Global burden of breast cancer and attributable risk factors in 195 countries and territories, from 1990 to 2017: results from the Global Burden of Disease Study 2017
Statistical data on the incidence, mortality, and burden of breast cancer and the relevant risk factors are valuable for policy-making. We aimed to estimate breast cancer incidence, deaths, and disability-adjusted life years (DALYs) by country, gender, age group, and social-demographic status between 1990 and 2017.
We extracted breast cancer data from the 2017 Global Burden of Disease (GBD) study from 1990 through 2017 in 195 countries and territories. Data about the number of breast cancer incident cases, deaths, DALYs, and the age-standardized rates were collected. We also estimated the risk factors attributable to breast cancer deaths and DALYs using the comparative risk assessment framework of the GBD study.
In 2017, the global incidence of breast cancer increased to 1,960,681 cases. The high social-development index (SDI) quintile included the highest number of breast cancer death cases. Between 2007 and 2017, the ASDR of breast cancer declined globally, especially in high SDI and high middle SDI countries. The related DALYs were 17,708,600 in 2017 with high middle SDI quintile as the highest contributor. Of the deaths and DALYs, alcohol use was the greatest contributor in most GBD regions and other contributors included high body mass index (BMI) and high fasting plasma glucose.
The increasing global breast cancer burden is mainly observed in lower SDI countries; in higher SDI countries, the breast cancer burden tends to be relieving. Therefore, steps against attributable risk factors should be taken to reduce breast cancer burden in lower SDI countries.
KeywordsBreast cancer Global cancer burden Disability-adjusted life years Alcohol use Incidence
Age-standardized death rate
Age-standardized incidence rate
Disability-adjusted life year
Estimated annual percentage change
Global burden of disease
Year lived with disability
Year of life lost
Breast cancer was the third highest incident cancer in 2017, with an estimated 1,960,681 (95% UI = 1,891,447–2,023,170) incident cases and a high prevalence in females. Breast cancer is the leading cause of cancer death in females and also a non-negligible cause of cancer death in males worldwide, claiming 181,004 lives and resulting in 17.7 million disability-adjusted life years (DALYs), making it one of the most severe burdensome cancer globally [1, 2]. Breast cancer incidence and mortality are still increasing, both in developing and developed countries . Although the survival rate in breast cancer has improved, it varies in different countries distinctly , due to factors such as lack of early-stage screening, detection, and cost-effective therapy . To better understand the enormous influence of this disease on public health, it is worthwhile to review and analyze the related global trends.
Thus far, several regional and national studies on breast cancer incidence and mortality have been performed, and the results of these multifarious studies from different parts of the world present an inclusive picture. Epidemiological studies from Arab countries , India , Latin America , and Europe  show an alarmingly rising burden with associated incidence and mortality. However, specific studies of breast cancer burden at a global level are lacking. The aim of this study was to describe the influence of geographical location, social-development index (SDI), age, and gender on the global trends in the incident cases, deaths, and DALYs of breast cancer based on data from the Global Burden of Disease from 1990 to 2017 in 195 countries and territories.
Materials and methods
Data for the disease burden of breast cancer were obtained from an online data source tool, the Global Health Data Exchange (GHDx) query tool (http://ghdx.healthdata.org/gbd-results-tool), which is an ongoing global collaboration that uses all available epidemiological data to provide a comparative assessment of health loss from 328 diseases across 195 countries and territories. From GBD study 2017, we obtained data on annual incidence, death, DALY, and respective age-standardized rate (ASR) of breast cancer from 1990 to 2017. The SDI, which is based on national-level income per capita, average years of education among persons older than 15, and total fertility rate, was used to categorize the countries into five SDI quintiles (high, high-medium, medium, low-medium, and low levels).
Estimation of cancer incidence was based on individual cancer registries or integrated databases of cancer registries. Systematic literature search was performed in PubMed to find the evidence for breast cancer deaths due to the attributable risk factors (alcohol use, high body mass index, high fasting plasma glucose, low physical activity, smoking, and secondhand smoke). For each included study, the proportions of breast cancer cases induced by the specific risk factors were calculated. The proportion data obtained from systematic literature review were applied to four independent DisMod-MR2.1 as inputs . Mortality data of breast cancer from vital registration systems and mortality estimates were used as input data into the CODEm (Cause of Death Ensemble Model) . The CODEm predicts mortality based on available data and covariates such as education, smoking, SDI, lagged distributive income, and alcohol use. Single cause estimates were adjusted to fit for the separately estimated all-cause mortality using the Cod-Correct algorithm [12, 13].
