Endometriosis Burden Among Adolescents and Young Adult Women in China and the Western Pacific Region from 1990 to 2023: A Systematic Analysis for the GBD 2023

Endometriosis can cause substantial pain and functional impairment during adolescence and young adulthood. It is a chronic, estrogen-dependent gynecological condition characterized by endometrial-like tissue outside the uterus, commonly causing dysmenorrhea, chronic pelvic pain, dyspareunia, fatigue, and infertility. These symptoms can restrict daily activities and affect education, employment, intimate relationships, psychological well-being, and quality of life. 1–3 Although endometriosis is rarely fatal, its chronic and recurrent course can result in sustained disability and continued use of reproductive health services. Symptoms often emerge during adolescence or early adulthood and their effects may extend across the reproductive life through pain management, fertility decisions, reproductive autonomy, and healthcare needs. 4–6 Adolescents and young women warrant separate consideration because early manifestations of endometriosis may be difficult to distinguish from common menstrual symptoms. 5 Dysmenorrhea and pelvic pain may be normalized by patients, families, teachers, and healthcare professionals. 7 Some young patients consequently receive repeated symptomatic treatment including non-steroidal anti-inflammatory drugs before gynecological assessment. Clinical presentation may also differ between adolescents and older women, further complicating recognition. 7 , 8 The period from 15 to 29 years encompasses changes in education, employment, independent healthcare use, sexual health, and fertility planning. Persistent pain or functional impairment during these years may have consequences beyond the immediate clinical symptoms. Assessing endometriosis burden in this age group requires consideration of both disease frequency and its disabling effects. Global Burden of Disease (GBD) estimates of incidence and prevalence are influenced by symptom recognition, healthcare-seeking behavior, access to gynecological assessment, diagnostic practice, and data availability. Years lived with disability (YLDs) provide an additional measure of non-fatal health loss, but they remain dependent on the prevalence estimates and disability weights. 9–12 These considerations are relevant to the Western Pacific Region (WPR), where countries and areas differ substantially in population size, geography, and health-system capacity. 13 China was examined separately because its large population contributes substantially to regional case numbers. Comparison with the GBD WPR aggregate provides a regional reference for China’s estimates. Analyzing the 31 countries and areas identifies where estimated rates are high and where case numbers are large. Regional averages may obscure these differences. Considering both measures may inform local assessments of capacity for pelvic pain evaluation, continuing care, and specialist referral for adolescents and young women. 9 , 14 Previous GBD studies have described endometriosis burden globally and across broad reproductive-age or fully modeled female populations. 2 , 14–18 Several identified the highest incidence during ages 20–24 years and the highest prevalence or disability burden during ages 25–29 years. 9 , 14 , 17 These findings suggest that analyses covering the entire reproductive age range may conceal important differences between adolescents, young adults, and older women. However, their findings on socioeconomic patterning have varied according to the GBD iteration, geographical unit, age range, and burden measure examined. Analyses based on broad reproductive-age groups may also conceal differences between adolescents, young adults, and older women. Evidence therefore remains limited on how age-specific patterns, national variation, socioeconomic development, and demographic change jointly shape the burden among adolescents and young women in this region. 2 , 5 , 19 , 20 This study provides a focused assessment of endometriosis burden among females aged 15–29 years in China and the WPR using GBD 2023 estimates. We examined incidence, prevalence, and YLDs from 1990 to 2023. Regional temporal trends were assessed using the GBD WPR, while geographical and socioeconomic comparisons were conducted across 31 WPR countries and areas, with China also analyzed separately. Age-specific analyses covering females aged 10–39 years were used to determine whether ages 15–29 represent a period in which burden becomes increasingly apparent. By examining age-standardized rates alongside absolute numbers, we sought to distinguish high-rate settings from populous settings with large numbers of affected individuals. We also decomposed changes in absolute burden into population-size, age-structure, and age-specific rate components. Together, these analyses were intended to provide regionally relevant evidence for reproductive health planning without treating modeled burden as a direct measure of diagnostic delay or individual disease progression.

This study was a secondary analysis of publicly available data from GBD 2023. GBD 2023 quantified health loss from 375 diseases and injuries and 88 risk factors in 204 countries and territories from 1990 to 2023. 12 , 21 Annual estimates for the GBD 2023 endometriosis cause category from 1990 to 2023 were obtained from the GBD Results Tool ( https://vizhub.healthdata.org/gbd-results/ ). We extracted incidence, prevalence, and YLDs among females for the GBD-provided WPR and 31 selected WPR countries and areas, including China. Individual-level information on clinical, surgical, or histological confirmation was unavailable in the aggregate GBD data. Data were available as age-specific numbers and rates per 100,000 population, with 95% uncertainty intervals (UIs). The primary study population comprised females aged 15–29 years, combining the 15–19, 20–24, and 25–29-year age groups. Additional analyses included ages 10–14, 30–34, and 35–39 years to describe burden across early reproductive life. Published GBD endometriosis methods defined reference cases by pelvic examination with confirmation by laparoscopy. Inputs included epidemiological studies, insurance claims, and hospital discharge records, with diagnostic codes used to identify cases in administrative data. The estimates used in this study did not provide individual-level information on clinical, surgical, or histological confirmation. Two WPR data structures were used. Analyses of overall regional burden, temporal trends, age-specific comparisons with China, Joinpoint trends, and demographic decomposition used the GBD WPR. Country-level mapping, geographical comparisons, and SDI analyses used estimates for the 31 countries and areas. For SDI-stratified age-specific analyses, numbers and corresponding population denominators were summed within each stratum before rates were recalculated. Estimates from the 31-location set were not treated as equivalent to the GBD WPR. China was analyzed separately but remained part of the WPR. Annual SDI values were used as a continuous measure in the association analyses. For stratified analyses, country- and area-level SDI values for 2023 were classified using the reference quintile boundaries supplied with the IHME GBD 2023: low, 0 ≤ SDI < 0.531564; low-middle, 0.531564 ≤ SDI < 0.630485; middle, 0.630485 ≤ SDI < 0.676662; high-middle, 0.676662 ≤ SDI < 0.718827; and high, 0.718827 ≤ SDI ≤ 1. For stratified analyses, the 31 countries and areas were assigned to low, low-middle, middle, high-middle, or high SDI, which was held constant across the study analyses. 22

