000001000010000 Japan 000000000000000000000000111111111111222111100 Lao People's Democratic Republic 000000000000000000000000000000000011100001000 Malaysia 000000000000000000000000000000000001111000000 Marshall Islands 000000000000000000000000000000000000000000100 Micronesia (Federated States of) 000000000000000000000000000000000000001000000 Mongolia 000000000000000000000000000000001112222111110 Nauru 000000000000000000000000000000000000010000000 New Zealand 000000000000000000000000000000000001111000000 Papua New Guinea 000000000000000000000000000000000010000000000 Philippines 000000000000000000000000011111111112222111100 Republic of Korea 000000000000000000000000000000000001111121100 Republic of Palau 000000000000000000000000100000000000001000000 Samoa 000000000000000000000000000000000000200001000 Singapore 000000000000000000000000000000000001010000000 Solomon Islands 000000000000000000000000000000000000010000000 Tonga 000000000000000000000000000000000000000000100 Tuvalu 000000000000000000000000000000000000000010000 Vanuatu 000000000000000000000000000000100010000000000 Viet Nam 000000000000000000000000000000010112331211110 1980 1990 2000 2010 2020 YearSurveys (N) 1  2  3  4  12 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 Because microdata do not include individual participant identifiers, data were probabilistically linked longitudinally using matching by household number, household sequence number, sex and year of birth. These were all official surveys conducted by national statistical offices and originally shared by countries with the ILO or European Commission Directorate Eurostat (https://ec.europa.eu/eurostat). A more detailed description of the database, including all source surveys, is provided elsewhere (see Pega et al. (48)). 3.2.2. Effect estimates Risk ratios for the effect of exposure to long working hours on ischaemic heart disease and stroke were sourced from the literature using systematic reviews and meta-analyses (38, 39). Table 4 summarizes the bodies of evidence from systematic reviews and meta-analyses of studies with estimates of the effect of long working hours on four different health outcomes, reporting the effect of long working hours on each health outcome by three different risk factor categories. For the two recently added pairs considered in this report, systematic reviews found sufficient evidence of harmfulness at the exposure level of ≥ 55 hours per week on both ischaemic heart disease and stroke. 3.2.3. Total burden of disease envelopes Estimates of the total numbers of deaths (Annexes 3 and 4) and DALYs as a result of ischaemic heart disease and stroke for the years 2000, 2010 and 2016 were sourced from the WHO Global Health Estimates (3). The total number of deaths by disease, sex and age group were available for 2016 for the 183 countries with populations larger than 90 000 in the year 2015. These estimates are openly available to browse via the WHO Global Health Observatory webpage (https://www.who.int/data/gho). 3.2.4. Other data sources (a) UN population estimates Estimates of the total populations by country, year, sex and age group for the years 1950–2018 were sourced from the UN global population estimates (50). (b) WHO life tables Estimates of probability of death by country, year, sex and age group were sourced from WHO life tables (51). DATA SOURCES 13 4. ESTIMATION METHODS 4.1. Established pairs We followed the CRA framework (2) to estimate the burden of disease attributable to exposure to occupational risk factors. We derived population attributable fractions (disaggregated by sex and age group) for the 39 established pairs of occupational risk factor and health outcome (Table 1) from the Global Burden of Disease Study. The estimates from which these population attributable fractions were calculated are openly available (http://ghdx.healthdata.org/). WHO and the ILO calculated point estimates of population attributable fractions from the Global Burden of Disease Study estimates as the attributable number of deaths or DALYs as a fraction of the total number of deaths or DALYs, by country, sex, age and health outcome for the years 2000, 2010 and 2016 as: P,GBD P O,GBD aBOD PAF = BOD (1) where PAF is the population attributable fraction, aBOD is the attributable burden of disease (number of deaths or DALYs) and BOD is the burden of disease envelope (total number of deaths or DALYs). The subscript P indicates variables relating to the numbers of deaths or DALYs resulting from a particular health outcome attributable to that occupational risk factor, for each cohort defined by country, sex and age group (attributable burden of disease for the specific risk factor). The subscript O indicates the numbers of deaths or DALYs resulting from a particular health outcome, for each cohort defined by country, sex and age group (total disease burden envelope). For each pair, point estimates of the respective WHO/ILO attributable burden of disease estimates were calculated by multiplying the WHO/ILO Global Health Estimates of the total numbers of deaths and DALYs by the corresponding population attributable fraction separately for each cohort defined by country, sex, age and health outcome for the years 2000, 2010 and 2016, that is: aBODP = BODO,GHE × PAFP (2) where BODO,GHE represents the burden of disease envelope defined by International Statistical Classification of Diseases and Related Health Problems (version 10) codes in the Global Health Estimates (3) in terms of numbers of deaths or DALYs. The 95% uncertainty ranges (URs) for aBODP were calculated assuming that it follows a normal distribution with expected value E equal to the point estimate and variance (var) calculated as: TECHNICAL REPORT WITH DATA SOURCES AND METHODS var(aBOD) = var(BOD)var(PAF) + var(BOD)[E(PAF)]2 var(PAF)[E(BOD)]3 (3) where it is assumed that PAF and BOD are independent random variables. In this case, the upper and lower URs are defined as E(aBOD)± z* var(aBOD) , where z* is the critical value calculated from the inverse of the cumulative distribution function of the standard normal distribution at 0.025. 4.2. Additional pairs To produce these estimates, data were required on the number of workers exposed to long working hours, as well as the risk ratio of exposure at the hazardous level (here defined as ≥ 55 hours per week) compared with exposure to the theoretical minimum risk exposure level (here defined as 35–40 hours per week). We estimated the number of the exposed population using multilevel models (52). Measures of exposure to long working hours were taken primarily from global databases of national official surveys, and the models contributed the percentage of the population (disaggregated by country, sex and age group; Annex 5) exposed to working ≥ 55 hours per week. Our estimates of exposure were combined with estimates of the risk ratios to generate the population attributable fractions for long working hours and each of the health outcomes of ischaemic heart disease and stroke, which were then used to estimate the attributable burden of disease (48). We made several modelling assumptions based on available evidence (Table 9), described in the following. A more detailed description of the estimation models for the recently added pairs is provided elsewhere (48). 4.2.1. Exposure to long working hours A three-model approach was developed to estimate the exposure to long working hours (Annex 5) (48). Model 1 estimated the prevalence of exposure to long working hours using a multilevel model. Model 2 calculated the transition probabilities between exposure categories during the time window of the exposure. Finally, Model 3 was a microsimulation model used to estimate the exposed population during the time window. (a) Model 1: Multilevel model An established multilevel model was used that predicts prevalence over time for a particular geographical region (52). WHO has used this model regularly to estimate exposure to environmental risk factors such as air pollution (64) and water sanitation and hygiene (65). The UN Statistical Commission considers this method to be established, and has approved it for producing several SDG indicators, including 3.9.1, 3.9.2 and 3.9.3 (https://unstats.un.org/unsd/statcom). In the microsimulation model, these estimates were used as the first set of input data (Input Data 1). For each year during 1980–2016, for each population defined by country, sex and age group, we produced estimates of the proportion of the population in each exposure category (i). We modelled Input Data 1 using the following multilevel model (Model 1): Proportioni = Ai + Bit (4) where Proportioni is the proportion of the population in exposure category i in a given group defined by country/area/territory, sex and age group, and t is the survey year. The intercept Ai and slope Bi of t, dependent on Proportioni, are calculated using a multilevel model with sex and age as fixed effects, and sex and age as random effects, nested in the countries, areas and territories within the region (with regions treated independently). Because Proportioni was strongly non-linearly dependent on age, we linearized age by fifth-order orthogonal polynomials to prevent collinearity. RESULTS 15 TABLE 9 MODELLING ASSUMPTIONS AND THEIR EVIDENCE BASE Assumption in main analysis (sensitivity analyses) Explanation and example Evidence base Lag time (a) of 10 years (8 and 12 years) For an outcome event in year t, the exposure is assumed to have occurred in the lag year (year t–a) For example, burden of disease in 2016 is attributable to exposure 10 years earlier, with the lag year being 2006 Theoretically, lag time varies according to the mechanism via which long working hours are associated with ischaemic heart disease and stroke, that is, either: directly (exposure has a direct effect on pathophysiology); indirectly (exposure impacts risk factors for ischaemic heart disease and/or stroke); as a trigger (exposure triggers events that lead to ischaemic heart disease and/or stroke events); and/or as a prognostic factor (exposure affects prognosis of coronary heart disease or cerebrovascular disease) (53, 54). If direct and indirect effects are the dominant mechanisms, then lag time