The global, regional, and national burden of benign prostatic hyperplasia in 204 countries and territories from 2000 to 2019: a systematic analysis for the Global Burden of Disease Study 2019

1Benign prostatic hyperplasia is a multifocal, non-malignant, hyperplastic, and progressive histopathological change in stromal and epithelial cells in the transitional zone of the prostate, resulting in discrete prostatic nodules, inflammation, fibrosis, and changes in smooth muscle activity, which can cause partial or complete obstruction of the urethra., The resulting bladder outlet obstruction, coupled with increased muscle tone of the bladder and secondary dysfunction of the detrusor, produce lower urinary tract symptoms.

2Benign prostatic hyperplasia is a common urological disease among older men. The age-specific prevalence of benign prostatic hyperplasia has been estimated from autopsy studies to be 8% in the fourth decade of life, 50% in the sixth decade of life, and 80% in the ninth decade of life., The annual prostatic volume increment with age, based on Krimpen and Baltimore's longitudinal study of ageing, is about 2·0–2·5% per year in older men., There is some evidence to suggest the prevalence varies by race and ethnicity. Other factors associated with benign prostatic hyperplasia include metabolic syndrome, obesity, increased BMI, dyslipidaemia, diabetes, cardiovascular disease, acute and chronic prostatic inflammation, functional bladder capacity, treatment for cardiac disease, post-void residual urine volume, educational level, antidepressant use, calcium antagonist use, erectile function or dysfunction, high concentrations of prostate disease-specific antigen, family history of bladder cancer, and family history of prostatic disease, whereas an inverse association has been observed with increased physical exercise, moderate alcohol consumption, and smoking.,,,,,,,,,,,,,

3Evidence before this study Given that the global population is both growing and ageing, addressing the burden of age-associated diseases, such as benign prostatic hyperplasia, has become a global health priority. Although a systematic review was not conducted before producing and reporting these estimates, we did identify many studies that have examined the prevalence of benign prostatic hyperplasia in different communities. However, varied case definitions, research methodologies, access to care, diagnostic modalities, and coding practices created a challenge of assessing and comparing the disease burden across different populations and over time. The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) assembles data from diverse sources worldwide and applies standardised data processing and modelling techniques to facilitate like-versus-like comparisons across time and space. Previous GBD publications on multiple diseases, and a previous study by Launer and colleagues based on GBD 2017 estimates, reported a rising burden of benign prostatic hyperplasia worldwide. Added value of this study The present study overcomes some of the limitations of epidemiological studies conducted in one or a few locations by making use of the GBD modelling methods to leverage international administrative data and systematically estimate the prevalence of, and disability-adjusted life-years (DALYs) associated with, benign prostatic hyperplasia in 204 countries and territories between 2000 and 2019. We extended and improved upon the results of GBD 2017 reported by Launer and colleagues by providing estimates for additional countries and territories as well as more recent estimation years, incorporating more location-years of prevalence data, applying a novel method to adjust for systematic bias in prevalence data sources, and incorporating the use of a validated instrument to estimate benign prostatic hyperplasia symptom severity. We explored both absolute and age-standardised rates of benign prostatic hyperplasia burden at different levels of geographical hierarchy, over time, and across countries with different Socio-demographic Index (SDI) rankings. We found that the absolute burden of benign prostatic hyperplasia is rising across all SDI quintiles, largely driven by demographic changes, but that the age-standardised rate of benign prostatic hyperplasia appears to be increasing at low SDI quintiles and appears to be decreasing at higher SDI quintiles. As the first formal report of the global benign prostatic hyperplasia burden prepared by the GBD Benign Prostatic Hyperplasia Collaborators, this study also provides the most nuanced discussion to date of the strengths and limitations of the available dataset and current methods for estimation of the benign prostatic hyperplasia burden, which should help guide primary data collection in the future. Implications of all the available evidence Our study shows that the burden of benign prostatic hyperplasia is increasing in many parts of the world. This increase was most notable in low-income and middle-income countries that are undergoing rapid demographic and epidemiological transitions, highlighting the importance of monitoring and planning for future health system strain. The comprehensive nature of this study provides important policy-relevant information to health-care professionals, policy makers, and international health organisations to assess the burden of benign prostatic hyperplasia at the global, regional, and country levels; prepare effective public health awareness campaigns; and implement effective diagnostic, prevention, and treatment strategies to manage the growing burden of benign prostatic hyperplasia.

4Evidence before this study

5Given that the global population is both growing and ageing, addressing the burden of age-associated diseases, such as benign prostatic hyperplasia, has become a global health priority. Although a systematic review was not conducted before producing and reporting these estimates, we did identify many studies that have examined the prevalence of benign prostatic hyperplasia in different communities. However, varied case definitions, research methodologies, access to care, diagnostic modalities, and coding practices created a challenge of assessing and comparing the disease burden across different populations and over time. The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) assembles data from diverse sources worldwide and applies standardised data processing and modelling techniques to facilitate like-versus-like comparisons across time and space. Previous GBD publications on multiple diseases, and a previous study by Launer and colleagues based on GBD 2017 estimates, reported a rising burden of benign prostatic hyperplasia worldwide.