At the GBD WPR aggregate level, the absolute burden of endometriosis among females aged 15–29 years decreased between 1990 and 2023. The number of prevalent cases dropped from 1,876,232 (95% UI: 1,104,449 to 2,841,601) in 1990 to 1,333,780 (95% UI: 794,707 to 1,989,545) in 2023, a reduction of 28.9%. Incident number fell from 497,014 (95% UI: 266,677 to 827,416) to 336,592 (95% UI: 180,434 to 560,705), a reduction of 32.3%, and YLDs decreased from 175,998 (95% UI: 95,653 to 294,723) to 125,411 (95% UI: 69,661 to 207,734), a reduction of 28.7% ( Supplementary Table 1 ). Age-standardized rates varied widely across the 31 countries and areas. In 2023, New Zealand had the highest point estimates for ASIR (343.20 [95% UI: 197.23 to 559.27]), ASPR (1,572.59 [95% UI: 901.76 to 2528.10]), and age-standardized YLDs rate per 100,000 population (143.93 [95% UI: 77.31 to 251.48]). Samoa, Papua New Guinea, Tonga, and the Solomon Islands also had relatively high age-standardized rates ( Figure 1 and Supplementary Figure 1 ). In contrast, China had the lowest age-standardized rates per 100,000 population among the 31 countries and areas: ASIR 140.75 (95% UI: 74.47 to 235.63); ASPR 524.20 (95% UI: 306.90 to 798.92); and age-standardized YLDs rate 49.46 (95% UI: 26.14 to 82.16) ( Table 1 ). Table 1 Trends for Endometriosis Burden Among Females Aged 15-29 years in 31 Western Pacific Region by Country/Area and SDI Category, 1990–2023 SDI Category Location Age-Standardized Incidence Rate Age-Standardized Prevalence Rate Age-Standardized Years Lived With Disability Rate 1990 per 100,000 (95% UI) 2023 per 100,000 (95% UI) 1990–2023 EAPC (95% CI) P value 1990 per 100,000 (95% UI) 2023 per 100,000 (95% UI) 1990–2023 EAPC (95% CI) P value 1990 per 100,000 (95% UI) 2023 per 100,000 (95% UI) 1990–2023 EAPC (95% CI) P value Overall Western Pacific Region 194.49 (103.95, 325.39) 174.49 (93.76, 289.79) −0.66 (−0.79, −0.52) <0.001 744.96 (440.51, 1124.88) 679.56 (403.49, 1016.33) −0.55 (−0.68, −0.43) <0.001 69.86 (38.06, 116.61) 63.91 (35.45, 106.15) −0.54 (−0.67, −0.42) <0.001 High SDI Australia 247.35 (124.48, 423.30) 210.25 (101.34, 359.28) −0.38 (−0.47, −0.30) <0.001 1020.28 (557.15, 1664.66) 868.00 (471.40, 1395.98) −0.39 (−0.47, −0.30) <0.001 93.75 (45.89, 173.90) 78.98 (37.07, 142.60) −0.42 (−0.50, −0.33) <0.001 Japan 260.17 (142.20, 428.90) 240.00 (125.50, 399.30) −0.78 (−1.01, −0.56) <0.001 1088.54 (633.26, 1748.78) 1013.53 (559.45, 1625.84) −0.54 (−0.69, −0.39) <0.001 101.60 (54.00, 175.47) 94.86 (48.83, 162.81) −0.54 (−0.69, −0.39) <0.001 New Zealand 394.58 (229.30, 633.95) 343.20 (197.23, 559.27) −0.37 (−0.66, −0.07) 0.021 1757.02 (1013.58, 2728.19) 1572.59 (901.76, 2528.10) −0.21 (−0.53, 0.10) 0.193 161.70 (85.75, 273.02) 143.93 (77.31, 251.48) −0.22 (−0.53, 0.09) 0.179 Republic of Korea 235.63 (114.37, 407.01) 198.86 (98.02, 339.62) −0.64 (−0.68, −0.59) <0.001 1043.54 (575.95, 1709.13) 855.30 (467.99, 1408.03) −0.71 (−0.76, −0.67) <0.001 97.16 (49.06, 173.38) 80.28 (38.34, 140.11) −0.69 (−0.74, −0.64) <0.001 Singapore 227.34 (109.84, 389.64) 194.37 (97.63, 328.07) −0.56 (−0.61, −0.52) <0.001 999.35 (542.91, 1612.67) 839.50 (462.44, 1340.45) −0.61 (−0.67, −0.56) <0.001 93.77 (45.30, 171.13) 79.01 (39.05, 145.59) −0.61 (−0.65, −0.56) <0.001 American Samoa 311.57 (174.83, 521.01) 228.86 (127.79, 381.99) −0.87 (−0.91, −0.82) <0.001 1272.07 (790.09, 1927.64) 905.84 (552.30, 1338.05) −0.96 (−1.01, −0.91) <0.001 120.39 (61.72, 211.62) 85.03 (42.91, 147.06) −0.98 (−1.03, −0.92) <0.001 Brunei Darussalam 276.83 (133.15, 477.88) 209.51 (101.34, 359.32) −0.99 (−1.06, −0.92) <0.001 1279.85 (716.38, 2047.37) 909.54 (503.32, 1499.87) −1.14 (−1.23, −1.05) <0.001 119.48 (59.10, 208.38) 84.49 (40.80, 150.09) −1.14 (−1.22, −1.05) <0.001 China 169.09 (88.00, 287.47) 140.75 (74.47, 235.63) −1.12 (−1.32, −0.92) <0.001 624.98 (361.98, 958.66) 524.20 (306.90, 798.92) −0.98 (−1.14, −0.81) <0.001 58.74 (30.83, 97.49) 49.46 (26.14, 82.16) −0.97 (−1.13, −0.80) <0.001 Cook Islands 253.87 (141.12, 423.00) 208.77 (115.86, 349.23) −0.65 (−0.70, −0.60) <0.001 982.37 (578.78, 1528.43) 803.75 (478.61, 1241.37) −0.69 (−0.74, −0.63) <0.001 92.74 (50.63, 160.60) 75.25 (38.81, 128.18) −0.71 (−0.77, −0.65) <0.001 Guam 265.91 (147.76, 434.58) 229.26 (124.07, 385.67) −0.41 (−0.48, −0.34) <0.001 1033.51 (618.79, 1579.46) 909.47 (546.44, 1396.01) −0.41 (−0.50, −0.32) <0.001 97.90 (51.13, 165.25) 85.83 (43.87, 147.23) −0.42 (−0.51, −0.33) <0.001 Malaysia 246.80 (139.75, 403.04) 195.24 (110.20, 323.91) −0.75 (−0.79, −0.71) <0.001 972.85 (582.09, 1460.97) 765.70 (453.26, 1162.89) −0.79 (−0.84, −0.74) <0.001 91.98 (48.42, 158.45) 72.06 (35.98, 124.04) −0.79 (−0.84, −0.74) <0.001 Niue 255.95 (143.44, 420.79) 215.48 (120.44, 367.86) −0.47 (−0.51, −0.44) <0.001 997.72 (601.15, 1548.17) 833.92 (504.78, 1263.11) −0.48 (−0.51, −0.45) <0.001 94.49 (49.09, 157.51) 78.28 (40.78, 134.00) −0.49 (−0.53, −0.45) <0.001 Northern Mariana Islands 247.33 (138.29, 409.83) 213.38 (117.54, 355.12) −0.43 (−0.45, −0.42) <0.001 946.56 (555.92, 1466.16) 828.74 (500.79, 1268.30) −0.42 (−0.43, −0.41) <0.001 89.47 (43.77, 151.25) 78.66 (40.96, 135.75) −0.42 (−0.44, −0.41) <0.001 Palau 233.21 (130.52, 377.40) 211.11 (115.26, 352.36) −0.29 (−0.31, −0.26) <0.001 888.59 (533.48, 1327.33) 814.63 (479.51, 1245.88) −0.26 (−0.28, −0.24) <0.001 84.07 (42.38, 146.60) 75.64 (39.19, 127.11) −0.29 (−0.32, −0.27) <0.001 Tokelau 265.51 (147.36, 448.30) 214.16 (120.51, 363.00) −0.62 (−0.65, −0.60) <0.001 1038.91 (627.24, 1590.13) 832.07 (498.24, 1287.94) −0.67 (−0.70, −0.64) <0.001 97.38 (51.88, 166.78) 77.76 (39.94, 133.23) −0.66 (−0.69, −0.64) <0.001 High-middle SDI Fiji 258.12 (142.70, 428.84) 222.46 (123.21, 374.59) −0.38 (−0.42, −0.34) <0.001 1007.36 (595.27, 1508.45) 872.85 (515.47, 1300.30) −0.37 (−0.42, −0.32) <0.001 94.24 (48.71, 161.74) 81.50 (41.77, 143.49) −0.37 (−0.42, −0.33) <0.001 Middle SDI Philippines 250.62 (141.00, 417.93) 216.94 (120.97, 362.10) −0.68 (−0.75, −0.61) <0.001 979.70 (598.71, 1469.53) 852.88 (518.33, 1275.65) −0.59 (−0.65, −0.54) <0.001 91.77 (51.24, 151.80) 80.23 (44.91, 133.10) −0.58 (−0.63, −0.52) <0.001 Tonga 336.38 (188.40, 559.90) 277.78 (152.46, 469.52) −0.50 (−0.54, −0.46) <0.001 1397.45 (833.52, 2153.13) 1145.27 (685.66, 1772.39) −0.52 (−0.56, −0.48) <0.001 131.30 (69.99, 219.70) 107.22 (58.87, 177.54) −0.52 (−0.56, −0.48) <0.001 Viet Nam 235.08 (132.50, 385.52) 204.04 (112.69, 339.80) −0.33 (−0.40, −0.25) <0.001 924.93 (571.71, 1414.83) 805.89 (478.15, 1244.56) −0.31 (−0.41, −0.21) <0.001 87.03 (47.70, 150.33) 76.02 (40.54, 128.86) −0.29 (−0.39, −0.19) <0.001 Mongolia 253.60 (145.60, 411.56) 217.62 (126.02, 348.27) −0.03 (−0.25, 0.18) 0.772 1080.22 (664.27, 1652.84) 922.15 (553.16, 1409.08) 0.01 (−0.23, 0.25) 0.924 101.13 (52.27, 175.83) 86.64 (46.27, 146.75) 0.03 (−0.21, 0.28) 0.797 Low-middle SDI Nauru 317.67 (177.09, 532.58) 256.86 (144.02, 423.50) −0.64 (−0.66, −0.62) <0.001 1299.44 (768.88, 2010.71) 1034.26 (616.67, 1560.38) −0.71 (−0.72, −0.69) <0.001 121.72 (64.79, 210.06) 