Article Text

Extended Report
The global burden of gout: estimates from the Global Burden of Disease 2010 study
  1. Emma Smith1,
  2. Damian Hoy2,
  3. Marita Cross1,
  4. Tony R Merriman3,
  5. Theo Vos2,4,
  6. Rachelle Buchbinder5,6,
  7. Anthony Woolf7,
  8. Lyn March1
  1. 1Department of Rheumatology, Royal North Shore Hospital, Northern Clinical School, Institute of Bone and Joint Research, University of Sydney, St Leonards, New South Wales, Australia
  2. 2School of Population Health, University of Queensland, Herston, Queensland, Australia
  3. 3School of Medical Sciences, University of Otago, Dunedin, New Zealand
  4. 4Institute for Health Metrics and Evaluation, University of Washington, Seattle, Washington, USA
  5. 5Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
  6. 6Monash Department of Clinical Epidemiology, Cabrini Hospital, Melbourne, Victoria, Australia
  7. 7Department of Rheumatology, Royal Cornwall Hospital, Truro, UK
  1. Correspondence to Lyn March, Northern Clinical School, Institute of Bone and Joint Research, University of Sydney, Department of Rheumatology, Royal North Shore Hospital, Clinical Administration Level 7C, St Leonards, NSW 2065 Australia; lyn.march{at}sydney.edu.au

Abstract

Objective Gout is the most common cause of inflammatory arthritis in men, but has not previously been included in Global Burden of Disease (GBD) studies. As part of the GBD 2010 Study, the Musculoskeletal Disorders and Risk Factors Expert Group estimated the global burden of gout.

Methods The American Rheumatism Association 1977 case definition of primary gout was used in the study. A series of systematic reviews were conducted to gather the age-specific and sex-specific epidemiological data for gout prevalence, incidence, mortality risk and duration. Two main disabling sequelae of gout were identified; acute episode gout and chronic polyarticular gout, and used in the surveys to collect data to derive disability weights. The epidemiological data together with disability weights were then used to calculate years of life lived with disability (YLDs) for gout, for 1990 and 2010. No evidence of cause-specific mortality associated with gout was found. Gout disability-adjusted life years (DALYs), therefore, have the same value as YLDs.

Results Global prevalence of gout was 0.08% (95% uncertainty interval (UI) 0.07 to 0.08). DALYs increased from 76 000 (95% UI 48 to 112) in 1990 to 114 000 (95% UI 72 to 167) in 2010. Out of all 291 conditions studied in the GBD 2010 Study, gout ranked 138th in terms of disability as measured by YLDs, and 173rd in terms of overall burden (DALYs).

Conclusions The burden of gout is rising. With increasing ageing populations globally, this evidence is a significant prompt to optimise treatment and management of gout at individual, community and national levels.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Gout is an inflammatory metabolic joint disease, characterised by formation of monosodium urate crystals in the synovial fluid of joints and in other tissues.1 The prevalence of gout and hyperuricaemia has been reported to be increasing worldwide.2 ,3 The disease is predominant in men, but is also prevalent in the older female population.4 Menopause increases the risk of gout among women.5 Gout recently passed rheumatoid arthritis to become the most common inflammatory arthritis in the USA.6 In the UK, gout is a common inflammatory joint disease affecting 1.4% of the population, with a prevalence as high as 7% in men aged over 75 years.7 Gout prevalence in some ethnic groups such as the New Zealand Ma¯ori is as high as 6.1% overall, and 30% in men over 75 years.8 Acute gout is a painful disabling condition while untreated chronic gout leads to chronic joint pain, joint erosion and damage, and formation of tophi which can rupture, become infected or cause other complications depending upon where they are situated. The disease, therefore, presents a significant burden on the individual and the community. Gout was included in the Global Burden of Disease (GBD) study for estimation of the overall burden globally for the first time. It was one of the five musculoskeletal (MSK) diseases included in the GBD 2010 Study.

