{"paper_id":"2b9b8957-699e-4364-83c4-4aec7a6906f1","body_text":"RESEARCH Open Access\n© The Author(s) 2025. Open Access  This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 \nInternational License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you \ngive appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the \nlicensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or \nother third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the \nmaterial. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or \nexceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit  h t t p  : / /  c r e a  t i  \nv e c  o m m  o n s .  o r  g / l  i c e  n s e s  / b  y - n c - n d / 4 . 0 /.\nLi et al. Reproductive Biology and Endocrinology           (2025) 23:88 \nhttps://doi.org/10.1186/s12958-025-01421-z\nReproductive Biology \nand Endocrinology\n*Correspondence:\nLing Zhang\nzhanglingxh@hust.edu.cn\nYi Liu\nliqun1994@hust.edu.cn\n1Department of Obstetrics and Gynecology, Union Hospital, Tongji \nMedical College, Huazhong University of Science and Technology,  \nWuhan 430022, China\nAbstract\nBackground endometriosis as a common gynecologic finding significantly affects the quality of life of many women. \nAn accurate understanding of the epidemiological characteristics of endometriosis is essential for disease control \nand prevention. We aimed to use the latest data from the Global Burden of Disease (GBD) 2021 to comprehensively \nanalyze the various epidemiological indicators of surgically confirmed endometriosis and their changing trends to \nbetter measure the disease burden and help improve health management.\nMethods We delineated incidence, prevalence, and years lived with disability (YLDs) of surgically confirmed \nendometriosis at the global, regional, and national levels. The estimated annual percentage change (EAPC) was \ncalculated to assess temporal trends in the age-standardized rate (ASR). In addition, we used joinpoint regression \nmodels to describe local trends in these indicators, assessed the correlation between disease burden and Socio-\ndemographic index (SDI) levels, and used decomposition analysis to quantitatively analyze the driving factors leading \nto changes in disease burden.\nResults Globally, the age-standardized rate of incidence, prevalence, and YLDs of surgically confirmed endometriosis \nall showed a decreasing trend from 1990 to 2021. The burden of surgically confirmed endometriosis is mainly \nconcentrated in women aged 20–30 years and declines with increasing SDI levels. The results of the decomposition \nanalysis indicated that population growth is the main driving factor for the upward in the number of incidence, \nprevalence, and YLDs cases of endometriosis worldwide.\nConclusions The overall burden of endometriosis has decreased globally from 1990 to 2021, but there are regional \ndisparities. Managing this condition remains a major challenge, and more refined policies and interventions are \nneeded to effectively address the burden of endometriosis.\nKeywords Endometriosis, Disease burden, Estimated annual percentage change, Incidence, Prevalence\nGlobal and regional trends in the burden \nof surgically confirmed endometriosis \nfrom 1990 to 2021\nRuijie Li1, Ling Zhang1* and Yi Liu1*\n\nPage 2 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nIntroduction\nEndometriosis is a prevalent gynecological finding \ndefined by the presence of endometrium-like epithe -\nlium and/or stroma (lesions) outside the endometrium \nand myometrium, usually with an associated inflamma -\ntory process [ 1]. According to surveys, over 176 million \nwomen worldwide suffer from this condition, constitut -\ning 5–10% of women of reproductive age [ 2]. Although \nendometriosis is considered benign, it exhibits malignant \nbehaviors such as invasion, implantation, and recurrence \n[3]. This chronic and refractory trait consumes a large \namount of social resources and causes a serious eco -\nnomic burden [4]. Endometriosis often leads to infertility \nand chronic pelvic discomfort [ 5], and patients are prone \nto experience depression and anxiety [ 6], which severely \naffect the physical health, mental health, and quality of \nlife of many women [7].\nAlthough the impact of endometriosis on women’s \nhealth urgently needs attention, global research data \non the burden of this disease are currently very limited. \nGiven the diagnostic barrier and delay, the exact preva -\nlence of endometriosis is unknown. This is due to the \nfact that although laparoscopy is the gold standard for \ndiagnosis, it is an invasive examination that is not rou -\ntinely used in women with a suspicion of endometriosis. \nSometimes doctors tend to give a “working diagnosis” \nof probable endometriosis and prompt early drug treat -\nment without waiting for a more definitive diagnosis, \nwhich makes accurate assessment difficult [ 8– 10]. Simi-\nlarly, there are no established criteria for the exact time \nof onset of endometriosis, as symptoms must be present \nand sufficiently disruptive to obtain a referral for a defini-\ntive diagnosis. Different nonspecific symptoms, clinician \nawareness of endometriosis, and economic and geo -\ngraphic access to care all contribute to an average delay \nof 7 years from symptom onset to surgical diagnosis [11].\nTherefore, it should be recognised that when attempt -\ning to conduct epidemiological studies on endometrio -\nsis, we need to take into account that it varies across \npopulations and across time and space. This requires a \nlarge amount of well-documented longitudinal data. To \naddress this issue, the study utilized the latest data from \nthe Global Burden of Disease (GBD) to conduct a com -\nprehensive and updated analysis of various epidemiologi -\ncal indicators of surgically confirmed endometriosis. In \nthis study, we described the long-term and partial spatio -\ntemporal trends of the incidence, prevalence, and YLDs \nof endometriosis at global and regional levels, and ana -\nlyzed the contributions of different factors to changes \nin the epidemiological indicators of endometriosis from \nmultiple perspectives. The aim is to better measure the \ncurrent burden of endometriosis to enhance women’s \nhealth awareness, help improve health management, \nand develop timely and effective prevention and control \nstrategies.\nMethods\nOverview\nThe GBD 2021 database (  h t t p  s : /  / g h d  x .  h e a  l t h  d a t a  . o  r g / g b \nd - 2 0 2 1), led by the Institute for Health Metrics and  E v a l u \na t i o n (IHME), provides the most comprehensive and up-\nto-date data assessment of the descriptive epidemiology \nof diseases in 21 regions and 204 countries and territo -\nries from 1990 to 2021, using all available data. All data \nis calculated by direct query and downloaded from the \nGBD results tool. A detailed description of the method \ncan be found on the help page of the database and other \npublications [12]. The GBD collects health data from life \nrecords, censuses, registers, health surveys, population \nsurveillance, administrative reports, scientific research, \ndischarge records, records of outpatient visits, and health \ninsurance claims, as well as many other sources. These \nare then input into an algorithm to generate an estimate \nof the burden of disease. In the GBD study, disease esti -\nmates were generated by age, year, and location using the \nBayesian meta-regression tool DisMod-MR 2.1 to ensure \nconsistent epidemiological parameters for the conditions \nunder study.