96.87 (48.45, 166.97) −0.70 (−0.72, −0.69) <0.001 Kiribati 316.81 (176.86, 517.22) 257.61 (143.26, 424.58) −0.64 (−0.65, −0.63) <0.001 1304.58 (783.44, 1944.93) 1041.70 (615.74, 1557.32) −0.69 (−0.71, −0.68) <0.001 121.71 (63.92, 204.03) 96.86 (48.86, 165.65) −0.70 (−0.71, −0.68) <0.001 Marshall Islands 300.24 (165.95, 499.27) 226.78 (124.98, 379.88) −0.88 (−0.90, −0.85) <0.001 1209.47 (727.99, 1861.60) 890.18 (534.55, 1349.73) −0.95 (−0.98, −0.93) <0.001 113.35 (59.80, 191.38) 83.35 (44.04, 146.87) −0.94 (−0.96, −0.91) <0.001 Micronesia (Federated States of) 310.50 (170.56, 508.11) 234.19 (130.55, 388.91) −0.82 (−0.88, −0.76) <0.001 1257.05 (741.89, 1916.10) 918.44 (551.31, 1367.94) −0.92 (−0.98, −0.86) <0.001 118.18 (62.57, 199.02) 86.52 (46.52, 152.02) −0.92 (−0.98, −0.85) <0.001 Samoa 335.41 (185.54, 557.76) 291.87 (160.62, 487.39) −0.34 (−0.37, −0.30) <0.001 1386.62 (841.71, 2110.97) 1227.24 (725.07, 1812.78) −0.30 (−0.33, −0.27) <0.001 130.78 (68.02, 218.39) 115.24 (58.99, 197.91) −0.31 (−0.34, −0.28) <0.001 Tuvalu 331.64 (186.90, 549.92) 247.99 (135.69, 417.22) −0.87 (−0.94, −0.79) <0.001 1359.70 (816.47, 2084.33) 988.74 (582.79, 1514.91) −0.94 (−1.02, −0.85) <0.001 127.60 (66.85, 212.17) 92.39 (47.76, 157.50) −0.93 (−1.02, −0.84) <0.001 Low SDI Cambodia 314.46 (176.95, 527.69) 218.47 (121.85, 366.03) −1.03 (−1.14, −0.92) <0.001 1327.98 (786.26, 2071.03) 883.97 (533.05, 1378.00) −1.17 (−1.30, −1.03) <0.001 122.87 (62.56, 214.76) 83.23 (44.00, 144.60) −1.12 (−1.25, −0.98) <0.001 Lao People’s Democratic Republic 324.88 (185.20, 535.28) 219.50 (122.54, 369.70) −1.32 (−1.44, −1.21) <0.001 1371.05 (820.46, 2128.52) 896.06 (524.80, 1409.56) −1.50 (−1.66, −1.34) <0.001 128.27 (65.39, 218.95) 84.65 (43.46, 147.28) −1.49 (−1.65, −1.33) <0.001 Papua New Guinea 320.62 (179.48, 531.47) 281.76 (156.79, 461.79) −0.40 (−0.42, −0.38) <0.001 1318.01 (805.86, 1962.36) 1157.09 (715.28, 1752.80) −0.40 (−0.42, −0.37) <0.001 123.37 (64.72, 202.21) 108.20 (57.05, 191.17) −0.40 (−0.43, −0.36) <0.001 Solomon Islands 372.48 (208.01, 635.34) 275.54 (152.20, 459.41) −0.96 (−1.01, −0.92) <0.001 1611.37 (965.47, 2432.78) 1137.47 (681.18, 1743.70) −1.14 (−1.21, −1.08) <0.001 151.86 (78.52, 262.03) 106.92 (55.19, 185.78) −1.14 (−1.20, −1.08) <0.001 Vanuatu 336.37 (189.26, 555.54) 247.72 (138.97, 414.84) −0.99 (−1.01, −0.97) <0.001 1396.71 (836.03, 2114.48) 992.17 (621.34, 1529.08) −1.10 (−1.13, −1.07) <0.001 131.78 (70.57, 222.00) 93.80 (51.22, 161.81) −1.09 (−1.11, −1.06) <0.001 Notes : ASRs were calculated for ages 15–29 years using GBD world standard population weights for ages 15–19, 20–24, and 25–29 years. Abbreviations : ASR, age-standardized rate; EAPC, estimated annual percentage change; UI, uncertainty interval; CI, confidence interval; SDI, Socio-demographic Index; YLDs, years lived with disability. Figure 1 Geographical distribution of endometriosis burden among females aged 15–29 years across 31 Western Pacific Region in 1990 and 2023. ( A ) 1990 ASPR; ( B ) 2023 ASPR; ( C ) 1990 age-standardized YLDs rate; ( D ) 2023 age-standardized YLDs rate. ASPR, age-standardized prevalence rate; YLDs, years lived with disability. The color gradient reflects the age-standardized rates per 100,000 population, with darker shades of red indicating higher rates. A) Map of Western Pacific in 1990 showing age-standardized prevalence rate (ASPR) per 100,000. Higher rates are concentrated in Australia, New Zealand and Papua New Guinea. B) Map of Western Pacific in 2023 showing ASPR per 100,000. New Zealand shows the highest rates, followed by Australia and Papua New Guinea. C) Map of Western Pacific in 1990 showing age-standardized years lived with disability (YLDs) rate per 100,000. Higher rates are seen in Australia and New Zealand. D) Map of Western Pacific in 2023 showing YLDs rate per 100,000. New Zealand has the highest rates, with significant rates in Australia and Papua New Guinea. The color gradient indicates higher rates with darker shades. Four maps of Western Pacific show endometriosis burden in 1990 and 2023 by ASPR and YLDs rates. Trends for Endometriosis Burden Among Females Aged 15-29 years in 31 Western Pacific Region by Country/Area and SDI Category, 1990–2023 Notes : ASRs were calculated for ages 15–29 years using GBD world standard population weights for ages 15–19, 20–24, and 25–29 years. Abbreviations : ASR, age-standardized rate; EAPC, estimated annual percentage change; UI, uncertainty interval; CI, confidence interval; SDI, Socio-demographic Index; YLDs, years lived with disability. Geographical distribution of endometriosis burden among females aged 15–29 years across 31 Western Pacific Region in 1990 and 2023. ( A ) 1990 ASPR; ( B ) 2023 ASPR; ( C ) 1990 age-standardized YLDs rate; ( D ) 2023 age-standardized YLDs rate. ASPR, age-standardized prevalence rate; YLDs, years lived with disability. The color gradient reflects the age-standardized rates per 100,000 population, with darker shades of red indicating higher rates. Absolute numbers showed a different distribution. China had the largest absolute burden throughout the study period. In 2023, it accounted for 150,427 incident cases, 567,881 prevalent cases, and 53,570 YLDs number, representing 44.7%, 42.6%, and 42.7% of the GBD WPR, respectively ( Supplementary Table 1 and Supplementary Figure 2 ). New Zealand, despite an ASPR approximately three times China’s, had 7847 prevalent cases. The Philippines, Japan, Viet Nam, the Republic of Korea, and Malaysia were also major contributors to absolute burden. Long-term changes in age-standardized rates also differed across the 31 countries and areas ( Table 1 and Supplementary Figure 3 ). Most countries and areas had negative EAPCs for age-standardized incidence, prevalence, and YLDs from 1990 to 2023, although the magnitude varied. The largest decreases were observed in the Lao People’s Democratic Republic, with EAPCs of −1.32% (95% CI:-1.44 to −1.21) for ASIR, −1.50% (95% CI: −1.66 to −1.34) for ASPR, and −1.49% (95% CI: −1.65 to −1.33) for the age-standardized YLDs rate. China also showed declining rates, with corresponding EAPCs of −1.12% (95% CI: −1.32 to −0.92), −0.98% (95% CI: −1.14 to −0.81), and −0.97% (95% CI: −1.13 to −0.80). Mongolia showed no statistically significant change in age-standardized rates. New Zealand had the highest ASPR and age-standardized YLDs rate point estimates in 2023, but the EAPCs for these two indicators were not statistically significant ( Table 1 ). Between 1990 and 2023, age-standardized rates among females aged 15–29 years decreased overall in both China and the WPR, although the trajectories were non-linear ( Figure 2 , Supplementary Figure 4 and Figure 3 ). In China, the ASIR per 100,000 population declined from 169.09 (95% UI: 88.00 to 287.47) in 1990 to 140.75 (95% UI: 74.47 to 235.63) in 2023, a reduction of 16.8%. The ASPR decreased from 624.98 (95% UI: 361.98 to 958.66) to 524.20 (95% UI: 306.90 to 798.92), and the age-standardized YLDs rate decreased from 58.74 (95% UI: 30.83 to 97.49) to 49.46 (95% UI: 26.14 to 82.16), corresponding to reductions of 16.1% and 15.8%, respectively. Figure 2 Trends in ASPR and age-standardized YLDs rate of endometriosis among females aged 15–29 years and all-ages in China and the Western Pacific Region, 1990–2023. ( A ) ASPR and ( B ) age-standardized YLDs rate among