Abstract
Background
Anxiety disorders are among the most prevalent mental disorders worldwide and impose a substantial burden on public health systems. This study aimed to comprehensively assess the global burden of anxiety disorders, examine long-term temporal trends, explore their associations with human resources for health (HRH), and project future disease burden.
Methods
Data were obtained from the Global Burden of Disease (GBD) 2021 study. We first described the global burden of anxiety disorders in 2021 and its geographical distribution across countries and regions. We then analyzed temporal trends in incidence, prevalence, and disability-adjusted life years (DALYs) from 1990 to 2021. In addition, correlations between anxiety disorder burden and HRH were assessed. Finally, future trends from 2022 to 2050 were projected using multiple statistical models.
Results
In 2021, anxiety disorders accounted for 53.91 million incident cases worldwide, with an age-standardized incidence rate of 678.25 per 100,000 population. The global number of prevalent cases reached 359.21 million, and the total number of DALYs was 42.51 million. China had the highest absolute number of cases, whereas Portugal had the highest age-standardized prevalence rate. Correlation analyses showed moderate associations between several HRH categories and the burden of anxiety disorders; traditional and complementary medicine practitioners were weakly negatively correlated with DALYs. Model-based projections suggest that the global burden of anxiety disorders may continue to increase over the next 25 years.
Conclusion
Anxiety disorders remain a major global public health challenge. These findings provide updated evidence on their epidemiological burden and may help inform health resource allocation and long-term mental health planning.
Keywords: Anxiety disorders, Age-standardized rate, Disability-adjusted life years, Global burden of disease, ARIMA, Human resources for health
Introduction
Anxiety disorders are characterized by persistent and impairing fear and worry, typically accompanied by behavioral, cognitive, and physical symptoms, and they can substantially disrupt daily functioning and quality of life [1]. Despite the availability of effective treatments, the treatment gap for anxiety disorders remains large across countries [2].These figures underline anxiety disorders’ importance as a public health and health-system challenge—not only because of their high prevalence, but also because unmet need remains substantial.
Within the Global Burden of Disease (GBD) framework, mental disorders contribute predominantly to years lived with disability (YLDs) rather than mortality. In GBD 2019, the global number of DALYs due to mental disorders increased from 80.8 million in 1990 to 125.3 million in 2019, and the proportion of global DALYs attributable to mental disorders rose from 3.1 to 4.9% over the same period, highlighting persistent and growing non-fatal health loss [3]. For anxiety disorders specifically, the combination of high prevalence and chronic or recurrent course implies a large disability burden and a high demand for sustained, accessible care [4].
The COVID-19 pandemic amplified this challenge. WHO concluded that global prevalence of anxiety and depression increased by approximately 25% in the first year of the pandemic [5]. GBD-affiliated modeling for 2020 estimated that anxiety disorders increased by 25.6% globally [3], corresponding to 76.2 million additional cases, with greater impacts among females and younger age groups, and with cross-national variation associated with pandemic indicators such as reduced human mobility and higher SARS‑CoV‑2 infection rates [3]. At the same time, it is critical to distinguish between elevated anxiety symptoms measured in surveys and formally diagnosed anxiety disorders recorded in clinical systems. Longitudinal evidence suggests that anxiety symptoms peaked early in the pandemic and then declined, with substantial heterogeneity and limited longitudinal evidence from some world regions [6]. In parallel, health-system constraints likely affected recorded diagnoses: a multinational network analysis of electronic health records and claims data reported immediate declines in incident mental health diagnoses following COVID-related restrictions, followed by gradual recovery toward historical baselines in many settings by 2021 [7].
Health-system capacity, particularly human resources for health (HRH), is therefore central to interpreting post-COVID burden patterns and to designing effective responses. A WHO rapid assessment found that 93% of countries reported disruptions in mental, neurological, and substance use services in mid‑2020, occurring alongside increased population mental health needs [8]. More recently, WHO’s Mental Health Atlas reporting highlights persistent structural constraints: mental health accounts for only 2% of health budgets, the global median mental health workforce is about 13 workers per 100,000 population, and fewer than 10% of countries have fully transitioned to community-based care models [9]. Such workforce and service gaps intersect with demand shocks like COVID-19 and may contribute to cross-national differences in disability burden and in the observed trajectory of service utilization. For example, analysis across nine countries found that primary-care mental health visit rates increased after pandemic onset in most settings, with anxiety/depression as the leading reason for visits and virtual care comprising up to 40% of mental health visits where recorded, indicating rapid adaptation but also raising questions about capacity, access, and equity [10].
Although HRH is widely recognized as essential for building resilient health systems, evidence linking workforce capacity to mental disorder burden can be nonlinear and context dependent. Recent cross-national work has found persistent inequality in psychologist density and a negative, nonlinear association between psychologist density and mental disorder DALY rates, with diminishing reductions in burden as workforce density increases and varying patterns across levels of social development [11]. Taken together, these findings suggest that understanding anxiety disorder burden in the post-COVID era requires simultaneous attention to epidemiologic trends and to the capacity and distribution of the health workforce that delivers prevention and care.
Using data from GBD 2021 and publicly available HRH indicators, the present study aims to: (1) estimate the global burden of anxiety disorders in 2021; (2) describe temporal trends from 1990 to 2021; (3) examine cross-national associations between anxiety disorder burden and HRH; and (4) project future burden from 2022 to 2050. By integrating post-pandemic burden dynamics with health workforce capacity, this study may provide evidence to inform mental health planning and more equitable resource allocation.
Methods
Overview
The GBD database represents the most comprehensive and detailed scientific endeavor to quantify global health levels and trends. The GBD2021 study represents the inaugural investigation to assess the impact of the COVID-19 pandemic and its cascading effects on global health. The database encompasses 371 diseases and 88 risk factors [3, 12].
We extracted data on anxiety disorders from 1990 to 2021, including incidence, prevalence, and disability-adjusted life years (DALYs), together with their corresponding age-standardized rates (ASRs). Estimates were available for 204 countries and territories, 54 GBD regions, and five sociodemographic index (SDI) categories. Data were further stratified by sex and age group. All analyses were performed using R software (version 4.4.1), and a two-sided p value < 0.05 was considered statistically significant.
Data acquisition and preprocessing
Data on anxiety disorders were obtained from the GBD 2021 study, including annual numbers and age-standardized rates of incidence, prevalence, and DALYs from 1990 to 2021. Population data for forecasting analyses from 1990 to 2050 were obtained from a published GBD-based population projection study [13]. For forecasting analyses, year and population variables were processed before model fitting. Specifically, variables were normalized, and derived features were generated, including lag terms and moving-average terms, to capture temporal patterns and reduce short-term fluctuations. Interaction terms between year and population were also considered in selected models to improve predictive performance.
Statistical analysis
Descriptive and temporal trend analyses
We first described the global burden of anxiety disorders in 2021 at the global, regional, and national levels, including incidence, prevalence, DALYs, and corresponding age-standardized rates. Subgroup analyses were conducted by sex, age group, and SDI category.
