Global, regional and national disparities and temporal trends of common autoimmune disease burdens among children and adolescents from 1990 to 2019

Abstract

Introduction

Previous evidence lacked a thorough review of the disparities of autoimmune diseases (AD) burdens among countries and regions, which led to an insufficient basis for developing country-specific developmental level relevant preventive measures. This study aimed to analyse disparities and trends of global, regional and national burden of common ADs in children and adolescents from 1990 to 2019 and to investigate the associations between specific ADs and varied country indexes.

Methods

All data for four major ADs were obtained from the Global Burden of Diseases Study 2019. Age period-cohort modelling was conducted to disentangle age, period and birth cohort effects on AD incidence from 1990 to 2019. Local regression smoothing models were used to fit the correlation between AD burdens and sociodemographic index (SDI). Pearson’s correlation was used to investigate varied country-level risk factors for disease burden.

Results

A global increase in four common ADs incidence was observed from 1.57 million to 1.63 million between 1990 and 2019 in the 0–24 age group. The age-standardised incidence rate of overall four ADs showed substantial regional and global variation with the highest incidence in high SDI regions. The age, period and cohort distributions of AD incidence varied significantly, especially in high SDI countries. Relative to the expected level of age-standardised incidence associated with SDI, the distribution varied by regions depending on the specific ADs. Countries with higher levels of socioeconomic development, better quality of life and easier access to healthcare and the healthcare system showed lower disease burdens of ADs.

Conclusions

The incidence patterns and disease burdens of ADs varied considerably according to age, time period and generational cohort, across the world between 1990 and 2019. Incidences of ADs in children and adolescents were significantly correlated with indexes involving risks of the environment, human rights and health safety and quality of life.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Previous studies have investigated the epidemiological trends of autoimmune diseases (ADs) during adolescence, but there is a lack of research on different types of ADs in children and adolescents.

WHAT THIS STUDY ADDS

• This study further analysed disparities and trends of global, regional and national burden of common ADs between children and adolescents from 1990 to 2019 and to investigate the association between specific ADs and varied country indexes.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Based on the specific developmental status, as well as the cultural and regional characteristics of each country, the combined enhancement of the socioeconomic development of countries, the quality of life of the population and the accessibility of healthcare services contribute to reducing unfavourable increasing trends of AD incidence and burden in children and adolescents at different age stages.

Introduction

Autoimmune diseases (ADs) encompass a clinically heterogeneous group of disorders characterised by multisystem involvement and chronic inflammation. These diseases are caused by the immune system attacking the body’s own tissues and organs due to a dysregulation in recognising self from non-self.1,3 A large population-based study has estimated that up to approximately 10% of the population was impacted by ADs during the study period (from 2000 to 2019)⁴ and the incidence is expected to keep increasing worldwide.⁵ Although significant progress has been made in the treatment of ADs, the health and economic burden remains tremendous. Previous work has shown that most ADs affect younger people⁶ with a potential rise trend in the future,⁷ and that these diseases are one of the leading causes of death in young women.⁸ Therefore, a comprehensive assessment of the prevalence of ADs in children and adolescents is important for a better understanding of the epidemiology of these diseases and for the development of public health policies.

ADs may occur at any age, but the age of onset is not the same for different diseases. Diabetes mellitus type 1 (DM1), one of the most common ADs, continues to be the predominant type of diabetes in children and adolescents,⁹ often developing in the 6–13 years age group. An updated report from the DIAMOND project has examined the trends in the increased incidence of DM1 among children aged 14 years or younger.¹⁰ It was assessed that the cost of treating diabetes mellitus is at least 3.2 times the per capita health expenditure,¹¹ indicating an increasing economic burden on children and adolescents with DM1. Besides, the disease burden of paediatric-onset inflammatory bowel disease (IBD), another type of ADs, continues to grow rapidly as well.¹² Similar to IBD, psoriasis is a common chronic immune-mediated inflammatory disorder, which begins in childhood in almost one-third of the cases. Previous studies have estimated that about 30%–50% of adults with psoriasis develop psoriasis before 20 years old.¹³ In addition, individuals younger than 18 years suffering from psoriasis have a higher prevalence of certain comorbidities, such as diabetes mellitus and rheumatoid arthritis (RA).¹⁴ It has been reported that RA has an increasing burden of disease to varying degrees in different countries in the 16+ age group.¹⁵ Here, we focus primarily on four major ADs in children and adolescents, including DM1, RA, IBD and psoriasis. Studying those ADs is relevant because of their common occurrence and the emergence of new drugs and therapeutic strategies in recent decades.

Although many studies have been published on the prevalence and associated factors of the above-mentioned four ADs, both globally or in specific regions,1014,16 the comprehensive analysis of the total incidence distribution of four ADs, along with the similarities and disparities among countries and regions remains insufficiently examined among children and adolescents. The Global Burden of Diseases (GBD), Injuries and Risk Factors Study offers a systematic approach to evaluate the burden of ADs in children and adolescents across 21 regions and 204 countries and territories, presenting a unique opportunity to comprehend the underlying trends over the last three decades. Data from large epidemiological studies worldwide indicate that sociodemographic factors may predict ADs,^{17 18} including economic status and so on. Moreover, regional differences in ADs among children and adolescents are equally likely to stem from differences between countries, including healthcare resources and the national environment as a whole. The Environmental, Social and Governance (ESG) index is an efficient tool to quickly assess specific risks in relation to the environment, human rights and health and safety, including three subindexes that operate independently of the others.¹⁹ The quality of life (QoL) index serves as a composite measure of overall well-being and life satisfaction in a country, with higher scores indicating a better QoL.²⁰ This estimation is derived from an empirical formula that incorporates several key factors that significantly impact QoL, including purchasing power index, pollution index, house price to income ratio, cost of living index, safety index, healthcare index, traffic commute time index and climate index.²⁰ Recent studies have pointed out that higher burdens of ADs may occur in regions or countries with higher levels of sociodemographic development,²¹ suggesting that epidemiological patterns of ADs are worth exploring. To shed light on differences in the disease burden of ADs in children and adolescents in different countries, it is necessary to investigate the interactions between the ESG index, QoL index and ADs.

Given the evolving healthcare needs of children and adolescents with ADs over their lifespan and medical advancements, this study aimed to (1) estimate the overall pattern and specific trends in AD incidence, disability and disease burden in children and adolescents, (2) identify global, regional and national trends in the incidence of specific ADs from 1990 to 2019, (3) distinguish the effects of age, period and birth cohort on the incidence of specific ADs from 1990 to 2019 and stratified by socioeconomic status and (4) determine the association between varied country indexes and the incidence of specific AD incidence worldwide in 2019.

Methods

Data sources

The GBD 2019 study is a comprehensive epidemiological dataset and is developed to offer global health estimates for 369 causes of death and injuries for 204 countries and territories from 1990 to 2019. Broader information relating to the methods for GBD 2019 has been detailed in previous studies, covering main data sources, disease modelling approaches and definitions of disease outcomes.^{22 23} We reviewed all causes included in GBD 2019 and their corresponding International Classification of Disease codes to identify four major AD in children and adolescents. DM1, RA, IBD and psoriasis were included within our definition of major AD in children and adolescents.

Data extraction

We used the Global Health Data Exchange query tool online to access data for the four investigated ADs. We extracted absolute numbers, rates and age-standardised rates per 100 000 population from 1990 to 2019 for the following metrics: incidence; disability (years lived with disability (YLDs)); and disease burden (measured as disability-adjusted life-years (DALYs)). Age-standardised rate (age-standardised incidence rate (ASIR) and age-standardised DALY rate (ASDR)) was collected according to related rates in the age group of children and adolescents.

Sociodemographic development status data

The sociodemographic index (SDI), a composite index of sociodemographic development status, is calculated as the geometric mean of 0–1 indices on the following metrics: total fertility rate among individuals under 25 years old, the average education level for those aged 15 years and above, and lag-distributed income per capita. The GBD system accessed SDI data to categorise 204 countries and territories into 5 regions at similar levels of sociodemographic development (low (SDI <0.46), low-middle (SDI: 0.46–0.60), middle (SDI: 0.61–0.69), high-middle (SDI: 0.70–0.81) and high (SDI>0.81)).²⁴ In this study, we extracted SDI and SDI classification for each country and region from the GBD online system.²⁴

The ESG index, healthcare index and QoL index

We obtained the ESG index,¹⁹ healthcare index²⁵ and QoL index²⁰ online. The healthcare index is a comprehensive index that reflects a country’s overall level of healthcare services, encompassing factors such as the skill and competency of medical staff, speed in completing examinations and reports, equipment for modern diagnosis and treatment, accuracy and completeness in filling out reports, friendliness and courtesy of the staff, responsiveness (waiting times) in medical institutions and the convenience of location for patients.²⁵ Further detailed information can be found online: https://www.numbeo.com/quality-of-life/rankings_by_country.jsp.

Statistical analysis

This analysis focuses on exploring patterns and temporal trends in AD during the developmental window of childhood and adolescence. We defined childhood as being younger than 10 years, while adolescence is considered to span from 10 to 24 years of age,26,30 in line with updated insights into neurodevelopment and evolving global patterns in the timing of significant social role changes (eg, completion of education and parenthood).³⁰

Estimated annual percentage changes analysis

We conducted estimated annual percentage changes (EAPC) analysis to estimate the ASIR trend over a specific time period in children and adolescents. EAPC is a widely accepted quantitative indicator that adopts a linear regression line to fit the natural logarithm of the ASIR trend with a logarithmic linear regression model. A regression model was set up to depict the connection between the natural logarithm (ln) of ASIR and time, represented as y=b₀+βx+c, y=ln (ASIR). In this equation, x=the calendar year, b₀ represents the constant term, c stands for the false term and signifies the interpretation of the negative or positive trend of the chosen age-standardised rate. The EAPC was determined using the formula: EAPC=100×(exp[β]−1). The 95% CI for the EAPC was derived from the linear regression model. The ASIR shows an increasing trend when the lower limits of an EAPC 95% CI exceed 0. Conversely, the ASIR demonstrates a decreasing trend when the upper limits of an EAPC 95% CI are below 0.

Age-period-cohort analysis

To explore hidden information about age-adjusted incidence, age-period-cohort (APC) analysis is widely used in long-term epidemiological trend studies to focus simultaneously on three time-related changes: age, period and cohort effects.^{31 32} The APC model is a statistical framework that is extensively used in health and social science fields, especially in epidemiological studies of non-communicable diseases,³³ to disentangle the effects of age, period and birth cohort, for instance, on the incidence of a particular disease. We carried out this analysis with an online analysis tool (https://analysistools.cancer.gov/apc/).³⁴ We applied the APC model to distinguish the age, period and birth cohort effects on AD incidence across socioeconomic status. We aggregated the data on incidence and population from six regions based on SDI (global, low SDI, low-middle SDI, middle SDI, high-middle SDI and high SDI regions) and organised them into six successive 5-year intervals spanning from 1990 to 2019. For age effects, we established 5 consecutive 5-year age intervals spanning from 0–5 to 20–24 years. Concerning cohort effects, data from 5 successive cohorts, encompassing individuals born from 1965 to 2019, were analysed over the 30-year study period.

The annual percentage change in disease burden, identified as the general temporal trend or ‘net drift’ by the APC model, embodies the combined impact of time progression and successive birth cohorts.³⁴ Conversely, the age-specific temporal trend, termed the ‘local drift’, delineates the annual percentage change within individual age groups.³⁴ Even a subtle yearly percentage change, denoting a minor drift value, can signal significant shifts in disease prevalence over a 30-year time frame. To evaluate the statistical significance of these trends, we employed the Wald χ² test. Age effects in the APC model manifest through age-specific rates across various birth cohorts, adjusted for period influences.³¹ Meanwhile, period and cohort effects are illustrated as relative risks (RRs) of disease burden, comparing age-specific rates across different periods or cohorts to an arbitrarily chosen reference point, which does not impact the interpretability of the results.³⁴

The associations between SDI and varied country indexes and ASIR

In the GBD system, a total of 204 countries and territories were separated into 21 regions in terms of geography.²⁴ To investigate the factors influencing ASIR in children and adolescents, we first conducted local regression smoothing models to fit the correlation between ASIR and SDI for 21 regions from 1990 to 2019. Additionally, we employed Pearson’s correlation to assess the associations between ESG index, QoL index, healthcare index and ASIR, respectively. All statistical analyses were performed using R software (V.4.3.2). A significance level of p<0.05, at a two-tailed level, was used to determine statistical significance.

Patient and public involvement

Patients and the public were not involved in this study.

Results

The overall pattern and temporal trends of the incidence of four ADs in children and adolescents

Total AD incidences: In 2019, there were about 1.63 million incidence cases from ADs globally among childhood and adolescents (aged 0–24 years) (table 1). Compared with males, females have modestly higher incidence (54.1 incidence per 100 000 population) (table 1). Total AD incidence cases were predominantly high in countries of low-middle sociodemographic development (0.39 million; online supplemental table S1a) and the 20–24 age group (0.79 million; online supplemental table S1b), respectively. Globally, the most commonly occurring ADs were psoriasis, followed by DM1, RA and IBD (online supplemental figure S1).

Table 1. Age-specific and sex-specific estimates of incident cases, YLDs, YLLs and DALYs for overall four autoimmune diseases in 2019.

	Population	Incident cases		Disability (YLDs)		Life lost (YLLs)		Disease burden (DALYs)
		Number of incident cases	Incidence/100 000	Number of YLDs	YLDs/100 000	Number of YLLs	YLLs/100 000	Number of DALYs	DALYs/100 000
Under 5 years
Female	320 443 934	120 320	37.5	24 926	7.8	133 200	41.6	158 126	49.3
Male	342 398 748	120 429	35.2	25 087	7.3	164 855	48.1	189 942	55.5
5–9 years
Female	316 754 487	169 825	53.6	67 436	21.3	65 980	20.8	133 416	42.1
Male	337 949 216	158 717	47.0	65 561	19.4	52 439	15.5	118 000	34.9
10–14 years
Female	310 852 519	182 211	58.6	105 442	33.9	67 889	21.8	173 331	55.8
Male	331 334 185	159 319	48.1	96 504	29.1	52 328	15.8	148 832	44.9
15–19 years
Female	301 758 869	181 898	60.3	142 316	47.2	71 678	23.8	213 994	70.9
Male	317 782 109	159 847	50.3	121 732	38.3	74 781	23.5	196 513	61.8
20–24 years
Female	295 776 247	202 676	61.6	184 379	62.3	89 812	30.4	274 191	92.7
Male	304 368 218	191 912	63.1	152 585	50.1	134 025	44.0	286 610	94.2
0–24 years
Female	1 545 586 057	836 465	54.1	524 499	33.9	428 559	27.7	953 058	61.7
Male	1 633 832 475	790 224	48.4	461 469	28.2	313 573	19.2	775 042	47.4
Female and male	3 179 418 532	1 626 689	51.2	985 968	31.0	742 132	23.3	1 728 100	54.4

DALYs, disability-adjusted life-yearsYLDs, years lived with disability; YLLs, years of life lost

DM1: A global increase in the proportion of the number of DM1 incident cases out of the total incidence of ADs was found from 1990 to 2019 in the 0–24 age group. The largest increase in the proportion was observed in middle SDI regions (online supplemental figure S1). The global incidence rate of DM1 in 2019 was highest in children aged 5–9 years (figure 1 and online supplemental table S1b). Globally, the ASIR of DM1 per 100 000 adolescents increased by 1.20% for the 10–24 age group and 0.69% for the 0–9 age group from 1990 to 2019 (online supplemental table S2). Among 204 countries and territories, France was the country with the largest EAPC of DM1 in the 0–9 age group (EAPC=3.39%, 95% CI: 3.15% to 3.62%) and Greece has the largest EAPC in the 10–24 age group (EAPC=4.16%, 95% CI: 3.72% to 4.60%) (online supplemental table S3).

Figure 1. The rate of incidence (A), YLDs (B) and DALYs (C) attributable to autoimmune diseases by age, sex and SDI throughout human lifespan in 2019. The turning point represents a change in the disease burden across age, sex and SDI groups. DALYs, disability-adjusted life-years; SDI, sociodemographic index; YLDs, years lived with disability.

RA: The number of RA incident cases in children and adolescents accounted for a relatively small proportion of the total incidence of ADs in 2019 (not exceeding 11.3%), but it has increased from 1990 to 2019. The global incidence rate of RA in 2019 was highest in the 20–24 age group (figure 1 and online supplemental table S1b). In the 15–19 and 20–24 age groups, the Central Latin America region has the highest number of incident cases in 1990 and 2019. For RA, countries with the greatest rise in EAPC in children and adolescents were concentrated in the low SDI and low-middle SDI regions, such as the Democratic People’s Republic of Korea (the 0–9 age group) and Bhutan (the 10–24 age group) (online supplemental tables S2 and S3).

IBD: The proportion of the global number of IBD incident cases in children and adolescents increased slightly from 1990 to 2019. The global incidence rate of IBD in 2019 was highest in the 20–24 age group (figure 1 and online supplemental table S1b). For those aged 15–19 and 20–24 years, the high-income Asia Pacific region showed a more significant increase (online supplemental figure S1). Among 204 countries and territories, Taiwan (Province of China) showed the largest EAPC for IBD in the 0–9 (EAPC=2.48%, 95 % CI: 2.01% to 2.95%) and 10–24 age groups (EAPC=4.59%, 95 % CI: 4.1% to 5.08%) (online supplemental table S3).

Psoriasis: A global decrease in the proportion of the number of psoriasis incident cases out of the total incidence of ADs was found from 1990 to 2019 in the 0–24 age group. For those aged 0–4 years, the smallest decrease in the proportion was observed in high SDI regions (from 83.7% to 80.3%), while the decline was the smallest in low SDI regions in other age groups (online supplemental figure S3). Globally, the incidence rate of psoriasis in 2019 was highest in the 20–24 age group (figure 1 and online supplemental table S1b). The North Africa and Middle East region has the largest decrease in the proportion of psoriasis incidence cases in those aged 10–14 years old (from 76.2% to 56.4%; online supplemental figure S1). In addition, there was the largest decrease in the high-income Asia Pacific region, with 19.8% in the 15–19 age group and 22.1% in the 20–24 age group (online supplemental figure S1). Among 204 countries and territories, the country with the greatest decline in EAPC in children and adolescents was in Equatorial Guinea of the middle SDI regions (the 0–9 age group: EAPC=−1.90%, 95 % CI: −1.97% to −1.82%; the 10–24 age group: EAPC=−1.84%, 95 % CI: −1.90% to −1.78%; online supplemental tables S2 and 3).

More details are available in online supplemental materials A—Results.

The overall pattern of disability and disease burden for four ADs in children and adolescents

Total AD disability and disease burden: Globally, there were about 0.99 million years of healthy life lost to disability measured by YLDs and 1.73 million years of total healthy life lost measured by DALYs in 2019 from ADs in the 0–24 age group (table 1). Females have modestly higher disease burden (table 1). There was significantly higher total AD burdens in countries of low-middle sociodemographic development and 20–24 years among children and adolescents, with 0.23 million YLDs and 0.52 million DALYs (online supplemental tables S4a and S5a) and 0.34 million YLDs and 0.56 million DALYs (online supplemental tables S4b and S5b), respectively.

DM1: For the DALYs rate of DM1 (figures1 2A1, A2), there was a significant decrease in middle SDI and high-middle SDI countries in the 10–24 age group, such as Colombia, Vietnam, Greece, Croatia and Belarus, compared with 0–9 years. However, low SDI (eg, Yemen) and high SDI countries (eg, Iceland and France) showed an increasing trend in the DALY rate in DM1 between 1990 and 2019 in the 10–24 years age group (figures1 2A2). The top 5 decreasing EAPC countries with the highest ASDRs of DM1 in 2019 were Uruguay (the 0–9 age group: ASDR=5.2, 95% UI: 3.57 to 7.52) and Burundi (the 10–24 age group: ASDR=47.73, 95% UI: 34.56 to 65.59) (online supplemental table S5)

Figure 2. The global map of DALY rate attributable to four autoimmune diseases in 204 countries and territories for both sexes combined in 1990 and 2019. DALYs, disability-adjusted life-years.

RA: For RA (figures1 2B2), there was a certain degree of upward trends of burden in some countries among those aged 10–24 years, such as Kazakhstan, Iran, Mauritania, Mali, Ghana and Botswana. The top 5 decreasing EAPC countries with highest ASDRs of RA in 2019 were Russian Federation (the 0–9 age group: ASDR=0.24, 95% UI: 0.18 to 0.32) and Italy (the 10–24 age group: ASDR=6.54, 95% UI: 4.36 to 9.28) (online supplemental table S5)

IBD: For IBD, the disease burden reduced in the 0–9 years age group globally (figures1 2C1), while DALY rates increased in middle SDI (eg, Peru) and high SDI (eg, Australia) countries for the 10–24 age group (figures1 2C2). The top 5 decreasing EAPC countries with highest ASDRs of IBD in 2019 were the Netherlands (the 0–9 age group: ASDR=3.92, 95% UI: 1.98 to 5.36) and Panama (the 10–24 age group: ASDR=8.96, 95% UI: 5.94 to 12.28) (online supplemental table S5)

Psoriasis: For psoriasis, although most countries showed stable trends in the DALY rates of psoriasis in the 0–9 age group (figure 2D1), decreases in burden were seen in several countries, such as China, Mongolia, Kazakhstan, Turkey, Colombia, Australia and Thailand. To the contrary, there was a significant increase trend of DALY rate in Australian aged 10–24 years old (figure 2D2). The top 5 decreasing EAPC countries with highest ASDRs of IBD in 2019 were Palestine (the 0–9 age group: ASDR=5.7, 95% UI: 3.47 to 8.78) and Bosnia and Herzegovina (the 10–24 age group: ASDR=30.67, 95% UI: 20.32 to 42.85) (online supplemental table S3)

A more detailed characterisation of the results is stored in online supplemental materials A—Results.

The age-period-cohort analysis of the incidence rate of four ADs and comparison of effect similarities and disparities among children and adolescents from 1990 to 2019

Online supplemental figure S2 presents the APC model-derived estimates of age-period-cohort effects across different SDI regions. In terms of DM1, compared with other countries, high-SDI countries showed an overall higher incidence across all age groups, and the incidence was lowest in those aged 15–19 years. For the same birth cohort, the incidence rate of IBD and RA showed similar increasing trends with age globally, with the highest risk in the 20–24 age group (online supplemental figure S3–’Age effects’). As for psoriasis, the decline in incidence in the high SDI area was more pronounced in the 5–9 age group, with the lowest incidence being in the 15–19 age group.

Period effects for DM1, IBD and RA generally showed an increasing risk of incidence but a declining risk of incidence for psoriasis across different SDI regions over the study period (online supplemental figure S3—‘Period effects’). However, period RRs for IBD trended downwards over the past two decades globally, indicating decreasing incidence. In high-SDI countries, the period RRs for psoriasis trended slowly downwards in the past 20 years.

The upward trend in cohort RRs for DM1 and RA was weaker among those born from 1990 to 2019 than among those born in the previous 20 years, while for psoriasis, they trended downwards slowly in high and high-middle SDI countries (online supplemental figure S3—‘Cohort effects’). Increasing cohort effects for DM1 and RA were more pronounced in higher-SDI countries. High-SDI countries had progressively increasing incidence of DM1 in those born after the 1990s, whereas the risk in low-SDI countries remained nearly constant over the past two decades.

