Abstract
Objective
Prior studies suggested that maternal systemic autoimmune rheumatic diseases (SARDs) may increase the risk of mental disorders (MDs) in offspring. However, evidence is limited by heterogeneous definitions of autoimmune rheumatic diseases, focusing on specific MDs and scarce data from Asian populations. We aimed to investigate the association between maternal SARDs and the risk of childhood MDs.
Methods
Our study included 1 949 968 maternal-child dyads identified in the Taiwan Maternal and Child Health Database between 2004 and 2015. Maternal SARD diagnoses were identified based on approved catastrophic illness certificates, while childhood MDs were identified by documentation of a relevant diagnostic code in the nationwide health insurance claims data. Cox proportional hazard models were used to examine the association between maternal SARDs and childhood MDs.
Results
After a mean follow-up of 11.7 years, there were 774 (19.8%) and 306 907 (15.8%) live births born to mothers with and without SARDs respectively, which developed childhood MDs. Maternal SARDs were associated with an increased risk of childhood MDs (HR, 1.30; 95% CI 1.19 to 1.41). Elevated risks of childhood MDs were observed among offspring of mothers with systemic lupus erythematosus (HR=1.26, 95% CI 1.13 to 1.41), primary Sjogren’s syndrome (HR=1.59, 95% CI 1.29 to 1.97), idiopathic inflammatory myositis (HR=1.73, 95% CI 1.06 to 2.83) and systemic sclerosis (HR=1.94, 95% CI 1.15 to 3.28), compared with mothers without SARDs.
Conclusion
In this nationwide, population-based study from an Asian population, maternal SARDs were associated with higher risks of childhood MDs. Our findings suggest early monitoring of MDs among offspring of mothers with SARDs as part of clinical practice.
Keywords: Epidemiology, Autoimmune Diseases, Psychology
WHAT IS ALREADY KNOWN ON THIS TOPIC
Maternal autoimmune rheumatic diseases have been linked to the risks of mental disorders in offspring.
Despite the higher prevalence and severity of certain systemic autoimmune rheumatic diseases among mothers of Asian descent, research on mental disorders in their offspring remains limited.
WHAT THIS STUDY ADDS
Risks of mental disorders in the offspring were significantly associated with maternal systemic lupus erythematosus, primary Sjogren’s syndrome, idiopathic inflammatory myositis and systemic sclerosis.
Mental disorders that were significantly associated with maternal systemic autoimmune rheumatic diseases include anxiety disorders, intellectual disability, attention-deficit/hyperactivity disorder, specific learning disorders and motor disorders.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Our findings suggest the crucial needs for monitoring mental disorders in offspring born to mothers with systemic autoimmune rheumatic diseases in clinical practice.
Introduction
Systemic autoimmune rheumatic diseases (SARDs) include systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), primary Sjogren’s syndrome (PSS), idiopathic inflammatory myositis (IIM), systemic sclerosis (SSc) and systemic vasculitis, which are characterised by immune system dysregulation that may lead to organ dysfunction, disability and even death.1 SARDs contribute to a substantial disease burden worldwide, with the global years lived with disability rate for RA increasing by 13.8% between 1990 and 2020, while other musculoskeletal disorders, including SLE and SSc, ranked as the sixth leading cause of global years lived with disability in 2020.2–4 SARDs predominantly affect women of childbearing age, and active maternal SARDs were associated with preterm birth and low birth weight5 6 among offspring, potentially aggravating the loading of caregivers as well as increasing healthcare spending. As a result, children’s health has been an important issue in clinical care among offspring born to mothers with SARDs.7
Mental disorders (MDs) remain a major cause of burden globally. The Global Burden of Diseases, Injuries, and Risk Factors Study in 2019 disclosed that the prevalence of MDs, including neurodevelopmental disorders (NDDs), was increasing every year, while disability-adjusted life-years for MDs were rising incrementally during childhood and adolescence, making childhood MDs a growing public health concern.8 Previous animal studies observed that maternal immune activation can induce alterations in brain morphologic features, abnormal neurotransmitter activity and behavioural deficits in offspring, which may be attributable to transplacental inflammatory cytokines and autoantibodies.9 10 Recently, a few population-based cohort studies have examined the association between maternal autoimmune rheumatic diseases (ARDs) and childhood MDs.11–14 However, these findings may not be directly comparable and remain inconclusive due to methodological limitations, including heterogeneous definitions of ARDs exposure by combining localised ARDs with SARDs, a narrow focus on specific NDDs rather than the full spectrum of childhood MDs and a lack of evidence from Asian populations. Therefore, we aimed to investigate the risk of full-spectrum childhood MDs among offspring born to mothers with SARDs in the Taiwanese population.
Materials and methods
Data source and study population
Our population-based retrospective cohort study extracted data from the Taiwan Maternal and Child Health Database (TMCHD), which included all live births born in Taiwan from 1 January 2004 to 31 December 2015. We linked the TMCHD to the Taiwanese Birth Registry (2004–2015), the Death Registry (2004–2021) and the registry of catastrophic illness patient files in the National Insurance Research Database (2002–2021), which includes in-patient and out-patient claims and prescription records.15 Thus, the combined registry files contain comprehensive information on newborns (including the child’s sex, gestational age at birth, birth weight, parity and birth date) as well as maternal factors (eg, demographics, medical history of disease and medication use during pregnancy).16 High concordance between the birth registry data and birth records has been reported.17 The datasets used in this study were accessed through the Health and Welfare Data Science Center, which is administered by the Ministry of Health and Welfare in Taiwan. Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research. Approvals for our study were received from the Institutional Review Board of National Cheng Kung University Hospital, Taiwan (IRB#B-ER-112-308).