Temporal trends from 1990 to 2021 were assessed using estimated annual percentage change (EAPC), calculated by fitting a linear regression model to the natural logarithm of the age-standardized rates over calendar year. GBD regions were further grouped according to the direction and magnitude of EAPC values to summarize patterns of change.
Human resources for health (HRH) data and analysis
Data on human resources for health (HRH) were obtained from the GBD 2019 Human Resources for Health Collaborators study. These data covered 22 categories of health workers across 204 countries and territories from 1990 to 2019, with workforce density expressed as the number of workers per 10,000 population. To visualize trends in HRH density over time, we applied LOESS (Locally Estimated Scatterplot Smoothing) smoothing. Spearman’s rank correlation analysis was performed to evaluate associations between HRH density and the burden of anxiety disorders, including age-standardized incidence, prevalence, and DALY rates. Because these analyses were observational, correlation coefficients were interpreted as measures of association rather than evidence of causality [14].
Model construction and evaluation
In our study, we developed and implemented eight distinct models to capture the complex dynamics of anxiety disorder trends. The Prophet model, developed by Facebook, was utilized with several customizations to enhance its performance. We employed grid search optimization for key parameters such as changepoint_prior_scale, n_changepoints, and growth mode. Data scaling to the 0.1–0.9 range was applied to improve numerical stability, and we used maximum likelihood estimation instead of MCMC sampling for faster computation. Growth constraints and smoothing techniques were implemented to enhance prediction stability, and we incorporated country-specific holiday effects to account for potential seasonal variations.
For the ARIMA model, we used automatic parameter selection with the auto.arima function, limiting the maximum order to 5 for both AR and MA terms to prevent overfitting. Non-linear external regressors, including population (linear and quadratic terms) and their interactions with time, were incorporated. We applied the Box-Cox transformation to address non-linearity and heteroscedasticity in the data and included a drift term to capture long-term trends more effectively. Dynamic smoothing constraints with exponential decay factors were implemented to ensure realistic future projections.
Our TBATS model implementation featured a simplified ETS (Error, Trend, Seasonal) model as the foundation, with additive error and trend components. We included Fourier terms to capture complex seasonal patterns, allowing for multiple seasonal periods. The Box-Cox transformation was applied for variance stabilization, and ARMA errors were used to model the residual autocorrelation. A damped trend component was incorporated to prevent unrealistic long-term forecasts.
The Elastic Net regression model combined L1 and L2 regularization, with feature engineering including non-linear terms such as quadratic year terms, log-transformed population, and year-population interactions. We used grid search optimization for the alpha and lambda parameters, with 3-fold cross-validation for each parameter combination. Input features were standardized to ensure equal penalization in the regularization process.
Our ETS model implementation used an additive seasonal ETS model (AAA) with fixed smoothing parameters. We incorporated non-linear adjustments including logarithmic trends and cyclical fluctuations modeled using sine and cosine functions. Smoothing constraints were applied to limit maximum prediction changes to 5% year-over-year (Table 1).
Table 1.
The number of incidence cases and the age-standardized incidence rate attributable to anxiety disorders in 1990 and 2021, and its trends from 1990 to 2021 globally
| Characteristics | 1990 | 2021 | 1990–2021 | ||
|---|---|---|---|---|---|
| Number of incidence cases(95%UI) | The age standardized incidence rate/100,000(95%UI) | Number of incidence cases(95%UI) | The age standardized. incidence rate/100,000(95%UI) | EAPC(95%CI) | |
| Global | 30,239,333 (25024017–37400424) | 562.48 (468.32–686.07) | 53,917,371 (44990958–65978475) | 678.25 (565.15–832.44) | 0.21 (0.06–0.35) |
| Sex | |||||
| Female | 17,406,305 (14305986–21528923) | 649.32 (537.68–795.75) | 30,729,973 (25426580–37775899) | 784.53 (649.26–966.02) | 0.2 (0.05–0.34) |
| Male | 12,833,028 (10648667–15663015) | 479.3 (402.19–579.75) | 23,187,398 (19477816–28113826) | 576.46 (486.27–701.84) | 0.22 (0.07–0.36) |
| Age | |||||
| < 5 years | 555,878 (375845–797368) | 89.67 (60.63–128.62) | 773,776 (514626–1119410) | 117.56 (78.19–170.08) | 0.37 (0.18–0.57) |
| 5–9 years | 3,194,280 (2171849–4549903) | 547.41 (372.19–779.72) | 4,695,377 (3151260–6720947) | 683.41 (458.66–978.23) | 0.24 (0.07–0.4) |
| 10–14 years | 4,076,313 (3108602–5308851) | 760.96 (580.31–991.04) | 6,166,868 (4693642–8089046) | 925.07 (704.08–1213.41) | 0.15 (−0.01–0.32) |
| 15–19 years | 3,620,228 (2331461–5039146) | 696.97 (448.86–970.14) | 5,430,290 (3454697–7622176) | 870.27 (553.66–1221.54) | 0.22 (0.03–0.4) |
| 20–24 years | 3,257,952 (1813694–4806943) | 662.07 (368.57–976.85) | 5,073,722 (2897232–7425000) | 849.64 (485.17–1243.39) | 0.28 (0.08–0.48) |
| 25–29 years | 2,991,754 (2130737–4065136) | 675.92 (481.39–918.43) | 4,964,911 (3605929–6723011) | 843.88 (612.9–1142.7) | 0.29 (0.11–0.46) |
| 30–34 years | 2,638,442 (1943137–3467579) | 684.56 (504.16–899.68) | 5,011,965 (3703285–6539013) | 829.14 (612.64–1081.76) | 0.28 (0.13–0.42) |
| 35–39 years | 2,408,953 (1600407–3300291) | 683.89 (454.35–936.93) | 4,675,206 (3147936–6400080) | 833.57 (561.26–1141.11) | 0.31 (0.19–0.44) |
| 40–44 years | 1,917,543 (1244671–2660657) | 669.34 (434.47–928.74) | 3,993,102 (2601033–5541327) | 798.22 (519.95–1107.71) | 0.25 (0.13–0.38) |
| 45–49 years | 1,452,085 (1057761–1939379) | 625.37 (455.55–835.24) | 3,399,980 (2484912–4542895) | 718.05 (524.79–959.42) | 0.14 (0.03–0.26) |
| 50–54 years | 1,230,359 (916556–1617765) | 578.8 (431.18–761.05) | 2,900,685 (2168884–3816414) | 651.95 (487.47–857.77) | 0.09 (−0.02–0.19) |
| 55–59 years | 992,786 (685863–1400130) | 536.06 (370.34–756.01) | 2,344,622 (1633366–3304974) | 592.48 (412.75–835.16) | 0.02 (−0.09–0.13) |
| 60–64 years | 777,987 (515651–1116309) | 484.4 (321.06–695.05) | 1,687,254 (1111109–2421783) | 527.19 (347.17–756.69) | −0.03 (−0.14–0.08) |
