Louisa Degenhardt PhD, Professor and NHMRC Principal Research Fellow, Harvey Whiteford MBBS, Kratzman Professor of Psychiatry & Population Health, Wayne D. Hall PhD, Professor and NHMRC Australia Fellow.
The Global Burden of Disease projects: What have we learned about illicit drug use and dependence and their contribution to the global burden of disease?
Article first published online: 20 NOV 2013
© 2013 Australasian Professional Society on Alcohol and other Drugs
Drug and Alcohol Review
Volume 33, Issue 1, pages 4–12, January 2014
How to Cite
Degenhardt, L., Whiteford, H. and Hall, W. D. (2014), The Global Burden of Disease projects: What have we learned about illicit drug use and dependence and their contribution to the global burden of disease?. Drug and Alcohol Review, 33: 4–12. doi: 10.1111/dar.12088
- Issue published online: 9 JAN 2014
- Article first published online: 20 NOV 2013
- Manuscript Accepted: 14 OCT 2013
- Manuscript Received: 26 SEP 2013
- an Australian National Health and Medical Research Council Principal Research Fellowship
- an National Health and Medical Research Council Australia Fellowship
- Australian Government under the Substance Misuse Prevention
- Service Improvements Grants Fund
- Queensland Department of Health
- illicit drugs;
- burden of disease;
The Global Burden of Disease (GBD) 2010 study updated the findings of earlier exercises. It provided regional and global estimates of the burden of disease attributable to diseases, injuries and risk factors. Here we provide a brief summary of the work for illicit drug use.
Systematic reviews were undertaken to estimate the major epidemiological parameters (incidence, prevalence, duration/remission and mortality) for each drug. Reviews evaluated the nature and quality of evidence for illicit drug use as a risk factor for many health outcomes, for the comparative risk assessment (CRA) exercise.
Substantial gaps existed in basic epidemiological parameters. Following modelling and imputation of missing data, it was estimated that opioid and amphetamine dependence were the most common forms of illicit drug dependence in 2010; opioid dependence was responsible for the greatest burden. Few putative consequences of illicit drug use had the quality or quantity of data required to be included in the CRA.
Estimates of the extent and distribution of disease burden are likely to shape global and regional health policy development. The GBD exercise will be repeated on an annual basis; GBD 2010 clearly demonstrated that although the illicit drug field is generating more and better epidemiological data on the health risks of drug use, there is still much work to be done to generate defensible estimates of the magnitude of risk, particularly for impactful and prevalent outcomes, such as injuries, violence and mental health complications. Until then, burden of disease attributable to illicit drugs will be underestimated. [Degenhardt L, Whiteford H, Hall WD. The Global Burden of Disease projects: What have we learned about illicit drug use and dependence and their contribution to the global burden of disease? Drug Alcohol Rev 2014;33:4–12]
There is good evidence that illicit drug use and mental disorders produce substantial loss of life and disability, but their impact upon population health needs to be better understood. Governments, policy-makers and funding bodies need information on the comparative population health impact of different diseases and risk factors when making decisions about where to focus policy, service and research planning, and implementation. It is important that illicit drugs and mental health are understood with that context.
‘Burden of disease’ studies have identified the large proportion of disease burden arising from mental disorders and illicit drug use [1,2]. The high prevalence and chronic nature of some drug use and mental disorders led to them being prominent in the league table of disorders ranked in order of burden in the first Global Burden of Disease (GBD) study . The often quoted finding that depression was a leading cause of disability in the world has been used to promote funding for programs to treat and prevent mental disorders . Burden of disease estimates have been even more important in countries where disease advocacy groups are not well established and where external agencies [such as the World Health Organization (WHO) and World Bank] have been influential in setting priorities for health spending.
Started in 2007, with capstone papers published in December 2012 [5–9] and more specific papers published during 2013 [10,11], the GBD 2010 study comprehensively update the findings of the first GBD exercise. It provided regional and global estimates of the burden of disease attributable to hundreds of diseases, injuries and their risk factors. In this commentary, we provide a brief summary of the work undertaken by the expert group examining illicit drug use and mental disorders, with a specific focus on illicit drugs. We draw on work that has already been published from both GBD 2010 and in the work leading up to the overview publications throughout this paper.
