Cumulative Social Disadvantage and Disease Activity in Juvenile Idiopathic Arthritis: A Childhood Arthritis and Rheumatology Research Alliance Registry Study

William Daniel Soulsby; John Boscardin; Andrea Knight; Daniel B Horton; Karine Toupin‐April; Emily von Scheven; the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry Investigators and for the CARRA Health Equity Work Group

doi:10.1002/acr.80051

. 2026 Apr 30;78(10):1552–1563. doi: 10.1002/acr.80051

Cumulative Social Disadvantage and Disease Activity in Juvenile Idiopathic Arthritis: A Childhood Arthritis and Rheumatology Research Alliance Registry Study

William Daniel Soulsby ^1,^✉, John Boscardin ¹, Andrea Knight ², Daniel B Horton ^3,^4,⁵, Karine Toupin‐April ^6,^7,⁸, Emily von Scheven ¹; the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry Investigators and for the CARRA Health Equity Work Group

PMCID: PMC13613256 PMID: 41931026

Abstract

Objective

Social determinants of health (SDOH) contribute to juvenile idiopathic arthritis (JIA) disparities, but most studies have assessed SDOH independently rather than cumulatively across individual, family, and neighborhood levels. Using a socioecological framework, we investigated the relationship among cumulative social disadvantage, neighborhood disadvantage, and JIA disease activity.

Methods

We conducted a retrospective cohort study using the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry (2015–2022). Individual‐ and family‐level SDOH were combined into a cumulative social disadvantage score (range: 0–3), with one point each for public or no insurance, family income less than $50,000 per year, and guardian education of high school level or less. Neighborhood disadvantage was measured using Area Deprivation Index (quartiles). The primary outcome was active disease by clinical Juvenile Arthritis Disease Activity‐10 (cJADAS‐10). Mixed effects models were generated to assess associations adjusted for demographic and clinical covariates, and causal mediation analysis evaluated whether cumulative social disadvantage mediated the relationship between neighborhood disadvantage and disease activity.

Results

Among 9,609 children with JIA with a median follow‐up time of 2.1 (interquartile range 0.3–5.8) years, 39.1% had a cumulative social disadvantage score greater than 0 with higher scores correlating with greater neighborhood disadvantage. In adjusted analysis, cumulative social disadvantage was associated with higher odds of active disease (adjusted odds ratio 2.10, 95% confidence interval [CI] 1.73–2.53 for a score of 3 vs 0). A total of 87% (95% CI 38–100) of the effect of neighborhood disadvantage was mediated through cumulative social disadvantage.

Conclusion

Cumulative social disadvantage is strongly associated with active JIA and mediates much of the effect of neighborhood disadvantage. Interventions addressing multilevel SDOH will be essential to reducing JIA health disparities.

graphic file with name ACR-78-1552-g002.webp

graphic file with name ACR-78-1552-g004.webp

INTRODUCTION

Juvenile idiopathic arthritis (JIA) is the most common rheumatic disease affecting children, with an estimated incidence between 1.6 and 23 cases per 100,000 children per year globally. ¹ Poorly treated JIA can lead to permanent joint deformities and predispose those with this condition to a lifetime of chronic pain with a significant loss in function. ² , ³ Improvements in therapies, including biologic disease modifying anti‐rheumatic drugs (bDMARD), have led to significant improvements in long‐term outcomes for children living with JIA. ⁴

SIGNIFICANCE & INNOVATIONS.

This is the first study in juvenile idiopathic arthritis (JIA) to evaluate social determinants of health (SDOH) across multiple socioecological levels (ie, individual, family, and community level) using a single analytic framework.
In a large US cohort of children with JIA, cumulative social disadvantage (measuring income, insurance, and education) was strongly associated with active JIA.
We identified that the effect of neighborhood‐level deprivation (as measured by Area Deprivation Index) is largely mediated by more proximal individual‐ and family‐level SDOH.
This work provides new evidence to guide the development of multilevel interventions aimed at reducing health disparities in JIA.

However, multiple health disparities have been demonstrated in children with JIA, including higher disease activity among individuals who are economically disadvantaged ⁵ , ⁶ , ⁷ , ⁸ , ⁹ and delays in treatment among those who are publicly insured. ¹⁰ In most health disparities research in JIA, social determinants of health (SDOH) are studied as independent risk factors. This has yielded inconsistent relationships between SDOH and JIA outcomes. However, SDOH more likely work together to influence patient outcomes; therefore, the presence of multiple risk factors associated with social disadvantage portends worse outcomes as compared with a single risk factor. ¹¹ , ¹² The socioecological model is a framework that acknowledges multiple levels of social influence, including individual, family, community, and societal levels, which converge and interact to create and perpetuate health disparities. ¹³ An analytic approach that acknowledges this multilevel complexity is needed to further elucidate causes of health disparities in JIA that can inform intervention strategies. We previously used such an approach to investigate cumulative social disadvantage in childhood arthritis in a nationally representative data set (the National Survey of Children's Health). ¹⁴ In that study, successive exposure to cumulative social disadvantage, comprised of individual‐ and family‐level SDOH, was associated with higher odds of an arthritis diagnosis and higher disease activity. However, childhood arthritis diagnoses and disease severity in this data set were guardian‐reported, limiting conclusions with respect to JIA specifically.

Therefore, we aimed to assess individual, family, and community‐level SDOH and their associations with disease activity among a cohort of patients with JIA in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry. We generated a cumulative social disadvantage score comprising household (including individual‐ and family‐level) SDOH and separately analyzed Area Deprivation Index (ADI) at the community level. We hypothesized that, among children with JIA, those with higher exposure to cumulative social disadvantage and higher neighborhood disadvantage (ADI) would have higher disease activity across long‐term follow‐up. Informed by associations demonstrated among the cumulative social disadvantage score, neighborhood disadvantage, and disease activity, we subsequently performed causal mediation analysis to investigate neighborhood disadvantage as a mediator of the effect of the cumulative social disadvantage score on active disease.

METHODS

Study design and setting

We performed a retrospective cohort study using data from the CARRA Registry, a prospective, North American multicenter registry for children with rheumatic diseases, including JIA. ¹⁵ We queried the registry for all patients with JIA and analyzed deidentified demographic and clinical data. See Appendix A for the CARRA Registry Site Principal Investigators, Sub‐Investigators, and Research Coordinator.

Analytic sample

The sample population includes all eligible children with JIA enrolled in the CARRA Registry between July 2015 and January 2022 who reside in the United States. Registry eligibility requires a diagnosis of JIA by a pediatric rheumatologist according to the International League of Associations for Rheumatology (ILAR) classification criteria. ¹⁶ Patients enrolled in the registry represent a convenience sample of patients treated at registry sites. Each patient in the CARRA Registry completes an enrollment visit followed by follow‐up visits every six months. Additional inclusion criteria for our study included at least one visit with a complete clinical Juvenile Arthritis Disease Activity‐10 (cJADAS‐10) score (see section “Primary Outcome”). All available registry visits per participant were included for longitudinal analyses.

Primary exposures

Our aim was to study SDOH available in the CARRA Registry that represented different social strata as conceptualized by our socioecological model (Figure 1A). We conceptualized individual‐ and family‐level risk factors as operating on the same strata (eg, household) given that traditional individual‐level risk factors, such as insurance and income, are informed by family‐level factors in a pediatric population (eg, parental income and insurance). One primary exposure variable was the cumulative social disadvantage score (range 0–3) that comprised available individual‐ and family‐level SDOH in the CARRA Registry, including insurance status (one point for public or no insurance), income level (one point for self‐reported family income level of <$50,000 per year), and guardian education (one point for high school level education or less). These three SDOH were chosen because of their availability in the registry and similarity to the parameters chosen in our previous work. ¹⁴ Adverse childhood experiences (ACEs) were omitted from this iteration of the cumulative social disadvantage score because they were not available in the registry. Our other primary exposure variable was ADI, a composite measure of social and economic deprivation at the neighborhood (census tract) level, ¹⁷ to represent neighborhood disadvantage. ADI was collected and grouped into quartiles to make comparisons between groups. This community‐level SDOH has already been studied in previous pediatric ¹⁸ and adult rheumatology ¹⁹ populations. Societal‐level factors, hypothesized to impact our outcomes, could not be measured. These exposures were assessed at the enrollment registry visit; therefore, changes in SDOH over time were not able to be assessed.

(Left) The socioecological model for SDOH applied to this study. Individual/family‐level factors (income, insurance, education) were combined into a cumulative social disadvantage score; community‐level factors were measured by the ADI, a composite measure of neighborhood‐level deprivation encompassing domains including housing, education, employment, and income and geocoded to the census‐tract level. Societal‐level factors were not measured in this study. (Right) A DAG demonstrating proposed relationships between community‐level SDOH (ADI) and individual/family‐level SDOH (cumulative social disadvantage) and active JIA (cJADAS‐10). The dashed line references the proposed mediating relationship under study. ADI, Area Deprivation Index; cJADAS‐10, Clinical Juvenile Arthritis Disease Activity Score 10; DAG, directed acyclic graph; JIA, juvenile idiopathic arthritis; SDOH, social determinants of health. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/doi/10.1002/acr.80051/abstract.

Primary outcome

Our primary outcome was the presence of active disease by the cJADAS‐10, ²⁰ a composite measure including Physician Global Assessment (PGA) (0–10), Parent or Patient Global Assessment (0–10), and active joint count (0–10). cJADAS‐10 scores are collected at a per visit level in the CARRA Registry with study visits occurring every six months. Active disease by cJADAS‐10 was defined as a score greater than 1 for oligoarticular JIA and greater than 2.5 for all other JIA subtypes. ²¹ All available cJADAS‐10 scores at enrollment and follow‐up visits were included in our analysis; visits in which cJADAS components were missing were not included.

Secondary outcome

The presence of functional disability at any point during follow‐up was defined by a Child Health Assessment Questionnaire (CHAQ) ²² score greater than 0. All available CHAQ scores at enrollment and follow‐up visits were included in our analysis.

Study covariates

Demographic covariates assessed included age at enrollment into the registry, self‐identified race and/or ethnicity (White, Black, Hispanic, Asian, other race, missing, or more than one race), and sex (female or male). Disease‐related covariates included (1) JIA subtype (by ILAR criteria); (2) time from symptom onset to JIA diagnosis; (3) time from diagnosis; (4) year of diagnosis; (5) the presence of any secondary rheumatologic disease; (6) ever use of a conventional synthetic DMARD (csDMARD), bDMARD (biologic DMARD), or JAK inhibitors at enrollment or across follow‐up; (7) any previous glucocorticoid use at enrollment or across follow‐up; and (8) current glucocorticoid use.

Statistical analysis

Summary statistics were calculated, including median and interquartile range (IQR), to describe the demographic, disease‐related, and SDOH variables of the cohort. Comparisons of these variables by the cumulative social disadvantage score were tested by chi‐squared testing or Kruskal‐Wallis testing, as appropriate. Univariable and multivariable (adjusting for all covariates—see section “Study Covariates”) mixed effects logistic regression models were generated to estimate the association of cumulative social disadvantage and neighborhood disadvantage with active disease as a binary outcome across registry follow‐up. Univariable models included fixed effects for the categorical cumulative social disadvantage score and a random effect for participant to account for intrasubject correlation across longitudinal measurements. Multivariable models additionally included fixed effects for all covariates, as specified previously (see section “Study Covariates”). All models were estimated with robust clustered SEs at the individual level to guard against model misspecification. ²³ Similar models were generated to estimate the association of cumulative social disadvantage and neighborhood disadvantage with the presence of functional disability across registry follow‐up. Because of significant missingness of individual components of the cumulative social disadvantage score (income 24.6%; guardian education: 15.7%; insurance: 0.2%) and ADI (17.2%), chained equation multiple imputation ²⁴ was used (N = 50 cycles) with the cumulative social disadvantage score and all multivariable model covariates in the data set used to inform our imputation. Final models used imputed data. We tested for interaction between the cumulative social disadvantage score and race and between the cumulative social disadvantage score and neighborhood disadvantage and found no significant interactions; therefore no interaction terms were included in the final models. All analyses were conducted in STATA version 18. ²⁵ The University of California, San Francisco's institutional review board determined our study to be exempt from review (#22‐36233).

