The global burden of depression in adolescents and young adults, 1990–2021: systematic analysis of the global burden of disease study

Abstract

Background

Depression is a leading cause of disability among adolescents and young adults, yet comprehensive assessments of its global burden and associated risk factors in this population remain limited. This study aimed to evaluate the global, regional, and national trends in depression burden among individuals under 30 years of age from 1990 to 2021, and to identify key risk factors and disparities.

Methods

We analyzed data from the Global Burden of Disease Study 2021, focusing on depression incidence and disability-adjusted life years (DALYs) among individuals aged under 30. Temporal trends were assessed using an age-period-cohort (APC) model. We further examined trends using estimated annual percentage change (EAPC), frontier analysis, and inequality metrics to provide a comprehensive evaluation.

Results

From 1990 to 2021, both the incidence of depression and related DALYs in individuals under 30 increased by over 50%. Males experienced a larger proportional increase, although females had an age-standardized incidence rate (ASIR) 23 times higher than males in 2021. The highest burden was observed in the 20–24 age group, while the fastest growth occurred among those aged 15–19. Individuals born after 2000 showed a higher prevalence of depression. High sociodemographic index (SDI) regions—particularly high-income North America and Greenland—had the highest ASIR and age-standardized DALY rate (ASDR). The most prominent risk factors were bullying victimization (61.8%), childhood sexual abuse (12.0%), and intimate partner violence (12.6%). Notable regional disparities were observed, with countries like Switzerland, Norway, Monaco, and Germany showing rapid increases in depression burden.

Conclusions

The burden of depression among young people has risen substantially over the past three decades, with disproportionate impacts on females, post-2000 birth cohorts, and those in high-SDI regions. Interventions targeting modifiable risk factors—particularly bullying, childhood abuse, and intimate partner violence—are essential to reduce the burden. Tailored, region-specific strategies are necessary to effectively address disparities and improve mental health outcomes in youth.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12888-025-07201-8.

Introduction

The rising prevalence of depression among adolescents and young adults presents a significant global public health challenge [1]. Over the past 15 years, mental health in this population has deteriorated significantly, with nearly half of females aged 16–24 reporting mental health disorders, particularly post-COVID-19 [2]. Approximately 75% of adolescents experience substantial anxiety or depression symptoms before age 18, and two-thirds face recurrent episodes [3]. In Germany, depression rates among young people have risen nearly 30% in five years, with a sharper increase in females [4]. Globally, between 2010 and 2019, the incidence, prevalence, and disability-adjusted life years (DALYs) of depression among individuals aged 10–24 steadily increased, disproportionately affecting adolescents [5]. These patterns highlight the critical necessity for focused prevention and intervention efforts.

The burden of major depressive disorder (MDD) in youth is shaped by social, familial, and individual risk factors. Adverse childhood experiences, including family violence, sexual abuse, and bullying, are key contributors [6]. Socioeconomic disparities, academic pressure, and societal expectations further exacerbate mental distress, while the COVID-19 pandemic has intensified these challenges through social isolation and disrupted education [7]. Biologically, genetic predisposition, hormonal changes, and neurotransmitter imbalances may heighten vulnerability in certain adolescent subgroups [8]. Limited social support and mental health stigma often delay diagnosis and treatment, while cultural and regional differences influence depression prevalence [9]. Low Sociodemographic Index (SDI) regions face higher risks due to restricted healthcare access, whereas high-SDI regions contend with greater psychological stress linked to social competition.

Depression in young people arises from complex interactions within broader social and biological systems [10]. Effective prevention and intervention require a multidimensional approach, addressing social stressors, enhancing mental health education, improving psychological care access, and facilitating early intervention for high-risk individuals [11].

Although modifiable risk factors play a crucial role, most existing research predominantly examines regional and gender disparities, with relatively little focus on long-term depression trends among young populations [12]. The Global Burden of Disease (GBD) Study offers a comprehensive framework for evaluating disease burden and associated risk factors. This study analyzes global, regional, and national trends in depression prevalence and DALYs among individuals under 30 from 1990 to 2021, utilizing data from GBD 2021. An age-period-cohort (APC) model will be employed to assess the impact of age, period, and birth cohort on depression burden, while differences by sex, age group, period, and region will also be examined. Findings will inform global mental health policies and support targeted interventions to reduce the depression burden and mitigate modifiable risk factors in adolescents and young adults.

Methods

Data sources

This study is a secondary analysis based on publicly available data from the Global Burden of Disease Study 2021 (GBD 2021), retrieved via the Global Health Data Exchange (GHDx) platform (http://ghdx.healthdata.org/gbd-2021). The GBD provides standardized estimates of disease burden for 204 countries and territories from 1990 to 2021.

We extracted data on depression in individuals aged ≤ 30 years. Depression was defined according to the International Classification of Diseases, 10th Revision (ICD-10), including major depressive disorder (MDD) and dysthymia (ICD-10 codes: F32–F33, F34.1) [13].