6Added value of this study

7The present study overcomes some of the limitations of epidemiological studies conducted in one or a few locations by making use of the GBD modelling methods to leverage international administrative data and systematically estimate the prevalence of, and disability-adjusted life-years (DALYs) associated with, benign prostatic hyperplasia in 204 countries and territories between 2000 and 2019. We extended and improved upon the results of GBD 2017 reported by Launer and colleagues by providing estimates for additional countries and territories as well as more recent estimation years, incorporating more location-years of prevalence data, applying a novel method to adjust for systematic bias in prevalence data sources, and incorporating the use of a validated instrument to estimate benign prostatic hyperplasia symptom severity. We explored both absolute and age-standardised rates of benign prostatic hyperplasia burden at different levels of geographical hierarchy, over time, and across countries with different Socio-demographic Index (SDI) rankings. We found that the absolute burden of benign prostatic hyperplasia is rising across all SDI quintiles, largely driven by demographic changes, but that the age-standardised rate of benign prostatic hyperplasia appears to be increasing at low SDI quintiles and appears to be decreasing at higher SDI quintiles. As the first formal report of the global benign prostatic hyperplasia burden prepared by the GBD Benign Prostatic Hyperplasia Collaborators, this study also provides the most nuanced discussion to date of the strengths and limitations of the available dataset and current methods for estimation of the benign prostatic hyperplasia burden, which should help guide primary data collection in the future.

8Implications of all the available evidence

9Our study shows that the burden of benign prostatic hyperplasia is increasing in many parts of the world. This increase was most notable in low-income and middle-income countries that are undergoing rapid demographic and epidemiological transitions, highlighting the importance of monitoring and planning for future health system strain. The comprehensive nature of this study provides important policy-relevant information to health-care professionals, policy makers, and international health organisations to assess the burden of benign prostatic hyperplasia at the global, regional, and country levels; prepare effective public health awareness campaigns; and implement effective diagnostic, prevention, and treatment strategies to manage the growing burden of benign prostatic hyperplasia.

10Previous studies have shown that benign prostatic hyperplasia contributes to increased health costs and decreased quality of life.,, It is associated with serious morbidities, including an increased risk of falls, depression, and diminished health-related quality of life based on indicators such as sleep, psychological condition, activities of daily living, and sexual wellbeing. The effects of benign prostatic hyperplasia are not only seen on the patient but also on the patient's family and on society at large., Beyond its immediate effect on morbidity, benign prostatic hyperplasia is also associated with complications such as urinary tract infection, acute urinary retention, urolithiasis, and acute renal failure.,,

11Benign prostatic hyperplasia has been identified as a major urological health problem in older men in many countries., The prevalence of benign prostatic hyperplasia from descriptive epidemiology studies ranges from 12% to 42%,,, and one study estimated the lifetime risk of benign prostatic hyperplasia to be 29%. A systematic review and meta-analysis by Lee and colleagues in 2017 identified 30 population-based, hospital-based, and community-based epidemiological studies in different countries, and yielded a 26% point prevalence of benign prostatic hyperplasia in older men for the years 1990–2016. Another meta-analysis done in China indicated that the pooled overall prevalence of benign prostatic hyperplasia among men aged 40 years or older was 36·6% during 1989–2014.

12The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) is the largest and most comprehensive scientific effort to produce estimates of health loss due to 369 diseases and injuries. As such, GBD overcomes some of the limitations of the epidemiological studies described above by utilising a large number of administrative datasets from around the world, processing them all in a comparable fashion, and using regional patterns and predictive covariates to provide the most precise annual estimates of disease burden possible for a large number of countries and territories, including those without primary data, over a long time series. The burden of benign prostatic hyperplasia has been estimated and included in comprehensive reports since GBD 2010,,,,,, and Launer and colleagues previously reported the global benign prostatic hyperplasia burden in GBD 2017. In this report, we extended and improved upon the GBD 2017 results reported by Launer and colleagues by providing estimates for additional countries and territories and more recent estimation years, incorporating more location-years of prevalence data, applying a novel method to adjust for systematic bias in prevalence data sources, and incorporating the use of a validated instrument to estimate benign prostatic hyperplasia symptom severity. We also provided a more detailed account of the dataset and current methods for estimation of the benign prostatic hyperplasia burden to stimulate substantive engagement on data collection priorities and future directions for improvement of estimations.