| 65–69 years | 526,090 (382429–715118) | 425.61 (309.38–578.53) | 1,262,754 (908799–1718928) | 457.78 (329.46–623.16) | −0.01 (−0.1–0.08) |
| 70–74 years | 307,366 (230231–405727) | 363.05 (271.94–479.24) | 800,663 (607196–1055735) | 388.97 (294.99–512.89) | 0.03 (−0.04–0.1) |
| 75–79 years | 182,597 (129911–249489) | 296.64 (211.05–405.31) | 423,228 (302629–582173) | 320.91 (229.47–441.43) | 0.1 (0.05–0.14) |
| 80–84 years | 80,360 (51243–118271) | 227.16 (144.85–334.33) | 215,528 (136205–319864) | 246.08 (155.52–365.21) | 0.15 (0.12–0.18) |
| 85–89 years | 24,046 (15338–35479) | 159.13 (101.5–234.79) | 78,104 (49374–116095) | 170.82 (107.99–253.92) | 0.12 (0.1–0.15) |
| 90–94 years | 4003 (2555–5933) | 93.41 (59.63–138.46) | 17,623 (11226–26291) | 98.51 (62.75–146.97) | 0.08 (0.06–0.11) |
| 95 + years | 313 (200–466) | 30.71 (19.65–45.81) | 1715 (1098–2563) | 31.46 (20.15–47.02) | 0.03 (0.01–0.05) |
| SDI region | |||||
| High SDI | 5,782,745 (4790685–7094478) | 658.87 (548.57–807.91) | 8,394,781 (6982548–10139111) | 841.56 (702.07–1030.93) | 0.3 (0.13–0.48) |
| High-middle SDI | 6,043,692 (5032322–7397528) | 559.35 (469.03–679.83) | 8,423,075 (6984883–10298263) | 669.84 (553.81–827.68) | 0.16 (0.02–0.3) |
| Middle SDI | 9,974,238 (8200142–12231278) | 568.58 (474.43–686.56) | 17,436,716 (14457652–21191691) | 700.88 (582.42–859.4) | 0.24 (0.09–0.39) |
| Low-middle SDI | 5,940,309 (4894485–7367106) | 530.66 (437.5–651.36) | 12,828,953 (10543452–15898000) | 649.4 (534.77–801.33) | 0.34 (0.19–0.49) |
| Low SDI | 2,469,940 (2020196–3082986) | 520.59 (425.18–649.77) | 6,788,499 (5512876–8440444) | 603.86 (496.22–752.22) | 0.22 (0.11–0.34) |
UI: uncertainty intervals; CI: confidence interval; SDI: socio-demographic index; EAPC: estimated annual percentage change
The VAR model included non-linear features such as log transformations of the dependent variables, squared terms for time, and interaction effects between variables. We used a first-order lag structure to capture immediate temporal dependencies while avoiding overfitting. Granger causality tests were performed to identify significant relationships between variables, and impulse response analysis was conducted to understand the system’s dynamic behavior.
We developed a modified Holt-Winters model, combining quadratic curve fitting and exponential smoothing. This non-linear prediction model used quadratic polynomial fitting for initial trend estimation, followed by exponential smoothing to combine historical trends with quadratic predictions. Change rate constraints were implemented to ensure smooth forecasts, and an adaptive error correction mechanism was included based on recent prediction accuracy.
Lastly, we enhanced the standard Theta model to incorporate seasonality and non-linear trends. This improved model decomposed the time series into trend and seasonal components, applied the standard Theta method to deseasonalized data, and incorporated periodic fluctuations simulating 12-month and 24-month seasonality using sine functions. A random growth trend component was introduced to capture long-term variability, and smoothing treatment using weighted moving averages was applied to reduce short-term fluctuations.For all models, we implemented robust error handling and fallback mechanisms to ensure continuity in our predictions. This multi-model approach allowed us to capture a wide range of potential trends and patterns in the data, providing a comprehensive view of possible future scenarios for anxiety disorder prevalence and incidence.
Results
Burden of anxiety disorders in 2021
Figure 1 shows the global distribution of the burden of anxiety disorders in 2021. Worldwide, there were 359.21 million prevalent cases of anxiety disorders in 2021, with an age-standardized prevalence rate of 4421.87 per 100,000 population (95% UI: 3768.28–5182.08). The number of incident cases was 53.91 million (95% UI: 44.99–65.98 million), corresponding to an age-standardized incidence rate of 678.25 per 100,000 population (95% UI: 565.15–832.44). The total number of DALYs attributable to anxiety disorders was 42.51 million (95% UI: 29.40–57.73 million), with an age-standardized DALY rate of 524.33 per 100,000 population (95% UI: 363.06–716.25).
Fig. 1.
Global burden of anxiety disorders by country in 2021. A Age-standardized incidence rate. B Age-standardized prevalence rate. C Age-standardized DALY rate. D Number of incident cases. E Number of prevalent cases. F Number of DALYs
At the national level, China had the largest absolute number of prevalent cases of anxiety disorders among the 204 countries and territories, with 53.13 million cases (95% UI: 45.54–61.64 million). China also had the greatest absolute number of DALYs, reaching 6.31 million (95% UI: 4.37–8.57 million). In contrast, Portugal had the highest age-standardized prevalence rate, at 9712.44 per 100,000 population (95% UI: 6864.89–13062.42), as well as the highest age-standardized DALY rate, at 1156.75 per 100,000 population (95% UI: 692.93–1723.82) (Fig. 1).
Temporal trends in the burden of anxiety disorders from 1990 to 2021
Figure 2 shows the global temporal trends in anxiety disorder burden from 1990 to 2021. Overall, the absolute numbers of incident cases, prevalent cases, and DALYs increased substantially over the study period. In contrast, age-standardized incidence, prevalence, and DALY rates remained relatively stable from 1990 to 2019, followed by a noticeable increase after 2019 (Fig. 2).
Fig. 2.
Global trends in the burden of anxiety disorders from 1990 to 2021. A The line indicates the age-standardized incidence rate, and the bars indicate the number of incident cases. B The line indicates the age-standardized prevalence rate, and the bars indicate the number of prevalent cases. C The line indicates the age-standardized DALY rate, and the bars indicate the number of DALYs
To further characterize temporal changes across countries and territories, estimated annual percentage change (EAPC) values of age-standardized rates were calculated and visualized in Fig. 3. For the age-standardized incidence rate, Mexico showed the highest EAPC from 1990 to 2021 (EAPC = 1.00, 95% CI: 0.77 to 1.24), whereas Japan showed the lowest (EAPC = −0.24, 95% CI: −0.45 to −0.03). For the age-standardized prevalence rate, the highest EAPC was also observed in Mexico (EAPC = 1.09, 95% CI: 0.85 to 1.33), while the lowest was observed in China (EAPC = −0.39, 95% CI: −0.51 to −0.26). For the age-standardized DALY rate, Mexico had the highest EAPC (EAPC = 1.10, 95% CI: 0.86 to 1.34), whereas China had the lowest (EAPC = −0.37, 95% CI: −0.49 to −0.24) (Fig. 2).