This extended report is part of the GBD 2010 Study recently published in The Lancet (Vol. 380 (9859) December 2012). The global burdens of low back pain, occupationally related low back pain, neck pain, osteoarthritis, rheumatoid arthritis, ‘other MSK conditions’ category, and low bone mineral density, included in the GBD 2010 Study, are reported in the issue.9–15 Reflecting on the global burden of MSK conditions and lessons learnt from the GBD 2010 study, from the point of view of the MSK Disorders and Risk Factors Expert Group, and suggestion for the next steps forward, are also reported in this issue.16

Methods

Full description of the methods used for the burden estimate of the MSK diseases, GBD 2010 Study, is reported separately in the issue.17

Case definition

The GBD 2010 Study used the most common case definition of primary gout given by the American College of Rheumatology (ACR), generally referred to as ARA 1977 (also known as ACR 1977) survey criteria requiring at least 6 of 11 gout symptoms to make a diagnosis.18 The International Classification of Diseases, ICD-10 code for gout is M10.

Systematic reviews

Systematic reviews of the prevalence, incidence and mortality risk of gout throughout the world were conducted. Search terms were established for gout. GBD regions and countries were also included as part of the search terms. Literature searches were then performed for years 1980 to 2009 on MEDLINE, EMBASE, CINAHL, CAB Abstracts, WHO Library and OpenSIGLE. Inclusion and exclusion criteria were set to identify eligible articles for relevant epidemiological data extraction (see online supplementary appendix 1 for details). Ninety-eight eligible articles were identified for prevalence and incidence, and three articles for mortality risk. A risk of bias assessment tool19 was adapted and used to assess the potential for bias of eligible articles. Additional literature searches were performed to identify articles that reported on disease duration, and the average number of episodes or attacks a person with gout experiences per year. The data were then gathered for gout severity estimation.

A number of studies, particularly prevalence studies, identified as being eligible for data extraction in the systematic review process, were excluded from the DisMod-MR model (see details below) during the analysis. Studies identified for prevalence included hyperuricaemia and gout, but only the extracted data on gout were included in the DisMod-MR model for the GBD 2010 Study. Additionally, some of the reported values of gout prevalence were excluded as they appeared to be outliers. Table 1 shows the details of the extracted data for prevalence, incidence, mortality risk and disease duration and number of gout episodes, included in the DisMod-MR model for gout.

Table 1

Details of gout data for prevalence, incidence, mortality risk, and disease duration and number of gout episodes, included in DisMod-MR analysis, GBD 2010 Study

DisMod-MR model

DisMod-MR was the newest iteration of the generic disease modelling system developed for the GBD 2010 Study. It has been redesigned as a Bayesian meta-regression tool. A detailed description of the tool was given in another article in this issue, which gave an overview of the methods for MSK conditions. For gout, in the prior settings of DisMod-MR model, it was assumed that there was no remission, as remission by the GBD definition means total cure without need for ongoing treatment. Because only data from the adult population were included in the analysis, the prior for prevalence was set to zero before age 15 years. Incidence data were set to ‘unusable’ after a considerable inconsistency was found between prevalence and incidence that could not be reconciled (see online supplementary appendix 2 for details).

A composite score of the risk of bias was added as a study level covariate. The large positive coefficient meant that DisMod-MR adjusted the data points from studies with high risk of bias downwards. Body mass index, red meat intake and per capita alcohol intake were initially included as country-level covariates based on associations widely reported in the epidemiological literature. Alcohol and red meat intake covariates, however, did not have an effect on the results, and were subsequently removed (see online supplementary appendix 2 for details).

Severity and disability weights

Sequelae refer to consequences of diseases and injuries. The MSK Expert Group chose two sequelae to characterise different levels of gout severity: an acute episode of gout; and chronic polyarticular gout. Both sequelae were described in lay terms, according to a specific set of health domains. The lay descriptions were then used in the household surveys and an open-access web-based survey by a Disability Weights (DWs) Group, established by the Core Team at IHME (Institute for Health Metrics and Evaluation, University of Washington, USA), to collect data to derive study-wide DWs20 (table 2).