The history of burden of disease studies
Until the early 1990s, guidance on public health funding allocations largely came from studies of population mortality. These studies ignored morbidity arising from disorders and injuries that were not fatal but nonetheless adversely affected a person's functioning . Measuring the impact of disease was revolutionised in 1993, when the World Bank provided estimates of causes of global disease burden using a new summary measure, the disability-adjusted life year (DALY) . The DALY is a summary measure of population health that integrates mortality with morbidity and disability information to produce a single measure of disease burden that enables the relative importance of health problems to be compared. One DALY represents the loss of one healthy year of life.
For each disease or injury, DALYs are calculated as the sum of years lost due to premature mortality and the years of lost health due to disability (see Figure 1). The DALY combines measures of premature mortality (years of life lost) and morbidity (years lived with disability) that are attributable to diseases (e.g. depression, cancer and heart disease). The DALY allows the mortality and morbidity of various diseases to be compared, with the aim of ‘disconnecting advocacy from epidemiology’ . It is used by the World Bank and the WHO to estimate disease burden and the cost-effectiveness of health interventions [14–16], that is, the cost and impact of interventions on the burden of a particular disease or injury. This information is critical for informed priority setting in health care.
A revised set of estimates was published in 1996 as part of the first GBD study . Regular updates have been included in the WHO's World Health Reports  and the Disease Control Priorities publications . In 2002, the World Health Report estimated disease burden attributable to various risk factors—the so-called ‘comparative risk assessment’ (CRA) exercise. This was finalised in 2006 . The CRA estimated that alcohol, tobacco and injecting drug use were important risk factors for global disease burden .
GBD 2010 was led by a core team of researchers from a consortium that included the Institute for Health Metrics and Evaluation at the University of Washington (Seattle, USA), the University of Queensland (Australia), Johns Hopkins University (USA), Harvard University (USA) and the WHO (Switzerland).
It was the first major effort since the original GBD study to produce systematic and comprehensive estimates of the burden of diseases and injuries. It updated the comparative estimates of the burden of risk factors for 1990, 2005 and 2010. The 1990 estimates were recalculated using the improved methods and data that have become available since the original study was undertaken. The new GBD study produced estimates for 21 regions of the world that were published in 2012 [5–9].
The study included epidemiological reviews of all diseases, injuries and risk factors and estimates of the mortality and cause of death for all countries in the world. This involved multiple systematic reviews to estimate the major epidemiological parameters (incidence, prevalence, duration/remission and mortality) for each disorder and a critical synthesis of the existing evidence. The intent of the GBD was to understand and incorporate all existing data on the epidemiology of diseases and risk factors. It did not conduct new studies to estimate these parameters.
Experts across the range of diseases and injuries included in the study provided input to this core team, with disease/disorder groups organised into ‘expert groups’. The expert groups were asked to synthesise existing data on the incidence, prevalence, duration and excess mortality of diseases, exposures to important risk factors, and then to critically assess the estimates of disease burden produced by the core project team. More information on the process is provided at http://www.gbd.unsw.edu.au.
Advances in our understanding of disease epidemiology allowed the GBD group to expand the number of mental disorders and illicit drugs included beyond those in the original GBD study. For example, a greater number of anxiety disorders and childhood mental disorders were included. The original GBD study focused on heroin dependence, because this was the form of illicit drug use for which there were the best prevalence estimates and mortality data . This time, estimates of disease burden were made for dependence on heroin and other illicit opioid use, cocaine, amphetamines and cannabis.
Other drugs were not separately estimated because of data limitations and lack of research on their risks of dependence and other harms. This does not mean that the use of these drugs is without risk to users. However, the inclusion of an ‘other drugs’ category will have captured some of this burden in a non-drug-specific manner .