13With the population rapidly ageing in many parts of the world, the burden of benign prostatic hyperplasia is expected to rise. Understanding the current burden of and recent trends in benign prostatic hyperplasia, the role of demographic and other factors in driving the change, and the strengths and limitations of existing datasets is necessary to fill data gaps and help health systems prepare for the challenges associated with this rising global burden.

14This manuscript was produced as part of the GBD Collaborator Network and in accordance with the GBD Protocol.

15A comprehensive description of GBD study aims, data sources, methodologies, and analytical tools has been reported previously. The methods specific to the estimation of health loss due to benign prostatic hyperplasia are summarised below. The analysis presented here complies with the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER) statement. All input data and the code used to execute all data processing and modelling described below can be found on the Global Health Data Exchange (GHDx) website.

16In this study, we defined a case of benign prostatic hyperplasia on the basis of an individual receiving that diagnosis in a clinical encounter, as ascertained from administrative data, using the codes of the International Classification of Diseases, version 9 (ICD-9), and version 10 (ICD-10). The ICD-9 codes used were 600, 600.0, 600.1, 600.2, 600.3, and 600.9, and the ICD-10 codes used were N40, N40.0, N40.1, N40.2, N40.3, and N40.9. Ascertainment via diagnostic codes in administrative data necessarily means that not all individuals have had a histological diagnosis of benign prostatic hyperplasia, and receipt of clinical care suggests these individuals could be properly described as having lower urinary tract symptoms due to benign prostatic obstruction; for consistency with previous GBD publications and visualisations, we refer to these as cases of benign prostatic hyperplasia in this Article. Complications of benign prostatic hyperplasia that can be mapped to other GBD-defined diseases, such as urinary tract infections, kidney stones, and chronic kidney disease, were not included in the analysis. Only non-fatal health loss was estimated in this analysis, because life-threatening complications of benign prostatic hyperplasia are classified in other GBD-defined diseases and mortality due to these complications should not be double-counted.

17We used international clinical administrative data to estimate the prevalence of benign prostatic hyperplasia. Clinical administrative data included claims from three locations—the USA, Taiwan (province of China), and Poland—and hospital inpatient admissions data from 45 locations. The claims data from the USA consisted of more than 12 billion claims records from the commercially insured population in 2000 and 2010–16. Claims data from Taiwan (province of China) were from the national insurance programme covering more than 99% of the population in 2016, and claims data from Poland were from the national insurance programme covering more than 90% of the population in 2015–17.,, The inpatient admission records came from 297 sources, each covering 1–5 years of data between 1980 and 2018 (figure 1). A complete list of prevalence data sources is available in appendix 2 (pp 22–30) and on the GHDx website.

18The processing of these data sources has been described in detail elsewhere. Briefly, for claims data, we used the unique enrollee identification numbers to link inpatient and outpatient claims to a single individual. An enrollee was then extracted as a prevalent case if they had at least one inpatient or two outpatient medical encounters with any of the defining ICD codes for benign prostatic hyperplasia in either the primary or secondary diagnostic position. Processing of inpatient admissions data involved converting benign prostatic hyperplasia admission counts into cause fractions, which are the number of hospital admissions specifically for benign prostatic hyperplasia divided by the total number of admissions in the data year. Then, these cause fractions were multiplied by the hospital admission rate per capita and the total population size for each unique source, age, sex, and year combination to convert data to population-level inpatient admission rates. Details of the data and methods used to estimate hospital admission rates and population sizes have been previously described. Population-level benign prostatic hyperplasia admission rates were then transformed to population prevalence data by applying a ratio of total benign prostatic hyperplasia cases to inpatient benign prostatic hyperplasia admissions modelled from claims data.

19We treated claims data and outpatient-adjusted inpatient admission data from Taiwan (province of China) and Poland as reference data, meeting our ICD-based case definition, and representing a general population defined only by year, age, sex and geographical location. The claims data from the USA, however, were adjusted to account for selection bias since these data come from a database of commercial insurance claims, and enrolment in commercial health insurance is generally associated with higher economic status. The adjustment coefficient was estimated with a Bayesian, regularised, trimmed meta-regression (MR-BRT) analysis. MR-BRT is a mixed-effects meta-regression tool that accounts for between-study heterogeneity and has been described previously.,, Once all data were standardised to the reference standard, and the uncertainty of the adjustment process was accounted for, we applied the median absolute deviation (MAD) exclusion criterion to systematically exclude unreasonably high or low datapoints as outliers. Specifically, this was done by calculating the MAD of the age-standardised prevalence of all data and marking any data series greater than two MAD from the median as outliers and excluding them from the analysis (8·8% of processed datapoints were marked as outliers on the basis of the MAD exclusion criterion).