Estimates were generated using DisMod-MR 2.1, a Bayesian meta-regression tool developed by the Institute for Health Metrics and Evaluation (IHME). This model produces internally consistent estimates of incidence, prevalence, remission, and excess mortality across age, sex, geography, and time. Key input parameters included study-level covariates, age-specific data, and regional random effects. The model addresses variations across countries and years by applying hierarchical modeling that borrows strength from data-rich settings to inform estimates in data-scarce locations [14].To assess development level, the Sociodemographic Index (SDI) was used. SDI is a composite indicator of income per capita, educational attainment, and total fertility rate, and it classifies countries into five categories: low, low-middle, middle, high-middle, and high SDI [15].

Outcomes and measures

The primary outcomes were: Incidence of depression per 100,000 population, Disability-adjusted life years (DALYs), comprising years of life lost (YLLs) and years lived with disability (YLDs), Age-standardized incidence rate (ASIR), and Age-standardized DALY rate (ASDR).Temporal trends were evaluated using estimated annual percentage change (EAPC), calculated from log-linear regression models. An age-period-cohort (APC) analysis was conducted to assess temporal effects across age groups, periods, and birth cohorts.

Risk factor attribution

We conducted a stratified analysis of depression burden attributable to modifiable risk factors among individuals aged ≤ 30 years. The population attributable fraction (PAF) for each risk factor was estimated using GBD 2021 methods and stratified by sex, region, and year. Risk factors analyzed included: Bullying victimization, Childhood sexual abuse, and Intimate partner violence (IPV).Although these risk factors may co-occur, each was evaluated independently in the GBD framework. Potential overlaps were not adjusted in PAF calculations, which may lead to slight overestimation due to double-counting. The burden attributable to each factor was estimated by multiplying its PAF by the number of incident cases and DALYs. All estimates are presented with 95% uncertainty intervals (UIs), derived from 1,000 posterior draws of the model.

The GBD analytical framework uses standardized ICD-coded definitions, rigorous model-based estimation (DisMod-MR 2.1), and uncertainty quantification to ensure cross-country and temporal comparability. UIs reflect both data and model uncertainty, and represent the 2.5th and 97.5th percentiles of 1,000 model simulations.

Inequality and frontier analyses

Geographical disparities in depression burden were evaluated using: The Slope Index of Inequality (SII) to quantify absolute inequality by regressing DALY rates on SDI rankings, The Concentration Index (CI) to assess relative inequality across the socioeconomic gradient.A frontier analysis was also conducted to explore the nonlinear relationship between SDI and ASDR, identifying countries with a burden that exceeds expectations based on development level. Such deviations may reflect unmet mental health needs, detection bias due to enhanced screening, or sociocultural factors influencing help-seeking and diagnosis.

Decomposition analysis

To determine the contributors to changes in depression burden between 1990 and 2021, we applied a decomposition model. Total changes in DALYs were disaggregated into contributions from: Population growth, Population aging, and Epidemiological transitions.Globally, 21.8% of the increase in DALYs was attributed to epidemiological changes, suggesting that rising incidence or severity—not just demographic shifts—played a significant role in the increasing burden.

Statistical analysis

All statistical analyses were performed using R version 4.3.3. EAPCs and their 95% confidence intervals (CIs) were used to assess trends in ASIR and ASDR. An increasing trend was defined as EAPC with a 95% CI lower bound > 0, and a decreasing trend as a 95% CI upper bound < 0. Statistical significance was set at p < 0.05.

Results

Worldwide impact of depression in adolescents and young adults

In 2021, the global ASIR and DALY rate for depression were 2,688.66 (95% UI: 1,893.–3,774.15) and 402.06 (95% UI: 253.25–603.55) per 100,000 population, respectively. EAPC analysis revealed minimal variations over time, with ASIR at 0.03 (95% CI: −0.43 to 0.49) and DALY rate at 0.06 (95% CI: −0.38 to 0.49).Analysis by SDI indicated that high-SDI regions exhibited the highest ASIR (4,261.36, 95% UI: 3,110.07–5,714.89) and DALY rate (618.69, 95% UI: 397.87–918.22). Conversely, middle-SDI regions recorded the lowest ASIR (2,195.94, 95% UI: 1,550.35–3,071.33) and DALY rate (331.35, 95% UI: 208.23–497.59). Geographically, depression burden was most pronounced in North America and certain European regions. At the national level, ASIR and DALY rates followed a U-shaped relationship with SDI (p < 0.05)(Fig. 1; Table 1).

Fig. 1.

EAPCs of the ASRs for depression among youth and adolescents from 1990 to 2021. A, EAPCs of the ASDR for depression among youth and adolescents in 204 countries and regions. B, EAPCs of the ASIR for depression among youth and adolescents in 204 countries and regions. ASRs, agestandardized rates; ASDR, agestandardized disabilityadjusted lifeyear rate; ASIR: age-standardized incidence rate; EAPC, estimated annual percentage change

Table 1.

