Pos0453 baseline prevalence and factors associated with interstitial lung disease in two rheumatoid arthritis inception cohorts in the uk

Background Rheumatoid arthritis (RA) is a chronic, inflammatory disease with interstitial lung disease (ILD) being an important extra-articular manifestation, associated with significant morbidity and mortality. Objectives To determine the prevalence of ILD at time of RA diagnosis and identify baseline independent factors associated with ILD. Methods Two early inception cohorts (1986-2001) were used: the early RA network (ERAN) and the early RA study (ERAS). Socio-demographic, clinical and laboratory measures were recorded at baseline, 6 months and yearly thereafter. Baseline variables collected and included in analyses were: seropositivity, body mass index (BMI), ethnicity, measures of deprivation, smoking status. Comorbidity burden was evaluated using the rheumatic disease comorbidity index (RDCI). Baseline prevalence (proportion of individuals with ILD at baseline) and 95% confidence interval was estimated. Multivariable regression analyses were performed to identify risk factors at baseline associated with RA-ILD development. Multiple imputation was used for missing data. Results Data from a total of 2701 patients were included in the study, from whom 101 patients were diagnosed with ILD. Follow up was up to 25 years (mean 6.8 years); 75% of patients followed for ≥10 years. Of these, 12% had ILD at baseline; 46% were diagnosed with ILD during follow up and 43% were diagnosed with ILD post-mortem. The estimated prevalence of ILD at baseline was 0.44% (95% CI 0.19% to 0.69%). Univariable analyses of baseline factors showed age at onset (p<0.001), ever smoking (p=0.010), seropositivity (p=0.002), RDCI (p<0.001) and lung disease (p<0.001) to be associated with developing RA-ILD. Multivariable logistic regression showed age at onset (OR 1.03, 95% CI 1.01 to 1.05), seropositivity (OR 2.39, 95% CI 1.27 to 4.50) and ever smoking (OR 2.01, 95% CI 1.16 to 3.50) to be associated with ILD development. An increase in RDCI predicted higher ILD (OR 1.46, 95% CI 1.16 to 1.82). More recent recruitment year into the cohort was associated with decreased odds of developing ILD. Table 1. Univariable and Multivariable Analyses. Baseline Variables Univariable analysis Multivariable analysis Odds Ratio (95% CI) p-value All comorbidities (n=2,689)Odds Ratio (95% CI) Lung Disease (n=2,689)Odds Ratio (95% CI) Age at onset (years ) 1.03 (1.02, 1.05) <0.001 1.03 (1.01, 1.05)** 1.03 (1.02, 1.05)** Female Gender 0.56 (0.37, 0.86) 0.009 0.74 (0.47, 1.16) 0.74 (0.47, 1.16) Minority Ethnicity 1.94 (0.76, 4.91) 0.201 - - IMD quintile 0.91 (0.77, 1.07) 0.243 - - Ever Smoked # 1.91 (1.15, 3.15) 0.010 2.01 (1.16, 3.50)* 2.01 (1.20, 3.38)** Recruitment year 0.98 (0.95, 1.01) 0.146 0.95 (0.93, 0.98)** 0.97 (0.94, 0.99)* Body Mass Index (kg/m 2 ) 0.99 (0.95, 1.04) 0.706 - - HAQ-DI 1.21 (0.92, 1.58) 0.172 - - DAS28 1.09 (0.93, 1.27) 0.279 - - Seropositive # 2.40 (1.29, 4.44) 0.002 2.39 (1.27, 4.50)** 2.57 (1.37, 4.85)** Time to DMARD (months ) 1.01 (1.00, 1.02) 0.183 - - RDCI 1.51 (1.24, 1.83) <0.001 1.46 (1.16, 1.82)** - Lung disease 5.29 (3.12, 8.96) <0.001 - 4.69 (2.72, 8.08)** IMD: Index of Multiple Deprivation, HAQ-DI: Health Assessment Questionnaire Disability Index, DAS28: Disease Activity Score 28 joint count, RDCI: Rheumatic Diseases Comorbidity Index **p<0.01, *p<0.05 # Imputed values Conclusion Baseline prevalence of ILD was very low. Seropositivity, smoking, higher comorbidity burden and age at onset increased the odds of subsequent ILD development. Identifying RA-ILD predictors early on is important to ensure appropriate screening, timely diagnosis and treatment. REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests Rositsa Dacheva: None declared, Amanda Busby: None declared, Patrick Kiely: None declared, Adam Young: None declared, David Walsh Consultant of: Consultancy through the University of Nottingham to GlaxoSmithKline plc, AbbVie Ltd, Pfizer Ltd, Eli Lilly and Company, AKL Research & Development Limited, Galapagos, and Reckitt Benckiser Health Limited (each non-personal, pecuniary). Contributed to educational materials through the University of Nottingham, supported by Medscape Education, New York, International Association for the Study of Pain and Osteoarthritis Research Society International (OARSI), each of which received financial support from commercial and non-commercial entities (each non-personal, pecuniary), Grant/research support from: Responsible for research funded by Pfizer Ltd, Eli Lilly and UCB Pharma (non-personal, pecuniary). Receives salary from the University of Nottingham, who have received funding for that purpose directly or indirectly from Sherwood Forest Hospitals NHS Foundation Trust, and UKRI/Versus Arthritis (personal, pecuniary), Daniel McWilliams Grant/research support from: Grant support from Pfizer. Works on collaborative projects between the University of Nottingham and Eli Lilly, UCB and Orion (separately) funded by the relevant pharmaceutical company, but not named as co-investigator, James Galloway: None declared, Elena Nikiphorou Speakers bureau: Speaker honoraria/advisory boards for Celltrion, Pfizer, Sanofi, Gilead, Galapagos, AbbVie, Lilly, Fresenius, Grant/research support from: Holds research grants from Pfizer and Lilly.