females aged 15–29 years. ( C ) ASPR and ( D ) age-standardized YLDs rate among all-ages female population. Red lines denote the estimates for China, and blue lines denote the estimates for the Western Pacific Region. Shaded regions depict the 95% uncertainty intervals. The multi-line graphs show trends in ASPR and age-standardized YLDs rates per 100,000 for China and the Western Pacific Region from 1990 to 2023. The x-axis represents years and the y-axis represents rates per 100,000. In each graph, the red line represents China and the blue line represents the Western Pacific Region. Image A shows ASPR rates, with China starting around 625 and decreasing to about 524, while the Western Pacific starts near 750 and ends around 700. Image B shows YLDs rates, with China decreasing from approximately 59 to 49 and the Western Pacific from about 75 to 65. Image C and D follow similar patterns. Notable trends include a general decline in rates for both regions, with China consistently lower than the Western Pacific. Peaks and troughs are visible around 2000 and 2010, indicating fluctuations in rates. Shaded areas depict 95 percent uncertainty intervals. Multi-line graphs compare ASPR and YLDs rates for China and Western Pacific from 1990 to 2023. Figure 3 Joinpoint analysis of endometriosis burden among females aged 15–29 years in the Western Pacific Region and China from 1990 to 2023. Joinpoint trends for ASPR ( A ) ASIR ( B ) and age-standardized YLDs rate ( C ) in China. ( D – F ) show the corresponding trends for Western Pacific Region. APC is shown for each fitted segment. Rates are expressed per 100,000. * indicates that the annual percent change (APC) is significantly different from zero at the alpha = 0.05 level. APC, annual percentage change. The multi-line graphs display trends in ASPR, ASIR and YLDs for females aged 15 to 29 years from 1990 to 2023. All panels share the x-axis labeled ′year′ and y-axis labeled with respective metrics per 100,000. Observed data points are marked, with fitted joinpoint segments showing annual percent change (APC). Panel A: ASPR in China. Observed decline from 620 to 445, then rise to 520. Joinpoints: 1990 to 2005 APC -0.6 percent, 2005 to 2010 APC -3.5 percent, 2010 to 2023 APC 1.5 percent. Panel B: ASIR in China. Observed decline from 165 to 121, then rise to 140. Joinpoints: 1990 to 2005 APC -0.7 percent, 2005 to 2010 APC -4.0 percent, 2010 to 2023 APC 1.3 percent. Panel C: YLDs in China. Observed decline from 54 to 42, then rise to 49. Joinpoints: 1990 to 2005 APC -0.5 percent, 2005 to 2010 APC -3.8 percent, 2010 to 2023 APC 1.8 percent. Panel D: ASPR in Western Pacific. Observed decline from 740 to 610, then rise to 675. Joinpoints: 1990 to 2005 APC -0.5 percent, 2005 to 2010 APC -3.2 percent, 2010 to 2023 APC 1.2 percent. Panel E: ASIR in Western Pacific. Observed decline from 192 to 155, then rise to 173. Joinpoints: 1990 to 2005 APC -0.6 percent, 2005 to 2010 APC -3.1 percent, 2010 to 2023 APC 1.0 percent. Panel F: YLDs in Western Pacific. Observed decline from 70 to 59, then rise to 65. Joinpoints: 1990 to 2005 APC -0.4 percent, 2005 to 2010 APC -2.9 percent, 2010 to 2023 APC 1.2 percent. Observed points are marked with dots and fitted segments are shown with lines. The graphs highlight the decline and subsequent rise in metrics, with varying APCs across regions and metrics. Graphs of ASPR, ASIR, YLDs trends for females 15 to 29 in China & Western Pacific, 1990 to 2023. Trends in ASPR and age-standardized YLDs rate of endometriosis among females aged 15–29 years and all-ages in China and the Western Pacific Region, 1990–2023. ( A ) ASPR and ( B ) age-standardized YLDs rate among females aged 15–29 years. ( C ) ASPR and ( D ) age-standardized YLDs rate among all-ages female population. Red lines denote the estimates for China, and blue lines denote the estimates for the Western Pacific Region. Shaded regions depict the 95% uncertainty intervals. Joinpoint analysis of endometriosis burden among females aged 15–29 years in the Western Pacific Region and China from 1990 to 2023. Joinpoint trends for ASPR ( A ) ASIR ( B ) and age-standardized YLDs rate ( C ) in China. ( D – F ) show the corresponding trends for Western Pacific Region. APC is shown for each fitted segment. Rates are expressed per 100,000. * indicates that the annual percent change (APC) is significantly different from zero at the alpha = 0.05 level. APC, annual percentage change. The GBD WPR showed smaller reductions. The ASIR per 100,000 population decreased from 194.49 (95% UI: 103.95 to 325.39) in 1990 to 174.49 (95% UI: 93.76 to 289.79) in 2023, a reduction of 10.3%. The ASPR decreased from 744.96 (95% UI: 440.51 to 1124. 88) to 679.56 (95% UI: 403.49 to 1016.33), and the age-standardized YLDs rate decreased from 69.86 (95% UI: 38.06 to 116.61) to 63.91 (95% UI: 35.45 to 106.15), reductions of 8.8% and 8.5%, respectively ( Figure 2 ). Joinpoint provided a segmented description of these changes ( Figure 3 and Supplementary Table 2 ). In China, negative AAPCs were found in ASIR [−0.57% (95% CI: −0.95 to −0.19; P = 0.004)], ASPR [−0.54% (95% CI: −0.89 to −0.19; P = 0.002)], and age-standardized YLDs rate [−0.52% (95% CI: −0.89 to −0.16; P = 0.005)] over the full period. ASIR declined gradually from 1990 to 2006 [APC: −0.64% (95% CI: −0.79 to −0.49; P<0.001)], followed by a sharper decrease from 2006 to 2010 [APC: −5.77% (95% CI: −7.55 to −3.97; P<0.001)]. It then increased from 2010 to 2015 [APC: 1.56% (95% CI: 0.36 to 2.78; P = 0.013)], declined from 2015 to 2021 [APC: −1.38% (95% CI: −2.21 to −0.54; P = 0.003)], and increased in the 2021–2023 segment [APC: 8.16% (95% CI: 4.10 to 12.38; P<0.001)]. ASPR and the age-standardized YLDs rate followed the same pattern. The long-term AAPCs in WPR were −0.36% (95% CI: −0.57 to −0.14; P = 0.001) for ASIR, −0.32% (95% CI: −0.37 to −0.26; P < 0.001) for ASPR, and −0.31% (95% CI: −0.36 to −0.26; P < 0.001) for the age-standardized YLDs rate. ASIR declined from 1990 to 2006 [APC: −0.54% (95% CI: −0.62 to −0.47; P<0.001)] and more rapidly from 2006 to 2009 [APC: −4.05% (−5.93 to −2.13; P<0.001)], before increasing from 2009 to 2021 [APC: 0.27% (95% CI: 0.14 to 0.40; P<0.001)] and during 2021–2023 [APC: 3.09% (1.08 to 5.15; P = 0.004)]. ASPR declined from 1990 to 1994 [APC: −1.15% (95% CI: −1.35 to −0.95; P<0.001)], 1994 to 2005 [APC: −0.21% (95% CI: −0.27 to −0.16; P<0.001)], and 2005 to 2010 [APC: −2.64% (95% CI: −2.83 to −2.44; P<0.001)], then increased from 2010 to 2021 [APC: 0.37% (95% CI: 0.33 to 0.42; P<0.001)] and 2021 to 2023 [APC: 2.94% (95% CI: 2.29 to 3.60; P<0.001)]. The age-standardized YLDs rate followed a similar sequence. In the sensitivity analysis with a stricter endpoint constraint, all six full-period AAPC estimates remained negative. However, the 95% CIs included zero for incidence in China [AAPC: −0.83% (95% CI: −1.69 to 0.03; P = 0.059) and YLDs in China [AAPC: −0.72% (95% CI: −1.44 to 0.01; P = 0.052)]. The intervals remained below zero for prevalence in China and all three outcomes in the WPR ( Supplementary Table 2 ). The relative gaps in ASR between China and the WPR widened between 1990 and 2023 among females aged 15–29 years. In 1990, China’s ASIR, ASPR, and the age-standardized YLDs rate were 13.1%, 16.1%, and 15.9% lower than those of the GBD WPR; by 2023, the corresponding