could be < 10 years; if exposure acts as a trigger or a prognostic factor, 10 years would be too long. If all four mechanisms contribute to risk of cardiovascular disease, an average lag of 10 years is an appropriate assumption. Previous examples of the use of lag times of around 10 years include: (i) in WHO/ILO systematic reviews and meta-analyses on ischaemic heart disease and stroke (38, 39); (ii) mean follow- up times in previous large systematic reviews and individual studies of around 9 and 8 years for ischaemic heart disease and stroke, respectively (55, 56); (iii) an incubation period of at least 10 years for coronary heart disease (57); and (iv) according to evidence from the CONSTANCES Cohort Study in France, only a lag time of ≥ 10 years was found to increase odds ratios of ischaemic heart disease and stroke (58, 59). For exposure to long working hours, direct measures are available in the WHO/ILO global databases, as described in Section 3.2.1. These provided the proportion of survey participants within the predefined categories of working hours (including ≥ 55 hours per week) that people in a group defined by country, sex and age group were exposed to, allowing exposure to be modelled directly. (b) Model 2: Transition probabilities Based on the methodology developed by Eurostat, pseudo-longitudinal data were derived from the European Union (EU) Labour Force Surveys by matching data from the annually overlapping samples, averaging over the four quarters per year, for the years 2010–2018. Because of the absence of personal identifiers, matching was based on household number, household sequence number, sex and year of birth. Longitudinal weights were derived by scaling the available target year weights to represent the correct labour market status (labour market inactive versus working) by sex and 10- year age group (66). The aggregated transition matrix for labour market status was then ranked to match initial and target year margins (with the exception of initial year inactivity status). 16 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 TABLE 9 (CONTINUED) MODELLING ASSUMPTIONS AND THEIR EVIDENCE BASE Assumption in main analysis (sensitivity analyses) Explanation and example Evidence base Time window of exposure (b) of 10 years (8 and 12 years) Rather than occurring in year t–a only, exposure occurs in any year during a “critical” time window of length b, and exposure within any year in this time window can still cause the disease outcome in year t For example, to estimate burden of disease in 2016, we model exposure over a 10-year time window As mentioned above, the four potential mechanisms (53, 54) are likely to have different lag times; a time window of exposure around the lag year accounts for some of this variability. Previous occupational burden of disease studies have also estimated exposure over a time window (60, 61). Evidence suggests that exposure (sometimes measured cumulatively) to long working hours during this 10-year time window contributes to a significant increase in cardiovascular disease incidence (57–59). Spacing of the time window (b) symmetrically around the lag year The time window of the exposure is equally spaced around the “lag year” of the average lag period (year t–a), so that the time window of exposure is defined as from year t–a–(b/2) to year t–a+(b/2) For example, to estimate burden of disease in 2016, we model exposure over the time window of 2001–2010 As mentioned above, the four potential mechanisms (53, 54) are likely to have different lag times; with the exception of the trigger mechanism, symmetrical spacing of the time window around the lag year is a reasonable assumption. If all four mechanisms contribute, symmetrical spacing is the most appropriate model. This is common practice in studies estimating burden of disease attributable to exposure to occupational risk factors; for example, a report on the burden of occupational cancer in the United Kingdom of Great Britain and Northern Ireland estimated the “peak latency” period for their outcomes of interest and spaced the time window of exposure symmetrically around this point (61). The highest exposure category in any year over the time window of exposure (the most common exposure category) For each worker, the highest exposure category they had in any year over the time window is assigned as their exposure category over the window For example, during 2001–2010, worker A was exposed to ≥ 55 hours work per week in 2001 and 2002, and to 49–54 hours work per week in 2003–2011; we therefore assign worker A the exposure category of ≥ 55 hours per week For diseases with long latency periods, which is possible for cardiovascular disorders, once the disease process has started, the worker continues to be at risk even if exposure levels are reduced. The assignment of the highest level of exposure observed over the time window is in line with assumptions made by other studies focusing on the effect of long working hours and ischaemic heart disease and stroke (55, 56, 58, 59, 62, 63). The “best” effect estimate For estimating numbers of deaths and DALYs and for all cohorts defined by country, sex, and age group, we assigned the same “best” effect estimate There is no evidence for effect modification by country (or WHO region), sex or age group in the subgroup analyses in the WHO/ ILO systematic reviews (38, 39); we therefore assigned the pooled effect estimate from the main analysis. This is the same approach used in previous WHO burden of disease studies (21). We systematically selected the “best” effect estimate, based on the pre-specified criteria (see Section 2.2). This is based on prioritizing mortality over morbidity and relatively higher strength of evidence over lower strength of evidence (Section 2.2 and Table 4). Source: Pega et al. (16). RESULTS 17 We estimated average transition probabilities between exposure categories of long working hours by running a weighted multinomial logit regression model, including sex and a function of age as regressors, and using a matched sample of all available longitudinal data. We then used fractional polynomials with automated model selection in Stata version 14 (StataCorp LLC, College Station, United States of America) to model age for each separate regression run for exposure categories h0, h1, ..., h5 (where the six categories refer to labour market inactive and working hours categories 0–35, 35–40, 41–48, 49–54 and ≥ 55 hours per week). From the coefficients estimated, we derived predicted probabilities of transitioning between exposure categories by group defined by sex and age group. Including age as a continuous function allowed strength to be borrowed from the distribution of age, in order to estimate age groups with limited numbers of observations. The final estimates, particularly for the highest and lowest age groups, are therefore driven by the choice of function. For each population group defined by country, sex and age group, we estimated the probability (Pj) of transitioning from long working hours exposure category i in year t to exposure category i in year t + 1. Here, j denotes one of the n possible transitions from one of the exposure categories in year t to a specific exposure category in year t + 1. We adopted methods developed by Eurostat for calculating these transition probabilities (66). Using Input Data 2, we scaled the survey weights for the target year (year t + 1) to represent the correct labour market status by country, sex and age group for the initial year (year t) and the target year. We then adjusted the complete sample in the target year to match margins for labour market status in both years, using iterative raking by sex. We did not match the exposure category i = 0 (labour market inactive) for the initial year. Input Data 2 was modelled using the multinomial logit regression model (Model 2): ( ) ( ) exp 1 exp j j j j X P Xαα β β = +∑ (5) where βj is the set of regression coefficients describing the longitudinal weights associated with transition j; Xj is a set of explanatory variables (sex and age as a fractional polynomial with maximal permitted degree of four associated with transition j); and the summation (index α) encompasses all possible transitions j (except the transition from i = 0 in year t to i = 0 in year t + 1, which was chosen as a pivot outcome). By modelling transition probabilities, we derived 15 900 transition probabilities for the 15 countries for which we hold data (listed in supplementary tables S3 and S5 of Pega et al. (48)) using Input Data 2. In addition, by modelling quarterly EU Labour Force Surveys data using transition probabilities, Eurostat derived 31 104 transition probabilities covering 27 countries and shared these transition probabilities with WHO and the ILO. For populations defined by country, sex and age group for which Pi could not be calculated (because the required longitudinal data were unavailable), Pi was imputed. The imputed Pi was the mean of all transition probabilities of the population defined by the same sex and age in the region, weighted by the number of observations contributing to the transition probabilities. (c) Model 3: Microsimulation A time window is the period in which an exposure can lead to health loss in the estimation year (67). As an example, we seek to estimate the burden of disease at year a that is attributable to past exposure to a risk factor. For this, we require estimates of the number of people exposed to the risk factor, at a particular level of exposure, throughout the time window (year t − (b/2) to year t + (b/2)). The known or assumed lag time is year t + (b/2) to year a. We then seek to estimate the number of people exposed to the risk factor at the highest level of exposure during the time window (Fig. 5). 