Table 2

Sequelae and DWs for gout, GBD 2010 Study

Estimates of the average number of episodes or attacks per year and the duration of episodes were pooled from six published sets of data.21–25 The small proportion of cases of gout (1.4% cases) with more than 26 gout episodes in a year, implied by the fitted curve, represented chronic cases. The average number of episodes per year was 3.9 (95% UI 2.9 to 5.2). The average duration of gout episodes, derived from the data reported by Gardner et al,26 was 6.8 days (95% UI 6.3 to 7.4). When multiplying the average duration of the episodes with the average number of episodes per year, a proportion of symptomatic time of 6.1% (95% UI 3.9% to 8.3%) was contributed by patients with acute episodes of gout, and a proportion of symptomatic time of 1.0% (95% UI 0.8% to 1.3%) was contributed by patients with chronic polyarticular gout. The average DW for an individual with gout was 0.023 (95% UI 0.014 to 0.034) (see online supplementary appendix 3 for details).

YLDs and DALYs estimates

The disability-adjusted life year (DALY) is the standard metric used to quantify burden,27 defined as years of healthy life lost. DALYs are the sum of years of life lost due to premature mortality and years of life lived with disability (YLD). One DALY equals one lost year of healthy life. As there was no evidence for cause-specific mortality associated with gout, YLDs and DALYs estimates for the disease reported in the GBD 2010 Study were the same. For the years 1990 and 2010, YLDs for gout were estimated by multiplying the age-standardised prevalence of gout by the DW for acute gout and the symptomatic time. The symptomatic time was a total of the proportion of symptomatic time contributed by patients with acute gout (6.1%, 95% UI 3.9% to 8.3%) and the proportion of symptomatic time contributed by patient with chronic polyarticular gout (1%, 95% UI 0.8% to 1.3%). Age-standardised prevalence estimates were presented using the 2001 WHO standard population.28 The uncertainty interval (UI) around each estimate was calculated systematically carrying forward uncertainty around data inputs and data manipulations, including the use of country and region fixed effects in DisMod-MR and the pooled estimate of average frequency of episodes. Uncertainty ranges are bounded by the 2.5 and 97.5 percentile values, which can be interpreted as a 95% UI. Further detail on how uncertainty was calculated can be found elsewhere.29

As DWs were derived for single health states, simple addition of YLDs for all conditions would assume that disability is additive if a person has comorbid health states. Thus, a person with a number of more severe health states could be awarded a cumulative DW that exceeds 1, which equates to greater health loss than ‘being dead’. Assuming a multiplicative function between DWs for comorbid health states assures that a combined DW could never be greater than 1. To make a correction for comorbidity, hypothetical populations were simulated for each age, sex, country and year. Individuals in these hypothetical populations were assigned to have none, one or more than one health state based on the prevalence figures for each health state. The multiplicative function was applied to any individual with comorbid health states, and the average DW for each component health state reduced proportionately. This allowed an estimate of the reduction in DW for any health state in an age and sex group by country and year. The process is referred to as the ‘comorbidity correction’.

Results

Description of analysed data

There were 496 data points included in the final DisMod-MR analysis. These were from 30 countries, and 15 of the 21 GBD 2010 regions. The majority of data were for prevalence across a broad age range in the adult population.

Prevalence

The global age-standardised prevalence of gout (from 0 years to 100 years of age) in 2010 was 0.076% (95% UI 0.072 to 0.082). It was greater in men (mean: 0.125%; 95% UI 0.116 to 0.136) than in women (mean: 0.032%; 95% UI 0.030 to 0.035). Gout prevalence did not change significantly when compared between 1990 and 2010. The prevalence increased steadily from the age of 30 years in both sexes, with the minimal prevalence value (less than 0.025%) observed in women before the age of 45 years. The increases were distinctly higher in 5 of the 21 GBD 2010 regions, especially for men. Higher age-standardised prevalence of gout was also noted in these five GBD regions. In 2010, the prevalences were 0.39% (95% UI 0.35 to 0.43) for Australasia, 0.28% (95% UI 0.15 to 0.49) for southern Latin America, 0.24% (95% UI 0.22 to 0.28) for North America High Income, 0.21% (95% UI 0.18 to 0.25) for western Europe, and 0.12% (95% UI 0.07 to 0.21) for Oceania (figure 1 and table 3).

Table 3

Gout age-standardised prevalence (%) and DALYs (all ages), with 95% uncertainty intervals, by region and sex for 2010, GBD 2010 Study

Figure 1

DisMod-generated 1990 and 2010 prevalence of gout by age, sex, year and region, Global Burden of Disease (GBD) 2010 Study.