differences were 19.3%, 22.9%, and 22.6%. Among females of all ages, these three indicators also declined in China and the WPR. All-age ASIR decreased by 32.9% in China and 26.0% in the WPR between 1990 and 2023. ASPR and the age-standardized YLDs rate showed the same overall direction of change ( Figure 2 and Supplementary Figure 4 ). Cross-sectional age-specific analyses in 2023 revealed a steep rise in endometriosis burden from adolescence into young adulthood in both China and the WPR ( Figure 4 ). In China, age-specific incidence rate per 100,000 population rose from 73.73 (95% UI: 33.37 to 128.20) at ages 15–19 years to 174.14 (95% UI: 104.30 to 263.54) at ages 20–24 years and 179.61 (95% UI: 88.67 to 324.26) at ages 25–29 years. Age-specific prevalence rate increased from 91.52 (95% UI: 43.45 to 160.43) to 572.38 (95% UI: 306.61to 916.01) and 947.13 (95% UI: 594.85 to 1375.77) across the same age groups, and the age-specific YLDs rate increased from 8.73 (95% UI: 4.05 to 16.53) to 54.18 (95% UI: 27.09 to 93.14) and 89.07 (95% UI: 49.30 to 142.53) per 100,000. Figure 4 Age-specific burden of endometriosis in China and the Western Pacific Region in 2023. ( A ) Incidence, ( B ) Prevalence, and ( C ) YLDs rate among females aged 10–39 years in China, and the corresponding measures in the GBD Western Pacific Region ( D – F ) in 2023. Bars indicate numbers; lines indicate rates per 100,000 population. The image A showing a bar and line graph. The x-axis label is, Age group (years), with categories 10 to 14, 15 to 19, 20 to 24, 25 to 29, 30 to 34, 35 to 39. The left y-axis label is, Number of incidence cases (thousands), ranging from 0.0 to 75.0. The right y-axis label is, Rate per 100,000, ranging from 0 to 200. Bars by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 25, 20 to 24 equals about 55, 25 to 29 equals about 65, 30 to 34 equals about 85, 35 to 39 equals about 85. Line points by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 80, 20 to 24 equals about 175, 25 to 29 equals about 180, 30 to 34 equals about 175, 35 to 39 equals about 160. The image B showing a bar and line graph. The x-axis label is, Age group (years), with categories 10 to 14, 15 to 19, 20 to 24, 25 to 29, 30 to 34, 35 to 39. The left y-axis label is, Number of prevalence cases (thousands), ranging from 0.0 to 600.0. The right y-axis label is, Rate per 100,000, ranging from 0 to 1000. Bars by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 50, 20 to 24 equals about 200, 25 to 29 equals about 350, 30 to 34 equals about 600, 35 to 39 equals about 650. Line points by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 100, 20 to 24 equals about 500, 25 to 29 equals about 800, 30 to 34 equals about 1000, 35 to 39 equals about 1050. The image C showing a bar and line graph. The x-axis label is, Age group (years), with categories 10 to 14, 15 to 19, 20 to 24, 25 to 29, 30 to 34, 35 to 39. The left y-axis label is, Number of YLDs (thousands), ranging from 0.0 to 60.0. The right y-axis label is, Rate per 100,000, ranging from 0 to 125. Bars by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 5, 20 to 24 equals about 20, 25 to 29 equals about 32, 30 to 34 equals about 55, 35 to 39 equals about 60. Line points by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 10, 20 to 24 equals about 60, 25 to 29 equals about 95, 30 to 34 equals about 110, 35 to 39 equals about 115. The image D showing a bar and line graph. The x-axis label is, Age group (years), with categories 10 to 14, 15 to 19, 20 to 24, 25 to 29, 30 to 34, 35 to 39. The left y-axis label is, Number of incidence cases (thousands), ranging from 0.0 to 125.0. The right y-axis label is, Rate per 100,000, ranging from 0 to 200. Bars by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 50, 20 to 24 equals about 110, 25 to 29 equals about 100, 30 to 34 equals about 120, 35 to 39 equals about 120. Line points by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 90, 20 to 24 equals about 220, 25 to 29 equals about 200, 30 to 34 equals about 180, 35 to 39 equals about 170. The image E showing a bar and line graph. The x-axis label is, Age group (years), with categories 10 to 14, 15 to 19, 20 to 24, 25 to 29, 30 to 34, 35 to 39. The left y-axis label is, Number of prevalence cases (thousands), ranging from 0.0 to 750.0. The right y-axis label is, Rate per 100,000, ranging from 0 to 1000. Bars by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 75, 20 to 24 equals about 350, 25 to 29 equals about 600, 30 to 34 equals about 850, 35 to 39 equals about 900. Line points by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 100, 20 to 24 equals about 650, 25 to 29 equals about 1000, 30 to 34 equals about 1100, 35 to 39 equals about 1120. The image F showing a bar and line graph. The x-axis label is, Age group (years), with categories 10 to 14, 15 to 19, 20 to 24, 25 to 29, 30 to 34, 35 to 39. The left y-axis label is, Number of YLDs (thousands), ranging from 0.0 to 75.0. The right y-axis label is, Rate per 100,000, ranging from 0 to 120. Bars by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 8, 20 to 24 equals about 35, 25 to 29 equals about 55, 30 to 34 equals about 80, 35 to 39 equals about 85. Line points by age group are approximately: 10 to 14 equals 0, 15 to 19 equals about 10, 20 to 24 equals about 70, 25 to 29 equals about 105, 30 to 34 equals about 115, 35 to 39 equals about 115. A set of six bar and line graphs showing endometriosis incidence, prevalence and YLDs by age group. Age-specific burden of endometriosis in China and the Western Pacific Region in 2023. ( A ) Incidence, ( B ) Prevalence, and ( C ) YLDs rate among females aged 10–39 years in China, and the corresponding measures in the GBD Western Pacific Region ( D – F ) in 2023. Bars indicate numbers; lines indicate rates per 100,000 population. A comparable age pattern was observed in the WPR. The incidence rate was highest at ages 20–24 years in the WPR, whereas the estimates for ages 20–24 and 25–29 years were similar in China. Prevalence and YLDs rate continued to increase through ages 25–29 years in both settings. Age-specific incidence rate per 100,000 population in WPR rose from 103.41 (95% UI: 52.34 to 176.13) at ages 15–19 years. Its point estimate was highest at ages 20–24 years (226.34 [95% UI:131.71 to 344.16]) and remained high at ages 25–29 years (198.80 [95% UI:99.97 to 358.02]). Age-specific prevalence rate increased from 129.24 (95% UI:69.03 to 220.08) to 781.45 (95% UI:426.78 to 1227.06) and 1175.74 (95% UI:744.76 to 1669.22), and age-specific YLDs rate increased from 12.24 (95% UI:6.08 to 22.40) to 73.74 (95% UI:40.20 to 129.82) and 110.22 (95% UI:62.65 to 173.28) per 100,000 ( Figure 4 ). Absolute numbers followed a similar age-related pattern, particularly for prevalence number and YLDs number. Although the magnitude and ranking of burden differed by SDI group, the age-related increase was also evident across SDI categories ( Supplementary Figure 5 ). Given this age-specific heterogeneity, we next examined whether endometriosis burden varied by socioeconomic development within the region. SDI-stratified analyses indicated that endometriosis burden differed by age group ( Supplementary Figure 5 ). At ages 15–19 years, low-middle-SDI areas had the highest incidence, prevalence, and YLDs rate, at 