18 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 For each population defined by country, sex and age group, we estimated the proportion (Proportionk) of the population in each exposure category throughout the time window (year t − (b/2) to year t + (b/2)). We defined k as the highest exposure category i in any year in the time window. Based on advice from the WHO/ILO Technical Advisory Group, we assumed burden of ischaemic heart disease and stroke in the estimation year could be the result of exposure to (hazardous) working hours during a time window of 5–15 years before the estimation year. For example, to estimate burden for the year 2016, we assume that the time window of exposure was 2001–2010. We used microsimulation, a method for generating micro-level estimates by combining individual- and aggregate-level datasets, and initiated a synthetic population for each country. We used input data on the probability of dying for the synthetic cohort to ensure representative sex and age distribution during the first year of the time window (i.e. estimation year minus 15 years), as well as the estimates output from the multilevel model to probabilistically assign each individual to a specific exposure category (i) in the first year. Using transition probabilities for each year over the entire time window, transitions from one exposure category to another were stochastically modelled to estimate each synthetic individual’s working hours category in each year. Using transition probabilities over the time window (Input Data 4), from the first year of the time window to the estimation year, each individual was stochastically assigned to the states of “dead” or “alive”. All synthetic individuals that reached the state “dead” before the estimation year were censored. Using this microsimulation method, Proportionk is derived using the model (Model 3): ,max1,..., ( ) Proportion k ll n k S n δ == ∑ (6) where the summation runs through individuals “alive” in the estimation year; δk,max is the Kronecker delta function; Sl is the sequence of the lth individual of all working hours categories (i) in each year in the time window; and max denotes the highest i that the lth individual experiences in the sequence assigned. 4.2.2. Burden of disease As for the established pairs, the CRA framework (2) was used to estimate the burden of disease attributable to exposure to occupational risk factors. We estimated the proportional reduction in death or disease that would occur if exposure was reduced to a level with a minimum risk (i.e. working 35– 40 hours per week), while other conditions remain unchanged. FIGURE 5 DEFINITION OF TIME WINDOW OF EXPOSURE. ADAPTED FROM PEGA ET AL. (48). Time window Lag time Estimation year time (year) year t year t+n/2 year t+n year a Be gi nn in g of ti m e w in do w En d of ti m e w in do w RESULTS 19 (a) Population attributable fractions We calculated population attributable fractions (Annexes 6 and 7) using prevalence estimates from the WHO/ILO global working hours databases for the recently added pairs of exposure to long working hours and both ischaemic heart disease and stroke, using estimates output from the microsimulation model (Model 3) and risk ratios for the exposure categories. To produce the attributable burdens of ischaemic heart disease and stroke for the risk factor of exposure to long working hours, we applied the estimation model: ( ) ( ) 1 1 Proportion RR 1 PAF Proportion RR 1 1 n k kk n k kk = = − = − + ∑ ∑ (7) where Proportionk is the proportion of the population in working hours category k; RRk is the risk ratio for the exposure category k; and n is the total number of long working hours categories. (b) Applying population attributable fractions to total disease burden envelopes Applying the population attributable fraction (Annexes 6 and 7) to the total mortality burden of the health outcome provides the total number of deaths from the disease or injury that can be attributed to the occupational risk factor. The population attributable fractions calculated for each additional pair of occupational risk factor and health outcome were applied to the total disease burden envelopes for the health outcome from the WHO Global Health Estimates for the years 2000–2016 (3). 4.2.3. Uncertainty range calculations When estimating the attributable burden of disease for selected occupational risk factors, several risks of bias and/or errors may exist, such as risk of selection bias, statistical error, or risk of confounding of underlying input parameters. A large body of literature exists regarding the estimation of statistical uncertainty of an estimate, which is itself a function of existing estimates. We calculated uncertainty ranges for exposure, death and DALY estimates using bootstrapping (68). One hundred estimates of prevalence were produced with starting parameters sampled independently from normal distributions, with the median equal to the corresponding point estimate and uncertainty ranges taken from those of the prevalence estimates per year. The 2.5% and 97.5% quantiles of the resulting random deviates of the exposures were then calculated and assigned as the lower and upper limits of the uncertainty range, respectively (48). 4.2.4. Sensitivity analyses We performed the following sensitivity analyses to test our assumptions (Annex 8): (i) we reduced the lag time to 8 years (2003–2012); (ii) we increased the lag time to 12 years (1999–2008); (iii) we reduced the time window for the exposure to 8 years (2002–2009); (iv) we increased the time window to 12 years (2000–2011); and (v) we assigned the long working hours category with the largest number of years in the time window (censoring years spent in labour market inactivity) (48). 4.3. Inequalities in work-related burden of disease To consider differences in the occupational burden of disease between regions, sexes and age groups, we used the number of deaths or DALYs per 100 000 population (i.e. death or DALY rate) for all regions, both sexes and for people of working age (≥ 15 years) as the reference. As an absolute measure of inequality, for each death rate and DALY rate for each category of region, sex and age group, we then calculated the difference from the reference rate (global rate) (58). As a relative measure of inequality, we also calculated the rate ratios as the fraction of the reference rate for each death rate and DALY rate, for each category (69). 20 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 5. DISCUSSION The WHO/ILO Joint Estimates have several strengths and limitations which should be considered when interpreting the estimates. 5.1. Strengths These official estimates of work-related burden of disease have been produced jointly by the UN Specialized Agencies for health and labour, improving partnerships for development and policy coherence across sectors. The exposure estimates were generated from large cross-sectional and longitudinal databases of direct exposure measurements collected primarily by producers of official statistics in countries, areas and territories. This is the largest exposure database for any occupational risk factor (comprising over 2300 official surveys, primarily Labour Force Surveys, and covering 77.4% (154/199) of the countries, areas and territories with at least one survey). These data are likely to be of high quality as they are produced primarily by national statistics offices using national statistical standards; once provided to WHO and the ILO, they are harmonized to international statistical standards. The estimates of burden of disease are generated using risk ratios calculated in tailor-made systematic reviews and meta-analyses of the latest bodies of evidence that were comprehensive, transparent and synthesized with the latest systematic review methods. The production of these WHO/ILO reviews was also supported by a large number of individual experts. The WHO/ILO Joint Estimates make use of new and improved modelling methods, including microsimulation models to estimate exposure over a time window using the longitudinal data of direct exposure. This is a versatile and flexible modelling approach for producing complex occupational exposure estimates that make use of official data produced by countries, areas and territories. Finally, the estimates of burden of disease are produced at the country, regional and global levels. They are also fully disaggregated by sex and age group, honouring the central premise of the SDGs to reduce inequalities (1). The estimates provide the basis for designing, planning, costing, TECHNICAL REPORT WITH DATA SOURCES AND METHODS implementing and evaluating actions to prevent work-related burden of disease, with a focus on improving population health and health equity among workers, both within and across sectors. 5.2. Limitations As for all estimates, several assumptions have been made during their modelling (Table 9). However, all assumptions have been transparently reported as stipulated in the GATHER guidelines (19) (Annex 1). The modelling assumptions are based on the best current evidence, but some assumptions may change as additional evidence becomes available and the evidence base improves. The estimates are not disaggregated by some important factors, such as occupation, industrial section and migration status. As more data become available, this additional disaggregation can be considered and added as appropriate, if and when feasible. 22 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 6. CONCLUSIONS The WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury (WHO/ILO Joint Estimates) provide a new approach for generating estimates of loss of life and health from exposure to occupational risk factors, the relative importance of different occupational risk factors and health outcomes leading to death or disability, global regional patterns, trends over time, and inequalities by sex and age group within and between countries. These estimates are obtained in several ways, from the use of large cross-sectional and longitudinal data, to new and innovative evidence synthesis and statistical modelling approaches that add accuracy to these estimates. This estimation cycle has benefited from the contributions of a large number of individual experts and a variety of data sources to calculate the burden of disease for the included pairs. The resulting estimates (as well as the processes undergone, methods developed and experience gained) can help advance understanding of the work-related burden of disease at the national, regional and global levels for the period 2000– 2016, and its distribution by sex and age group. These estimates provide the base for policy and practice in occupational and workers’ health and safety nationally, regionally and globally. TECHNICAL REPORT WITH DATA SOURCES AND METHODS 1. United Nations. Transforming our world: The 2030 agenda for Sustainable Development; 2015 (https://sustainabledevelopment.un.org/ post2015/transformingourworld/publication, accessed 28 July 2021). 