20We used DisMod-MR version 2.1 to estimate the age-specific, year-specific, and location-specific prevalence of benign prostatic hyperplasia. DisMod-MR is a Bayesian mixed-effects meta-regression tool that was designed for disease modelling by the Institute for Health Metrics and Evaluation (Seattle, WA, USA).,, The tool uses a compartmental model with a series of age-integrated differential equations to estimate a set of epidemiological measures (prevalence, incidence, remission, excess mortality rate, relative risk, and cause-specific mortality rate) that are internally consistent with one another. Estimation occurs at each level of a geographical cascade (ie, global, seven GBD super-regions, 21 regions, and 204 countries and territories), in which each subsequent model borrows information from the previous model in the cascade via a Bayesian prior. A Gaussian data likelihood function is used at each step in the cascade. We provided value priors on incidence, remission, and excess mortality. First, we assumed a benign prostatic hyperplasia incidence of zero in men younger than 40 years. The maximum disease duration above the age of 40 years was set at 10 years, and we assumed excess mortality rates of zero for all ages. We included age-standardised prevalence of diabetes as a predictive covariate because of its known association with benign prostatic hyperplasia. In previous rounds of GBD, we had tested mean BMI as a predictive covariate. This was dropped in GBD 2019, however, because it did not have a significant association with benign prostatic hyperplasia and did not improve prediction.,,,,

21The estimates from the most granular level of location, either at the country or subnational level, were aggregated up along the geographical cascade to obtain the final regional, super-regional, and global prevalence of benign prostatic hyperplasia. Age-specific prevalence was estimated for 12 age groups: 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, 65–69 years, 70–74 years, 75–79 years, 80–84 years, 85–89 years, 90–94 years, and 90 years and older. Age-standardised prevalence was estimated across all age groups through the direct method and the GBD reference population. Input data, outlier designations, and model fit for the benign prostatic hyperplasia DisMod MR model can be viewed on the Epi Visualization Viz Hub.

22The disability weight represents the severity of health loss associated with a generic health state, described in lay language. It ranges from 0 (perfect health) to 1 (death). Disability weights for generic health states were estimated by use of nine large population-based surveys and one open-access internet survey where participants were asked to compare pairwise combinations of health states.,, Two health states were assigned to benign prostatic hyperplasia: asymptomatic and symptomatic. The symptomatic health state has a disability weight of 0·067 (95% CI 0·043–0·097). The asymptomatic health state has a disability weight of zero, assuming no 95% CIs.

23To determine what proportion of the estimated benign prostatic hyperplasia prevalence to assign to the symptomatic health state, we used data from four community-based surveys from Japan, the USA, Scotland, and France that recruited men aged 40–84 years. The surveys used the International Prostate Symptom Score (I-PSS), a validated questionnaire, to measure the severity of lower urinary tract symptoms among men.,, We modelled the cumulative distribution of the I-PSS scores in survey participants using MR-BRT to estimate the mean proportion of individuals with symptomatic lower urinary tract symptoms. The symptomatic and asymptomatic proportions were applied to the prevalence estimates from DisMod-MR to produce estimates of the prevalence of symptomatic and asymptomatic benign prostatic hyperplasia.

24YLDs due to benign prostatic hyperplasia were estimated by multiplying the disability weights by the prevalence of symptomatic and asymptomatic benign prostatic hyperplasia and summing them together. Because we did not estimate mortality for benign prostatic hyperplasia, we did not produce years of life lost (YLLs) for benign prostatic hyperplasia. Therefore, DALYs equalled YLDs. Final estimates by year, age, and location, including those included in the text, figures, tables, and appendix 2 of this Article, can be viewed on the GBD Compare Viz Hub.

25The Socio-demographic Index (SDI) is a summary measure that describes a country's development. The SDI is derived from a country's total fertility rate for women younger than 25 years, educational attainment in adults, and lag-distributed income per capita. Details of the SDI estimation methods are available elsewhere. We grouped all GBD countries and territories in five SDI quintiles on the basis of their SDI values in 2019. For this analysis, we used the most recent year's SDI groupings when describing trends between 2000 and 2019. The list of locations and their assigned SDI quintile can be found in appendix 2 (pp 8–13).

26The 95% uncertainty intervals (UIs) were estimated by taking 1000 draws of the distribution of every modelling and computation process. The final mean estimate was calculated by taking the mean value of the 1000 draws, and the 95% UI was set by finding the 25th and 975th of their ordered values. Data are presented to three significant figures.

27The funder of the study had no role in study design, data collection, data analysis, data interpretation, or the writing of the report.

28In 2019, there were 94·0 million (95% UI 73·2 to 118) prevalent cases of benign prostatic hyperplasia globally among men aged 40 years and older, corresponding to an age-standardised prevalence of 2480 (1940 to 3090) per 100 000 (table). This was a 70·5% (95% UI 68·6 to 72·7) increase from 51·1 million (43·1 to 69·3) cases in 2000. The global age-standardised prevalence, however, remained largely unchanged during this period (–0·770% [–0·0912 to 1·56] difference).

