Pos1089 psychological distress over 12 months post-diagnosis in a national early inflammatory arthritis cohort

Background In inflammatory arthritis patients, the concomitant decline of their mental wellbeing is an increasing concern [1,2] . It is important to not only describe the trajectory of psychological distress in early disease stages, but also understand which clinical outcome measures are most associated with these changes. Objectives Using data from the National Early Inflammatory Arthritis Audit (NEIAA), we assessed trends in psychological wellbeing over 12 months after initial diagnosis and mapped these against clinical outcomes to identify significant associations. Methods NEIAA collects data from patients referred with suspected early inflammatory arthritis in rheumatology services in England and Wales. We used data provided by 20,472 patients eligible for follow-up (diagnosis of inflammatory arthritis) between May 1st, 2018, and April 1st, 2022. Data items included baseline demographics e.g., age and gender, and clinical variables e.g., rheumatic disease comorbidity index (RDCI), DAS28, and patient reported outcomes. Psychological distress was measured by the sum score of Patient Health Questionnaire Anxiety and Depression Screener (PHQ4ADS). Using mixed effects regression models, we analysed the co-variability of PHQ4ADS with demographic factors and clinical outcomes over 12 months. Time was included as a dummy-coded covariant. Results The analysis included 36% of patients (7,378 out of 20,472) who completed the baseline patient outcome survey. In this cohort, PHQ4ADS scores decreased from a baseline average of 4.7 (CI: [4.6, 4.8]) to 2.62 (CI: [2.5, 2.8]) at 12 months post-diagnosis. The proportion of patients screening positive decreased from 50.0% (CI: [48.9, 51.1]) at baseline to 23.8% (CI: [21.8, 25.9]) at 12 months. At baseline, psychological distress correlated significantly with age, gender, ethnicity, RDCI, prior depression diagnosis, and baseline DAS28 (Figure 1). No significant correlations were found between psychological distress and working diagnosis, seropositivity, or the assessment being recorded after the start of the COVID-19 pandemic. Younger ages were nonlinearly associated with higher distress levels (coefficient per decade: -0.006; p<0.001; CI: [-0.009, -0.003]) (Figure 1a). Distress levels in females were higher than that of males (coefficient: 0.5; p<0.001; CI: [0.4, 0.7]) (Figure 1b). White patients reported lower PHQ4ADS scores compared to non-white patients (coefficient: -0.7; p<0.001; CI: [-1.0, -0.4]) (Figure 1c). Higher distress levels were also associated with higher RDCI (coefficient: 0.2; p<0.001; CI: [0.1, 0.3]) and prior diagnosis of depression (coefficient: 1.8; p<0.001; CI: [1.5, 2.2]) (Figure 1d, 1e). Furthermore, higher baseline DAS28 scores correlated with more severe psychological distress (coefficient: 0.8; p<0.001; CI: [0.7, 0.8]) (Figure 1f). By 12-months, psychological distress decreased significantly overall, which correlated significantly with ethnicity (coefficient: 0.8; p=0.005; CI: [0.3, 1.4]) and baseline DAS28 (coefficient: -0.5; p<0.001; CI: [-0.6, -0.4]). Compared to white patients, the reduction was significantly greater for non-white patients, but the level of distress was no longer different at 12 months (Figure 1c). While those with higher baseline DAS28 showed a greater reduction in psychological distress, the distress levels remained higher at 12 months (Figure 1f). Figure 1. Changes in psychological distress correlated with age, gender, ethnicity, RDCI, prior depression diagnosis, and baseline DAS28. Conclusion In this early inflammatory arthritis cohort, mental health burden was high. Age, gender, ethnicity, RDCI, prior depression diagnosis and baseline DAS28 significantly correlated with psychological distress at baseline. Supporting mental health should be a focus of clinical care for this population and it may be beneficial to use an approach that is culturally valid for non-white patients and accounts for multimorbidity. References [1]Euesden, J, et al. Psychosomatic medicine 79.6 (2017): 638. [2]Lwin, MN, et al. Rheumatology and therapy 7.3 (2020): 457-471. Acknowledgements The authors would like to thank the Healthcare Quality Improvement Partnership (HQIP) as the commisioner of NEIAA, British Society for Rheumatology as the audit providers, Net Solving as the audit platform developers, and the Wellcome Trust (ST12406) for funding to support L.Z.. Disclosure of Interests Lucy Zhao: None declared, James Galloway Speakers bureau: Has received honoraria from AbbVie Celgene, Chugai, Gillead, Janssen, Eli Lilly, Pfizer, Roche, and UCB, Jo Ledingham: None declared, Sarah Gallagher: None declared, Neena Garnavos: None declared, Paul Amlani-Hatcher: None declared, Nicky Wilson: None declared, Lewis Carpenter Consultant of: Statistical consultancy for Pfizer, Kirsty Bannister: None declared, Sam Norton Speakers bureau: Has received honoraria from Janssen and Pfizer.