2. Ezzati M, Lopez AD, Rogers A, Murray CJL, editors. Comparative quantification of health risks: global and regional burden of disease attributable to selected major risk factors. Geneva: World Health Organization; 2004 (https://apps.who.int/iris/handle/10665/42770, accessed 28 July 2021). 3. Global Health Estimates. Geneva: World Health Organization; 2018 (https://www.who.int/data/global-health-estimates, accessed 03 August 2021). 4. Wolf J, Prüss-Ustün A, Ivanonv I, Mugdal S, Corvalán C, Bos R, et al. Preventing disease through a healthier and safer workplace. Geneva: World Health Organization; 2018 (https://apps.who.int/iris/handle/10665/272980, accessed 29 July 2021). 5. Concha-Barrientos M, Nelson D I, Driscoll T, Steenland K N, Punnet L, Fingerhut M, Prüss-Üstün A, Leigh J, Woo Tak S, Corvalan C. Chapter 21. Selected occupational risk factors. In Ezzati M, Lopez AD, Rogers A, Murray CJL, editors. Comparative quantification of health risks: global and regional burden of disease attributable to selected major risk factors. Geneva: World Health Organization; 2004. 6. Concha-Barrientos M, Nelson DI, Fingerhut M, Driscoll T, Leigh J. The global burden due to occupational injury. Am J Ind Med. 2005;48:470–81. https://doi.org/10.1002/ajim.20226 PMID:16299709 7. Driscoll T, Nelson DI, Steenland K, Leigh J, Concha-Barrientos M, Fingerhut M, et al. The global burden of disease due to occupational carcinogens. Am J Ind Med. 2005;48:419–31. doi:10.1002/ajim.20209 PMID:16299703 8. Driscoll T, Nelson DI, Steenland K, Leigh J, Concha-Barrientos M, Fingerhut M, et al. The global burden of non-malignant respiratory disease due to occupational airborne exposures. Am J Ind Med. 2005;48:432–45. doi:10.1002/ajim.20210 PMID:16299701 9. Driscoll T, Takala J, Steenland K, Corvalan C, Fingerhut M. Review of estimates of the global burden of injury and illness due to occupational exposures. Am J Ind Med. 2005;48(6):491–502. doi:10.1002/ajim.20194 PMID:16299705 10. Nelson DI, Concha-Barrientos M, Driscoll T, Steenland K, Fingerhut M, Punnett L, et al. The global burden of selected occupational diseases and injury risks: Methodology and summary. Am J Ind Med. 2005;48:400–18. doi:10.1002/ajim.20211 PMID:16299700 11. Nelson DI, Nelson RY, Concha-Barrientos M, Fingerhut M. The global burden of occupational noise-induced hearing loss. Am J Ind Med. 2005;48:446–58. doi:10.1002/ajim.20223 PMID:16299704 12. Punnett L, Prüss-Üstün A, Nelson DI, Fingerhut MA, Leigh J, Tak S, et al. Estimating the global burden of low back pain attributable to combined occupational exposures. Am J Ind Med. 2005;48:459–69. doi:10.1002/ajim.20232 PMID:16299708 13. Prüss-Üstün A, Rapiti E, Hutin Y. Estimation of the global burden of disease attributable to contaminated sharps injuries among health-care workers. Am J Ind Med. 2005;48:482–90. doi:10.1002/ajim.20230 PMID:16299710 14. Nenonen N, Hämäläinen P, Takala J, Saarela KL, Yong E, Lim SL, et al. Global estimates of occupational accidents and work-related illnesses 2014. Tampere: Tampere University of Technology; Singapore: Workplace Safety & Health Institute; Tampere: VTT Technical Research Centre (https://www.wshi.gov.sg/-/media/wshi/past-publications/2014/global-estimates-of-occupational-accidents-and-work-related-illness-2014. pdf?la=en&hash=CC0F4E20BDDEB6189A8B1BB4786D7973, accessed 4 August 2021). 15. Hämäläinen P, Takala J, Kiat TB. Global estimates of occupational accidents and work-related illnesses 2017. Helsinki: Ministry of Social Affairs and Health, Finland (http://www.icohweb.org/site/images/news/pdf/Report%20Global%20Estimates%20of%20Occupational%20Accidents%20 and%20Work-related%20Illnesses%202017%20rev1.pdf, accessed 4 August 2021). 16. Indicator proposal: Indicator 8.8.3: Mortality rate from diseases attributed to occupational risk factors, by disease, risk factor, sex, and age group. Submission to the Inter-agency and Expert Group on SDG Indicators. Geneva: World Health Organization and International Labour Organization; 2019. 17. Murray CJ, Lopez AD. On the comparable quantification of health risks: lessons from the Global Burden of Disease Study. Epidemiology. 1999;10:594–605. PMID:10468439 18. Prüss-Ustün A, Wolf J, Corvalán C, Bos R, Neira M. Preventing disease through healthy environments: a global assessment of the burden of disease from environmental risks. Geneva: World Health Organization (https://www.who.int/publications/i/item/9789241565196 accessed 3 August 2021). REFERENCES 24 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 19. Stevens GA, Alkema L, Black RE, Boerma JT, Collins GS, Ezzati M, et al. Guidelines for accurate and transparent health estimates reporting: the GATHER statement. Lancet. 2016;388:e19–e23 .doi:10.1016/S0140-6736(16)30388-9 PMID:27371184 20. WHO methods and data sources for global burden of disease estimates 2000-2016. Global Health Estimates Technical Paper WHO/HIS/IER/ GHE/2018.4. WHO: Geneva; 2018. 21. Murray CJL, Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S. Comparative quantification of health risks: conceptual framework and methodological issues. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, editors. Comparative quantification of health risks: global and regional burden of disease attributable to selected major risk factors. Geneva: World Health Organization; 2004:1–39. 22. Prüss-Ustün A, Wolf J, Corvalán C, Neville T, Bos R, Neira M. Diseases due to unhealthy environments: an updated estimate of the global burden of disease attributable to environmental determinants of health. J Public Health. 2017;39:464–75. doi:10.1093/pubmed/fdw085 PMID:27621336 23. GBD 2017 Risk Factor 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. Lancet. 2018;392:1923–94. doi:10.1016/S0140-6736(18)32225-6 PMID:30496105 24. Thirteenth International Conference of Labour Statisticians (Geneva, 18-29 October 1982). Geneva: International Labour Office; 1982 (https:// labordoc.ilo.org/discovery/fulldisplay?docid=alma992208263402676&context=L&vid=41ILO_INST:41ILO_V2&lang=en&adaptor=Local%20 Search%20Engine, accessed 29 July 2021). 25. Sixteenth International Conference of Labour Statisticians, Geneva, 6-15 October 1998. Geneva: International Labour Organization; 1998 (https://ilo.primo.exlibrisgroup.com/discovery/fulldisplay?docid=alma993316683402676&context=L&vid=41ILO_INST:41ILO_ V2&lang=en&adaptor=Local%20Search%20Engine, accessed 29 July 2021). 26. Global Monitoring Report: WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury, 2000–2016. Geneva: World Health Organization; Geneva: International Labour Organization. 27. Pega F, Momen NC, Ujita Y, Driscoll T, Whaley P. Systematic reviews and meta-analyses for the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2021;155:106605. doi:10.1016/j.envint.2021.106605 PMID:34051644 28. Descatha A, Sembajwe G, Baer M, Boccuni F, Di Tecco C, Duret C, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of exposure to long working hours and of the effect of exposure to long working hours on stroke. Environ Int. 2018;119:366–78. doi:10.1016/j.envint.2018.06.016 PMID:30005185 29. Li J, Brisson C, Clays E, Ferrario MM, Ivanov ID, Landsbergis P, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of exposure to long working hours and of the effect of exposure to long working hours on ischaemic heart disease. Environ Int. 2018;119:558–69. doi:10.1016/j.envint.2018.06.022 PMID:30125833 30. Godderis L, Boonen E, Cabrera Martimbianco AL, Delvaux E, Ivanov ID, Lambrechts MC, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of exposure to long working hours and of the effect of exposure to long working hours on alcohol consumption and alcohol use disorders. Environ Int. 2018;120:22–33. doi:10.1016/j.envint.2018.07.025 PMID:30055358 31. Rugulies R, Ando E, Ayuso-Mateos JL, Bonafede M, Cabello M, Di Tecco C, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of exposure to long working hours and of the effect of exposure to long working hours on depression. Environ Int. 2019;125:515–28.doi:10.1016/j.envint.2018.11.011 PMID:30737040 32. Hulshof CTJ, Colosio C, Daams JG, Ivanov ID, KC P, Kuijer PPFM, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of exposure to occupational ergonomic risk factors and of the effect of exposure to occupational ergonomic risk factors on osteoarthritis of hip or knee and selected other musculoskeletal diseases. Environ Int. 2019;125:554–66. doi:10.1016/j.envint.2018.09.053 PMID:30583853 33. Mandrioli D, Schlunssen V, Adam B, Cohen RA, Colosio C, Chen W, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of occupational exposure to dusts and/or fibres and of the effect of occupational exposure to dusts and/or fibres on pneumoconiosis. Environ Int. 2018;119:174–85. doi:10.1016/j.envint.2018.06.005 PMID:29958118 34. Paulo MS, Adam B, Akagwu C, Akparibo I, Al-Rifai RH, Bazrafshan S, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of occupational exposure to solar ultraviolet radiation and of the effect of occupational exposure to solar ultraviolet radiation on melanoma and non-melanoma skin cancer. Environ Int. 2019;126:804–15. doi:10.1016/j.envint.2018.09.039 PMID:30792021 35. Teixeira LR, Azevedo TM, Bortkiewicz A, Correa da Silva DT, de Abreu W, de Almeida MS, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of exposure to occupational noise and of the effect of exposure to occupational noise on cardiovascular disease. Environ Int. 2019;125:567–78. doi:/10.1016/j.envint.2018.09.040 PMID:30683322 36. Tenkate T, Adam B, Al-Rifai RH, Chou BR, Gobba F, Ivanov ID, et al. WHO/ILO work-related burden of disease and injury: protocol for systematic reviews of occupational exposure to solar ultraviolet radiation and of the effect of occupational exposure to solar ultraviolet radiation on cataract. Environ Int. 2019;125:542–53. doi:10.1016/j.envint.2018.10.001 PMID:30737039 REFERENCES 25 