Sensitivity analysis

As a sensitivity analysis, we generated similar univariable and multivariable mixed effects models estimating the association between cumulative social disadvantage and neighborhood disadvantage on active disease at different cJADAS‐10 thresholds, including moderate disease activity or higher (oligoarticular JIA with cJADAS >4 and in all other JIA categories >5) and severe disease activity (oligoarticular JIA with cJADAS >12 and in all other JIA categories >16). ²¹

Causal mediation analysis

Based on our theoretical framework of the relationships among cumulative social disadvantage, neighborhood disadvantage (ADI), and disease activity (Figure 1B), we performed a causal mediation analysis to investigate if there was a significant mediating relationship between the cumulative social disadvantage score (individual‐ and family‐level SDOH) and neighborhood disadvantage (community‐level SDOH) on disease activity. Therefore, we tested whether the cumulative social disadvantage score was a mediator of the effect of neighborhood disadvantage on disease activity. The average causal mediation effects, along with the total effects and the proportion of the effect via the proposed mediator, were estimated using linear probability models ²⁶ with robust clustered SEs at the subject level. This causal mediation analysis adjusted for other demographic and disease‐related covariates in our models. To understand if specific components of the cumulative social disadvantage score (eg, low income, underinsured status, or guardian education level) were individual mediators of the relationship between neighborhood disadvantage and disease activity, analogous causal mediation analyses were performed that substituted these components as the proposed mediator and adjusted for the other components.

RESULTS

Demographic and clinical characteristics of the cohort

In total, 9,609 participants meeting inclusion criteria were identified (Table 1) among 10,225 who were eligible in the registry. Most of the cohort had a cumulative social disadvantage score of 0 (60.9%) with lower proportions for a score of 1, 2, and 3 (21.3%, 12.9%, and 4.9%, respectively). The median time from diagnosis was 2.1 years (IQR 0.3–5.8 years). The most common subtypes of arthritis represented were oligoarticular JIA (35.6%) closely followed by rheumatoid factor (RF) negative polyarticular JIA (29.5%). bDMARD and csDMARD use were common across the cohort (68.6% and 77.9%, respectively). Glucocorticoid use was uncommon overall but increased in frequency both at enrollment and at any point across follow‐up among those reporting the highest cumulative social disadvantage score (use at enrollment 13.0% and at any time across follow‐up 42.6% for a score of 3). Of the parameters of the cumulative social disadvantage score, having public or no insurance was the most common, with 26.3% of the cohort reporting, followed by low income (19.1%) and low guardian education (16.4%). Yearly household income had a high degree of missingness (24.6%) as did guardian education (15.7%), with little missingness in insurance status (0.2%). Those who had a higher cumulative social disadvantage score were more likely to live in an area of higher neighborhood disadvantage (51.6% of those with a score of 3 lived in the highest ADI quartile, as compared with only 11.7% of those with a score of 0). Participants identifying as Black or Hispanic were more likely to have higher cumulative social disadvantage scores (11.9% Black with a score of 3 vs 1.8% with a score of 0; 21.3% Hispanic with a score of 3 vs 3.4% with a score of 0).

Table 1.

Demographic and clinical characteristics of pediatric patients with JIA in the CARRA Registry (2015–2022)^*

Demographic and clinical characteristics	Cumulative Social Disadvantage Score ^a
Demographic and clinical characteristics	0	1	2	3	Total	Test ^b
Patients, N (%)	5,849 (60.9)	2,048 (21.3)	1,243 (12.9)	469 (4.9)	9,609 (100.0)
Median age at enrollment (IQR), y	11.7 (7.3–15.1)	12.0 (7.6–15.4)	11.6 (7.2–15.1)	10.5 (6.1–14.2)	11.7 (7.3–15.1)	<0.001
Female, n (%)	4,117 (70.4)	1,433 (70.0)	850 (68.4)	317 (67.6)	6,717 (69.9)	0.360
Male, n (%)	1,732 (29.6)	615 (30.0)	393 (31.6)	152 (32.4)	2,892 (30.1)
Race and/or ethnicity, n (%)
White	4,783 (81.8)	1,442 (70.4)	726 (58.4)	246 (52.5)	7,197 (74.9)	<0.001
Hispanic	199 (3.4)	212 (10.4)	223 (17.9)	100 (21.3)	734 (7.6)
Multiracial	340 (5.8)	138 (6.7)	108 (8.7)	39 (8.3)	625 (6.5)
Black	108 (1.8)	105 (5.1)	109 (8.8)	56 (11.9)	378 (3.9)
Asian	183 (3.1)	45 (2.2)	21 (1.7)	6 (1.3)	255 (2.7)
Other	114 (1.9)	52 (2.5)	45 (3.6)	22 (4.7)	233 (2.4)
Refuse/missing	122 (2.1)	54 (2.6)	11 (0.9)	0 (0.0)	187 (1.9)
National ADI ^c quartiles, n (%)
Q1 (lowest)	1,701 (29.1)	269 (13.1)	102 (8.2)	23 (4.9)	2,095 (21.8)	<0.001
Q2	1,334 (22.8)	411 (20.1)	183 (14.7)	61 (13.0)	1,989 (20.7)
Q3	1,027 (17.6)	489 (23.9)	313 (25.2)	91 (19.4)	1,920 (20.0)
Q4 (highest)	687 (11.7)	546 (26.7)	481 (38.7)	242 (51.6)	1,956 (20.4)
Missing	1,100 (18.8)	333 (16.3)	164 (13.2)	52 (11.1)	1,649 (17.2)
Median national ADI score (IQR)	29.0 (15.0–47.0)	46.0 (27.0–67.0)	57.0 (36.0–77.0)	66.0 (41.0–84.0)	37.0 (20.0–60.0)	<0.001
Family income level, n (%)
<$25,000	0 (0.0)	110 (5.4)	370 (29.8)	243 (51.8)	723 (7.5)	<0.001
$25,000–49,999	0 (0.0)	354 (17.3)	531 (42.7)	226 (48.2)	1,111 (11.6)
$50,000–74,999	676 (11.6)	405 (19.8)	85 (6.8)	0 (0.0)	1,166 (12.1)
$75,000–99,999	790 (13.5)	238 (11.6)	33 (2.7)	0 (0.0)	1,061 (11.1)
$100,000–150,000	1,417 (24.3)	241 (11.8)	17 (1.4)	0 (0.0)	1,675 (17.4)
Above $150,000	1,378 (23.6)	113 (5.5)	5 (0.4)	0 (0.0)	1,496 (15.6)
Prefer not to answer	789 (13.5)	305 (14.9)	99 (8.0)	0 (0.0)	1,193 (12.4)
Unknown	792 (13.6)	280 (13.7)	103 (8.3)	0 (0.0)	1,175 (12.2)
Insurance status, n (%)
Private	5,056 (86.4)	900 (43.9)	191 (15.4)	20 (4.3)	6,167 (64.2)	<0.001
Public	0 (0.0)	859 (41.9)	952 (76.6)	437 (93.2)	2,248 (23.4)
Other insurance	792 (13.5)	212 (10.4)	62 (5.0)	0 (0.0)	1,066 (11.1)
Uninsured	0 (0.0)	64 (3.1)	35 (2.8)	12 (2.6)	111 (1.2)
Missing/unknown	1 (0.0)	13 (0.6)	3 (0.2)	0 (0.0)	17 (0.2)
Guardian education level, n (%)
Elementary/middle school (grade 1 − 8)	0 (0.0)	41 (2.1)	47 (4.0)	42 (9.0)	130 (1.4)	<0.001
Some high school (grade 9–11)	0 (0.0)	40 (2.0)	72 (6.1)	65 (13.9)	177 (1.9)
Graduated high school (grade 12 or GED)	0 (0.0)	489 (25.0)	420 (35.9)	362 (77.2)	1,271 (13.9)
College (1–4 y college, junior college or technical school)	2,657 (48.0)	848 (43.3)	438 (37.4)	0 (0.0)	3,943 (43.2)
Graduate school (eg, Masters, Doctorate, or professional degree)	1,889 (34.1)	234 (12.0)	51 (4.4)	0 (0.0)	2,174 (23.8)
Prefer not to answer	986 (17.8)	305 (15.6)	143 (12.2)	0 (0.0)	1,434 (15.7)
Components of the Cumulative Social Disadvantage Score, n (%)
Income <$50,000/year	0 (0.0)	464 (22.7)	901 (72.5)	469 (100.0)	1,834 (19.1)	<0.001
Guardian education ≤ high school	0 (0.0)	570 (27.8)	539 (43.4)	469 (100.0)	1,578 (16.4)	<0.001
Public or no insurance	0 (0.0)	1,014 (49.5)	1,046 (84.2)	469 (100.0)	2,529 (26.3)	<0.001
Median year of diagnosis (IQR)	2015 (2012–2018)	2016 (2012–2018)	2016 (2012–2018)	2016 (2013–2018)	2015 (2012–2018)	0.003
Median days from symptom onset to diagnosis (IQR)	137.0 (62.0–312.0)	150.0 (63.0–342.5)	157.0 (66.0–371.0)	132.0 (61.0–330.0)	143.0 (63.0–328.0)	0.526
Median years from diagnosis (IQR)	2.1 (0.3–5.9)	2.1 (0.3–5.9)	1.8 (0.2–5.4)	1.7 (0.2–5.1)	2.1 (0.3–5.8)	0.043
JIA subtype, n (%)
Enthesitis related arthritis	661 (11.3)	209 (10.2)	111 (8.9)	30 (6.4)	1,011 (10.5)	<0.001
Oligoarthritis	2,214 (37.9)	692 (33.8)	362 (29.1)	148 (31.6)	3,416 (35.6)
Polyarthritis (RF+)	302 (5.2)	144 (7.0)	115 (9.3)	49 (10.4)	610 (6.3)
Polyarthritis (RF−)	1,679 (28.7)	587 (28.7)	404 (32.5)	160 (34.1)	2,830 (29.5)
Psoriatic arthritis	441 (7.5)	161 (7.9)	84 (6.8)	17 (3.6)	703 (7.3)
Systemic arthritis	386 (6.6)	179 (8.7)	123 (9.9)	57 (12.2)	745 (7.8)
Undifferentiated arthritis	166 (2.8)	76 (3.7)	44 (3.5)	8 (1.7)	294 (3.1)
Any secondary rheumatic disease	5,611 (95.9)	1957 (95.6)	1,191 (95.8)	448 (95.5)	9,207 (95.8)	0.888
Medication use, n (%)
Any bDMARD use	3,895 (66.6)	1,448 (70.7)	902 (72.6)	349 (74.4)	6,594 (68.6)	<0.001
Any csDMARD use	4,517 (77.2)	1,585 (77.4)	1,007 (81.0)	381 (81.2)	7,490 (77.9)	0.008
Any JAK inhibitor use	250 (4.3)	83 (4.1)	37 (3.0)	20 (4.3)	390 (4.1)	0.214
Glucocorticoid use at enrollment visit	435 (7.4)	181 (8.8)	141 (11.3)	61 (13.0)	818 (8.5)	<0.001
Any glucocorticoid use during follow‐up	1,874 (32.0)	712 (34.8)	482 (38.8)	200 (42.6)	3,268 (34.0)	<0.001
Components of the cJADAS‐10
Median Physician Global Assessment (IQR)	1.0 (0.0–3.0)	1.0 (0.0–3.0)	2.0 (0.0–4.0)	1.5 (0.0–3.5)	1.0 (0.0–3.0)	<0.001
Median Patient/Parent Global Assessment (IQR)	1.0 (0.0–3.0)	1.0 (0.0–4.0)	2.0 (0.0–5.0)	2.0 (0.0–5.0)	1.0 (0.0–4.0)	<0.001
Active joint count (IQR)	0.0 (0.0–2.0)	0.0 (0.0–2.0)	1.0 (0.0–4.0)	1.0 (0.0–3.0)	0.0 (0.0–2.0)	<0.001
Median cJADAS score at enrollment (IQR)	3.5 (0.5–8.0)	4.0 (1.0–9.0)	6.0 (1.5–12.0)	5.0 (1.5–12.0)	4.0 (1.0–9.0)	<0.001
Active disease by cJADAS at enrollment n (%)	3,460 (59.2)	1,293 (63.1)	883 (71.0)	325 (69.3)	5,961 (62.0)	<0.001
Median CHAQ score at enrollment (IQR)	0.0 (0.0–0.5)	0.2 (0.0–0.8)	0.4 (0.0–1.0)	0.4 (0.0–1.0)	0.1 (0.0–0.6)	<0.001
Functional disability by CHAQ at enrollment n (%)	2,772 (48.9)	1,166 (58.5)	819 (67.5)	312 (67.7)	5,069 (54.3)	<0.001