We included all singleton live births born between 2004 and 2015 in Taiwan. Detailed demographic and gestational information were retrieved. A flowchart illustrating patient inclusion is shown in figure 1. Briefly, we excluded children from multiple births (n=71 101), mothers who did not register in the National Health Insurance (NHI) programme (n=65 059) or stillbirths (n=30), resulting in 2 093 865 maternal–child dyads.
Figure 1. Flowchart of offspring born to maternal systemic autoimmune rheumatic diseases in Taiwan and their counterparts, 2004–2015. ARD, Autoimmune rheumatic diseases; SARDs, Systemic autoimmune rheumatic diseases; NHIRD, National Health Insurance Research Database.
Definition of maternal SARDs and childhood MDs
We used the diagnosis code of the International Classification of Diseases, Ninth and Tenth Revision, Clinical Modification (ICD-9 and ICD-10 CM) to categorise disease diagnoses. Maternal SARDs, including SLE, RA, PSS, IIM, SSc and systemic vasculitis, were confined to those that were approved for the relevant critical illness certificate before delivery by using the registry of catastrophic illness patient files in the NHI programme. To assist families that face financial hardship due to medical bills, the NHI itemises 31 groups of catastrophic illnesses, including all SARDs. A patient’s attending physician is required to provide relevant clinical and laboratory information to obtain a critical illness certificate. The application is examined by a review committee based on the classification criteria for individual diseases, thus validating the diagnosis. Patients receiving a critical illness certificate are exempted from copayment, which provides reassurance that the SARDs information is complete. Among 2 093 865 maternal–child dyads, 3907 offspring were born to mothers with SARDs. Mothers without a critical illness certificate approval for their SARDs nor diagnosis codes of other organ-specific ARDs before delivery, including endocrine, gastrointestinal, blood, nervous system, skin and immune arthritis, were classified into the comparison group, resulting in 1 946 061 offspring born to mothers without any ARDs (online supplemental table 1).
The primary outcome of interest was diagnoses of childhood MDs. We selected various MDs with peak incidence during infancy and childhood listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth edition,18 after taking follow-up length into consideration. These MDs included depressive disorders, anxiety disorders, trauma-related and stressor-related disorders, dissociative disorders, feeding and eating disorders, elimination disorders, gender dysphoria, disruptive, impulse-control and conduct disorders as well as NDDs (intellectual disability, autism spectrum disorder, attention deficit hyperactivity disorder, specific learning disorders and motor disorders). Relevant ICD-9 and ICD-10 codes for MDs are shown in online supplemental table 1. Offspring were defined as having MDs if they had at least one documentation of a relevant ICD code in the inpatient record or at least two times in outpatient records. For each MD, the onset date was defined as the first day of the first medical service contact at either a clinic visit or hospitalisation record during the follow-up period (until end of 2021). Validation of common MD diagnosis coding in the NHI programme has been done, revealing generally acceptable accuracy.19 All offspring were censored at the earliest of death, loss of insurance coverage or the end of the study period.
Covariates
The selection of covariates for adjustment was based on a literature review.11–14 Variables included maternal characteristics: age at delivery (<20, 20–<34, ≥34 years), psychiatric history (yes/no) and tobacco use (yes/no) (a proxy for smoking status), which was defined by ICD codes for nicotine dependence (ICD-9-CM code: 305.1; ICD-10-CM code: F17.200) and medication prescriptions for treating nicotine dependence (Anatomical Therapeutic Chemical codes: N07BA and N06AX12) before delivery, urbanicity of residence at delivery (urban, suburban or rural),20 family income and anti-rheumatic medication prescription records during the entire pregnancy (yes/no); characteristics of offspring including parity (1, 2, ≥3), sex (male or female) and birth cohort year. Monthly family household income was categorised into quartiles in US dollars (<$1037, 1037–<1540, 1540–<2310 and ≥2310); these estimates were based on health insurance premiums, as insurance premium per family is dependent on family income.
Antirheumatic medications may play a role in the relationship between maternal SARDs and childhood MDs, so we further classified antirheumatic medications into five categories: corticosteroids (Anatomical Therapeutic Chemical code: H02A), conventional disease-modifying antirheumatic drugs (Anatomical Therapeutic Chemical code: A07EC, P01BA02), biologic and targeted-synthetic disease-modifying antirheumatic drugs (Anatomical Therapeutic Chemical code: L04A), non-biologic nor targeted-synthetic immunosuppressant/immunomodulator (Anatomical Therapeutic Chemical code: L04A), non-steroidal anti-inflammatory drugs (Anatomical Therapeutic Chemical code: M01A) and acetaminophen (Anatomical Therapeutic Chemical code: N02BE01).
Statistical analysis
We used Cox proportional hazard models to examine the association between maternal SARDs and childhood MDs. HRs were estimated to examine the relationships between maternal SARDs (both overall and by specific type) and childhood MDs (including both overall and subtype-specific outcomes). In all models, we adjusted for maternal age, maternal psychiatric history, paternal psychiatric history, maternal tobacco use, parity, urbanicity of residence, family income, birth sex and cohort year.