37. Pega F, Chartres N, Guha N, Modenese A, Morgan RL, Martinez-Silveira MS, et al. The effect of occupational exposure to welding fumes on trachea, bronchus and lung cancer: a protocol for a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2020;145:106089. doi:/10.1016/j.envint.2020.106089 PMID:32950789 38. Descatha A, Sembajwe G, Pega F, Ujita Y, Baer M, Boccuni F, et al. The effect of exposure to long working hours on stroke: a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2020;142:105746. doi:10.1016/j.envint.2020.105746 PMID:32505015 39. Li J, Pega F, Ujita Y, Brisson C, Clays E, Descatha A, et al. The effect of exposure to long working hours on ischaemic heart disease: a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2020;142:105739. doi:10.1016/j.envint.2020.105739 PMID:32505014 40. Pachito DV, Pega F, Bakusic J, Boonen E, Clays E, Descatha A, et al. The effect of exposure to long working hours on alcohol consumption, risky drinking and alcohol use disorder: a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2021;146:106205. doi/10.1016/j.envint.2020.106205 PMID:33189992 41. Rugulies R, Sørensen K, Di Tecco C, Bonafede M, Rondinone BM, Ahn S, et al. The effect of exposure to long working hours on depression: a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-Related Burden of Disease and Injury. Environ Int. 2021;155:106629. Epub 2021 Jun 15. doi:10.1016/j.envint.2021.106629 PMID:34144478 42. Hulshof CTJ, Pega F, Neupane S, Colosio C, Daams JG, KC P, et al. The effect of occupational exposure to ergonomic risk factors on osteoarthritis of hip or knee and selected other musculoskeletal diseases: a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2021;146:106349. doi:10.1016/j.envint.2020.106349 PMID:33546919 43. Hulshof CTJ, Pega F, Neupane S, van der Molen HF, Colosio C, Daams JG, et al. The prevalence of occupational exposure to ergonomic risk factors: a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2021;146:106157. doi:10.1016/j.envint.2020.106157 PMID:33395953 44. Teixeira LR, Pega F, de Abreu W, de Almeida MS, F. dACA, Azevedo TM, et al. The prevalence of occupational exposure to noise: a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2021;154:106380. doi:10.1016/j.envint.2021.106380 45. Teixeira LR, Pega F, Dzhambov AM, Bortkiewicz A, da Silva DTC, de Andrade CAF, et al. The effect of occupational exposure to noise on ischaemic heart disease, stroke and hypertension: a systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-Related Burden of Disease and Injury. Environ Int. 2021;154:106387. doi:10.1016/j.envint.2021.106387 PMID:33612311 46. Pega F, Norris SL, Backes C, Bero LA, Descatha A, Gagliardi D, et al. RoB-SPEO: A tool for assessing risk of bias in studies estimating the prevalence of exposure to occupational risk factors from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2020;135:105039. doi:10.1016/j.envint.2019.105039 PMID:31864023 47. Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJ; Comparative Risk Assessment Collaborating Group. Selected major risk factors and global and regional burden of disease. Lancet. 2002;360:1347–60. doi:10.1016/S0140-6736(02)11403-6 PMID:12423980 48. Pega F, Náfrádi B, Momen NC, Ujita Y, Streicher KN, Prüss-Üstün AM, et al. Global, regional, and national burdens of ischemic heart disease and stroke attributable to exposure to long working hours for 194 countries, 2000-2016: a systematic analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2021;154:106595. Epub 2010 May 17. doi:10.1016/j.envint.2021.106595 PMID:34011457 49. Woodruff TJ, Sutton P. The Navigation Guide systematic review methodology: a rigorous and transparent method for translating environmental health science into better health outcomes. Environ Health Perspect. 2014;122:1007–14. doi:10.1289/ehp.1307175 PMID:24968373 50. World Population Prospects 2019, Online Edition. Geneva: United Nations; 2019 (https://population.un.org/wpp/, accessed 03 August 2021). 51. Global Health Observatory (GHO) data: Life tables. Geneva: World Health Organization; 2020 (https://www.who.int/gho/mortality_burden_ disease/life_tables/life_tables/en/, accessed 03 August 2021). 52. Leyland AH, Goldstein H. Multilevel modelling of health statistics. Chicester: Wiley; 2001. 53. Kivimäki M, Steptoe A. Effects of stress on the development and progression of cardiovascular disease. Nat Rev Cardiol. 2018;15:215–29. doi:10.1038/nrcardio.2017.189 PMID:29213140 54. Steptoe A, Kivimäki M. Stress and cardiovascular disease. Nat Rev Cardiol. 2012;9:360–70. doi:10.1038/nrcardio.2012.45 PMID:22473079 55. Hannerz H, Albertsen K, Burr H, Nielsen ML, Garde AH, Larsen AD, et al. Long working hours and stroke among employees in the general workforce of Denmark. Scand J Public Health. 2018;46:368–74. doi:10.1177/1403494817748264 PMID:29251227 56. Kivimäki M, Jokela M, Nyberg ST, Singh-Manoux A, Fransson EI, Alfredsson L. Long working hours and risk of coronary heart disease and stroke: a systematic review and meta-analysis of published and unpublished data for 603,838 individuals. Lancet. 2015;386:1739–46. doi:/10.1016/ S0140-6736(15)60295-1 PMID:26298822 26 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 57. Rose G. Incubation period of coronary heart disease. Br Med J (Clin Res Ed) 1982;284:1600–01. doi:10.1136/bmj.284.6329.1600 58. Fadel M, Sembajwe G, Gagliardi D, Pico F, Li J, Ozguler A, et al. Association between reported long working hours and history of stroke in the CONSTANCES Cohort. Stroke. 2019;50:1879–82. doi:10.1161/STROKEAHA.119.025454 PMID:31216962 59. Fadel M, Li J, Sembajwe G, Gagliardi D, Pico F, Ozguler A, et al. Cumulative exposure to long working hours and occurrence of ischemic heart disease: evidence from the CONSTANCES Cohort at inception. J Am Heart Assoc. 2020;9:e015753. doi:/10.1161/JAJA.119.015753 PMID:32476603 60. Dalboge A, Frost P, Andersen JH, Svendsen SW. Surgery for subacromial impingement syndrome in relation to intensities of occupational mechanical exposures across 10-year exposure time windows. Occup Environ Med. 2018;75:176–82. doi:/10.1136/oemed-2017-104511 PMID:28823987 61. Rushton L, Bagga S, Bevan R, Brown T, Cherrie J, Holmes P, et al. The burden of occupational cancer in Great Britain: overview report. London: Health and Safety Executive; 2012 (https://www.hse.gov.uk/research/rrhtm/rr595.htm, accessed 3 August 2021). 62. Gardner MJ, Pannett B, Winter PD, Cruddas AM. A cohort study of workers exposed to formaldehyde in the British chemical industry: an update. Br J Ind Med. 1993;50:827–34. doi:10.1136/oem.50.9.827 PMID:8398877 63. Meyers AR, Pinkerton LE, Hein MJ. Cohort mortality study of garment industry workers exposed to formaldehyde: update and internal comparisons. Am J Ind Med. 2013;56:1027–39. doi:10.1002/ajim.22199 PMID:23788124 64. Bonjour S, Adair-Rohani H, Wolf J, Bruce NG, Mehta S, Prüss-Üstün A, et al. Solid fuel use for household cooking: country and regional estimates for 1980-2010. Environ Health Perspect. 2013;121:784–90. doi:10.1289/ehp.1205987 PMID:23674502 65. Wolf J, Bonjour S, Prüss-Üstün A. An exploration of multilevel modeling for estimating access to drinking-water and sanitation. J Water Health. 2013;11:64–77. doi:10.2166/wh.2012.107 PMID:23428550 66. Labour market flow statistics in the EU. Eurostat; 2021. ( https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Labour_market_ flow_statistics_in_the_EU, accessed 3 August 2021). 67. Kuh D, Ben-Shlomo Y, Lynch J, Hallqvist J, Power C. Life course epidemiology. J Epidemiol Community Health. 2003;57:778–83. doi:10.1136/ jech.57.10.778 PMID:14573579 68. Efron B. Bootstrap methods: Another look at the jackknife. Ann Stat. 1979;7:1–26. 69. Handbook on health inequality monitoring with a special focus on low- and middle-income countries. Geneva: World Health Organization; 2013 (https://www.who.int/docs/default-source/gho-documents/health-equity/handbook-on-health-inequality-monitoring/handbook-on-health- inequality-monitoring.pdf?sfvrsn=d27f8211_2, accessed 29 July 2021). REFERENCES 27 ANNEXES ANNEX 1. GATHER (GUIDELINES FOR ACCURATE AND TRANSPARENT HEALTH ESTIMATES REPORTING) CHECKLIST OF INFORMATION THAT SHOULD BE INCLUDED IN NEW REPORTS OF GLOBAL HEALTH ESTIMATES Item no. Checklist item Page no. Objectives and funding 1 Define the indicator(s), populations (including age, sex and geographic entities) and time period(s) for which estimates were made. Abstract, 3 2 List the funding sources for the work. Abstract, 3 Data inputs: for all data inputs from multiple sources that are synthesized as part of the study 3 Describe how the data were identified and how the data were accessed. 8 4 Specify the inclusion and exclusion criteria. Identify all ad hoc exclusions. – 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. 8 6 Identify and describe any categories of input data that have potentially important biases (e.g. based on characteristics listed in item 5). 10 Data inputs: for data inputs that contribute to the analysis but were not synthesized as part of the study 7 Describe and give sources for any other data inputs. Table 7; Sections 3.2.3 and 3.2.4 Data inputs: for all data inputs 8 Provide all data inputs in a file format from which data can be efficiently extracted (e.g. a spreadsheet rather than a PDF), including all relevant meta-data listed in item 5. For any data inputs that cannot be shared because of 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. https://www.who.int/ teams/environment- climate-change-and- health/monitoring/who-ilo- joint-estimates Data analysis 9 Provide a conceptual overview of the data analysis method. A diagram may be helpful. 7–10 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). 8–9 11 Describe how candidate models were evaluated and how the final model(s) were selected. 8–9 12 Provide the results of an evaluation of model performance, if done, as well as the results of any relevant sensitivity analysis. 10, Annex 7 13 Describe methods for calculating uncertainty of the estimates. State which sources of uncertainty were, and were not, accounted for in the uncertainty analysis. 