Open in a new tab

ANOVA, analysis of variance; ADI, Area Deprivation Index; bDMARD, biologic disease modifying anti‐rheumatic drug; CARRA, Childhood Arthritis and Rheumatology Research Alliance; csDMARD, conventional synthetic DMARD; CHAQ, Child Health Assessment Questionnaire; cJADAS‐10, clinical Juvenile Arthritis Disease Activity‐10; GED, General Education Development Test; IQR, interquartile range; JIA, juvenile idiopathic arthritis; Q, quartile; RF, rheumatoid factor.

^{^a}

Cumulative social disadvantage score (range 0–3) is calculated as follows: 1 point for self‐reported family income level <$50,000/year, 1 point for public or no insurance, and 1 point for guardian level education of high school or less.

^{^b}

Computed using descriptive statistics, such as chi‐squared testing, one‐way ANOVA, or Kruskal‐Wallis testing, as appropriate.

^{^c}

ADI is a composite measure of neighborhood‐level deprivation that comprises 17 indicators of neighborhood deprivation, including income, housing, employment, and education using US Census data and geocoded to the census‐tract level.

Disease activity by cJADAS‐10

The median cJADAS‐10 score at enrollment was 4.0 (IQR 1.0–9.0), with 62% having active cJADAS‐10 score at enrollment (Table 1). In unadjusted analyses of the effects of individual‐ and family‐level SDOH, there was a successive increase in the odds of active disease across the registry follow‐up (highest for a cumulative social disadvantage score of 3: odds ratio [OR] 2.15, 95% confidence interval [CI] 1.73–2.68) (Table 2). After adjustment for all covariates, the relationship among cumulative social disadvantage and disease activity remained strong (highest for a cumulative social disadvantage score of 3: adjusted odds ratio [aOR] 2.17, 95% CI 1.77–2.65). After adjustment for all covariates and neighborhood disadvantage, these associations were minimally changed (highest for a cumulative social disadvantage score of 3: aOR 2.10, 95% CI 1.73–2.53). In unadjusted analyses of the effects of community‐level SDOH, there was a successive increase in the odds of active disease across registry follow‐up with successive increases in neighborhood disadvantage that corresponded to increased neighborhood‐level disadvantage (OR for ADI quartile 4: 1.45, 95% CI 1.18–1.77). After adjustment for all covariates, these associations were minimally changed (aOR for ADI quartile 4: 1.41, 95% CI 1.16–1.71). After adjustment for all covariates and the cumulative social disadvantage score, there was no longer a statistically significant relationship between neighborhood disadvantage and active disease (aOR for ADI quartile 4: 1.11, 95% CI 0.92–1.35).

Table 2.

Unadjusted and adjusted logistic regression models assessing the association between the cumulative social disadvantage score, race and/or ethnicity, and ADI and any disease activity (by cJADAS‐10) across registry follow‐up^*

Social disadvantage, race/ethnicity, ADI, and disease activity	Unadjusted OR (95% CI)	Adjusted ^a OR (95% CI)	Final adjusted ^b OR (95% CI)
Cumulative social disadvantage score
0	Ref	Ref	Ref
1	1.34 (1.19–1.51)	1.32 (1.18–1.48)	1.30 (1.16–1.45)
2	1.81 (1.58–2.07)	1.84 (1.61–2.10)	1.79 (1.57–2.04)
3	2.15 (1.73–2.68)	2.17 (1.77–2.65)	2.10 (1.73–2.53)
ADI
1 (lowest)	Ref	Ref	Ref
2	1.14 (1.00–1.31)	1.15 (1.01–1.32)	1.09 (0.95–1.24)
3	1.24 (1.05–1.46)	1.23 (1.04–1.45)	1.08 (0.91–1.28)
4 (highest)	1.45 (1.18–1.77)	1.41 (1.16–1.71)	1.11 (0.92–1.35)

Open in a new tab

cJADAS‐10 defined in participants with oligoarticular JIA as >1.1 and in all other JIA categories as >2.5. ADI, Area Deprivation Index; bDMARD, biologic disease modifying anti‐rheumatic drug; CI, confidence interval; cJADAS‐10, clinical Juvenile Arthritis Disease Activity Score‐10; csDMARD, conventional synthetic DMARD; JIA, juvenile idiopathic arthritis; OR, odds ratio; Ref, reference.

^{^a}

Adjusted for age at enrollment, time from symptom onset to diagnosis, year of diagnosis, sex, JIA category, presence of secondary rheumatic disease, medication use during follow‐up (bDMARD, csDMARD, small molecule), current glucocorticoid use, any glucocorticoid use during follow‐up.

^{^b}

Adjusted for the same covariates, including either the cumulative social disadvantage score or ADI.

To understand if specific components of the cumulative social disadvantage score were more strongly associated with persistent disease activity across registry follow‐up, we generated univariable and multivariable models in which score components were individually assessed. In these multivariable models, low insurance (aOR 1.53, 95% CI 1.35–1.75), low guardian education (aOR 1.35, 95% CI 1.22–1.49), and low income (aOR 1.69, 95% CI 1.51–1.88) were associated with persistent disease activity (Table 3). Neighborhood disadvantage remained statistically significant in adjusted analysis in these multivariable models (aOR for ADI quartile 4: 1.41, 95% CI 1.16–1.71).

Table 3.

Unadjusted and adjusted logistic regression models assessing the association between the individual components of the cumulative social disadvantage score, race and/or ethnicity, and ADI and any active disease (by cJADAS‐10) across registry follow‐up^*

Social disadvantage components, race/ethnicity, ADI, and disease activity	Unadjusted OR (95% CI)	Adjusted ^a OR (95% CI)
Low insurance	1.62 (1.41–1.85)	1.53 (1.35–1.75)
Low guardian education	1.51 (1.33–1.70)	1.35 (1.22–1.49)
Low income	1.73 (1.53–1.96)	1.69 (1.51–1.88)
ADI
1 (lowest)	Ref	Ref
2	1.14 (1.00–1.31)	1.15 (1.01–1.32)
3	1.24 (1.05–1.46)	1.23 (1.04–1.45)
4 (highest)	1.45 (1.18–1.77)	1.41 (1.16–1.71)

Open in a new tab

^{^a}

As a sensitivity analysis, similar multivariable regression models were generated to estimate the association of the cumulative social disadvantage score and neighborhood disadvantage with persistent disease activity using different disease activity cut‐offs, including moderate disease activity or higher and severe disease activity or higher (Table 4). There remained a successive increase in the odds of the cumulative social disadvantage score and disease activity using these thresholds with no qualitative differences identified (for a score of 3, any disease activity: aOR 2.10, 95% CI 1.73–2.53; moderate disease activity or higher: aOR 2.16, 95% CI 1.82–2.56; severe disease activity or higher: aOR 2.09, 95% CI 1.61–2.72). The association between neighborhood disadvantage and disease activity was attenuated in multivariable analysis at every threshold (for ADI quartile 4, any disease activity: aOR 1.11, 95% CI 0.92–1.35; moderate disease activity or higher: aOR 1.03, 95% CI 0.88–1.20; severe disease activity or higher: aOR 1.07, 95% CI 0.83–1.38). As a separate sensitivity analysis, our univariable and multivariable primary models were generated using our final imputed data set versus complete case series analysis with no significant differences noted in our measures of association (Supplemental Table 1).

Table 4.

Adjusted logistic regression models assessing the association between the cumulative social disadvantage score, race and/or ethnicity, and ADI and disease activity (by varying cJADAS‐10 cutoffs) across registry follow‐up)^*

Social disadvantage, race/ethnicity, ADI, and disease activity by cJADAS‐10 Cutoffs	Any active disease, ^a N = 5,961, Adjusted OR (95% CI)	Moderate disease activity or higher, ^b N = 4,154, Adjusted OR (95% CI)	Severe disease activity or higher, ^c N = 1,026, Adjusted OR (95% CI)
Cumulative social disadvantage score
0	Ref	Ref	Ref
1	1.30 (1.16–1.45)	1.33 (1.18–1.50)	1.33 (1.11–1.58)
2	1.79 (1.57–2.04)	1.91 (1.69–2.17)	1.83 (1.49–2.24)
3	2.10 (1.73–2.53)	2.16 (1.82–2.56)	2.09 (1.61–2.72)
ADI
1 (lowest)	Ref	Ref	Ref
2	1.09 (0.95–1.24)	1.06 (0.93–1.22)	1.14 (0.93–1.40)
3	1.08 (0.91–1.28)	1.09 (0.94–1.26)	1.07 (0.87–1.33)
4 (highest)	1.11 (0.92–1.35)	1.03 (0.88–1.20)	1.07 (0.83–1.38)

Open in a new tab

OR adjusted for age at enrollment, time from symptom onset to diagnosis, year of diagnosis, sex, JIA category, presence of secondary rheumatic disease, medication use during follow‐up (bDMARD, csDMARD, small molecule), current glucocorticoid use, any glucocorticoid use during follow‐up. ADI, Area Deprivation Index; bDMARD, biologic disease modifying anti‐rheumatic drug; CI, confidence interval; cJADAS‐10, clinical Juvenile Arthritis Disease Activity Score‐10; csDMARD, conventional synthetic DMARD; JIA, juvenile idiopathic arthritis; OR, odds ratio; Ref, reference.

^{^a}

Defined in participants with oligoarticular JIA as cJADAS >1.1 and in all other JIA categories as >2.5.

^{^b}

Defined in participants with oligoarticular JIA as cJADAS >4 and in all other JIA categories as >5

^{^c}

Definied in participants with oligoarticular JIA as cJADAS >12 and in all other JIA categories as >16

Causal mediation analysis

Comparing the most disadvantaged with least disadvantaged neighborhoods (ADI quartile 4 vs quartile 1), the total effect of neighborhood disadvantage on disease activity (4.8%, 95% CI 2.7–7.1) can be separated into a 0.7% (95% CI −1.9 to 3.2) direct effect and 4.2% (95% CI 3.0–5.4) indirect effect mediated through the cumulative social disadvantage score. Therefore, 87% (95% CI 38–100) of the effect of neighborhood disadvantage on disease activity is mediated through the cumulative social disadvantage score (P < 0.001) (Figure 2). When performing similar mediation analyses substituting the individual components of the score, we found that low income mediated 72% (95% CI 30–100) of the effect of neighborhood disadvantage on disease activity (P = 0.001), whereas public or no insurance mediated 52% (95% CI 21–83) of this relationship (P = 0.001), and guardian education mediated 27% (95% CI 5–48) of this relationship (P = 0.016).