29Globally, men aged 65–74 years shared the greatest absolute burden of benign prostatic hyperplasia (figure 2), accounting for 42% of the total prevalent cases among men aged 40 years and older. The age-specific prevalence was highest in men aged 75–79 years, at 24 300 (95% UI 18 600–31 500) per 100 000, followed by the those aged 80–84 years, at 23 500 (17 800–30 400) per 100 000, and those aged 70–74 years, at 22 200 (16 100–29 400) per 100 000. Between 2000 and 2019, the number of prevalent cases of benign prostatic hyperplasia increased rapidly in all age groups (appendix 2 p 31). In men aged 40–44 years, the percentage increase was 22·6% (16·7–26·8). For men aged 80 years and older, the percentage increase was 173% (166–179).

30There was substantial geographical variation in the prevalence of benign prostatic hyperplasia in 2019 (figure 3A). The highest age-standardised prevalence was observed in eastern Europe (6480 [95% UI 5130–8080] per 100 000), followed by central Latin America (4140 [3340–5090] per 100 000) and Andean Latin America (3610 [2700–4610] per 100 000). The lowest age-standardised prevalence was recorded in north Africa and the Middle East (987 [732–1320] per 100 000) and three sub-Saharan African regions: eastern sub-Saharan Africa (1160 [852–1540] per 100 000), western sub-Saharan Africa (1120 [817–1490] per 100 000), and central sub-Saharan Africa (1110 [812–1480] per 100 000). The age-standardised prevalence in five high-income regions (western Europe, high-income North America, high-income Asia Pacific, Australasia, and southern Latin America) ranged from 2250 (1800–2790) per 100 000 in western Europe to 1180 (906–1550) per 100 000 in the high-income Asia Pacific in 2019 (table).

31Prevalent cases of benign prostatic hyperplasia increased in all 21 GBD regions between 2000 and 2019, with the percentage change ranging from 19·3% to 101%. The greatest increase was noted in central Latin America, with a 101% change (95% UI 96·5–105), followed by tropical Latin America (98·4% [93·5–104]), east Asia (97·5% [93·3–102]), and south Asia (91·4% [87·4–96·3]). The smallest increase was noted in three European regions (eastern, central, and western Europe). Despite this steady increase in the number of prevalent cases, 17 regions had a less than 2% change in age-standardised prevalence during the same period (figure 4). Of the remaining four regions, only two had change estimates that excluded zero in their uncertainty intervals: the high-income Asia Pacific, which saw a 3·92% (1·48–6·37) decrease, and high-income North America, which saw a 2·93% (1·42–4·44) increase (table).

32The global distribution of the benign prostatic hyperplasia burden, as represented by age-standardised DALY rates, was the same as that for prevalence (figure 3A, B). In 2019, benign prostatic hyperplasia was responsible for 1·86 million (95% UI 1·13–2·78) DALYs globally, equating to an age-standardised DALY rate of 48·9 (29·7–72·6) per 100 000. Consistent with the findings for prevalence, the majority of the disease burden was concentrated in countries in eastern Europe, central Latin America, Andean Latin America, and southeast Asia. In 2019, the age-standardised DALY rate was 128 (76·5–190) per 100 000 in eastern Europe, 81·8 (50·2–121) per 100 000 in central Latin America, 71·9 (43·1–109) per 100 000 in Andean Latin America, and 69·7 (41·8–104) per 100 000 in southeast Asia (appendix pp 32–48). The temporal variation of DALYs was the same as that for prevalence.

33At the national level, the age-standardised prevalence ranged from 949 per 100 000 to 6910 per 100 000 across countries and territories in 2019 (table). The highest age-standardised prevalences of benign prostatic hyperplasia were observed in Lithuania (6910 [95% UI 5830 to 7940] per 100 000), Russia (6510 [5110 to 8130] per 100 000), and Ukraine (6450 [5030 to 8050] per 100 000), whereas the lowest age-standardised prevalences were observed in Yemen (949 [702 to 1280] per 100 000), Syria (949 [694 to 1290] per 100 000), Sudan (958 [703 to 1300] per 100 000), and Lebanon (959 [706 to 1310] per 100 000). As noted for many regional estimates above, the percentage change in the age-standardised prevalence of individual countries during the 2000–19 period was small and often non-significant; the highest percentage increases in age-standardised prevalence were noted in Mauritius (9·81% [1·51 to 18·1]), followed by Vanuatu (5·53% [–1·73 to 13·6]), and the Solomon Islands (5·35% [–1·68 to 15·1]), whereas the greatest decreases were recorded in Brunei (4·00% [–5·61 to 12·4]), Belgium (3·91% [–6·96 to 20·1]), Ecuador (3·45% [–4·54 to 10·2]), and Singapore (3·18% [–4·46 to 11·1]; table).

