Diagnostic Patterns and Subtype Overlap in Idiopathic Inflammatory Myopathies: A Retrospective Claims-Based Cohort Study
Author(s)
Parisa Fatemeh Asad, Sr., PhD1, Andre Gladiator, PhD2, Navya Koneripalli, MPH3, Sol Han, MPH3, Charlotte E. Ward, PhD4, Shreyas Jarmale, BS5, Clémence Arvin-Berod, PharmD2.
1argenx BV, Basel (BS), Switzerland, 2argenx BV, Zwijnaarde, Belgium, 3ZS Associates, Boston, MA, USA, 4Consultant, ZS Associates, Concord, NH, USA, 5ZS Associates, Bangalore, India.
1argenx BV, Basel (BS), Switzerland, 2argenx BV, Zwijnaarde, Belgium, 3ZS Associates, Boston, MA, USA, 4Consultant, ZS Associates, Concord, NH, USA, 5ZS Associates, Bangalore, India.
OBJECTIVES: Idiopathic inflammatory myopathies (IIMs) are rare, heterogeneous autoimmune diseases that are challenging to distinguish in real-world data due to overlapping features and non-specific coding. This study explored the feasibility of identifying IIM subtypes, particularly dermatomyositis (DM) and polymyositis (PM), using claims data, aiming to highlight diagnostic ambiguity and inform future identification strategies.
METHODS: Using US-based Komodo claims data (2015-2024), we identified patients with ≥2 primary-position ICD-10 codes for DM or PM, occurring 30-365 days apart. Patients were assigned to four cohorts based on initial subtype coding: PM, DM, PM/DM with unspecified myositis, and those with all three codes. Diagnostic coding patterns, comorbidities, overlap syndromes (≥2 claims for other autoimmune conditions), and specialty involvement were assessed over a 2-year follow-up period.
RESULTS: Among incident patients, 1,258 had ≥2 PM-coded visits and 1,892 had ≥2 DM-coded visits. During follow-up, half of PM patients also received codes for other IIM subtypes, suggesting ongoing diagnostic uncertainty. Among those with both PM and DM codes, most had more DM-coded visits (56% in PM group; 49% in DM group). Diagnostic ambiguity was greatest among patients with all three codes (n = 78; 3%), nearly half of whom had no dominant subtype. Twenty percent of PM and DM patients had overlap syndromes, most commonly rheumatoid arthritis, lupus, or systemic sclerosis. Rheumatologists were most involved overall, while dermatologists were more active in DM and neurologists in unspecified IIM. Common comorbidities included skin disorders, hypertension, osteoarthritis, and diabetes. Patients with multiple subtypes had consistently higher comorbidity burden.
CONCLUSIONS: Findings reveal meaningful diagnostic variation and coding patterns that underscore the complexity of identifying IIM subtypes in claims data. While ambiguity exists, the observed trends in coding, comorbidities, and specialty involvement highlight valuable signals that can inform the development of more sophisticated algorithms for improved subtype identification.
METHODS: Using US-based Komodo claims data (2015-2024), we identified patients with ≥2 primary-position ICD-10 codes for DM or PM, occurring 30-365 days apart. Patients were assigned to four cohorts based on initial subtype coding: PM, DM, PM/DM with unspecified myositis, and those with all three codes. Diagnostic coding patterns, comorbidities, overlap syndromes (≥2 claims for other autoimmune conditions), and specialty involvement were assessed over a 2-year follow-up period.
RESULTS: Among incident patients, 1,258 had ≥2 PM-coded visits and 1,892 had ≥2 DM-coded visits. During follow-up, half of PM patients also received codes for other IIM subtypes, suggesting ongoing diagnostic uncertainty. Among those with both PM and DM codes, most had more DM-coded visits (56% in PM group; 49% in DM group). Diagnostic ambiguity was greatest among patients with all three codes (n = 78; 3%), nearly half of whom had no dominant subtype. Twenty percent of PM and DM patients had overlap syndromes, most commonly rheumatoid arthritis, lupus, or systemic sclerosis. Rheumatologists were most involved overall, while dermatologists were more active in DM and neurologists in unspecified IIM. Common comorbidities included skin disorders, hypertension, osteoarthritis, and diabetes. Patients with multiple subtypes had consistently higher comorbidity burden.
CONCLUSIONS: Findings reveal meaningful diagnostic variation and coding patterns that underscore the complexity of identifying IIM subtypes in claims data. While ambiguity exists, the observed trends in coding, comorbidities, and specialty involvement highlight valuable signals that can inform the development of more sophisticated algorithms for improved subtype identification.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
MSR76
Topic
Clinical Outcomes, Methodological & Statistical Research, Real World Data & Information Systems
Disease
Rare & Orphan Diseases, Sensory System Disorders (Ear, Eye, Dental, Skin), Systemic Disorders/Conditions (Anesthesia, Auto-Immune Disorders (n.e.c.), Hematological Disorders (non-oncologic), Pain)