Functional disability by CHAQ

Functional disability as defined by a CHAQ score greater than 0 was common at enrollment at 54.3% (Table 1). Across registry follow‐up in both unadjusted and adjusted analyses, an increase in the cumulative social disadvantage score was associated with a successive increase in the odds of functional disability (for a score of 3: aOR 2.86, 95% CI 2.14–3.81). In unadjusted analysis, a similar relationship was found between neighborhood disadvantage and persistent functional disability (for ADI quartile 4: aOR 2.32, 95% CI 1.74–3.10). This relationship was attenuated but remained statistically significant in adjusted analysis (for ADI quartile 4: aOR 1.38, 95% CI 1.03–1.83) (Supplemental Table 2).

DISCUSSION

In the largest cohort study of US children with JIA to date focused on health disparities, we found that cumulative social disadvantage, a composite measure of individual‐ and family‐level SDOH, was strongly and dose‐dependently associated with persistent disease activity and functional disability across CARRA registry follow‐up. This association persisted after adjustment for demographic and disease‐related factors, suggesting that cumulative social disadvantage exerts an independent effect on JIA outcomes. Additionally, we identified neighborhood disadvantage as a statistically significant predictor of disease activity across registry follow‐up, although most of the effect of this community‐level SDOH on disease activity is mediated through more proximal individual‐ and family‐level SDOH. These findings build on our previous work in a nationally representative data set ¹⁴ by replicating this relationship in a large, rheumatologist‐confirmed JIA cohort and confirm that cumulative social disadvantage is a clinically relevant construct for JIA risk stratification and intervention design.

SDOH rarely act in isolation. Social disadvantages, including poverty, underinsured status, and low education level, may accumulate and have been consistently linked to poorer childhood health, including having a chronic medical condition, and having functional limitations. ¹¹ Previous studies had used combined scoring systems to investigate this concept of cumulative social disadvantage on health outcomes. ²⁷ In a previous study, we created a cumulative social disadvantage score that comprised measures of childhood poverty, insurance status, low guardian education, and high ACE scores, and found an increase in the odds of a childhood arthritis diagnosis among a representative population of US children and higher reported disease severity. ¹⁴ Limitations of this study were its cross‐sectional nature, which could not assess if such disparities were sustained over time, as well as the self‐reported nature of both arthritis diagnosis and severity, which could have been subject to misclassification bias. This study extends our understanding by demonstrating that social disadvantage appears to operate additively or synergistically over time, even within the context of subspecialty care. This underscores the need to address both individual‐level and family‐level risk factors and the overall burden of social disadvantage in pediatric rheumatology.

When considering potential targets for intervention, it is important to consider the ways in which cumulative social disadvantage may operate to worsen JIA health outcomes. Our mediation analysis provides important insight into potential mechanisms. Most of the effect (87%) of neighborhood disadvantage (as measured by ADI) on disease activity was mediated through more proximal individual and family‐level factors. Of these, low family income emerged as the strongest mediator, followed by public or no insurance and low guardian education. Although this suggests that addressing proximal needs may have the greatest impact and will be necessary to successful JIA interventions, it does not completely diminish the importance of community context. Community‐level poverty, defined as 20% or more of the population of a geographic area living below the federal poverty limit, has been demonstrated to be associated with delays in time to first rheumatology appointment and persistent disease activity and functional disability among children with polyarticular JIA. ²⁸ , ²⁹ We hypothesize that community‐level determinants remain critical in shaping JIA outcomes. Pediatric rheumatology care in the United States is concentrated in urban academic centers, ³⁰ often requiring families, especially those in rural and high‐deprivation areas, to travel significant distances to specialty care. ³¹ Such geographic and resource barriers may exacerbate delays in diagnosis, ³² limit timely follow‐up, and constrain access to health care and other ancillary services, such as physical therapy, mental health support, or social work. Moreover, neighborhood‐level factors not captured by ADI, including availability of transportation, local health care infrastructure, community health literacy, and exposure to environmental stressors, may independently influence disease management and outcomes. ³³ Future studies incorporating broader community‐level metrics and health care access geography will be important to further assess the role between community and health disparities in JIA.

The findings of our study underscore the need for multilevel interventions extending beyond the clinic. At a systems level, potential strategies to mitigate the impact of social disadvantage include expanding telehealth infrastructure, ³⁴ efforts to increase recruitment of pediatric rheumatologists to address workforce shortages, ³⁵ establishing satellite clinics or mobile care units to reduce travel burden, ³⁶ and embedding social care services into pediatric rheumatology practices. ³⁷ , ³⁸ Presently among CARRA registry sites, only 31% of pediatric rheumatology clinics have a social worker or psychologist. ³⁹ Patient navigator programs have been successful in diseases such as Type I diabetes ⁴⁰ and piloted in adult rheumatoid arthritis ⁴¹ , ⁴² and may provide both individualized support and stronger links between tertiary care centers and patients’ home communities. Additionally, policy initiatives that improve insurance coverage (including expansion of Medicaid and the Children's Health Insurance Program ⁴³ ), transportation assistance, ⁴⁴ and broadband access ⁴⁵ , ⁴⁶ may also help bridge the gap for families in high‐deprivation areas. Building community partnerships with schools, primary care practices, and community‐based organizations ⁴⁷ could further reinforce disease management support where children live.

The persistence of disparities among patients receiving care at tertiary and quaternary centers (ie, CARRA Registry sites) highlights that access to specialty care alone is insufficient, and delays to diagnosis or pediatric rheumatology care do not fully explain JIA health disparities. Routine SDOH screening in pediatric rheumatology should be a priority, with intervention models designed for scalability across diverse practice settings. The feasibility of SDOH screening in rheumatology clinics has already been demonstrated ⁴⁸ but, to our knowledge, not yet applied to or studied in a pediatric rheumatology population. Our mediation results suggest that pairing community‐level strategies with targeted individual and family support will be essential. Programs such as the Novel Interventions in Child Health model ⁴⁹ , ⁵⁰ or patient navigation programs warrant further evaluation in JIA.

There are several limitations of this study to consider. We had a high degree of missingness of our predictor variables. However, we addressed this with multiple imputations and performed sensitivity analyses comparing complete case versus imputed data models with no substantive quantitative differences. SDOH variables and ADI were measured only once at the enrollment registry visit, preventing assessment of SDOH changes over time. Unmeasured confounding is possible, particularly from societal‐level factors that are difficult to measure, including structural racism and policy environments. Although we chose cJADAS and CHAQ as our outcomes of interest, given their broad applicability to the JIA cohort at large and their use in other JIA trials, they may not fully capture patient experience. Future work will integrate more patient‐reported outcomes. Finally, generalizability may be limited because the CARRA Registry is a convenience sample of patients with access to specialty care and who consent to registry participation and excludes those managed in adult rheumatology or primary care settings.

Despite these limitations, this is the largest North American cohort of patients with JIA analyzed longitudinally with validated measures of disease activity and function focusing on health disparities. Additionally, the CARRA Registry allowed for the analysis of individually reported household‐level SDOH, including family income and guardian education, which are not routinely available in large data sets. Neighborhood‐level proxies for household‐level SDOH, such as ADI, may be more useful in data sets in which these SDOH are not available. Our application of the socioecological model, combined with causal mediation analysis, advances our understanding of how multilevel SDOH influence pediatric rheumatology outcomes. Even with limited heterogeneity in SDOH within this cohort, we identified a clear gradient effect of cumulative social disadvantage on JIA outcomes.

In conclusion, we identified cumulative social disadvantage, comprising individual‐ and family‐level SDOH, as strongly associated with persistent disease activity among a large cohort of US children with JIA. We also identified a mediating relationship between neighborhood disadvantage and cumulative social disadvantage on disease activity. These findings demonstrate that although community‐level SDOH are implicated in JIA health disparities, their influence is primarily mediated through more proximal SDOH and highlights the need for individual‐level intervention to mitigate JIA health disparities. Future directions will expand measurement of community‐level determinants and leverage these findings to develop interventions targeting SDOH through patient navigation, care coordination, and community partnerships. We also plan to create and validate a JIA‐specific social disadvantage screening tool for integration into routine clinical practice. Given the persistence of disparities in JIA despite therapeutic advances, it is imperative that interventions address the broader social context in which children live and receive care. Pediatric rheumatology should move toward integrated care models that recognize and mitigate social disadvantage to achieve equitable outcomes for all children with JIA.

AUTHOR CONTRIBUTIONS

All authors contributed to at least one of the following manuscript preparation roles: conceptualization AND/OR methodology, software, investigation, formal analysis, data curation, visualization, and validation AND drafting or reviewing/editing the final draft. As corresponding author, Dr Soulsby confirms that all authors have provided the final approval of the version to be published and takes responsibility for the affirmations regarding article submission (eg, not under consideration by another journal), the integrity of the data presented, and the statements regarding compliance with institutional review board/Declaration of Helsinki requirements.

Supporting information

Disclosure form.

ACR-78-1552-s003.pdf^{(325.7KB, pdf)}

Supplemental Table 1: Unadjusted and adjusted logistic regression models assessing the association between the cumulative social disadvantage score, race and/or ethnicity, and area deprivation index and disease activity and functional disability (Child Health Assessment Questionnaire [CHAQ] >0) across Registry follow‐up:

ACR-78-1552-s001.docx^{(15.3KB, docx)}

Supplemental Table 2: Unadjusted and adjusted logistic regression models assessing the association between the cumulative social disadvantage score, race and/or ethnicity, and area deprivation index and any disease activity (by Clinical Juvenile Arthritis Disease Activity Score‐10 [cJADAS‐10¹]) across Registry follow‐up using complete case (non‐imputed) data:

ACR-78-1552-s002.docx^{(15.4KB, docx)}

ACKNOWLEDGMENTS

The authors wish to acknowledge CARRA and the ongoing Arthritis Foundation financial support of CARRA. This study used (data and/or biospecimens) collected in the CARRA Registry. The views expressed are the authors’ and do not necessarily represent the view of CARRA. This work could not have been accomplished without the aid of the following organizations: The NIH's National Institute of Arthritis and Musculoskeletal and Skin Diseases and the Arthritis Foundation. We would also like to thank all participants and hospital sites that recruited patients for the CARRA Registry. The authors thank the CARRA Registry site principal investigators, sub‐investigators and research coordinators. Members are listed in Appendix A.