34We also assessed the disease burden in five SDI quintiles (figure 5). Between 2000 and 2019, the majority of the absolute DALY burden of benign prostatic hyperplasia was concentrated in the high-middle and middle SDI quintiles, with the fewest DALYs in the low SDI quintile. All five SDI quintiles observed an increase in the absolute DALY burden between 2000 and 2019. The most rapid increases were seen in the middle SDI quintile (94·7% [95% UI 91·8–97·6]), the low-middle SDI quintile (77·3% [74·1–81·2]), and the low SDI quintile (77·7% [72·9–83·2]). The high SDI quintile saw a 55·2% (52·6–58·2) increase and the high-middle SDI quintile saw a 52·8% (49·4–56·3) increase. During the same period, the high and high-middle SDI quintiles saw a small decrease in the age-standardised DALY rate, whereas the middle, low-middle, and low SDI quintiles saw small increases. The greatest increase in the age-standardised DALY rate was seen in the low SDI quintile (5·28% [2·41–8·36]). In contrast to the patterns we observed in the absolute DALY burden, the low-middle SDI quintile had the highest age-standardised DALY rate in 2019, surpassing that of the high-middle SDI quintile in 2016.

35We present a comprehensive assessment of the temporal and geographical patterns of the benign prostatic hyperplasia burden from GBD 2019. Our findings are consistent with previous reports, showing that the absolute disease burden is rising in many parts of the world. The global number of prevalent cases almost doubled in the past 20 years. Despite the increase in the absolute benign prostatic hyperplasia burden, the global age-standardised prevalence and DALY rates remained largely unchanged during the study period, suggesting that population growth and ageing have a greater impact on driving the increased prevalence of, and DALYs associated with, benign prostatic hyperplasia at the global level than other risk factors for benign prostatic hyperplasia do.

36Our study shows that the peak absolute benign prostatic hyperplasia burden occurred in men aged 65–69 years and the age-specific prevalence was highest in men aged 75–79 years. This trend contrasts with the age trend found in autopsy studies, where the histological prevalence continues to rise with advancing age,,,, but was similar to the age trend found in community-based studies, where the diagnosis of benign prostatic hyperplasia was made on the basis of lower urinary tract symptoms and prostatic enlargement in clinical practice.,,, Geographically, the age-standardised prevalence and DALY rates were lowest in countries in north Africa and the Middle East and sub-Saharan Africa, and highest in countries in eastern Europe. Although this geographical variation could be attributable to the varying stages at which each country is undergoing demographic and epidemiological transitions, it could also partially be explained by differences in the underlying risk factors in these populations.

37Our study suggests a close and nuanced link between the benign prostatic hyperplasia burden and national sociodemographic status, and the potential for intervention to make an impact. As noted, the absolute benign prostatic hyperplasia burden has increased globally between 2000 and 2019, but age-standardised prevalence and DALY rates were more stable. This pattern of rising absolute burden with stability or only small changes in age-standardised rates was seen in most regions and many countries, reflecting the major role of widespread population growth and ageing in the substantial increase in benign prostatic hyperplasia cases. A rising absolute burden of benign prostatic hyperplasia was seen across all SDI quintiles, and the middle SDI quintile in particular carried the greatest absolute DALY burden by 2019. Notably, however, countries in the lowest three SDI quintiles (low, low-middle, and middle) had the largest percentage change in absolute DALYs between 2000 and 2019, and also had age-standardised DALY rates that overall trended upwards over the study period. Countries in the highest two SDI quintiles (high-middle and high) had somewhat smaller relative increases in absolute DALY counts and had age-standardised rates that overall trended downwards. Although population growth and ageing are the two most important factors contributing to the rising burden of benign prostatic hyperplasia worldwide, divergent trends in age-standardised rates suggest some influence from other risk factors for benign prostatic hyperplasia, such as metabolic syndrome, obesity, diabetes, and acute and chronic prostatic inflammation.,,,,,,, Rising age-standardised DALY rates in the bottom three SDI quintiles could reflect increased detection and diagnosis, or a true increase in disease frequency driven by rising levels of upstream risk factors. Downward trends in age-standardised DALY rates could reflect increased treatment initiation, advancement in surgical care and access, or improved control of upstream risk factors. Although the age-standardised prevalence and DALY rates declined in many high-income countries during the study period, we saw increasing age-standardised rates in the USA. This finding was consistent with the rising prevalence of major comorbidities associated with benign prostatic hyperplasia, such as diabetes, hypertension, cardiac disease, and hyperlipidaemia in the USA compared with other high-income countries., This observation indicates that even high-income countries with similar advancement in economic development and similar age structures could have varying levels of benign prostatic hyperplasia depending on the prevalence of the underlying causes of benign prostatic hyperplasia in the population. This finding emphasises the broader relevance of benign prostatic hyperplasia to other non-communicable diseases and their control measures and serves as an urgent call for countries to strengthen efforts to address these public health challenges together.