Detail on results of systematic reviews, by drug type and parameter
There are major gaps in the most basic of epidemiological parameters for illicit drug dependence. Systematic reviews conducted for each of the four major drug types revealed that the greatest amount of data were available for cannabis and opioids, with data on amphetamines and cocaine more sparse (Table 1). Incidence estimates were extremely rare. Estimates of ‘remission’ from dependence were very uncommon ; and mortality had been most widely studied among regular or dependent opioid users . When examining prevalence, use estimates were far more common than dependence estimates; and school surveys were more common than adult surveys .
|Evidence that use or dependence occurs (no. of countries)||Estimates of prevalence of use (no. of countries)||Estimates of dependence prevalence (no. of countries)||Prospective studies examining remission||Prospective studies examining mortality|
Although we found improvements over time in the quality and scope of data on the epidemiology of drug use disorders, nonetheless huge gaps remained. In most countries, there was only a single measure of prevalence and limited knowledge of the natural history of these disorders; in others, such as countries in the Caribbean and Pacific regions and Africa, there were often no data. Expert opinion and advice was sought to produce the most plausible estimates and uncertainty bounds for countries without data. In all stages of data extraction, all available study details were extracted into an access database specifically designed for the project. These databases facilitated the use of modelling and regression techniques to impute data and estimate error around the estimates.
Drug dependence prevalence and disease burden
The results of the epidemiological modelling undertaken in GBD 2010 produced novel results about levels of dependent drug use across the globe; the results presented here have been reported in detail elsewhere previously [10,25, Degenhardt et al., unpublished data]. Opioid and amphetamine dependence were the two most common forms of illicit drug dependence globally in 2010 (15.4 million and 17.2 million estimated cases, respectively; Table 2). There were 13.1 million cannabis-dependent and 6.9 million cocaine-dependent persons. Males formed the majority of cases (64% each for cannabis and amphetamines and 70% each for opioids and cocaine).
|n||%||95% CI||n||%||95% CI||n||%||95% CI||n||%||95% CI|
|Asia Pacific, high income||390 000||0.28||(0.18–0.41)||372 000||0.24||(0.17–0.34)||257 000||0.06||(0.05–0.07)||456 000||0.28||(0.17–0.44)|
|Asia Central||197 000||0.22||(0.17–0.29)||203 000||0.23||(0.18–0.29)||52 000||0.02||(0.01–0.02)||209 000||0.24||(0.18–0.33)|
|Asia East||2 402 000||0.17||(0.09–0.28)||2 634 000||0.18||(0.12–0.26)||234 000||0.16||(0.11–0.24)||2 180 000||0.14||(0.08–0.24)|
|Asia South||2 649 000||0.15||(0.13–0.18)||3 993 000||0.24||(0.16–0.37)||1 086 000||0.07||(0.04–0.10)||4 331 000||0.26||(0.22–0.31)|
|Asia South East||977 000||0.15||(0.11–0.19)||2 724 000||0.42||(0.34–0.54)||114 000||0.02||(0.01–0.02)||956 000||0.15||(0.11–0.20)|
|Australasia||154 000||0.68||(0.60–0.78)||98 000||0.41||(0.29–0.56)||32 000||0.14||(0.09–0.20)||110 000||0.46||(0.41–0.53)|
|Caribbean||69 000||0.16||(0.12–0.21)||88 000||0.20||(0.16–0.25)||143 000||0.33||(0.26–0.42)||109 000||0.26||(0.18–0.36)|