9–10 14 State how analytic or statistical source code used to generate estimates can be accessed. Pega et al. (1), supplementary file 2 28 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 Item no. Checklist item Page no. Results and discussion 15 Provide published estimates in a file format from which data can be efficiently extracted. https://www.who.int/ teams/environment- climate-change-and- health/monitoring/who-ilo- joint-estimates 16 Report a quantitative measure of the uncertainty of the estimates (e.g. uncertainty intervals). 10–13 17 Interpret results in light of existing evidence. If updating a previous set of estimates, describe the reasons for changes in estimates. 14 18 Discuss limitations of the estimates. Include a discussion of any modelling assumptions or data limitations that affect interpretation of the estimates. 14–15 Reference 1. Pega F, Nafradi B, Momen NC, Ujita Y, Streicher KN, Pruss-Ustun AM, et al. Global, regional, and national burdens of ischemic heart disease and stroke attributable to exposure to long working hours for 194 countries, 2000-2016: A systematic analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environ Int. 2021:106595. https://doi.org/10.1016/j.envint.2021.106595 PMID:34011457 ANNEXES 29 ANNEX 2. COUNTRIES, AREAS AND TERRITORIES INCLUDED WITHIN THE DIFFERENT REGIONS WHO region Countries, areas and territories included in regional grouping African Region (47) Algeria; Angola; Benin; Botswana; Burkina Faso; Burundi; Cabo Verde; Cameroon; Central African Republic; Chad; Comoros; Congo; Côte d’Ivoire; Democratic Republic of the Congo; Equatorial Guinea; Eritrea; Eswatini; Ethiopia; Gabon; Gambia; Ghana; Guinea; Guinea-Bissau; Kenya; Lesotho; Liberia; Madagascar; Malawi; Mali; Mauritania; Mauritius; Mozambique; Namibia; Niger; Nigeria; Rwanda; Sao Tome and Principe; Senegal; Seychelles; Sierra Leone; South Africa; South Sudan; Togo; Uganda; United Republic of Tanzania; Zambia; Zimbabwe Region of the Americas (36) Antigua and Barbuda; Argentina; Bahamas; Barbados; Belize; Bolivia (Plurinational State of); Brazil; Canada; Chile; Colombia; Costa Rica; Cuba; Dominicaa; Dominican Republic; Ecuador; El Salvador; Grenada; Guatemala; Guyana; Haiti; Honduras; Jamaica; Mexico; Nicaragua; Panama; Paraguay; Peru; Puerto Ricoa,b; Saint Kitts and Nevisa; Saint Lucia; Saint Vincent and the Grenadines; Suriname; Trinidad and Tobago; United States of America; Uruguay; Venezuela (Bolivarian Republic of) South-East Asia Region (11) Bangladesh; Bhutan; Democratic People’s Republic of Korea; India; Indonesia; Maldives; Myanmar; Nepal; Sri Lanka; Thailand; Timor-Leste European Region (53) Albania; Andorraa; Armenia; Austria; Azerbaijan; Belarus; Belgium; Bosnia and Herzegovina; Bulgaria; Croatia; Cyprus; Czechia; Denmark; Estonia; Finland; France; Georgia; Germany; Greece; Hungary; Iceland; Ireland; Israel; Italy; Kazakhstan; Kyrgyzstan; Latvia; Lithuania; Luxembourg; Malta; Monacoa; Montenegro; Netherlands; North Macedonia; Norway; Poland; Portugal; Republic of Moldova; Romania; Russian Federation; San Marinoa; Serbia; Slovakia; Slovenia; Spain; Sweden; Switzerland; Tajikistan; Turkey; Turkmenistan; Ukraine; United Kingdom; Uzbekistan Eastern Mediterranean Region (22) Afghanistan; Bahrain; Djibouti; Egypt; Iran (Islamic Republic of); Iraq; Jordan; Kuwait; Lebanon; Libya; Morocco; occupied Palestinian territory, including east Jerusalemb; Oman; Pakistan; Qatar; Saudi Arabia; Somalia; Sudan; Syrian Arab Republic; Tunisia; United Arab Emirates; Yemen Western Pacific Region (30) Australia; Brunei Darussalam; Cambodia; China; China, Hong Kong Special Administrative Regionb; Taiwan, Chinab; Cook Islandsa; Fiji; Japan; Kiribati; Lao People’s Democratic Republic; Malaysia; Marshall Islandsa; Micronesia (Federated States of); Mongolia; Naurua; New Zealand; Niuea; Palaua; Papua New Guinea; Philippines; Republic of Korea; Samoa; Singapore; Solomon Islands; Tokelaua; Tonga; Tuvalua; Vanuatu; Viet Nam a WHO/ILO Joint Estimates not calculated because population was < 90 000 in 2015 or because of a lack of disease burden envelope. b No burden of disease calculated. 30 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 AN N EX 3 . NU M BE R OF D EA TH S FR OM IS CH AE M IC H EA RT D IS EA SE IN 2 00 0, 2 01 0 AN D 20 16 , A ND M EA N PE RC EN TA GE C HA NG E FO R 20 00 –2 01 0, 2 01 0– 20 16 A ND 2 00 0– 20 16 , B Y W HO R EG IO N AN D SE X, 1 83 C OU NT RI ES W HO re gi on Bo th s ex es M al es Fe m al es No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) 20 00 20 10 20 16 20 00 -2 01 0 20 10 -2 01 6 20 00 -2 01 6 20 00 20 10 20 16 20 00 -2 01 0 20 10 -2 01 6 20 00 -2 01 6 20 00 20 10 20 16 20 00 -2 01 0 20 10 -2 01 6 20 00 -2 01 6 Af ric an Re gi on 37 1 39 7 (3 61 0 67 to 38 1 72 6) 44 0 27 6 (4 28 2 38 to 45 2 31 3) 51 1 58 3 (4 94 9 76 to 52 8 19 0) 18 .6 (1 4. 0 to 23 .2 ) 16 .2 (1 1. 4 to 21 .1 ) 37 .8 (3 1. 9 to 43 .9 ) 18 6 49 4 (1 79 5 51 to 19 3 43 8) 21 8 12 7 (2 09 9 50 to 22 6 30 4) 25 4 67 3 (2 43 2 70 to 26 6 07 5) 17 .0 (1 0. 9 to 23 .4 ) 16 .8 (1 0 to 23 .8 ) 36 .6 (2 8. 8 to 44 .9 ) 18 4 90 2 (1 77 2 55 to 19 2 55 0) 22 2 14 9 (2 13 3 14 to 23 0 98 3) 25 6 91 0 (2 44 8 36 to 26 8 98 5) 20 .1 (1 3. 5 to 27 .2 ) 15 .7 (8 .6 to 22 .8 ) 38 .9 (3 0. 3 to 48 .0 ) Re gi on o f th e Am er ic as 1 08 2 45 1 (1 0 62 7 48 to 1 10 2 15 3) 1 01 9 44 2 (1 0 03 0 76 to 1 03 5 80 9) 1 09 1 05 7 (1 0 72 1 47 to 1 10 9 96 8) -5 .8 (- 8. 1 to -3 .5 ) 7.0 (4 .5 to 9. 5) 0. 8 (-1 .6 to 3. 4) 55 4 60 5 (5 41 9 67 to 56 7 24 3) 54 4 72 7 (5 33 5 81 to 55 5 87 2) 60 2 47 3 (5 88 8 17 to 61 6 13 0) -1 .8 (- 4. 7 to 1. 3) 10 .6 (7 .2 to 14 .2 ) 8. 6 (5 .2 to 12 .1 ) 52 7 84 6 (5 12 7 30 to 54 2 96 1) 47 4 71 6 (4 62 7 30 to 48 6 70 1) 48 8 58 4 (4 75 5 03 to 50 1 66 5) -1 0. 1 (-1 3. 4 to -6 .5 ) 2. 9 (- 0. 8 to 6 .8 ) -7 .4 (-1 1. 0 to -3 .7 ) So ut h- Ea st As ia R eg io n 1 28 8 83 9 (1 2 25 5 07 to 1 35 2 17 1) 1 84 0 81 0 (1 7 52 5 91 to 1 92 9 02 9) 2 23 2 82 7 (2 1 09 9 56 to 2 35 5 69 8) 42 .8 (3 3. 3 to 53 .0 ) 21 .3 (1 2. 7 to 30 .5 ) 73 .2 (6 1. 2 to 86 .5 ) 73 9 59 6 (6 87 9 07 to 79 1 28 5) 1 06 7 87 7 (9 95 8 36 to 1 13 9 91 9) 1 32 6 19 2 (1 2 24 0 99 to 1 42 8 28 4) 44 .4 (3 0. 8 to 5 9. 3) 24 .2 (1 2. 4 to 37 .6 ) 79 .3 (6 1. 5 to 98 .9 ) 54 9 24 4 (5 12 6 49 to 58 5 83 9) 77 2 93 3 (7 22 0 15 to 82 3 85 0) 90 6 63 5 (8 38 2 66 to 97 5 00 5) 40 .7 (2 8. 1 to 54 .6 ) 17 .3 (5 .9 to 30 .1 ) 65 .1 (4 9. 2 to 82 .5 ) Eu ro pe an Re gi on 2 58 6 29 3 (2 5 58 7 22 to 2 61 3 86 5) 2 48 2 97 4 (2 4 54 5 08 to 2 51 1 43 9) 2 34 2 22 2 (2 3 11 3 15 to 2 37 3 12 8) -4 .0 (- 5. 5 to - 2. 5) -5 .7 (- 7.3 to -4 .0 ) -9 .4 (-1 1. 0 to -7 .9 ) 1 22 1 19 1 (1 2 04 3 64 to 1 23 8 01 8) 1 17 1 24 8 (1 1 54 4 55 to 1 18 8 04 1) 1 10 2 24 4 (1 0 84 4 48 to 1 12 0 04 0) -4 .1 (- 6. 0 to -2 .1 ) -5 .9 (- 7.9 to -3 .9 ) -9 .7 (-1 1. 6 to -7 .8 ) 1 36 5 10 3 (1 3 43 2 61 to 1 38 6 94 4) 1 31 1 72 6 (1 2 88 7 42 to 1 33 4 71 0) 1 23 9 97 8 (1 21 4 70 8 to 1 26 5 24 7) -3 .9 (- 6. 2 to -1 .7 ) -5 .5 (- 8. 0 to -2 .9 ) -9 .2 (-1 1. 5 to -6 .8 ) Ea st er n M ed ite rr an ea n Re gi on 57 3 16 8 (5 57 61 1 to 58 8 72 5) 72 2 54 9 (7 02 8 98 to 74 2 20 1) 83 3 64 3 (8 08 2 25 to 85 9 06 1) 26 .1 (2 1. 3 to 31 .0 ) 15 .4 (1 0. 7 to 20 .1 ) 45 .4 (3 9. 6 to 51 .7 ) 32 5 36 5 (3 12 9 94 to 33 7 73 6) 41 0 34 4 (3 94 71 6 to 42 5 97 2) 47 1 88 5 (4 51 7 82 to 49 1 98 8) 26 .1 (1 9. 6 to 33 .0 ) 15 .0 (8 .5 to 21 .6 ) 45 .0 (3 6. 9 to 53 .6 ) 24 7 80 3 (2 38 3 70 to 25 7 23 6) 31 2 20 5 (3 00 2 91 to 32 4 12 0) 36 1 75 8 (3 46 2 03 to 37 7 31 3) 26 .0 (1 9. 4 to 33 .0 ) 15 .9 (9 .4 to 22 .7 ) 46 .0 (3 7.9 to 54 .3 ) W es te rn Pa ci fic Re gi on 1 09 7 55 2 (1 0 40 61 6 to 1 15 4 48 8) 1 91 9 19 2 (1 8 10 1 43 to 2 02 8 24 1) 2 39 0 46 7 (2 2 34 7 83 to 2 54 6 15 2) 74 .9 (6 1. 9 to 88 .3 ) 24 .6 (1 4. 2 to 36 .0 ) 11 7.8 (1 00 .2 to 13 6. 4) 55 8 58 0 (5 18 3 35 to 59 8 82 6) 94 6 06 0 (8 71 71 0 to 1 02 0 41 0) 1 17 9 45 9 (1 0 73 5 90 to 1 28 5 32 7) 69 .4 (5 2. 2 to 88 .5 ) 24 .7 (1 0. 5 to 40 .7 ) 11 1. 2 (8 8. 0 to 13 7.0 ) 53 8 97 2 (4 98 6 97 to 57 9 24 7) 97 3 13 2 (8 93 3 59 to 1 05 2 90 5) 1 21 1 00 9 (1 0 96 8 62 to 1 32 5 15 6) 80 .6 (6 1. 9 to 10 1. 1) 24 .4 (9 .4 to 41 .3 ) 12 4. 7 (9 8. 3 to 15 2. 5) UR , u nc er ta in ty ra ng e. So ur ce : W HO G lo ba l H ea lth E st im at es ANNEXES 31 AN N EX 4 . NU M BE R OF D EA TH S FR OM S TR OK E IN 2 00 0, 2 01 0 AN D 20 16 , A ND M EA N PE RC EN TA GE C HA NG E FO R 20 00 –2 01 0, 2 01 0– 20 16 A ND 2 00 0– 20 16 , B Y W HO R EG IO N AN D SE X, 18 3 CO UN TR IE S W HO re gi on Bo th s ex es M al es Fe m al es No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) 20 00 20 10 20 16 20 00 -2 01 0 20 10 -2 01 6 20 00 -2 01 6 20 00 20 10 20 16 20 00 -2 01 0 20 10 -2 01 6 20 00 -2 01 6 20 00 20 10 20 16 20 00 -2 01 0 20 10 -2 01 6 20 00 -2 01 6 Af ric an Re gi on 28 8 86 4 (2 81 1 03 to 29 6 62 6) 32 5 11 9 (3 16 5 87 to 33 3 65 2) 36 5 79 0 (3 54 4 13 to 37 7 16 7) 12 .6 (8 .2 to 17 .0 ) 12 .5 (8 .0 to 17 .3 ) 26 .6 (2 1. 4 to 31 .9 ) 12 4 42 2 (1 19 8 01 to 12 9 04 3) 13 9 37 6 (1 34 2 37 to 14 4 51 5) 15 8 70 4 (1 51 71 3 to 16 5 69 6) 12 .0 (6 .4 to 18 .2 ) 13 .9 (7 .4 to 20 .5 ) 27 .6 (2 0. 5 to 35 .0 ) 16 4 44 2 (1 58 2 06 to 17 0 67 8) 18 5 74 3 (1 78 9 32 to 19 2 55 5) 20 7 08 6 (1 98 1 11 to 21 6 06 1) 13 .0 (7 .2 to 19 .1 ) 11 .5 (5 .4 to 18 .0 ) 25 .9 (1 8. 9 to 33 .3 ) Re gi on o f t he Am er ic as 43 6 76 4 (4 29 1 04 to 44 4 42 4) 40 8 74 4 (4 02 1 20 to 41 5 36 9) 43 4 66 6 (4 26 7 50 to 44 2 58 2) -6 .4 (- 8. 6 to -4 .2 ) 6. 3 (3 .8 to 9. 0) -0 .5 (- 2. 9 to 2. 1) 19 6 89 7 (1 92 5 05 to 20 1 28 9) 18 8 84 2 (1 84 8 49 to 19 2 83 5) 20 3 92 6 (1 99 1 19 to 20 8 73 3) -4 .1 (- 7.0 to -1 .1 ) 8. 0 (4 .6 to 11 .5 ) 3. 6 (0 .3 to 7.0 ) 23 9 86 8 (2 33 5 92 to 24 6 14 4) 21 9 90 3 (2 14 61 7 to 22 5 18 8) 23 0 74 0 (2 24 4 51 to 23 7 02 9) -8 .3 (-1 1. 5 to -5 .0 ) 4. 9 (1 .3 to 8. 8) -3 .8 (- 7.4 to -0 .1 ) So ut h- Ea st As ia R eg io n 95 2 32 4 (9 07 4 57 to 99 7 19 0) 1 13 2 81 8 (1 0 82 4 32 to 1 18 3 20 5) 1 24 7 15 7 (1 1 85 2 83 to 1 30 9 03 2) 19 .0 (1 1. 5 to 26 .9 ) 10 .1 (2 .8 to 17 .4 ) 31 .0 (2 2. 5 to 40 .3 ) 49 0 77 5 (4 59 1 37 to 52 2 41 3) 59 8 34 6 (5 62 1 47 to 63 4 54 4) 67 5 66 5 (6 29 6 50 to 72 1 67 9) 21 .9 (1 1. 7 to 33 .4 ) 12 .9 (3 .1 to 23 .6 ) 37 .7 (2 4. 