38With medical, social, and economic advances, people are living longer worldwide. Many countries are undergoing rapid changes in their population size, as well as the proportion of older people in their population. Consequently, addressing the burden of ageing-related diseases, such as benign prostatic hyperplasia, has to become one of the top global health priorities. In addition to imposing a substantial health burden, as shown in the present analysis, benign prostatic hyperplasia imposes substantial economic costs on societies. An analysis of the National Health and Nutrition Examination Survey-III done in the USA revealed that there were close to 8 million clinic visits for a primary or secondary diagnosis of benign prostatic hyperplasia in 2000, resulting in a direct cost of US$1·1 billion for treatment, excluding outpatient medication costs. The estimated economic burden of benign prostatic hyperplasia in the global population of men older than 65 years was $73·8 billion per year. Thus, our study findings have important implications for health service structure, human resource capacity building, and economic burden prediction. Although well documented, evidence-based prevention of benign prostatic hyperplasia is limited, disability and complications related to benign prostatic hyperplasia can be mitigated. In particular, there are several medical and surgical therapy options to reduce disability due to benign prostatic hyperplasia. Medical therapy involves the use of alpha blockers, 5-alpha reductase inhibitors, phosphodiesterase-5 inhibitors, anticholinergic agents, beta-3-agonists, or therapy with a combination of the above. Surgical therapy can be used for selected patients, such as those with renal insufficiency associated with benign prostatic hyperplasia, refractory urinary retention, recurrent urinary tract infection, recurrent bladder stone, gross haematuria, or failed medical therapy., With the rising number of benign prostatic hyperplasia cases, the demand for diagnostic tools, medications, and hospital care will increase enormously. Therefore, the health structure and human resource capacity-building of a nation should be organised to meet these increasing demands. Despite its burden and increasing trends, efforts from the global community to design prevention strategies for benign prostatic hyperplasia are inadequate. Global, regional, and national efforts should begin and be integrated into broader non-communicable disease control efforts to prevent health loss due to benign prostatic hyperplasia.

39This study has several limitations, which can be broadly organised into limitations of prevalence inputs, limitations of severity inputs, and analytical considerations. First, data to estimate the prevalence of benign prostatic hyperplasia are sparse and heterogeneous, and theycarry with them the inherent biases of administrative records from medical facilities and claims. With regard to scarcity, despite the international administrative data we used in our analyses, we did not have prevalence data for many countries, especially in sub-Saharan Africa, Australasia, south Asia, Andean Latin America, and eastern Europe. We partially overcame prevalence data scarcity by using regional estimates and predictive covariates to produce estimates of the prevalence of benign prostatic hyperplasia in locations without local data. Our predictive covariates are, themselves, estimated for all year, age, and location combinations, generally with much stronger input databases than are available for benign prostatic hyperplasia, but with some uncertainty in estimation. Although the uncertainty intervals we report with our final prevalence estimates include uncertainty due to sampling and bias adjustment of input data, and uncertainty in the model fitting itself, the uncertainties in the covariate estimation processes are not reflected; future rounds of GBD should better account for covariate uncertainty. With regard to heterogeneity, the data we do have might differ on the basis of health-care-seeking behaviours and access to quality health care, rather than differences in underlying disease. This is partially addressed by processing hospital data as admission cause fractions, applying the fractions to estimates of health-care utilisation modelled from large household surveys, and then applying estimates of outpatient cases to inpatient admissions modelled from individual-level data using the Healthcare Access and Quality (HAQ) index as a predictor. Heterogeneity related to access was further accounted for in the USA by adjusting commercial claims data towards a general population reference standard through MR-BRT methods. Nonetheless, our ratios of outpatient cases to inpatient admissions were modelled from US data and might reflect a relationship between inpatient and outpatient care that is unique to the USA, and we did not have sufficient data to identify, quantify, and develop MR-BRT adjustments to account for all instances of heterogeneity due to access worldwide. We attempted to account for the most egregious heterogeneity by excluding outliers more than 2 MAD above or below the median, but this approach does not distinguish heterogeneity in the sources from heterogeneity in underlying disease. Regarding the general level of benign prostatic hyperplasia ascertainment in administrative data, these data sources might miss undiagnosed cases of benign prostatic hyperplasia that have not been seen by a medical provider. According to the Multinational Survey of Aging Male (MSAM-7) study conducted in the UK, the USA, France, Germany, the Netherlands, Italy, and Spain, only 19% of men with lower urinary tract symptoms sought care for urinary problems and only 10·2% had been medically treated. Another community-based study done in Singapore found that more than 70% of study participants with moderate-to-severe lower urinary tract symptoms did not seek care from a health-care provider. These studies suggest that we could be underestimating the prevalence of benign prostatic hyperplasia by relying on administrative data from clinical care encounters. In future rounds of GBD, we should augment our prevalence input data set via a systematic review of population-based studies, both to close gaps in countries without data and to facilitate nuanced quantification of the association between provider-diagnosed and overall benign prostatic hyperplasia prevalence and more accurately correct administrative data sources from diverse settings.