Both preterm birth (defined as gestational age <37 weeks) and low birth weight (defined as birth weight <2500 g) have been proposed as potential mediators of the effect of maternal SARDs on childhood MDs.21 22 Therefore, causal mediation analyses were performed to assess these pathways. In addition, we performed several sensitivity analyses. For example, considering that child mortality is a competing event for an MD diagnosis, we used the Fine-Gray hazard model, which estimated the cumulative incidence function in the presence of competing risks, thus preventing overestimation of childhood MD risks. Furthermore, we also performed sensitivity analyses by comparing the associations between prenatal corticosteroid use (±the median daily 7.5 mg prednisolone equivalent during the entire pregnancy) and other steroid-sparing agents, including biologic and targeted-synthetic disease-modifying antirheumatic drugs, non-biologic nor targeted-synthetic immunosuppressant/immunomodulator and overall childhood MDs. The median daily 7.5 mg prednisolone equivalent is a so-called ‘low-dose’ and has been established as a treatment target in SARDs.23 All analyses were performed with SAS V.9.4 (SAS Institute, Cary, North Carolina).
Results
From 2004 to 2015, we included 1 949 968 singleton live births in this study (median age at the end of the study, 10.9 years; 52.1% male; median follow-up years were 10.3 years and 11.6 years in maternal SARDs group and in comparison, respectively), of which 0.2% were born to mothers with SARDs diagnosed before childbirth. In comparison to mothers without ARDs, mothers with SARDs were older at delivery (31.9 years vs 30.3 years), had a higher percentage of underlying psychiatric history (10.3% vs 5.6%) and tobacco use disorder before delivery (4.0% vs 1.8%) as well as had a higher proportion of all kinds of prenatal antirheumatic medication use (table 1). Compared with their counterparts, offspring born to mothers with any type of SARD were more likely to be born preterm (19.5% vs 7.1%), with low birth weight (21.3% vs 6.2%), born in urban areas and with a greater family income (table 1). The number of newborns born to mothers with any SARD increased each year.
Table 1. Baseline characteristics of mothers and offspring by maternal autoimmune rheumatic diseases status.
| Characteristics | With maternal SARDs (n=3907) n (%) |
Without maternal ARDs (n=1 946 061) n (%) |
||
|---|---|---|---|---|
| Maternal characteristics | ||||
| Maternal age at delivery (year) | ||||
| Mean±SD, median | 31.9±4.6, 31.9 | 30.3±4.8, 30.5 | ||
| < 20 | 38 | (1.0) | 39 322 | (2.0) |
| 20–<34 | 2623 | (67.1) | 1 476 172 | (75.9) |
| ≥34 | 1246 | (31.9) | 430 567 | (22.1) |
| Maternal psychiatric disorders (yes)* | 403 | (10.3) | 109 018 | (5.6) |
| Paternal psychiatric disorders (yes)*† | 162 | (4.2) | 70 184 | (3.6) |
| Tobacco use (yes)* | 158 | (4.0) | 34 673 | (1.8) |
| Parity | ||||
| 1 | 2759 | (70.6) | 1 337 663 | (68.7) |
| 2 | 1033 | (26.4) | 535 497 | (27.5) |
| ≥3 | 115 | (3.0) | 72 901 | (3.8) |
| Urbanicity of residence | ||||
| Urban | 2203 | (56.4) | 1 037 363 | (53.3) |
| Suburban | 1421 | (36.4) | 730 806 | (37.6) |
| Rural | 283 | (7.2) | 177 892 | (9.1) |
| Family income (US$)† | ||||
| <1037 | 553 | (14.2) | 364 926 | (18.8) |
| 1037–<1540 | 687 | (17.6) | 365 106 | (18.8) |
| 1540–<2310 | 758 | (19.4) | 365 231 | (18.8) |
| ≥2310 | 839 | (21.5) | 365 148 | (18.8) |
| Anti-rheumatic medications during pregnancy (n, %) | ||||
| Corticosteroids (mg/d) | ||||
| Mean±SD, median | 5.9±5.5, 4.7 | 0.4±0.8, 0.2 | ||
| No | 2363 | (60.5) | 1 905 475 | (97.9) |
| Yes | 1544 | (39.5) | 40 535 | (2.1) |
| Conventional DMARD | ||||
| No | 1986 | (50.8) | 1 944 548 | (99.9) |
| Yes | 1921 | (49.2) | 1462 | (0.1) |
| Biologic and targeted-synthetic DMARD | ||||
| No | 3850 | (98.5) | 1 946 007 | (100.00) |
| Yes | 57 | (1.5) | 3 | (0.00) |
| Non-biologic nor targeted-synthetic immunosuppressants/immunomodulators | ||||
| No | 3423 | (87.6) | 1 945 693 | (100.00) |
| Yes | 484 | (12.4) | 317 | (0.00) |
| NSAID | ||||
| No | 2221 | (56.8) | 1 340 400 | (68.9) |
| Yes | 1686 | (43.2) | 605 610 | (31.1) |
| Acetaminophen (paracetamol) | ||||
| No | 1742 | (44.6) | 1 012 781 | (52.0) |
| Yes | 2165 | (55.4) | 933 139 | (48.0) |
| Offspring characteristics | ||||
| Birth sex | ||||
| Male | 2043 | (52.3) | 1 013 901 | (52.1) |
| Female | 1864 | (47.7) | 932 160 | (47.9) |
| Gestational age at birth (wks) | ||||
| <37 | 761 | (19.5) | 138 575 | (7.1) |
| ≥37 | 3146 | (80.5) | 1 807 486 | (92.9) |
| Birth weight (g) | ||||
| <2500 | 832 | (21.3) | 121 303 | (6.2) |
| ≥2500 | 3075 | (78.7) | 1 824 758 | (93.8) |
| Birth cohort year | ||||
| 2004 | 259 | (6.6) | 175 284 | (9.0) |
| 2005 | 233 | (6.0) | 166 498 | (8.6) |
| 2006 | 257 | (6.6) | 165 944 | (8.5) |
| 2007 | 284 | (7.3) | 167 518 | (8.6) |
| 2008 | 298 | (7.6) | 161 913 | (8.3) |
| 2009 | 309 | (7.9) | 158 545 | (8.2) |
| 2010 | 250 | (6.4) | 126 965 | (6.5) |
| 2011 | 343 | (8.8) | 153 230 | (7.9) |
| 2012 | 383 | (9.8) | 171 936 | (8.8) |
| 2013 | 394 | (10.1) | 159 544 | (8.2) |
| 2014 | 406 | (10.4) | 163 827 | (8.4) |
| 2015 | 491 | (12.6) | 174 857 | (9.0) |
| Number of offspring with diagnoses of neurodevelopmental and mental disorders during follow-up | 774 | (19.8) | 306 907 | (15.8) |
Time period to define these variables was between 2004 and before delivery.