|Europe Central||249 000||0.23||(0.18–0.29)||365 000||0.31||(0.27–0.37)||63 000||0.05||(0.04–0.06)||230 000||0.19||(0.15–0.26)|
|Europe Eastern||432 000||0.22||(0.15–0.33)||298 000||0.14||(0.11–0.19)||117 000||0.05||(0.04–0.07)||607 000||0.27||(0.17–0.44)|
|Europe Western||1,141 000||0.34||(0.28–0.41)||938 000||0.26||(0.24–0.28)||641 000||0.18||(0.16–0.19)||1 318 000||0.35||(0.32–0.39)|
|Latin America, Andean||62 000||0.11||(0.08–0.15)||76 000||0.14||(0.12–0.17)||145 000||0.26||(0.20–0.34)||153 000||0.28||(0.18–0.42)|
|Latin America, Central||220 000||0.09||(0.07–0.13)||710 000||0.30||(0.23–0.39)||274 000||0.12||(0.09–0.14)||572 000||0.24||(0.17–0.35)|
|Latin America, Southern||169 000||0.28||(0.19–0.43)||153 000||0.26||(0.20–0.33)||184 000||0.30||(0.21–0.42)||208 000||0.35||(0.22–0.54)|
|Latin America, Tropical||286 000||0.14||(0.08–0.23)||708 000||0.33||(0.26–0.43)||920 000||0.43||(0.30–0.59)||491 000||0.23||(0.12–0.39)|
|North Africa/Middle East||735 000||0.14||(0.12–0.18)||1 145 000||0.24||(0.20–0.28)||691 000||0.14||(0.11–0.17)||1 374 000||0.29||(0.22–0.37)|
|North America, High Income||1 755 000||0.60||(0.53–0.68)||717 000||0.23||(0.18–0.28)||1 604 000||0.53||(0.39–0.72)||959 000||0.30||(0.25–0.36)|
|Oceania||21 000||0.20||(0.13–0.31)||25 000||0.26||(0.18–0.37)||3 000||0.03||(0.02–0.05)||19 000||0.20||(0.12–0.31)|
|Sub-Saharan Africa Central||151 000||0.16||(0.11–0.23)||207 000||0.24||(0.17–0.34)||40 000||0.05||(0.03–0.07)||118 000||0.15||(0.09–0.23)|
|Sub-Saharan Africa East||589 000||0.16||(0.13–0.20)||798 000||0.24||(0.20–0.29)||105 000||0.03||(0.03–0.04)||488 000||0.15||(0.12–0.19)|
|Sub-Saharan Africa South||149 000||0.18||(0.12–0.28)||188 000||0.24||(0.17–0.34)||37 000||0.05||(0.03–0.07)||157 000||0.21||(0.13–0.35)|
|Sub-Saharan Africa West||276 000||0.08||(0.06–0.11)||742 000||0.24||(0.19–0.32)||149 000||0.05||(0.04–0.07)||435 000||0.15||(0.11–0.20)|
|Females||4 696 000||0.14||(0.12–0.16)||6 256 000||0.18||(0.16–0.22)||2 090 000||0.06||(0.05–0.07)||4 698 000||0.14||(0.12–0.16)|
|Males||8 377 000||0.23||(0.20–0.27)||10 928 000||0.31||(0.27–0.37)||4 801 000||0.14||(0.12–0.16)||10 781 000||0.31||(0.27–0.35)|
|Overall||13 073 000||0.19||(0.17–0.21)||17 184 000||0.25||(0.22–0.28)||6 891 000||0.10||(0.09–0.11)||15 479 000||0.22||(0.20–0.25)|
The geographic distribution of cases reflected variations in prevalence and country populations (Table 2). An estimated 57.8% of amphetamine dependence cases were found across the Asian regions (9.3 million cases), but the highest prevalence estimates were for Southeast Asia [0.42%; 95% uncertainty interval (UI) 0.34–0.54%] and Australasia (0.41%; 95% UI 0.29–0.56%). North America High-Income was estimated to contain 13.4% of cannabis-dependent people, with a high prevalence (0.6%; 0.5–0.7%). The highest levels of cocaine dependence were estimated in North America High-Income (0.53%; 95% UI 0.39–0.72%) and Latin America. Australasia had among the highest levels of opioid dependence (0.46%; 95% UI 0.41–0.53%), although the largest populations were in East and South Asia. Estimated levels of illicit drug dependence were generally lower in African and Asian regions.
Drug use disorders directly accounted for 20.0 million DALYs in 2010 (95% UI 15.3–25.4 million; Table 3). This was 0.8% (0.6%–1.0%) of global all-cause DALYs. This was an increase of 52% from estimates for 1990 (using the same methods), when the estimated direct burden was 13.1 million DALYs or 0.5% (0.4–0.7%) of all-cause DALYs.