Years lived with disability (YLDs) were calculated by multiplying the prevalence of each sequela by its disability weight and by adding the procedure-related morbidity associated with breast cancer treatment. YLLs due to breast cancer were calculated using standard global life expectancy and the number of deaths according to age . Breast cancer DALYs were calculated as the sum of YLDs and YLLs.
The GBD study incorporated the comparative risk assessment framework previously  to quantify the burden of several causes and impairments attributable to 84 environmental, occupational, metabolic, and behavioral risk factors. Briefly, after assessing the casual evidence in each risk-outcome pair, we selected 2 components to model the attributable burden of causes to risks, including deaths and DALYs.
(ai, where i denotes the ith age class, and the number of persons (or weight) (wi) in the same age subgroup i of the selected reference standard population.)
Moreover, trends in ASIR reflect the alterations in human disease patterns and risk factors. The concept of EAPC is introduced to describe the trends in ASR within a specified time interval, as it is assumed that the natural logarithm of ASR is linear along with time. Thus, Y = α+βX+ε, where Y refers to ln(ASR), X represents calendar year, and ε represents error term. Based on this formula, β determines the positive or negative trends of ASR. The formula for calculating EAPC is EAPC = 100 × (exp(β) − 1) and 95% confidence intervals are obtained from the linear model [16, 17]. It is shown that when EAPC and the lower boundary of the confidence interval are positive, then ASR is in an upward trend. Conversely, when EAPC and the upper boundary of the confidence interval are negative, the ASR is in a descending trend. All statistical analyses were performed using the R program (Version 3.5.2, R core team). A p value of less than 0.05 was considered statistically significant.
Breast cancer incidence burden
The incidence cases and age-standardized incidence of breast cancer in 1990 and 2017, and its temporal trends from 1990 to 2017
No. × 103 (95% UI)
ASR per 100,000
No. (95% UI)
No. × 103 (95% UI)
ASR per 100,000
No. (95% UI)
No. (95% CI)
− 0.13 (− 0.25–0.01)
− 0.10 (− 0.29–0.09)
High-income Asia Pacific
High-income North America
− 1.04 (− 1.20–0.88)
Southern Latin America
0.02 (− 0.14–0.19)
Andean Latin America
Central Latin America
Tropical Latin America
North Africa and Middle East
Central Sub-Saharan Africa
Eastern Sub-Saharan Africa
Southern Sub-Saharan Africa
Western Sub-Saharan Africa
Breast cancer deaths and DALY burden
Breast cancer death cases, DALYs, and related age-standardized rates
All ages × 103 (95% UI)
Age-standardized rate/100,000 (95% UI)
2017 no. male + female
2017 no. male
2017 no. female
% 1990–2017 change
2017 male + female
% 1990–2017 change
− 11.66 (− 20.06–6.02)
− 9.24(− 19.51–1.88)
− 32.23 (− 34.22–30.42)
− 31.67 (− 34.00–29.53)
High middle SDI
− 8.37(− 22.48–2.52)
− 12.17(− 26.10–5.59)
10.59 (− 11.08–22.71)
9.12 (− 12.08–22.40)
24.10 (− 8.76–52.76)
24.63 (− 9.51–57.13)
26.93 (− 8.42–56.46)
22.74 (− 12.23–54.50)
Regionally, a number of breast cancer deaths were the highest in South Asia during the study period, reaching 108,966 (95% UI = 93,488–131,457) cases in 2017. Meanwhile, South Asia showed the largest increase of breast cancer deaths between 1990 and 2017. Only in high-income North America, Western Europe, and Australasia, ASDR decreased. The countries with the largest populations, including China, India, and the USA, had the most deaths in 2017 while the ASDR was highest in Fiji, Tonga, and the Bahamas (Fig. 2b). South Asia, East Asia, and Western Europe were the areas with the highest breast cancer-related DALYs in 2017. Among the 21 GBD regions, only Western Europe showed a declined trend of DALY cases. The age-standardized DALY rate increased in 12 regions, while it decreased in 9 GBD regions. The highest number of DALY cases was observed in China, India, and the USA, whereas the Bahamas, Nigeria, and Tonga were the 3 countries with the highest age-standardized DALY rate (Fig. 2c).
As shown in Additional file 2: Figure S2a and b, the proportion of incidence in the three age groups remained stable between 1990 and 2017. Regionally, young breast cancer incidence was relatively higher in Oceania and Southeast Asia, and elderly breast cancer incidence was higher in Western Europe and the USA. However, young breast cancer incidence decreased, but elderly breast cancer incidence increased visibly in high-income Asia Pacific region.