9 to 50 .8 ) 46 1 54 8 (4 29 7 36 to 49 3 36 0) 53 4 47 3 (4 99 4 23 to 56 9 52 3) 57 1 49 3 (5 30 1 27 to 61 2 85 8) 15 .8 (5 .1 to 27 .3 ) 6. 9 (- 3. 3 to 18 .0 ) 23 .8 (1 2. 1 to 36 .8 ) Eu ro pe an Re gi on 1 32 9 76 7 (1 3 12 1 18 to 13 47 41 6) 1 10 9 31 6 (1 0 94 3 51 to 1 12 4 28 2) 98 5 95 2 (9 70 5 61 to 1 00 1 34 3) -1 6. 6 (-1 8. 1 to -1 5. 0) -1 1. 1 (-1 3. 0 to -9 .3 ) -2 5. 9 (- 27 .4 to -2 4. 4) 52 4 74 7 (5 15 7 75 to 53 3 71 8) 45 5 50 7 (4 47 7 80 to 46 3 23 4) 41 0 06 0 (4 02 2 34 to 41 7 88 7) -13 .2 (-1 5. 2 to -1 1. 1) -1 0. 0 (-1 2. 2 to -7 .7 ) -2 1. 9 (- 23 .8 to -1 9. 8) 80 5 02 0 (7 89 8 22 to 82 0 21 9) 65 3 80 9 (6 40 9 93 to 66 6 62 5) 57 5 89 2 (5 62 6 39 to 58 9 14 4) -1 8. 8 (- 21 .0 to -1 6. 6) -1 1. 9 (-1 4. 5 to -9 .2 ) -2 8. 5 (- 30 .6 to -2 6. 3) Ea st er n M ed ite rr an ea n Re gi on 24 5 02 1 (2 37 1 11 to 25 2 93 0) 29 3 15 4 (2 83 71 6 to 30 2 59 2) 32 3 36 8 (3 11 5 64 to 33 5 17 1) 19 .6 (1 4. 4 to 25 .3 ) 10 .3 (5 .1 to 15 .9 ) 32 .0 (2 5. 8 to 38 .5 ) 12 2 73 4 (1 16 9 03 to 12 8 56 5) 14 4 99 1 (1 38 3 58 to 15 1 62 3) 15 9 75 0 (1 51 5 14 to 16 7 98 5) 18 .1 (1 0. 7 to 26 .2 ) 10 .2 (2 .9 to 18 .2 ) 30 .2 (2 1. 3 to 39 .6 ) 12 2 28 7 (1 16 9 41 to 12 7 63 2) 14 8 16 3 (1 41 4 48 to 15 4 87 8) 16 3 61 8 (1 55 1 62 to 17 2 07 4) 21 .2 (1 3. 7 to 29 .1 ) 10 .4 (3 .1 to 18 .3 ) 33 .8 (2 5. 0 to 43 .3 ) W es te rn Pa ci fic Re gi on 1 86 9 98 8 (1 7 55 7 20 to 1 98 4 25 6) 2 13 4 25 5 (2 0 05 0 39 to 2 26 3 47 0) 2 39 0 35 6 (2 2 27 1 07 to 2 55 3 60 4) 14 .1 (4 .7 to 24 .5 ) 12 .0 (2 .4 to 22 .8 ) 27 .8 (1 6. 7 to 39 .8 ) 96 5 22 6 (8 79 7 82 to 1 05 0 66 9) 1 11 5 95 6 (1 0 18 1 78 to 1 21 3 73 4) 1 26 6 69 3 (1 1 42 0 81 to 1 39 1 30 6) 15 .6 (2 .2 to 30 .8 ) 13 .5 (- 0. 4 to 29 .5 ) 31 .2 (1 4. 7 to 49 .5 ) 90 4 76 2 (8 28 8 91 to 98 0 63 4) 1 01 8 29 9 (9 33 8 23 to 1 10 2 77 4) 1 12 3 66 2 (1 0 18 2 02 to 1 22 9 12 3) 12 .6 (0 .1 to 26 .8 ) 10 .4 (- 2. 6 to 25 .0 ) 24 .2 (9 .2 to 41 .2 ) UR , u nc er ta in ty ra ng e. So ur ce : W HO G lo ba l H ea lth E st im at es 32 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 AN N EX 5 . PR OP OR TI ON O F PO PU LA TI ON E XP OS ED TO L ON G W OR KI NG H OU RS (≥  5 5  HO UR S PE R W EE K) F OR 2 00 0, 2 01 0 AN D 20 16 , A ND M EA N PE RC EN TA GE C HA NG E FO R 20 00 –2 01 0, 2 01 0– 20 16 AN D 20 00 –2 01 6, B Y W HO R EG IO N AN D SE X, 1 83 C OU NT RI ES W HO re gi on Bo th s ex es M al es Fe m al es No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) No . d ea th s (9 5% U R) Pe rc en t c ha ng e (9 5% U R) 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 00 –2 01 6 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 00 –2 01 6 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 10 –2 01 6 Af ric an Re gi on 17 .3 (1 6. 9 to 17 .7 ) 13 .5 (1 3. 3 to 13 .6 ) 11 .4 (1 1. 2 to 11 .5 ) −2 2. 3 (− 24 .2 to −2 0. 4) −1 5. 6 (− 17 .3 to −1 3. 9) −3 4. 4 (− 36 .1 to −3 2. 6) 21 .5 (2 0. 9 to 22 .1 ) 17 .4 (1 7.1 to 17 .7 ) 15 .3 (1 5. 0 to 1 5. 6) −1 8. 9 ( −2 1. 4 to −1 6. 2) −1 2. 4 ( −1 4. 6 to −1 0. 1) −2 8. 9 ( −3 1. 3 to −2 6. 5) 13 .3 (1 2. 8 to 13 .7 ) 9. 6 (9 .4 to 9. 8) 7.5 (7 .3 to 7.7 ) −2 7.7 (− 30 .4 to −2 4. 7) −2 1. 5 (− 24 .1 to −1 8. 8) −4 3. 2 (− 45 .6 to −4 0. 8) Re gi on o f t he Am er ic as 8. 8 (8 .7 to 8 .9 ) 7.3 (7 .2 to 7.4 ) 6. 4 (6 .3 to 6. 6) −1 7. 2 (− 19 .0 to −1 5. 4) −1 2. 1 (− 14 .9 to −9 .2 ) −2 7. 2 (− 29 .3 to −2 5. 1) 13 .2 (1 3. 1 to 13 .4 ) 10 .8 (1 0. 6 to 11 .0 ) 9. 4 (9 .1 to 9. 7) −1 8. 4 (− 20 .2 to −1 6. 5) −1 2. 9 (− 16 .0 to −9 .8 ) −2 8. 9 (− 31 .1 to −2 6. 7) 4. 6 (4 .5 to 4. 8) 4. 0 (3 .8 to 4. 1) 3. 6 (3 .4 to 3 .8 ) −1 4. 3 (− 18 .5 to −1 0. 0) −9 .8 (− 15 .9 to −3 .6 ) −2 2. 7 (− 27 .3 to −1 7. 9) So ut h- Ea st As ia R eg io n 11 .5 (1 0. 5 to 12 .5 ) 11 .7 (1 1. 0 to 12 .3 ) 11 .7 (1 0. 8 to 12 .5 ) 1. 4 (− 8. 6 to 12 .8 ) 0. 1 ( −9 .2 to 10 .0 ) 1. 5 (− 9. 5 to 13 .9 ) 17 .9 (1 6. 2 to 19 .6 ) 18 .2 (1 7.0 to 19 .3 ) 18 .1 (1 6. 5 to 19 .6 ) 1. 5 (− 9. 2 to 14 .1 ) −0 .4 (− 10 .8 to 10 .7 ) 1. 1 (− 11 .0 to 14 .8 ) 4. 9 (3 .9 to 5. 9) 4. 9 (4 .2 to 5 .6 ) 5. 0 (4 .2 to 5 .8 ) 0. 8 (− 21 .2 to 31 .3 ) 2. 2 (− 18 .0 to 26 .8 ) 3. 0 (− 19 .7 to 35 .1 ) Eu ro pe an Re gi on 4. 6 (4 .6 to 4. 6) 3. 9 (3 .9 to 4. 0) 3. 5 (3 .5 to 3. 6) −1 4. 8 (− 16 .1 to −1 3. 6) −9 .8 (− 11 .4 to −8 .1 ) −2 3. 1 (− 24 .5 to −2 1. 8) 7.5 (7 .4 to 7.5 ) 6. 2 (6 .1 to 6 .2 ) 5. 4 (5 .3 to 5. 5) −1 7. 5 ( −1 8. 9 to −1 5. 9) −1 1. 9 (− 13 .9 to −9 .9 ) −2 7. 3 (− 28 .8 to −2 5. 7) 2. 0 (2 .0 to 2 .0 ) 1. 9 (1 .8 to 1. 9) 1. 8 (1 .8 to 1. 9) −6 .5 (− 8. 8 to −4 .1 ) −3 .5 (− 6. 3 to −0 .7 ) −9 .8 ( −1 2. 3 to −7 .3 ) Ea st er n M ed ite rr an ea n Re gi on 9. 8 (9 .6 to 10 .0 ) 10 .7 (1 0. 5 to 10 .9 ) 11 .4 (1 1. 1 to 11 .6 ) 9. 4 (5 .9 to 13 .0 ) 5. 9 (2 .8 to − 9. 2) 15 .9 (1 2. 0 to 20 .0 ) 17 .4 (1 7.0 to 17 .9 ) 19 .5 (1 9. 1 to 20 .0 ) 21 .0 (2 0. 5 to 21 .5 ) 12 .1 (8 .4 to 16 .0 ) 7.5 (4 .0 to 11 .0 ) 20 .4 (1 6. 2 to 24 .8 ) 1. 8 (1 .7 to 1. 9) 1. 3 (1 .3 to 1. 4) 1. 0 (1 .0 to 1. 1) −2 6. 4 (− 32 .5 to −1 9. 5) −2 1. 3 (− 27 .3 to −1 5. 0) −4 2. 1 ( −4 7. 3 to −3 6. 3) W es te rn Pa ci fic Re gi on 3. 9 (3 .5 to 4. 4) 5. 9 (5 .5 to 6 .4 ) 8. 4 (7 .8 to 9. 0) 51 .7 (3 2. 1 to 74 .9 ) 42 .0 (2 7.8 to 57 .9 ) 11 5. 4 (8 8. 7 to 14 8. 3) 4. 5 (3 .9 to 5. 2) 7.4 (6 .6 to 8. 1) 11 .1 (1 0. 1 to 12 .1 ) 62 .8 (3 7.6 to 9 5. 8) 50 .6 (3 2. 1 to 72 .0 ) 14 5. 2 (1 07 .9 to 19 2. 5) 3. 3 (2 .6 to 3 .9 ) 4. 4 (3 .9 to 5. 0) 5. 7 ( 5. 0 to 6 .3 ) 36 .0 (8 .6 to 74 .9 ) 27 .6 (7 .7 to 51 .7 ) 73 .5 (3 9. 9 to 12 0. 4) UR , u nc er ta in ty ra ng e. ANNEXES 33 AN N EX 6 . PO PU LA TI ON A TT RI BU TA BL E FR AC TI ON S FO R IS CH AE M IC H EA RT D IS EA SE D EA TH S AN D DA LY S AT TR IB UT AB LE TO E XP OS UR E TO L ON G W OR KI NG H OU RS (≥  5 5  HO UR S PE R W EE K) F OR 2 00 0, 20 10 A ND 2 01 6, A ND M EA N PE RC EN TA GE C HA NG E FO R 20 00 –2 01 0, 2 01 0– 20 16 A ND 2 00 0– 20 16 , B Y W HO R EG IO N AN D SE X , 1 83 C OU NT RI ES W HO re gi on Bo th s ex es M al es Fe m al es PA F (% ) ( 95 % UR ) Pe rc en t c ha ng e (9 5% U R) PA F (% ) ( 95 % UR ) Pe rc en t c ha ng e (9 5% U R) PA F (% ) ( 95 % UR ) Pe rc en t c ha ng e (9 5% U R) 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 00 –2 01 6 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 00 –2 01 6 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 00 –2 01 6 De at hs Af ric an Re gi on 3. 4 (3 .2 to 3 .6 ) 3. 3 (3 .1 to 3 .6 ) 3. 3 (3 .1 to 3 .5 ) −2 .4 (− 10 .8 to 6 .7 ) −0 .6 (− 9. 7 to 9 .4 ) 2. 9 (− 11 .7 to 6 .9 ) 4. 6 (4 .2 to 4 .9 ) 4. 6 (4 .2 to 4 .9 ) 4. 6 (4 .2 to 5 .0 ) 0. 0 (− 11 .0 to 1 2. 6) 0. 6 (− 11 .0 to 1 3. 5) 0. 6 (− 11 .3 to 1 3. 7) 2. 3 (2 .1 to 2 .5 ) 2. 1 (1 .9 to 2 .3 ) 2. 1 (1 .8 to 2 .3 ) −5 .8 (− 17 .8 to 8 .4 ) −3 .5 (− 16 .9 to 1 1. 8) −9 .1 (− 21 .6 to 5 .0 ) Re gi on o f t he Am er ic as 2. 2 (2 .0 to 2 .3 ) 2. 3 (2 .1 to 2 .5 ) 2. 3 (2 .1 to 2 .5 ) 6. 0 (− 5. 8 to 1 8. 9) 0. 5 (− 10 .4 to 1 2. 4) 6. 4 (5 .2 to 1 9. 5) 3. 6 (3 .3 to 4 .0 ) 3. 7 (3 .4 to 4 .1 ) 3. 6 (3 .3 to 3 .9 ) 1. 7 (− 11 .2 to 1 6. 0) −2 .9 (− 14 .8 to 1 0. 4) −1 .2 (− 13 .6 to 1 3. 4) 0. 6 (0 .5 to 0 .7 ) 0. 7 (0 .6 to 0 .7 ) 0. 7 (0 .6 to 0 .8 ) 10 .5 (− 3. 6 to 2 6. 9) 4. 8 (− 8. 0 to 1 9. 3) 15 .8 (1 .2 to 3 3. 1) So ut h- Ea st A si a Re gi on 7.2 (6 .2 to 8 .4 ) 7.2 (6 .1 to 8 .3 ) 7.2 (6 .0 to 8 .4 ) −0 .2 (− 19 .9 to 2 4. 5) 0. 7 (− 20 .4 to 2 4. 9) −0 .9 (− 20 .9 to 2 3. 7) 9. 3 (7 .5 to 1 1. 2) 9. 3 (7 .6 to 1 1. 2) 9. 2 (7 .4 to 1 1. 1) 0. 0 (− 24 .7 to 3 3. 2) −1 .0 (− 25 .8 to 3 2. 0) −1 .0 (− 26 .2 to 3 2. 8) 4. 4 (3 .5 to 5 .3 ) 4. 3 (3 .4 to 5 .1 ) 4. 1 (3 .3 to 5 .0 ) −2 .4 (− 26 .5 to 2 9. 9) −3 .9 (− 28 .9 to 2 8. 7) −6 .2 (− 30 .7 to 2 6. 1) Eu ro pe an Re gi on 1. 8 (1 .6 to 1 .9 ) 1. 6 (1 .4 to 1 .7 ) 1. 4 (1 .3 to 1 .5 ) −1 1. 5 (− 21 .0 to − 1. 2) −8 .2 (− 18 .0 to 2 .5 ) 18 .7 (− 27 .1 to − 9. 6) 3. 1 (2 .8 to 3 .4 ) 2. 7 (2 .4 to 2 .9 ) 2. 5 (2 .3 to 2 .7 ) −1 2. 6 (− 23 .6 to − 0. 3) −8 .0 (− 19 .5 to 4 .9 ) −1 9. 6 (− 29 .4 to − 8. 4) 0. 6 (0 .5 to 0 .6 ) 0. 6 (0 .5 to 0 .6 ) 0. 5 (0 .5 to 0 .6 ) −6 .1 (− 20 .5 to 0 .4 ) −7 .9 (− 21 .7 to 8 .9 ) −1 3. 5 (− 26 .0 to 1 .4 ) Ea st er n M ed ite rr an ea n Re gi on 6. 0 (5 .5 to 6 .5 ) 5. 7 (5 .2 to 6 .2 ) 5. 5 (5 .0 to 6 .1 ) −5 .1 ( −1 6. 5 to 7. 8) −2 .5 (− 14 .9 to 1 1. 4) 7.5 ( −1 9. 2 to 5 .4 ) 8. 4 (7 .5 to 9 .3 ) 8. 1 (7 .2 to 9 .0 ) 7.9 (7 .0 to 8 .9 ) −3 .6 (− 17 .9 to 1 2. 9) −1 .7 (− 16 .6 to 1 5. 9) −5 .3 (− 19 .7 to 1 1. 4) 2. 9 (2 .5 to 3 .2 ) 2. 6 (2 .3 to 2 .9 ) 2. 4 (2 .1 to 2 .8 ) −1 0. 7 (− 24 .7 to 5 .5 ) −4 .8 (− 20 .4 to 1 3. 5) −1 5. 0 (− 28 .6 to 0 .7 ) W es te rn P ac ifi c Re gi on 3. 3 (2 .8 to 3 .8 ) 2. 8 (2 .4 to 3 .3 ) 2. 7 (2 .3 to 3 .2 ) −1 4. 5 (− 30 .9 to 5 .6 ) −2 .9 (− 23 .2 to 2 2. 9) −1 7. 0 (− 33 .9 to 3 .2 ) 4. 4 (3 .6 to 5 .2 ) 3. 7 (3 .0 to 4 .5 ) 3. 6 (2 .8 to 4 .4 ) −1 4. 6 (− 35 .6 to 1 3. 8) −3 .8 (− 30 .4 to 3 0. 8) −1 7. 8 (− 39 .8 to 9 .8 ) 2. 2 (1 .7 to 2 .6 ) 1. 9 (1 .5 to 2 .4 ) 1. 9 (1 .4 to 2 .4 ) −1 1. 5 (− 35 .7 to 2 0. 1) −1 .4 (− 30 .9 to 3 9. 1) −1 2. 7 (− 38 .4 to 2 0. 2) DA LY s Af ric an Re gi on 3. 8 (3 .6 to 4 .0 ) 3. 8 (3 .6 to 4 .0 ) 3. 9 (3 .7 to 4 .1 ) 0. 1 (− 6. 7 to 7. 4) 1. 0 ( −5 .7 to 8 .3 ) 1. 2 (− 5. 5 to 8 .3 ) 5. 0 (4 .7 to 5 .3 ) 5. 1 (4 .8 to 5 .4 ) 5. 1 (4 .8 to 5 .5 ) 1. 6 (− 7. 0 to 1 1. 1) 1. 6 (− 6. 7 to 1 0. 9) 3. 3 (− 5. 2 to 1 2. 7) 2. 4 (2 .3 to 2 .6 ) 2. 4 (2 .2 to 2 .5 ) 2. 3 (2 .2 to 2 .5 ) −2 .3 (− 11 .8 to 8 .3 ) −1 .3 (− 11 .1 to 9 .3 ) −3 .6 (− 12 .8 to 6 .9 ) Re gi on o f t he Am er ic as 3. 3 (3 .0 to 3 .6 ) 3. 4 (3 .2 to 3 .7 ) 3. 4 (3 .1 to 3 .7 ) 4. 8 ( −6 .2 to 1 7. 6) −1 .4 (− 11 .4 to 9 .9 ) 3. 4 ( −7 .9 to 1 5. 7) 4. 9 (4 .4 to 5 .3 ) 5. 0 (4 .5 to 5 .4 ) 4. 8 (4 .4 to 5 .2 ) 2. 1 (− 10 .3 to 1 6. 6) −3 .3 (− 14 .7 to 9 .6 ) −1 .2 (− 13 .4 to 1 2. 5) 1. 0 (0 .9 to 1 .1 ) 1. 1 (1 .0 to 1 .2 ) 1. 1 (1 .0 to 1 .2 ) 7.2 (− 5. 3 to 2 1. 9) 1. 3 (− 10 .1 to 1 4. 1) 8. 5 (− 4. 2 to 2 3. 2) So ut h- Ea st A si a Re gi on 8. 3 (7 .2 to 9 .4 ) 8. 4 (7 .3 to 9 .6 ) 8. 5 (7 .3 to 9 .7 ) 1. 6 (− 15 .8 to 2 3. 0) 0. 6 (− 16 .9 to 21 .7 ) 2. 2 (− 15 .9 to 2 3. 6) 10 .1 (8 .4 to 1 1. 8) 10 .2 (8 .5 to 1 1. 9) 10 .2 (8 .4 to 1 2. 0) 0. 9 (− 20 .6 to 2 7.8 ) 0. 0 (− 22 .1 to 2 8. 0) 0. 9 (− 21 .5 to 2 9. 3) 5. 5 (4 .5 to 6 .4 ) 5. 6 (4 .6 to 6 .6 ) 5. 