APPENDIX A. CARRA Registry Site Principal Investigators, Sub‐Investigators, and Research Coordinators

The CARRA Registry site principal investigators, sub‐investigators, and research coordinators are as follows: R. Aamir, K. Abulaban, A. Adams, C. Aguiar Lapsia, H. Ahmed, S. Akoghlanian, A. AlBijadi, E. Allenspach, M. Alpizar, G. Amarilyo, M. Amoruso, S. Angeles‐Han, S. Ardoin, S. Armendariz, N. Aviran Dagan, I. Balboni, S. Balevic, S. Ballinger, S. Baluta, L. Barillas‐Arias, M. Basiaga, K. Baszis, M. Becker, A. Begezda, E. Beil, H. Bell‐Brunson, H. Benham, S. Benseler, L. Bermudez‐Santiago, W. Bernal, T. Bigley, C. Bingham, B. Binstadt, C. Black, B. Blackmon, M. Blakley, J. Bohnsack, A. Boneparth, H. Bradfield, J. Bridges, E. Brooks, M. Brothers, D. Brown, H. Brunner, L. Buckley, Mary Buckley, Meredith Buckley, H. Bukulmez, D. Bullock, A. Cancino, S. Canna, L. Cannon, S. Canny, V. Cartwright, E. Chalom, Johanna Chang, Joyce Chang, M. Chang, A. Chang‐Hoftman, A. Chen, P. Chiraseveenuprapund, K. Ciaglia, M. Cidon, D. Co, E. Cohen, R. Connor, K. Cook, A. Cooper, J. Cooper, K. Corbin, C. Correll, R. Cron, M. Curry, A. Dagci, A. Dalrymple, E. Datyner, T. Davis, D. De Ranieri, J. Dean, C. DeCoste, F. Dedeoglu, M. DeGuzman, N. Delnay, E. L. DeSantis, R. Devine, M. Dhalla, A. Dhanrajani, D. Dissanayake, B. Dizon, J. Drew, K. Driest, Q. Du, E. Duncan, K. Dunnock, D. Durkee, J. Dvergsten, A. Eberhard, K. Ede, B. Edelheit, C. Edens, M. Elder, Y. Elzaki, C. Failing, D. Fair, L. Favier, B. Feldman, J. Fennell, I. Ferguson, P. Ferguson, C. Figueroa, E. Flanagan, L. Fogel, E. Fox, M. Fox, L. Franklin, R. Fuhlbrigge, J. Fuller, T. Futch‐West, S. Gagne, M. Geiszler, D. Gerstbacher, M. Gilbert, A.C. Gironella, D. Glaser, I. Goh, S. Gorry, N. Goswami, B. Gottlieb, T. Graham, S. Grevich, T. Griffin, A. Grim, A. Grom, M. Guevara, L. Guzman, T. Hahn, O. Halyabar, E. Hammelev, T. Hammond, S. Haro, J. Harris, O. Harry, J. Hausmann, A. Hay, K. Hays, K. Hayward, L. Henderson, M. Henrickson, A. Hersh, L. Hiraki, M. Hiskey, P. Hobday, C. Hoffart, M. J. Holland, M. Hollander, S. Hong, D. Horton, J. Hsu, A. Huber, J. Huggins, J. Hui‐Yuen, M. Ibarra, A. Imlay, L. Imundo, C. Inman, A. Jackson, K. James, G. Janow, Y. Jiang, L. Johnson, N. Johnson, J. Jones, D. Kafisheh, K. Kaidar, S. Kasinathan, R. Kaur, E. Kessler, B. Kienzle, S. Kim, Y. Kimura, D. Kingsbury, M. Kitcharoensakkul, J. Klauss, K. Klein, M. Klein‐Gitelman, A. Knight, L. Kovalick, D. Krajewski, C. Kremer, T. LaFlam, B. Lang, S. Lapidus, B. Lapin, A. Lasky, E. Lawson, R. Laxer, A. Lee, Patricia Lee, Pui Lee, T. Lee, E. Leisinger, L. Lentini, M. Lerman, Y. Levinsky, D. Levy, S. Li, S. Lieberman, L. Lim, E. Limenis, C. Lin, N. Ling, G. Lionetti, R. Livny, M. Lo, A. Long, M. Lopez‐Peña, D. Lovell, S. Lvovich, A. Lytch, M. Ma, A. Machado, J. MacMahon, J. Madison, M. Mannion, C. Manos, L. Mansfield, B. Marston, K. Marzan, T. Mason, S. Matossian, L. McAllister, K. McBrearty, D. McCurdy, K. McDaniels, J. McDonald, L. McIntosh, E. Meidan, E. Mellins, Z. Mian, P. Miettunen, M. Miller, D. Milojevic, R. Mitacek, R. Modica, S. Mohan, K. Moore, T. Moore, L. Moorthy, J. Moreno, E. Morgan, A. Moyer, B. Murante, A. Murphy, E. Muscal, O. Mwizerwa, A. Najafi, K. Nanda, L. Nassi, S. Nativ, M. Natter, J. Neely, L. Newhall, A. Nuyen, P. Nigrovic, J. Nocton, B. Nolan, A. Nowakowski, K. Nowicki, R. Oakes, E. Oberle, S. Ogbonnaya‐Whittesley, E. Ogbu, M. Oliver, R. Olveda, K. Onel, A. Orandi, J. Padam, N. Pan, J. Pandya, S. Panupattanapong, A. Pappo Toledano, J. Patel, P. Patel, A. Patrick, S. Patrizi, S. Paul, J. Perfetto, M. Perron, M. Peskin, C. Pinotti, L. Ponder, R. Pooni, S. Prahalad, M. Quinlan‐Waters, J. Rafko, H. Rahimi, S. Ramsey, R. Randell, L. Ray, Ann Reed, Annelle Reed, H. Reid, D. Reiff, I. Reyhan, B. Richard, M. Riebschleger, E. Rife, M. Riskalla, A. Robinson, L. Robinson, L. Rodgers, M. Rodriquez, D. Rogers, T. Ronis, A. Rosado, M. Rosenkranz, N. Rosenwasser, H. Rothermel, D. Rothman, E. Rothschild, K. Rouster ‐ Stevens, T. Rubinstein, N. Ruth, S. Sabbagh, R. Sadun, L. Santiago, V. Saper, A. Sarkissian, L. Scalzi, J. Schahn, K. Schikler, A. Schlefman, B. Schlichting, H. Schmeling, E. Schmitt, G. Schulert, C. Schutt, C. Seper, B. Shaham, R. Sheets, A. Shehab, S. Shenoi, M. Sherman, J. Shirley, M. Shishov, N. Singer, V. Sivaraman, E. Sloan, C. Smith, J. Smith, E. Smitherman, J. Soep, M. B. Son, C. Spencer, L. Spiegel, J. Spitznagle, H. Srinivasalu, H. Stapp, A. Stephens, Y. Sterba Rakovchik, S. Stern, B. Stevens, R. Stevenson, C. Stingl, M. Stoll, E. Stringer, S. Sule, J. Sullivan, R. Sundel, M. Sutter, C. Swaffar, N. Swayne, T. Symington, G. Syverson, A.M. Szymanski, S. Taber, R. Tal, A. Tambralli, A. Taneja, T. Tanner, S. Tarvin, A. Taxter, M. Tesher, T. Thakurdeen, A. Theisen, G. Thieroff, B. Thomas, L. Thomas, N. Thomas, L. Timmerman, T. Ting, C. Todd, D. Toib, K. Torok, H. Tory, M. Toth, E. Treemarcki, S. Tse, T. Tse, C. Tsin, J. Twachtman‐Bassett, M. Twilt, T. Valcarcel, R. Valdovinos, A. Vallee, H. Van Mater, S. Vandenbergen, C. Varghese, N. Vasquez, P. Vega‐Fernandez, J. Verbsky, R. Verstegen, E. von Scheven, S. Vora, L. Wagner‐Weiner, D. Wahezi, S. Wakefield, B. Walker, S. Wallgren, H. Walters, M. Waterfield, J. Weiss, P. Weiss, E. Wershba, V. Westheuser, K. Widrick, C. Williams, S. Wong, S. Wooldridge, L. Woolnough, T. Wright, E. Wu, A. Yalcindag, R. Yeung, K. Yomogida, A. Zeft, Y. J. Zhang, Y. D. Zhao, Z. Zheng, A. Zhu, C. Zic.

This study used data collected in the CARRA Registry. The views expressed are the authors’ and do not necessarily represent the view of CARRA or the National Institutes of Health (NIH).

Supported by the National Center for Advancing Translational Sciences, NIH, through University of California, San Francisco (UCSF)‐Clinical and Translational Science Institute grant UL1‐TR‐001872. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. Dr Soulsby's work was supported by an Arthritis Foundation Diversity, Equity, and Inclusion Research in Health Outcomes award and is currently supported by a CARRA‐Arthritis Foundation Career Development Award and UCSF Society of Hellman Fellows award. Dr Knight's work was supported by a Canadian Research Chair in Mental Health and Chronic Disease of Childhood and a Lupus Research Alliance Career Development Award. Dr Horton's work was supported by funding from the National Institute of Arthritis and Musculoskeletal and Skin Diseases, NIH (R01‐AR‐074436), Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01‐HD‐109335), National Center for Advancing Translational Sciences (UL1‐TR‐003017, UM1‐TR‐004789), CARRA, and the Arthritis Foundation. Dr von Scheven's work receives support from CARRA.

¹University of California, San Francisco; ²The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada; ³Robert Wood Johnson Medical School at Rutgers University, New Brunswick, New Jersey; ⁴Institute for Health, Health Care Policy and Aging Research, New Brunswick, New Jersey; ⁵Rutgers School of Public Health, Piscataway, New Jersey; ⁶University of Ottawa, Ottawa, Ontario, Canada; ⁷Children's Hospital of Eastern Ontario Research Institute, Ottawa, Ontario, Canada; ⁸Institut du savoir Montfort, Ottawa, Ontario, Canada.

Additional supplementary information cited in this article can be found online in the Supporting Information section (https://acrjournals.onlinelibrary.wiley.com/doi/10.1002/acr.80051).

Author disclosures and graphical abstract are available at https://onlinelibrary.wiley.com/doi/10.1002/acr.80051.

Contributor Information

William Daniel Soulsby, Email: daniel.soulsby@ucsf.edu.

the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry Investigators and for the CARRA Health Equity Work Group:

R. Aamir, K. Abulaban, A. Adams, C. Aguiar Lapsia, H. Ahmed, S. Akoghlanian, A. AlBijadi, E. Allenspach, M. Alpizar, G. Amarilyo, M. Amoruso, S. Angeles‐Han, S. Ardoin, S. Armendariz, N. Aviran Dagan, I. Balboni, S. Balevic, S. Ballinger, S. Baluta, L. Barillas‐Arias, L. Barillas‐Arias, M. Basiaga, K. Baszis, M. Becker, A. Begezda, E. Beil, H. Bell‐Brunson, H. Benham, S. Benseler, L. Bermudez‐Santiago, W. Bernal, T. Bigley, C. Bingham, B. Binstadt, C. Black, B. Blackmon, M. Blakley, J. Bohnsack, A. Boneparth, H. Bradfield, J. Bridges, E. Brooks, M. Brothers, D. Brown, H. Brunner, L. Buckley, Mary Buckley, Meredith Buckley, H. Bukulmez, D. Bullock, A. Cancino, S. Canna, L. Cannon, S. Canny, V. Cartwright, E. Chalom, Johanna Chang, Joyce Chang, M. Chang, A. Chang‐Hoftman, A. Chen, P. Chiraseveenuprapund, K. Ciaglia, M. Cidon, D. Co, E. Cohen, R. Connor, K. Cook, A. Cooper, J. Cooper, K. Corbin, C. Correll, R. Cron, M. Curry, A. Dagci, A. Dalrymple, E. Datyner, T. Davis, D. De Ranieri, J. Dean, C. DeCoste, F. Dedeoglu, M. DeGuzman, N. Delnay, E. L. DeSantis, R. Devine, M. Dhalla, A. Dhanrajani, D. Dissanayake, B. Dizon, J. Drew, K. Driest, Q. Du, E. Duncan, K. Dunnock, D. Durkee, J. Dvergsten, A. Eberhard, K. Ede, B. Edelheit, C. Edens, M. Elder, Y. Elzaki, C. Failing, D. Fair, L. Favier, B. Feldman, J. Fennell, I. Ferguson, P. Ferguson, C. Figueroa, E. Flanagan, L. Fogel, E. Fox, M. Fox, L. Franklin, R. Fuhlbrigge, J. Fuller, T. Futch‐West, S. Gagne, M. Geiszler, D. Gerstbacher, M. Gilbert, A.C. Gironella, D. Glaser, I. Goh, S. Gorry, N. Goswami, B. Gottlieb, T. Graham, S. Grevich, T. Griffin, A. Grim, A. Grom, M. Guevara, L. Guzman, T. Hahn, O. Halyabar, E. Hammelev, T. Hammond, S. Haro, J. Harris, O. Harry, J. Hausmann, A. Hay, K. Hays, K. Hayward, L. Henderson, M. Henrickson, A. Hersh, L. Hiraki, M. Hiskey, P. Hobday, C. Hoffart, M. J. Holland, M. Hollander, S. Hong, D. Horton, J. Hsu, A. Huber, J. Huggins, J. Hui‐Yuen, M. Ibarra, A. Imlay, L. Imundo, C. Inman, A. Jackson, K. James, G. Janow, Y. Jiang, L. Johnson, N. Johnson, J. Jones, D. Kafisheh, K. Kaidar, S. Kasinathan, R. Kaur, E. Kessler, B. Kienzle, S. Kim, Y. Kimura, D. Kingsbury, M. Kitcharoensakkul, J. Klauss, K. Klein, M. Klein‐Gitelman, A. Knight, L. Kovalick, D. Krajewski, C. Kremer, T. LaFlam, B. Lang, S. Lapidus, B. Lapin, A. Lasky, E. Lawson, R. Laxer, A. Lee, Patricia Lee, Pui Lee, T. Lee, E. Leisinger, L. Lentini, M. Lerman, Y. Levinsky, D. Levy, S. Li, S. Lieberman, L. Lim, E. Limenis, C. Lin, N. Ling, G. Lionetti, R. Livny, M. Lo, A. Long, M. Lopez‐Peña, D. Lovell, S. Lvovich, A. Lytch, M. Ma, A. Machado, J. MacMahon, J. Madison, M. Mannion, C. Manos, L. Mansfield, B. Marston, K. Marzan, T. Mason, S. Matossian, L. McAllister, K. McBrearty, D. McCurdy, K. McDaniels, J. McDonald, L. McIntosh, E. Meidan, E. Mellins, Z. Mian, P. Miettunen, M. Miller, D. Milojevic, R. Mitacek, R. Modica, S. Mohan, K. Moore, T. Moore, L. Moorthy, J. Moreno, E. Morgan, A. Moyer, B. Murante, A. Murphy, E. Muscal, O. Mwizerwa, A. Najafi, K. Nanda, L. Nassi, S. Nativ, M. Natter, J. Neely, L. Newhall, A. Nuyen, P. Nigrovic, J. Nocton, B. Nolan, A. Nowakowski, K. Nowicki, R. Oakes, E. Oberle, S. Ogbonnaya‐Whittesley, E. Ogbu, M. Oliver, R. Olveda, K. Onel, A. Orandi, J. Padam, N. Pan, J. Pandya, S. Panupattanapong, A. Pappo Toledano, J. Patel, P. Patel, A. Patrick, S. Patrizi, S. Paul, J. Perfetto, M. Perron, M. Peskin, C. Pinotti, L. Ponder, R. Pooni, S. Prahalad, M. Quinlan‐Waters, J. Rafko, H. Rahimi, S. Ramsey, R. Randell, L. Ray, Ann Reed, Annelle Reed, H. Reid, D. Reiff, I. Reyhan, B. Richard, M. Riebschleger, E. Rife, M. Riskalla, A. Robinson, L. Robinson, L. Rodgers, M. Rodriquez, D. Rogers, T. Ronis, A. Rosado, M. Rosenkranz, N. Rosenwasser, H. Rothermel, D. Rothman, E. Rothschild, K. Rouster ‐ Stevens, T. Rubinstein, N. Ruth, S. Sabbagh, R. Sadun, L. Santiago, V. Saper, A. Sarkissian, L. Scalzi, J. Schahn, K. Schikler, A. Schlefman, B. Schlichting, H. Schmeling, E. Schmitt, G. Schulert, C. Schutt, C. Seper, B. Shaham, R. Sheets, A. Shehab, S. Shenoi, M. Sherman, J. Shirley, M. Shishov, N. Singer, V. Sivaraman, E. Sloan, C. Smith, J. Smith, E. Smitherman, J. Soep, M. B. Son, C. Spencer, L. Spiegel, J. Spitznagle, H. Srinivasalu, H. Stapp, A. Stephens, Y. Sterba Rakovchik, S. Stern, B. Stevens, R. Stevenson, C. Stingl, M. Stoll, E. Stringer, S. Sule, J. Sullivan, R. Sundel, M. Sutter, C. Swaffar, N. Swayne, T. Symington, G. Syverson, A.M. Szymanski, S. Taber, R. Tal, A. Tambralli, A. Taneja, T. Tanner, S. Tarvin, A. Taxter, M. Tesher, T. Thakurdeen, A. Theisen, G. Thieroff, B. Thomas, L. Thomas, N. Thomas, L. Timmerman, T. Ting, C. Todd, D. Toib, K. Torok, H. Tory, M. Toth, E. Treemarcki, S. Tse, T. Tse, C. Tsin, J. Twachtman‐Bassett, M. Twilt, T. Valcarcel, R. Valdovinos, A. Vallee, H. Van Mater, S. Vandenbergen, C. Varghese, N. Vasquez, P. Vega‐Fernandez, J. Verbsky, R. Verstegen, E. von Scheven, S. Vora, L. Wagner‐Weiner, D. Wahezi, S. Wakefield, B. Walker, S. Wallgren, H. Walters, M. Waterfield, J. Weiss, P. Weiss, E. Wershba, V. Westheuser, K. Widrick, C. Williams, S. Wong, S. Wooldridge, L. Woolnough, T. Wright, E. Wu, A. Yalcindag, R. Yeung, K. Yomogida, A. Zeft, Y. J. Zhang, Y. D. Zhao, Z. Zheng, A. Zhu, and C. Zic