\nData source\nData on the global burden of surgically confirmed endo -\nmetriosis were obtained from published sources using \nthe Global Health Data Exchange Query Tool. GBD 2021 \ndefines endometriosis cases according to the ACOG \nguidelines as cases diagnosed by pelvic exam confirmed \nby laparoscopy or laparotomy. This study obtained global, \nregional, territorial, and socio-demographic index (SDI) \nquintile data on incidence, prevalence, and years lived \nwith disability (YLDs) of endometriosis from 1990 to \n2021 from the GBD 2021. Incidence is the frequency \nof new cases of a disease in a population over a certain \nperiod of time. Prevalence is the ratio of new and old \ncases of a disease in the entire population over a certain \nperiod of time. YLDs are years of life lost due to disability \ncaused by the disease, estimated as the product of preva -\nlence estimate and disability weight for health states of \neach mutually exclusive sequela adjusted for comorbidity. \nThe age range is limited to between 15 and 54 years old, \ndivided into eight 5-year-old age groups. GBD divides \nthe SDI of 21 regions and 204 countries and territories \ninto five components (high, high-middle, middle, low-\nmiddle, and low) based on the lag-distributed income per \ncapita, average years of schooling, and the fertility rate in \nfemales younger than 25 years for a given location. SDI \nranges from 0 to 1, with higher values indicating higher \nincome and years of schooling, and lower fertility. In \n\nPage 3 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \naddition, GBD regions are not actual geopolitical units, \nbut groupings of countries created for analysis.\nStatistical analyzes\nCalculation of the estimated annual percentage change\nAge-standardized rates of incidence, prevalence, and \nYLDs from 1990 to 2021 were used to assess the burden \nof endometriosis. Temporal trends of burden over thirty \nyears are reflected by the estimated annual percentage \nchange (EAPC). The EAPC is a widely used measure of \nthe age-standardized rate trend over a specified time \ninterval [ 13]. We fitted a regression line to the natural \nlogarithm of the age-standardized rate to calculate the \nEAPC:\n y = α + βx + ε\nwhere y = ln (age-standardized rate), x = calendar year, β \nis the regression coefficient, and ε is the error term in the \nregression model (also known as residual). This was then \nexpressed as a percentage: EAPC = 100*(exp(β)-1). The \n95% confidence interval (CI) of the EAPC was calculated \nto reflect the temporal trend in the age-standardized rate \n(ASR). An upward trend in the age-standardized rate was \nindicated when the EAPC and the lower boundary of the \n95% CI were positive, whereas a downward trend was \nindicated when the EAPC and the upper boundary of the \n95% CI were negative.\nJoinpoint regression analysis\nIn order to detect changes in parameter trends of endo -\nmetriosis health metrics, the joinpoint regression model \nwas utilized. Joinpoint regression model partitions a \nlong-term trend line into several segments through \nmodel-fitting, using permutation tests to identify points \n(joinpoints) where linear trends change significantly in \ndirection or magnitude (e.g., zero joinpoints indicate \na straight line). This model’s calculating approach is to \nestimate the changing rule of illness rates using the least \nsquare method, avoiding the non-objectivity of typical \ntrend analyses based on linear trends [ 14]. Therefore, \nwe analyzed the age-standardized rate of endometriosis \nincidence, prevalence and, YLDs by different SDI regions, \ncalculated the number of junction points and the position \nof each junction point by Monte Carlo permutation test, \nand the corresponding test statistic P value (α = 0.05). For \nconvenience of understanding, slopes are often converted \nto annual percentage changes (APCs) and average annual \npercent change (AAPC); that is, the estimated annual \npercentage change from one connection point to the next \n[15].\nSocio-demographic index\nThe association between the burden of surgically con -\nfirmed endometriosis and SDI, for global and the 21 GBD \nregions from 1990 to 2021, were assessed using smooth -\ning splines models. The SDI ranges from 0 (less devel -\noped) to 1 (most developed) and is comprised of the: \n(1) lag-distributed income per capita, which is the gross \ndomestic product per capita smoothed over the preced -\ning decade; (2) average years of schooling for the popula -\ntion older than 15 years of age; and (3) the fertility rate in \nfemales younger than 25 years for a given location. The \nstatistical analyses were conducted using R software, ver -\nsion 4.3.2.\nDecomposition analysis\nDecomposition analysis refers to the breakdown of a \ncomposite indicator (e.g., incidence, prevalence, etc.) into \nmultiple components in order to gain a clearer under -\nstanding of the contribution of each factor to the overall \noutcome. Specifically, the disease burden can be decom -\nposed into different influencing factors, such as age, \npopulation, and epidemiological changes, to quantify the \nimpact of each factor on the total change. We employed \nthe decomposition analysis proposed by Das Gupta com -\nbined with an improved method proposed by Cheng \nand colleagues in 2020 to disentangle alterations in the \nburden of endometriosis into three group-level determi -\nnants: population aging, population growth, and epide -\nmiological change [16, 17]. The approaches can be briefly \nsummarized as follows.\nThe number of disease burden indicators(X) at each \nlocation was obtained from the following formula:\n \nX ay,py,ey=\n∑17\n(i=1)\n(ai,y ∗ p y ∗ ei,y )\nWhere X ay, py, ey  represented disease burden indicators \nbased on the factors of age structure, population, and \nASR for specific year y; a i y represents the proportion of \npopulation for the age category i in given year y; p y repre-\nsents the total population in given year y; and e i, y repre-\nsents ASR given age category i in year y. The contribution \nof each factor to the change of disease burden from 1990 \nto 2021 was defined by the effect of one factor changing \nwhile the other factors were held constant. And the sum \nof the effects of each driving factor should exactly equal \nthe total change in the disease burden indicator.\nResults\nThe overall burden of surgically confirmed endometriosis \nfrom 1990 to 2021\nFrom 1990 to 2021, there was a downward trend in the \nglobal age-standardized incidence, prevalence and YLDs \nrates of surgically confirmed endometriosis. The number \n\nPage 4 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nof incidence, prevalence and YLDs cases rose during this \ntimeframe. In 2021, the incident cases of surgically con -\nfirmed endometriosis worldwide were 3.45  million, the \nprevalent cases arrived at 22.28  million, and the YLDs \ncases increased to 2.05 million (Table 1, S1, S2) (Fig. 1).\nAs far as SDI regions are concerned, the most signifi -\ncant annual decrease in age-standardized incidence, \nprevalence and YLDs rates from 1990 to 2021 occurred \nin low-middle SDI regions(EAPC of ASIR: -1.5, 95%UI: \n-1.52 to -1.47; EAPC of ASPR: -1.67, 95%UI: -1.7 to \n-1.64; EAPC of YLDs: -1.65, 95%UI: -1.68 to -1.62), \nwhile the smallest decrease was seen in high-middle \nSDI regions(EAPC of ASIR: -0.72, 95%UI: -0.82 to -0.62; \nEAPC of ASPR: -0.71, 95%UI: -0.81 to -0.62; EAPC of \nYLDs: -0.71, 95%UI: -0.8 to -0.61)(Table 1, S1, S2) (Fig. 1).