5 (4 .5 to 6 .5 ) 1. 5 (− 20 .8 to 2 9. 7) −1 .2 (− 22 .9 to 2 7.4 ) 0. 3 (− 22 .1 to 2 8. 4) Eu ro pe an Re gi on 2. 6 (2 .4 to 2 .8 ) 2. 4 (2 .2 to 2 .6 ) 2. 3 (2 .1 to 2 .5 ) −8 .9 (− 18 .7 to 1 .7 ) −4 .1 (− 13 .7 to 6 .8 ) −1 2. 7 (− 21 .4 to − 2. 4) 3. 9 (3 .5 to 4 .2 ) 3. 5 (3 .2 to 3 .8 ) 3. 4 (3 .1 to 3 .7 ) −9 .2 (− 20 .0 to 3 .3 ) −4 .2 (− 15 .3 to 8 .5 ) −1 3. 0 (− 23 .3 to − 1. 3) 0. 9 (0 .8 to 1 .0 ) 0. 9 (0 .8 to 1 .0 ) 0. 8 (0 .8 to 0 .9 ) −4 .1 (− 17 .2 to 1 1. 3) −3 .3 (− 16 .2 to 1 2. 1) −7 .2 (− 19 .4 to 7. 1) Ea st er n M ed ite rr an ea n Re gi on 6. 9 (6 .4 to 7. 4) 6. 8 (6 .3 to 7. 4) 6. 9 (6 .3 to 7. 4) −0 .7 (− 10 .8 to 1 0. 5) 0. 2 (− 10 .7 to 1 2. 5) −0 .5 (− 10 .9 to 1 0. 9) 9. 1 (8 .3 to 9 .9 ) 9. 1 (8 .2 to 1 0. 0) 9. 2 (8 .3 to 1 0. 0) −0 .1 (− 12 .7 to 1 3. 8) 0. 6 (− 12 .7 to 1 5. 7) 0. 6 (− 12 .1 to 1 5. 0) 3. 6 (3 .2 to 3 .9 ) 3. 4 (3 .0 to 3 .7 ) 3. 3 (3 .0 to 3 .7 ) −5 .1 (− 17 .6 to 9 .0 ) −1 .5 (− 15 .5 to 1 4. 5) −6 .6 (− 19 .5 to 7. 9) W es te rn P ac ifi c Re gi on 4. 8 (4 .2 to 5 .3 ) 4. 3 (3 .8 to 4 .9 ) 4. 3 (3 .7 to 4 .9 ) −9 .4 (− 24 .5 to 7. 6) −1 .3 (− 19 .2 to 2 0. 2) −1 0. 6 (− 25 .8 to 7. 6) 6. 0 (5 .1 to 6 .9 ) 5. 5 (4 .6 to 6 .4 ) 5. 4 (4 .4 to 6 .4 ) −8 .6 (− 27 .7 to 1 5. 1) −1 .7 (− 23 .7 to 2 5. 5) −1 0. 1 (− 30 .1 to 1 3. 6) 3. 3 (2 .7 to 3 .8 ) 3. 0 (2 .4 to 3 .6 ) 3. 0 (2 .4 to 3 .6 ) −8 .2 (− 30 .1 to 1 8. 4) −0 .1 (− 24 .9 to 3 3. 5) −8 .3 (− 30 .2 to 1 9. 8) DA LY s, d is ab ili ty -a dj us te d lif e ye ar s; P AF , p op ul at io n at tr ib ut ab le fr ac tio n; U R, u nc er ta in ty ra ng e. 34 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016 AN N EX 7 . PO PU LA TI ON A TT RI BU TA BL E FR AC TI ON S FO R ST RO KE D EA TH S AN D DA LY S AT TR IB UT AB LE TO E XP OS UR E TO L ON G W OR KI NG H OU RS (≥  5 5  HO UR S PE R W EE K) F OR 2 00 0, 2 01 0 AN D 20 16 , AN D M EA N PE RC EN TA GE C HA NG E FO R 20 00 –2 01 0, 2 01 0– 20 16 A ND 2 00 0– 20 16 , B Y W HO R EG IO N AN D SE X , 1 83 C OU NT RI ES W HO re gi on Bo th s ex es M al es Fe m al es PA F (% ) ( 95 % UR ) Pe rc en t c ha ng e (9 5% U R) PA F (% ) ( 95 % UR ) Pe rc en t c ha ng e (9 5% U R) PA F (% ) ( 95 % UR ) Pe rc en t c ha ng e (9 5% U R) 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 00 –2 01 6 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 00 –2 01 6 20 00 20 10 20 16 20 00 –2 01 0 20 10 –2 01 6 20 00 –2 01 6 De at hs Af ric an Re gi on 6. 2 (5 .9 to 6 .5 ) 6. 3 (5 .9 to 6 .6 ) 6. 3 (5 .9 to 6 .7 ) 0. 9 ( −6 .5 to 9 .0 ) 0. 6 (− 7.4 to 8 .9 ) 1. 5 (6 .4 to 1 0. 2) 8. 6 (8 .0 to 9 .3 ) 8. 8 (8 .2 to 9 .5 ) 9. 0 (8 .3 to 9 .7 ) 2. 6 (− 7.7 to 1 3. 9) 1. 4 (− 9. 4 to 1 3. 1) 4. 0 (− 7.1 to 1 6. 1) 4. 4 (4 .0 to 4 .7 ) 4. 3 (4 .0 to 4 .7 ) 4. 2 (3 .9 to 4 .6 ) −1 .3 (− 12 .0 to 1 0. 6) −1 .7 (− 13 .2 to 1 0. 9) −3 .0 (− 14 .4 to 9 .6 ) Re gi on o f t he Am er ic as 4. 5 (4 .2 to 4 .8 ) 4. 4 (4 .1 to 4 .7 ) 4. 2 (3 .9 to 4 .5 ) −2 .7 (− 11 .7 to 7. 1) −3 .6 (− 12 .0 to 5 .7 ) 6. 2 (− 14 .7 to 3 .1 ) 8. 0 (7 .3 to 8 .6 ) 7.5 (6 .9 to 8 .1 ) 7.2 (6 .6 to 7 .7 ) −5 .2 (− 15 .4 to 6 .4 ) −5 .0 (− 14 .8 to 6 .3 ) −9 .9 (− 19 .7 to 1 .3 ) 1. 6 (1 .5 to 1 .8 ) 1. 6 (1 .5 to 1 .8 ) 1. 6 (1 .5 to 1 .7 ) −0 .1 (− 11 .5 to 1 2. 5) −2 .6 (− 13 .1 to 9 .2 ) −2 .7 (− 13 .6 to 9 .5 ) So ut h- Ea st A si a Re gi on 12 .6 (1 1. 1 to 1 4. 1) 12 .7 (1 1. 3 to 1 4. 1) 12 .8 (1 1. 3 to 1 4. 3) 0. 7 (− 14 .7 to 1 8. 9) 0. 6 (− 14 .5 to 1 8. 3) 1. 3 (− 14 .9 to 2 0. 4) 17 .0 (1 4. 4 to 1 9. 8) 17 .1 (1 4. 7 to 1 9. 6) 17 .1 (1 4. 5 to 1 9. 8) 0. 5 (− 18 .7 to 2 4. 4) −0 .2 (− 19 .5 to 2 3. 2) 0. 3 (− 19 .8 to 2 5. 7) 7.9 (6 .5 to 9 .3 ) 7.7 (6 .4 to 9 .1 ) 7.6 (6 .4 to 9 .0 ) −2 .1 (− 24 .2 to 2 6. 3) −1 .0 (− 23 .0 to 2 5. 8) −3 .0 (− 24 .0 to 2 4. 5) Eu ro pe an Re gi on 2. 8 (2 .6 to 3 .0 ) 2. 7 (2 .5 to 2 .9 ) 2. 5 (2 .3 to 2 .6 ) −5 .2 (− 14 .7 to 5 .3 ) −7 .5 (− 16 .1 to 1 .9 ) 12 .4 (− 20 .9 to − 2. 5) 5. 2 (4 .7 to 5 .8 ) 4. 9 (4 .5 to 5 .3 ) 4. 5 (4 .1 to 4 .9 ) −7 .1 (− 19 .0 to 6 .4 ) −7 .9 (− 18 .5 to 3 .8 ) −1 4. 5 (− 25 .0 to − 2. 2) 1. 2 (1 .1 to 1 .3 ) 1. 1 (1 .0 to 1 .2 ) 1. 0 (0 .9 to 1 .1 ) −8 .0 (− 20 .3 to 6 .3 ) −9 .2 (− 20 .3 to 3 .7 ) −1 6. 5 (− 27 .2 to − 3. 4) Ea st er n M ed ite rr an ea n Re gi on 10 .4 (9 .5 to 1 1. 4) 9. 8 (8 .9 to 1 0. 6) 9. 5 (8 .6 to 1 0. 4) −6 .2 ( −1 7.1 to 6 .1 ) 2. 3 ( −1 4. 1 to 1 1. 1) −8 .4 (− 19 .7 to 4 .5 ) 15 .1 (1 3. 4 to 1 6. 9) 14 .5 (1 2. 9 to 1 6. 2) 14 .2 (1 2. 6 to 1 5. 9) −4 .4 (− 18 .4 to 12 .7 ) −1 .8 (− 16 .9 to 1 5. 0) −6 .1 (− 20 .6 to 1 1. 1) 5. 7 (5 .0 to 6 .3 ) 5. 1 (4 .5 to 5 .8 ) 5. 0 (4 .3 to 5 .6 ) −9 .1 (− 23 .6 to 7. 5) −3 .5 (− 19 .7 to 1 6. 0) −1 2. 3 (− 26 .8 to 4 .5 ) W es te rn P ac ifi c Re gi on 6. 1 (5 .2 to 7 .2 ) 6. 0 (5 .0 to 6 .9 ) 6. 0 (5 .0 to 7 .1 ) −3 .1 (− 23 .3 to 2 2. 3) 0. 7 (− 20 .8 to 2 7.0 ) −2 .4 (− 24 .0 to 2 4. 1) 7.6 (6 .0 to 9 .4 ) 7.2 (5 .7 to 8 .8 ) 7.2 (5 .6 to 9 .0 ) −5 .4 (− 31 .2 to 3 0. 5) −0 .6 (− 29 .2 to 3 8. 6) −6 .0 (− 33 .0 to 3 0. 6) 4. 5 (3 .6 to 5 .6 ) 4. 6 (3 .6 to 5 .6 ) 4. 7 (3 .6 to 5 .8 ) 0. 3 (− 27 .5 to 3 7. 6) 2. 1 (− 27 .5 to 4 1. 6) 2. 4 (− 26 .9 to 4 2. 6) DA LY s Af ric an Re gi on 6. 8 (6 .5 to 7. 1) 7.0 (6 .7 to 7. 2) 7.1 (6 .8 to 7. 3) 2. 6 ( −2 .7 to 8 .1 ) 1. 3 (− 4. 0 to 6 .9 ) 4. 0 (− 1. 5 to 9 .8 ) 9. 3 (8 .8 to 9 .7 ) 9. 6 (9 .1 to 1 0. 0) 9. 8 (9 .3 to 1 0. 2) 3. 3 (− 3. 7 to 1 0. 9) 2. 0 (− 5. 0 to 9 .3 ) 5. 4 (− 1. 9 to 1 3. 2) 4. 7 (4 .5 to 5 .0 ) 4. 8 (4 .5 to 5 .1 ) 4. 8 (4 .5 to 5 .0 ) 1. 4 (− 6. 1 to 9 .3 ) −0 .5 (− 8. 3 to 7. 9) 0. 9 (− 7. 0 to 9 .5 ) Re gi on o f t he Am er ic as 6. 2 (5 .8 to 6 .6 ) 5. 9 (5 .6 to 6 .3 ) 5. 8 (5 .4 to 6 .1 ) −3 .9 (− 12 .0 to 4 .8 ) −2 .9 (− 10 .6 to 5 .5 ) −6 .7 (− 14 .3 to 1 .9 ) 9. 9 (9 .1 to 1 0. 7) 9. 4 (8 .8 to 1 0. 1) 9. 1 (8 .4 to 9 .7 ) −4 .8 (− 14 .4 to 6 .2 ) −4 .0 (− 13 .2 to 6 .3 ) −8 .6 (− 17 .7 to 1 .7 ) 2. 5 (2 .3 to 2 .7 ) 2. 4 (2 .3 to 2 .6 ) 2. 4 (2 .2 to 2 .5 ) −3 .2 (− 13 .0 to 7. 8) −2 .7 (− 12 .0 to 7. 5) −5 .9 (− 15 .0 to 4 .1 ) So ut h- Ea st A si a Re gi on 14 .4 (1 3. 1 to 1 5. 7) 14 .7 (1 3. 5 to 1 6. 0) 14 .9 (1 3. 6 to 1 6. 2) 2. 1 (− 10 .2 to 1 5. 9) 1. 2 (− 10 .5 to 1 4. 6) 3. 3 (− 9. 2 to 1 7.4 ) 18 .1 (1 6. 0 to 2 0. 3) 18 .4 (1 6. 3 to 2 0. 4) 18 .4 (1 6. 4 to 2 0. 5) 1. 3 (− 13 .9 to 1 9. 2) 0. 5 (− 14 .5 to 1 8. 2) 1. 8 (− 13 .8 to 1 9. 2) 10 .1 (8 .7 to 1 1. 5) 10 .2 (8 .8 to 1 1. 5) 10 .2 (8 .9 to 1 1. 5) 0. 9 (− 17 .0 to 2 3. 0) 0. 1 (− 17 .1 to 1 9. 9) 1. 1 (− 16 .7 to 21 .5 ) Eu ro pe an Re gi on 4. 1 (3 .8 to 4 .4 ) 4. 0 (3 .7 to 4 .2 ) 3. 8 (3 .6 to 4 .1 ) −2 .6 (− 11 .8 to 7. 7) −3 .8 (− 12 .2 to 5 .6 ) −6 .3 (− 14 .9 to 3 .3 ) 6. 7 (6 .0 to 7. 3) 6. 4 (5 .8 to 6 .9 ) 6. 1 (5 .6 to 6 .5 ) −4 .5 (− 15 .4 to 8 .4 ) −4 .5 (− 15 .0 to 7. 2) −8 .8 (− 19 .3 to 3 .1 ) 1. 9 (1 .7 to 2 .1 ) 1. 8 (1 .6 to 1 .9 ) 1. 7 (1 .6 to 1 .9 ) −4 .0 (− 15 .9 to 9 .7 ) −3 .3 (− 14 .3 to 8 .9 ) −7 .2 (− 18 .4 to 6 .1 ) Ea st er n M ed ite rr an ea n Re gi on 11 .7 (1 0. 9 to 1 2. 5) 11 .5 (1 0. 7 to 1 2. 2) 11 .5 (1 0. 7 to 1 2. 2) −2 .0 (− 10 .5 to 7. 4) 0. 2 (− 8. 7 to 9 .8 ) −1 .8 (− 10 .5 to 8 .0 ) 16 .3 (1 4. 9 to 1 7.6 ) 16 .2 (1 4. 9 to 1 7.6 ) 16 .3 (1 4. 9 to 1 7.7 ) −0 .5 (− 11 .8 to 1 2. 0) 0. 6 (− 10 .7 to 1 3. 6) 0. 1 (− 11 .0 to 1 2. 9) 6. 9 (6 .3 to 7. 6) 6. 6 (6 .0 to 7. 2) 6. 6 (6 .0 to 7. 2) −4 .6 (− 16 .4 to 8 .6 ) −0 .8 (− 12 .8 to 1 3. 3) −5 .4 (− 16 .9 to 7. 6) W es te rn P ac ifi c Re gi on 8. 6 (7 .5 to 9 .6 ) 8. 5 (7 .4 to 9 .5 ) 8. 6 (7 .4 to 9 .8 ) −1 .1 (− 16 .9 to 1 7. 9) 1. 3 (− 16 .5 to 2 2. 0) 0. 1 (− 17 .1 to 2 0. 6) 10 .4 (8 .6 to 1 2. 2) 10 .1 (8 .4 to 1 1. 8) 10 .1 (8 .3 to 1 2. 0) −2 .4 (− 23 .5 to 2 4. 3) 0. 3 (− 22 .2 to 2 9. 0) −2 .0 (− 24 .4 to 2 6. 3) 6. 5 (5 .3 to 7. 7) 6. 6 (5 .3 to 7. 8) 6. 7 (5 .4 to 8 .1 ) 1. 0 (− 21 .8 to 3 1. 0) 2. 6 (− 21 .9 to 3 4. 9) 3. 6 (− 20 .8 to 3 5. 0) DA LY s, d is ab ili ty -a dj us te d lif e ye ar s; P AF , p op ul at io n at tr ib ut ab le fr ac tio n; U R, u nc er ta in ty ra ng e. ANNEXES 35 AN N EX 8 . RE SU LT S FR OM S EN SI TI VI TY A NA LY SE S FO R DE AT HS A ND D AL YS A TT RI BU TA BL E TO E XP OS UR E TO L ON G W OR KI NG H OU RS (9 5% U NC ER TA IN TY R AN GE S) , 1 83 C OU NT RI ES , F OR TH E YE AR 2 01 6 Se ns iti vi ty a na ly si s Is ch ae m ic h ea rt d is ea se St ro ke No . d ea th s (9 5% U R) No . D AL Ys (m ill io ns ) (9 5% U R) No . d ea th s (9 5% U R) No . D AL Ys (m ill io ns ) (9 5% U R) M ai n an al ys is 34 6  75 3 (3 19  6 58 to 3 73  8 48 ) 10 .6 6 (9 .8 7 to 1 1. 44 ) 39 8  44 1 (3 69  8 26 to 4 27  0 56 ) 12 .6 0 (1 1. 82 to 1 3. 39 ) Re du ce d la g tim e to 8 y ea rs ( 20 03 –2 01 2) 32 4  34 6 (2 98  5 08 to 3 50  1 85 ) 10 .2 2 (9 .4 6 to 1 0. 97 ) 36 8  31 7 (3 42  6 19 to 3 94  0 15 ) 11 .9 5 (1 1. 22 to 1 2. 68 ) In cr ea se d la g tim e to 1 2 ye ar s (1 99 9– 20 08 ) 37 6  81 4 (3 47  8 92 to 4 05  7 35 ) 11 .1 8 (1 0. 37 to 1 1. 99 ) 43 9  16 9 (4 06  7 39 to 4 71  6 00 ) 13 .4 3 (1 2. 57 to 1 4. 28 ) Re du ce d tim e w in do w o f e xp os ur e to 8 y ea rs (2 00 2– 20 09 ) 30 7  32 9 (2 82  8 67 to 3 31  7 92 ) 9. 51 (8 .8 0 to 1 0. 22 ) 35 1  91 1 (3 26  8 00 to 3 77  0 22 ) 11 .2 1 (1 0. 51 to 1 1. 91 ) In cr ea se d tim e w in do w o f e xp os ur e to 1 2 ye ar s (2 00 0– 20 11 ) 38 6  33 9 (3 56  8 10 to 4 15  8 68 ) 11 .71 3 (1 0. 88 to 1 2. 54 ) 44 6  17 1 (4 14  0 58 to 4 78  2 84 ) 13 .9 3 (1 3. 07 to 1 4. 80 ) As si gn ed m os t c om m on e xp os ur e ca te go ry o ve r t im e w in do w ( re st ric te d to y ea rs in la bo ur m ar ke t a ct iv ity ) 17 9  36 0 (1 66  7 55 to 1 91  9 66 ) 5. 01 (4 .7 0 to 5 .3 1) 21 9  51 7 (2 03  9 99 to 2 35  0 34 ) 6. 47 (6 .0 6 to 6 .8 8) UR , u nc er ta in ty ra ng e. 36 WHO/ILO JOINT ESTIMATES OF THE WORK-RELATED BURDEN OF DISEASE AND INJURY, 2000-2016

For further information please contact: Department of Environment, Climate Change and Health World Health Organization 20 avenue Appia CH-1211 Geneva 27 Switzerland jointestimates@who.int https://www.who.int/teams/environment-climate-change- and-health/monitoring/who-ilo-joint-estimates Governance and Tripartism Department International Labour Organization 4 route des Morillons CH-1211 Geneva 22 Switzerland labadmin-osh@ilo.org https://www.ilo.org/global/topics/safety-and-health-at-work/ programmes-projects/WCMS_674797/lang--en/index.htm