The incidence and dalys of MDD in 1990 and 2021 by gender, SDI regions and GBD regions, and their EAPC

Risk Factor	1990					2021					1990–2021
	Incidence (95% UI)	DALYs (95% UI)	ASIR (95% UI)	ASDR (95% UI)		Incidence (95% UI)	DALYs (95% UI)	ASIR (95% UI)	ASDR (95% UI)		EAPC of ASIR (95% CI)	EAPC of ASDR (95% CI)
Global	69,264,983.67(50,177,073.07,95,181,306.86)	10,532,779.07(6,744,715.17,15,636,225.34)	2165.78(1569.31,2975.92)	329.55(211.12,489.12)		107,151,780.82(75,553,859.91,150,378,069.48)	16,048,941.72(10,115,148.53,24,082,959.97)	2688.66(1893.94,3774.15)	402.06(253.25,603.55)		0.03(-0.43,0.49)	0.06(-0.38,0.49)
Gender
Men	26,031,511.17(18,734,181.49,35,873,061.13)	43,233,472.50(31,362,130.54,59,487,507.59)	1610.43(1159.78,2218.73)	249.17(158.97,371.34)		42,253,182.64(29,849,320.85,59,181,344.49)	64,898,598.19(45,727,608.22,91,424,539.69)	2079.67(1468.14,2913.62)	315.22(197.48,476.11)		0.11(-0.13,0.35)	0.133(-0.078,0.345)
Women	43,233,472.50(31,362,130.54,59,487,507.59)	6,510,599.47(4,170,170.31,9,636,250.33)	2733.97(1982.80,3761.95)	411.70(263.75,609.21)		64,898,598.19(45,727,608.22,91,424,539.69)	10,532,779.07(6,744,715.17,15,636,225.34)	3322.78(2338.32,4681.60)	492.38(309.50,735.33)		-0.04(-0.28,0.21)	-0.003(-0.219,0.213)
Andean Latin America	392,451.10(268,073.61,559,606.82)	60,265.41(36,629.26,91,987.56)	1640.60(1125.07,2334.91)	253.95(154.83,387.02)		886,665.16(589,592.86,1,302,096.57)	132,314.15(78,730.60,207,616.16)	2358.62(1561.67,3466.81)	350.03(207.84,549.54)		0.20(-0.47,0.86)	0.18(-0.40,0.77)
Australasia	459,981.43(333,540.61,611,914.25)	67,745.06(43,396.74,99,770.02)	4325.87(3118.28,5776.31)	632.65(403.52,933.36)		631,337.27(426,395.53,890,398.24)	92,421.72(57,048.89,141,679.50)	4883.36(3281.13,6915.08)	708.13(435.50,1088.71)		0.23(-0.32,0.78)	0.23(-0.30,0.76)
Caribbean	575,790.22(400,923.92,807,228.62)	85,210.96(52,799.19,130,020.63)	2571.45(1790.04,3606.76)	380.58(235.75,580.79)		694,381.89(462,983.19,1,036,291.99)	101,647.61(60,991.17,158,501.71)	2823.57(1875.87,4216.68)	411.47(246.35,642.05)		-0.34(-1.05,0.37)	-0.33(-0.97,0.31)
Central Asia	789,040.19(546,473.89,1,109,132.58)	123,615.86(76,384.75,186,353.98)	1887.45(1306.98,2653.31)	295.63(182.73,445.68)		1,162,785.43(786,427.55,1,682,289.28)	178,037.21(109,692.27,273,793.79)	2405.40(1621.91,3485.21)	366.51(225.41,564.11)		0.20(-0.12,0.52)	0.19(-0.10,0.47)
Central Europe	886,617.61(630,809.33,1,227,879.47)	143,079.22(90,301.09,212,155.88)	1475.63(1050.56,2043.80)	238.46(150.52,353.52)		775,011.05(536,164.29,1,108,141.40)	121,829.21(74,912.80,184,138.58)	1911.28(1313.86,2740.09)	297.39(182.07,450.57)		-0.24(-0.69,0.22)	-0.19(-0.59,0.21)
Central Latin America	1,801,462.47(1,271,901.94,2,519,667.94)	271,941.74(170,391.30,410,047.04)	1712.58(1214.08,2394.07)	260.21(163.31,391.88)		3,720,376.60(2,632,323.91,5,200,975.33)	544,225.58(340,400.63,824,016.61)	2662.63(1879.84,3726.25)	388.44(242.70,588.45)		0.80(0.20,1.40)	0.72(0.19,1.26)