YLDs and DALYs

While the global prevalence remained relatively similar, the burden of gout has increased by 49%. Gout DALYs increased from 76 000 (95% UI 48 000 to 112 000) in 1990 to 114 000 (95% UI 72 000 to 167 000) in 2010 (table 3). In 2010, at a global level, the DALYs estimates were higher in men (89 000; 95% UI 57 000 to 130 000) than in women (25 000; 95% UI 16 000 to 37 000). Among the 21 GBD 2010 regions, the highest overall burden of gout was in western Europe (30 000; 95% UI 19 000 to 44 000), followed by North America High Income (25 000; 95% UI 15 000 to 37 000) and East Asia (21 000; 95% UI 13 000 to 32 000) (table 3).

Out of all 291 conditions studied in the GBD 2010 Study, for 2010, gout ranked 138th in terms of disability as measured by YLDs, and 173rd in terms of overall burden (measured in DALYs) (table 4). The rankings were slightly higher than in 1990 when the disease YLD estimate was ranked 141st and it was 174th for its DALY. Overall, the highest ranking for gout was in Australasia, for YLDs and DALYs and for the years 1990 and 2010 (table 4). In Australasia, North American High Income, western Europe and southern Latin America, gout YLDs were ranked within the top 100 disease estimates for 2010. It was also noticeable that gout disability burden was becoming more prominent in Asia Pacific High Income over the past two decades.

Table 4

Regional gout YLD and DALY rankings in 2010 (out of 291 conditions), GBD 2010 Study

Discussion

Estimates of the global burden of gout

Gout is known to be more prevalent in men. The global burden of gout has been estimated for the first time in the GBD study. In clinical terms, gout can be cured or go into remission by urate lowering therapy. However, by the GBD 2010 Study definition set by the Core Team at IHME, remission means total cure, that is, people with the disease stop having the disease, and not the clinician’s notion of an improvement that is less than a total cure. For the GBD purposes of gout, this would be remission without need for ongoing treatment and it was deemed that this did not occur. Therefore, in the DisMod-MR model, gout was determined as having no remission after the disease onset.

Gout prevalence progressively increases from the age of 30 years regardless of gender; however, relative to men, there is a low prevalence of gout in women before the age of 45 years, which could be explained by the link between menopause and gout.5 Due to ageing and rising obesity, the burden of gout is increasing around the world, especially in high-income countries. For example, in the Asia Pacific High Income region, gout burden has become more prominent over the past two decades, with a 56% increase in DALYs, while the age-standardised prevalence has increased by just 6% (−20 to +47%). With improved child survival and ageing populations throughout the world,30 especially in low-income and middle-income countries, the number of people living with the disability from gout will continue to increase over coming decades, especially in those areas where effective treatment is not available.

Although a recent review of gout epidemiology1 showed that gout prevalence was rising over the past few decades, the GBD 2010 Study found that the global age-standardised prevalence of gout did not change significantly when compared between 1990 and 2010. It is possible that the heterogeneity in the available data analysed may not allow the trend to be detected even if a rising trend could be expected due to an increase in exposure to some risk factors, including increased alcohol and sugar-sweetened beverage consumption, changes in dietary habits, more modernised lifestyle, obesity and metabolic syndrome, as per the report by Lim et al.31 Additionally, at a glance, gout YLD and DALY rankings may seem relatively low among the other 290 diseases and injuries studied. But as the disease burden continues to rise with age, and together with world population growth, decreased death rates and increased average age of the world's population, the need to address the disability from gout is becoming more and more important. An emphasis on better management of gout in primary and secondary care as well as public health education and promotion of healthy lifestyles should also be prioritised along with other chronic MSK diseases. With the effective treatment of gout currently available, the disease burden could be reduced dramatically.

Strengths and limitations

The extensive series of systematic reviews that were undertaken to obtain data for making the estimates, together with considering the risk of bias of included studies are the greatest strengths of the gout burden estimates in the GBD 2010 Study. Inclusion criteria were strictly employed to identify population-based epidemiological studies that based the presence of gout on the ARA 1977 criteria.18 The newly developed set of DWs20 and use of a new, more advanced version of DisMod, DisMod-MR, in the analysis ensured a robust analysis and credible estimates.