Fig. 3.
Changes in the burden of anxiety disorders from 1990 to 2021. A Change in the number of incident cases. B Change in the number of prevalent cases. C Change in the number of DALYs
Associations between anxiety disorder burden and healthcare workforce categories
Figure 4 shows the correlations between the burden of anxiety disorders and different healthcare workforce categories in 1990 and 2019. Overall, the observed correlations were generally weak to moderate in magnitude. In 2019, for age-standardized incidence rates, the strongest positive correlations were observed for dieticians and nutritionists (r = 0.48, P < 0.001), environmental health officers (r = 0.44, P < 0.001), and dentists (r = 0.41, P < 0.001). For age-standardized prevalence rates, the strongest positive correlations were found for dental assistants (r = 0.48, P < 0.001), dieticians and nutritionists (r = 0.43, P < 0.001), and dentistry personnel (r = 0.42, P < 0.001). Similar patterns were observed for age-standardized DALY rates, for which dieticians and nutritionists (r = 0.49, P < 0.001), environmental health officers (r = 0.44, P < 0.001), and dentistry personnel (r = 0.43, P < 0.001) showed the strongest positive correlations. Traditional and complementary practitioners showed a weak negative correlation with age-standardized DALY rates (r = −0.17, P = 0.016). These findings indicate statistical associations between HRH density and anxiety disorder burden, but they should not be interpreted as evidence of causality (Fig. 4A–C).
Fig. 4.
Associations between age-standardized anxiety disorder burden and healthcare workforce categories in 204 countries in 1990 and 2019. A Correlations between age-standardized incidence rates and different HRH categories. B Correlations between age-standardized prevalence rates and different HRH categories. C Correlations between age-standardized DALY rates and different HRH categories. D Associations of age-standardized incidence, prevalence, and DALY rates with traditional and complementary practitioners. E Associations of age-standardized incidence, prevalence, and DALY rates with dieticians and nutritionists
Country-level comparisons further illustrated the heterogeneity of anxiety disorder burden and healthcare workforce composition. Portugal showed high age-standardized rates of anxiety disorders in both 1990 and 2019, whereas Uzbekistan remained among the countries with relatively low rates across incidence, prevalence, and DALYs. In contrast, New Zealand had a relatively high density of traditional and complementary practitioners but still ranked among the countries with a high burden of anxiety disorders. These descriptive comparisons suggest that the relationship between healthcare workforce composition and anxiety disorder burden is complex and may be influenced by multiple socioeconomic, cultural, and health-system factors beyond workforce density alone (Fig. 4D–E).
Multi-model forecasting of anxiety disorder burden
To improve the stability of future estimates, we compared the predictive performance of eight forecasting models for both age-standardized rates and absolute numbers of anxiety disorder burden. Model performance was evaluated using multiple indicators, including R2 and mean absolute percentage error (MAPE). As shown in Fig. 5A, the Theta model showed the best overall performance for age-standardized incidence, prevalence, and DALY rates, with relatively higher R2 values (0.71, 0.81, and 0.81, respectively) and lower MAPE values than the other models. For the absolute numbers of incident cases, prevalent cases, and DALYs, the ARIMA model performed best overall, with R2 values close to 0.99 and comparatively low prediction errors (Fig. 5C). These results suggest that different models may be more appropriate for different outcome measures and support the use of a multi-model framework for forecasting anxiety disorder burden (Table 2).
Table 2.
The number of prevalence cases and the age-standardized prevalence rate attributable to anxiety disorders in 1990 and 2021, and its trends from 1990 to 2021 globally
| Characteristics | 1990 | 2021 | 1990–2021 | ||
|---|---|---|---|---|---|
| Number of prevalence cases (95%UI) | The age standardized prevalence rate/100,000 (95%UI) | Number of prevalence cases (95%UI) | The age standardized prevalence rate/100,000 (95 %UI) |
EAPC (95%CI) | |
| Global | 192,936,134 (164639914–226523809) | 3746.36 (3234.69–4368.51) | 359,213,270 (307185425–419966576) | 4421.87 (3768.28–5182.08) | 0.13 (−0.01–0.27) |
| Sex | |||||
| Female | 120,227,844 (102878960–141298590) | 4674.19 (4018.03–5451.66) | 224,809,615 (192230867–263052153) | 5535.61 (4730.42–6495.3) | 0.11 (−0.03–0.26) |
| Male | 72,708,290 (61996057–85847346) | 2807.4 (2410.04–3284.79) | 134,403,655 (114699149–157908850) | 3308.52 (2830.39–3887.81) | 0.17 (0.04–0.31) |
| Age | |||||
| < 5 years | 532,534 (361902–767814) | 85.9 (58.38–123.85) | 741,923 (492589–1080004) | 112.72 (74.84–164.09) | 0.37 (0.18–0.57) |
| 5–9 years | 7,556,842 (5222201–10850901) | 1295.02 (894.93–1859.53) | 11,127,687 (7537913–16039904) | 1619.62 (1097.14–2334.59) | 0.23 (0.07–0.4) |
| 10–14 years | 18,067,424 (12656244–24635548) | 3372.79 (2362.64–4598.91) | 27,076,415 (18730807–37517456) | 4061.65 (2809.75–5627.88) | 0.11 (−0.05–0.27) |
| 15–19 years | 23,053,160 (17761413–29621083) | 4438.22 (3419.45–5702.68) | 33,044,677 (25155221–42792690) | 5295.79 (4031.42–6858.02) | 0.11 (−0.05–0.27) |
| 20–24 years | 22,633,064 (16836006–29782002) | 4599.39 (3421.34–6052.17) | 33,826,064 (24956762–44703689) | 5664.51 (4179.26–7486.07) | 0.18 (0.01–0.36) |
| 25–29 years | 20,650,116 (15408048–27432043) | 4665.43 (3481.1–6197.66) | 33,891,272 (25079228–45013328) | 5760.46 (4262.69–7650.86) | 0.22 (0.05–0.4) |
| 30–34 years | 18,200,540 (14055482–23335716) | 4722.23 (3646.77–6054.58) | 34,509,155 (26618310–44568218) | 5708.9 (4403.5–7372.98) | 0.21 (0.06–0.37) |