Central Sub-Saharan Africa	1,371,130.07(903,910.94,2,012,968.31)	195,957.27(117,851.71,307,034.41)	4181.24(2776.95,6121.99)	603.03(364.56,943.66)		3,787,973.56(2,402,968.18,5,706,442.14)	542,608.41(318,579.95,849,721.11)	4483.35(2866.34,6733.98)	648.81(383.55,1012.64)		-0.06(-0.35,0.23)	-0.02(-0.30,0.27)
East Asia	12,565,269.76(9,228,597.94,16,998,872.71)	1,966,506.19(1,264,898.94,2,917,375.52)	1537.49(1124.65,2082.05)	239.35(153.73,355.46)		5,383,481.06(3,880,225.73,7,328,451.23)	898,183.15(570,868.06,1,325,792.13)	993.22(713.73,1354.26)	164.37(104.23,242.93)		-1.58(-2.06,-1.10)	-1.36(-1.83,-0.89)
Eastern Europe	2,128,626.43(1,483,172.78,3,009,936.49)	328,993.15(206,136.02,489,106.93)	1974.97(1371.93,2794.19)	303.23(189.63,451.77)		2,023,395.49(1,377,764.72,2,850,971.13)	299,186.28(185,804.62,451,330.18)	2821.96(1918.83,3974.80)	416.34(258.50,628.72)		0.02(-0.23,0.28)	0.03(-0.20,0.26)
Eastern Sub-Saharan Africa	3,146,537.76(2,148,682.77,4,494,078.08)	477,400.83(295,062.39,719,000.97)	2736.65(1882.88,3901.62)	421.17(261.60,632.50)		8,733,490.29(5,836,970.71,12,622,046.65)	1,314,233.84(795,284.59,2,014,904.26)	3150.62(2118.85,4546.26)	478.98(291.25,733.06)		-0.20(-0.50,0.10)	-0.14(-0.45,0.17)
High-income Asia Pacific	1,572,989.67(1,167,851.51,2,081,845.42)	243,933.05(156,297.67,358,243.84)	1733.26(1282.51,2299.78)	267.84(171.21,393.76)		1,443,239.73(1,044,974.21,1,931,010.94)	219,694.98(140,011.02,324,169.60)	2377.77(1706.19,3196.41)	356.83(226.40,528.12)		0.64(0.21,1.07)	0.57(0.19,0.96)
High-income North America	4,620,877.21(3,451,176.45,6,103,460.65)	715,909.89(464,080.08,1,049,284.07)	3295.39(2443.32,4367.76)	503.35(325.20,740.05)		9,545,088.55(7,158,448.55,12,443,183.28)	1,374,857.24(900,493.84,2,018,571.45)	6132.88(4588.67,8003.08)	875.97(573.15,1287.32)		1.00(0.35,1.66)	0.88(0.25,1.51)
High-middle SDI	12,513,906.75(9,228,496.35,16,824,596.17)	1,929,481.48(1,240,943.02,2,833,796.05)	2019.01(1483.29,2718.78)	309.26(198.55,454.57)		11,926,083.12(8,291,519.25,16,798,037.58)	1,810,986.63(1,118,109.00,2,715,870.76)	2362.96(1633.73,3330.67)	355.86(219.00,535.02)		-0.08(-0.57,0.40)	-0.07(-0.53,0.39)
High SDI	11,990,384.97(9,018,502.07,15,706,712.35)	1,845,365.55(1,200,721.46,2,685,906.56)	2702.57(2018.11,3555.25)	411.20(266.65,600.05)		17,772,705.11(13,038,026.16,23,809,843.66)	2,610,469.72(1,682,075.74,3,867,730.83)	4261.36(3110.07,5714.89)	618.69(397.87,918.22)		0.76(0.18,1.34)	0.67(0.13,1.22)
Low-middle SDI	16,411,222.00(11,550,259.67,23,227,631.90)	2,447,569.12(1,543,116.33,3,706,222.99)	2355.99(1665.72,3329.50)	353.95(223.74,535.46)		31,124,440.83(21,476,508.09,44,196,179.47)	4,650,942.33(2,901,008.17,6,997,890.19)	2714.06(1871.95,3854.69)	405.48(252.87,610.14)		-0.33(-0.76,0.10)	-0.24(-0.64,0.16)
Low SDI	7,406,630.67(5,049,543.12,10,654,813.80)	1,108,685.23(681,714.05,1,676,885.82)	2541.58(1744.46,3651.65)	385.04(237.89,580.69)		20,039,128.00(13,430,882.83,28,896,011.80)	2,994,439.96(1,835,441.34,4,574,161.81)	2845.30(1920.84,4098.67)	429.43(264.46,654.44)		-0.18(-0.53,0.17)	-0.11(-0.45,0.23)