162.08, 194.93, and 18.55 per 100,000 population, respectively, whereas high-SDI areas had the lowest corresponding rates, at 89.15, 114.60, and 10.89 per 100,000 in 2023. These differences persisted at ages 20–24 years. Disease burden in low-middle-SDI settings was approximately 1.8 times that in high-SDI settings (Incidence, 348.83 vs 194.15 per 100,000; Prevalence, 1233.57 vs 666.83 per 100,000; YLDs, 115.80 vs 62.93 per 100,000). At ages 25–29 years, low-middle-SDI settings continued to have the highest rates, with incidence, prevalence, and YLDs rate of 284.67, 1862.30, and 173.93 per 100,000, respectively. The corresponding rates in high-SDI settings remained the lowest, at 180.82, 1029.02, and 96.52 per 100,000. Across all five SDI categories, prevalence and YLDs rate increased from ages 15–19 to 25–29 years, whereas incidence peaked at ages 20–24 years. The high-middle-SDI estimates represented Fiji alone. Across country-year observations, SDI was negatively correlated with ASIR (ρ = −0.53), ASPR (ρ = −0.43), and the age-standardized YLDs rate (ρ = −0.43) ( Figure 5 and Supplementary Figure 6 ). P values were <0.001 for all three measures, but annual observations within countries were not independent. We based all statistical inference on a single observation per country. In the 2023 cross-section of 31 countries and areas, the coefficients were −0.53 (P = 0.002) for ASIR, −0.41 (P = 0.023) for ASPR, and −0.42 (P = 0.019) for the age-standardized YLDs rate. Analyses using each country’s mean value over 1990–2023 showed associations in the same direction ( Supplementary Table 3 ). The LOESS fits suggested non-linear relationships. ASIR generally decreased as SDI increased, then flattened at higher SDI values. ASPR and the age-standardized YLDs rate reached their lowest fitted values at an SDI of approximately 0.69 before rising modestly. China lay below the fitted prevalence and YLDs curves at its observed SDI value. Figure 5 Trends between SDI and age-standardized YLDs rate among females aged 15–29 years in 31 Western Pacific countries and areas from 1990 to 2023. YLDs, years lived with disability; SDI, Socio-demographic index. Dark curves indicate loess fits with 95% confidence intervals. A scatter plot with many connected point sequences and a smooth fitted curve. The x-axis label is Socio-demographic Index (SDI), ranging from 0.3 to 0.9 with labeled ticks at 0.4, 0.6 and 0.8. The y-axis label is age-standardized YLDs rate (per 100,000), ranging from 40 to 160 with labeled ticks at 40, 80, 120 and 160. Text near the upper left reads rho equals minus 0.43, P less than 0.001. A dark smooth curve with a shaded band declines from about 125 at SDI 0.3 to about 85 near SDI 0.65 to 0.7, then rises slightly to about 90 to 95 by SDI 0.9. Numerous country trajectories form downward sloping and looping paths between roughly 70 and 140 across SDI 0.35 to 0.85. One labeled point, China, appears near SDI about 0.7 and y about 50. A legend titled Location lists many country and area names with matching markers. Scatter plot: Socio-demographic index vs age-standardized disability rate, curving down then up. Trends between SDI and age-standardized YLDs rate among females aged 15–29 years in 31 Western Pacific countries and areas from 1990 to 2023. YLDs, years lived with disability; SDI, Socio-demographic index. Dark curves indicate loess fits with 95% confidence intervals. Decomposition analysis showed that reductions in the absolute burden of endometriosis among females aged 15–29 years between 1990 and 2023 were attributable mainly to population change, with epidemiological change making a smaller contribution ( Figure 6 and Supplementary Table 4 ). In China, incident cases decreased by 155.2 thousand, with contributions of −113.7 thousand from population, −40.8 thousand from epidemiological change, and −0.8 thousand from aging. Prevalent cases decreased by 546.2 thousand; population and epidemiological change contributed −421.3 thousand and −144.7 thousand, respectively, while aging offset part of the reduction (+19.9 thousand). YLDs decreased by 51.2 thousand, with contributions of −39.7 thousand from population and −13.3 thousand from epidemiological change, partly offset by aging (+1.8 thousand). Figure 6 Decomposition of changes in endometriosis burden among females aged 15–29 years in China and the Western Pacific Region from 1990 to 2023. ( A ) Prevalence number; ( B ) Incidence number; ( C ) YLDs number values are expressed in thousands. Image A: Horizontal bar graph with X-axis from -150 to 0 (thousands) and Y-axis categories China and Western Pacific Region. Legend: Aging, Population, Epidemiological change. China: Population 0 to -100, Epidemiological change -100 to -150, Aging near 0. Western Pacific: Population 0 to -100, Epidemiological change -100 to -150, Aging near 0. Image B: X-axis from -600 to 0 (thousands). China: Population 0 to -500, Epidemiological change -500 to -600, Aging near 0. Western Pacific: Population 0 to -500, Epidemiological change -500 to -600, Aging larger near 0. Image C: X-axis from -60 to 0 (thousands). China: Population 0 to -50, Epidemiological change -50 to -60, Aging near 0. Western Pacific: Population 0 to -50, Epidemiological change -50 to -60, Aging larger near 0. Horizontal bar graph of endometriosis burden change in China and Western Pacific. Decomposition of changes in endometriosis burden among females aged 15–29 years in China and the Western Pacific Region from 1990 to 2023. ( A ) Prevalence number; ( B ) Incidence number; ( C ) YLDs number values are expressed in thousands. The GBD WPR showed a similar pattern. Incident cases decreased by 160.4 thousand, contributions of −115.9 thousand from population and −44.8 thousand from epidemiological change; aging made a minimal positive contribution (+0.3 thousand). Prevalent cases decreased by 542.5 thousand, reflecting contributions of −448.1 thousand from population and −146.5 thousand from epidemiological change, partly offset by aging (+52.1 thousand). YLDs decreased by 50.6 thousand, with contributions of −42.1 thousand from population and −13.3 thousand from epidemiological change, while aging offset the reduction by 4.8 thousand. Thus, population change accounted for the largest share of the decline in absolute burden, although epidemiological change also contributed.

Among females aged 15–29 years in China and the WPR, estimated endometriosis rates and absolute burden declined overall between 1990 and 2023, with fluctuations over time. In 2023, age-specific prevalence and YLDs rate increased from adolescence into young adulthood. Incidence rate estimates were highest at ages 20–24 years in the WPR and 25–29 years in China. Geographical and socioeconomic differences persisted: several Pacific settings had high standardized rates, while populous countries, particularly China, accounted for large case numbers. These patterns may inform local assessment of reproductive-health services, but declining burden estimates alone cannot establish a reduction in clinical need. Future studies linking population-level estimates with individual clinical and health system data are needed to distinguish variation in underlying disease occurrence from differences in recognition and diagnostic capture, and to guide reproductive health planning across the WPR.