References

1. Thierry S, Fautrel B, Lemelle I, et al. Prevalence and incidence of juvenile idiopathic arthritis: a systematic review. Joint Bone Spine 2014;81(2):112–117. doi: 10.1016/j.jbspin.2013.09.003 [DOI] [PubMed] [Google Scholar]
2. Glerup M, Rypdal V, Arnstad ED, et al; Nordic Study Group of Pediatric Rheumatology. Long‐term outcomes in juvenile idiopathic arthritis: eighteen years of follow‐up in the population‐based Nordic juvenile idiopathic arthritis cohort. Arthritis Care Res (Hoboken) 2020;72(4):507–516. doi: 10.1002/acr.23853 [DOI] [PubMed] [Google Scholar]
3. Oliveira‐Ramos F, Eusébio M, Martins M F, et al. Juvenile idiopathic arthritis in adulthood: fulfilment of classification criteria for adult rheumatic diseases, long‐term outcomes and predictors of inactive disease, functional status and damage. RMD Open 2016;2(2):e000304. doi: 10.1136/rmdopen-2016-000304 [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Stoll ML, Cron RQ. Treatment of juvenile idiopathic arthritis: a revolution in care. Pediatr Rheumatol Online J 2014;12(1):13. doi: 10.1186/1546-0096-12-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Brunner HI, Taylor J, Britto MT, et al. Differences in disease outcomes between Medicaid and privately insured children: possible health disparities in juvenile rheumatoid arthritis. Arthritis Rheum 2006;55(3):378–384. doi: 10.1002/art.21991 [DOI] [PubMed] [Google Scholar]
6. Verstappen SMM, Cobb J, Foster HE, et al. The association between low socioeconomic status with high physical limitations and low illness self‐perception in patients with juvenile idiopathic arthritis: results from the Childhood Arthritis Prospective Study. Arthritis Care Res (Hoboken) 2015;67(3):382–389. doi: 10.1002/acr.22466 [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Rocha FAC, Landim JIVD, Aguiar MG, et al. Evaluation of disease activity in a low‐income juvenile idiopathic arthritis cohort. Rheumatol Int 2019;39(1):67–71. doi: 10.1007/s00296-018-4128-8 [DOI] [PubMed] [Google Scholar]
8. Yang G, Bykerk VP, Boire G, et al; CATCH Investigators. Does socioeconomic status affect outcomes in early inflammatory arthritis? Data from a Canadian multisite suspected rheumatoid arthritis inception cohort. J Rheumatol 2015;42(1):46–54. doi: 10.3899/jrheum.131382 [DOI] [PubMed] [Google Scholar]
9. Akinsete AM, Woo JMP, Rubinstein TB. Disparities in pediatric rheumatic diseases. Rheum Dis Clin North Am 2022;48(1):183–198. doi: 10.1016/j.rdc.2021.09.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Roberts JE, Williams K, Dallas J, et al; CARRA Registry Investigators. Insurance status and tumor necrosis factor inhibitor initiation among children with juvenile idiopathic arthritis in the CARRA Registry. J Rheumatol 2023;50(8):1047–1057. doi: 10.3899/jrheum.220871 [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Bauman LJ, Silver EJ, Stein REK. Cumulative social disadvantage and child health. Pediatrics 2006;117(4):1321–1328. doi: 10.1542/peds.2005-1647 [DOI] [PubMed] [Google Scholar]
12. Seabrook JA, Avison WR. Socioeconomic status and cumulative disadvantage processes across the life course: implications for health outcomes. Can Rev Sociol 2012;49(1):50–68. doi: 10.1111/j.1755-618X.2011.01280.x [DOI] [PubMed] [Google Scholar]
13. Kilbourne AM, Switzer G, Hyman K, et al. Advancing health disparities research within the health care system: a conceptual framework. Am J Public Health 2006;96(12):2113–2121. doi: 10.2105/AJPH.2005.077628 [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Soulsby WD, Lawson E, Pantell MS. Cumulative social disadvantage associated with childhood arthritis: a cross‐sectional analysis of the National Survey of Children's Health. Arthritis Care Res (Hoboken) 2023;75(1):3–8. doi: 10.1002/acr.24991 [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Beukelman T, Kimura Y, Ilowite NT, et al; CARRA Registry Investigators. The new Childhood Arthritis and Rheumatology Research Alliance (CARRA) registry: design, rationale, and characteristics of patients enrolled in the first 12 months. Pediatr Rheumatol Online J 2017;15(1):30. doi: 10.1186/s12969-017-0160-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Petty RE, Southwood TR, Manners P, et al; International League of Associations for Rheumatology. International League of Associations for Rheumatology classification of juvenile idiopathic arthritis: second revision, Edmonton, 2001. J Rheumatol 2004;31(2):390–392. [PubMed] [Google Scholar]
17. Kind AJH, Buckingham WR. Making neighborhood‐disadvantage metrics accessible ‐ the neighborhood atlas. N Engl J Med 2018;378(26):2456–2458. doi: 10.1056/NEJMp1802313 [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Soulsby WD, Olveda R, He J, et al; CARRA Registry Investigators. Racial disparities and achievement of the low lupus disease activity state: a CARRA registry study. Arthritis Care Res (Hoboken) 2025;77(1):38–49. doi: 10.1002/acr.25443 [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Mbonu I, Tai S, Bartels CM, et al. Association between neighborhood deprivation and number of rheumatology providers. Arthritis Care Res (Hoboken) 2023;75(1):9–13. doi: 10.1002/acr.25036 [DOI] [PubMed] [Google Scholar]
20. Consolaro A, Negro G, Gallo MC, et al. A30: assessment of construct and discriminative validity of the 3‐variable JADAS in relation of parent‐reported outcomes. Arthritis Rheumatol 2014;66(S3):S46–S46. doi: 10.1002/art.38446 [DOI] [Google Scholar]
21. Trincianti C, Van Dijkhuizen EHP, Alongi A, et al; Paediatric Rheumatology International Trials Organisation. Definition and validation of the American College of Rheumatology 2021 juvenile arthritis disease activity score cutoffs for disease activity states in juvenile idiopathic arthritis. Arthritis Rheumatol 2021;73(11):1966–1975. doi: 10.1002/art.41879 [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Singh G, Athreya BH, Fries JF, et al. Measurement of health status in children with juvenile rheumatoid arthritis. Arthritis Rheum 1994;37(12):1761–1769. doi: 10.1002/art.1780371209 [DOI] [PubMed] [Google Scholar]
23. Neuhaus JM. Statistical methods for longitudinal and clustered designs with binary responses. Stat Methods Med Res 1992;1(3):249–273. doi: 10.1177/096228029200100303 [DOI] [PubMed] [Google Scholar]
24. White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med 2011;30(4):377–399. doi: 10.1002/sim.4067 [DOI] [PubMed] [Google Scholar]
25. Soulsby WD. Journal club review of “Avacopan for the treatment of ANCA‐associated vasculitis.” ACR Open Rheumatol 2022;4(7):558–561. doi: 10.1002/acr2.11412 [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Cheung YB. A modified least‐squares regression approach to the estimation of risk difference. Am J Epidemiol 2007;166(11):1337–1344. doi: 10.1093/aje/kwm223 [DOI] [PubMed] [Google Scholar]
27. Pantell MS, Prather AA, Downing JM, et al. Association of social and behavioral risk factors with earlier onset of adult hypertension and diabetes. JAMA Netw Open 2019;2(5):e193933. doi: 10.1001/jamanetworkopen.2019.3933 [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Balmuri N, Soulsby WD, Cooley V, et al; CARRA Registry Investigators. Community poverty level influences time to first pediatric rheumatology appointment in polyarticular juvenile idiopathic arthritis. Pediatr Rheumatol Online J 2021;19(1):122. doi: 10.1186/s12969-021-00610-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Soulsby WD, Balmuri N, Cooley V, et al; CARRA Registry Investigators. Social determinants of health influence disease activity and functional disability in polyarticular juvenile idiopathic arthritis. Pediatr Rheumatol Online J 2022;20(1):18. doi: 10.1186/s12969-022-00676-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Correll CK, Ditmyer MM, Mehta J, et al. 2015 American College of Rheumatology Workforce Study and Demand Projections of Pediatric Rheumatology Workforce, 2015‐2030. Arthritis Care Res (Hoboken) 2022;74(3):340–348. doi: 10.1002/acr.24497 [DOI] [PubMed] [Google Scholar]
31. Mayer ML. Are we there yet? Distance to care and relative supply among pediatric medical subspecialties. Pediatrics 2006;118(6):2313–2321. doi: 10.1542/peds.2006-1570 [DOI] [PubMed] [Google Scholar]
32. Neely J, Shalen J, Sturrock H, et al; Childhood Arthritis and Rheumatology Research Alliance Investigators. Access to care and diagnostic delays in juvenile dermatomyositis: results from the Childhood Arthritis and Rheumatology Research Alliance Legacy Registry. ACR Open Rheumatol 2021;3(5):349–354. doi: 10.1002/acr2.11246 [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Kranjac AW, Denney JT, Kimbro RT, et al. Neighborhood and social environmental influences on child chronic disease prevalence. Popul Environ 2018;40(2):93–114. doi: 10.1007/s11111-018-0303-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Pooni R, Lee T. The emerging telehealth landscape in pediatric rheumatology. Rheum Dis Clin North Am 2022;48(1):259–270. doi: 10.1016/j.rdc.2021.08.005 [DOI] [PubMed] [Google Scholar]
35. Correll CK, Klein‐Gitelman MS, Henrickson M, et al. Child health needs and the pediatric rheumatology workforce: 2020‐2040. Pediatrics 2024;153(suppl 2):e2023063678R. doi: 10.1542/peds.2023-063678R [DOI] [PubMed] [Google Scholar]
36. Vestal E, Newman S, Phillips S. Barriers and facilitators to accessing pediatric specialty care for rural‐dwelling children with complex chronic conditions: an integrative review. J Pediatr Nurs 2024;77:e385–e393. doi: 10.1016/j.pedn.2024.05.001 [DOI] [PubMed] [Google Scholar]
37. Treemarcki EB, Danguecan AN, Cunningham NR, et al. Mental health in pediatric rheumatology: an opportunity to improve outcomes. Rheum Dis Clin North Am 2022;48(1):67–90. doi: 10.1016/j.rdc.2021.09.012 [DOI] [PubMed] [Google Scholar]
38. Cunningham NR, Danguecan AN, Ely SL, et al; CARRA Mental Health Workgroup. American College of Rheumatology guidance statements for addressing mental health concerns in youth with pediatric rheumatologic diseases. Arthritis Care Res (Hoboken) 2025;77(8):953–964. doi: 10.1002/acr.25519 [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Davis AM, Rubinstein TB, Rodriguez M, et al. Mental health care for youth with rheumatologic diseases ‐ bridging the gap. Pediatr Rheumatol Online J 2017;15(1):85. doi: 10.1186/s12969-017-0214-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Horný M, Glover W, Gupte G, et al. Patient navigation to improve diabetes outpatient care at a safety‐net hospital: a retrospective cohort study. BMC Health Serv Res 2017;17(1):759. doi: 10.1186/s12913-017-2700-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Feldman CH, Wohlfahrt A, Campos A, et al. Can patient navigators improve adherence to disease‐modifying antirheumatic drugs? Quantitative findings from a six‐month single‐arm pilot intervention. Arthritis Care Res (Hoboken) 2018;70(9):1400–1405. doi: 10.1002/acr.23302 [DOI] [PubMed] [Google Scholar]
42. Feldman CH, Hicks LS, Norton TL, et al. Assessing the need for improved access to rheumatology care: a survey of Massachusetts community health center medical directors. J Clin Rheumatol 2013;19(7):361–366. doi: 10.1097/RHU.0b013e3182a6a490 [DOI] [PMC free article] [PubMed] [Google Scholar]
43. Kusma JD, Raphael JL, Perrin JM, Hudak ML; Committee on Child Health Financing. Medicaid and the Children's Health Insurance Program: optimization to promote equity in child and young adult health. Pediatrics 2023;152(5):e2023064088. doi: 10.1542/peds.2023-064088 [DOI] [PubMed] [Google Scholar]
44. Grant R, Goldsmith G, Gracy D, et al. Better transportation to health care will improve child health and lower costs. Adv Pediatr 2016;63(1):389–401. doi: 10.1016/j.yapd.2016.04.003 [DOI] [PubMed] [Google Scholar]
45. Drake C, Zhang Y, Chaiyachati KH, et al. The limitations of poor broadband internet access for telemedicine use in rural America: an observational study. Ann Intern Med 2019;171(5):382–384. doi: 10.7326/M19-0283 [DOI] [PubMed] [Google Scholar]
46. Bauerly BC, McCord RF, Hulkower R, et al. Broadband access as a public health issue: the role of law in expanding broadband access and connecting underserved communities for better health outcomes. J Law Med Ethics 2019;47(suppl 2):39–42. doi: 10.1177/1073110519857314 [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Henize AW, Beck AF, Klein MD, et al. A road map to address the social determinants of health through community collaboration. Pediatrics 2015;136(4):e993–e1001. doi: 10.1542/peds.2015-0549 [DOI] [PubMed] [Google Scholar]
48. Feldman CH, Summit R, Retzel K, et al. Rheumatology‐based social determinants of health screening to uncover and address needs [abstract]. In: Achieving Equity in Lupus Care & Outcomes. Lupus Sci Med. 2024;11(suppl 2):A1.2–A4. doi: 10.1136/lupus-2023-lupus21century.2 [DOI] [Google Scholar]
49. Harris MA, Spiro K, Heywood M, et al. Novel Interventions in Children's Health Care (NICH): innovative treatment for youth with complex medical conditions. Clin Pract Pediatr Psychol 2013;1(2):137–145. doi: 10.1037/cpp0000016 [DOI] [Google Scholar]
50. Harris MA, Wagner DV, Wilson AC, et al. Novel Interventions in Children's Healthcare for youth hospitalized for chronic pain. Clin Pract Pediatr Psychol 2015;3(1):48–58. doi: 10.1037/cpp0000088 [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

Disclosure form.

ACR-78-1552-s003.pdf^{(325.7KB, pdf)}

ACR-78-1552-s001.docx^{(15.3KB, docx)}

ACR-78-1552-s002.docx^{(15.4KB, docx)}

[acr80051-bib-0001] 1. Thierry S, Fautrel B, Lemelle I, et al. Prevalence and incidence of juvenile idiopathic arthritis: a systematic review. Joint Bone Spine 2014;81(2):112–117. doi: 10.1016/j.jbspin.2013.09.003 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0002] 2. Glerup M, Rypdal V, Arnstad ED, et al; Nordic Study Group of Pediatric Rheumatology. Long‐term outcomes in juvenile idiopathic arthritis: eighteen years of follow‐up in the population‐based Nordic juvenile idiopathic arthritis cohort. Arthritis Care Res (Hoboken) 2020;72(4):507–516. doi: 10.1002/acr.23853 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0003] 3. Oliveira‐Ramos F, Eusébio M, Martins M F, et al. Juvenile idiopathic arthritis in adulthood: fulfilment of classification criteria for adult rheumatic diseases, long‐term outcomes and predictors of inactive disease, functional status and damage. RMD Open 2016;2(2):e000304. doi: 10.1136/rmdopen-2016-000304 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0004] 4. Stoll ML, Cron RQ. Treatment of juvenile idiopathic arthritis: a revolution in care. Pediatr Rheumatol Online J 2014;12(1):13. doi: 10.1186/1546-0096-12-13 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0005] 5. Brunner HI, Taylor J, Britto MT, et al. Differences in disease outcomes between Medicaid and privately insured children: possible health disparities in juvenile rheumatoid arthritis. Arthritis Rheum 2006;55(3):378–384. doi: 10.1002/art.21991 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0006] 6. Verstappen SMM, Cobb J, Foster HE, et al. The association between low socioeconomic status with high physical limitations and low illness self‐perception in patients with juvenile idiopathic arthritis: results from the Childhood Arthritis Prospective Study. Arthritis Care Res (Hoboken) 2015;67(3):382–389. doi: 10.1002/acr.22466 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0007] 7. Rocha FAC, Landim JIVD, Aguiar MG, et al. Evaluation of disease activity in a low‐income juvenile idiopathic arthritis cohort. Rheumatol Int 2019;39(1):67–71. doi: 10.1007/s00296-018-4128-8 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0008] 8. Yang G, Bykerk VP, Boire G, et al; CATCH Investigators. Does socioeconomic status affect outcomes in early inflammatory arthritis? Data from a Canadian multisite suspected rheumatoid arthritis inception cohort. J Rheumatol 2015;42(1):46–54. doi: 10.3899/jrheum.131382 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0009] 9. Akinsete AM, Woo JMP, Rubinstein TB. Disparities in pediatric rheumatic diseases. Rheum Dis Clin North Am 2022;48(1):183–198. doi: 10.1016/j.rdc.2021.09.014 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0010] 10. Roberts JE, Williams K, Dallas J, et al; CARRA Registry Investigators. Insurance status and tumor necrosis factor inhibitor initiation among children with juvenile idiopathic arthritis in the CARRA Registry. J Rheumatol 2023;50(8):1047–1057. doi: 10.3899/jrheum.220871 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0011] 11. Bauman LJ, Silver EJ, Stein REK. Cumulative social disadvantage and child health. Pediatrics 2006;117(4):1321–1328. doi: 10.1542/peds.2005-1647 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0012] 12. Seabrook JA, Avison WR. Socioeconomic status and cumulative disadvantage processes across the life course: implications for health outcomes. Can Rev Sociol 2012;49(1):50–68. doi: 10.1111/j.1755-618X.2011.01280.x [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0013] 13. Kilbourne AM, Switzer G, Hyman K, et al. Advancing health disparities research within the health care system: a conceptual framework. Am J Public Health 2006;96(12):2113–2121. doi: 10.2105/AJPH.2005.077628 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0014] 14. Soulsby WD, Lawson E, Pantell MS. Cumulative social disadvantage associated with childhood arthritis: a cross‐sectional analysis of the National Survey of Children's Health. Arthritis Care Res (Hoboken) 2023;75(1):3–8. doi: 10.1002/acr.24991 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0015] 15. Beukelman T, Kimura Y, Ilowite NT, et al; CARRA Registry Investigators. The new Childhood Arthritis and Rheumatology Research Alliance (CARRA) registry: design, rationale, and characteristics of patients enrolled in the first 12 months. Pediatr Rheumatol Online J 2017;15(1):30. doi: 10.1186/s12969-017-0160-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0016] 16. Petty RE, Southwood TR, Manners P, et al; International League of Associations for Rheumatology. International League of Associations for Rheumatology classification of juvenile idiopathic arthritis: second revision, Edmonton, 2001. J Rheumatol 2004;31(2):390–392. [PubMed] [Google Scholar]