Middle SDI	20,885,810.87(15,044,690.42,28,699,951.44)	3,192,945.68(2,042,375.43,4,745,989.70)	1849.44(1332.22,2541.29)	283.01(181.13,420.55)		26,209,973.37(18,548,667.27,36,628,954.52)	3,970,195.92(2,498,573.29,5,958,003.03)	2195.94(1550.35,3071.33)	331.35(208.23,497.59)		-0.11(-0.57,0.36)	-0.07(-0.51,0.37)
North Africa and Middle East	7,130,728.11(4,845,769.79,10,263,122.96)	1,046,774.99(634,841.00,1,600,486.43)	3366.82(2302.23,4841.88)	498.47(303.45,760.09)		14,115,012.23(9,269,298.97,20,827,710.80)	2,057,458.14(1,209,612.50,3,215,534.80)	4164.26(2734.74,6143.97)	607.20(356.99,949.02)		0.33(-0.14,0.80)	0.32(-0.12,0.76)
Oceania	84,310.52(57,185.78,123,315.80)	12,732.18(7795.26,19,608.19)	2042.42(1391.06,2982.59)	310.74(190.70,477.74)		178,624.47(116,228.09,269,867.67)	26,995.51(15,617.65,42,301.99)	2160.94(1408.30,3262.01)	327.25(189.55,512.25)		-0.13(-0.24,-0.02)	-0.11(-0.24,0.02)
South Asia	14,485,726.11(10,372,178.97,20,200,883.94)	2,163,370.37(1,374,189.15,3,265,037.61)	2225.09(1597.40,3098.36)	334.05(212.56,504.02)		27,395,463.51(19,286,747.77,38,227,079.30)	4,122,613.82(2,593,725.88,6,183,888.13)	2464.25(1731.77,3441.34)	370.01(232.59,555.24)		-0.63(-1.05,-0.21)	-0.49(-0.88,-0.10)
Southeast Asia	4,430,606.82(3,095,488.37,6,241,556.43)	699,487.18(437,682.67,1,049,805.60)	1480.26(1037.33,2083.75)	235.10(147.38,352.40)		6,979,868.68(4,743,158.43,9,945,844.76)	1,084,238.36(667,466.39,1,633,069.93)	1893.72(1282.26,2701.80)	292.22(179.47,440.77)		0.12(-0.15,0.39)	0.13(-0.15,0.40)
Southern Latin America	809,248.79(591,589.24,1,088,245.67)	119,477.75(76,130.12,177,836.57)	2969.95(2173.55,3993.25)	439.32(280.08,653.52)		1,280,700.67(891,986.28,1,769,370.75)	187,383.28(115,495.15,280,115.95)	3799.73(2631.49,5265.70)	551.44(338.83,826.76)		0.11(-0.56,0.78)	0.09(-0.53,0.70)
Southern Sub-Saharan Africa	772,580.56(555,171.12,1,056,534.79)	118,577.47(75,503.61,175,922.38)	2311.22(1667.73,3155.73)	357.10(227.88,529.39)		1,521,509.36(1,076,777.67,2,120,226.31)	225,611.48(142,133.18,338,150.92)	3300.73(2334.13,4600.05)	488.98(307.88,732.92)		0.64(0.25,1.03)	0.57(0.20,0.93)
Tropical Latin America	2,453,791.69(1,775,587.79,3,348,845.20)	356,537.89(228,718.27,529,380.93)	2558.63(1855.26,3487.57)	372.97(239.46,553.48)		3,564,984.49(2,578,785.84,4,847,495.09)	514,156.28(329,962.72,770,778.54)	3107.55(2238.04,4237.88)	444.87(284.77,667.71)		-0.44(-1.25,0.39)	-0.39(-1.14,0.37)
Western Europe	6,314,004.01(4,801,197.36,8,250,778.33)	956,389.48(620,818.57,1,384,077.47)	3374.79(2543.71,4427.93)	504.57(326.14,732.86)		6,643,383.24(4,542,546.03,9,418,664.33)	979,750.01(601,045.35,1,493,408.95)	4188.31(2843.83,5946.59)	612.67(374.27,936.24)		0.06(-0.55,0.68)	0.05(-0.52,0.63)
Western Sub-Saharan Africa	2,473,213.13(1,687,213.71,3,551,625.93)	378,873.13(233,045.51,570,997.19)	2178.53(1495.47,3124.59)	337.74(208.55,507.72)		6,685,008.10(4,524,163.90,9,745,877.47)	1,031,495.45(632,043.97,1,574,402.52)	2193.90(1495.28,3196.15)	343.13(211.02,522.30)		-0.16(-0.57,0.25)	-0.10(-0.48,0.29)