| 35–39 years | 16,608,619 (13079472–20756601) | 4715.08 (3713.18–5892.67) | 32,319,011 (25202432–40486181) | 5762.34 (4493.48–7218.51) | 0.24 (0.09–0.38) |
| 40–44 years | 13,758,718 (10254449–17677844) | 4802.65 (3579.44–6170.67) | 28,791,627 (21310409–36521138) | 5755.44 (4259.94–7300.57) | 0.19 (0.04–0.34) |
| 45–49 years | 11,090,469 (8142173–14389420) | 4776.35 (3506.6–6197.11) | 26,176,334 (19104780–33600352) | 5528.21 (4034.76–7096.1) | 0.12 (−0.01–0.24) |
| 50–54 years | 9,960,583 (7756725–12602806) | 4685.76 (3649–5928.74) | 23,849,909 (18610697–30200913) | 5360.46 (4182.9–6787.9) | 0.1 (−0.02–0.22) |
| 55–59 years | 8,592,950 (6838996–10779747) | 4639.82 (3692.76–5820.6) | 20,786,011 (16743649–25919322) | 5252.6 (4231.1–6549.78) | 0.08 (−0.04–0.2) |
| 60–64 years | 7,427,372 (5712238–9863843) | 4624.5 (3556.61–6141.52) | 16,488,480 (12689396–21796379) | 5151.88 (3964.84–6810.35) | 0 (−0.12–0.13) |
| 65–69 years | 5,763,148 (4303235–7826249) | 4662.38 (3481.31–6331.43) | 13,714,726 (10262695–18579787) | 4971.95 (3720.5–6735.66) | −0.09 (−0.2–0.02) |
| 70–74 years | 3,864,882 (2963777–5158882) | 4565.12 (3500.75–6093.56) | 9,952,077 (7605506–13304902) | 4834.88 (3694.88–6463.74) | −0.13 (−0.23–0.04) |
| 75–79 years | 2,800,754 (2166514–3598698) | 4549.96 (3519.61–5846.26) | 6,192,709 (4813402–7973794) | 4695.56 (3649.71–6046.05) | −0.15 (−0.23–0.06) |
| 80–84 years | 1,566,543 (1228639–1985401) | 4428.27 (3473.09–5612.29) | 3,986,239 (3115792–5033857) | 4551.38 (3557.53–5747.52) | −0.13 (−0.2–0.06) |
| 85–89 years | 622,446 (466899–798199) | 4119.14 (3089.78–5282.21) | 1,921,456 (1451236–2446391) | 4202.5 (3174.06–5350.6) | −0.1 (−0.17–0.03) |
| 90–94 years | 155,221 (114146–206551) | 3622.26 (2663.74–4820.12) | 655,505 (484190–876850) | 3664.22 (2706.58–4901.51) | −0.08 (−0.13–0.02) |
| 95 + years | 30,750 (21006–42764) | 3020.4 (2063.28–4200.42) | 161,993 (111207–225102) | 2972.18 (2040.38–4130.08) | −0.14 (−0.18–0.1) |
| SDI region | |||||
| High-middle SDI | 40,436,065 (34976337–46922812) | 3708.62 (3214.52–4297.07) | 59,670,393 (50848624–69798383) | 4325.31 (3667.44–5079.84) | 0.06 (−0.07–0.2) |
| High SDI | 42,393,966 (36509082–49627139) | 4544.37 (3905.75–5298.24) | 64,890,613 (55471021–75492082) | 5639.51 (4807.41–6672.65) | 0.18 (−0.01–0.36) |
| Low-middle SDI | 34,433,302 (28899675–41107705) | 3376.19 (2864.56–3983.67) | 78,736,534 (66913314–93335234) | 4110.28 (3502.94–4813.65) | 0.35 (0.22–0.48) |
| Low SDI | 13,933,696 (11509002–17010067) | 3300.55 (2770.42–3963.42) | 38,513,782 (31398707–47545721) | 3782.31 (3110.34–4583.98) | 0.2 (0.09–0.3) |
| Middle SDI | 61,553,503 (52828726–71975118) | 3710.46 (3222.16–4290.05) | 117,094,164 (100351714–136044666) | 4556.61 (3889.42–5294.11) | 0.22 (0.08–0.36) |
UI: uncertainty intervals; CI: confidence interval; SDI: socio-demographic index; EAPC: estimated annual percentage change
Based on comparative model performance, the Theta model was used to project age-standardized rates, whereas the ARIMA model was used to project absolute numbers from 2022 to 2050. For incidence, the age-standardized rate was projected to increase from 680.87 per 100,000 population in 2022 to 734.37 per 100,000 in 2050, corresponding to an increase of 7.86%. Over the same period, the number of incident cases was projected to rise from 51.44 to 68.06 million, representing an increase of 32.31%. For prevalence, the age-standardized rate was projected to increase from 4438.92 per 100,000 population in 2022 to 4788.22 per 100,000 in 2050, while the number of prevalent cases was projected to increase from 342.85 to 445.97 million, an increase of 30.08%. For DALYs, the age-standardized rate was projected to rise from 526.36 per 100,000 population in 2022 to 567.78 per 100,000 in 2050, whereas the total number of DALYs was projected to increase from 40.59 to 52.56 million, representing an increase of 29.49% (Fig. 5B, D).
Fig. 5.
Performance of forecasting models and projected trends in the burden of anxiety disorders from 2022 to 2050. A Radar plots comparing the performance of eight models for age-standardized incidence, prevalence, and DALY rates. B Projected age-standardized incidence, prevalence, and DALY rates from 2022 to 2050 using eight forecasting models. C Radar plots comparing the performance of eight models for the absolute numbers of incident cases, prevalent cases, and DALYs. D Projected numbers of incident cases, prevalent cases, and DALYs from 2022 to 2050 using eight forecasting models
Overall, both age-standardized rates and absolute numbers were projected to increase over the study period. The greater increases observed in absolute numbers than in age-standardized rates suggest that population growth and demographic change may contribute importantly to the future burden of anxiety disorders. These projections should be interpreted cautiously, as they are model-based estimates rather than observed future trends.
Discussion
Several previous studies have quantified the burden of anxiety disorders, although most have focused on individual countries or regions [15,16]. Only a limited number of studies have examined anxiety disorders at the global level [17,18]. Previous analyses based on GBD data showed that anxiety disorders contributed substantially to the global burden of disease and that women generally experienced a higher burden than men [17,18]. However, these studies were based on GBD 2019 data and did not evaluate associations with human resources for health (HRH) or project future trends. Using GBD 2021 data, the present study provides updated global estimates of anxiety disorder burden, characterizes long-term temporal trends, examines associations with healthcare workforce composition, and offers model-based projections of future burden. Taken together, these analyses extend previous descriptive epidemiological work and provide a broader framework for understanding anxiety disorders from both population health and health-system perspectives (Table 3).
Table 3.
The number of DALYs cases and the age-standardized DALYs rate attributable to anxiety disorders in 1990 and 2021, and its trends from 1990 to 2021 globally
| Characteristics | 1990 | 2021 | 1990–2021 | ||
|---|---|---|---|---|---|
| Number of DALYs cases(95%UI) | The age standardized DALYs rate/100,000(95%UI) | Number of DALYs cases(95%UI) | The age standardized DALYs rate/100,000(95%UI) | EAPC(95%CI) | |
| Global | 22,996,732 (15814116–31547276) | 443.73 (306.21–603.28) | 42,509,645 (29396723–57729820) | 524.33 (363.06–716.25) | 0.14 (0–0.27) |
| Sex | |||||
| Female | 14,236,491 (9856232–19377107) | 550.7 (382.99–745.49) | 26,395,988 (18390434–35710763) | 652.17 (453.44–887.15) | 0.11 (−0.03–0.26) |
| Male | 8,760,241 (6016981–12171231) | 335.85 (231.23–460.99) | 16,113,657 (11067291–22164952) | 396.91 (272.28–547.66) | 0.18 (0.05–0.32) |
| Age | |||||
| < 5 years | 66,350 (38912–101086) | 10.7 (6.28–16.31) | 92,783 (52877–145776) | 14.1 (8.03–22.15) | 0.39 (0.2–0.59) |
| 5–9 years | 942,648 (562013–1416463) | 161.54 (96.31–242.74) | 1,392,657 (810716–2131730) | 202.7 (118–310.27) | 0.24 (0.07–0.41) |
| 10–14 years | 2,238,635 (1350319–3327223) | 417.9 (252.07–621.12) | 3,358,105 (2013418–5072153) | 503.74 (302.03–760.86) | 0.12 (−0.04–0.28) |
| 15–19 years | 2,831,156 (1787407–4215876) | 545.06 (344.11–811.65) | 4,054,128 (2545178–6004786) | 649.72 (407.89–962.34) | 0.11 (−0.04–0.27) |
| 20–24 years | 2,752,731 (1738657–4102815) | 559.4 (353.32–833.76) | 4,111,484 (2580955–6110620) | 688.51 (432.21–1023.28) | 0.19 (0.02–0.36) |
| 25–29 years | 2,489,434 (1562418–3701749) | 562.43 (352.99–836.33) | 4,084,697 (2535558–6043162) | 694.27 (430.97–1027.15) | 0.23 (0.06–0.4) |
| 30–34 years | 2,178,950 (1402556–3152139) | 565.34 (363.9–817.84) | 4,133,466 (2671031–5951479) | 683.81 (441.87–984.56) | 0.22 (0.07–0.37) |
| 35–39 years | 1,975,554 (1297494–2802441) | 560.85 (368.35–795.6) | 3,841,611 (2521935–5437237) | 684.94 (449.65–969.44) | 0.24 (0.1–0.38) |
| 40–44 years | 1,624,225 (991051–2389155) | 566.96 (345.94–833.96) | 3,394,028 (2066561–4956306) | 678.47 (413.11–990.76) | 0.19 (0.05–0.34) |
| 45–49 years | 1,299,316 (809783–1912509) | 559.58 (348.75–823.66) | 3,064,476 (1907197–4471868) | 647.19 (402.78–944.42) | 0.12 (0–0.25) |
| 50–54 years | 1,157,361 (751158–1647252) | 544.46 (353.37–774.92) | 2,769,385 (1803914–3913933) | 622.44 (405.44–879.69) | 0.11 (−0.01–0.22) |
| 55–59 years | 988,321 (667102–1411065) | 533.65 (360.21–761.91) | 2,387,418 (1612834–3400518) | 603.3 (407.56–859.31) | 0.08 (−0.03–0.2) |
| 60–64 years | 842,752 (551854–1240341) | 524.72 (343.6–772.27) | 1,866,300 (1226816–2738308) | 583.13 (383.32–855.59) | 0 (−0.12–0.12) |
| 65–69 years | 642,651 (414025–953909) | 519.9 (334.95–771.71) | 1,525,959 (991153–2236585) | 553.2 (359.32–810.82) | −0.09 (−0.2–0.01) |
| 70–74 years | 422,209 (280273–609405) | 498.71 (331.05–719.82) | 1,084,363 (723663–1556565) | 526.8 (351.57–756.2) | −0.14 (−0.23–0.05) |
| 75–79 years | 299,429 (206647–428715) | 486.44 (335.71–696.47) | 659,446 (452470–929706) | 500.02 (343.08–704.94) | −0.15 (−0.24–0.07) |
| 80–84 years | 163,512 (112833–230020) | 462.21 (318.95–650.22) | 414,769 (281467–584195) | 473.57 (321.37–667.02) | −0.13 (−0.2–0.07) |
| 85–89 years | 63,195 (41213–89689) | 418.2 (272.73–593.53) | 194,484 (125462–275323) | 425.36 (274.4–602.17) | −0.11 (−0.17–0.04) |
| 90–94 years | 15,346 (9824–21532) | 358.12 (229.25–502.47) | 64,575 (41583–90619) | 360.97 (232.45–506.55) | −0.08 (−0.13–0.03) |
| 95 + years | 2956 (1862–4310) | 290.34 (182.89–423.34) | 15,511 (9839–22325) | 284.6 (180.53–409.61) | −0.15 (−0.19–0.11) |
| SDI region | |||||
| High SDI | 5,013,827 (3447647–6768039) | 540 (370.86–737.42) | 7,587,224 (5295258–10197786) | 668.99 (463.37–917.34) | 0.17 (−0.01–0.35) |
| High-middle SDI | 4,827,853 (3320397–6587790) | 441.91 (306.74–605.24) | 7,053,070 (4756266–9536400) | 517.28 (350.45–711.25) | 0.08 (−0.06–0.21) |
| Middle SDI | 7,391,037 (5124269–10205740) | 440.96 (304.66–599.62) | 13,889,164 (9600906–18924071) | 541.94 (374.72–741.61) | 0.22 (0.08–0.36) |
| Low-middle SDI | 4,089,572 (2796842–5650816) | 396.3 (274.46–536.18) | 9,340,395 (6459975–12701317) | 484.76 (337.96–655.59) | 0.37 (0.25–0.5) |
| Low SDI | 1,652,386 (1120557–2294992) | 386.5 (264.09–525.82) | 4,603,364 (3027207–6403847) | 446.36 (296.14–609.76) | 0.23 (0.13–0.34) |
UI: uncertainty intervals; CI: confidence interval; SDI: socio-demographic index; EAPC: estimated annual percentage change
Anxiety disorders are not only highly prevalent mental disorders but also important contributors to healthcare utilization and economic burden. Previous studies have shown that anxiety disorders are associated with increased medical expenditure, reduced productivity, and greater demand for healthcare services [19]. In addition, anxiety disorders frequently coexist with physical health conditions, which may further impair quality of life and amplify their overall health burden [20]. These observations place our findings in a broader public health context: the significance of anxiety disorders lies not only in the large number of individuals affected, but also in their wider consequences for healthcare systems and society. Accordingly, updated global evidence is important for informing prevention, improving early recognition, and supporting more effective allocation of mental health resources.
Our subgroup analyses showed a consistently higher burden of anxiety disorders among women than among men, in keeping with previous epidemiological studies [21, 22]. This concordance with prior evidence strengthens confidence in the sex-specific pattern observed in the present study. The underlying reasons are likely multifactorial. Previous research suggests that reporting patterns, social roles, emotional processing, physiological responses, and differences in help-seeking behavior may all contribute to the higher burden observed among women [23]. Broader social and economic inequalities may also be relevant [24, 25]. This finding has important implications for public health practice, as it suggests that prevention and service delivery strategies may need to account for sex-specific patterns of vulnerability rather than approaching anxiety disorders as a uniform condition across populations.
From 1990 to 2019, age-standardized incidence, prevalence, and DALY rates remained relatively stable, whereas all three measures increased after 2019. This increase was likely associated, at least in part, with the COVID-19 pandemic. In addition, our age-group analyses indicate that anxiety disorders impose a relatively high burden among adolescents and young adults. Previous studies have identified adolescence and early adulthood as critical periods for the onset of anxiety disorders, potentially influenced by both biological and psychosocial factors [26, 27]. Younger adults may also be more likely than older adults to recognize symptoms and seek professional help [28. Social stressors, including financial pressure, work-related stress, and family responsibilities, may further contribute to the elevated burden observed in this age group [29].Taken together, these findings suggest that anxiety disorder burden is sensitive to major societal disruption and that younger populations may represent a particularly important target for mental health promotion, screening, and early intervention [30, 31].
Several mechanisms may account for the increase in anxiety disorder burden after 2019. Fear of infection[ [32], pandemic-related uncertainty [33], economic disruption [34], and reduced access to healthcare [35] may all have contributed to worsening mental health during this period. Increased awareness of mental health symptoms and greater attention to psychiatric disorders during the pandemic may also have enhanced help-seeking and reporting, although over-reporting cannot be excluded [36]. Among healthcare workers, uncertainty regarding pandemic control and sleep-related problems have also been identified as important contributors to anxiety symptoms [37, 38]. These observations suggest that the post-2019 increase likely reflects both a real increase in psychological distress and changes in symptom recognition and reporting. More broadly, they underscore the importance of integrating mental health support into preparedness and response strategies for large-scale public health emergencies. At the same time, these interpretations should remain cautious, as the present study was not designed to establish causal pathways.
A distinctive contribution of the present study lies in its incorporation of HRH into the analysis of anxiety disorder burden. Our findings suggest that cross-national variation in the burden of anxiety disorders may be related, in part, to differences in healthcare workforce composition, although the observed associations were heterogeneous across workforce categories and countries. In particular, the negative correlation observed for traditional and complementary practitioners and the positive correlation observed for dieticians and nutritionists should be understood within the broader context of healthcare systems, cultural practices, service accessibility, and socioeconomic development, rather than as isolated explanatory factors. The value of this analysis lies not in identifying any single workforce category as a determinant of anxiety burden, but in introducing a systems-level perspective into the study of global mental health. The contrasting examples of Portugal and New Zealand further suggest that workforce density alone is insufficient to explain cross-national differences in anxiety disorder burden, and that broader structural and contextual determinants must also be considered. In this sense, the HRH analysis expands the interpretation of anxiety disorder burden beyond epidemiological description and provides a useful basis for future research and more context-sensitive mental health workforce planning.
Model-based projections suggested that the global burden of anxiety disorders may continue to increase from 2022 to 2050. By 2050, the number of incident cases was projected to reach 68.06 million, the number of prevalent cases 445.97 million, and the total number of DALYs 52.56 million. Notably, the increase in absolute numbers was greater than that in age-standardized rates, suggesting that population growth and demographic change may contribute substantially to the future burden of anxiety disorders. This distinction is important from a policy perspective. Even if changes in age-standardized rates remain relatively modest, the total number of affected individuals and the absolute health loss may still increase considerably, thereby placing growing pressure on healthcare systems and mental health services worldwide. Although these projections should be interpreted with caution because they are model-based rather than observed estimates, they nevertheless provide useful evidence for long-term mental health planning, workforce preparation, and resource allocation.
Several limitations should be considered when interpreting the findings of this study. First, the lack of high-quality civil registration and vital statistics systems in many low- and middle-income countries introduces uncertainty into population and disease burden estimates. Second, analyses of finer subpopulations, such as subnational areas, remain constrained by data availability [39]. Third, the analyses of healthcare workforce and anxiety disorder burden were ecological and correlational in nature and therefore do not permit causal inference. Fourth, the future burden estimates were derived from statistical forecasting models and should therefore be interpreted cautiously. Finally, the present study did not estimate the burden of specific anxiety disorder subtypes, such as panic disorder or social anxiety disorder [17]. Future research should investigate these subtypes in greater detail, examine contextual determinants more deeply, and support the development of more targeted prevention and treatment strategies across countries and regions. Despite these limitations, the present study provides updated and globally comparable evidence on the burden, temporal trends, health-system correlates, and projected future burden of anxiety disorders, thereby contributing to a more comprehensive understanding of this important public health challenge.
Conclusion
This study provides updated global evidence on the burden of anxiety disorders using GBD 2021 data. Anxiety disorders remain a substantial public health challenge worldwide, with an increasing burden observed after 2019 and further increases projected in the coming decades. By integrating epidemiological trends, healthcare workforce associations, and future projections, this study offers a broader perspective for understanding anxiety disorders and may help inform mental health policy, resource allocation, and long-term planning.
Acknowledgements
The authors gratefully acknowledge the Global Burden of Disease (GBD) 2021 collaborators for providing the publicly available data used in this study.
Abbreviations
- DALYs
Disability adjusted life years
- SDI
socio-demographicindex
- GBD
Global burden of disease study
- ASR
Age-standardized rate
- EAPC
Estimated annual percentage change
- UI
Uncertainty intervals
- CI
Confidence interval
Author contributions
Z.W. performed formal analysis, visualization, and wrote the original draft.Y.C. contributed to conceptualization, methodology, formal analysis, and writing— review and editing.X.Z. conducted investigation, provided resources, performed data curation, and formal analysis.J.W. contributed to data curation and wrote the original draft.B.L. contributed to formal analysis and methodology.L.L. contributed to formal analysis and wrote the original draft.S.L. contributed to investigation and methodology.J.S. contributed to conceptualization, methodology, and resources.Q.Z. contributed to project administration, validation, supervision, and funding acquisition.All authors reviewed and approved the final manuscript.
Funding
This research received no external funding.
Data availability
The data used in this study are publicly available from the Global Burden of Disease (GBD) 2021 study and its associated data sources.
Declarations
Ethical approval and consent to participate
This study was based exclusively on publicly available, de-identified data from the Global Burden of Disease (GBD) 2021 study. No individual-level identifiable information was used. Therefore, ethical approval was not required. Not applicable, because this study did not involve direct participation of human subjects.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Zihao Wang and Yuzhou Cai contributed equally to this work.
Qingguo Zhang and Jiaduo Sun contributed equally to this work.
Contributor Information
Jiaduo Sun, Email: 616285037@qq.com.
Qingguo Zhang, Email: zjyyzqg@126.com.