40Second, data used for estimating the symptomatic proportion of benign prostatic hyperplasia prevalence and for estimating disability weights were more limited than prevalence data. Disability weights associated with health state descriptions used in GBD are derived from a series of face-to-face, telephone, and internet surveys conducted over several years and in nine countries, and reported in a series of publications.,, If these nine countries are poorly representative of the values surrounding health in other countries, this would misrepresent the disability globally. Estimation of the proportion of doctor-diagnosed symptomatic versus asymptomatic benign prostatic hyperplasia cases rests on an even smaller database; we made use of four community-based surveys of I-PSS scores to calculate the pooled proportion of symptomatic cases of benign prostatic hyperplasia. This approach makes two important assumptions: that the distribution of I-PSS scores in community-based samples is similar to the distribution of I-PSS scores among cases ascertained from administrative data, and that the distribution of I-PSS scores from these four surveys done in Japan, the USA, Scotland, and France is reflective of the global distribution.

41Third, we acknowledge a pair of analytical limitations. Because of the GBD principle of assigning every death in our estimation framework to a single underlying cause of death, we elected to assign deaths related to benign prostatic hyperplasia to other diseases in the cascade of events that lead to death, and mortality related to benign prostatic hyperplasia was thus accounted for in various complications (eg, urinary tract infection or urolithiasis) and not included in the estimates for benign prostatic hyperplasia. Additionally, GBD estimation to date has largely focused on producing estimates for general populations defined only by year, age, sex, and location. We acknowledge the important matter of the disparate burden by race and ethnicity within these geographically defined populations. Future rounds of GBD should attempt to estimate the proportion of other deaths due to urological diseases that can be reasonably attributed to benign prostatic hyperplasia as an upstream risk factor and should disaggregate estimates of burden by race and ethnicity within populations.

42The burden of benign prostatic hyperplasia is rising throughout the world, primarily due to population growth and ageing. Consequently, the male burden on the existing health-care system is expected to grow substantially in the coming years. This growth could be modified by control of upstream risk factors, and technologies exist to treat and mitigate the symptoms of benign prostatic hyperplasia. Coordinated and collaborative efforts from global, regional, and national policy makers, researchers, and advocates are needed to tackle this challenge.

43To download the source data and analytic code used in these analyses, please visit the Global Health Data Exchange GBD 2019 website.

44B Bikbov reports grants or contracts from the Lombardy Region outside the submitted work to their institution. T Garg reports support for attending meetings or travel from Siemens Healthineers outside the submitted work. N E Ismail reports unpaid leadership or fiduciary roles in board, society, committee, or advocacy groups with the Malaysian Academy of Pharmacy as a council member outside the submitted work. N Perico reports support for attending meetings or travel from and participation on a data safety monitoring board or advisory board with Bayer AG outside the submitted work. J A Singh reports consulting fees from Crealta Horizon, Medisys, Fidia, PK Med, Two Labs, Adept Field Solutions, Clinical Care Options, Clearview Healthcare Partners, Putnam Associates, Focus Forward, Navigant Consulting, Spherix, MedIQ, Jupiter Life Science, UBM, Trio Health, Medscape, WebMD, Practice Point Communications, the National Institutes of Health, and the American College of Rheumatology; payment or honoraria for lectures, presentations, speakers' bureaus, manuscript writing or educational events from Simply Speaking; support for attending meetings or travel from the steering committee of OMERACT; participation on a data safety monitoring board or advisory board with the US Food and Drug Administration Arthritis Advisory Committee; a leadership or fiduciary role in a board, society, committee or advocacy group, paid or unpaid, with OMERACT as a steering committee member, with the Veterans Affairs Rheumatology Field Advisory Committee as Chair (unpaid), and with the UAB Cochrane Musculoskeletal Group Satellite Center on Network Meta-analysis and editor and director (unpaid); stock or stock options in TPT Global Tech, Vaxart Pharmaceuticals, Atyu Biopharma, Adaptimmune Therapeutics, GeoVax Labs, Pieris Pharmaceuticals, Enzolytics, Seres Therapeutics, Tonix Pharmaceuticals and Charlotte's Web Holdings; and having previously owned stock options in Amarin, Viking, and Moderna Pharmaceuticals, all outside the submitted work. All other authors declare no competing interests.