|Lower CI||Mean||Upper CI||Lower CI||Mean||Upper CI||Lower CI||Mean||Upper CI|
|YLDs||1 348 000||2 057 000||2 929 000||849 000||1 323 000||1 936 000||481 000||734 000||1 063 000|
|DALYs||1 348 000||2 057 000||2 929 000||849 000||1 323 000||1 936 000||481 000||734 000||1 063 000|
|YLDs||1 460 000||2 596 000||3 957 000||928 000||1 657 000||2 562 000||522 000||939 000||1 502 000|
|YLLs||6 000||21 000b||15 000||4 000||15 000b||13 000||1 000||5 000b||4 000|
|DALYs||1 470 000||2 617 000||4 109 000||933 000||1 673 000||2 653 000||524,000||944 000||1 520 000|
|YLDs||633 000||1 085 000||1 639 000||443 000||760 000||1 168 000||187 000||325 000||503 000|
|YLLs||7 000||25 000b||22 000||5 000||18 000b||17 000||2 000||6 000b||5 000|
|DALYs||645 000||1 110 000||1 727 000||452 000||778 000||1 200 000||189 500||331 800||518 700|
|YLDs||5 143 000||7 170 000||9 258 000||3 550 000||5 017 000||6 536 000||1 484 000||2 153 000||2 877 000|
|YLLs||1 233 000||1 981 000||3 133 000||771 000||1 460 000||2 419 000||287 000||522 000||792 000|
|DALYs||7 066 000||9 152 000||11 443 000||4 860 000||6 477 000||8 298 000||1 963 000||2 675 000||3 453 000|
|Other drug use disorders|
|YLDs||2 108 000||3 503 000||5 170 000||1 380 000||2 306 000||3 439 000||723 000||1 198 000||1 821 000|
|YLLs||1 008 000||1 555 000||2 552 000||590 000||1 114 000||1 941 000||249 000||441 000||739 000|
|DALYs||3 555 000||5 059 000||7 042 000||2 390 000||3 420 000||4 798 000||1 128 000||1 639 000||2 348 000|
|YLDs||11 837 000||16 411 000||21 584 000||7 934 000||11 063 000||14 572 000||3 763 000||5 349 000||7 095 000|
|YLLs||2 225 000||3 582 000||5 683 000||1 340 000||2 607 000||4 409 000||538 000||975 000||1 510 000|
|DALYs||15 255 000||19 995 000||25 367 000||10 214 000||13 670 000||17 454 000||4 715 000||6 324 000||8 199 000|
Much of the change over time could be attributed to population growth. The exception was opioid dependence, where 42% of the increase was attributed to increased prevalence between 1990 and 2010; overall opioid dependence burden increased by 74% across the period.
Two-thirds (69.3%) of all drug disorder DALYs were explained by years lived with disability and 30.7% by years of life lost. Opioid dependence accounted for the highest proportion (46%) of illicit drug burden (9.2 million DALYs, 95% UI 7.1–11.4 million). Cocaine dependence accounted for the smallest burden (5.5% of illicit drug burden; 1.1 million DALYs, 95% UI 0.65–1.7 million). Cannabis dependence was not estimated to cause any years of life lost but contributed 2.1 million DALYs in the form of years lived with disability (95% UI 1.3–2.9 million; 10.3% of illicit drug burden).
It is important to note that we did not estimate harmful use/abuse of illicit drugs (as defined by WHO's International Classification of Diseases (ICD) and American Psychiatric Association's Diagnostic and Statistical Manual (DSM) of mental disorders) in GBD 2010. The same decision was made by the GBD alcohol expert group. The reasons for this decision were the limited data on these disorders, ongoing debate about the validity of these diagnoses and the likely small disability associated with such disorders. Future iterations of GBD might reconsider this decision.
The health consequences of illicit drug use
As mentioned previously, one component of GBD 2010 was the CRA exercise, which examines risk factors for health outcomes, including illicit drug use. The adverse health effects of illicit drug use can be considered conceptually under four headings : acute toxic effects, including overdose; acute effects of intoxication, such as injuries and violence; dependence on the drug; and adverse health effects of sustained regular use, such as chronic physical disease (e.g. cardiovascular disease and cirrhosis), blood-borne infections, and mental disorders.
