POS1197 VALIDATION OF PATIENT-REPORTED OUTCOMES MEASUREMENT INFORMATION SYSTEM (PROMIS) QUESTIONNAIRES FOR CHILDREN WITH CHRONIC NONBACTERIAL OSTEOMYELITIS USING THE CHOIR DATA (2024)

Abstract

Background: Chronic nonbacterial osteomyelitis (CNO) is an autoinflammatory bone disease that mainly affects children and adolescents. Disease monitoring is challenging as reported pain is not reliable and imaging is not always obtained at all clinic visits. To date, patient reported outcomes used in CNO research have not yet been validated in this population. The Patient-reported Outcomes Measurement Information System (PROMIS) questionnaires, validated in other pediatric rheumatic diseases, are administered to patients enrolled in the prospective multisite CHronic nonbacterial Osteomyelitis International Registry (CHOIR)¹ since 2018.

Objectives: To assess the convergent and responsive validity of the PROMIS instruments in patients with CNO.

Methods: Children or young adults with CNO were consented and enrolled into CHOIR. Self-reported PROMIS questionnaires of fatigue, pain interference (PI), pain behavior (PB), mobility, upper extremity (UE), physical activity (PA) and strength impact (SI) were administered to patients 8 years and older in English or Spanish at each clinical visit. Demographic, clinical, and imaging data were prospectively collected. The T score was calculated. External validation surveys were administered to assess patients’ perception of difficulty of use of limb/back/jaw, fatigue, sadness and worry on a 0-10 scale, disease status (inactive, mild, moderate, severe), and status change (unchanged, worsened, improved). Improvement of clinical disease activity score (CDAS) of 2.5 was defined as meaningful change. Descriptive statistics were used for demographic and clinical characteristics. Log-transformed linear mixed effect models with random participant intercepts were performed to assess PROMIS score changes after treatment. Wilcoxon signed-rank test with continuity correction was performed to determine the change of the PROMIS scores among the improved, worsened, and unchanged groups. Spearman rank correlation test was performed to determine the relationship between the PROMIS scores and reported disease status by patient/families.

Results: More than 1,000 clinical visits from 184 patients were associated with self-reported PROMIS questionnaire entries in English. The median age at disease onset, diagnosis, and enrollment were 9.4 (IQR 7.5 – 11.3), 10.4 (IQR 8.4 -12.3), and 11.4 (IQR 9.5 – 13.8) years respectively. Seventy (38%) were males and 153 (83%) were White. All PROMIS scores correlated significantly (p<0.01) with patient reported variables and physician global assessment (PHGA). The correlation with function and PHGA was good (0.4-0.6) for Mobility, PB, and PI. All PROMIS scores, except physical activity, correlated significantly (p<0.05) with patient reported disease status. The correlation between patient-reported disease status and PHGA (0.75) was strong, moderate with Mobility (-0.53), PB (0.57), PI (0.5), and Fatigue (0.36), and weak with, SI (-0.23), UE (-0.23), and PA (-0.03). The changes of PROMIS scores over time for Mobility (p=0.015), PB (p<0.001), PA (p=0.019), and PI (p=0.011) compared to the self-reported status change (unchanged, improved, worsened) was significant. However, PROMIS score changes for Fatigue (p=0.055), SI (p=0.878), and UE (p=0.086) did not differ across various self-reported status change groups. After effective treatment when clinical disease activity score improved by at least 2.5 points (n=18), the change of PROMIS score from Mobility, PB, PI, UE was significant (p<0.05), whereas the change of PROMIS score from Fatigue, SI, and PA were not.

Conclusion: PROMIS Questionnaires provide valuable information about disease status of children with CNO and correlate well with self-reported functional and other psychosocial domains. Mobility, PI, and PB show sensitivity to change after effective treatment or with disease status change. These instruments are useful for CNO clinical disease monitoring and research.

REFERENCES: [1] Wu EY, Oliver M, Scheck J, et al. J Rheumatol. 2023 PMID: 37399459; PMCID: PMC10543471.

Acknowledgements: The authors thank the CHOIR participants, research assistants and volunteers at all sites including Teresa Dickson, Corinne Lawler, Sumaya Aden, Thuan Bui, Kyra Shelton, Esha Mahal, Annie Xu, Kellen James, Shayla Nguyen, Zheng Xu, Ava Klein, Chessie Snider, Mabel Ho, Trang Pham, Anna Saack, Paige Trunnell, Emily Deng, Ana Park, Cailey Karshmer, Emma Leisinger, Mary Ellen Riordan, and Justine Griswold. We appreciate the help from Ingrid Goh, Mariana Correia Marques, and Min-Lee Chang for the testing of the registry database and the training material. In addition to the authors, the following CARRA CNO workgroup members participated in the February 2021 meeting: Ingrid Goh, Brian Nolan, Tzielan Lee, Annette Jansson, Aleksander Lenert, Lina Jaberi, David Cabral, Lauren Potts, Arielle Hay, Karine Toupin-April, Akaluck Thatayatikom, Ingram Chang, Piya Lahiry, Anja Schnabel, Mikhail Kostik, Nathan Rogers, Achille Marino, Dita Cebecauerova, Phillip Mease, Lindsey Bergstrom, Suzanne Li, Deborah McCurdy, Alex Theos, Matthew Hollander, Samira Nazzar, Farzana Nuruzzaman, Beverley Shea, and Chris Obrien. Statistical analysis was supported by CRMO Warriors Guild and Kaila’s Komfort generous donations.