Risk factors attributable to breast cancer burden
High body mass index led to 4.61% (95% UI = 1.52–8.83%) of global breast cancer DALYs in 2017 with an increasing trend between 1990 and 2017. Except the gentle trend of high BMI attributed DALYs in the high SDI quintile, the other quintiles showed increased trends, especially in the high middle SDI and the middle SDI countries (Fig. 6a).
In 2017, 6.07% of breast cancer DALYs was attributable to high fasting plasma glucose (95% UI = 1.15–13.53%) with a slight increase. In the high SDI quintile, the ASR has been decreasing since 1990, when it was the highest among the five quintiles. However, the rate was of great fluctuation between 1990 and 2017 while the low SDI, low middle SDI, and the middle SDI countries exhibited distinct increase trends (Fig. 6a).
The deaths attributable to risk factors mirrored the same pattern of attributable DALYs (Fig. 6b). With regard to risk factors, there was some difference in gender. In female breast cancer population, the global disease burden database indicated 6 risk factors of DALYs and deaths including alcohol use, high body mass index, high fasting plasma glucose, low physical activity, smoking, and second hand smoke. However, in males, the risk factors included alcohol use and second hand smoke, and the rate of contribution in 2017 was 20.18% (15.37–25.05%) and 1.51% (0.35–2.61%), respectively.
As far as we know, this GBD-based study reveals the most up-to-date trends and patterns of the incidence, mortality, and DALYs associated with breast cancer worldwide and the most relevant risk factors. Our analysis revealed that with the 1,960,681 new cases in 2017, the global incident cases of breast cancer increased by 123% between 1990 and 2017, but changes in ASIR (16% increasing) were less prominent. According to the global cancer burden 2016 , the changes in breast cancer incident cases were mainly attributable to population growth (12.4%) and aging (15.7%) rather than age-specific incidence rates (0.9%). Specifically, in the high SDI and the high middle SDI countries, the incident cases attributable to change of age-standardized incidence rate were close to none or even negative. On the contrary, changes in incidence rate contributed a lot in the middle SDI and the low middle SDI countries while in the low SDI quintile, incident cases due to population growth accounted for the largest proportion.
The trend of global age-standardized DALY rate was considerably varied in females and males, between 1990 and 2017. There was a gentle decline of age-standardized DALY rate in females, while the rate was increasing in males in the same study period. This discrepancy partially reflected the consequence of different risk factors between the genders. The age-standardized DALY rate also differed in the SDI quintiles. In the low SDI, low middle SDI, and the middle SDI countries, the rate was increasing both in males and females, while in the high middle SDI countries, the DALY rate had the similar global pattern and in the high SDI countries, the rate was declining in both the genders. The global number of breast cancer deaths increased substantially during the study period in both genders. This increase was consistent with the increase in breast cancer incident cases. Global ASDR declined since 1990 which was mainly due to the decrease in the high SDI and the high middle SDI quintiles, while the ASDR of other three quintiles slightly increased, between 1990 and 2017. Much of these geographic and gender disparities could be explained by heterogeneity in the prevalence of risk factors, with alcohol use being the chief factor.
Alcohol use is the most important of the risk factors contributing to breast cancer deaths and DALYs, though the attributable DALYs have been declining from 1990 to 2017. Consistently, the global daily prevalence of alcohol use declined significantly during the study period, which was more pronounced in the high and the high middle SDI countries . This finding was consistent with the observed decrease in breast cancer ASDR and age-standardized DALY rate. The perniciousness of alcohol is instantiated both in the genetic level (enhancement of DNA damage, interference with mitochondrial function) and in the epigenetic level (affecting DNA methylation status and histone modification) . Notably, there is a possible dose-response relationship between alcohol drinking and breast cancer .
Moreover, high body mass index and high fasting plasma glucose were also identified as potential risk factors attributable to breast cancer deaths and DALYs, which showed an increasing contribution trend of the two risk factors globally, especially in the high SDI and the high middle SDI quintiles. According to Hyuna Sung’s study , the most conspicuous obesity increase occurred among males in high-income Western countries and among females in Central Asia, the Middle East, and North Africa, which is considered to be caused by global food system changes promoting high-calorie, low-nutrient foods, accompanied by decreased physical activity. Increase of BMI is associated with increased breast cancer risk, and it is more risky in Asians, when compared with North Americans and Europeans. Though acceleration of national wealth has been in accordance with an increasing in body weight , prosperity is not always correlated with high body mass: obesity rate is quite low in high-income Asian Pacific countries, which is likely due to their traditional low-calorie dietary habits and physical activities such as daily walking [24, 25]. Nevertheless, obesity prevalence is extremely high in some low-income countries, such as some Pacific Island nations and Egypt .