The main limitation of the global burden of gout estimates relates to a paucity of rigorous epidemiological data for some regions of the world, and missing and unusable data. Only two out of 496 data points concerned estimates from the four sub-Saharan GBD 2010 Study regions while data from Oceania and the Caribbean were also sparse (4 out of 496 data points). Consequently, for these regions, the data had to be statistically inferred from data of other regions considered to be theoretically comparable and are, therefore, likely to be less precise. This is a guiding principle in the GBD 2010 Study to make estimates even when the data are sparse. Only a small amount of data could be identified to derive estimates of the number of episodes per year, and the duration of each episode. None of the published gout incidence data were usable because of a substantial inconsistency between prevalence and incidence that could not be reconciled. This could be speculated as a result of the chronic episodic cases with a repeat episode being counted as a new case.

Some of the relevant data for the burden estimates at a country level were identified as outliers at the regional and global levels in DisMod-MR analysis, and were subsequently removed from the final gout DisMod model. These were the data of exceptionally high gout prevalence in Aboriginal Australian,32 New Zealand Ma¯ori33 and Tokelauan34 populations. An attempt was made to adjust other relatively high gout prevalence data reported in sizeable ethnic groups;35–37 New Zealand Ma¯ori and Taiwanese Aborigines, to the proportion of the total population to derive national average, but this led to too low prevalence values. The Core Team at IHME has been made aware of the outcome of this data manipulation, and has planned to improve the data transformation to ascertain a more accurate average of the data for the estimates in the future iteration of GBD.

Different ways of reporting the methods and findings, especially in relation to the demographics of participants, prevalence period, reported values specific to the disease such as prevalence and incidence, and their uncertainty ranges, contributed to the difficulties in extracting the data needed for the analysis. This resulted in some studies being excluded. Concerning the health states, body functions and structures (eg, vision), and more complex human operations (eg, mobility) were the main components of gout sequelae. Other components such as participation, wellbeing, carer burden, economic impact and burden of disease from the individual's perspective, which are also relevant, were not included. This additional information would allow a more complete assessment of the impact of gout in the population.

Suggested further research

Further population-based studies on gout prevalence, incidence, levels of severity including the duration and number of the episodes per year, are justified in regions identified as lacking data for gout burden estimates. Sharing of unpublished data should be encouraged, especially for those needed for the burden of disease study for gout, by making them available through peer reviewed publications.

Conclusion

Gout was one of the five MSK disorders included in the GBD 2010 Study. The burden of gout continues to rise with the increased aging population throughout the world. Better management of gout is, therefore, needed in most regions of the world. An emphasis on public health education on diet and alcohol intake, and weight loss, in addition to management of gout in primary and secondary care, would help with minimising the burden, and could potentially prevent the disease onset. Better documentation of frequency and severity of attacks of gout at a population level, globally, is required to more accurately reflect the real disease burden.

Acknowledgments

The authors thank Dr Zeinab Slim for providing the unpublished Lebanon COPCORD 2008 data upon request, and Dr Joanna Makovey and Ms Jia Tian for their help with the translation of articles in Russian and Chinese, respectively. The authors are grateful for the expert advice of the following individuals, at various times during the course of the study: Professor Fernando Perez-Ruiz, A/Professor William Taylor, Dr Jed Blore, Dr Cesar Diaz-Torne, Dr Rebecca Grainger, Dr Neil McGill and Dr Michele Meltzer. The MSK Gout Expert Group comprised of the following individuals: ES, DH, MC, TRM, RB, AW and LM. The GBD Core Team member included on this paper is: TV.

References

View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • Handling editor Tore K Kvien

  • Contributors ES drafted and finalised the manuscript. All authors commented and gave feedback on the manuscript.

  • Funding Supported by the Bill and Melinda Gates Foundation (to DH and TV), the Australian Commonwealth Department of Health and Ageing (to ES and LM), University of Sydney Institute of Bone and Joint Research (to MC and ES), the Australian National Health and Medical Research Council (Postgraduate Scholarship 569772 to DH and Practitioner Fellowship to RB).

  • Competing interests None.

  • Patient consent No.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The authors thank Dr Zeinab Slim for providing the unpublished Lebanon COPCORD 2008 data upon request.

Linked Articles