Endometriosis is a leading cause of chronic pelvic pain and infertility, posing a substantial but often under-recognized burden on global healthcare systems. 16 Utilizing GBD 2023 data, we examined its burden among females aged 15–29 years in the WPR, and 31 WPR countries and areas including China from 1990 to 2023. By narrowing our focus to early reproductive life and jointly examining temporal, demographic, and socioeconomic patterns, we found that estimated age-standardized rates and absolute endometriosis burden declined overall among females aged 15–29 years in China and the WPR between 1990 and 2023, although the temporal patterns were non-linear. Substantial geographical differences remained, with New Zealand having the highest standardized rates and China the largest absolute burden. Decomposition analysis indicated that population change made the largest contribution to declining absolute burden, with epidemiological change also contributing. Within this overall pattern, the 2023 age-specific estimates showed a marked increase in prevalence and YLDs rate from ages 15–19 to 25–29 years. Incidence was highest at ages 20–24 years in the WPR, while China’s point estimate was slightly higher at ages 25–29 years. These findings distinguish long-term changes in estimated burden from cross-sectional differences between age groups. This research provides crucial insights distinct from previous GBD studies, allowing for a more profound understanding of endometriosis burden during early reproductive life in the WPR. The overall decline in age-standardized rates was broadly consistent with previous GBD studies. Zhang et al reported declining global ASIR, ASPR and age-standardized YLDs rate using GBD 2017 estimates across 195 countries and territories. 10 Shen et al confirmed similar declines through 2019 27 . Subsequent GBD 2021 analyses of women aged 15–49 years or broader age ranges generally reported the same long-term direction. 9 , 14 , 17 , 18 , 25–28 In the present study, Joinpoint analysis showed that these declines were not uniform. Periods of decrease alternated with periods of increase, including a short-fitted rise from 2021 to 2023. Similar non-linear patterns have been reported in other global analyses by Li et al and Wang et al 27 , 29 Greater awareness of menstrual symptoms and improved access to gynecological assessment may increase case ascertainment. Financial or geographical barriers may have the opposite effect. Changes in referral practices, insurance coverage, and diagnostic coding could also influence the data used in GBD models. These influences may coincide with changes in disease occurrence, but their effects may differ in direction. We did not measure when or to what extent these factors changed. The fitted joinpoints varied and our findings cannot attribute them to specific social or policy events. The sensitivity analysis retained negative full-period AAPC estimates, although the 95% CIs included zero for incidence and YLDs in China. Evidence for these two declines was less precise under the stricter endpoint constraint. The longer terminal segments showed increasing rates in the WPR, whereas 95% CIs included zero in China. The increases fitted over 2021–2023 should therefore be interpreted cautiously and do not establish a sustained reversal of the long-term pattern. The geographical analysis showed that age-standardized rates and absolute numbers describe different dimensions of regional burden. Previous global studies of women aged 15–49 years or broader female populations identified Oceania as an area with comparatively high age-standardized endometriosis burden. 9 , 14 , 18 , 30 Our analysis identified high rates in New Zealand and several Pacific Island settings within the WPR. China, by contrast, had comparatively low standardized rates but accounted for more than 42% of WPR prevalent cases and YLDs in 2023. High rates in smaller populations may indicate a substantial burden relative to population size, while large absolute numbers in populous settings indicate the potential scale of service demand. Regional averages can obscure both patterns. Both measures are relevant to service planning, but neither provides a direct estimate of healthcare use or unmet need. For service planning, these estimates need to be considered alongside evidence on symptoms, care-seeking, and access to care. The decomposition findings further explain why absolute burden declined in this population despite increases reported elsewhere. Several global analyses of women aged 15–49 years or broader modeled female populations reported increasing absolute endometriosis burden despite declining age-standardized rates. 14 , 16 , 17 , 27–29 Population growth was commonly identified as a major contributor to those increases. In contrast, burden measures decreased among females aged 15–29 years in both China and the WPR. Population made the largest negative contribution, consistent with demographic change in this younger female population. Epidemiological change also contributed to the reduction, in line with the negative EAPCs and AAPCs. Aging, which describes changes in age composition within the 15–29-year population, made a smaller contribution that varied in direction across outcomes. The contrast with global studies needs to be considered in light of differences in geographical scope and age coverage. The reduction in absolute burden reflected both demographic change and declining modeled age-specific rates; it cannot establish a proportional reduction in individual disease risk or clinical need. The age-specific findings place the 15–29-year period within the broader reproductive-age distribution of endometriosis. Xu et al analyzed women aged 15–49 years across 204 countries and territories and reported that incidence was highest at ages 20–24 years, whereas prevalence and Disability-Adjusted Life Years (DALYs) peaked at ages 25–29 years. 14 Yan et al examined global incidence and DALYs across GBD age groups from 10 to older than 55 years and reported similar peaks at ages 20–24 and 25–29 years, respectively. 9 Other GBD 2021 analyses of women aged 15–49 years or the full modeled age range reported similar age ordering. 25–27 , 30 Feng et al found that the age effect on incidence in China increased among women aged 15–24 years and declined thereafter. 31 Our findings extend this evidence by examining females aged 15–29 years specifically within the WPR. Prevalence and YLDs rate increased sharply from ages 15–19 to 20–29 years, while incidence rose rapidly and was highest at ages 20–24 or 25–29 years. This pattern supports treating adolescents and young women as a distinct analytical group. However, these were cross-sectional age comparisons, not observations of individuals as they grew older. They do not establish when disease began or demonstrate progression, diagnostic delay, or changing risk within individuals. The observed gradient may reflect differences in disease occurrence, symptom recognition, healthcare use, diagnostic practice, or data capture. The relationship between SDI and endometriosis burden was more complex. Previous studies differed in their age definitions and outcomes. Using GBD 2017 data across 195 countries and territories, Zhang et al found that all ASIR, ASPR and the age-standardized YLDs rate declined in most SDI quintiles but increased in the high-SDI quintile. 10 Xu et al examined women aged 15–49 years across 204 countries and territories and reported the highest ASIR, ASPR, and age-standardized DALYs rate in the low-SDI group in 2021 14 . Yan et al similarly reported high ASIR and age-standardized DALYs rate in low-SDI settings across a broader modeled age range. 9 Liu et al analyzed women aged 15–55 years across 204 countries and territories and found a greater age-standardized burden in lower-SDI settings. 28 In our analysis, higher SDI was associated with lower burden, although the LOESS curves suggested non-linearity. In our age-specific analysis, the low-middle-SDI group had the highest age-specific rates and the high-SDI group the lowest at ages 15–19, 20–24, and 25–29 years. These findings are not directly comparable with previous reports of endometriosis burden across broader reproductive-age populations. Differences in geographical coverage and SDI classification should also be considered. Limited access to assessment and treatment may contribute to persistent symptoms among young women in lower-SDI settings. Delayed recognition and interrupted care are possible explanations for the higher prevalence and YLDs rate. Restricted diagnostic access can reduce ascertainment as well as delay care. Higher YLDs rate are also not independent evidence of disease progression, because YLDs estimates depend on modeled prevalence and disability weights. In addition, the high-middle-SDI estimates represent Fiji alone and should not be generalized to other settings in that category. China lay below the fitted LOESS curves, particularly for prevalence and YLDs. This may reflect differences in disease occurrence as well as access to care, diagnostic practice, and data capture. This pattern alone cannot determine whether endometriosis is underdiagnosed. The findings have implications for reproductive health services. Persistent dysmenorrhea, pelvic pain, and symptoms that interfere with education, employment, or daily activities warrant appropriate clinical assessment. 1 , 2 , 4 , 5 Service pathways may include symptom assessment, appropriate empirical management, follow-up of persistent symptoms, and specialist referral when indicated. In Pacific Island settings, small and geographically dispersed populations and limited local specialist capacity may require referral arrangements that include regional or overseas care. 32 The organization of these pathways should reflect local clinical need and health-system capacity, neither of which can be measured directly from GBD estimates. In China, the large absolute burden provides a reason to consider how existing services accommodate adolescents and young women with persistent menstrual and pelvic pain. Within a tiered healthcare system, primary-care services could provide initial assessment and follow-up, with referral to gynecological services for persistent or complex presentations. School and university health services could help young women reach appropriate care, while communication between specialist and local providers could support follow-up. Referral waiting times, symptom-related functional impairment, and patient-reported experience could be used to evaluate such arrangements. These proposals require local service data and their effectiveness cannot be inferred from our burden estimates. This study covered more than three decades, focused on an age range often incorporated into broader reproductive-age estimates, and considered standardized rates alongside absolute burden. The combined analysis distinguished temporal, geographical, socioeconomic, and demographic patterns. However, several limitations should be considered. First, GBD estimates depend on the availability, comparability, and structure of underlying data. Symptom recognition, access to assessment, diagnostic practice, and coding influence the burden captured by those sources. Although GBD modeling extends beyond recorded cases, it may not fully capture unrecognized disease. Geographical differences may therefore reflect both variation in disease occurrence and ascertainment. These estimates cannot establish the true biological incidence of endometriosis or distinguish lesion phenotypes. Second, SDI analyses were ecological. Country-level sensitivity analyses addressed repeated observations within countries, but residual confounding and ecological fallacy remain possible. Third, uncertainty for the derived 15–29-year ASR was approximated from age-specific bounds because posterior draws were unavailable. Full GBD uncertainty was not propagated through pooled SDI-specific rates or the decomposition components. Fourth, China is part of the WPR, so comparisons between China and the WPR were descriptive rather than statistically independent. Finally, the data could not assess symptom onset, diagnostic delay, treatment, recurrence, infertility, quality of life, or clinical phenotype. Future studies should combine population-level burden estimates with individual clinical and health system data. Longitudinal data could clarify the timing of symptom onset, diagnostic pathways, treatment, and patient-reported outcomes among adolescents and young women. Linkage with claims data, hospital records, and reproductive health surveys may help distinguish changes in disease occurrence from changes in recognition and diagnostic capture. Country-level service studies are also needed to examine why estimates differ among locations with similar SDI values. Such studies would provide a firmer basis for assessing whether available care meets the needs of young women in both high-rate and high-number settings.

