Background/Purpose: Youths with childhood-onset chronic rheumatic diseases (ChildCRD), including juvenile arthritis (JA) and systemic autoimmune rheumatic diseases (SARD), have worse grade 12 standards tests results compared to their peers. SARD includes systemic lupus erythematous, dermatomyositis, Sjogren’s syndrome and systemic sclerosis. We aimed to test if ChildCRD disease severity predicted the performance of ChildCRD patients on grade 12 standards tests. 

Methods: Population-based longitudinal cohort study from one Canadian province. All ChildCRD patients had been captured in a registry since 1984. Data from ChildCRD patients born 1979-1998 (grade 12, 1996-2015) were linked to the administrative health (hospitalizations, physician billings, medications), education (grade 12 standards tests and enrollment) and social data (income assistance, child welfare involvement) housed in the provincial population data repository. Outcomes: Validated language and arts achievement index (LAI) and Maths achievement index (MAI) derived from grade 12 standards tests results and enrollment data. Key predictor and covariates : We constructed a universal disease severity indicator using latent class trajectory analysis (including disease groups: SARD, oligo- and non-oligoarticular JA, as membership covariate) of all physician visit within the first 3.5 years after diagnosis (early disease). Model covariates included an area-based socioeconomic factor index (SEFI2), maternal age at first childbirth, family ever on income assistance or involved with child welfare services, psychiatric morbidities pre-diagnosis and in the 12-months preceding tests. Model:  Effects of the severity indicator and covariates were tested in linear regressions for LAI and MAI; standardized coefficients (b) with standard errors (SE) and scaled deviance (DEV) were reported.

Results: 541 participants (474 JA, 44 SARD), of whom 70% were females were studied. The best fitting latent class model of disease severity, showed 3 latent trajectories: high (9%), moderate (54%) and low numbers of visit (37%) corresponding to severe, moderate and mild disease severities (Figure 1). The membership probabilities for the groups were 0.98 (severe), 0.97 (moderate) and 0.96 (mild). The severe group predicted LAI (b=  –0.44, SE=0.17, p< 0.01) and MAI (b= –0.54, SE= 0.17, p< 0.01) in univariable models. After covariate adjustments, the severe disease group did not predict LAI, (b= –0.29, SE= 0.16, p=0.06, DEV=1.03) but predicted MAI (b= –0.41, SE=0.15, p< 0.01, DEV=1.03).

Conclusion: Longitudinal patterns of disease severity in early disease predicted grade 12 Maths standards tests results of ChildCRD patients, after covariate adjustment. Recognizing severe disease in early disease, in association with sociodemographic and psychiatric histories can help identify ChildCRD patients who are at risk of reduced academic achievement and who may benefit from increased education support.

FIg1_Universal severity indicator

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/are-patterns-of-early-disease-severity-predictive-of-grade-12-academic-achievement-in-patients-with-childhood-onset-chronic-rheumatic-diseases/