It is worth mentioning that risk factors differed in the subtypes of breast cancer , for instance, high body mass index was associated with an increased risk of triple negative breast cancer, while low physical activity contributed to attributable risk of ER+/PR+ subtypes. On the one hand, this discrepancy possibly resulted from the conversion of androgen to estrogen in adipose tissue, which had a more severe influence on hormone receptor (HR)–positive breast cancer types . On the other hand, leptin and other hormones could have exerted stimulating effects on HR–negative breast cancer cell proliferation, invasion, and angiogenesis, either directly or indirectly .
Policy-makers require country-specific information on the burden of different cancers to assess the impact of cancer control programs, benchmark progress in their nation, and allocate limited resources in their health care systems. Given the fact that existing data in many countries are of low quality or missing, the GBD study results can be used by stakeholders to study the trends of different diseases in their respective locations.
GBD studies provide high-quality estimates of global cancer burden, yet there exist several limitations. One inevitable limitation is the uncertainty of GBD estimates in cases in which actual data on disease burden are unavailable, and the GBD estimates fill the vacancies in this occasion. Besides, differences in data collecting and coding, as well as quality of data sources, remain inevitable in this analysis pattern. Moreover, the fluctuations in incidence and mortality rates may partly reflect the detection bias related to adjustments in screening protocols instead of real changes in age-specific rates.
The global burden of breast cancer has been increasing continuously between 1990 and 2017, although in some SDI quintiles, the ASDR and age-standardized DALY rate has been declining. In recent years, disease reduction was observed in higher SDI regions while lower SDI regions had carried an incremental burden of breast cancer, and there may be a widening in disparities in the years to come. Consequently, steps against attributable risk factors should be taken to reduce breast cancer burden especially in lower SDI countries, to prevent acceleration of these disparities, because underdeveloped countries are affected to a greater degree by increases in health burden.
We highly appreciate the works by the Global Burden of Disease Study 2017 collaborators.
LN contributed to study design, data collection, data analysis, data interpretation, and writing of the manuscript. YJ D contributed study conduct, data collection, and writing of the manuscript. LH Z contributed data collection, analysis, and interpretation. TT contributed review, interpretation, and analysis. SY contributed data collection, data analysis, and data interpretation. YW contributed data collection, data analysis, and data interpretation. YZ contributed data collection, data analysis, and data interpretation. ZZ contributed data collection, data analysis, and data interpretation. QH contributed data collection, data analysis, and data interpretation. DL S contributed data collection, data analysis, and data interpretation. DZ contributed data collection, data analysis, and data interpretation. HF K contributed in writing and editing of the manuscript. ZJ D contributed study concept and design, chairing of steering committee, oversight of study implementation, extensive data analysis and interpretation, and writing and approval of final version of manuscript. The other authors read and approved the final manuscript.
This study was supported by National Natural Science Foundation of China (No. 81471670); The Key research and development plan, Shaanxi Province, China (2017ZDXM-SF-066).
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
- 3.Collaboration GBoDC. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2016: a systematic analysis for the Global Burden of Disease Study Global Burden of Cancer, 1990 to 2016Global Burden of Cancer, 1990 to 2016. JAMA oncology. 2018;4:1553–68.CrossRefGoogle Scholar
- 18.Collaboration GBoDC. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 32 cancer groups, 1990 to 2015: a systematic analysis for the Global Burden of Disease Study Global Burden of Cancer 2015Global Burden of Cancer 2015. JAMA oncology. 2017;3:524–48.CrossRefGoogle Scholar
- 20.Zakhari S, Hoek JB. Epidemiology of moderate alcohol consumption and breast cancer: association or causation? Cancers. 2018;10: undefined.Google Scholar
- 22.Sung H, Siegel RL, Torre LA, et al. Global patterns in excess body weight and the associated cancer burden. CA: a cancer journal for clinicians. 2019;69:88–112.Google Scholar
- 28.Althuis MD, Fergenbaum JH, Garcia-Closas M, Brinton LA, Madigan MP, Sherman ME. Etiology of hormone receptor-defined breast cancer: a systematic review of the literature. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2004;13:1558–68.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.