According to the 2021 GBD study, Greenland (9,772.21, 95% UI: 6,428.13–14,211.27), Palestine (6,379.27, 95% UI: 3,977.35–9,939.96), and the United States (6,330.05, 95% UI: 4,746.95–8,260.89) had the highest ASIR. These countries also had the highest ASDR, with Greenland (1,383.99, 95% UI: 811.53–2,189.55), Palestine (916.45, 95% UI: 513.04–1,523.86), and the United States (902.23, 95% UI: 593.08–1,322.24) leading(Figure S1,Table S1).

Sex disparities in 2021 showed significantly higher ASIR in females (3,322.78, 95% UI: 2,338.32–4,681.60) compared to males (2,079.67, 95% UI: 1,468.14–2,913.62)(Figure S2 and Table 1). Additionally, incidence and DALY rates increased with age. Trends from 1990 to 2021 remained relatively stable across age groups, apart from a slight increase in adolescents and young adults around 2020. Incidence rates were comparable between the 20–24 and 25–29 age groups, though the latter exhibited a marginally higher DALY rate(Figure S3).

Among 21 GBD regions, High-income North America recorded the highest ASDR (875.97, 95% UI: 573.15–1,287.32) and ASIR (6,132.88, 95% UI: 4,588.67–8,003.08), whereas East Asia had the lowest ASDR (164.37, 95% UI: 104.23–242.93) and ASIR (993.22, 95% UI: 713.73–1,354.26). Across SDI classifications, High-SDI regions consistently had the highest ASDR and ASIR, whereas Middle-SDI regions had the lowest rates(Figure S4).

1990–2021 trends in EAPC of ASDR and ASIR

The greatest increase in EAPC of ASDR and ASIR was observed in High-income North America, with ASDR at 0.88 (95% CI: 0.25–1.51) and ASIR at 1.00 (95% CI: 0.35–1.66). The most significant decrease was noted in East Asia, with ASDR (− 1.36, 95% CI: −1.83 to − 0.89) and ASIR (− 1.58, 95% CI: −2.06 to − 1.10) declining. High-SDI regions exhibited the largest increase, while Low-middle SDI regions showed the greatest reduction(Figure S5).

Proportional burden of major depressive disorder and dysthymia in 2021

MDD accounted for the majority of DALYs and incidence, particularly in High-SDI regions, where MDD constituted 90.60% of DALYs and 96.9% of incidence. In contrast, Low-SDI regions had the lowest proportions (84.90% and 94.60%, respectively). Among GBD regions, High-income North America had the highest proportion of DALYs (92.90%) and incidence (97.60%) attributable to MDD, while East Asia had the lowest DALY proportion (74.70%)(Figure S6).

Association between EAPC of ASDR and DALYs across 204 countries in 2021

A significant positive correlation was identified between EAPC of ASDR and DALYs (r = 0.20, p = 0.35 × 10^− 2), as well as between EAPC of ASDR and incidence (r = 0.21, p = 0.21 × 10^− 2), indicating that regions with increasing EAPC tended to have a higher disease burden. However, no significant association was observed between EAPC and SDI(Figure S7).

Temporal trends of ASDR and ASIR in global and SDI regions

Between 1990 and 2015, ASDR and ASIR remained stable, followed by a moderate increase from 2015 to 2020, then stabilization. High-SDI regions consistently reported the highest DALY and incidence rates, whereas Low-SDI and Low-middle SDI regions had rates exceeding the global average(Figure S8).

Effects of age, period, and birth cohort on adolescent and youth depression

APC modeling indicated peak Net and Local drift rates for DALYs and incidence in the 5–10-year age group, followed by declines with age. Period effects peaked in 2005 and declined post-2010. Age effects for DALYs and incidence increased before 20 years and stabilized between 20 and 25 years. Cohort effects were lowest between 1980 and 1990, rising thereafter(Fig. 2).

Fig. 2.

APC analysis of depression among youth and adolescents of global DALYs and incidence attributable from 1990 to 2021. Including Net drifts and local drifts, age effects, period effects and cohort effects in DALYs, age-specific rates by period and cohort-specific rates. APC: The age-period-cohort; DALYs: Disability-adjusted life years

Risk factors for depression in adolescents and young adults

The leading contributors to depression-related DALYs were bullying victimization, child sexual abuse, and intimate partner violence (IPV). Bullying victimization was most impactful in the 10–24 age group, while IPV had greater influence in young adults (20–29 years)(Fig. 3).

Fig. 3.

The proportion of risk factors of depression among youth and adolescents in different age groups of global DALYs in 2021

Frontier and decomposition analysis

Frontier analysis highlighted increased depression burden in high-SDI nations (e.g., Switzerland, Norway, Monaco, and Germany) and declining trends in low-SDI countries (e.g., Burundi, South Sudan, Mali, Chad, and Somalia). Decomposition analysis indicated that global DALYs increased due to aging (50.09%), population growth (28.10%), and epidemiological shifts (21.81%)(Fig. 4,Table S2). This indicates that demographic factors alone do not account for the growing burden; rising incidence and/or severity of depression also play an essential role.

Fig. 4.

The frontier analysis based on age-standardized DALY rates for depression among youth and adolescents in 21 regions and 5 SDI regions. The color scale ranges from dark blue (1990) to light blue (2021). Examples of frontier countries with low-SDI (0.85) and relatively high effective difference are labeled in red. Red dots indicate a decrease in ASDR, while blue dots indicate an increase in ASDR from 1990 to 2021. DALYs, disability adjusted life-years; SDI, sociodemographic index

The greatest aging-related burden was in High-middle SDI regions (603.20%), while population growth contributed most in Middle-SDI regions (82.70%) and epidemiological changes in High-SDI regions (114.73%)(Fig. 5,Table S3).

Fig. 5.

Changes in DALYs for depression among youth and adolescents at the population level, driven by determinants such as aging, population growth, and epidemiological transitions, from 1990 to 2021 globally and across the five SDI regions

Inequality analysis

Depression burden disparities widened over time. The slope index of inequality (SII) for ASDR increased from 0.01 (95% CI: −0.02 to 0.04) in 1990 to 0.04 (95% CI: 0.01 to 0.06) in 2021, while the concentration index (CI) rose from − 13.69 (95% CI: −63.46 to 36.08) to 67.08 (95% CI: 7.60 to 126.55), highlighting growing health disparities between low-SDI and high-SDI regions(Fig. 6,Table S4).