Missing: for with maternal connective tissue diseases group, there were missing information for paternal psychiatric disorder (n=1109) and family income (n=1070); for without maternal autoimmune diseases group, there were missing information for paternal psychiatric disorder (n=509 253) and family income (n=485 650).
DMARD, disease-modifying anti-rheumatic drug; NSAID, non-steroidal anti-inflammatory drug; SARD, systemic autoimmune rheumatic diseases.
Increased HRs of MDs were observed for offspring born to any SARD (HR=1.30, 95% CI 1.19 to 1.41), SLE (HR=1.26, 95% CI 1.13 to 1.41), PSS (HR=1.59, 95% CI 1.29 to 1.97), IIM (HR=1.73, 95% CI 1.06 to 2.83) and SSc (HR=1.94, 95% CI 1.15 to 3.28) (table 2). On the other hand, any maternal SARD was associated with childhood MDs other than NDDs (HR=1.30, 95% CI 1.07 to 1.58), including anxiety disorders (HR=1.55, 95% CI 1.17 to 2.06). Any maternal SARD also increased the risk of childhood NDDs (HR=1.28, 95% CI 1.17 to 1.40), including intellectual disability (HR=1.29, 95% CI 1.12 to 1.49), attention deficit hyperactivity disorder (HR=1.21, 95% CI 1.07 to 1.37), specific learning disorders (HR=1.31, 95% CI 1.07 to 1.61) and motor disorders (HR=1.37, 95% CI 1.19 to 1.57) (table 3).
Table 2. Associations between maternal systemic autoimmune rheumatic diseases and childhood mental disorders.
| Maternal CTDs | Maternal SARDs n (%) | Offspring MDs among maternal with SARDs (n=3907) n (%) |
Crude model | Adjusted model* |
|---|---|---|---|---|
| HR (95% CI) | HR (95% CI) | |||
| Overall maternal SARDs | 3907 | 774 (19.8) | 1.36 (1.27 to 1.46) | 1.30 (1.19 to 1.41) |
| SLE | 2200 (56.3) | 424 (10.9) | 1.29 (1.17 to 1.42) | 1.26 (1.13 to 1.41) |
| RA | 1009 (25.8) | 182 (4.7) | 1.22 (1.05 to 1.41) | 1.15 (0.97 to 1.36) |
| Sjogren’s syndrome | 550 (14.1) | 128 (3.3) | 1.80 (1.52 to 2.14) | 1.59 (1.29 to 1.97) |
| IIM | 90 (2.3) | 20 (0.5) | 1.60 (1.03 to 2.47) | 1.73 (1.06 to 2.83) |
| SSc | 67 (1.7) | 20 (0.5) | 2.10 (1.36 to 3.25) | 1.94 (1.15 to 3.28) |
| SV | 67 (1.7) | 11 (0.3) | 1.18 (0.66 to 2.12) | 0.96 (0.46 to 2.01) |
Overall: Offspring born to maternal connective tissue diseases, including SLE, RA, Sjogren’s syndrome, IIM, SSc and SV.
Adjusted for maternal age, maternal psychiatric history, paternal psychiatric history, maternal tobacco use, parity, urbanicity of residence, family income, birth sex and cohort year.
IIM, idiopathic inflammatory myositis; MDs, mental disorders; RA, rheumatoid arthritis; SARDs, systemic autoimmune rheumatic diseases; SLE, systemic lupus erythematosus; SSc, systemic sclerosis; SV, systemic vasculitis.
Table 3. Associations between maternal systemic autoimmune rheumatic diseases and overall/specific childhood mental disorders.