Onlinesupplemental figure S4 tables S6 show the net drift and local drift across SDI regions, age groups and causes. The interpretation of the results is presented in online supplemental materials A−Results.

The association between varied country indexes and total AD incidence in 2019

Figure 3 shows the observed regional ASIRs in relation to SDI versus the expected level for each region on the basis of SDI. In the 0–9 years age group, Central Asia, Southern Sub-Saharan Africa, Andean Latin America and Australasia closely followed expected trends over the study period in DM1. However, in many middle SDI regions, such as East Asia and Southeast Asia, the observed patterns varied and remained below expected levels throughout the study period, with age-standardised rates showing little change. Regions above expected levels demonstrated fluctuating or decreasing ASIR.

Figure 3. Age-standardised rate of incidence attributable to autoimmune disease per 100 000 persons for sociodemographic index by 21 regions among children and adolescents, 1990–2019. The black line represents expected values based on socio-demographic index and disease rates across 21 global burden of disease regions; each point shows the observed age standardised rate of incidence for specified global burden of disease region from 1990 to 2019.

Table 2 shows the associations between varied country indexes and the ASIR and ASDR of each AD, separately for children and adolescents. For overall ADs, ASIRs of children and adolescents were significantly negatively correlated with the ESG Index, positively correlated with the healthcare index and positively correlated with the QoL index (more in online supplemental materials A—Results 4).

Table 2. Associations between age-standardised incidence rate and DALY rate of autoimmune disease and varied country indexes in children and adolescents.

	Childhood (0–9 years)		Adolescence (10–24 years)
	rASIR (95% CI)	rASDR (95% CI)	rASIR (95% CI)	rASDR (95% CI)
N=171
ESG Index
Diabetes mellitus type 1	−0.61 (−0.70 to −0.51)	0.66 (0.56 to 0.73)	−0.54 (−0.64 to −0.42)	0.33 (0.19 to 0.46)
Rheumatoid arthritis	−0.62 (−0.71 to −0.52)	−0.29 (−0.42 to −0.14)	−0.57 (−0.67 to −0.46)	−0.64 (−0.72 to −0.54)
Inflammatory bowel disease	−0.45 (−0.57 to −0.33)	0.26 (0.11 to 0.39)	−0.66 (−0.73 to −0.56)	−0.57 (−0.66 to −0.45)
Psoriasis	−0.62 (−0.71 to −0.52)	−0.62 (−0.70 to −0.51)	−0.65 (−0.73 to −0.55)	−0.65 (−0.73 to −0.55)
N=85
Healthcare index
Diabetes mellitus type 1	0.37 (0.17 to 0.54)	−0.36 (−0.53 to −0.16)	0.24 (0.03 to 0.43)	−0.34 (−0.51 to −0.13)
Rheumatoid arthritis	0.34 (0.14 to 0.51)	0.29 (0. 08 to 0.47)	0.07 (−0.14 to 0.28)	0.23 (0.02 to 0.42)
Inflammatory bowel disease	0.24 (0.03 to 0.44)	−0.16 (−0.36 to 0.06)	0.42 (0.23 to 0.58)	0.33 (0.12 to 0.50)
Psoriasis	0.36 (0.16 to 0.53)	0.37 (0.17 to 0.54)	0.33 (0.12 to 0.50)	0.35 (0.14 to 0.52)
N=77
Quality of Life Index
Diabetes mellitus type 1	0.64 (0.49 to 0.76)	−0.38 (−0.55 to −0.17)	0.65 (0.50 to 0.77)	−0.14 (−0.35 to 0.08)
Rheumatoid arthritis	0.53 (0.35 to 0.68)	0.26 (0.03 to 0.45)	0.24 (0.01 to 0.44)	0.40 (0.20 to 0.57)
Inflammatory bowel disease	0.37 (0.16 to 0.55)	−0.14 (−0.35 to 0.09)	0.65 (0.49 to 0.76)	0.57 (0.40 to 0.70)
Psoriasis	0.59 (0.42 to 0.72)	0.58 (0.41 to 0.71)	0.60 (0.44 to 0.73)	0.60 (0.44 to 0.73)

ASDRage-standardised DALYs rateASIRage-standardised incidence rateDALYsdisability-adjusted life-yearsESGEnvironmental, Social and Governance Index

Discussion

Based on the worldwide data extracted from GBD 2019, this study provides a comprehensive estimation of the incidence, disability and disease burden of overall ADs and investigates their temporal trend for four specific ADs in children and adolescents.

Our results showed that the most commonly occurring ADs worldwide were psoriasis in children and adolescents. Previous research reported that psoriasis begins in childhood in almost one-third of the cases,^{14 16} and psoriatic skin lesions typically exhibit a chronic pattern of relapse and remission.¹⁶ Compared with children without psoriasis, those suffering from the condition exhibit a heightened incidence of comorbidities, encompassing obesity, diabetes mellitus, hypertension, RA, etc.^{14 35} Meanwhile, psoriasis has more triggering factors than other types of ADs, such as psychosocial stress, infections and being obese or overweight, which could be potential explanations for this phenomenon as well.¹⁶ We also found there was an obvious increase in the psoriasis burden in Australian aged 10–24 years old, compared with other regions. A systematic review included 54 studies and revealed that there seems to be a certain relationship between the burden of psoriasis and a country’s distance from the equator.³⁶ The exact reasons for geographical variation are still unclear, but it is likely driven by a combination of factors, including climate, genetics and environmental influences such as antigen exposure, sunlight exposure and vitamin D levels.^{36 37} In addition, the treatment of psoriasis continues to pose a significant challenge due to the scarcity of guidelines and the fact that most systemic treatments have not been approved for children and adolescents,^{38 39} which may help explain the stably high disease burden observed in this study from 1990 to 2019. It underscores the urgent need for effective prevention strategies and treatment guidelines for children and adolescents to reduce the burden of psoriasis cases.

We observed that only seven countries and territories showed decreasing trends in the ASIR of DM1 for the 0–9 age group, accounting for about 3.4% of the 204 countries (Finland, Greenland, Israel, Republic of Korea, Singapore, Sweden and Uruguay). These above seven countries belong to the high or high-middle SDI region and share commonalities in terms of diversity in economic models, medical technology innovation, as well as a strong emphasis on education and culture. For instance, Finland is a welfare state with the improvement of social and living conditions and the reduction of poverty rate in recent years,⁴⁰ which may explain the decline trend of DM1 incidence in the country’s younger age groups. However, this study showed that the DM1 burden of Finland has been stable at higher levels from 1990 to 2019. Finland has a relatively high prevalence rate of DM1,⁴¹ and the higher early mortality rate and elevated risk of comorbidities among children with DM1 may further exacerbate the disease burden in the younger population.⁴² Besides, the interaction between changes in living environment, lifestyle and the disease process may also accelerate the increase in the burden of DM1.⁴³ Therefore, it is necessary to develop targeted policies to reduce the burden of DM1 by taking into account the differences in lifestyle and environment across various regions, to ensure the efficiency of the healthcare system.

In both children and adolescents, the incidence of IBD increased markedly in Taiwan (Province of China) from high-SDI regions. The similar results were consistent with previous research, which found that the incidence of IBD in Taiwan increased in stages between 2001 and 2015, and the number of newly diagnosed IBD cases in individuals under 20 years old rose from 67 during 2001–2005 to 118 during 2011–2015.⁴⁴ Our results also show that the number of IBD incidence cases is rising rapidly in the high-income Asia Pacific region primarily. Although the Asia-Pacific region has been marked as an area with a low incidence of IBD compared with Europe and North America, there is a fact that the incidence and prevalence rates of paediatric IBD in many industrialised and affluent areas in the Asia-Pacific region are increasing rapidly.^{45 46} However, the currently available guidelines, predominantly originating from Europe and North America, may not fully pertain to clinicians treating children and adolescents with IBD in this region.⁴⁷ Therefore, there is an urgent need to take into account the unique disease characteristics and available financial resources in the Asia-Pacific region and to provide evidence-based guidelines for the treatment of children and adolescents with IBD in response to the rapidly increasing disease burden.

Regarding RA, higher levels of the increasing trend in incidence and disease burden were observed in low-middle or low SDI regions (eg, Mauritania and Mali). It is commonly believed that a high disease burden stems from the combined effects of underdeveloped sociodemographic factors, as well as the lack of access to and low effectiveness of the healthcare systems.¹⁸ RA that occurs in childhood or adolescence has a negative impact on QoL and increases the burden of medical expenditure for families and society, with poor treatment prognosis. Therefore, disease prevention is particularly essential in high-risk individuals living in underdeveloped countries, such as interventions targeting modifiable factors (eg, lifestyle). Early treatment during the preclinical phase of RA is also crucial.⁴⁸ However, for individuals in the preclinical stage who have developed arthralgia, there are currently no formal treatment recommendations, and early treatment with rituximab and abatacept only delays the onset of full-blown RA.⁴⁹ It is worth emphasising the urgent need for more efficient prevention strategies to address the rising RA burden.

The study explored the relative impact of age, period and birth cohort on AD prevalence trends. Age effects in IBD showed similar patterns across five SDI regions, with risk increasing with age. Due to the clinical diversity and sometimes ambiguous symptoms of children aged 0–9 years with IBD,⁵⁰ which often occurs in 21% of paediatric IBD,⁵¹ delays in diagnosis are common and may contribute to a greater magnitude of increased risk in adolescents. The age effect of DM1 again had a similar pattern across SDI regions, but the risk decreased with age in the 5–19 age group. Globally, unfavourable period effects of DM1 and RA and favourable period effects of IBD and psoriasis were observed. Period risks kept increasing in DM1 and RA over the last few years, which may be correlated to the cumulative risk factors, especially harmful environment exposure,⁵² genetic susceptibility,⁵³ access to medical healthcare systems,⁵⁴ etc, demonstrating the significant deterioration in recent years and the urgent need to control these diseases. In terms of cohort effects, individuals born later had a higher overall risk of DM1 and IBD and a lower overall risk of developing psoriasis than those born earlier. On the one hand, individuals born earlier may have more cumulative risk factors over their lifetime.⁵⁵ On the other hand, individuals born later in life are likely to experience high-quality early life-stage exposure management and, therefore, are at lower risk compared with individuals born earlier in life.⁵⁵

We also identified associations between the ESG index, the QoL index and the healthcare index and ASIR in 2019. Beyond reinforcing the importance of socioeconomic development for children and adolescent health, these findings indicate that countries experiencing rapid socioeconomic growth must act quickly to improve the QoL of those with ADs. To meet future challenges, optimising healthcare services is crucial for reducing the burden of ADs in children and adolescents. Regarding the relationships between varied country indexes and incidence of the four specific ADs, our findings indicate a positive association of AD incidence with both higher levels of national development and healthcare level, yet with differences in disease burden between the 0–9 years and 10–24 years age group. We found that lower combined risk in relation to the environment, human rights and health and safety and higher levels of overall healthcare services and QoL are related to an increase in the incidence of DM1 but a decrease in the burden of DM1 in both 0–9 years and 10–24 years age group, and the same association is found for the burden of IBD in children aged 0–9 years. In countries with comprehensive healthcare and a high QoL, DM1 has received widespread attention, and its diagnostic tools, testing and response to acute and chronic complications are timelier and more accurate than in countries with inadequate healthcare, potentially lowering the burden of disease among children and adolescents. A large number of people from low-income and middle-income countries suffer from chronic diseases, including about 291 million people with diabetes.⁵⁶ The macroeconomic impact of chronic diseases is enormous in these countries. For every 10% increase in chronic disease mortality, a country’s annual economic growth is estimated to be reduced by 0.5 percentage points.⁵⁷ Our findings suggested that ADs among children and adolescents may be underestimated, especially in low-income or low-QoL countries. There is an urgent need to pay more attention to DM1 in low-income and middle-income countries, and that various healthcare measures to reduce the burden of disease need to be further followed up. However, countries with high levels of development have a higher burden of RA in the child-adolescent population, as well as IBD in the 10–24 years age group, which could be explained by the accumulation of risk factors⁵² and the lack of necessary attention to RA in the child-adolescent population and IBD in adolescents aged 10–24 years.

To our best knowledge, this is the first study to comprehensively analyse the pattern and temporal trends of characteristic ADs and the effect of varied country indexes on AD incidence in children and adolescents at global, regional and national levels, compared with previous GBD publications regarding ADs.^{58 59} The major merit of the present study is that our findings generate valuable insights for AD epidemiology and health policy-making in countries with different healthcare and social development levels. However, there are several limitations in the present study. First, the use of large-scale statistical modelling methods has led to a lack of primary data, especially at the country level. There is therefore a large range of uncertainty in GBD estimates for 204 countries and territories, which may influence the accuracy of estimates of age, period and birth cohort effects. Second, because of the low level of healthcare in some underdeveloped countries, there may be misdiagnosis and underdiagnosis of diseases, leading to underestimation. Third, local characteristics of prevalence trends could not be explored because local data with sufficient precision were not available.

Implications

Despite these limitations, our study provides valuable insights into the policy implications for countries or territories dealing with a growing burden of ADs in children and adolescents. This analysis documents the substantial unmet needs in ADs control in children and adolescents before the COVID-19 pandemic. Future research should study the changes in AD incidence after the beginning of the pandemic, as there appears to be new onset of ADs due to COVID-19.⁶⁰

Conclusions

The unfavourable increasing trends in overall AD incidences among children and adolescents have continued to be observed globally over the past 30 years. Although we observed favourable decreasing trends of psoriasis incidence in children and adolescents, unfavourable increasing trends have appeared in low socioeconomic status regions. Our findings imply that the wide variation in the burden of ADs across geographically diverse countries should be addressed by considering the cultural and regional characteristics and development of each country and by designing and implementing mental health initiatives.

The founders had no role in the study design, data collection and analysis, decision to publish or preparation of the manuscript.

Data availability statement

Data are available on reasonable request.