For GBD 2010, it was necessary to evaluate the nature and quality of evidence for illicit drug use as a risk factor for many health outcomes . In order for risk outcomes to be eligible for inclusion in the CRA component, a number of eligibility criteria needed to be met (see Table 4). In order to make a causal inference, it is necessary to document an association, confirm that drug use preceded the outcome and exclude alternative explanations of the association, such as reverse causation and confounding .
|Inclusion criteria for each risk-outcome pair:|
The results of these reviews were sobering, in that very few putative consequences of illicit drug use had anywhere near the quality or quantity of data, or enumeration of effect sizes, required to be eligible for inclusion in the CRA (Table 5). Many studies report associations between illicit drug use and various health-related harms, but it has been more challenging to decide whether these are causal relationships. Our review of the availability of evidence, the quality of evidence and the strength of associations observed for each drug type for a range of putative acute and chronic outcomes revealed several things : (i) the risks of cannabis use are much more modest than those of other illicit drugs, largely because cannabis does not produce fatal overdoses and it cannot easily be injected; (ii) the quality of evidence varies widely across drug and health outcomes: there are more data on cannabis use from prospective population-based cohorts, and for the use of other drug types, more data from selected treatment cohorts; and (iii) the magnitude of the effect is often poorly quantified. In the end, GBD 2010 only included the outcomes of drug use listed in Table 6, despite considering dozens more .
|Acute toxic effects (fatal overdose)||✗||✓||✓||✓|
|Acute intoxication effects|
|Accidental injury||?||✓||One of the most common causes of death among opioid users, however not included as it was decided confounding was not adequately addressed in existing cohorts.||?||Plausible, however too few data to assess||?||Plausible, however too few data to assess|
|Motor vehicle accidents||✓||Evidence suggests an association, but existing epidemiological studies not thought to have adequately controlled for confounding||?||Plausible, however too few data to assess||?||Plausible, however too few data to assess||?||Plausible, however too few data to assess|
|Drug-induced psychotic symptoms||✓||✗||✓||Limited controlled data on risk||✓||Limited controlled data on risk|
|Violence||✗||✗||✓||Plausible, however too few data to assess||✓||Plausible, however too few data to assess|
|Myocardial infarction||?||Emerging evidence, considered too limited at present. Not included in CRA||✗||✓||Plausible, however too few data to assess||✓||Plausible, however too few data to assess|
|Dependence||✓||Included in CRA||✓||Included in CRA||✓||Included in CRA||✓||Included in CRA|
|Adverse health effects of chronic use|
|Cardiovascular pathology||?||Emerging evidence, considered too limited at present. Not included in CRA||?||Evidence largely cross-sectional studies of users or case series. Poor control for confounding. Not included.|
|Liver disease||✗||✓||Pathology in chronic users, however poor control for confounding||✓||Pathology in chronic users, however poor control for confounding||✓||Pathology in chronic users, however poor control for confounding|
|Pulmonary disease||?||Emerging evidence, considered too limited at present. Not included in CRA||?||Pathology in chronic users, however poor control for confounding||?||?|
|Cancers||?||Emerging evidence, considered too limited at present. Not included in CRA||?||Some evidence of increased risk but confounding not controlled||?||?|
|Neurotoxic effects||?||Emerging evidence, considered too limited at present. Not included in CRA||✗||✓||Some evidence of increased risk but confounding not controlled||✓||Some evidence of increased risk but confounding not controlled|
|Psychotic disorders||✓||Included in CRA||✗||✓||Plausible, but Insufficient controlled prospective data. Not included in CRA||✓||Plausible, but Insufficient controlled prospective data. Not included in CRA|
|Common mental disorders||?||Inconsistent evidence on the association with depression and anxiety, although clearer for depression. Not included.||✓||Depression and anxiety elevated among this group, although few prospective data examining risks. Not included.||✓||Depression and anxiety elevated among this group, although few prospective data examining risks. Not included.||✓||Depression and anxiety elevated among this group, although few prospective data examining risks. Not included.|
|Suicide||✗||Few epidemiological data, poor control for confounding, inconsistent results.||✓||Consistent finding of elevations In suicide. Included in CRA||✓||Consistent finding of elevations In suicide. Included in CRA||✓||Consistent finding of elevations In suicide. Included in CRA|
|Consequences of unsafe drug injection||Effect|
|HIV||✓||Included in CRA|
|HCV||✓||Included in CRA|
|HBV||✓||Included in CRA|
|Infective endocarditis||✓||A likely outcome of unsafe injection, however risk rarely quantified. Not included in CRA|
|Tuberculosis||✓||Has been noted as prevalent in some countries among injectors, particularly as an HIV co-infection, but few data on prevalence across countries. Not included in CRA|
Where to from here?