| Offspring MDs | Without maternal ARDs (n=1 946 061) n (%) |
With maternal SARDs (n=3907) n (%) |
Crude | Adjusted model* |
|---|---|---|---|---|
| HR (95% CI) | HR (95% CI) | |||
| Overall childhood MDs | 306 907 (15.8) | 774 (19.8) | 1.36 (1.27 to 1.46) | 1.30 (1.19 to 1.41) |
| MDs other than NDDs | 59 497 (3.1) | 133 (3.4) | 1.28 (1.08 to 1.51) | 1.30 (1.07 to 1.58) |
| Depressive disorders | 10 493 (0.5) | 17 (0.4) | 1.06 (0.66 to 1.71) | 1.11 (0.66 to 1.88) |
| Anxiety disorders | 23 965 (1.2) | 62 (1.6) | 1.49 (1.16 to 1.91) | 1.55 (1.17 to 2.06) |
| Trauma-related and stressor-related disorders | 17 798 (0.9) | 30 (0.8) | 1.00 (0.70 to 1.43) | 0.95 (0.63 to 1.44) |
| Dissociative disorders | 189 (0.0) | ≦3 (0.0)† | 3.36 (0.47 to 23.90) | 4.69 (0.66 to 33.58) |
| Feeding and eating disorders | 296 (0.0) | 0 (0.0) | NA | NA |
| Elimination disorders | 10 254 (0.5) | 26 (0.7) | 1.34 (0.91 to 1.97) | 1.48 (0.96 to 2.30) |
| Gender dysphoria | 21 (0.0) | 0 (0.0) | NA | NA |
| Disruptive, impulse-control and conduct disorders | 8158 (0.4) | 20 (0.5) | 1.35 (0.87 to 2.10) | 1.02 (0.57 to 1.85) |
| NDDs | 279 326 (14.4) | 714 (18.3) | 1.36 (1.26 to 1.46) | 1.28 (1.17 to 1.40) |
| Intellectual disability | 95 575 (4.9) | 264 (6.8) | 1.42 (1.26 to 1.60) | 1.29 (1.12 to 1.49) |
| Autism spectrum disorders | 21 969 (1.1) | 50 (1.3) | 1.17 (0.89 to 1.55) | 0.95 (0.68 to 1.33) |
| ADHD | 157 365 (8.1) | 377 (9.6) | 1.27 (1.15 to 1.40) | 1.21 (1.07 to 1.37) |
| Specific learning disorders | 52 383 (2.7) | 128 (3.3) | 1.24 (1.04 to 1.48) | 1.31 (1.07 to 1.61) |
| Motor disorders | 92 737 (4.8) | 275 (7.0) | 1.56 (1.39 to 1.75) | 1.37 (1.19 to 1.57) |
Adjusted for maternal age, maternal psychiatric history, paternal psychiatric history, maternal tobacco use, parity, urbanicity of residence, family income, birth sex and cohort year.
The results≤3 cannot be displayed.
ADHD, attention deficit hyperactivity disorder; ARDs, autoimmune rheumatic disorders; MDs, mental disorders; NA, not applicable; NDDs, neurodevelopmental disorders; SARDs, systemic autoimmune rheumatic diseases.
An estimated mediating effect via low birth weight was observed (natural indirect effect=1.05, 95% CI 1.04 to 1.05; proportion mediated=18.7%), but not for preterm birth (natural indirect effect=0.96, 95% CI 0.92 to 1.02; proportion mediated=−16.13%) (online supplemental table 2). The competing risk analysis for premature maternal death using the Fine and Gray model revealed similar results to the main analysis (online supplemental table 3). The associations between maternal SARDs and childhood MDs were all attenuated after additionally adjusting for antirheumatic medication prescriptions during pregnancy (online supplemental tables 4 and 5). We found elevated risks of MDs among offspring born to mothers with overall and specific SARDs with lower dose prenatal prednisolone usage, especially for those without concurrent steroid-sparing agent use (online supplemental table 6).
Discussion
In this Taiwanese population-based birth cohort study with a median follow-up of 11.7 years, we found that offspring born to mothers with SARDs, including SLE, PSS, IIM and SSc, were associated with an increased risk of childhood MDs compared with their counterparts. Children were at risk of various childhood MDs, especially anxiety disorders.
Our findings echoed those seen in previous studies.11 12 Possible biologic mechanisms include shared genetic factors between SARDs and MDs as well as transplacental autoantibodies and proinflammatory cytokines that interfere with fetal brain development.9 10 Neuropsychiatric SLE is complex and multifactorial, encompassing releases of various inflammatory cytokines, formation of autoantibodies as well as immune complexes and dysfunction of the blood–brain barrier, resulting in intracranial vasculopathy and neuronal cell damage. The current study suggested that maternal SLE was associated with any childhood MDs, which is supported by an established mouse model, indicating that transplacental neurotoxic autoantibodies of maternal SLE impair fetal cortical function.24 Patients with RA present with symmetric and erosive polyarthritis, which is distinguished by marked elevation of multiple inflammatory cytokines. However, a meta-analysis observed that the disease activity declined in 60% of mothers with RA during pregnancy,25 thus may decrease the risk of childhood MDs. The declined disease activity may be attributable to pregnancy-related hormones, which modulate innate and adaptive immune responses by suppressing the activation of macrophages as well as lowering serum proinflammatory cytokine production.26
PSS is well known for dry eyes and dry mouth. The glandular secretory dysfunction is related to serum levels of anti-Ro/SSA antibody. We found that maternal PSS was associated with overall childhood MDs, in which anti-Ro/SSA antibody may play a role. This antibody can be found both in peripheral blood and cerebrospinal fluid of patients with PSS, suggesting its ability to penetrate the blood–brain barrier and contribute to neurological complications.27 Moreover, an association has been reported between childhood neuropsychiatric symptoms and prenatal exposure to anti-Ro/SSA antibody,28 which supports our findings. The hallmark symptom of IIM is symmetrical proximal muscle weakness, resulting from muscle inflammation. Our study demonstrated an association between maternal IIM and overall childhood MDs, which may be attributable to IL-18, a recently recognised key cytokine both in IIM as well as in severe mental illness,29 providing a possible explanation for our observations.