References
- 1.Szuhany KL, Simon NM. Anxiety Disorders: A Review. JAMA. 2022;328:2431–45. [DOI] [PubMed] [Google Scholar]
- 2.Alonso J, et al. Treatment gap for anxiety disorders is global: Results of the World Mental Health Surveys in 21 countries. Depress Anxiety. 2018;35:195–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Santomauro DF, et al. Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic. Lancet. 2021;398:1700–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.GBD2019MDC. Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry. 2022;9:137–50. [DOI] [PMC free article] [PubMed]
- 5.Mental Health. and COVID-19: Early evidence of the pandemic’s impact: Scientific brief, 2 March 2022. https://www.who.int/publications/i/item/WHO-2019-nCoV-Sci_Brief-Mental_health-2022.1
- 6.The global evolution of mental health. problems during the COVID-19 pandemic: A systematic review and meta-analysis of longitudinal studies. J Affect Disord. 2022;315:70–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Chai Y, et al. Incidence of mental health diagnoses during the COVID-19 pandemic: a multinational network study. Epidemiol Psychiatric Sci. 2024;33:e9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.World Health Organization. The impact of COVID-19 on mental, neurological and substance use services: results of a rapid assessment. Geneva: WHO; 2020. [Google Scholar]
- 9.World Health Organization. Mental health atlas 2024. Geneva: WHO; 2025. [Google Scholar]
- 10.Effect of the COVID-. 19 pandemic on mental health visits in primary care: an interrupted time series analysis from nine INTRePID countries. eClinicalMedicine. 2024;70:102533. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Chen Y, et al. The Vulnerability Paradox in Global Mental Health: Psychologist Density, Social Development, and the Disease Burden of Mental Disorder. Int J Ment Health Addict. 2025;23:3975–87. [Google Scholar]
- 12.Schumacher AE, et al. Global age-sex-specific mortality, life expectancy, and population estimates in 204 countries and territories and 811 subnational locations, 1950–2021, and the impact of the COVID-19 pandemic: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. Lancet. 2024;403:1989–2056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Vollset SE, et al. Fertility, mortality, migration, and population scenarios for 195 countries and territories from 2017 to 2100: a forecasting analysis for the Global Burden of Disease Study. Lancet. 2020;396:1285–306. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.GBD 2019 Human Resources for Health Collaborators. Measuring the availability of human resources for health and its relationship to universal health coverage for 204 countries and territories from 1990 to 2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2022;399:2129–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Bereza BG, et al. A Markov Model Approach Assessing the Cost of Illness of Generalized Anxiety Disorder in Canada. Neurol Ther. 2012;1:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.DiBonaventura M, Toghanian S, Järbrink K, Locklear J. Economic and humanistic burden of illness in generalized anxiety disorder: an analysis of patient survey data in Europe. CEOR. 2014. 10.2147/CEOR.S55429 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Yang X, et al. Global, regional and national burden of anxiety disorders from 1990 to 2019: results from the Global Burden of Disease Study 2019. Epidemiol Psychiatr Sci. 2021;30:e36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Baxter AJ, Vos T, Scott KM, Ferrari AJ, Whiteford HA. The global burden of anxiety disorders in 2010. Psychol Med. 2014;44:2363–74. [DOI] [PubMed] [Google Scholar]
- 19.Marciniak MD, et al. The cost of treating anxiety: the medical and demographic correlates that impact total medical costs. Depress Anxiety. 2005;21:178–84. [DOI] [PubMed] [Google Scholar]
- 20.Roy-Byrne PP, et al. Anxiety disorders and comorbid medical illness. Gen Hosp Psychiatry. 2008;30:208–25. [DOI] [PubMed] [Google Scholar]
- 21.McLean CP, Asnaani A, Litz BT, Hofmann SG. Gender differences in anxiety disorders: prevalence, course of illness, comorbidity and burden of illness. J Psychiatr Res. 2011;45:1027–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bandelow B, Michaelis S. Epidemiology of anxiety disorders in the 21st century. Dialogues Clin Neurosci. 2015;17:327–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Asher M, Asnaani A, Aderka IM. Gender differences in social anxiety disorder: A review. Clin Psychol Rev. 2017;56:1–12. [DOI] [PubMed] [Google Scholar]
- 24.Lun J, Kesebir S, Oishi S. On Feeling Understood and Feeling Well: The Role of Interdependence. J Res Pers. 2008;42:1623–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Klipfel KM, Sweet JJ, Nelson NW, Moberg PJ. Gender and ethnic/racial diversity in clinical neuropsychology: Updates from the AACN, NAN, SCN 2020 practice and ‘salary survey’. Clin Neuropsychol. 2023;37:231–85. [DOI] [PubMed] [Google Scholar]
- 26.Narmandakh A, Roest AM, de Jonge P, Oldehinkel AJ. Psychosocial and biological risk factors of anxiety disorders in adolescents: a TRAILS report. Eur Child Adolesc Psychiatry. 2021;30:1969–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.de Lijster JM, et al. The Age of Onset of Anxiety Disorders. Can J Psychiatry. 2017;62:237–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hadjimina E, Furnham A. Influence of age and gender on mental health literacy of anxiety disorders. Psychiatry Res. 2017;251:8–13. [DOI] [PubMed] [Google Scholar]
- 29.Gondek D, Bernardi L, McElroy E, Comolli CL. Why do Middle-Aged Adults Report Worse Mental Health and Wellbeing than Younger Adults? An Exploratory Network Analysis of the Swiss Household Panel Data. Appl Res Qual Life. 2024. 10.1007/s11482-024-10274-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mayo-Wilson E, et al. Psychological and pharmacological interventions for social anxiety disorder in adults: a systematic review and network meta-analysis. Lancet Psychiatry. 2014;1:368–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Pilling S et al. Recognition, assessment and treatment of social anxiety disorder: summary of NICE guidance. BMJ. 2013; 346. [DOI] [PubMed]
- 32.Puia A, et al. Coping Strategies Among Healthcare Workers During the COVID-19 Pandemic: Emotional Responses, Challenges, and Adaptive Practices. Med (Kaunas). 2025;61:311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Moreno C, et al. How mental health care should change as a consequence of the COVID-19 pandemic. Lancet Psychiatry. 2020;7:813–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Gong Y et al. COVID-19 Induced Economic Slowdown and Mental Health Issues. Front Psychol. 2022; 13 [DOI] [PMC free article] [PubMed]
- 35.Challenges facing mental. health systems arising from the COVID-19 pandemic: Evidence from 14 European and North American countries. Health Policy. 2023;136:104878. [DOI] [PubMed] [Google Scholar]
- 36.Wu T, et al. Prevalence of mental health problems during the COVID-19 pandemic: A systematic review and meta-analysis. J Affect Disord. 2021;281:91–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Omrane A, et al. Quality of sleep among healthcare workers treating patients with coronavirus disease-19. Eur Psychiatry. 2023;66:S777–777. [Google Scholar]
- 38.Parchani A, et al. Fear, Anxiety, Stress, and Depression of Novel Coronavirus (COVID-19) Pandemic Among Patients and Their Healthcare Workers – A Descriptive Study. PRBM. 2021;14:1737–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.GBD 2021 Demographics Collaborators. Global age-sex-specific mortality, life expectancy, and population estimates in 204 countries and territories and 811 subnational locations, 1950–2021, and the impact of the COVID-19 pandemic: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. Lancet. 2024;403:1989–2056. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data used in this study are publicly available from the Global Burden of Disease (GBD) 2021 study and its associated data sources.