Fig. 6.

The inequality slope index and concentration index for DALYs of depression among youth and adolescents worldwide in 1990 and 2021. A, The absolute inequality relationship of the ASDR of depression among youth and adolescents; B, The relative inequality relationship of the ASDR of depression among youth and adolescents. Inequality slope index, depicting the relationship between SDI and age standardized DALYs rates for each condition, with points representing individual countries sized by population. The concentration index, which quantifies relative inequalities by integrating the area under the Lorenz curve, aligning DALYs distribution with population distribution by SDI. Blue represents data from 1990, and red represents data from 2021. DALYs, disability-adjusted life-years; SDI, sociodemographic index

Discussion

In the early 21st century, the World Health Organization (WHO) launched the World Mental Health Survey (WMHS) to assess the global prevalence and impact of mental disorders, including depression [16]. The expansion of WMHS provided epidemiological data across multiple countries, informing policy initiatives aimed at improving mental health services and reducing the burden of depression [17].

This study found that the most notable decline in ASIR and ASDR occurred between 2005 and 2010, potentially reflecting the impact of global mental health initiatives, including WMHS [18]. The burden of depression varied significantly across countries. For example, Germany exhibited the highest ASR for adolescent and young adult depression, which may be linked to its healthcare system’s emphasis on early intervention, thereby increasing diagnostic rates [19]. Despite this, surveys indicate that depression prevalence in Germany remains high due to a complex interplay of social, psychological, and biological risk factors [20]. Conversely, Malaysia reported lower ASR than more than half of the countries and regions globally, yet exhibited the most pronounced increase in ASR from 1990 to 2021. Studies from Malaysia suggest a rising trend in mental health issues among children aged 5 to 15 years, with depression closely linked to academic stress, family environment, social support, and broader psychosocial challenges [21].

At the national scale, the reported depression burden in this study varies from earlier findings. For instance, while GBD 2019 identified Uganda as having the highest ASIR, our study found Germany to be the leading country in this regard [22]. Future research is needed to explore potential explanations for these discrepancies.From 1990 to 2021, the global burden of depression in individuals under 30 displayed significant spatial and temporal variations. Although ASIR and ASDR generally trended upward, the total number of depression cases and DALYs continued to grow, underscoring an ongoing public health concern, especially among at-risk populations.

To estimate the proportion of depression burden attributable to specific risk factors, we used population attributable fractions (PAFs) based on the GBD 2021 comparative risk assessment framework. The GBD study applies a counterfactual approach to quantify the burden that could be avoided if exposure to a given risk factor were reduced to an ideal level.

This study identified a marked increase in depression burden among individuals under 30, with the most pronounced rise occurring in those aged 10–14 years. Compared to other age groups, adolescent depression is often overlooked due to atypical symptom presentation [23]. Emotional distress may manifest as irritability, academic decline may be misinterpreted as laziness, and social withdrawal might go unnoticed, resulting in delayed diagnosis and treatment. Such delays heighten the risk of long-term mental health consequences, including substance use, self-harm, and suicidality. Among young adults (18–24 years), the heightened depression risk is likely associated with transitional life stressors, such as academic pressures, career uncertainty, financial burdens, and evolving social relationships. Additionally, the widespread use of social media has emerged as a significant risk factor. Evidence suggests that social media addiction exacerbates depression risk through mechanisms such as sleep disruption, social isolation, negative self-comparison, and emotional detachment [24]. Prolonged social media exposure has been linked to poorer sleep quality, while frequent “social comparison” may foster feelings of inadequacy, anxiety, and loneliness, further increasing vulnerability to depression [25].

Consistent with previous research, this study found that depression burden was significantly higher among adolescent and young adult females than males [26]. This gender disparity likely results from a combination of biological, social, and psychological factors [27]. Hormonal fluctuations during puberty, particularly changes in estrogen and progesterone, may influence mood regulation, increasing vulnerability to depression [28]. Socially, adolescent girls are more likely to experience gender role pressures, body image concerns, and adverse life events, further contributing to depression risk [29]. Additionally, greater emotional sensitivity and stress reactivity among females may heighten their susceptibility [30]. These findings highlight the need for gender-sensitive mental health interventions, particularly during adolescence and early adulthood, focusing on resilience-building and targeted support.

One notable finding is the higher burden of depression in high-SDI regions. This may reflect greater awareness, improved diagnosis, and better access to mental health care, which increases case detection [31]. Additionally, social stressors such as academic competition, workplace pressure, and high expectations may amplify mental health risks in affluent societies. Hence, the observed burden may be a combination of true prevalence and detection bias.