Estimates of the extent and distribution of disease burden for different disorders are likely to shape global and regional health policy development. Existing estimates were used in debates to evaluate (and justify) WHO's global funding distribution [30,31]. They have been used in multiple discussions of funding allocation and priorities [32–34]. There have been consistent increases in the extent of funding for non-communicable diseases since 1990 (http://go.worldbank.org/851WC143G0).
There are limitations and controversies surrounding the GBD methodology, arising from the way in which ‘disability’ is estimated, the lack of consideration of social, economic and crime aspects of ‘burden’; and the inherent limitations of any enumerative exercise that is constrained by limited and potentially low quality data. Nonetheless, the GBD 2010 results have already been widely discussed and used, with demand from countries for specific country-level data to assist in their health service and policy planning [35,36]. These data will no doubt be used in future to inform funding allocation across multiple sectors, perhaps also including the drug and alcohol field.
The GBD team (now GBD 2.0) will continue to collect new data, and generate improved models and outputs, on an annual basis (see http://www.healthmetricsandevaluation.org/gbd/2013/protocol). At present, data are being collected for the 2013 round of GBD estimates. An important component of this ongoing work is the formation of the GBD Scientific Council. One of the roles of this council will be to evaluate proposals for new risk-outcome pairs, in a transparent and rigourous process that adheres to the principles for inclusion of such pairs in the CRA (Table 4). GBD 2010 clearly demonstrated that although the field is generating more and better epidemiological data on the health risks of drug use in more recent years, there is still much work to be done to generate defensible estimates of the magnitude of risk, particularly for impactful and prevalent outcomes, such as injuries, violence and mental health complications drug use. Until such data are generated, estimates of the burden of disease attributable to illicit drug use will be vast underestimates.
Professor Louisa Degenhardt is supported by an Australian National Health and Medical Research Council Principal Research Fellowship. Professor Wayne Hall is supported by an National Health and Medical Research Council Australia Fellowship. The National Drug and Alcohol Research Centre at the University of NSW is supported by funding from the Australian Government under the Substance Misuse Prevention and Service Improvements Grants Fund. Professor Harvey Whiteford is affiliated with the Queensland Centre for Mental Health Research, which receives its core funding from the Queensland Department of Health. More information can be found at: http://www.gbd.unsw.edu.au and http://www.healthmetricsandevaluation.org/gbd.
Conflict of interest
Professor Louisa Degenhardt has received untied educational grants from Reckitt Benckiser for the conduct of post-marketing surveillance studies of the diversion and injection of opioids prescribed for opioid substitution therapy. That funder had no knowledge of this paper.
- 3The global burden of disease: a comprehensive assessment of mortality and disability from diseases, injuries and risk factors in 1990 and projected to 2020. Boston, MA: Harvard University Press, 1996., , eds.
- 5A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380:2224–2260., , , et al.
- 6Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380:2095–2128., , , et al.
- 7Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380:2197–2223., , , et al.
- 8Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010. Lancet 2012;380:2129–2143., , , et al.
- 9Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380:2163–2196., , , et al.
- 13World Bank. World development report 1993: investing in health. New York: Oxford University Press, 1993.
- 14Design, content and financing of an essential national package of health services. In: Murray CJL , Lopez AD , eds. Global comparative assessments in the health sector: disease burden, expenditures and intervention packages. Geneva: World Health Organization, 1994:171–180., , , .
- 15on behalf of WHO-CHOICE. Choosing cost-effective interventions in psychiatry: results from CHOICE programme of the World Health Organization. World Psychiatry 2005;4:37–44.,
- 16Jamison DT , Breman JG , Measham AR , et al., eds. Disease control priorities in developing countries, Second edn. New York: Oxford University Press, 2006.
- 17World Health Organization. The world health report 2008: primary health care now more than ever. Geneva: World Health Organization, 2008.
- 26Babor TF , Caulkins J , Edwards G , et al., eds. Drug policy and the public good. Oxford: Oxford University Press, 2010.
- 27Extent of illicit drug use and dependence, and their contribution to the global burden of disease. Lancet 2012;379:55–70., .
- 28World Cancer Research Fund/American Institute for Cancer Research. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. Washington, DC: AICR, 2007.