SSc is characterised by multiple organ fibrosis. With few effective treatments of SSc-related organ involvement, patients with SSc have a generally worse prognosis compared with other SARDs. Maternal SSc noticeably increased overall childhood MDs. Though the underlying mechanism remains elusive, shared inflammatory cytokines have been implicated in both SSc and psychiatric disorders.30 A recent study reported that offspring born to mothers with SSc had normal cognitive and adaptive function, but had a higher proportion of psychological and social impairments, suggesting that the mothers’ severe and irreversible illness may cause emotional distress in children and restrict emotional and social development.31
Childhood MDs are heterogeneous conditions with potentially distinct causal pathways, which may explain why we observed associations with specific childhood MDs rather than all MDs. For example, we observed the strongest association for anxiety disorders, whereas associations with other childhood MDs were weaker or absent. Anxiety disorder is the most prevalent MD among Taiwanese children and may have gained awareness to a certain extent, while other childhood MDs may go unrecognised during childhood.32 Offspring born to mothers with SARDs had similar risks across specific childhood NDDs, which may result from disease co-occurrence.33 Previous research found that childhood intellectual disability is related to elevated proinflammatory cytokine levels in umbilical blood.34 Attention deficit hyperactivity disorder is the second most prevalent MD among Taiwanese children. Maternal SARD-related transplacental cytokine may cause dopaminergic system dysfunction, potentially leading to the development of attention deficit hyperactivity disorder.35 We found an association between maternal SARDs and childhood-specific learning disorders, supported by a published study observing that offspring born to maternal SLE were at risk of a learning disability.36 To the best of our knowledge, this is the first population study to report that all-cause maternal SARDs increased the risk of motor disorders among offspring, which is supported by previous mouse models demonstrating offspring motor behaviours induced by maternal immune activation.9 Our study did not find a link between maternal SARDs and childhood autism spectrum disorder, which was in line with previous study and may be partially explained by small case numbers, and diagnostic overshadowing because of existent other NDDs among children in Taiwanese clinical settings.37
As reported previously, our mediation analyses suggested that the influence of maternal SARDs on childhood MDs was partly mediated by low birth weight (online supplemental table 2).22 Our sensitivity analyses suggested that prenatal antirheumatic medications may play a role in the association between maternal SARDs and childhood MDs, mirroring previous literature (online supplemental tables 4 and 5).38–40 Elevated risks of MDs were observed among offspring born to mothers with SARDs with lower dose prednisolone usage before delivery, particularly for those without concurrent steroid-sparing agent use (online supplemental table 6). These observations may be explained by flare-ups of maternal SARDs due to inadequate medical control. However, we lack data on the actual disease progression of maternal SARDs to further evaluate this plausible explanation. Future studies that incorporate data on disease progression and medication use in mothers affected by SARDs are warranted.
Our data showed annual increases in the number of newborns born to mothers with SARDs, emphasising the rising disease burden of maternal SARDs in Taiwan. Of these, SLE comprised the largest percentage, at 56.3%. In comparison to a previous population-based study of maternal SARDs and child mental health conducted in Denmark, the prevalence rate of maternal SLE and Sjogren’s syndrome was higher in Taiwan, while the prevalence rate of maternal RA, IIM and SSc was lower (online supplemental table 7).12
To our knowledge, this is the largest population-based cohort study in Asia to show robust evidence of associations between childhood MDs and maternal SARDs. First, among the strengths of our study is the large-scale population cohort, providing the opportunity for the study of these rare conditions. Second, the mean follow-up length of nearly 12 years enabled us to comprehensively capture multiple childhood MDs with a peak incidence falling within this period.
Several limitations of the study should be noted. First, despite 11.7 years of median follow-up, certain MDs have a peak incidence in later childhood, which may underestimate the case numbers. Second, we lacked access to disease activities of maternal SARDs during pregnancy to investigate whether the transplacental inflammatory cytokines and autoantibodies that are associated with the SARDs impact childhood mental health. Third, despite we used critical illness certificate-related diagnoses in the claims data for maternal SARDs, misclassification bias may exist. Fourth, because childhood MDs were identified using claims data, the incidence of childhood MDs may have been underestimated, which would likely bias our estimates towards the null. Fifth, we did not exclude mothers with overlapping SARDs; though the numbers were small (eg, the most common overlapping SARDs were SLE and RA in our study, accounting for 1.1% among all mothers with SARDs), we were unable to look at how each maternal SARD affects each childhood MD. Sixth, information on some potential confounders, such as maternal obesity and alcohol consumption, was not available in our registry data. In addition, maternal smoking information was unavailable, and nicotine dependence was used as a proxy for tobacco use during pregnancy, which is likely to underestimate the true prevalence of smoking. Last but not least, there may be uncontrolled or residual confounding factors, including maternal genetics, lifestyles and environmental factors, thus the causality of the associations has not been established.
Conclusion
Maternal SARDs such as SLE, PSS, IIM and SSc are associated with various types of MDs in offspring. Our findings suggest the need for monitoring MDs in offspring born to mothers with SARDs in clinical practice with multidisciplinary care. Future research is required to elucidate the underlying pathogenesis and the roles played by prenatal antirheumatic medications in childhood MDs.
Supplementary material
Acknowledgements
An earlier version of this paper entitled 'Maternal connective tissue disease and the risk of childhood mental disorder: A population-based cohort in Taiwan' was orally presented by Y-CH at the Asia Pacific League of Associations for Rheumatology 2025 Congress in Fukuoka, Japan, September 2025. The authors are grateful to the session chair and the audience for their valuable comments and suggestions, which significantly improved this manuscript. (2026), Oral Presentation Abstracts. Int J Rheum Dis, 29:e70439.
Footnotes
Funding: US National Institutes of Health (grant number R03CA273608), Alex’s Lemonade Stand Foundation (grant number 17-01882), Taiwan National Science and Technology Council (NSTC 112-2314-B-006 -068 -MY3), National Science and Technology Council (NSTC 113-2629-H-006-002-SS2) and the National Health Research Institutes (NHRI-EX114-11410PI).
Provenance and peer review: Not commissioned; externally peer-reviewed.
Patient consent for publication: Not applicable.
Data availability free text: The dataset supporting the conclusions of this article is available in the National Health Insurance Research Database (NHIRD) published by the Taiwan National Health Insurance (NHI) Bureau. Due to 'Personal Information Protection Act', data cannot be made publicly available. Requests for data can be sent as a formal proposal to the NHIRD (http://nhird.nhri.org.tw).
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
Data may be obtained from a third party and are not publicly available.
References
- 1.Gabriel SE, Michaud K. Epidemiological studies in incidence, prevalence, mortality, and comorbidity of the rheumatic diseases. Arthritis Res Ther. 2009;11:229. doi: 10.1186/ar2669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Carter EE, Barr SG, Clarke AE. The global burden of SLE: prevalence, health disparities and socioeconomic impact. Nat Rev Rheumatol. 2016;12:605–20. doi: 10.1038/nrrheum.2016.137. [DOI] [PubMed] [Google Scholar]
- 3.Safiri S, Kolahi AA, Hoy D, et al. Global, regional and national burden of rheumatoid arthritis 1990-2017: a systematic analysis of the Global Burden of Disease study 2017. Ann Rheum Dis. 2019;78:1463–71. doi: 10.1136/annrheumdis-2019-215920. [DOI] [PubMed] [Google Scholar]
- 4.Tian J, Kang S, Zhang D, et al. Global, regional, and national incidence and prevalence of systemic sclerosis. Clin Immunol. 2023;248:109267. doi: 10.1016/j.clim.2023.109267. [DOI] [PubMed] [Google Scholar]
- 5.Sugawara E, Kato M, Fujieda Y, et al. Pregnancy outcomes in women with rheumatic diseases: a real-world observational study in Japan. Lupus. 2019;28:1407–16. doi: 10.1177/0961203319877258. [DOI] [PubMed] [Google Scholar]
- 6.Tsai Y-C, Chang H-C, Chiou M-J, et al. Fetal-neonatal and maternal pregnancy outcomes in women with rheumatoid arthritis: a population-based cohort study. BMJ Open. 2022;12:e059203. doi: 10.1136/bmjopen-2021-059203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Marder W, Littlejohn EA, Somers EC. Pregnancy and autoimmune connective tissue diseases. Best Pract Res Clin Rheumatol. 2016;30:63–80. doi: 10.1016/j.berh.2016.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.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: 10.1016/S2215-0366(21)00395-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Aavani T, Rana SA, Hawkes R, et al. Maternal immune activation produces cerebellar hyperplasia and alterations in motor and social behaviors in male and female mice. Cerebellum. 2015;14:491–505. doi: 10.1007/s12311-015-0669-5. [DOI] [PubMed] [Google Scholar]
- 10.Luchicchi A, Lecca S, Melis M, et al. Maternal Immune Activation Disrupts Dopamine System in the Offspring. Int J Neuropsychopharmacol. 2016;19:pyw007. doi: 10.1093/ijnp/pyw007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Nielsen TC, Nassar N, Shand AW, et al. Association of Maternal Autoimmune Disease With Attention-Deficit/Hyperactivity Disorder in Children. JAMA Pediatr. 2021;175:e205487. doi: 10.1001/jamapediatrics.2020.5487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.He H, Yu Y, Liew Z, et al. Association of Maternal Autoimmune Diseases With Risk of Mental Disorders in Offspring in Denmark. JAMA Netw Open. 2022;5:e227503. doi: 10.1001/jamanetworkopen.2022.7503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tsai P-H, Yu K-H, Chou I-J, et al. Risk of autism spectrum disorder in children born to mothers with systemic lupus erythematosus and rheumatoid arthritis in Taiwan. Joint Bone Spine. 2018;85:599–603. doi: 10.1016/j.jbspin.2017.11.005. [DOI] [PubMed] [Google Scholar]
- 14.Chiu H-J, Sun C-K, Tsai S-J, et al. A nationwide study of the risks of major mental disorders among the offspring of parents with rheumatoid arthritis. Sci Rep. 2022;12:4962. doi: 10.1038/s41598-022-08834-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.National Health Insurance Administration. Ministry of Health and Welfare . Taiwan: R.O.C; 2014. [31-Dec-2016]. National health insurance annual report 2014-2015.https://www.nhi.gov.tw/English/Content_List.aspx?n=8162ED7256773F45&topn=BCB2B0D2433F6491 Available. accessed. [Google Scholar]
- 16.Li CY, Chen LH, Chiou MJ, et al. Set-up and future applications of the Taiwan maternal and child health database (TMCHD) Taiwan J Public Health. 2016 [Google Scholar]
- 17.Lin C-M, Lee P-C, Teng S-W, et al. Validation of the Taiwan Birth Registry using obstetric records. J Formos Med Assoc. 2004;103:297–301. [PubMed] [Google Scholar]
- 18.American Psychiatric Association. DSM-5 Task Force . Diagnostic and Statistical Manual of Mental Disorders: DSM-5. 5th. American Psychiatric Publishing, Inc; 2013. edn. [Google Scholar]
- 19.Wu C-S, Kuo C-J, Su C-H, et al. Using text mining to extract depressive symptoms and to validate the diagnosis of major depressive disorder from electronic health records. J Affect Disord. 2020;260:617–23. doi: 10.1016/j.jad.2019.09.044. [DOI] [PubMed] [Google Scholar]
- 20.Heck JE, Lee P-C, Wu C-K, et al. Gestational risk factors and childhood cancers: A cohort study in Taiwan. Int J Cancer. 2020;147:1343–53. doi: 10.1002/ijc.32905. [DOI] [PubMed] [Google Scholar]