[acr80051-bib-0017] 17. Kind AJH, Buckingham WR. Making neighborhood‐disadvantage metrics accessible ‐ the neighborhood atlas. N Engl J Med 2018;378(26):2456–2458. doi: 10.1056/NEJMp1802313 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0018] 18. Soulsby WD, Olveda R, He J, et al; CARRA Registry Investigators. Racial disparities and achievement of the low lupus disease activity state: a CARRA registry study. Arthritis Care Res (Hoboken) 2025;77(1):38–49. doi: 10.1002/acr.25443 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0019] 19. Mbonu I, Tai S, Bartels CM, et al. Association between neighborhood deprivation and number of rheumatology providers. Arthritis Care Res (Hoboken) 2023;75(1):9–13. doi: 10.1002/acr.25036 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0020] 20. Consolaro A, Negro G, Gallo MC, et al. A30: assessment of construct and discriminative validity of the 3‐variable JADAS in relation of parent‐reported outcomes. Arthritis Rheumatol 2014;66(S3):S46–S46. doi: 10.1002/art.38446 [DOI] [Google Scholar]

[acr80051-bib-0021] 21. Trincianti C, Van Dijkhuizen EHP, Alongi A, et al; Paediatric Rheumatology International Trials Organisation. Definition and validation of the American College of Rheumatology 2021 juvenile arthritis disease activity score cutoffs for disease activity states in juvenile idiopathic arthritis. Arthritis Rheumatol 2021;73(11):1966–1975. doi: 10.1002/art.41879 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0022] 22. Singh G, Athreya BH, Fries JF, et al. Measurement of health status in children with juvenile rheumatoid arthritis. Arthritis Rheum 1994;37(12):1761–1769. doi: 10.1002/art.1780371209 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0023] 23. Neuhaus JM. Statistical methods for longitudinal and clustered designs with binary responses. Stat Methods Med Res 1992;1(3):249–273. doi: 10.1177/096228029200100303 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0024] 24. White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med 2011;30(4):377–399. doi: 10.1002/sim.4067 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0025] 25. Soulsby WD. Journal club review of “Avacopan for the treatment of ANCA‐associated vasculitis.” ACR Open Rheumatol 2022;4(7):558–561. doi: 10.1002/acr2.11412 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0026] 26. Cheung YB. A modified least‐squares regression approach to the estimation of risk difference. Am J Epidemiol 2007;166(11):1337–1344. doi: 10.1093/aje/kwm223 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0027] 27. Pantell MS, Prather AA, Downing JM, et al. Association of social and behavioral risk factors with earlier onset of adult hypertension and diabetes. JAMA Netw Open 2019;2(5):e193933. doi: 10.1001/jamanetworkopen.2019.3933 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0028] 28. Balmuri N, Soulsby WD, Cooley V, et al; CARRA Registry Investigators. Community poverty level influences time to first pediatric rheumatology appointment in polyarticular juvenile idiopathic arthritis. Pediatr Rheumatol Online J 2021;19(1):122. doi: 10.1186/s12969-021-00610-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0029] 29. Soulsby WD, Balmuri N, Cooley V, et al; CARRA Registry Investigators. Social determinants of health influence disease activity and functional disability in polyarticular juvenile idiopathic arthritis. Pediatr Rheumatol Online J 2022;20(1):18. doi: 10.1186/s12969-022-00676-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0030] 30. Correll CK, Ditmyer MM, Mehta J, et al. 2015 American College of Rheumatology Workforce Study and Demand Projections of Pediatric Rheumatology Workforce, 2015‐2030. Arthritis Care Res (Hoboken) 2022;74(3):340–348. doi: 10.1002/acr.24497 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0031] 31. Mayer ML. Are we there yet? Distance to care and relative supply among pediatric medical subspecialties. Pediatrics 2006;118(6):2313–2321. doi: 10.1542/peds.2006-1570 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0032] 32. Neely J, Shalen J, Sturrock H, et al; Childhood Arthritis and Rheumatology Research Alliance Investigators. Access to care and diagnostic delays in juvenile dermatomyositis: results from the Childhood Arthritis and Rheumatology Research Alliance Legacy Registry. ACR Open Rheumatol 2021;3(5):349–354. doi: 10.1002/acr2.11246 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0033] 33. Kranjac AW, Denney JT, Kimbro RT, et al. Neighborhood and social environmental influences on child chronic disease prevalence. Popul Environ 2018;40(2):93–114. doi: 10.1007/s11111-018-0303-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0034] 34. Pooni R, Lee T. The emerging telehealth landscape in pediatric rheumatology. Rheum Dis Clin North Am 2022;48(1):259–270. doi: 10.1016/j.rdc.2021.08.005 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0035] 35. Correll CK, Klein‐Gitelman MS, Henrickson M, et al. Child health needs and the pediatric rheumatology workforce: 2020‐2040. Pediatrics 2024;153(suppl 2):e2023063678R. doi: 10.1542/peds.2023-063678R [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0036] 36. Vestal E, Newman S, Phillips S. Barriers and facilitators to accessing pediatric specialty care for rural‐dwelling children with complex chronic conditions: an integrative review. J Pediatr Nurs 2024;77:e385–e393. doi: 10.1016/j.pedn.2024.05.001 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0037] 37. Treemarcki EB, Danguecan AN, Cunningham NR, et al. Mental health in pediatric rheumatology: an opportunity to improve outcomes. Rheum Dis Clin North Am 2022;48(1):67–90. doi: 10.1016/j.rdc.2021.09.012 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0038] 38. Cunningham NR, Danguecan AN, Ely SL, et al; CARRA Mental Health Workgroup. American College of Rheumatology guidance statements for addressing mental health concerns in youth with pediatric rheumatologic diseases. Arthritis Care Res (Hoboken) 2025;77(8):953–964. doi: 10.1002/acr.25519 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0039] 39. Davis AM, Rubinstein TB, Rodriguez M, et al. Mental health care for youth with rheumatologic diseases ‐ bridging the gap. Pediatr Rheumatol Online J 2017;15(1):85. doi: 10.1186/s12969-017-0214-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0040] 40. Horný M, Glover W, Gupte G, et al. Patient navigation to improve diabetes outpatient care at a safety‐net hospital: a retrospective cohort study. BMC Health Serv Res 2017;17(1):759. doi: 10.1186/s12913-017-2700-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0041] 41. Feldman CH, Wohlfahrt A, Campos A, et al. Can patient navigators improve adherence to disease‐modifying antirheumatic drugs? Quantitative findings from a six‐month single‐arm pilot intervention. Arthritis Care Res (Hoboken) 2018;70(9):1400–1405. doi: 10.1002/acr.23302 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0042] 42. Feldman CH, Hicks LS, Norton TL, et al. Assessing the need for improved access to rheumatology care: a survey of Massachusetts community health center medical directors. J Clin Rheumatol 2013;19(7):361–366. doi: 10.1097/RHU.0b013e3182a6a490 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0043] 43. Kusma JD, Raphael JL, Perrin JM, Hudak ML; Committee on Child Health Financing. Medicaid and the Children's Health Insurance Program: optimization to promote equity in child and young adult health. Pediatrics 2023;152(5):e2023064088. doi: 10.1542/peds.2023-064088 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0044] 44. Grant R, Goldsmith G, Gracy D, et al. Better transportation to health care will improve child health and lower costs. Adv Pediatr 2016;63(1):389–401. doi: 10.1016/j.yapd.2016.04.003 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0045] 45. Drake C, Zhang Y, Chaiyachati KH, et al. The limitations of poor broadband internet access for telemedicine use in rural America: an observational study. Ann Intern Med 2019;171(5):382–384. doi: 10.7326/M19-0283 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0046] 46. Bauerly BC, McCord RF, Hulkower R, et al. Broadband access as a public health issue: the role of law in expanding broadband access and connecting underserved communities for better health outcomes. J Law Med Ethics 2019;47(suppl 2):39–42. doi: 10.1177/1073110519857314 [DOI] [PMC free article] [PubMed] [Google Scholar]

[acr80051-bib-0047] 47. Henize AW, Beck AF, Klein MD, et al. A road map to address the social determinants of health through community collaboration. Pediatrics 2015;136(4):e993–e1001. doi: 10.1542/peds.2015-0549 [DOI] [PubMed] [Google Scholar]

[acr80051-bib-0048] 48. Feldman CH, Summit R, Retzel K, et al. Rheumatology‐based social determinants of health screening to uncover and address needs [abstract]. In: Achieving Equity in Lupus Care & Outcomes. Lupus Sci Med. 2024;11(suppl 2):A1.2–A4. doi: 10.1136/lupus-2023-lupus21century.2 [DOI] [Google Scholar]

[acr80051-bib-0049] 49. Harris MA, Spiro K, Heywood M, et al. Novel Interventions in Children's Health Care (NICH): innovative treatment for youth with complex medical conditions. Clin Pract Pediatr Psychol 2013;1(2):137–145. doi: 10.1037/cpp0000016 [DOI] [Google Scholar]

[acr80051-bib-0050] 50. Harris MA, Wagner DV, Wilson AC, et al. Novel Interventions in Children's Healthcare for youth hospitalized for chronic pain. Clin Pract Pediatr Psychol 2015;3(1):48–58. doi: 10.1037/cpp0000088 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Cumulative Social Disadvantage and Disease Activity in Juvenile Idiopathic Arthritis: A Childhood Arthritis and Rheumatology Research Alliance Registry Study

William Daniel Soulsby, MD

John Boscardin, PhD

Andrea Knight, MD, MSCE

Daniel B Horton, MD, MSCE

Karine Toupin‐April, PhD

Emily von Scheven, MD, MAS

Abstract

Objective

Methods

Results

Conclusion

INTRODUCTION

SIGNIFICANCE & INNOVATIONS.

METHODS

Study design and setting

Analytic sample

Primary exposures

Figure 1.

Primary outcome

Secondary outcome

Study covariates

Statistical analysis

Sensitivity analysis

Causal mediation analysis

RESULTS

Demographic and clinical characteristics of the cohort

Table 1.

Disease activity by cJADAS‐10

Table 2.

Table 3.

Table 4.

Causal mediation analysis

Figure 2.

Functional disability by CHAQ

DISCUSSION

AUTHOR CONTRIBUTIONS

Supporting information

ACKNOWLEDGMENTS

APPENDIX A. CARRA Registry Site Principal Investigators, Sub‐Investigators, and Research Coordinators

Contributor Information

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases