Search

ObjectiveTo assess the feasibility of using smartwatches in people with knee osteoarthritis (OA) to determine the day-to-day variability of pain and the relationship between daily pain and step count.DesignObservational, feasibility study.SettingIn July 2017, the study was advertised in newspapers, magazines and, on social media. Participants had to be living/willing to travel to Manchester. Recruitment was in September 2017 and data collection was completed in January 2018.Participants26 participants aged>50 years with self-diagnosed symptomatic knee OA were recruited.Outcome measuresParticipants were provided with a consumer cellular smartwatch with a bespoke app that triggered a series of daily questions including two times per day questions about level of knee pain and one time per month question from the pain subscale of the Knee Injury and Osteoarthritis Outcome Score (KOOS) questionnaire. The smartwatch also recorded daily step counts.ResultsOf the 25 participants, 13 were men and their mean age was 65 years (standard deviation (SD) 8 years). The smartwatch app was successful in simultaneously assessing and recording data on knee pain and step count in real time. Knee pain was categorised into sustained high/low or fluctuating levels, but there was considerable day-to-day variation within these categories. Levels of knee pain in general correlated with pain assessed by KOOS. Those with sustained high/low levels of pain had a similar daily step count average (mean 3754 (SD 2524)/4307 (SD 2992)), but those with fluctuating pain had much lower step count levels (mean 2064 (SD 1716)).ConclusionsSmartwatches can be used to assess pain and physical activity in knee OA. Larger studies may help inform a better understanding of causal links between physical activity patterns and pain. In time, this could inform development of personalised physical activity recommendations for people with knee OA.

Background:Long-term studies in rheumatoid arthritis (RA) have reported low inflammation yet high disability over time. It is important to determine which factors are driving this disparity, so appropriate interventions can be used to reduce this gap.Objectives:To identify a subgroup of people with RA with low inflammation yet high disability over 10 years, and describe their characteristics.Methods:Data came from two cohorts of inflammatory arthritis with regular assessments over 10 years: the Norfolk Arthritis Register (NOAR, inclusion: ≥2 swollen joints for ≥4 weeks) from the UK and the Etude et Suivi des Polyarthrites Indifférenciées Récentes study (ESPOIR, inclusion: early RA) from France. Participants provided demographic data and completed patient reported outcomes (PROs, including the Health Assessment Questionnaire [HAQ]). The 2-component Disease Activity Score (DAS28-2C)1, a measure of inflammation, was calculated from swollen joint counts and C-reactive protein level. Inclusion criteria for this analyis: Results:1001 NOAR and 767 ESPOIR participants were included. In both cohorts, a four group trajectory model had the best fit (Figure). Two subgroups were identified in each cohort that demonstrated the hypothesised relationship: similar DAS28-2C but differing HAQ scores (red trajectories in Figure), titled “High HAQ” and “Low HAQ” (mean difference in HAQ over follow-up [95% confidence interval (CI)]: NOAR 0.76 [0.73, 0.80]; ESPOIR 0.89 [0.82, 0.96]). At baseline, the High HAQ groups in both NOAR and ESPOIR were older, had a higher proportion of women, and had higher levels of fatigue (NOAR: odds ratio [OR] 1.16 [95% CI 1.06, 1.28]; ESPOIR: OR 1.20 [95% CI 1.05, 1.36] [Table]) and pain (NOAR only).Table 1.Baseline characteristics / logistic regression analysisNOARESPOIRVariableLow HAQ trajectory, mean (SD)High HAQ trajectory, mean (SD)OR (95% CI)Low HAQ trajectory, mean (SD)High HAQ trajectory, mean (SD)OR (95% CI)N (%)343 (59%)239 (41%)-131 (55%)108 (45%)-Age, years54.9 (14.2)62.1 (13.8)1.07 (1.05, 1.08)47.8 (13.3)51.8 (11.2)1.04 (1.01, 1.06)Women, N (%)224 (65.3%)176 (73.6%)1.82 (1.12, 2.78)100 (76.3%)95 (88.0%)2.73 (1.20, 6.23)Symptom duration, months7.8 (5.1)8 (5.4)1.10 (0.98, 1.05)3.4 (1.8)3.6 (1.8)1.16 (0.98, 1.37)Current smoker, N (%)77 (22.4%)50 (20.9%)1.19 (0.71, 2.00)61 (46.6%)52 (48.1%)1.52 (0.82, 2.83)DAS28-2C3.14 (1.46)3.21 (1.56)-4.65 (1.31)4.41 (1.35)-HAQ0.8 (0.6)1.4 (0.5)-1.1 (0.6)1.6 (0.6)-Pain (0-10)3.7 (2.4)4.6 (2.5)1.16 (1.07, 1.26)4.1 (2.8)5.1 (2.6)1.07 (0.95, 1.20)Fatigue (0-10)4.3 (2.8)5.3 (2.5)1.16 (1.06, 1.28)5.0 (2.6)6.5 (2.5)1.20 (1.05, 1.36)AIMS anxiety3.99 (1.96)4.42 (1.99)1.06 (0.88, 1.29)4.9 (2.26)5.98 (2.25)1.10 (0.94, 1.29)AIMS depression2.85 (1.87)3.38 (1.87)1.10 (0.94, 1.29)3.96 (1.99)5.08 (2.32)1.12 (0.94, 1.33)RF+, N (%)142 (41.4%)106 (44.4%)0.94 (0.60, 1.46)79 (60.3%)50 (46.3%)0.77 (0.34, 1.75)Anti-CCP+, N (%)113 (32.9%)86 (36.0%)1.35 (0.84, 2.17)76 (58.0%)45 (41.7%)0.89 (0.39, 2.05)Conclusion:There is a group of people with RA with high levels of disability, despite low inflammation. These results underline the potential need for pain and fatigue management in people with RA, even when inflammation is low.References:[1]Hensor et al (2019). Rheumatology (Oxford) 58(8)Acknowledgements:Thanks to the participants of NOAR and ESPOIR and those working in the recruiting centresESPOIR Funding:An unrestricted grant from Merck Sharp and Dohme (MSD) was allocated for the first 5 years of the cohort study. Two additional grants from INSERM supported part of the biological database. The French Society of Rheumatology, Abbvie, Pfizer, Lilly and more recently Fresenius and Biogen supported the ESPOIR cohort study.Disclosure of Interests:James Gwinnutt Grant/research support from: Research grant from Bristol Myers Squibb unrelated to this project, Sam Norton Consultant of: Pfizer and AstraZeneca, Kimme Hyrich Consultant of: Abbvie, Grant/research support from: Pfizer and BMS, Mark Lunt: None declared, Bernard Combe: None declared, Nathalie Rincheval: None declared, Adeline Ruyssen-Witrand: None declared, Bruno Fautrel: None declared, Jacqueline Chipping: None declared, Alex MacGregor: None declared, Suzanne Verstappen: None declared

Background:Prescribing behaviour of physicians has been described as a key driver of rising opioid prescriptions and long-term opioid use. However, the effect of prescribers requires interpretation within context. No studies have investigated the extent to which regions, practices and prescribers, vary in opioid prescribing accounting for case-mix by considering this hierarchy together.Objectives:(i) Quantify and identify risk factors for the transition from new-user to long-term opioid user (ii) Quantify variation of long-term use attributed to region, practice and prescriber, accounting for patient mix and chance variation.Methods:We conducted a retrospective observational UK study between 2006-2017 using Clinical Practice Research Datalink. Opioids new users, ≥18 years, without cancer were identified. Long-term opioid use was defined as ≥3 opioid prescriptions issued within a 90-day period from index date, or ≥1 opioid prescription lasting at least 90 days in the first year. A multi-level random-effects logistic regression model was used to examine the association of patient characteristics with the odds of becoming a long-term opioid user. To examine variation in opioid use amongst prescribers, GP practices and region after adjusting for patient case-mix, we used a nested random-effect structure. A ‘high-risk’ region, prescriber or practice was defined as those where the entire adjusted 95% CI lay above the population average.Results:1,968,742 new opioid users were included; 14.6% patients transitioned to long-term use. In the fully adjusted model, factors associated with higher odds of long-term opioid use included high morphine milligram equivalents (MME)/day at first prescription, older age, deprivation, fibromyalgia, rheumatological conditions, major surgery (Table). After adjustment for case-mix, the North-West, Yorkshire and South-West were found to be high-risk regions for long-term use. 103 practices (25.6%) and 540 prescribers (3.5%) were associated with a significantly higher risk of long-term use. The odds of becoming a long-term user for a patient belonging to these prescribers reached up to >3.5 times than the population average.Conclusion:Prescribing factors, age, deprivation and conditions including fibromyalgia and rheumatological conditions were associated with higher odds of long-term opioid use. In the first UK study evaluating long-term opioid prescribing with adjustment for patient-level characteristics, variation in regions and especially practices and prescribers were observed. Our findings support greater calls for action to reduce practice and prescriber variation by promoting safe practice in opioid prescribing.Table.Factors associated with long-term opioid use using a multi-level model accounting for clustering of individuals within prescriber, practice and regionIndividual factorsAdjusted Odds Ratio (95% CI) *Prescribing factorsIndex daily MME >2007.59 (6.29, 9.16)Index daily MME 100-2001.12 (1.03, 1.21)Index daily MME 50-1001.58 (1.49, 1.68)Index daily MME RefGabapentinoid use2.51 (2.43, 2.60)Psychotropic use1.28 (1.17, 1.40)Age>754.35 (4.26, 4.45)65-753.57 (3.50, 3.65)55-653.03 (2.96, 3.09)35-551.91 (1.88, 1.95)Age RefDeprivation (Townsend score)Quintile 5 (Most deprived)1.54 (1.51, 1.57)Quintile 41.34 (1.31, 1.36)Quintile 31.20 (1.18, 1.22)Quintile 21.09 (1.07, 1.11)Quintile 1 (Least deprived)RefPre-existing conditions/ prior proceduresFibromyalgia1.81 (1.49, 2.20)Substance use disorder1.76 (1.70, 1.83)Suicide and self-harm1.56 (1.51, 1.61)Rheumatological conditions Ψ1.54 (1.49, 1.59)Alcohol abuse1.50 (1.45, 1.55)Depression1.28 (1.26, 1.30)Major Surgery1.09 (1.06, 1.13)Abbreviations: MME, Morphine Milligram Equivalent; *pDisclosure of Interests:Meghna Jani Speakers bureau: Grifols, Belay Birlie Yimer: None declared, Therese Sheppard: None declared, Mark Lunt: None declared, William Dixon Consultant of: Bayer and Google

Background:There is considerable debate regarding the most appropriate outcome measures for clinical trials in systemic lupus erythematosus (SLE). We propose that studying changes in disease activity over time may yield important information above and beyond fixed endpoints at 12 months.Objectives:To compare outcomes and predictors of response using i) traditional response definitions at a fixed time point and ii) trajectories of disease activity over time in a clinical trial dataset.Methods:Data from the phase 3 clinical trial of Epratuzumab (EMBODY) in patients with active SLE were reanalysed. Response at 12 months was defined as: reduction to ≤1 BILAG B score, no increase in SLEDAI-2K score and no increase in prednisolone dose. Latent class mixed models (LCMM) were constructed using the numerical BILAG-2004 (nBILAG) score (where A=12, B=8, C=1, D/E=0) from the screening visit until trial end. Cumulative prednisolone and Epratuzumab exposure was determined by estimating the area under the dose-time or concentration-time curves (AUC) respectively.Results:1202 patients were included with a median disease duration of 5.6 [1.9, 12.7] years and baseline nBILAG score of 20 [16, 24]. At 12 months there were 564/1202 (46.9%) responders. There was no association between receiving Epratuzumab and response. Responders were older, with lower baseline nBILAG score and baseline steroid dose. Using a cubic spline LCMM in the whole cohort, 5 trajectories were identified: Non-responders (NR, 33.9%), rapid responders (RR, 36.7%), slow responders (SR, 20.7%), flare (F, 4.7%) and high disease activity (HDA, 3.9%) (figure). Of the 564 12-month responders, 113 (20.4%) were in the NR latent class whilst 417 (73.9%) were in RR or SR classes. Whilst there were no differences in study drop-out due to adverse events, withdrawal due to inefficacy was 20.6% in NR compared to 8.4% in RR and 7.6% in SR.In multivariable logistic regression models adjusted for age, gender and study drop-out, being a responder (RR or SR) was significantly associated with activity in constitutional, musculoskeletal and cardiorespiratory BILAG domains, higher baseline and cumulative steroid exposure, and receiving active drug.There was a trend towards higher baseline CD19+ counts in RR compared to NR or SR (p=0.108) but in a population-averaged multivariable linear model only baseline CD19+ count, Epratuzumab and steroid dose were associated with number of CD19+ cell over time (pConclusion:Compared to a fixed end-point at 12 months, LCMM identified trajectories of disease activity and an association between clinical improvement and receiving active drug. Predictors of response vary significantly according to the response definition used.Table.Comparison of predictors of response between the 2 modelsResponse at 12 monthsResponse latent call (RR or SR)AgeYesNoBaseline gBILAG scoreYes (negative)Yes (positive)Baseline steroid doseYesYesActive diseaseConstitutionalNoYesMusculoskeletalNoYesCardiorespiratoryNoYesGastroenterologicalYesNoAnti-SmithYesNoLow C3YesNoEpratuzumabNoYesAcknowledgments:Presented on behalf of the MASTERPLANS ConsortiumDisclosure of Interests:John Reynolds: None declared, Jennifer Prattley: None declared, Mark Lunt: None declared, Nophar Geifman: None declared, Ian N. Bruce Grant/research support from: Genzyme Sanofi, GSK, and UCB, Consultant of: Eli Lilly, AstraZeneca, UCB, Iltoo, and Merck Serono, Speakers bureau: UCB

Background:Illness perceptions at treatment onset are known to be important predictors of treatment response in rheumatoid arthritis (RA). Yet it is unknown how these perceptions change over time after the initiation of treatment, or which factors are associated with changing perceptions.Objectives:To identify groups of patients with early RA who have similar changes in illness perceptions over the first year following treatment, and assess predictors of these changes.Methods:Patients starting methotrexate (MTX) for the first time were recruited to the Rheumatoid Arthritis Medication Study (RAMS), a one-year prospective cohort. The DAS28 was calculated and patients completed a questionnaire at baseline and 12 months, reporting demographics and completing the HAQ, the Hospital Anxiety and Depression Scale (HADS), pain and fatigue visual analogue scales (VAS) and the Brief Illness Perception Questionnaire (B-IPQ). The B-IPQ consists of eight Likert scales: five represent cognitive illness perceptions (B-IPQ1-5), two represent emotional representations (B-IPQ6 & 8) and one represents illness comprehensibility (B-IPQ7). Change in illness perceptions and EULAR response were calculated over 12 months in those with data at both timepoints. Latent profile analysis was used to identify profiles of patients with similar changes in illness perceptions. Candidate predictors of profile membership were assessed using logistic regression. The association between profile and EULAR response was assessed using ordered logistic regression.Results:In total 1188 patients were included (mean [SD] age: 59.8 [12.7], 781 [65.7%] women). On average, illness perceptions for the whole cohort improved over 12 months, other than patients’ perception of longevity of arthritis (B-IPQ2) and of treatment helpfulness (B-IPQ4). Three profiles were identified: Small Improvers (N=900), Small Deteriorators (N=78) and Large Improvers (N=210) (Figure). Small Improvers improved on all B-IPQ items other than their perception of longevity of arthritis (B-IPQ2) and of treatment helpfulness (B-IPQ4). All B-IPQ items improved in the Large Improvers group to a greater extent than the Small Improvers, other than arthritis longevity (B-IPQ2). The perceptions of Small Deteriorators all worsened, other than arthritis comprehensibility (B-IPQ7). Higher baseline pain was associated with greater odds of being in both the Small Deteriorators and Large Improvers compared to Small Improvers (Small Deteriorators: OR 1.56 per standard deviation (SD) increase in pain [95% CI 1.11, 2.18]; Large Improvers: OR 1.46 per SD increase in pain [95% CI 1.15, 1.85]). Odds of better EULAR response were greater in the Large Improvers (OR 4.37 [95% CI 3.01, 6.33]) and worse in the Small Deteriorators (OR 0.50 [95% CI 0.29, 0.87]) compared to Small Improvers.Conclusion:In general, illness perceptions improved over the first year of MTX treatment and improvements were associated with better treatment response. Worsening illness perceptions may be driven by poor treatment response. These poor illness perceptions at follow-up may compound poor treatment response in the future. Greater understanding of patients’ initial and subsequent illness perceptions is crucial, given the association with treatment response.Figure:Disclosure of Interests:James Gwinnutt Grant/research support from: BMS, Sam Norton: None declared, Kimme Hyrich Grant/research support from: Pfizer, UCB, BMS, Speakers bureau: Abbvie, Mark Lunt: None declared, Anne Barton Consultant of: AbbVie, Lis Cordingley Grant/research support from: Unrestricted award from Pfizer unrelated to current abstract, Speakers bureau: Janssen, AbbVie, Celgene, Sanofi, Eli Lilly, Novartis all unrelated to current abstract, Suzanne Verstappen Grant/research support from: BMS, Consultant of: Celltrion, Speakers bureau: Pfizer

Background:Rituximab (RTX) is B cell-depleting biological therapy used in the treatment of systemic lupus erythematosus (SLE). Response rates are typically around 50% and there are currently no robust biomarkers to identify which patients are more likely to respond.Objectives:To determine whether proteomic analysis of baseline plasma samples could identify novel biomarkers of response to RTX in patients with active SLE.Methods:Patients with SLE (≥ 4 ACR criteria) receiving RTX for active disease (BILAG A or 2 BILAG B scores) from the UK-based BILAG-BR register were included. Improvement at 12 months was defined as a reduction to ≤ 1 BILAG B score with no increase in either SLEDAI score or steroid dose. Proteomic analysis was performed on baseline plasma samples using sequential window acquisition of all theoretical fragment ion spectra mass spectrometry (SWATH-MS). Data were aligned and normalised to give relative protein abundance. A Lasso penalised logistic regression, with an outcome of improvement at 12 months, was used to perform the variable selection procedure using the R package ‘glmnet’. The penalty metric was set at the value which minimised deviance in a 10-fold cross-validation procedure. Confounding variables of sex, age, ethnicity, oral steroid dose, and SLEDAI score had no penalty applied and were thus forcibly retained in the model. Protein-protein interaction networks were visualised using STRING v11.0.Results:70 patients of whom 63/70 (90%) were female with a median (IQR) age of 41 (32, 51) years and disease duration of 11 (6, 20) years at baseline were included. Patients had active disease with a median SLEDAI score of 10 (6, 16). All patients received RTX according to NHS England guidelines. Of 829 proteins, 815 were detected in at least 1 patient. We identified 9 proteins which were associated with clinical improvement at 12 months. A literature search identified that these proteins have relevance to T cell function, complement activation and lupus nephritis. Gender had the greatest weight in the model followed by Serpin Family A Member 10 (SERPINA10) and complement factor B (CFB). Protein interaction networks linked SERPINA10 to CFB by C3 complement (figure).Conclusion:Plasma proteome analysis has identified a protein signature associated with response to RTX in SLE. These findings will be validated in independent cohorts and may offer the ability to predict response to RTX in lupus patients.Protein Interaction Network of the 9 proteins from the multivariable model (with one additional level added). Cyan/magenta lines shown known interactions. Other colours show predicted interactions.Disclosure of Interests: :John Reynolds: None declared, Jennifer Prattley: None declared, Mark Lunt: None declared, Ian N. Bruce Grant/research support from: Genzyme Sanofi, GSK, and UCB, Consultant of: Eli Lilly, AstraZeneca, UCB, Iltoo, and Merck Serono, Speakers bureau: UCB

BackgroundPatients with rheumatoid arthritis (RA) have an increased risk of certain solid cancers, in particular lung cancer, compared to the general population. Treatment with tumour necrosis factor (TNF) inhibitors (TNFi) may further enhance this risk.ObjectivesTo compare the risk of solid cancer in patients with RA treated with TNFi to that in patients treated with non-biologic (synthetic) disease modifying antirheumatic drugs (sDMARDs).MethodsPatients with a physician diagnosis of RA enrolled in the British Society for Rheumatology Biologics Register, a national prospective cohort study established in 2001 to monitor the long-term safety of TNFi, were followed via record linkage with the national cancer registries until first solid cancer, death, for 5 years, or until 2011. Rates of solid cancers in 11 767 patients without prior cancer who received TNFi were compared to those in 3249 patients without prior cancer treated with sDMARDs.Results427 solid cancers were reported in 52 549 patient-years follow-up for the TNFi group (81 (95% CI 74 to 89) per 10 000 patient-years) and 136 cancers were reported in 11 672 patient-years in the sDMARD cohort (117 (95% CI 98 to 138) per 10 000 patient-years). After adjusting for differences in baseline characteristics there was no difference in risk of solid cancer for TNFi compared to sDMARD treated patients: HR 0.83 (95% CI 0.64 to 1.07). There was no difference in the relative risk of cancer for any of the individual TNFi drugs.ConclusionsThe addition of TNFi to sDMARD does not alter the risk of cancer in RA patients selected for TNFi in the UK.

Objectives This study aimed to evaluate whether the early achievement of clinical remission influences overall survival in an inception cohort of patients with inflammatory polyarthritis (IP). Methods Consecutive early IP patients, recruited to a primary care based inception cohort from 1990 to 1994 and from 2000 to 2004 were eligible for this study. Remission was defined as absence of clinically detectable joint inflammation on a 51-joint count. In sensitivity analyses, less stringent definitions of remission were used, based on 28-joint counts. Remission was assessed at 1, 2 and 3 years after baseline. All patients were flagged with the national death register. Censoring was set at 1 May 2011. The effect of remission on mortality was analysed using the Cox proportional hazard regression model, and presented as HRs and 95% CIs. Results A total of 1251 patients were included in the analyses. Having been in remission at least once within the first 3 years of follow-up was associated with a significantly lower risk of death: HR 0.72 (95% CI 0.55 to 0.94). Patients who were in remission 1 year after the baseline assessments and had persistent remission over time had the greatest reduction in mortality risk compared with patients who never achieved remission within the first 3 years of follow-up: HR 0.58 (95% CI 0.37 to 0.91). Remission according to less stringent definitions was associated with progressively lower protective effect. Conclusions Early and sustained remission is associated with decreased all-cause mortality in patients with IP. This result supports clinical remission as the target in the management of IP.

Objectives To evaluate the risk of gastrointestinal perforation (GIP) in subjects with rheumatoid arthritis (RA) treated with antitumour necrosis factor (anti-TNF) therapy compared with non-biological disease-modifying antirheumatic drugs (nbDMARDs). Methods Using data from the British Society for Rheumatology Biologics Register, we compared the incidence of GIPs between 11 881 anti-TNF-treated and 3393 nbDMARD-treated RA patients using Cox regression modelling. Hazard ratios (HRs) with confidence intervals (CI) were calculated. Adjustment was made for potential confounders including current steroid use. The study covered the time period between 2001 and 2011. Results There were 42 (upper 20, lower 22) GI perforations: five in the nbDMARD cohort and 37 in the anti-TNF cohort. After adjustment, treatment with TNF antagonists was associated with an HR of 1.6 (95% CI 0.4 to 6.0) for all GIPs, 2.7 (95% CI 0.4 to 18.1) for lower GIPs and 0.9 (95% CI 0.1 to 5.8) for upper GIPs. Current use of steroids was the single most important predictor of GI perforation with an adjusted HR of 2.9 (95% CI 1.5 to 5.4), but this risk was confined to lower GIPs (HR 8.0, 95% CI 2.6 to 24.1). Conclusions We have not found a statistically significant association between anti-TNF treatment and the risk of GIP.

Objectives To identify demographic and clinical predictors of obstructive lung disease (OLD) and restrictive lung disease (RLD) in patients with established inflammatory polyarthritis (IP) and to compare the prevalence of respiratory symptoms in patients with IP and the general population. Method A total of 421 patients with IP underwent a spirometry test 15 years after inclusion in the Norfolk Arthritis Register (NOAR). Logistic regression analyses were performed to assess the predictive ability of demographic and clinical characteristics obtained at inclusion in NOAR and to assess their association with OLD or RLD at 15 years (age- and gender-adjusted). In addition, the prevalence of OLD and RLD was compared with a matched population (1:4) of people participating in the European Prospective Investigation of Cancer-Norfolk, a representative sample of the general population in Norfolk, UK. Results In this IP population, current smoking was the strongest predictor for OLD and functional disability for RLD. In the comparison study, 11.6% had OLD in the IP population and 4.9% in the general population (adjOR 2.01, 95% CI 1.26 to 3.22). The prevalence of RLD was not statistically different between the IP population and the general population (14.6% vs 17.5%; adjOR 0.76, 95% CI 0.53 to 1.10). Conclusions OLD, but not RLD, is more prevalent in the IP population than in the general population. Functional disability is especially associated with RLD whereas smoking is associated with OLD. The latter finding, and the known association between smoking and a poor disease prognosis, underlines the importance of smoking cessation in patients with IP.

Background The metabolic syndrome (MetS) may contribute to increased cardiovascular risk in systemic lupus erythematosus (SLE). We aimed to examine the association of demographic factors, lupus phenotype and therapy exposure with the presence of MetS. Methods The Systemic Lupus International Collaborating Clinics Registry for Atherosclerosis inception cohort enrolled recently diagnosed (

Objectives To compare the risk of keratinoctye skin cancer (basal cell carcinoma (BCC) and squamous cell carcinoma (SCC)) in patients treated for rheumatoid arthritis (RA) compared with the general population, and to determine whether anti-tumour necrosis factor (TNF) therapy exacerbates this risk. Methods Patients with RA enrolled in the British Society for Rheumatology Biologics Register, a prospective national cohort established in 2001 to monitor the safety of anti-TNF, were followed until 2008. 11 881 patients treated with anti-TNF were compared with 3629 patients receiving non-biological disease-modifying antirheumatic drugs (nbDMARD). Standardised incidence ratios (SIR) were calculated for each cohort and rates between cohorts were compared using Cox proportional HR, adjusted using inverse probability of treatment weighting. Results SIR for skin cancer was increased in both cohorts compared with the English population: SIR 1.72 (95% CI 1.43 to 2.04) anti-TNF; 1.83 (95% CI 1.30 to 2.50) nbDMARD only. In patients without previous skin cancer, BCC incidence per 100 000 patient-years was 342 (95% CI 290 to 402) after anti-TNF and 407 (95% CI 288 to 558) after nbDMARD. HR after anti-TNF adjusted for treatment weighting was 0.95 (95% CI 0.53 to 1.71). SCC incidence per 100 000 patient-years: anti-TNF 53 (95% CI 33 to 79); nbDMARD 43 (95% CI 12 to 110); adjusted HR 1.16 (95% CI 0.35 to 3.84). Conclusions Skin cancers were increased among treated patients with RA. No evidence was found that anti-TNF therapy exacerbates the risk of BCC or SCC but this cannot be excluded. Patients with RA should use sun protection and be monitored for skin cancer.

Objective Previous evidence suggests that women with a history of adverse pregnancy outcomes (APOs) may be at greater risk of developing rheumatoid arthritis. Additionally, one study reported that female patients with rheumatoid arthritis with a history of preonset APOs showed a worse 2-year radiographic outcome than did patients with no APOs. The authors9 aim was to investigate the relationship between preonset APOs (spontaneous abortion or stillbirth) and disease outcome in women with inflammatory polyarthritis (IP). Methods The Norfolk Arthritis Register (NOAR) is a primary-care-based cohort of patients with recent-onset IP; 1586 gravid women who joined NOAR during 1990–2004 were included in this analysis. The authors examined the relationship between patient-reported preonset APOs and disease outcome, measured using the Health Assessment Questionnaire (HAQ) and disease activity score in 28 joints (DAS28 CRP ) (for a subgroup of patients), using linear random effects analysis, adjusted for age and other factors. Results In a predominantly parous cohort (99%), 397 (25%) women reported ≥1 APO before symptom onset. The rates of APOs in NOAR were comparable to the general population. On average, women with a history of ≥2 APOs had significantly higher HAQ and DAS28 scores over time than women with no APOs (mean difference in HAQ 0.13 (95% CI 0.002 to 0.26); DAS28, 0.56 (95% CI 0.01 to 1.11)). This relationship was more pronounced in women with ≥3 APOs (mean difference in HAQ 0.23 (95% CI 0.02 to 0.43); DAS28, 0.98 (95% CI 0.23 to 1.74)). Conclusion Women with two or more APOs before IP onset had a worse disease outcome than women with no APOs.

ObjectivesPast studies have reported conflicting rates of venous thrombotic events (VTEs) in rheumatoid arthritis (RA). The current study aimed to compare (1) the rates of VTEs in patients with RA treated with anti-tumour necrosis factor (anti-TNF) therapy versus those treated with non-biological disease-modifying antirheumatic drugs (nbDMARDs) alone and (2) the rates between each individual anti-TNF agent and nbDMARDs.MethodsUsing data from the British Society for Rheumatology Biologics Register, a national prospective observational cohort study of biological safety in patients with RA, the authors compared the incidence of VTEs between 11 881 anti-TNF- and 3673 nbDMARD-treated patients. Analysis was limited to the first VTE per person. HRs were calculated using Cox modelling. Adjustment was made for potential confounders including surgery performed during follow-up.ResultsA total of 196 first VTEs were reported (151 anti-TNF, 45 nbDMARD). Overall there was no difference in the rates of VTEs between anti-TNF- and nbDMARD-treated patients (adjusted HR 0.8 (95% CI 0.5 to 1.5)). The risk was similar across all anti-TNF agents. Rates of postoperative VTEs did not significantly differ between groups.ConclusionsThese data suggest that anti-TNF therapy is not associated with an increased risk of VTEs in RA patients.

ObjectiveTo evaluate the effect of different concomitant disease modifying antirheumatic drugs (DMARDs) on the persistence with antitumour necrosis factor (anti-TNF) therapies in patients with rheumatoid arthritis (RA).MethodThis analysis included 10 396 patients with RA registered with the British Society for Rheumatology Biologics Register, a prospective observational cohort study, who were starting their first anti-TNF therapy and were receiving one of the following DMARD treatments at baseline: no DMARD (n=3339), methotrexate (MTX) (n=4418), leflunomide (LEF) (n=610), sulfasalazine (SSZ) (n=308), MTX+SSZ (n=902), MTX+ hydroxychloroquine (HCQ) (n=401) or MTX+SSZ+HCQ (n=418). Kaplan–Meier survival analysis was used to study the persistence with anti-TNF therapy in each DMARD subgroup up to 5 years. Multivariate Cox proportional hazard models, stratified by anti-TNF used and start year and adjusted for a number of potential confounders, were used to compare treatment persistence overall and according to the reason for discontinuation between each of the DMARD subgroups, using MTX as reference.ResultsOne-year drug survival (95% CI) for the first anti-TNF therapy was 71% (71% to 72%) but this dropped to 42% (41% to 43%) at 5 years. Compared with MTX, patients receiving no DMARD, LEF or SSZ were more likely to discontinue their first anti-TNF therapy while patients receiving MTX in combination with other DMARDs showed better treatment persistence.ConclusionsThese results support the continued use of background DMARD combinations which include MTX. Consideration should be given to the discontinuation of LEF and SSZ monotherapy at the time anti-TNF therapies are started, with the possible exception of the SSZ+ETN combination.

Objective To examine the influence of post-symptom-onset pregnancy on disease outcome in women with inflammatory polyarthritis (IP). Methods A total of 631 women, aged

Objectives\ud \ud To investigate the influence of early disease-modifying antirheumatic drug (DMARD) treatment on long-term functional outcome in patients with recent-onset inflammatory polyarthritis (IP), and the impact of the duration of first and subsequent DMARD treatment.\ud \ud \ud \ud Methods\ud \ud 642 subjects from a primary care registry of patients with new-onset IP, recruited 1990–4, were followed up for 10 years. Mean change in Health Assessment Questionnaire (HAQ) scores between baseline and 10 years were compared by time to, and time receiving, first DMARD treatment and total time receiving treatment, using linear regression. Adjustment for time-dependent confounders and censoring was performed using marginal structural weights.\ud \ud \ud \ud Results\ud \ud When adjusted for baseline and subsequent disease severity, those treated early ( 3 years experienced an improvement (adj_MDIC in HAQ −0.37; 95% CI −0.77 to 0.04).\ud \ud \ud \ud Conclusions\ud \ud The importance of time to, and response to, first DMARD treatment and total duration of DMARD treatment in modifying the 10-year function in patients with IP has been demonstrated.

Purpose:Methotrexate (MTX) is the first choice conventional disease-modifying antirheumatic drug (DMARD) for rheumatoid arthritis. It is not universally effective, however; although to date it is not possible to predict with any accuracy which patients will respond to treatment. The aim of this analysis was to examine whether clinical and genetic variables could be used to predict response to MTX.Methods:Patients recruited to the Norfolk Arthritis Register (NOAR), a primary care based inception cohort of patients with inflammatory polyarthritis, were eligible for this analysis if they were commenced on MTX as their first DMARD within 3 months of their baseline visit and had at least 2 years of follow-up data. Outcome on MTX was defined as: (1) stopped for adverse events; (2) stopped for inefficacy or second DMARD added; (3) stopped for other reasons; or (4) remained on MTX monotherapy. Multiple logistic regression was used to establish which variables (including demographics, disease activity and Health Assessment Questionnaire score) predicted stopping monotherapy for inefficacy or adverse event (with those remaining on treatment taken as the referent category). The area under the Receiver Operating Characteristic curves (AUC ROC), were used to determine how accurate the model was at predicting outcome.Results:309 patients were included in this analysis. At 1 year (2 years), 34 (46) patients had stopped for adverse events and 25 (49) had either stopped monotherapy for inefficacy or had a second DMARD added. 231 (188) patients remained on MTX monotherapy. The strongest predictor of inefficacy at both time points was shared epitope positivity: odds ratios (OR) 5.8 (95% confidence intervals (CI) 1.3 to 25.6) at 1 year, OR 3.0 (95% CI 1.3 to 7.3) at 2 years. High Health Assessment Questionnaire score (OR 1.84 95% CI 1.12 to 3.01) and female gender (OR 2.2, 95% CI 0.92 to 5.28) were associated with adverse events on MTX at 1 year. However, even the most optimal combinations of the factors analysed were only weakly predictive of treatment outcome: AUC ROC for adverse events 0.68 (95% CI 0.58 to 0.78) and for inefficacy AUC ROC 0.71 (95% CI 0.6 to 0.81).Conclusions:Within this cohort, routine clinical and laboratory factors were poor at predicting outcome of treatment with MTX. Given the major therapeutic advantage to be derived from accurate prediction of treatment outcome, further studies will need to investigate novel biological and other markers.

Background: Lumbar disc degeneration is characterised radiologically by the presence of osteophytes, end-plate sclerosis and disc space narrowing. Aim: To determine the strength of the association between increasing severity of combinations of these features in a population sample of men and women. Methods: Men and women aged ⩾50 years were recruited from a primary care-based community health index in Aberdeen, UK. Participants had lateral spinal radiographs performed according to a standard protocol. The intervertebral disc spaces (L1/2–L4/5) were evaluated for the presence of anterior osteophytes, end-plate sclerosis and disc space narrowing using a graded semiquantitative score (grade 0–3). Log linear modelling was used to determine the associations (pairwise) between increasing severity of these features, with the results expressed as β coefficients and 95% confidence intervals (CIs). Results: There were 286 men (mean age 65.3 years) and 299 women (mean age 65.2 years) with spinal radiographs, yielding a total of 2340 assessable lumbar vertebral levels. In all, 73% of vertebral levels had evidence of osteophytes, 26% of sclerosis and 37% of disc space narrowing. Increasing severity of osteophyte grade was associated with an increasing severity both of sclerosis and of disc space narrowing, whereas the severity of sclerosis was associated with the severity of narrowing. This was true at all vertebral levels. The strongest association, however, was between osteophytes and sclerosis (β coefficient = 2.7, 95% CI 2.4 to 3.1). For sclerosis and narrowing the β coefficient was 1.9 (95% CI 1.7 to 2.1), whereas for osteophytes and narrowing the β coefficient was much weaker at 1.2 (95% CI 1.1 to 1.3). There was no important influence of vertebral level on any of these associations. Conclusion: The association between increasing severity of osteophytes and end-plate sclerosis is stronger than for other combinations of radiographic features of lumbar disc degeneration.

Objective: To compare the clinical and functional outcome at 2 and 5 years in patients with inflammatory polyarthritis treated with either methotrexate (MTX) or sulfasalazine (SSZ) as the first disease-modifying antirheumatic drug (DMARD). Methods: Patients recruited to a primary-care-based inception cohort of patients with inflammatory polyarthritis were eligible for this analysis if they were started on either SSZ (n = 331) or MTX (n = 108) as their first DMARD within 3 months. Outcomes assessed included the Disease Activity Score (DAS)28, Health Assessment Questionnaire, radiological erosions (Larsen Score) and cumulative mortality with the proportions still on the original treatment. To overcome potential bias in allocation to these two treatments, a propensity score was calculated based on baseline disease status variables. Results are expressed as the mean difference between MTX and SSZ, both unadjusted and adjusted for propensity score. Results: The baseline differences between the two groups disappeared after adjusting for propensity score. At 2 and 5 years there were few differences in the clinical outcomes, either unadjusted or after adjustment for propensity. By contrast, at 5 years the proportion that was erosive was lower in the MTX group: odds ratio 0.3 (95% confidence interval 0.1 to 0.8), with a 31% lower Larsen Score after adjustment. At both time points, those treated with MTX were at least twice as likely to remain on that drug as those treated with SSZ. Conclusion: Long-term clinical outcome is similar in patients prescribed MTX and SSZ, although it would seem that MTX has greater potential to suppress erosions, which supports it being the first DMARD of choice.

Objective: To determine the risk of lymphoma in a primary care derived cohort of new onset cases of inflammatory polyarthritis and assess the contribution of disease severity and standard immunosuppressive treatment. Design: Prospective cohort study. Methods: 2105 subjects with new onset inflammatory polyarthritis were recruited to the Norfolk Arthritis Register (NOAR) and followed annually for (median) 8.4 years. Occurrence of lymphoma was determined by annual morbidity review and linkage to the central hospital database serving the NOAR area. Cases of lymphoma were verified by record review. Standardised incidence ratios (SIRs) for lymphoma were calculated compared with the local, age, sex, and calendar year expected rates. Stratified analyses were undertaken for various markers of disease severity and treatment history. Results: There were 11 cases of lymphoma during 15 548 person years of follow up, the majority of which were of large B cell type. Compared with the local population the SIR was 2.4 (95% confidence interval, 1.2 to 4.2). The risks in cases classified as rheumatoid arthritis, ever rheumatoid factor positive, or ever treated with DMARDs were all higher, the highest risk group being those treated with methotrexate: SIR = 4.9 (1.8 to 10.6). Conclusions: There was a doubling in risk of lymphoma in new onset cases of inflammatory polyarthritis. Patients with the most severe disease were twice as likely as other patients to develop lymphoma. These results need to be taken into account when considering reported increased risks of lymphoma compared to background population risk in users of new biological agents.

OBJECTIVES: To estimate the incidence of severe lung disease in patients with scleroderma and identify the combination(s) of features present at first assessment which would be useful to predict future risk of severe lung disease. METHODS: Data were analysed on 561 patients with disease onset occurring on or after 1 January 1982 and disease duration of less than five years before the first assessment. Detailed clinical and laboratory assessments were undertaken at the initial visit. End stage lung disease was defined as pulmonary hypertension requiring continuous ambulatory iloprost, or pulmonary fibrosis requiring continuous oxygen, or death from a scleroderma related lung disease. Patient status was determined at 31 December 1997. The best subset of predictors was identified by Cox regression analysis. RESULTS: In all, 24 patients reached end stage lung disease. The cumulative incidences were 4%, 6%, and 12% at five, seven, and 14 years respectively. As expected, the lung function tests at baseline, including being in the lowest third of either diffusing lung capacity (hazard ratio (HR) = 18.2, 95% confidence interval (CI) 3.5 to 93.8) or of forced vital capacity (HR=4.1, 95% CI 1.1 to 15.2), were highly significant predictors of end stage lung disease. Interestingly, apart from the presence of proteinuria, none of the other baseline variables, including the extent of skin disease and serological markers, were predictive of severe lung disease. CONCLUSION: End stage lung disease was infrequent in this large cohort, but the cumulative incidence increased importantly with time. The risk can be predicted from baseline assessment of pulmonary function. In particular, those with normal pulmonary function at baseline are at very low risk

Background: There is a wide variation in responses to standard disease modifying antirheumatic drug (DMARD) treatment in rheumatoid arthritis (RA). Whether multidrug resistance, failure to respond to several DMARDs, is a specific entity over and above that expected by chance alone is unclear. Objective: To identify patients with RA who demonstrate a multidrug resistant phenotype and to determine what proportion of the variance in drug responses is due to patient related factors. Methods: Patients with RA (1987 American College of Rheumatology criteria) were identified from clinics at Manchester Royal Infirmary and through the Arthritis Research Campaign National RA Repository. The clinic records were reviewed and multidrug resistance was defined as stopping three or more DMARDs owing to lack of efficacy after an adequate trial of the drug. Logistic regression measured by a random effects model was used to determine the relative contribution of the drug and subject related differences to the multidrug resistance. Results: 265 patients (210 (79.3%) female) were studied. The mean (SD) age and disease duration were 52.2 (12.9) and 10.7 (8.8) years, respectively. Patients had a median (range) of 2 (1–8) DMARD courses. Failure of at least one DMARD due to inefficacy occurred in 105 (40%) and 13 (5%) were multidrug resistant. Overall, 35% of the variance in drug responses was due to between-subject differences (p=0.02). Rheumatoid factor (RF) status contributed significantly to this (OR=2.15, 95% confidence interval (95% CI) 1.00 to 4.62) but explained only 3% of the total variance in drug inefficacy. Conclusion: Multidrug resistance occurs in an uncommon (5%) but important subgroup of patients with RA. The between-subject variance is not fully explained by demographics and RF status. Understanding the biological mechanisms that contribute to multidrug resistance may suggest new therapeutic approaches and targets in RA.

Background Methotrexate (MTX) is the most commonly prescribed disease modifying anti-rheumatic drug in the treatment of rheumatoid arthritis (RA). Patients taking MTX are advised to restrict their alcohol consumption because of a theoretical hepatotoxic interaction between alcohol and MTX. However data are limited to support this advice. Objectives To quantify the association between alcohol consumption and abnormal serum liver function tests (LFTs) in patients with RA taking MTX, using routinely collected clinical data. Methods Patients with RA in the Clinical Practice Research Database (CPRD) starting MTX between 1987 and 2011 were studied. Patients were included if they had alcohol consumption details recorded in CPRD and ≥6 LFTs per year, indicating adequate monitoring. Patients were grouped by reported weekly alcohol consumption. Crude rates of abnormal LFTs (defined as alanine transaminase or aspartate aminotransferase level ≥3 times the upper limit of normal) per 1000 person-years follow up were calculated. Cox proportional hazards models described the association between alcohol consumed and development of abnormal LFTs whilst taking MTX univariately, then adjusting for age and gender. Patients were censored at time of first abnormal LFT, death or the end of follow up (31/12/2011). Results 8801 patients were included in the study, 6285 (71%) female, mean age (SD) was 58 (14) years. There were 241 abnormal LFTs in 38000 person-years follow up. There was no difference in rates of abnormal LFTs between drinkers and non-drinkers, adjusted hazard ratio (HR) (95% confidence interval (CI)) 1.12 (0.82–1.51). Crude rates of abnormal LFTs appeared to increase with increasing levels alcohol consumption (table 1); similarly when treated as a continuous variable, each increased unit of alcohol consumed was associated with a higher risk of abnormal LFTs, adjusted HR (95%CI) 1.01 (1.00–1.02). In the adjusted Cox model, moderate alcohol consumption (≤14 units per week) was not associated with a statistically significant risk of developing abnormal LFTs (table 1). There was a non-significant trend to higher HR with higher levels of alcohol consumption, however power was limited in the patients consuming >14 units per week. Conclusions In patients with RA taking MTX, increasing alcohol consumption is associated with an increased risk of developing abnormal LFTs. The clinical importance of this increased risk may be small when drinking ≤14 units per week. Disclosure of Interest None declared

Background There is concern that women with rheumatoid arthritis (RA) are at increased risk of high-grade cervical dysplasia (HGCD) detected via cervical smear. There is also controversy regarding the effect of chronic disease/disability on routine cancer screening attendance in this population. Objectives To (1) compare uptake of cervical screening for women with RA in the British Society for Rheumatology Biologics Register for Rheumatoid Arthritis (BSRBR-RA) with the general population and (2) compare frequency of HGCD between RA and the general population and between patients with RA ever versus never exposed to biologics. Methods The BSRBR-RA, a prospective cohort study, was linked to the NHS Cervical Screening Programme England (CSPE). Data from 2004–2014 was analysed in line with CSPE methods who publish coverage analysis annually; coverage is defined as the number of women who have been adequately screened at the defined age-dependent review period out of the total number of women eligible for screening in this period. Annual coverage was calculated for the eligible BSRBR-RA population and compared to national statistics. Within the BSRBR-RA, coverage was also compared between those patients with mild/moderate disability (mean Health Assessment Questionnaire (HAQ) scores Rates of HGCD (moderate dyskariosis or greater) and low grade cervical dysplasia (LGCD, borderline or mild dyskariosis) were compared to rates of negative smears using risk difference calculations between the BSRBR-RA and national statistics, and between the biologic naive (standard DMARD therapy) and biologic exposed cohorts within the BSRBR-RA. Results Of 13565 linked records, 12939 women had RA of which 12785 had cervical screening data, 2132 (17%) on nbDMARD therapy and 10653 (83%) with biologic exposure. Between 2004–2014 mean five-year coverage was higher in the BSRBR-RA population at 82.5% (95% CI 81.9, 83.1) compared to 79.0% (95% CI 78.5, 79.6) nationally. This mostly reflects significantly higher age appropriate coverage in the 25–49 year group for the BSRBR-RA population: 77.2% (95% CI 75.9, 78.6) compared to 71.4% (95% CI 70.2, 72.6) nationally. Coverage was higher among women with HAQ The risk of HGCD was lower in the BSRBR (0.9%) compared to the general population (1.3%) with a risk difference of -0.004 (95% CI -0.005, -0.002). The risk of LGCD was higher in the BSRBR (7.5%) compared to the general population (5.4%) with a risk difference of 0.021 (95% CI 0.017, 0.025). There was no significant risk difference for HGCD between the biologic and nbDMARD cohorts at 0.009 (95% CI -0.003, 0.021). Conclusions Women in the BSRBR-RA in England have higher rates of cervical screening attendance than the general population, although high HAQ scores had a negative effect on screening attendance. Those aged 25–49 years had markedly higher attendance; this may reflect familiarity with accessing healthcare. Cytology analysis suggests lower risk for HGCD in the RA group but higher risk for LGCD; this may reflect increased screening attendance. Women receiving biologic therapy were not at an increased risk of HGCD compared to those receiving standard DMARD therapy. Disclosure of Interest None declared

Background Early diffuse cutaneous systemic sclerosis (dcSSc) has high morbidity and mortality, and there is currently no cure. Objectives The main objective of ESOS (European Scleroderma Observational Study) was to compare the effectiveness of currently used immunosuppressant treatments. Methods ESOS was a prospective observational cohort study, capturing entry and outcome data in a standard way. Four recommended treatment protocols were decided by the steering committee: methotrexate (MTX), mycophenolate mofetil (MMF), cyclophosphamide, and “no immunosuppressant”. Patients were assessed at baseline, and then 3-monthly for 24 months or (for those entering the study subsequent to September 2013) at least 12 months. The primary outcome measure was the change in modified Rodnan skin score (mRSS – 17 sites) between 0 (baseline) and 12 months. Secondary endpoints largely reflected routine clinical practice and are not reported here. Effects of treatment on the skin score were assessed using linear regression with baseline treatments as predictors. Confounding by indication at baseline was accounted for using inverse probability of treatment (IPT) weights, while censoring weights adjust for bias due to patient attrition. Survival in each protocol was estimated using Kaplan-Meier curves and survival differences were tested based on the results of an IPT-weighted Cox regression model. Results Target recruitment was achieved, with 326 patients from 50 centres (median age 52.3 years, 72% female, median disease duration 11.1 months). Because this was not a randomized study, the number of patients starting on each treatment protocol differed: 65 MTX, 117 MMF, 87 cyclophosphamide, and 57 no immunosuppressant treatment. 273 patients (84%) completed 12 months follow-up and 179 (55%) 24 months follow-up. Median mRSS at baseline was 21.0 units. After 12 and 24 months, 153 (56%) and 139 (78%) of remaining patients had an improved skin score with respect to baseline. The mean change in mRSS after 12 and 24 months was -2.9 and -6.8 units. Predicted decreases in mRSS in each of the protocols are summarized in Figure 1. In both weighted and unadjusted models, the differences between the treatment groups are non-significant (p=0.27 and p=0.16). At 24 months, Kaplan-Meier predicted survival rates (adjusted for confounding) are 94% (MTX), 90% (MMF), 91% (cyclophosphamide) and 87% (no immunosuppressant). However, differences are non-significant (p=0.42). Conclusions 1. All groups showed improvement in MRSS which was not significantly different between groups although there was a trend in favour of methotrexate and MMF. 2. Survival was similar between groups although the no immunosuppressant group had lower survival (but it is important to stress that this was not a true “control” group). Disclosure of Interest None declared

Background Corticosteroid (CS) use has been identified as a key factor associated with damage in SLE but detailed analyses of patterns of use in large international cohorts have not previously been reported. Objectives To describe oral (PO) and parenteral CS use in a large international SLE cohort. Methods The SLICC Inception Cohort recruited patients within 15 months of developing ≥4 1997 American College of Rheumatology (ACR) criteria for SLE, between 2000 and 2011. Details of PO and parenteral steroid use were captured at enrolment and each annual assessment, including type, dose and course dates. All CS data was converted to prednisolone equivalents and analysed using Stata 13. Results 1848 patients have been recruited to the Cohort. Those with follow up data (1686) are included in this analysis, providing data on 10,572 annual follow-up (FU) intervals. The mean (SD) total FU is 6.72 (3.60) years; 1072 (63.5%) patients contributed ≥5 years FU; 343 (20.3%) contributed ≥10 years FU. The proportion of time spent on PO CS can be calculated for 10,098 (95.5%) FU intervals. For the majority of FU intervals (n=9103 (90.1%)), NONE (4494 (44.5%)) or ALL (4609 (45.6%)) of the time was spent on PO CS. In 995 (9.9%) FU intervals, only a proportion of the time was spent on PO CS. In those FU intervals where PO CS have been received (n=5604) the median (IQR) “average daily PO dose” was 6.5 (4.75–10) mg. The average daily CS dose decreased in later FU intervals: median (IQR) at FU 1=10 (5–15) mg and at FU 5=5.5 (4.8–10.0) mg. 558 (34.7%) patients have been on PO CS for ALL of their time in the study, 720 (44.8%) for some of it and 330 (20.5%) have never been on PO CS. 228 patients (13.5%) received parenteral CS prior to and/or at enrolment at a median (IQR) total dose of 1.5 (0.5–3.0) g. CS pulses were given during 429 (4.37%) FU intervals at a median (IQR) total dose of 679.5 (120–2500) mg. This compares to a median (IQR) cumulative total PO dose between FUs of 2565 (1575–4462) mg. PO CS use between FUs (using various definitions - see table 1) was higher in those FU intervals in which pulsed CS have also been given. Conclusions CS are widely prescribed in this international SLE cohort. 79.5% have been on CS at some point and 1 in 3 patients have taken CS throughout their entire disease duration. PO CS tend to be given at low daily doses for long periods of time but with a higher average daily dose earlier in the disease. Parenteral steroid use is associated with higher overall PO CS use, however PO CS contribute most to the total cumulative dose that patients receive. Current management of SLE still relies on CS use as a mainstay of therapy in a high proportion of patients. Acknowledgement The Systemic Lupus International Collaborating Clinics (SLICC) acknowledge funding support of UCB Pharma for this study. Disclosure of Interest None declared

Background Corticosteroids (CS) have been identified as a key factor associated with long term damage in systemic lupus erythematosus (SLE). The Systemic Lupus International Collaborating Clinics/American College of Rheumatology (SLICC/ACR) Damage Index (SDI) has been extensively validated and a previous analysis (Gladman et al J Rheumatol 2003;30:1955–9) suggested that SDI items can be grouped according to their likely association with CS. Objectives To generate international expert consensus opinion on which SDI items have an association with the use of CS. Methods We performed a Delphi exercise amongst SLICC members and associates. A survey requested expert opinions on how each SDI item was influenced by CS use on a Likert scale. The scale ranged from -3 to +3 where -3 = I believe that CS have a strong protective link with this item, 0 = I believe that CS have no effect on this item, +3 = I believe that CS have a strong causal link with this item. A second Delphi round presented each participant with their previous rating and the group median for that item, providing the opportunity for revision. A preliminary univariate analysis of CS exposure against damage was performed within the SLICC Inception Cohort to assess the appropriateness of these groupings and if any other items needed reclassifying. Results Of 43 members/associates, 36 respondents returned 26 complete forms (60.5%). All 26 completed the second round (100%) after which adequate consensus was reached. Four items (cataract, osteoporosis, avascular necrosis and diabetes mellitus) were assessed to be highly likely to be related to CS (median & mode =3, Table 1). Ten other items were “probably”/“possibly” CS related by expert opinion. The remaining items were felt not to be associated with CS. The univariate analysis supported these groupings and identified a significant association between CS exposure and the muscle atrophy/weakness item (e.g. OR (95% CI) 6.81 (1.91–24.27) if on steroids for all of intervening follow up period c.f. some or none of the period). It was agreed that muscle atrophy/weakness should be reclassified into the 9definite association9 group. Conclusions A consensus exercise based on expert opinion and initial analysis in an international inception cohort has generated a preliminary classification of the 42 SDI items according to their likely association with CS. This classification system requires further validation but will facilitate studies examining the impact of CS exposure on SLE populations. Acknowledgement The Systemic Lupus International Collaborating Clinics (SLICC) acknowledge funding support of UCB Pharma for this study. Disclosure of Interest None declared

Background Oral glucocorticoids (GCs) are commonly used to treat rheumatoid arthritis (RA) with GC prescriptions often issued in primary care. Some general practitioners (GPs) may have a greater tendency to prescribe GCs, which could affect the likelihood of a patient receiving a GC prescription. Patient characteristics such as age and comorbidities may influence GC prescribing differently, depending on GC prescribing tendency. Objectives To a) determine whether GC prescribing tendency influences GC prescriptions for RA patients and b) whether baseline patient characteristics affect GC prescribing differently in patients seen by prescribers with high vs low GC prescribing tendencies. Methods Patients with incident RA during a defined follow up period were identified from the Clinical Practice Research Datalink, a large UK general practice database. The primary (i.e. most frequent) prescriber of all medications was determined for each patient. For each primary prescriber, the mean proportion of time their patients spent on GCs during follow up was calculated. The median of these values was used as a cut-off for “low” or “high” GC prescribers. Logistic regression was used to determine the effect of prescriber tendency on the likelihood of receiving a GC prescription during follow up and to test the interaction between prescriber tendency and a range of baseline patient characteristics, including patient demographics, comorbidities that are also indications for GC use and known steroid-related comorbidities. Results 16 536 patients with RA were identified and 3270 GPs were assigned as primary prescribers. The mean proportion of time their patients spent on GCs ranged from 0 to 100% (median 10.2%, IQR 0.1%– 24.4%), and GPs were categorised as ‘high’ prescribers if their patients spent a mean of ≥10.2% of follow-up on GCs. 6,372 (38.5%) patients were assigned a “low” GC prescriber and 10,164 (61.5%) were assigned a “high” GC prescriber. The odds of a patient receiving a GC prescription during follow up was 3.1x greater if they were seen by a “high” GC prescriber compared to a “low” GC prescriber (95%CI 2.88-3.29). The probability of receiving a GC prescription increased with age, but the effect differed significantly between the prescriber groups: OR 1.14 (95%CI 1.10, 1.18) per decade in the “low” GC prescriber group, OR 1.27 (95% CI 1.23-1.30) in the “high” GC prescriber group. Baseline comorbidities that influenced the probability of receiving a GC prescription and that were significantly different between GC prescriber groups were asthma (“low” OR 2.65 CI 2.29-3.06, “high” OR 1.76 CI 1.57-1.98) lower respiratory tract infection (“low” OR 1.75 CI 1.53-2.00, “high” OR 1.45 CI 1.31-1.61), hypertension (“low” OR 1.21 CI 1.03-1.42, “high” OR 1.30 CI 1.18-1.43), and inflammatory bowel disease (“low” OR 0.85 CI 0.46-1.56), “high” OR 1.92 CI 1.21-3.06). Conclusions Prescriber tendency towards high GC use was associated with an increased likelihood of RA patients receiving GCs. The effect of certain patient characteristics on GC prescription differed according to prescriber tendency. Disclosure of Interest None declared

Background The introduction of tocilizumab (TCZ) has provided an expanded choice of both first and subsequent biologic therapies for patients with rheumatoid arthritis (RA). Understanding when in the treatment pathway TCZ is being used as well as describing its real world effectiveness will provide further evidence on which to frame healthcare decisions. The aims of this analysis are to describe and compare the baseline characteristics and 6 months treatment effectiveness between patients (1) starting TCZ or anti-tumour necrosis factor alpha (anti-TNF) therapy as their first biologic and (2) starting TCZ as a first or as a subsequent biologic. Methods The British Society for Rheumatology Biologics Register for RA (BSRBR-RA) is an observational prospective cohort study collecting detailed information on RA patients starting biologic therapies in the UK. This analysis included all patients with RA starting either TCZ or an anti-TNF as their first biologic and patients starting TCZ following prior biologic therapy between 1/1/2010 and 31/12/2013. Baseline characteristics were compared between cohorts. Treatment effectiveness was described using the absolute change in DAS28-ESR and EULAR response at 6 months. Ordinal logistic regression models, adjusted by estimated propensity score using inverse probability of treatment weighting (IPTW), were used to compare 6 months EULAR response between cohorts. Missing data were estimated using multiple imputation. Results 1457 patients starting their first biologic (92 intravenous TCZ, 442 etanercept, 26 infliximab, 301 adalimumab, 596 certolizumab) and 531 patients starting intravenous TCZ as a subsequent biologic therapy were included. There were no significant differences between patients starting TCZ or anti-TNF as their first biologic, with the exception of a shorter disease duration and higher prevalence of pulmonary fibrosis and cancer history among patients starting TCZ (Table). Compared to first line users, patients starting TCZ as a subsequent biologic had longer disease duration, higher HAQ scores, higher corticosteroid use but lower DAS28-ESR. There was no difference in 6 months EULAR response between patients starting TCZ or anti-TNF as their first line therapy (adjusted OR 1.08 (95% CI: 0.67, 1.73). Similar response rates were also seen when TCZ was used as first or subsequent biologic after allowing for differences in baseline characteristics (adjusted OR 0.98 (95% CI: 0.54-1.76). Conclusions Since 2010, it appears that TCZ is primarily used in the UK as a second-line or later biologic. When TCZ was used as first line, treatment response at 6 months is comparable to anti-TNF. Non-response rates were slightly higher among subsequent versus first line users of TCZ, however, these differences could be accounted for by differences in baseline disease severity. Disclosure of Interest M. Kihara: None declared, L. Kearsley-Fleet: None declared, R. Davies: None declared, K. Watson: None declared, M. Lunt: None declared, D. Symmons: None declared, K. Hyrich Grant/research support from: Abbvie and Pfizer

Patients with rheumatoid arthritis (RA) and inflammatory polyarthritis (IP) are at increased risk of premature death, with cardiovascular disease (CVD) forming the primary cause.1–3 However, epidemiological studies have shown that traditional CVD risk factors do not fully account for the increased risk.4 ,5 We have previously reported that IP patients with baseline C reactive protein (CRP) concentrations ≥5 mg/l are at greatest risk of CVD mortality.6 Previous studies have shown that single nucleotide polymorphisms (SNPs) at the CRP gene correlate with CRP levels in a healthy population,7 and haplotypes of these SNPs are associated with CRP levels following acute coronary syndrome.8 Therefore, if CRP SNPs/haplotypes were associated with CVD in individuals with IP, it would implicate CRP in the pathogenesis of CVD (Mendelian randomisation). The aims of this study were (1) to determine the relationship between SNP markers (rs1205, rs1800947 and rs3091244) in the CRP gene, selected …

Tumour necrosis factor (TNF) plays a pivotal role in the control of bacterial intracellular infection.1 The introduction of anti-TNF therapy to treat rheumatoid arthritis (RA) prompted the investigation of infection rates within these patients.2 Following an early signal of an increased risk of tuberculosis in anti-TNF treated patients,3 there have been many reports suggesting an increased risk of other intracellular bacterial infections, including listeria and salmonella.4 ,5 Indeed, the BSR Biologics Register (BSRBR) has reported an increase of such infections within anti-TNF treated patients compared with patients treated with non-biologic disease-modifying antirheumatic drugs.6 In response to these findings, patient information leaflets for anti-TNF agents (etanercept (ETN), infliximab (INF), adalimumab (ADA)) were modified in the UK in January 2006 advising patients to avoid high risk foods, such as raw eggs and poultry, which are associated with an increased risk of these infections.7–9 Given the addition of this advice, we investigated the rate of bacterial intracellular infection, before and after these guidelines were introduced, in RA patients treated with anti-TNF therapy within the BSRBR.10 A total of 11 723 patients with RA starting their first anti-TNF on enrolment in the BSRBR were included. This comprised 9376 patients starting therapy …

While many studies have investigated the risk of cancer following exposure to tumour necrosis factor (TNF) inhibitors,1–3 data are scarce with respect to people with cancer prior to treatment.2 ,4 To our knowledge, no studies have looked at cancer risk following exposure to anti-TNF in patients with a history of premalignant conditions, such as cervical dysplasia and Barrett's oesophagus. Guidelines from the British Society for Rheumatology (BSR) advise caution when using anti-TNF in subjects with premalignant conditions.5 Given the uncertainties regarding the use of anti-TNF in people with premalignant conditions, we investigated the outcome of women with rheumatoid arthritis (RA) with a previous history of carcinoma in situ (CIS) of the cervix registered with the BSR Biologics Register.6 The BSR Biologics Register, established in 2001, is following two cohorts of patients: one cohort of patients with RA …

Background Mortality rates in children with juvenile idiopathic arthritis (JIA) are increased compared to the general population. Systemic arthritis, a more severe subtype of JIA, is expected to be linked to an increase in mortality compared to other subtypes although evidence is so far lacking. Objectives To calculate the standardised mortality rates (SMR) in children with severe JIA, defined as requiring biologic therapy. Methods The study population comprised JIA subjects recruited to the BSPAR Etanercept Cohort Study. Deaths were identified through follow-up via hospital questionnaires and links to the national death register. All patients were followed from registration until death, last follow-up, or 31/12/2012, whichever came first. Poisson regression calculated crude incidence rates per 1,000 person years (pyears). SMRs were calculated using age- and gender-specific mortality rates in the UK population. Results 693 patients were registered and followed over 4475 pyears (mean age 11 years, median disease duration at registration 4 years, mean follow-up per patient 6.4 years). Seven deaths were reported, with an expected rate of 0.96. Four of the deaths occurred in systemic JIA patients. The incidence rate was 1.6/1,000 pyears with an SMR of 7.3 (95% CI) 2.9, 15.0). Patients with systemic JIA had an SMR of 21.7 (95% CI 5.9, 55.4) with non-systemic SMR of 3.9 (0.8, 11.3) (Table 1) Causes of death included macrophage activation syndrome, sepsis, and 2 in association with stem-cell transplants. Conclusions In a select group of patients with severe JIA, there appears to be a high mortality rate, particularly among severe systemic JIA, compared to a healthy paediatric population. Whilst death is still a rare outcome, this study highlights the importance of the diagnosis of JIA. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.3241

Missing data is very common problem in epidemiological studies, and this problem can affect both the accuracy and the precision of any estimates generated from the study data. The magnitude of the problem will depend to some extent on whether the missing data refer to outcomes, exposures or confounders. There are a number of simple methods commonly used for analysing incomplete data, but which rely on unrealistic assumptions about the mechanism causing the data to be missing. Multiple imputation is a more sophisticated method which can produce precise, reliable estimates in a wide range of circumstances, but still relies on unverifiable assumptions about the missing data. I will outline the problems caused by missing data, and explain when simpler methods of analysing incomplete data would be appropriate. I will also explain the idea behind multiple imputation, and the situations in which it can be used effectively. I will illustrate the differences between conventional analyses and multiple imputations analyses, and show how to interpret the results of multiple imputation. My aim is that you will leave able to confidently decide whether it would be appropriate to use multiple imputation to deal with missing data in a given situation, and to understand the statistical output it generates. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.6211

Background The Health Assessment Questionnaire (HAQ) is the most widely used measure of function in studies of inflammatory polyarthritis (IP) and its subset rheumatoid arthritis (RA). Previous research has suggested the mean HAQ over time is j-shaped with an initial improvement followed by an insidious decline. Few studies have attempted to identify distinct subgroups with common HAQ trajectories in IP and RA samples, or considered their validity across cohorts. Objectives To identify common trajectories of HAQ progression in two large prospective observational studies. Methods Data of two large inception cohorts, the Early Rheumatoid Arthritis Study (ERAS) and the Norfolk Arthritis Register (NOAR), were used. ERAS recruited patients with RA from 9 hospital clinics in England between 1986 and 1997: N=1460, mean age 55yrs, 66% female, mean symptom duration 8mths. NOAR is a primary care based inception cohort of early IP in Norwich, England. Patients recruited between 1990 and 1994 were included: N=1027, mean age 53yrs, 66% female, mean symptom duration 9mths. In both cohorts, disease activity and HAQ-scores are measured at baseline and at subsequent follow-up visits. To determine trajectories of HAQ progression over time, latent class growth models (LCGM) were applied to the datasets separately. Age, sex, baseline DAS, symptom duration, rheumatoid factor, and fulfillment of ACR criteria were included as predictors of class membership. Results In both cohorts a 4 class LCGM was selected as providing the best fit. The classes identified were similar in terms of the shape of the trajectories and distribution of patients between classes (see figure). Three classes exhibited a j-shaped trajectory. A fourth class experienced persistently high HAQ that increased early in the course of the disease. In both cohorts older age, female sex, longer symptom duration, fulfillment of ACR criteria and higher DAS were associated with increased likelihood of membership of classes with worse HAQ progression. Conclusions Four subgroups with common HAQ trajectories were derived from the ERAS and NOAR cohorts. That the subgroups identified were nearly identical supports their validity. Identifying distinct groups of patients at risk of poor functional outcome may help to target therapy to those most likely to benefit. Disclosure of Interest None Declared

Background The prognosis for patients with inflammatory polyarthritis (IP) varies and it is difficult to identify at onset which patients will ever achieve remission. Objectives To examine the disease characteristics at i) baseline and ii) five years of patients who persistently, intermittently, or never achieve remission within the Norfolk Arthritis Register (NOAR) over five years. Methods NOAR is a primary-care-based cohort of adults with recent onset IP (≥2 swollen joints lasting for >4 weeks). Disease related information collected over 5 years includes 51- swollen and tender joint counts, HAQ, and DAS28. This study included patients recruited between 2000 to 2004 with a symptom duration of Results Group NR had the highest disease severity at baseline (Table). Group PR were significantly more likely to be on DMARDs at baseline, but the duration on DMARDs at baseline did not differ significantly between groups. Patients with higher HAQ scores (adjs=0.44, 95% CI 0.07, 0.80) or DAS28 scores (adjs=0.47, 95% CI 0.26, 0.68) at baseline, were more likely to be in the IR or NR group. There was a significant effect of remission group on 5th year HAQ scores (p Conclusions A small proportion of patients achieved persistent remission, and patients in this group were more likely to have had DMARD treatment at baseline. Acknowledgements Funded by Arthritis UK and GlaxoSmithKline. Disclosure of Interest None Declared

Background Glucocorticoid (GC) therapy is used by more than one in two patients with Rheumatoid Arthritis (RA)[1][1]. GCs are effective [2][2] but have side effects including Diabetes Mellitus (DM) [3][3]. Objectives The aim of this retrospective cohort study was to quantify the risk of incident type II DM in patients with RA treated with oral GCs, and its relationship with dose. Methods Adult patients with RA were identified from a large UK primary care research database, the Clinical Practice Research Database (CPRD) using a validated algorithm during the study period 01/92-12/09. Patients with prevalent DM at the time of their first code for RA were excluded. GC exposure from first code for RA was considered using several models including a time-varying binary indicator of ever or current use, current daily dose, average daily dose and cumulative dose. Incident DM was defined as a READ code for type II DM, at least two oral anti-diabetic prescriptions or abnormal blood results (blood sugar, HbA1C or glucose tolerance test). Gender, age, BMI, smoking status, family history of DM, hypertension, prior cumulative dose of oral GC, current DMARDs and ever NSAID use were potential confounders. Multiple imputations were used to impute missing data for BMI and smoking status. Incidence rates for type II DM were calculated for different patterns of GC exposure. Crude and adjusted Hazard Ratios (HR) were estimated using Cox regression. Results 23,736 adult RA patients were included. 70% were female with a median age of 59 years (IQR 49-71). Median time at risk per patient was 5.39 years (Range: 0.003-18.0). 2,462 patients were diagnosed with type II DM during follow-up: incidence 14.0 events/1000 person years (pyrs) in unexposed patients and 21.9 events/1000pyrs in time following GC exposure. The crude HR was 1.53 (95%CI 1.41-1.66) in ever GC users compared with non-use. After adjusting for all covariates, the HR reduced to 1.38 (95%CI 1.27-1.51). This equates to one additional case of DM per year for every 185 patients currently receiving GCs. Each 5mg increase of current oral GC was associated with a 14% increased risk of DM (HR:1.14; 95%CI 1.11-1.17). Patients currently taking between 10-30mg/day had an adjusted HR of 1.95 (95%CI 1.62-2.34) compared to non-use, equating to one additional case of DM for every 67 patients treated. A 5mg increase in average daily dose was associated with a 32% increased risk (HR 1.32; 95%CI 1.26-1.39), suggesting prolonged exposure increased risk. Conclusions Oral GC therapy is a significant and clinically important risk factor for incident Type II DM in patients with RA. Screening for DM might be warranted in patients taking oral GC therapy. References 1. Caplan L, Wolfe F, Russell AS, Michaud K. Corticosteroid use in rheumatoid arthritis: prevalence, predictors, correlates, and outcomes. J Rheumatol. 2007;34(4):696-705. 2. Gotzsche PC, Johansen HK. Meta-analysis of short-term low dose prednisolone versus placebo and non-steroidal anti-inflammatory drugs in rheumatoid arthritis. Bmj 1998;316(7134):811-8. 3. Hoes, J. N., M. C. van der Goes, et al. (2011). “Glucose tolerance, insulin sensitivity and β-cell function in patients with rheumatoid arthritis treated with or without low-to-medium dose glucocorticoids.” Annals of the Rheumatic Diseases 70(11): 1887-1894. Disclosure of Interest None Declared [1]: #p-7 [2]: #p-8 [3]: #p-9

Background Rheumatoid Arthritis (RA) is associated with accelerated atherosclerosis and inflammation associated with RA may increase the propensity to develop dyslipidaemia and insulin resistance. In established RA, HDL-C (HDL) and ApoA-1 are negatively correlated with disease activity and insulin resistance is increased in RA patients. A few studies have suggested that lipid and glucose handling may be impaired prior to the onset of RA with Apo-B and triglyceride levels being found to be higher and HDL levels lower in those individuals who later develop inflammatory polyarthritis (IP). Objectives To investigate the association between lipid levels, diabetes and glycated haemoglobin on future risk of IP. Methods In the European Prospective Investigation of Cancer in Norfolk (EPIC-Norfolk) cohort, all participants completed a questionnaire at baseline and provided a non-fasting blood sample for routine lipid analysis. In subsets, additional biomarkers were measured including HbA1c, Apo-A1 and ApoB. Individuals with incident IP were identified by linkage with the Norfolk Arthritis Register (NOAR), a primary-care based disease register with an overlapping catchment area. We assessed the association of HDL, Apo-A1, ApoB and HbA1c with future risk of IP using Cox proportional hazard models. Confounders included age, gender, smoking status, smoking intensity, self-reported diabetes (DM) and Body Mass Index (BMI) and were controlled for in the final analyses. Results 25,455 EPIC participants aged 40-79 years were followed for a median (IQR) of 14.2 (12.9, 15.3) years (342,916 person-years of follow-up). 184 developed incident IP (69.6% women). Univariately, HbA1c [HR=1.3 (95% CI: 1.09, 1.55) per 1% increase] was associated with an increased risk of IP, remaining significant after further adjustment for confounders [HR=1.24 (95% CI: 1.05, 1.47) per 1% increase]. Univariately, neither Apo-A1, Apo-B nor HDL levels predicted IP onset. When multiple biomarkers were included in a single multivariable model, both HDL and HbA1c were significantly associated with incident IP [HR 0.76 (95%CI: 0.57, 1.00) per 0.4mmol/l and 1.24 (95% CI: 1.05, 1.47) per 1% respectively]. Conclusions In a prospective population-based cohort study low HDL and high HbA1c were identified as being associated with subsequent IP development. Several anti-inflammatory properties of HDL may be of relevance to its potential protective role. Glycated haemoglobin reflects insulin resistance which can promote a pro-inflammatory state in an at-risk population. Disclosure of Interest None Declared

Background The patient acceptable symptom state (PASS) is a patient reported outcome used to identify the health state patients find acceptable or satisfactory. Limited data is available about the contribution of demographic, disease and patient measures indicating a satisfactory health state according to the patient. Objectives To determine the association between demographic and clinical variables with a satisfactory patient rated health state. Methods Consecutive patients from the Rheumatoid Arthritis Medication Study (RAMS) were included. RAMS is a national multi-centre observational study including patients with rheumatoid arthritis or undifferentiated arthritis who are starting methotrexate (MTX) for the first time. Patients were clinically assessed by a research nurse at time of MTX start (baseline). Patients (n=163) recruited to date with complete baseline data including PASS, pain (100mm VAS), fatigue (100mm VAS), physician global assessment (100mm VAS), DAS28, HAQ-score and quality of life (Qol, EQ-5D) were included in this study. PASS was formulated as ”Is your current condition satisfactory, when you take your general functioning and your current pain into consideration” (yes/no). First, univariate logistic regression analyses were applied to determine the association between each of the demographic and clinical variables with PASS separately. Data are presented as odds ratios (OR, 95%CI) per change in SD. Secondly, forward stepwise logistic regression was performed (pe=0.2 and pr=0.1) and subsequently the total contribution of all variables included in the final model was calculated (AUC). Results Median [IQR] age was 59 [46 – 67] yrs and median symptom duration was 12 [4 – 29] months; 111 (68%) patients were female. Sixty-six (40%) responded that they were satisfied with their condition. Satisfied patients were compared to unsatisfied patients (median; OR 95%CI): those satisfied with their health state were more likely to be older (62 vs 56; 1.59, 1.13 to 2.25) and less likely to be women (34% vs 66%; 0.45, 0.23 to 0.87). In addition, all disease related measures were significantly better in patients who were satisfied compared to those who were not: VAS pain (30 vs 66; OR 0.23, 95%CI 0.15 to 0.37), VAS fatigue (33 vs 67; OR 0.27, 95%CI 0.17 to 0.41), VAS physician (29 vs 57, OR 0.33, 95%CI 0.22 to 0.50), Qol (0.727 vs 0.516, OR 4.76, 95%CI 2.54 to 8.91), HAQ-score (0.56 vs 1.63; OR 0.29, 95%CI 0.19 to 0.44) and DAS28 (4.4 vs 5.6, OR 0.41, 95%CI 0.27 to 0.60). Symptom duration was the only variable not statistically significant. In the final model the strongest predictor of a satisfactory health state was VAS pain ( Conclusions In this cohort of patients starting MTX for the first time, forty percent found their health state satisfactory. Patient reported outcomes and older age mainly explain fulfilment of this state, with many still having moderate levels of disease activity. Disclosure of Interest None Declared

Background Remission has become an achievable goal when treating patients with inflammatory polyarthritis (IP) and its subset rheumatoid arthritis. Previous studies suggest that early and sustained remission prevents long-term disability in patients with IP. Whether the achievement of the status of remission may also affect survival needs to be investigated. Objectives To estimate the relationship between clinical remission and overall survival in an inception cohort of patients with IP. Methods Consecutive patients with early IP from a primary-care based inception cohort, recruited between 1990 and 1994 (first cohort) and between 2000 and 2004 (second cohort), were eligible for this study. Patients were assessed yearly after inclusion by a research nurse and flagged with the national UK death register. Baseline assessment included a 51 tender and swollen joint count (JC), the HAQ-score; blood was collected to determine CRP, RF and ACPA. The 51- tender and swollen JC were also assessed at 1, 2 and 3 yr after registration. Remission was defined as the absence of clinically detectable joint inflammation (swollen 51-JC=0 and tender 51-JC=0). Only patients with at least 3 years of follow-up were included in the present study. Different variables for remission were evaluated: ever remission, number of assessments in remission and time of first remission within the first 3 years. All-cause death was the main outcome of this study. Censoring was set at 1st May 2011. The relationship between remission and mortality was analysed using the Cox proportional hazard regression model (start-date of survival was set at 3 yrs after inclusion). Multivariate analyses were applied to adjust for pre-specified relevant predictors, including demographics, cohort, baseline disease activity and severity, and cumulative treatment variables. Multiple imputation of predictors was used to optimize available information. The results are shown as hazard ratios (HR) and 95%CI. Results A total of 2,769 patients were eligible for the analyses, 1,604 from the first and 1,165 from the second cohort, with a median follow-up of 105 months. 578 subjects died during the observation time. 962/2,589 (37.2%) fulfilled the predefined 51-JC remission criteria at least once within the first 3 yrs. Having been in remission at least once within the first 3 years of follow-up was associated with a significant better survival: crude HR [95%CI] 0.81 [0.67, 0.96] and adjusted HR [95%CI] 0.80 [0.66, 0.96]. The number of times in remission was also associated with a significantly decreased risk of all-cause mortality: per each additional time spent in remission crude HR [95%CI] was 0.90 [0.82, 0.99] and adjusted HR [95%CI] was 0.90 [0.81, 0.99]. Looking at the effect of time lag until the first remission, those patients who were in remission 1 year after the first assessment had the greatest reduction in mortality risk compared to patients who never achieved remission within the first 3 years (adjusted HR 0.61 [0.43, 0.87]), while no significant association was found for patients who achieved remission at a later time point, i.e. at second (adjusted HR [95%CI] 0.80 [0.60, 1.07]) and third year (HR [95%CI] 0.97 [0.72, 1.32]). Conclusions Early remission and sustained remission are associated with a decreased all-cause mortality in patients with IP. Disclosure of Interest None Declared

Background A rheumatoid arthritis (RA) susceptibility region between the TNF receptor-associated factor-1 and complement component 5 genes ( TRAF1/C5 ) has previously been reported to associate with radiological damage in two studies. Objectives We aimed to test rheumatoid arthritis (RA) genetic susceptibility markers for association with a continuous measure of radiological damage over time using longitudinal modelling techniques in a discovery cohort of patients with inflammatory polyarthritis (IP). An independent inception cohort of UK IP patients was used as a replication cohort, and also combined together with the discovery cohort in order to increase sample size and power. Methods Sixty-seven RA susceptibility variants were genotyped in 474 patients from the Early Rheumatoid Arthritis Study (ERAS) using Sequenom MassArray technology. Correlation between genetic markers and Larsen score, as a time dependent trait, was assessed longitudinally using zero-inflated negative binomial regression to include repeat measurements in the same individual at different time points. Genetic markers associated with radiological damage in ERAS were tested using the same modelling techniques on previously published data from the Norfolk Arthritis Register (NOAR). Finally, the model was adapted to accommodate differences in disease progression between the two cohorts and a combined analysis was performed. Results rs2900180 at the TRAF1/C5 locus was the single marker associated longitudinally with Larsen score in the ERAS cohort (p=0.02; time averaged effect size: 2.96 Larsen units) and in the NOAR cohort (p=0.04; effect size: 1.56). Analysis of individual time points in ERAS showed that TRAF1/C5 polymorphisms display their effect primarily on the extent of Larsen score early in the disease course (rs2900180, year 2: coef. 0.35 95%CI 0.04, 0.66, p=0.025; rs10760130, year 2: coef. 0.34 95%CI 0.03, 0.65, p=0.030). Combined longitudinal analysis of the two cohorts suggests further association of several loci with Larsen score ( KIF5A, PTPN22, AFF3, TAGAP ). Conclusions The genetic marker rs2900180 is associated with extent of erosions, as measured by the Larsen score, in the ERAS cohort. This represents the third independent study correlating genotype at the TRAF1/C5 locus with radiological severity in RA. Replication in a large dataset is required to establish the role of other RA susceptibility loci in disease severity. Disclosure of Interest None Declared

Background Juvenile Arthritis Disease Activity Score (JADAS) is a 4 variable composite disease activity (DAS) score for juvenile idiopathic arthritis (JIA), including active 10, 27 or 71 joint count (AJC), physician global (PGA), parent/child global (PGE) and ESR. We have shown in previous work that the ESR is not routinely assessed in all patients with JIA; this reduces the feasibility of JADAS use in clinical practice. Objectives To determine whether exclusion of the ESR from JADAS (labelled JADAS3 followed by 10, 27 or 71 respectively) influences correlation with single markers of disease activity (DA), in a prospective inception cohort of UK children presenting with new onset inflammatory arthritis. Methods The JADAS 10, 27 and 71 and JADAS3-10, -27 and -71 were determined for all children in the Childhood Arthritis Prospective Study (CAPS) with complete data available at baseline. Correlation of JADAS 71 and JADAS3-71 with single markers of DA was determined for all ILAR subtypes (Spearman’s rank statistic). Results 354/1041 (34%) children had all four variables and 562/1041 (54%) children had three variables (ie without ESR) available at baseline. 994 (95%) had AJC available, 811 (78%) PGA, 694 (67%) PGE and 647 (62%) ESR. Median age at disease onset was 6.8 years (IQR 2.8-10.8), median disease duration at diagnosis was 5.2 mths (IQR 2.5-10.9) and 64% were female. Correlation for the whole cohort between JADAS3-71 and both JADAS3-27 and -10 was 0.99. Median JADAS3-71 ranged from 5.9-16.7/91 across the subtypes (Table 1). Correlation of JADAS 71 with JADAS3-71 was high, both for the whole cohort (0.96) and within individual ILAR subtypes (range 0.91 to 0.99). Correlation of JADAS 71 and JADAS3-71 with single markers of DA was similar between subtypes with the exception of the ESR; this may simply be a reflection of the variation in the frequency of measurement of the ESR between subgroups. Conclusions Correlation of single DA markers with JADAS3-71 was moderate to good across all ILAR subtypes. JADAS3 is potentially a valid surrogate for JADAS in the clinical setting when the ESR is unavailable. Further validation studies are required to determine whether JADAS and JADAS3 reflect changes in DA over time. Disclosure of Interest None Declared

Background Increased alcohol intake and fructose consumption may be associated with a lower risk of inflammatory arthritis. The mechanism of action may be related to a rise in serum uric acid (SUA), a circulating anti-oxidant. Objectives To examine the association between intake of alcohol, red wine, beer and fructose, and SUA with risk of inflammatory polyarthritis (IP). Methods Lifestyle data were obtained from 25,455 adults without IP and aged 40-79 using a health and lifestyle questionnaire (HLQ) in the population-based European Prospective Investigation of Cancer (EPIC-Norfolk) study from 1994-97. Over 342,916 person-years of follow-up, 184 individuals who subsequently developed IP were identified by linkage with the Norfolk Arthritis Register (NOAR), a primary-care based disease register. Controls were taken from a random sample of 4000 EPIC participants (case-cohort design), which due to overlap between cases and availability of 7-day diet diaries (7DD) resulted in a dataset of 4098 participants, of which 171 had IP. Results Correlation between the HLQ and 7DD for alcohol intake was good (ρ =0.8). The median (IQR) SUA for cases and unaffected cohort was 280.6 (237.7, 356.4) and 287.7 (235.7, 345.6) μmol/L respectively. In linear regression SUA showed a weak but significant association with alcohol intake in men (β coeff 4.17 (2.61,5.73), p 2 =0.02) and a weak negative association with fructose intake (mainly from fruits) in women (β coeff -0.37 (-0.75,0.00), p=0.05, adj r 2 Conclusions Fructose and alcohol intake may have gender specific influences on the development of future IP. Such effects are, however, unlikely to be mediated by the uric acid-allantoin anti-oxidant pathway. Disclosure of Interest None Declared

Background The development of new classification criteria for rheumatoid arthritis (RA) calls for a re-estimation of RA incidence rates. The new criteria purport to have increased sensitivity to identify RA in patients with early inflammatory arthritis (EIA). Objectives To estimate the age and sex-specific incidence of RA in Norfolk, England using the new criteria Methods This study included all patients aged ≥16 notified to the Norfolk Arthritis Register (NOAR), a primary-care based inception cohort of patients with EIA, from 1990-5 with symptom onset in 1990. The Norwich Health Authority population based on the 1991 census was the denominator. Age and sex specific incidence rates using both 1987 and 2010 classification criteria were calculated at baseline visit, annually for the first 3 years and at 5 years. At each follow up both criteria sets were applied cross-sectionally and cumulatively, i.e. once a patient had satisfied a particular criterion that result was carried forward to future assessments. Results 283 patients were notified to NOAR with symptom onset in 1990. 23 were excluded as an alternative diagnosis was made by their rheumatologist. The overall incidence rate (IR) when applying the 2010 criteria at baseline was 54/100 000 for women and 25/100 000 for men, compared to 44/100 000 for women and 19/100 000 for men using 1987 criteria at baseline. Age and sex-specific IRs using the 2010 classification criteria at baseline were similar to cumulative IRs applying the 1987 criteria up to 5 years. IRs using both criteria sets converged when applied cumulatively. However, 10 patients (4%) satisfied the 1987 criteria cumulatively after 5 years but never satisfied the new criteria, and 37 (14%) satisfied the new criteria but never satisfied the 1987 criteria. Conclusions The 2010 classification criteria aim to identify early, those patients with EIA who, in the absence of appropriate treatment, would go on to develop persistent, erosive and disabling RA. This study shows that the new criteria classify at baseline similar numbers of patients as having RA, as the previous criteria would have taken up to 5 years to identify. A small proportion of patients (18%) satisfied only one set of classification criteria over 5 years. Those patients who satisfied the 1987 criteria but not the 2010 had more hand joint involvement, symmetrical arthritis and morning stiffness. These results provide important information for health economics evaluation and healthcare planning. Disclosure of Interest None Declared

Objectives To evaluate the risk of gastrointestinal perforation (GIP) in subjects with RA treated with anti-TNF therapy compared to non-biological disease-modifying antirheumatic drugs (nbDMARDs). Methods The analysis was conducted in the British Society for Rheumatology Biologics Register (BSRBR), a prospective cohort study. Patients with RA starting treatment with the TNF inhibitors etanercept (ETA), infliximab (INF) or adalimumab (ADA) and a biologic-naive comparison cohort exposed to nbDMARDs were recruited between 2001-2009. Subjects were followed until 31/3/2011 or death, whichever came first. Events were ascribed to anti-TNF if they occurred while the patient was receiving anti-TNF therapy or within 90 days of the first missed dose. Events were attributed to the most recent drug received in patients who switched anti-TNF therapy. Cox proportional hazards model were developed to compare the rates of GIP between cohorts. A propensity score was used to balance covariates between treatment groups and included age, gender, weight, comorbidity, smoking, indexes of disease severity, use of steroids, NSAIDs, and gastroprotective drugs. Stratification by deciles of propensity score for treatment was used to adjust for confounding between the groups. Current steroid use was entered into the model as a time-varying covariate. Results There were 58 (upper: 25, lower: 33) GI perforations: 5 in the nbDMARD cohort, 37 in the anti-TNF cohort (on drug + 90 days), and 16 in the anti-TNF cohort (off drug). In univariate analyses higher HAQ, history of hypertension, and smoking were associated with the risk of lower GIP, and use of nonselective NSAIDs at baseline and low weight with the risk of upper GIP, irrespective of anti-TNF exposure. After adjustment, treatment with TNF antagonists was associated with hazard ratio (HR) of 1.6 (95% CI 0.4 to 6.0) for all perforations, 2.7 (95% CI 0.4 to 18.1) for lower and 0.9 (95% CI 0.1 to 5.8) for upper GI perforations. Current use of steroids was the single most important predictor of GIP with adjusted HR of 2.9 (95% CI 1.5 to 5.4), but this increase of risk was confined to lower GIPs (HR 8.0, 95% CI 2.6 to 24.1). Conclusions Anti-TNF therapy was not associated with statistically significant increase of all GI perforations, although there was a signal of possible increased risk of lower GI perforations. Current use of steroids was the single most important risk factor for lower GI perforations. Disclosure of Interest None Declared

Background The Metabolic Syndrome (MetS) is a clustering of metabolic abnormalities associated with an increased risk of developing diabetes and atherosclerosis, and may add to the increased cardiovascular risk seen in SLE. Inflammatory disease activity and therapeutic exposures, particularly to corticosteroids, may contribute to the development of MetS in SLE. Objectives We examined the impact of inflammation and corticosteroid exposure on the prevalence of MetS in a multicentre international inception cohort of patients with SLE. Methods The Systemic Lupus International Collaborating Clinics Registry for Atherosclerosis (SLICC-RAS) inception cohort enrolled recently diagnosed (

Background Juvenile Arthritis Disease Activity Score (JADAS) is a 4 variable disease activity (DA) score for JIA (including active 10, 27 or 71 joint count (AJC), physician global assessment (PGA), parent/child global evaluation (PGE) and ESR). Much of the original validation data were derived from clinical trials (predominance of polyarticular course disease). The validity of JADAS for all ILAR subtypes of JIA in the routine clinical setting is unknown. Objectives To investigate the construct validity of JADAS in all ILAR JIA subtypes in a prospective inception cohort of UK children with new onset inflammatory arthritis at the time of initial presentation. Methods JADAS 10, 27 and 71 were determined for all children in the Childhood Arthritis Prospective Study (CAPS) with complete data available at baseline. Correlation of JADAS 10, 27 and 71 with single DA markers was determined for all ILAR subtypes of JIA (Spearman’s rank statistic). Results 354/1041 (34%) children had all 4 variables available at baseline. 994/1041 (95%) had AJC available, 811 (78%) PGA, 694 (67%) PGE and 647 (62%) ESR. Median age at disease onset was 6.8 years (IQR 2.8-10.8), median disease duration at diagnosis was 5.2 mths (IQR 2.5-10.9) and 64% were female. Correlation for the whole cohort between JADAS 71 and both JADAS 27 and 10 was 0.99, approaching 1 for all ILAR subtypes. There was a significant difference in median JADAS scores between subtypes (p Correlation of JADAS71 with single markers of DA ranged from 0.48-0.75 for the whole cohort (see Table 1). Within the individual subtypes, the strongest correlation was with AJC (range 0.71-0.94). Correlation with PGA and PGE was moderate (range 0.29-0.75) and correlation with the ESR was strongest for undifferentiated and systemic JIA (0.76 and 0.59) but low to moderate for all other subtypes of JIA (range 0.2-0.41). Conclusions JADAS71 correlated well with AJC, PGE and PGA but did not correlate well with the ESR in all ILAR subtypes. Correlation of JADAS71 with measures of disease severity is moderate, with some variation between subtypes. The ESR is not routinely measured in all children with JIA in the UK, as illustrated by its availability in just 62% of this cohort at baseline. Given the high frequency of unmeasured ESR assessments, we suggest further work is required to assess the validity of JADAS without inclusion of the ESR. Disclosure of Interest None Declared

Background Previous research has suggested that the severity of rheumatoid arthritis (RA) has decreased over time. Objectives To examine the pattern of disease severity in patients with inflammatory polyarthritis (IP) at time of registration into the Norfolk Arthritis Register (NOAR). Methods NOAR is a primary-care-based cohort of adults with recent onset IP (≥2 swollen joints lasting for >4 weeks). Disease related information collected at baseline includes 51- swollen and tender joint counts, HAQ, and DAS28. Blood is collected and C - reactive protein (CRP), Rheumatoid Factor (RF), and Anti-Citrullinated Peptide Antibodies (ACPA) are measured. In this study, patients with a symptom duration of Results In total 3045 patients were evaluated: Cohort 1=1022; Cohort 2=883; Cohort 3=631; Cohort 4=509 (Table). Calendar year of assessment was a significant predictor of lower DAS28 scores over time for the total sample (adjs= -0.0007, 95% CI -0.001, -0.0001), and each sub-sample, indicating improved disease activity over time. Within the total sample, and the DMARD naive sub-sample, there was a significant effect of ‘Cohort’, and a significant ‘Calendar year by Cohort’ interaction indicating the rate of change in DAS28 scores over time differed between cohorts. Calendar year of assessment was not a significant predictor of HAQ scores for the total sample (adjs=0.0002, 95% CI -0.0002, 0.001), and was also a significant predictor of increased HAQ scores for the sub-sample who met the ACR/EULAR 2010 RA criteria, indicating no improvement in functional disability over time. Conclusions Baseline disease activity has decreased over time and may not be solely attributable to the increased use of DMARDs. However, there has been no improvement in levels of functional disability. Acknowledgements Funded by Arthritis UK and GlaxoSmithKline Disclosure of Interest None Declared

Background Recent advances in the identification of loci associated with susceptibility to complex disease have led to methods being developed that incorporate this information into genetic screening models to identify individuals at high risk of disease. Objectives Here we present the first risk prediction model including all 46 known genetic loci associated with rheumatoid arthritis (RA) in Caucasian populations. Methods Weighted genetic risk scores (wGRS) were created using odds ratios from 45 RA susceptibility SNPs [[1][1]] and a HLA-DRB1 tag SNP or imputed HLA-DRB1 alleles. The wGRS were tested in 11,370 RA cases and 15,536 healthy controls of known genotype. The risk of developing RA was assessed using logistic regression by dividing the wGRS into 5 quintiles. The ability of the wGRS to discriminate between cases and controls was assessed by receiver operator curve (ROC) analysis in all patients and patients with anti-CCP positive disease. Results HLA-DRB1 imputation accuracy was assessed by checking concordance with ~1800 HLA-DRB1 typed samples: 2 digit (97%), 4 digit (91%). Using a model containing all 45 loci and imputed 4 digit HLA-DRB1 genotypes the odds of developing RA for those in the highest risk quintile compared to all others was 10.01 (95% CI 9.16-10.94) and this was further increased in anti-CCP positive individuals (OR 26.23 95% CI 22.88-30.08). The area under the ROC (AUC) showed that the ability of the model to discriminate correctly between individuals at risk and those not at risk was 0.72, improving in anti CCP positive individuals (AUC 0.78). Net reclassification tests showed that including 4 digit HLA-DRB1 alleles produced an overall reclassification improvement of 49% over a HLA-DRB1 tag SNP between models. The wGRS model had a Sensitivity of 44% and specificity of 90% in CCP positive individuals, which resulted in an accuracy of 76%. Conclusions Our study has shown the increased accuracy when using 4-digit HLA-DRB1 alleles in risk prediction models. We have shown that in RA the prediction performance remains modest when including all known genetic variants and it is too soon for these tests to be implemented in clinic or to have great utility in population screening. 1. Eyre S, Bowes J, Diogo D, Lee A, Barton A, Martin P, Zhernakova A, Stahl E, Viatte S, McAllister K et al.: High-density genetic mapping identifies new susceptibility loci for rheumatoid arthritis. Nat Genet 2012, 44:1336-1340. Acknowledgements This work was supported by the IMI JU funded project BeTheCure, contract no 115142-2. Disclosure of Interest None Declared [1]: #p-8