- 21.Johnson S, Marlow N. Preterm birth and childhood psychiatric disorders. Pediatr Res. 2011;69:11R–8R. doi: 10.1203/PDR.0b013e318212faa0. [DOI] [PubMed] [Google Scholar]
- 22.Nomura Y, Wickramaratne PJ, Pilowsky DJ, et al. Low birth weight and risk of affective disorders and selected medical illness in offspring at high and low risk for depression. Compr Psychiatry. 2007;48:470–8. doi: 10.1016/j.comppsych.2007.04.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Buttgereit F, da Silva JAP, Boers M, et al. Standardised nomenclature for glucocorticoid dosages and glucocorticoid treatment regimens: current questions and tentative answers in rheumatology. Ann Rheum Dis. 2002;61:718–22. doi: 10.1136/ard.61.8.718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lee JY, Huerta PT, Zhang J, et al. Neurotoxic autoantibodies mediate congenital cortical impairment of offspring in maternal lupus. Nat Med. 2009;15:91–6. doi: 10.1038/nm.1892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Jethwa H, Lam S, Smith C, et al. Does Rheumatoid Arthritis Really Improve During Pregnancy? A Systematic Review and Metaanalysis. J Rheumatol. 2019;46:245–50. doi: 10.3899/jrheum.180226. [DOI] [PubMed] [Google Scholar]
- 26.Straub RH, Buttgereit F, Cutolo M. Benefit of pregnancy in inflammatory arthritis. Ann Rheum Dis. 2005;64:801–3. doi: 10.1136/ard.2005.037580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Tetsuka S, Suzuki T, Ogawa T, et al. Anti-Ro/SSA Antibodies May Be Responsible for Cerebellar Degeneration in Sjogren’s Syndrome. J Clin Med Res. 2021;13:113–20. doi: 10.14740/jocmr4429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Askanase AD, Izmirly PM, Katholi M, et al. Frequency of neuro-psychiatric dysfunction in anti-SSA/SSB exposed children with and without neonatal lupus. Lupus. 2010;19:300–6. doi: 10.1177/0961203309354542. [DOI] [PubMed] [Google Scholar]
- 29.Szabo A, O’Connell KS, Ueland T, et al. Increased circulating IL-18 levels in severe mental disorders indicate systemic inflammasome activation. Brain Behav Immun. 2022;99:299–306. doi: 10.1016/j.bbi.2021.10.017. [DOI] [PubMed] [Google Scholar]
- 30.Scala E, Pallotta S, Frezzolini A, et al. Cytokine and chemokine levels in systemic sclerosis: relationship with cutaneous and internal organ involvement. Clin Exp Immunol. 2004;138:540–6. doi: 10.1111/j.1365-2249.2004.02642.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Galli J, Loi E, Lazzaroni MG, et al. Neurodevelopmental profile in children born to mothers affected by systemic sclerosis. Early Hum Dev. 2024;191:105988. doi: 10.1016/j.earlhumdev.2024.105988. [DOI] [PubMed] [Google Scholar]
- 32.Chen Y-L, Chen WJ, Lin K-C, et al. Prevalence of DSM-5 mental disorders in a nationally representative sample of children in Taiwan: methodology and main findings. Epidemiol Psychiatr Sci. 2019;29:e15. doi: 10.1017/S2045796018000793. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.McClain MB, Hasty Mills AM, Murphy LE. Inattention and hyperactivity/impulsivity among children with attention-deficit/hyperactivity-disorder, autism spectrum disorder, and intellectual disability. Res Dev Disabil. 2017;70:175–84. doi: 10.1016/j.ridd.2017.09.009. [DOI] [PubMed] [Google Scholar]
- 34.von Ehrenstein OS, Neta GI, Andrews W, et al. Child Intellectual Development in Relation to Cytokine Levels in Umbilical Cord Blood. Am J Epidemiol. 2012;175:1191–9. doi: 10.1093/aje/kwr393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Dunn GA, Nigg JT, Sullivan EL. Neuroinflammation as a risk factor for attention deficit hyperactivity disorder. Pharmacol Biochem Behav. 2019;182:22–34. doi: 10.1016/j.pbb.2019.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ross G, Sammaritano L, Nass R, et al. Effects of mothers’ autoimmune disease during pregnancy on learning disabilities and hand preference in their children. Arch Pediatr Adolesc Med. 2003;157:397–402. doi: 10.1001/archpedi.157.4.397. [DOI] [PubMed] [Google Scholar]
- 37.Wei H-T, Hsu J-W, Huang K-L, et al. Timing of the Diagnoses of Attention Deficit Hyperactivity Disorder and Autism Spectrum Disorder in Taiwan. J Autism Dev Disord. 2021;51:790–7. doi: 10.1007/s10803-018-3655-1. [DOI] [PubMed] [Google Scholar]
- 38.Clift DE, Thorn RJ, Passarelli EA, et al. Effects of embryonic cyclosporine exposures on brain development and behavior. Behav Brain Res. 2015;282:117–24. doi: 10.1016/j.bbr.2015.01.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Marder W, Ganser MA, Romero V, et al. In utero azathioprine exposure and increased utilization of special educational services in children born to mothers with systemic lupus erythematosus. Arthritis Care Res (Hoboken) 2013;65:759–66. doi: 10.1002/acr.21888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Räikkönen K, Gissler M, Kajantie E. Associations Between Maternal Antenatal Corticosteroid Treatment and Mental and Behavioral Disorders in Children. JAMA. 2020;323:1924–33. doi: 10.1001/jama.2020.3937. [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.
Supplementary Materials
Data Availability Statement
Data may be obtained from a third party and are not publicly available.