This study also underscores the role of sociocultural and environmental factors in adolescent and young adult depression. In low- SDI regions, economic hardship, limited education, and inadequate healthcare contribute to a rising depression burden [32]. In contrast, high-SDI regions, despite greater material resources, exhibit elevated depression risk due to academic stress, competitive environments, and high parental expectations. Furthermore, family and community settings significantly influence adolescent depression. Adverse family conditions, such as domestic violence and poor parent-child relationships, along with limited community-based mental health support, exacerbate the problem [33]. These findings emphasize the importance of targeted interventions addressing modifiable risk factors, including improving family dynamics, expanding community mental health services, and enhancing social and educational support systems to mitigate depression burden.

This study uses GBD super regions, which group countries based on geographic and epidemiological similarity. The seven super regions include High-income; Central Europe, Eastern Europe, and Central Asia; Latin America and Caribbean; North Africa and Middle East; South Asia; Southeast Asia, East Asia, and Oceania; and Sub-Saharan Africa. These groupings aid in interpreting regional variations in disease burden by reflecting socioeconomic and health system similarities.

Globally, depression prevention and intervention strategies should be tailored to the socioeconomic and cultural contexts of different populations. Research indicates that school-based mental health education, family support programs, and community mental health services are effective in reducing depression burden [34]. For example, school-based programs that teach emotional regulation and stress management can help adolescents cope with academic and social pressures [35]. Family support initiatives that foster open communication and reduce conflict may enhance adolescent well-being. Additionally, expanding community mental health services, particularly in low-SDI regions, can provide essential resources such as counseling, crisis intervention, and social support networks, ultimately improving mental health outcomes. Effective implementation of these measures requires culturally and contextually appropriate policies and interventions.

Future directions

Global mental health strategies should prioritize adolescents and young adults, with a focus on early identification and intervention targeting modifiable risk factors. Governments and non-governmental organizations should increase investments in mental health resources, integrating mental health education into schools, communities, and family settings. Moreover, targeted interventions for high-risk groups, including adolescent females, should be central to public health policies.The rationale for the study is rooted in the increasing global concern for youth mental health, especially post-COVID-19. Depression is among the top contributors to disability globally. Recent literature highlights rising incidence among adolescents, gender disparities, and growing risk from social determinants. Our study builds on these findings using the latest GBD 2021 data, offering updated and comprehensive global trends and risk factor analyses to guide policy.

Limitations

This study has several limitations. First, GBD 2021 estimates may be constrained by insufficient epidemiological data on depression in some countries, particularly low-income regions. Diagnostic capabilities in underdeveloped healthcare systems may have evolved over time, affecting the accuracy of GBD data and leading to potential misclassification and underreporting. Additionally, only bullying victimization, childhood sexual abuse, and intimate partner violence were considered as depression-related risk factors in the GBD database, limiting the scope of risk factor analysis. Other potential contributors, such as social media addiction and family structure changes, warrant further investigation. Future research should incorporate longitudinal cohort studies to elucidate the complex etiology of adolescent and young adult depression and its long-term consequences.

Conclusion

Between 1990 and 2021, the burden of depression among adolescents and young adults exhibited an upward trend, with increasing cases, incidence rates, ASIR, and ASDR. Findings highlight that adolescents and young adults, particularly females, remain a high-risk population. Early prevention and targeted interventions, particularly in mitigating bullying victimization, childhood sexual abuse, and intimate partner violence, are critical in reducing depression burden. These results provide evidence-based insights for policymakers and healthcare authorities to optimize mental health strategies and enhance global mental health outcomes through interdisciplinary collaboration.Future research should explore additional modifiable risk factors, including digital media use, substance abuse, and school climate, and assess intervention effectiveness in diverse contexts. Longitudinal studies may better elucidate causality and inform age-specific strategies.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Z.L. wrote the main manuscript text and M.K. prepared Figs. 1, 2, 3, 4, 5 and 6. All authors reviewed the manuscript.

Funding

Project of Hunan Provincial Natural Science Foundation (2024JJ8137), This research was supported by the 2023 Outstanding Innovative Youth Program of Changsha City (KQ2306028), and the Project of the Hunan Provincial Health Commission (B202303108228), National Natural Science Youth Fund (82104839), and Hunan Provincial Natural Science Foundation (2025JJ80081).

Data availability

Data is provided within the manuscript or supplementary information files.This study used publicly available data from the Global Burden of Disease Study 2021, accessible at https://vizhub.healthdata.org/gbd-results/.

Declarations

Ethics approval and consent to participate

This study did not involve human participants, clinical samples, or live animals and is based entirely on literature review and secondary data analysis. Accordingly, institutional review board approval and written informed consent were not required. All data were obtained from publicly accessible sources and analyzed in aggregate form, ensuring no individual person’s privacy or welfare was affected.Clinical trial number: not applicable.Clinical trial number: not applicable.

Consent to participate

This study was approved by the all author. All participants were informed of the study’s purpose, procedures, potential risks, and benefits.

Consent to publish

All participants involved in this study provided written consent for the publication of the research findings, including any associated data, images, and personal details presented in anonymized form.

Competing interests

The authors declare no competing interests.