Absolute burden among females aged 15–29 years was calculated by summing numbers for ages 15–19, 20–24, and 25–29 years. To compare burden of populations with different age structures over time, we calculated the age-standardized rates (ASR) with corresponding 95% UIs in 15–29 years old based on the world standard population reported in the GBD. 23 The ASR, which accounts for incidence, prevalence, and YLDs, was calculated as: \documentclass[12pt]{minimal} \usepackage{wasysym} \usepackage[substack]{amsmath} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage[mathscr]{eucal} \usepackage{mathrsfs} \DeclareFontFamily{T1}{linotext}{} \DeclareFontShape{T1}{linotext}{m}{n} {linotext }{} \DeclareSymbolFont{linotext}{T1}{linotext}{m}{n} \DeclareSymbolFontAlphabet{\mathLINOTEXT}{linotext} \begin{document}$${{\rm ASR = }}{{\sum _{{{\rm i = 1}}}^{{\rm 3}}\,{{{\rm a}}_{{\rm i}}}{{{\rm w}}_{{\rm i}}}} \over {\sum _{{{\rm i = 1}}}^{{\rm 3}}\,{{{\rm w}}_{{\rm i}}}}}{{\rm \times 100,000}}$$\end{document} where a i is the age-specific rate and w i is the corresponding standard-population weight. Rates are expressed per 100,000 population. The GBD standard-population weights for ages 15–19, 20–24, and 25–29 years were renormalized to sum to one within the study age range (0.350021, 0.329419, and 0.320560, respectively). Because draw-level estimates were unavailable, bounds for derived 15–29-year rates were calculated by applying these weights separately to the reported age-specific lower and upper bounds. These bounds approximate, but do not fully propagate, GBD uncertainty across age groups. Temporal trends in the age-standardized incidence rate (ASIR), age-standardized prevalence rate (ASPR), and age-standardized YLDs rate from 1990 to 2023 were assessed using the estimated annual percentage change (EAPC). For each location and indicator, a log-linear regression model was fitted: \documentclass[12pt]{minimal} \usepackage{wasysym} \usepackage[substack]{amsmath} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage[mathscr]{eucal} \usepackage{mathrsfs} \DeclareFontFamily{T1}{linotext}{} \DeclareFontShape{T1}{linotext}{m}{n} {linotext }{} \DeclareSymbolFont{linotext}{T1}{linotext}{m}{n} \DeclareSymbolFontAlphabet{\mathLINOTEXT}{linotext} \begin{document}${\mathrm{ln}}\left({{\mathrm{AS}}{{\mathrm{R}}_{\mathrm{y}}}} \right){\mathrm{ = \alpha + \beta\ y + }}\varepsilon $\end{document} . EAPC was calculated as: \documentclass[12pt]{minimal} \usepackage{wasysym} \usepackage[substack]{amsmath} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage[mathscr]{eucal} \usepackage{mathrsfs} \DeclareFontFamily{T1}{linotext}{} \DeclareFontShape{T1}{linotext}{m}{n} {linotext }{} \DeclareSymbolFont{linotext}{T1}{linotext}{m}{n} \DeclareSymbolFontAlphabet{\mathLINOTEXT}{linotext} \begin{document}${\mathrm{EAPC = 100 \times [exp(\beta) - 1]}}$\end{document} , and its 95% confidence interval (CI) was derived from the standard error of β. A trend was classified as increasing when the EAPC estimate and the lower CI bound exceeded zero, decreasing when the estimate and upper CI bound were below zero, and stable when the CI included zero. Segmented temporal trends were evaluated separately for China and the WPR using the Joinpoint Software (version 6.1.0; National Cancer Institute, USA). Log-linear models were fitted to annual ASRs using standard errors derived from the reported 95% UIs. Model selection used a Monte Carlo permutation test and a two-sided significance level of 0.05. Parametric 95% CIs were calculated. Annual percentage change (APC) was estimated for each segment, and average annual percentage change (AAPC) was calculated for 1990–2023 as follows: \documentclass[12pt]{minimal} \usepackage{wasysym} \usepackage[substack]{amsmath} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage[mathscr]{eucal} \usepackage{mathrsfs} \DeclareFontFamily{T1}{linotext}{} \DeclareFontShape{T1}{linotext}{m}{n} {linotext }{} \DeclareSymbolFont{linotext}{T1}{linotext}{m}{n} \DeclareSymbolFontAlphabet{\mathLINOTEXT}{linotext} \begin{document}$${{\rm AAPC = }}\left \{ {{{\rm exp}}\left({{{\mathop \sum \nolimits^ {{{\rm w}}_{{\rm i}}}{{{\rm b}}_{{\rm i}}}} \over {\mathop \sum \nolimits^ {{{\rm w}}_{{\rm i}}}}}} \right){{\rm - 1}}} \right\}{{\rm \times 100}}$$\end{document} where w i indicates the length of each segment in the range of years, while the b i is the slope coefficient for each segment in the desired range of years. An AAPC was classified as increasing when its estimate and lower 95% CI bound exceeded zero, decreasing when its estimate and upper bound were below zero, and stable when the CI included zero. CI for EAPC and Joinpoint estimates reflect uncertainty from the fitted regression models. EAPC and AAPC summarize the same period under different model structures. EAPC provided a log-linear summary of the 1990–2023 trend for comparisons across the 31 countries and areas. Joinpoint analysis was used to characterize changes in trend within China and the WPR, with APC describing individual segments and AAPC summarizing the fitted segments over the full period. Associations between SDI and endometriosis burden were examined across the 31 countries and areas. Annual SDI values were linked to ASIR, ASPR, and age-standardized YLDs rate from 1990 to 2023. Spearman correlation coefficients were used to describe the direction and strength of association, and locally estimated scatter plot smoothing (LOESS) with 95% confidence bands was used to characterize possible non-linearity. 11 Changes in absolute burden between 1990 and 2023 were decomposed into aging, population, and epidemiological change, respectively. Decomposition was conducted separately for China and the GBD WPR using the Das Gupta method. 24 Total burden was expressed as: \documentclass[12pt]{minimal} \usepackage{wasysym} \usepackage[substack]{amsmath} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage[mathscr]{eucal} \usepackage{mathrsfs} \DeclareFontFamily{T1}{linotext}{} \DeclareFontShape{T1}{linotext}{m}{n} {linotext }{} \DeclareSymbolFont{linotext}{T1}{linotext}{m}{n} \DeclareSymbolFontAlphabet{\mathLINOTEXT}{linotext} \begin{document}$${{\mathrm{B}}_{\mathrm{y}}}{\mathrm{ = }}{\sum _{\mathrm{i}}}\,{{\mathrm{P}}_{\mathrm{y}}}{{\mathrm{a}}_{{\mathrm{i,y}}}}{{\mathrm{r}}_{{\mathrm{i,y}}}}$$\end{document} where P y is the total female population aged 15–29 years, a i,y is the proportion in age group i, and r i,y is the corresponding age-specific rate in year y. Symmetric decomposition averaged contributions across all possible factor order permutations. Results were expressed in thousands. All statistical analyses were performed using R version 4.3.3. Data visualization was primarily created using the ggplot2. All statistical tests were two-sided, and P < 0.05 was considered statistically significant.