\nThe burden due to surgically confirmed endometrio -\nsis varied significantly across different regions. In 2021, \nthe age-standardized incidence, prevalence and YLDs \nrates were highest in Oceania, Eastern Europe, Western \nSub-Saharan Africa, and North Africa and Middle East. \nHigh-income North America, Central Latin America, \nEast Asia, and Southern Latin America observed the low-\nest rates (Fig. 2, S1, S2). The annual percentage change \nin age-standardized rates varied from 1990 to 2021, with \nthe highest decrease observed in High-income North \nAmerica(EAPC of ASIR: -2.07, 95%UI: -2.24 to -1.9; \nEAPC of ASPR: -1.91, 95%UI: -2.08 to -1.75; EAPC of \nYLDs: -1.92, 95%UI: -2.09 to -1.76) and the most signifi -\ncant increase found in Eastern Europe (EAPC of ASIR: \n0.32, 95%UI: 0.15 to 0.5; EAPC of ASPR: 0.32, 95%UI: \n0.15 to 0.5; EAPC of YLDs: 0.33, 95%UI: 0.15 to 0.51) \n(Table 1, S1, S2) (Fig. 1).\nTable 1 The incidence of endometriosis in 1990 and 2021 and Temporal trends between 1990 and 2021\nLocation 1990 2021 1990–2021 (%)\nIncidence_Num-\nber_1000 (95%UI)\nASIR per 100,000 \n(95%UI)\nIncidence_Num-\nber_1000 (95%UI)\nASIR per 100,000 \n(95%UI)\nEAPC of inci-\ndence rate\nGlobal 3330.2 (2308.6 to 4507) 119.6 (83.5 to 160.5) 3447.1 (2436.3 to \n4611.5)\n88.5 (62.5 to 119.5) -1(-1.05 to -0.96)\nSDI region\n High SDI 437.9 (305.8 to 594.9) 96.3 (67.2 to 130.6) 358 (260.6 to 467.1) 75.4 (54.1 to 99.2) -0.92(-0.99 to -0.85)\n High-middle SDI 593.6 (412.1 to 799.1) 103.9 (72.5 to 141.1) 488.9 (353.4 to 647.9) 83.1 (59.4 to 110.5) -0.72(-0.82 to -0.62)\n Middle SDI 1049.8 (715 to 1440.1) 112.3 (77.5 to 151.1) 1008.1 (709 to 1348.9) 82.4 (58.1 to 111.2) -1.04(-1.12 to -0.97)\n Low-middle SDI 863 (597 to 1197.1) 149.5 (105.2 to 200.1) 981.6 (680.6 to 1336.9) 94.1 (65.8 to 127.3) -1.5(-1.52 to -1.47)\n Low SDI 383.2 (264.5 to 531.7) 160.3 (112.7 to 215.3) 607.9 (420.3 to 848.8) 103.6 (72 to 141.1) -1.41(-1.46 to -1.36)\nGBD region\n Andean Latin America 20.9 (14.2 to 29.3) 104.3 (72.5 to 140.7) 25 (17.3 to 34.5) 70.9 (49.1 to 97.5) -1.19(-1.26 to -1.12)\n Australasia 10.7 (7.5 to 14.7) 100.7 (70.1 to 137.7) 12 (8.3 to 16) 88.4 (61.3 to 119.6) -0.25(-0.33 to -0.18)\n Caribbean 18.4 (12.7 to 25.9) 94.3 (65.4 to 128.5) 17.1 (11.8 to 23.4) 70.8 (48.8 to 97.5) -0.88(-0.91 to -0.86)\n Central Asia 39.4 (27 to 55) 111.1 (77.3 to 150.1) 42.3 (29.5 to 56.3) 89 (61.8 to 119) -0.39(-0.57 to -0.21)\n Central Europe 50.9 (35.8 to 68.9) 84.5 (59.4 to 113.9) 36.1 (25.6 to 48.8) 76.3 (53.7 to 103) -0.23(-0.37 to -0.09)\n Central Latin America 88.2 (60.2 to 124) 99 (68.5 to 133.5) 89.8 (62.5 to 121.7) 65.3 (45.3 to 88.8) -1.3(-1.39 to -1.21)\n Central Sub-Saharan Africa 41.9 (28.6 to 58.5) 157.4 (109.2 to 212.8) 68.8 (46.6 to 96.8) 99 (67.6 to 135.7) -1.42(-1.54 to -1.31)\n East Asia 704.5 (477.8 to 971.7) 102.9 (70.7 to 142.1) 444.7 (320.3 to 592.9) 67 (47.9 to 89.3) -1.51(-1.68 to -1.34)\n Eastern Europe 148.3 (102.9 to 200.5) 136.4 (96.5 to 184) 117.9 (83.5 to 157) 133.6 (93.2 to 180.3) 0.32(0.15 to 0.5)\n Eastern Sub-Saharan Africa 130.3 (89.7 to 181.6) 139.6 (96.6 to 189.4) 195.6 (135.2 to 273.4) 85.2 (58.6 to 117) -1.59(-1.64 to -1.54)\n High-income Asia Pacific 110.5 (75.5 to 148.2) 120.2 (82 to 161.4) 77.2 (54.7 to 103.2) 104.4 (73.4 to 137.1) -0.53(-0.63 to -0.42)\n High-income North America 133.5 (88.4 to 188.2) 86.8 (58.3 to 122.9) 91.2 (65.3 to 118.7) 54.1 (38.6 to 71.1) -2.07(-2.24 to -1.9)\n North Africa and Middle East 260 (178.5 to 368.6) 154.1 (107.8 to 210.2) 341.4 (238.6 to 462.3) 106.7 (74.7 to 144.4) -1.19(-1.29 to -1.09)\n Oceania 5.9 (4.1 to 8.1) 181.5 (127.8 to 243.9) 10.5 (7.4 to 14.6) 148.3 (104.5 to \n204.3)\n-0.63(-0.65 to -0.61)\n South Asia 801.1 (554.3 to 1098.6) 149.1 (104.3 to 200.6) 936.8 (643.3 to 1268.5) 92.2 (63.8 to 124.6) -1.57(-1.6 to -1.55)\n Southeast Asia 341.5 (236.6 to 465.5) 136.1 (95.6 to 182.7) 386.1 (271.9 to 518.4) 105 (73.7 to 141.2) -0.79(-0.82 to -0.76)\n Southern Latin America 21.1 (14.6 to 28.3) 84.1 (58.1 to 113) 24.6 (17.4 to 32.1) 69.9 (49.4 to 91.5) -0.47(-0.56 to -0.38)\n Southern Sub-Saharan Africa 33.5 (22.9 to 46.1) 117.2 (80.8 to 157.7) 40.5 (27.6 to 55.2) 92.1 (63.3 to 125.3) -0.75(-0.77 to -0.72)\n Tropical Latin America 82.4 (55.4 to 114.8) 99.5 (67.9 to 137.4) 90.7 (63.4 to 120.4) 76 (52.6 to 102) -1.2(-1.36 to -1.04)\n Western Europe 140.9 (96.9 to 191.9) 74.7 (51.7 to 102.2) 126.3 (89.6 to 167.3) 72.4 (50.5 to 98) 0.02(-0.02 to 0.07)\n Western Sub-Saharan Africa 146.2 (100.5 to 203.8) 154.9 (107.6 to 209.1) 272.5 (186.8 to 380.7) 105.4 (72.7 to 142.5) -1.22(-1.28 to -1.16)\nEAPC, estimated annual percentage change; ASIR, age-standardized incidence rate; SDI, socio-demographic index\n\nPage 5 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nFig. 2 The ASIR of endometriosis in 204 countries and territories in 2021. ASIR, age-standardized incidence rate\n \nFig. 1 The EAPC for ASIR, ASPR, and Age Standardized YLDs Rate at the regional level. EAPC, estimated annual percentage change; ASIR, age-standardized \nincidence rate; ASPR, age-standardized prevalence rate; YLDs, years lived with disability\n \n\nPage 6 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nJoinpoint regression analysis of surgically confirmed \nendometriosis burden\nAge-standardized incidence, prevalence and YLDs rates \nshowed a year-by-year decline in global and most SDI \nregions (Table  2). However, the change trend in high-\nmiddle SDI and middle SDI was relatively more complex. \nThe tendency of ASIR and ASPR in high-middle SDI and \nmiddle SDI decreased from 1990 to 2010 and 2015 to \n2021, but increased from 2010 to 2015 (Fig. 3, S3). The \nage-standardized YLDs rate in high-middle SDI showed \nan upward trend from 2010 to 2015 and a downward \ntrend for the rest of the years (Fig. S4).\nAge-based description of the burden of surgically \nconfirmed endometriosis\nWe analyzed the burden of surgically confirmed endo -\nmetriosis according to age groups in 2021(Fig. 4). The \nglobal age-specific incidence rate and the number of \nincident cases peaked in the 20–24 age group (Fig. 4 A). \nThe global prevalence rate and number of prevalent cases \nTable 2 The trends in endometriosis burden by joinpoint regression\nSDI factor Index Incidence Prevalence YLDs\nPeriod Estimate(%)(95%UI) Period Estimate(%)(95%UI) Period Estimate(%)(95%UI)\nGlobal APC 1990–1992 -1.27(-1.51 to -1.03) 1990–1994 -1.3(-1.4 to -1.19) 1990–1994 -1.27(-1.37 to -1.17)\n1992–2003 -0.99(-1 to -0.97) 1994–2006 -1(-1.02 to -0.97) 1994–2006 -0.98(-1 to -0.96)\n2003–2006 -1.12(-1.36 to -0.89) 2006–2009 -2.14(-2.45 to -1.83) 2006–2009 -2.13(-2.43 to -1.84)\n2006–2009 -2.09(-2.32 to -1.85) 2009–2012 -0.67(-0.98 to -0.35) 2012–2019 -0.64(-0.93 to -0.34)\n2009–2018 -0.48(-0.5 to -0.45) 2012–2018 -0.43(-0.5 to -0.36) 2019–2021 -0.45(-0.5 to -0.4)\n2018–2021 -0.87(-0.99 to -0.76) 2018–2021 -0.77(-0.93 to -0.61)\nAAPC 1990–2021 -0.97 (-1.01 to -0.93) 1990–2021 -0.98 (-1.03 to -0.94) 1990–2021 -0.98 (-1.03 to -0.94)\nHigh SDI APC 1990–1996 -0.24(-0.33 to -0.15) 1990–1996 -0.29(-0.36 to -0.22) 1990–1996 -0.28(-0.32 to -0.24)\n1996–2004 -1.63(-1.69 to -1.56) 1996–2001 -1.36(-1.47 to -1.25) 1996–2001 -1.37(-1.44 to -1.29)\n2004–2008 -0.99(-1.21 to -0.76) 2001–2004 -2.06(-2.38 to -1.73) 2001–2004 -2.05(-2.27 to -1.83)\n2008–2021 -0.48(-0.51 to -0.46) 2004–2009 -0.95(-1.04 to -0.85) 2004–2009 -0.93(-1 to -0.87)\n2009–2014 -0.12(-0.21 to -0.03) 2009–2015 -0.12(-0.17 to -0.08)\n2014–2021 -0.26(-0.3 to -0.21) 2015–2021 -0.32(-0.36 to -0.29)\nAAPC 1990–2021 -0.8 (-0.83 to -0.76) 1990–2021 -0.71 (-0.75 to -0.66) 1990–2021 -0.71 (-0.74 to -0.68)\nHigh-middle SDI APC 1990–1995 -1.61(-1.74 to -1.49) 1990–1994 -1.93(-2.03 to -1.83) 1990–1994 -1.89(-2.02 to -1.77)\n1995–2005 -0.62(-0.67 to -0.57) 1994–1999 -0.96(-1.06 to -0.86) 1994–1999 -0.95(-1.07 to -0.82)\n2005–2010 -2.04(-2.21 to -1.87) 1999–2005 -0.41(-0.48 to -0.34) 1999–2005 -0.41(-0.49 to -0.32)\n2010–2015 0.81(0.64 to 0.98) 2005–2010 -1.95(-2.05 to -1.86) 2005–2010 -1.95(-2.06 to -1.83)\n2015–2021 -0.22(-0.31 to -0.13) 2010–2015 0.66(0.57 to 0.76) 2010–2015 0.69(0.57 to 0.8)\n2015–2021 -0.23(-0.28 to -0.18) 2015–2021 -0.27(-0.34 to -0.21)\nAAPC 1990–2021 -0.7 (-0.75 to -0.66) 1990–2021 -0.74 (-0.77 to -0.71) 1990–2021 -0.74 (-0.77 to -0.7)\nMiddle SDI APC 1990–1993 -1.6(-1.74 to -1.45) 1990–1995 -1.52(-1.62 to -1.42) 1990–1995 -1.49(-1.6 to -1.39)\n1993–2005 -0.96(-0.98 to -0.94) 1995–2005 -0.88(-0.92 to -0.84) 1995–2005 -0.87(-0.91 to -0.83)\n2005–2010 -2.3(-2.39 to -2.21) 2005–2010 -2.3(-2.44 to -2.17) 2005–2010 -2.29(-2.43 to -2.14)\n2010–2015 0.1(0.01 to 0.19) 2010–2015 0.01(-0.12 to 0.15) 2010–2016 -0.03(-0.13 to 0.08)\n2015–2021 -0.52(-0.57 to -0.47) 2015–2021 -0.42(-0.49 to -0.34) 2016–2021 -0.52(-0.62 to -0.43)\nAAPC 1990–2021 -0.98 (-1.01 to -0.96) 1990–2021 -0.98 (-1.02 to -0.94) 1990–2021 -0.98 (-1.02 to -0.94)\nLow-middle SDI APC 1990–1998 -1.63(-1.66 to -1.61) 1990–1999 -1.75(-1.76 to -1.73) 1990–1999 -1.71(-1.74 to -1.69)\n1998–2006 -1.45(-1.48 to -1.42) 1999–2006 -1.61(-1.65 to -1.58) 1999–2006 -1.58(-1.62 to -1.55)\n2006–2009 -1.84(-2.07 to -1.61) 2006–2009 -2.2(-2.4 to -1.99) 2006–2009 -2.17(-2.39 to -1.96)\n2009–2014 -1.46(-1.53 to -1.38) 2009–2014 -1.68(-1.75 to -1.62) 2009–2014 -1.67(-1.74 to -1.6)\n2014–2019 -1.12(-1.19 to -1.04) 2014–2019 -1.1(-1.17 to -1.04) 2014–2019 -1.08(-1.15 to -1.01)\n2019–2021 -1.4(-1.64 to -1.17) 2019–2021 -1.49(-1.69 to -1.28) 2019–2021 -1.63(-1.84 to -1.41)\nAAPC 1990–2021 -1.48 (-1.51 to -1.45) 1990–2021 -1.63 (-1.66 to -1.6) 1990–2021 -1.61 (-1.64 to -1.59)\nLow SDI APC 1990–1995 -0.89(-0.93 to -0.85) 1990–1995 -0.7(-0.76 to -0.65) 1990–1996 -0.73(-0.77 to -0.68)\n1995–2006 -1.2(-1.22 to -1.19) 1995–2003 -1.05(-1.09 to -1.02) 1996–2006 -1.06(-1.09 to -1.04)\n2006–2014 -1.67(-1.68 to -1.66) 2003–2006 -1.18(-1.44 to -0.92) 2006–2019 -1.76(-1.78 to -1.74)\n2014–2019 -2(-2.18 to -1.82) 2006–2019 -1.79(-1.8 to -1.77) 2019–2021 -2.22(-2.49 to -1.95)\n2019–2021 -0.89(-0.93 to -0.85) 2019–2021 -2.11(-2.36 to -1.85)\nAAPC 1990–2021 -1.4 (-1.42 to -1.39) 1990–2021 -1.39 (-1.42 to -1.35) 1990–2021 -1.37 (-1.39 to -1.34)\nYLDs, years lived with disability; SDI, socio-demographic index; APC, annual percentage changes; AAPC, average annual percent change\n\nPage 7 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nFig. 3 Joinpoint regression analysis in ASIR of endometriosis from 1990 to 2021 by SDI region. ASIR, age-standardized incidence rate; SDI, socio-demo -\ngraphic index; APC, annual percentage change; * P < 0.05\n \n\nPage 8 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \npeaked in the 25–29 age group (Fig. 4B), while the global \nYLDs rate and the number of YLDs similarly peaked in \nthe 25–29 years age group (Fig. 4 C). In all SDI regions, \nthe highest incidence rate was concentrated in the 20–24 \nage group (Fig. 4 A). In low SDI, low-middle SDI, middle \nSDI, and high-middle SDI regions, high prevalence rate \nand high YLD rate were concentrated in the 25-29-year-\nolds. But in the high SDI region, the highest prevalence \nrate and YLD rate were concentrated in the 40–44 age \ngroup (Fig. 4B, C).\nFig. 4 Age-specific burdens on incidence (A), prevalence (B), and YLDs (C) of endometriosis in 2021. YLDs, years lived with disability. The y-axis of Fig. 4B \nand C: “e” represents “multiply by a power of 10” . For example, in 2e + 05, 2 is the base (valid numeric part), and e + 05 means “multiply by 5 powers of 10” , \ni.e. 2e + 05 = 2 × 100,000 = 200,000\n \n\nPage 9 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nBurden of surgically confirmed endometriosis and SDI \nlevel estimates\nFig. 5 showed the relationship between the SDI levels and \nthe estimated burden of surgically confirmed endome -\ntriosis globally and in 21 GBD regions from 1990 to 2021. \nAge-standardized rates of incidence, prevalence, and \nYLDs of endometriosis all illustrated a broadly negative \ncorrelation with SDI levels. Overall, the average expected \nvalues of the estimated burden rates of endometriosis \ndecreased with increasing SDI levels. The global endo -\nmetriosis burden consistently remained higher than \nexpected between 1990 and 2021 (Fig. 5, S5, S6).\nDecomposition analysis of the changes in the number of \nsurgically confirmed endometriosis incidence, prevalence \nand YLDs between 1990 and 2021\nTable 3 presented the decomposition analysis results con-\ncerning changes in the number of incidence, prevalence, \nand YLDs cases, influenced by three population-levels \ndeterminants: population aging, population growth, \nand epidemiological changes at the global level, five SDI \nstrata and GBD regions (Table  3). Globally, from 1990 to \n2021, population growth resulted in a 1,211.68% increase \nof incident cases, epidemiological changes accounted \nfor an 892.28% reduction, and population aging led to a \n219.4% decrease. The increase in prevalence and YLDs \nwas also mainly attributed to population growth, while \nepidemiological changes were an important reason for \nlimiting the increase. Among the five SDI regions, the \nincidence, prevalence and YLDs cases decreased in high \nand high-middle SDI regions, mainly due to epidemio -\nlogical changes. While the numbers increased in low and \nlow-middle SDI regions, and population growth played \nthe most important role in it. At the regional level, South \nAsia saw the most significant increase in all incidence \n(135,743.42 cases), prevalence (936,710.51 cases), and \nYLDs (87,635.03 cases), followed by Western Sub-Saha -\nran Africa and North Africa and Middle East, primarily \ndriven by population growth. The most notable decrease \noccurred in East Asia, followed by High-income North \nAmerica, mainly due to epidemiological changes (Fig. 6, \nS7, S8).\nDiscussion\nThis study provides a comprehensive analysis of the \nglobal burden of endometriosis from 1990 to 2021 using \ndata from GBD. Our study indicated that during the \nperiod of 1990–2021, the global ASIR, ASPR, and age-\nstandardized YLDs rate for endometriosis showed a \nwidely decreasing trend, but EAPC varied across differ -\nent SDI regions and GBD regions. The burden of surgi -\ncally confirmed endometriosis is mainly concentrated in \nwomen aged 20–30 years, and declines with increasing \nSDI levels. The result of decomposition analysis reveals \nthe global numbers of incidence, prevalence, and YLDs \nof endometriosis significantly increased over the past 30 \nFig. 5 Coevolution of ASIR with SDI globally and for GBD regions of endometriosis, 1990–2021. Colored lines show global and regional values for age-\nstandardized burden estimate rates. Each point in a line represents 1 year starting in 1990 and ending in 2021. The black line represents the average \nexpected relationship between SDI and burden estimate rates for endometriosis based on values from each region. Regions above the solid line have \nhigher than expected burdens, while those below the line have lower than expected burdens\n \n\nPage 10 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nLocation Incidence Prevalence YLDs\nOverall \ndifference\nAging (Per-\ncent %)\nPopulation \n(Percent %)\nEpidemiological \nchange (Percent \n%)\nOverall \ndifference\nAging \n(Percent \n%)\nPopulation \n(Percent \n%)\nEpidemio-\nlogical change \n(Percent %)\nOverall \ndifference\nAging \n(Percent \n%)\nPopulation \n(Percent %)\nEpidemiolog-\nical change \n(Percent %)\nGlobal 116925.82 -256537.99\n(-219.4%)\n1416764.98\n(1211.68%)\n-1043301.16\n(-892.28%)\n2405819.57 236055.64\n(9.81%)\n8777925.95\n(364.86%)\n-6608162.02\n(-274.67%)\n218757.57 17529.95\n(8.01%)\n807097.8\n(368.95%)\n-605870.17\n(-276.96%)\nSDI region\nHigh SDI -79889.08 -26852.49\n(33.61%)\n45157.9\n(-56.53%)\n-98194.49\n(122.91%)\n-289065.55 -11146.77\n(3.86%)\n299551.78\n(-103.63%)\n-577470.55\n(199.77%)\n-27631.81 -1404.11\n(5.08%)\n27570.99\n(-99.78%)\n-53798.69\n(194.7%)\nHigh-middle \nSDI\n-104723.65 -63865.95\n(60.99%)\n81469.2\n(-77.79%)\n-122326.9\n(116.81%)\n-216141.17 78286.99\n(-36.22%)\n529932.53\n(-245.18%)\n-824360.69\n(381.4%)\n-21170.94 6033.07\n(-28.5%)\n48974.05\n(-231.33%)\n-76178.06\n(359.82%)\nMiddle SDI -41768.87 -118430.37\n(283.54%)\n400751.22\n(-959.45%)\n-324089.72\n(775.91%)\n657080.81 213368.95\n(32.47%)\n2407440.02\n(366.38%)\n-1963728.17\n(-298.86%)\n58349.92 17615.19\n(30.19%)\n222240.64\n(380.88%)\n-181505.9\n(-311.06%)\nLow-middle \nSDI\n118524.52 -42078.69\n(-35.5%)\n610259.79\n(514.88%)\n-449656.57\n(-379.38%)\n866487.36 145185.82\n(16.76%)\n3751133.49\n(432.91%)\n-3029831.95\n(-349.67%)\n80703.72 12445.7\n(15.42%)\n343166.82\n(425.22%)\n-274908.81\n(-340.64%)\nLow SDI 224730.03 -511.24\n(-0.23%)\n457893.76\n(203.75%)\n-232652.48\n(-103.53%)\n1385798.85 -468.78\n(-0.03%)\n2784468.73\n(200.93%)\n-1398201.11\n(-100.89%)\n128358.07 -97.12\n(-0.08%)\n253975.67\n(197.86%)\n-125520.48\n(-97.79%)\nGBD region\nAndean Latin \nAmerica\n4143.7 -1752.74\n(-42.3%)\n15189.97\n(366.58%)\n-9293.53\n(-224.28%)\n39761.37 5065.26\n(12.74%)\n88516.64\n(222.62%)\n-53820.53\n(-135.36%)\n3637.8 439.31\n(12.08%)\n8154.74\n(224.17%)\n-4956.26\n(-136.24%)\nAustralasia 1267.38 -1114.7\n(-87.95%)\n3922.55\n(309.5%)\n-1540.47\n(-121.55%)\n13452.44 -1389.4\n(-10.33%)\n25431.33\n(189.05%)\n-10589.5\n(-78.72%)\n1256.02 -133.5\n(-10.63%)\n2325.82\n(185.17%)\n-936.31\n(-74.55%)\nCaribbean -1349.18 -1562.38\n(115.8%)\n5351.71\n(-396.66%)\n-5138.51\n(380.86%)\n2132.86 1041.3\n(48.82%)\n33405.64\n(1566.24%)\n-32314.07\n(-1515.06%)\n156.82 71.39\n(45.52%)\n3063.26\n(1953.34%)\n-2977.82\n(-1898.87%)\nCentral Asia 2873.81 -3620.45\n(-125.98%)\n15727.6\n(547.27%)\n-9233.35\n(-321.29%)\n39964.28 2658.79\n(6.65%)\n99194.34\n(248.21%)\n-61888.85\n(-154.86%)\n3630.44 189.65\n(5.22%)\n9188.07\n(253.08%)\n-5747.28\n(-158.31%)\nCentral Europe -14819.64 -4163.73\n(28.1%)\n-6277.71\n(42.36%)\n-4378.2\n(29.54%)\n-69770.89 -889.11\n(1.27%)\n-42768.23\n(61.3%)\n-26113.55\n(37.43%)\n-6524.41 -150.42\n(2.31%)\n-3953.38\n(60.59%)\n-2420.6\n(37.1%)\nCentral Latin \nAmerica\n1548.96 -9049.13\n(-584.21%)\n48965.33\n(3161.18%)\n-38367.24\n(-2476.97%)\n77108.97 12598.99\n(16.34%)\n293150.27\n(380.18%)\n-228640.29\n(-296.52%)\n6901.97 990.49\n(14.35%)\n27003.9\n(391.25%)\n-21092.42\n(-305.6%)\nCentral Sub-\nSaharan Africa\n26915.85 -418.36\n(-1.55%)\n55564.8\n(206.44%)\n-28230.59\n(-104.88%)\n158618.07 260.8\n(0.16%)\n327707.23\n(206.6%)\n-169349.96\n(-106.77%)\n14656.26 11.6\n(0.08%)\n29750.28\n(202.99%)\n-15105.62\n(-103.07%)\nEast Asia -259803.1 -73729.5\n(28.38%)\n54827.8\n(-21.1%)\n-240901.4\n(92.72%)\n-886838.3 298450.19\n(-33.65%)\n343417.25\n(-38.72%)\n-1528705.74\n(172.38%)\n-82995.71 26017.94\n(-31.35%)\n31951.92\n(-38.5%)\n-140965.57\n(169.85%)\nEastern Europe -30429.52 -8105.03\n(26.64%)\n-19265.52\n(63.31%)\n-3058.97\n(10.05%)\n-155473.58 5268.56\n(-3.39%)\n-139412.01\n(89.67%)\n-21330.12\n(13.72%)\n-14637.08 227.97\n(-1.56%)\n-12809.25\n(87.51%)\n-2055.8\n(14.05%)\nEastern Sub-\nSaharan Africa\n65294.56 -1086.29\n(-1.66%)\n154463.89\n(236.56%)\n-88083.05\n(-134.9%)\n410760.63 7767.14\n(1.89%)\n888345.13\n(216.27%)\n-485351.63\n(-118.16%)\n38048.21 668.96\n(1.76%)\n81220.16\n(213.47%)\n-43840.91\n(-115.22%)\nHigh-income \nAsia Pacific\n-33235.49 -9633.04\n(28.98%)\n-11335.09\n(34.11%)\n-12267.36\n(36.91%)\n-153915.47 11272.06\n(-7.32%)\n-75590.18\n(49.11%)\n-89597.35\n(58.21%)\n-14426.14 920.48\n(-6.38%)\n-7000.96\n(48.53%)\n-8345.66\n(57.85%)\nHigh-income \nNorth America\n-42320.16 -7363.28\n(17.4%)\n19334.07\n(-45.69%)\n-54290.95\n(128.29%)\n-203400.14 -20642.22\n(10.15%)\n119859.64\n(-58.93%)\n-302617.57\n(148.78%)\n-19196.96 -1976.65\n(10.3%)\n10948.19\n(-57.03%)\n-28168.5\n(146.73%)\nTable 3 Incidence, prevalence and YLDs changes, decomposed by three population-level determinants: aging, population and epidemiological change\n\nPage 11 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nLocation Incidence Prevalence YLDs\nOverall \ndifference\nAging (Per-\ncent %)\nPopulation \n(Percent %)\nEpidemiological \nchange (Percent \n%)\nOverall \ndifference\nAging \n(Percent \n%)\nPopulation \n(Percent \n%)\nEpidemio-\nlogical change \n(Percent %)\nOverall \ndifference\nAging \n(Percent \n%)\nPopulation \n(Percent %)\nEpidemiolog-\nical change \n(Percent %)\nNorth Africa \nand Middle \nEast\n81390.06 -28897.69\n(-35.51%)\n228090.84\n(280.24%)\n-117803.08\n(-144.74%)\n696104.36 61262.06\n(8.8%)\n1389352.8\n(199.59%)\n-754510.5\n(-108.39%)\n62972.61 5016\n(7.97%)\n127006.26\n(201.68%)\n-69049.64\n(-109.65%)\nOceania 4643.25 -244.91\n(-5.27%)\n6644.45\n(143.1%)\n-1756.29\n(-37.82%)\n32789.81 1753.34\n(5.35%)\n41208.17\n(125.67%)\n-10171.7\n(-31.02%)\n3020.72 153.69\n(5.09%)\n3793.6\n(125.59%)\n-926.58\n(-30.67%)\nSouth Asia 135743.42 -40917.33\n(-30.14%)\n618570.69\n(455.69%)\n-441909.94\n(-325.55%)\n936710.51 119008.42\n(12.7%)\n3853516.61\n(411.39%)\n-3035814.52\n(-324.09%)\n87635.03 10048.38\n(11.47%)\n351699.72\n(401.32%)\n-274113.07\n(-312.79%)\nSoutheast Asia 44672.55 -29112.38\n(-65.17%)\n170824.49\n(382.39%)\n-97039.57\n(-217.22%)\n488059.78 64870.98\n(13.29%)\n1015017.25\n(207.97%)\n-591828.45\n(-121.26%)\n45101.2 5470.57\n(12.13%)\n94183.47\n(208.83%)\n-54552.84\n(-120.96%)\nSouthern Latin \nAmerica\n3461.2 -554.81\n(-16.03%)\n8278.13\n(239.17%)\n-4262.12\n(-123.14%)\n31537.37 4869.99\n(15.44%)\n51669.95\n(163.84%)\n-25002.57\n(-79.28%)\n2829.94 429.49\n(15.18%)\n4737.41\n(167.4%)\n-2336.96\n(-82.58%)\nSouthern Sub-\nSaharan Africa\n7004.47 -2915.59\n(-41.62%)\n19129.7\n(273.11%)\n-9209.64\n(-131.48%)\n66252.34 8294.31\n(12.52%)\n115780.21\n(174.76%)\n-57822.18\n(-87.28%)\n5803.79 690.38\n(11.9%)\n10547.19\n(181.73%)\n-5433.78\n(-93.62%)\nTropical Latin \nAmerica\n8246.96 -8424.02\n(-102.15%)\n40,779\n(494.47%)\n-24108.02\n(-292.33%)\n150810.21 11,238\n(7.45%)\n253965.95\n(168.4%)\n-114393.75\n(-75.85%)\n13563.06 855.27\n(6.31%)\n23178.58\n(170.89%)\n-10470.79\n(-77.2%)\nWestern \nEurope\n-14597.53 -12318.95\n(84.39%)\n2532.94\n(-17.35%)\n-4811.52\n(32.96%)\n-60966.04 -28304.25\n(46.43%)\n17662.72\n(-28.97%)\n-50324.51\n(82.55%)\n-5781.27 -2742.36\n(47.44%)\n1622.46\n(-28.06%)\n-4661.37\n(80.63%)\nWestern Sub-\nSaharan Africa\n126274.28 -3675.81\n(-2.91%)\n215682.9\n(170.81%)\n-85732.81\n(-67.89%)\n792120.98 -2834.02\n(-0.36%)\n1247366.32\n(157.47%)\n-452411.32\n(-57.11%)\n73105.26 -332.99\n(-0.46%)\n114080.01\n(156.05%)\n-40641.76\n(-55.59%)\nYLDs, years lived with disability; SDI, socio-demographic index\nTable 3 (continued)\n \n\nPage 12 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nyears, and these upwards are primarily driven by popu -\nlation growth. Conversely, epidemiological changes, \nreflecting reductions in incidence, prevalence, and YLDs, \nmitigated these increases.\nThe joinpoint regression analysis shows that from \n1990 to 2021, the burden of endometriosis has decreased \nyear by year in most SDI regions. However, high-middle \nand middle SDI regions experienced fluctuating trends: \ndeclines from 1990 to 2010 and 2015 to 2021, but an \nincrease from 2010 to 2015. This fluctuation may be due \nto changes in medical therapeutic patterns or therapy \nguidelines. For example, a German study indicated that \nthe proportion of endometriosis patients treated with \ndienogest significantly increased between 2010 and 2019, \nand during the same period, the prevalence of endome -\ntriosis also significantly increased [18].\nThe analysis of the burden of surgically confirmed \nendometriosis by age groups in 2021 reveals critical \ninsights into the demographic distribution of the disease. \nThe global age-specific incidence rate and the number of \nincident cases peaked in the 20–24 age group. The global \nprevalence rate and the number of prevalent cases, as \nwell as the YLDs rate and the number of YLDs, peaked \nin the 25–29 age group in most SDI regions. The peak \nburden in women aged 20–30 highlights its impact on \nquality of life and reproductive health. Thus, this period \nis a critical time for intervention. We should focus on the \nthe prevention and comprehensive management of the \n20–30 age group, and improve the ability of early diag -\nnosis and treatment. However, in the high SDI region, the \nhighest prevalence and YLDs rates were concentrated in \nthe 40–44 age group. Previous studies have shown that \ncesarean section and induced abortion are important risk \nfactors that cannot be ignored for developing endometri -\nosis [ 19, 20]. The average childbearing age in developed \ncountries is higher than in less developed countries, and \nthe prevalence of cesarean and abortion procedures is \nFig. 6 Changes in endometriosis incidence, decomposed by three population-level determinants: aging, population and epidemiological change\n \n\nPage 13 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \nvery high. This may be an important reason why the age \nof onset in high SDI regions is later than in other regions.\nBy analyzing the relationship between SDI levels \nand endometriosis burden, we found that the aver -\nage expected values of the burden of endometriosis \ndecreased with increasing SDI levels. This contrasts with \nsome previous studies [ 21, 22], but also shows the same \ntrend as the conclusions of others [ 23, 24]. In fact, the \ntrue incidence of endometriosis is difficult to determine, \nbecause the gold standard for the diagnosis of endome -\ntriosis is the combination of laparoscopy visualization \nand histologic confirmation of the presence of endome -\ntrial glands and/or stroma [25, 26]. However, laparoscopy \nis an invasive procedure, and clinicians in some regions \ntypically prefer other non-invasive techniques, such as \nultrasound and magnetic resonance imaging (MRI), to \nidentify endometriosis, but their accuracy is limited [ 27]. \nFurthermore, the nonspecific nature of endometriosis \nsymptoms and the tendency to normalise them may con -\ntribute to the delay in diagnosis. For example, non-spe -\ncific symptoms such as dysmenorrhea have often been \ntreated with hormonal drugs without consideration of \nendometriosis [ 28]. Thus, clinicians’ skills, awareness of \nendometriosis, and economic and geographic access to \ncare will all affect diagnostic outcomes [ 29, 30]. In high-\nlevel SDI regions, the increased medical management of \nendometriosis reduce the need for surgical treatment, \nthereby decreasing the incidence of surgically confirmed \ndiagnoses. Moreover, the operative treatment and diag -\nnostic procedures concerning fibroids in particular, and \nalso female sterilization and infertility, have decreased \nduring the years, decreasing the possibility to diagnose \nendometriosis as an incidental finding. These chang -\ning treatment trends may reduce the incidence of sur -\ngically validated endometriosis [ 23]. Additionally, oral \ncontraceptives have been proven to significantly reduce \nmenstrual flow and may prevent the occurrence of endo -\nmetriosis by interfering with the implantation of ret -\nrograde endometrial cells [ 31, 32]. The use of the pill is \nmore widespread in developed regions, which reduces \nthe incidence of endometriosis to some extent [ 33– 35]. \nMeanwhile, multiple studies have shown that environ -\nmental toxicants such as dioxins, phthalates, bisphenol \nA, or organochlorinated pollutants play a significant role \nin the development of endometriosis [ 36– 38]. Compared \nto high-level SDI regions, low-level SDI regions gener -\nally have poorer environmental governance. Activities \nsuch as waste incineration or metal smelting release large \namounts of dioxins, increasing the likelihood of exposure \nto harmful chemicals. These warn us that regions with \nlower SDI may face more severe challenges. Address -\ning these disparities requires multifaceted approaches, \nincluding promoting access to healthcare, enhancing \nhealth education, improving living environments, and \nimplementing targeted public health strategies.\nFurthermore, we employed decomposition analysis to \ndisentangle the contributions of population aging, popu -\nlation growth, and epidemiological changes to the disease \nburden. High SDI and High-middle SDI regions expe -\nrienced declines in numbers of incidence, prevalence, \nand YLDs, primarily driven by favorable epidemiologi -\ncal changes. This indicates that disease prevention and \nhealth promotion can effectively mitigate the challenges \nposed by demographic changes (population growth and \npopulation aging) to endometriosis. Low SDI and Low-\nmiddle SDI regions saw increases in incident cases, \nprevalent cases, and YLDs cases, with population growth \nbeing the dominant factor. Statistically, the 47 least devel-\noped countries are among the fastest growing countries \nin the world, and many of them are expected to double \ntheir populations from 2019 to 2050 [ 34]. Overall, in the \ncoming decades, population growth will have a greater \nimpact on some low SDI and lower-middle SDI regions, \nleading to a continued increase in the burden of endo -\nmetriosis, while the impact on high-level SDI regions \nwill stabilize. Therefore, when formulating or adjusting \nhealth prevention measures, international organizations \nor national governments should consider the poten -\ntial impact of population growth on health in different \nregions.\nTo the best of our knowledge, this study is the first to \ncomprehensively analyze the global burden of surgi -\ncally confirmed endometriosis from 1990 to 2021, using \nrobust statistical methods to assess trends and correla -\ntions. With each iteration of the GBD, the disease clas -\nsification methods have become more standardized, and \nmore in-depth systematic evaluation methods have been \nused to obtain country-specific information, provid -\ning reliable data sources for this study. Methodological \nadvancements have enabled GBD 2021 to produce esti -\nmates more easily than in previous iterations; however, as \nwith any study of this scope, there are several important \nlimitations to acknowledge. First of all, inconsistencies in \nthe availability of primary epidemiological data remain \na limitation and source of instability within GBD analy -\nses. The estimates of disease burden depend on the out-\nof-sample predictive validity of modelling processes in \ncases where data are insufficient to produce burden esti -\nmates for all 204 countries and territories (by year, sex, \nand age). Although this approach cannot fully replace \nhigh quality primary data, it ensures that populations \nor causes with no or little data are not excluded from \nimportant benchmarking exercises intended for bur -\nden estimation. In addition, with any given GBD release, \nthere might be extant data not identified or incorporated, \nwhich is a key part of the rationale for ongoing cycles \nof releases, rather than a single update. For the primary \n\nPage 14 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \ndata available, the data processing methods account for \nknown sources of variation wherever possible, but fully \ndisentangling variation in estimates is not always possible \ndue to measurement error and reporting inaccuracies. \nThere are problems with the quality and collection of pri-\nmary data, such as flawed methodologies and potential \nunder-reporting of illnesses, which is a recurring limi -\ntation for GBD that can be continually improved on by \nstrengthening data-collection systems [ 39]. This study \nalso has some limitations. First, sparsity of data or unreli-\nability of data from specific regions, time periods, or age \ngroups can influence the accuracy of the endometriosis \nburden estimates, particularly poor data quality and cov -\nerage from western, eastern, southern, and central sub-\nSaharan Africa and south Asia [ 40]. Second, the disease \nburden may be underestimated in some low- and mid -\ndle-income regions due to limited data or lack of gold-\nstandard diagnostics like laparoscopy. Third, as our study \nspans three decades, changes in the diagnostic criteria \nfor the disease may impact the temporal trend analysis. \nFinally, in the decomposition analysis, the selection of \ndriving factors may not be comprehensive enough, attrib-\nuting only to population aging, population growth, and \nepidemiological changes. Due to the lack of relevant data, \nother influencing factors such as environment, diet, life -\nstyle, or genetic susceptibility were not included tempo -\nrarily, future research may focus on this issue.\nConclusions\nDespite the global age-standardized rates of incidence, \nprevalence, and YLDs have shown a decreasing trend \nin the past 30 years, endometriosis will continue to be a \nmajor public health burden due to the increasing num -\nber of cases worldwide. Managing this condition remains \na significant challenge and requires better allocation \nof healthcare resources and more targeted interven -\ntions. Our study has comprehensively assessed the bur -\nden of surgically confirmed endometriosis and offered \nepidemiological evidence, which will provide valuable \nsolutions for relevant policymakers to improve health \nmanagement.\nSupplementary Information\nThe online version contains supplementary material available at  h t t p  s : /  / d o i  . o  r \ng /  1 0 .  1 1 8 6  / s  1 2 9 5 8 - 0 2 5 - 0 1 4 2 1 - z.\nSupplementary Material 1: Table S1. The prevalence of endometriosis in \n1990 and 2021 and temporal trends between 1990 and 2021.\nSupplementary Material 2: Table S2. The YLDs of endometriosis in 1990 \nand 2021 and temporal trends between 1990 and 2021.\nSupplementary Material 3: Figure S1. The ASPR of endometriosis in 204 \ncountries and territories in 2021. ASPR, age-standardized prevalence rate.\nSupplementary Material 4: Figure S2. The age-standardized YLDs rate of \nendometriosis in 204 countries and territories in 2021. YLDs, years lived \nwith disability.\nSupplementary Material 5: Figure S3. Joinpoint regression analysis in ASPR \nfrom 1990 to 2021 by SDI region, * P<0.05. ASPR, age-standardized preva-\nlence rate; SDI, socio-demographic index.\nSupplementary Material 6: Figure S4. Joinpoint regression analysis in age-\nstandardized YLDs rate from 1990 to 2021 by SDI region, * P<0.05. YLDs, \nyears lived with disability; SDI, socio-demographic index.\nSupplementary Material 7: Figure S5. Coevolution of ASPR with SDI glob-\nally and for GBD regions of endometriosis, 1990–2021. ASPR, age-stan-\ndardized prevalence rate; SDI, socio-demographic index.\nSupplementary Material 8: Figure S6. Coevolution of age-standardized \nYLDs rate with SDI globally and for GBD regions, 1990–2021. YLDs, years \nlived with disability; SDI, socio-demographic index.\nSupplementary Material 9: Figure S7. Changes in endometriosis preva-\nlence, decomposed by three population-level determinants: aging, \npopulation, and epidemiological change.\nSupplementary Material 10: Figure S8. Changes in endometriosis YLDs, \ndecomposed by three population-level determinants: aging, population, \nand epidemiological change. YLDs, years lived with disability.\nAuthor contributions\nRuijie Li: Methodology, Data curation, Formal analysis, Writing– original draft. \nLing Zhang: Writing– review & editing. Yi Liu: Conceptualization, Project \nadministration, Writing– review & editing.\nFunding\nThis study was financially supported by the National Natural Science \nFoundation of China (grant numbers: 82371681)and by Hubei Provincial \nNatural Science Foundation of China (2024AFB675).\nData availability\nNo datasets were generated or analysed during the current study.\nDeclarations\nEthical approval\nThe requirement for ethical approval and informed consent was not \napplicable because the data in this study were secondary data and did not \ncontain any data which could identify individuals.\nCompeting interests\nThe authors declare no competing interests.\nReceived: 14 November 2024 / Accepted: 19 May 2025\nReferences\n1. International Working Group of AAGL E, Tomassetti C, Johnson NP , et al. \nAn international terminology for endometriosis, 2021. Hum Reprod Open. \n2021;2021(4):hoab029.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 9 3  / h  r o p e n / h o a b 0 2 9.\n2. Taylor HS, Kotlyar AM, Flores VA. Endometriosis is a chronic systemic disease: \nclinical challenges and novel innovations. Lancet. 2021;397(10276):839–52.  h t \nt p  s : /  / d o i  . o  r g /  1 0 .  1 0 1 6  / S  0 1 4 0 - 6 7 3 6 ( 2 1 ) 0 0 3 8 9 - 5.\n3. Dai Y, Li X, Shi J, Leng J. A review of the risk factors, genetics and treatment \nof endometriosis in Chinese women: a comparative update. Reprod Health. \n2018;15(1):82.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 1 8 6  / s  1 2 9 7 8 - 0 1 8 - 0 5 0 6 - 7.\n4. GBD 2017 DALYs and, Collaborators HALE. Global, regional, and National \ndisability-adjusted life-years (DALYs) for 359 diseases and injuries and \nhealthy life expectancy (HALE) for 195 countries and territories, 1990–2017: \na systematic analysis for the global burden of disease study 2017. Lancet. \n2018;392(10159):1859–922.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 1 6  / S  0 1 4 0 - 6 7 3 6 ( 1 8 ) 3 2 3 3 5 - 3.\n5. Shafrir AL, Farland LV, Shah DK, et al. Risk for and consequences of endome-\ntriosis: A critical epidemiologic review. Best Pract Res Clin Obstet Gynaecol. \n2018;51:1–15.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 1 6  / j  . b p  o b g  y n . 2  0 1  8 . 0 6 . 0 0 1.\n\nPage 15 of 15\nLi et al. Reproductive Biology and Endocrinology            (2025) 23:88 \n6. Friedl F, Riedl D, Fessler S, et al. Impact of endometriosis on quality of life, \nanxiety, and depression: an Austrian perspective. Arch Gynecol Obstet. \n2015;292(6):1393–9.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 0 7  / s  0 0 4 0 4 - 0 1 5 - 3 7 8 9 - 8.\n7. M G, B A, R J, R A. Physical pain and emotion regulation as the main predictive \nfactors of health-related quality of life in women living with endometriosis. \nHum Reprod (Oxford England). 2017;32(7).  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 9 3  / h  u m r e p / d \ne x 0 9 1.\n8. Becker CM, Bokor A, Heikinheimo O, et al. ESHRE guideline: endometriosis. \nHum Reprod Open. 2022;2022(2):hoac009.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 9 3  / h  r o p e n / h \no a c 0 0 9.\n9. Kuznetsov L, Dworzynski K, Davies M, Overton C. Diagnosis and manage -\nment of endometriosis: summary of NICE guidance. BMJ. 2017;358:j3935.  h t t \np  s : /  / d o i  . o  r g /  1 0 .  1 1 3 6  / b  m j . j 3 9 3 5.\n10. Jones R, Barraclough K, Dowrick C. When no diagnostic label is applied. BMJ. \n2010;340:c2683.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 1 3 6  / b  m j . c 2 6 8 3.\n11. Farland LV, Horne AW. Disparity in endometriosis diagnoses between Racial/\nethnic groups. BJOG. 2019;126(9):1115–6.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 1 1 1  / 1  4 7 1 - 0 5 2 8 . \n1 5 8 0 5.\n12. Vos T, Lim SS, Abbafati C, et al. Global burden of 369 diseases and injuries in \n204 countries and territories, 1990–2019: a systematic analysis for the global \nburden of disease study 2019. Lancet. 2020;396(10258):1204–22.  h t t p  s : /  / d o i  . o  \nr g /  1 0 .  1 0 1 6  / S  0 1 4 0 - 6 7 3 6 ( 2 0 ) 3 0 9 2 5 - 9.\n13. Analysis of global. prevalence, DALY and trends of inflammatory bowel \ndisease and their correlations with sociodemographic index: data from 1990 \nto 2019. Autoimmunity reviews. Published Online Oct. 2024;2:103655.  h t t p  s : /  \n/ d o i  . o  r g /  1 0 .  1 0 1 6  / j  . a u  t r e  v . 2 0  2 4  . 1 0 3 6 5 5.\n14. Kim HJ, Fay MP , Feuer EJ, Midthune DN. Permutation tests for joinpoint \nregression with applications to cancer rates. Stat Med. 2000;19(3):335–51.  h t t \np  s : /  / d o i  . o  r g / 1 0 . 1 0 0 2 / ( S I C I ) 1 0 9 7 - 0 2 5 8 ( 2 0 0 0 0 2 1 5 ) 1 9 : 3 % 3 C 3 3 5 : : A I D - S I M 3 3 6 % 3 E \n3 . 0 . C O ; 2 - Z.\n15. Li HZ, Du LB. [Application of joinpoint regression model in cancer epidemio-\nlogical time trend analysis]. Zhonghua Yu Fang Yi Xue Za Zhi. 2020;54(8):908–\n12.  h t t p  s : /  / d o i  . o  r g /  1 0 .  3 7 6 0  / c  m a .  j . c  n 1 1 2  1 5  0 - 2 0 2 0 0 6 1 6 - 0 0 8 8 9.\n16. X C, Y Y, Dc S, et al. Population ageing and mortality during 1990–2017: A \nglobal decomposition analysis. PLoS Med. 2020;17(6).  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 3 7 1  \n/ j  o u r  n a l  . p m e  d .  1 0 0 3 1 3 8.\n17. P DG. Standardization and decomposition of rates from cross-classified data. \nGenus. 1994;50(3–4). Accessed June 19, 2024.  h t t p  s : /  / p u b  m e  d . n  c b i  . n l m  . n  i h . g \no v / 1 2 3 1 9 2 5 6 /\n18. Göhring J, Drewes M, Kalder M, Kostev K. Germany endometriosis pattern \nchanges; prevalence and therapy over 2010 and 2019 years: A retrospective \nCross- sectional study. Int J Fertil Steril. 2022;16(2):85–9.  h t t p  s : /  / d o i  . o  r g /  1 0 .  2 2 \n0 7  4 /  I J F  S . 2  0 2 1 .  5 2  8 3 9 7 . 1 1 1 3.\n19. Ananias P , Luenam K, Melo JP et al. Cesarean section: A potential and forgot-\nten risk for abdominal wall endometriosis. Cureus 13(8):e17410.  h t t p  s : /  / d o i  . o  r \ng /  1 0 .  7 7 5 9  / c  u r e u s . 1 7 4 1 0\n20. Jenabi E, Salimi Z, Bashirian S, Khazaei S, Ayubi E. The risk factors associated \nwith placenta previa: an umbrella review. Placenta. 2022;117:21–7.  h t t p  s : /  / d o i  \n. o  r g /  1 0 .  1 0 1 6  / j  . p l  a c e  n t a .  2 0  2 1 . 1 0 . 0 0 9.\n21. Zhang S, Gong TT, Wang HY, Zhao YH, Wu QJ. Global, regional, and National \nendometriosis trends from 1990 to 2017. Ann N Y Acad Sci. 2021;1484(1):90–\n101.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 1 1 1  / n  y a s . 1 4 4 6 8.\n22. Gylfason JT, Kristjansson KA, Sverrisdottir G, Jonsdottir K, Rafnsson V, Geirsson \nRT. Pelvic endometriosis diagnosed in an entire Nation over 20 years. Am J \nEpidemiol. 2010;172(3):237–43.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 9 3  / a  j e / k w q 1 4 3.\n23. Saavalainen L, Tikka T, But A, et al. Trends in the incidence rate, type and treat-\nment of surgically verified endometriosis– a nationwide cohort study. Acta \nObstet Gynecol Scand. 2018;97(1):59–67.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 1 1 1  / a  o g s . 1 3 2 4 4.\n24. Morassutto C, Monasta L, Ricci G, Barbone F, Ronfani L. Incidence and esti-\nmated prevalence of endometriosis and adenomyosis in Northeast Italy: A \ndata linkage study. PLoS ONE. 2016;11(4):e0154227.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 3 7 1  / j  \no u r  n a l  . p o n  e .  0 1 5 4 2 2 7.\n25. Kennedy S, Bergqvist A, Chapron C, et al. ESHRE guideline for the diagnosis \nand treatment of endometriosis. Hum Reprod. 2005;20(10):2698–704.  h t t p  s : /  / \nd o i  . o  r g /  1 0 .  1 0 9 3  / h  u m r e p / d e i 1 3 5.\n26. Dunselman GaJ, Vermeulen N, Becker C, et al. ESHRE guideline: management \nof women with endometriosis. Hum Reprod. 2014;29(3):400–12.  h t t p  s : /  / d o i  . o  \nr g /  1 0 .  1 0 9 3  / h  u m r e p / d e t 4 5 7.\n27. Moore J, Copley S, Morris J, Lindsell D, Golding S, Kennedy S. A systematic \nreview of the accuracy of ultrasound in the diagnosis of endometriosis. \nUltrasound Obstet Gynecol. 2002;20(6):630–4.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 4 6  / j  . 1 4  6 9 \n-  0 7 0 5  . 2  0 0 2 . 0 0 8 6 2 . x.\n28. Horne AW, Missmer SA. Pathophysiology, diagnosis, and management of \nendometriosis. BMJ. 2022;379:e070750.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 1 3 6  / b  m j - 2 0 2 2 - 0 7 \n0 7 5 0.\n29. Zondervan KT, Yudkin PL, Vessey MP , et al. Chronic pelvic pain in the com-\nmunity–symptoms, investigations, and diagnoses. Am J Obstet Gynecol. \n2001;184(6):1149–55.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 6 7  / m  o b . 2 0 0 1 . 1 1 2 9 0 4.\n30. Olive DL, Schwartz LB, Endometriosis. N Engl J Med. 1993;328(24):1759–69.  h t \nt p  s : /  / d o i  . o  r g /  1 0 .  1 0 5 6  / N  E J M  1 9 9  3 0 6 1  7 3  2 8 2 4 0 7.\n31. de Ziegler D, Borghese B, Chapron C. Endometriosis and infertility: patho-\nphysiology and management. Lancet. 2010;376(9742):730–8.  h t t p  s : /  / d o i  . o  r g /  \n1 0 .  1 0 1 6  / S  0 1 4 0 - 6 7 3 6 ( 1 0 ) 6 0 4 9 0 - 4.\n32. Teal S, Edelman A. Contraception selection, effectiveness, and adverse effects: \nA review. JAMA. 2021;326(24):2507–18.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 0 1  / j  a m a . 2 0 2 1 . 2 1 \n3 9 2.\n33. Spira A. Contraception by the end of the 20th century. Hum Reprod. \n1994;9(3):445–7.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 9 3  / o  x f o  r d j  o u r n  a l  s . h u m r e p . a 1 3 8 5 2 5.\n34. Haakenstad A, Angelino O, Irvine CMS, et al. Measuring contraceptive \nmethod mix, prevalence, and demand satisfied by age and marital status in \n204 countries and territories, 1970–2019: a systematic analysis for the global \nburden of disease study 2019. Lancet. 2022;400(10348):295–327.  h t t p  s : /  / d o i  . o  \nr g /  1 0 .  1 0 1 6  / S  0 1 4 0 - 6 7 3 6 ( 2 2 ) 0 0 9 3 6 - 9.\n35. Sai FT. Political and economic factors influencing contraceptive uptake. Br \nMed Bull. 1993;49(1):200–9.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 9 3  / o  x f o  r d j  o u r n  a l  s . b m b . a 0 7 2 \n5 9 8.\n36. Porpora MG, Resta S, Fuggetta E, et al. Role of environmental organochlo-\nrinated pollutants in the development of endometriosis. Clin Exp Obstet \nGynecol. 2013;40(4):565–7.\n37. Buck Louis GM, Peterson CM, Chen Z, et al. Bisphenol A and phthalates and \nendometriosis: the endometriosis: natural history, diagnosis and outcomes \nstudy. Fertil Steril. 2013;100(1):162–9. e1-2.\n38. Epidemiology of endometriosis and its comorbidities. Eur J Obstet Gynecol \nReproductive Biology. 2017;209:3–7.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 1 6  / j  . e j  o g r  b . 2 0  1 6  . 0 4 . \n0 2 1.\n39. Global incidence. prevalence, years lived with disability (YLDs), disability-\nadjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 \ndiseases and injuries in 204 countries and territories and 811 subnational \nlocations, 1990–2021: a systematic analysis for the global burden of disease \nstudy 2021. Lancet. 2024;403(10440):2133–61.  h t t p  s : /  / d o i  . o  r g /  1 0 .  1 0 1 6  / S  0 1 4 \n0 - 6 7 3 6 ( 2 4 ) 0 0 7 5 7 - 8.\n40. Global burden of. 288 Causes of death and life expectancy decomposition \nin 204 countries and territories and 811 subnational locations, 1990–2021: \na systematic analysis for the global burden of disease study 2021. Lancet. \n2024;403(10440):2100–32.  h t t p s :   /  / d o  i .  o r  g  /  1 0  . 1 0   1 6  / S 0  1 4 0 - 6 7 3 6 ( 2 4 ) 0 0 3 6 7 - 2.\nPublisher’s note\nSpringer Nature remains neutral with regard to jurisdictional claims in \npublished maps and institutional affiliations.","source_license":"CC0","license_restricted":false}