James M. Gwinnutt

Twenty-year outcome and association between early treatment and mortality and disability in an inception cohort of patients with rheumatoid arthritis: Results from the Norfolk Arthritis Register

To describe the outcome in patients with rheumatoid arthritis (RA) over 20 years from symptom onset, and to assess the association between early treatment (with disease‐modifying antirheumatic drugs/steroids) and mortality and disability during follow‐up.

Predictors of and outcomes following orthopaedic joint surgery in patients with early rheumatoid arthritis followed for 20 years

Objectives To analyse predictors and outcomes of major orthopaedic surgery in a cohort of RA patients followed for 20 years. Methods Patients were recruited to the Norfolk Arthritis Register from 1990 to 1994. Demographic and clinical variables (including the HAQ and swollen and tender joint counts) were assessed at baseline; the 2010 ACR/EULAR RA classification criteria were applied. Patients reported incident comorbidities and major orthopaedic joint surgery (replacement, synovectomy, fusion, excision) when reassessed at years 1, 2, 3, 5, 7, 10, 15 and 20. Baseline and time-varying predictors of orthopaedic surgery were assessed using a conditional risk set model, a type of multiple-failure survival analysis. Change in disability after surgery was assessed using weighted mixed-effects linear regression. Results Of 589 RA patients [median age 56 years (IQR 45-68); 66.7% women] recruited to the Norfolk Arthritis Register with at least one follow-up, 102 reported a total of 180 major surgeries, with hip replacement being the most common (n = 68/180). Patients reporting major surgery had worse functional disability at all time points, but similar swollen/tender joint counts to those without major surgery. Each unit increase in HAQ score was associated with a doubling of the patients risk of having surgery by the next assessment [hazard ratio 2.11 per unit increase in HAQ (95% CI 1.64, 2.71)]. Patients had worse HAQ scores after surgery than patients not undergoing surgery [β = 0.17 (95% CI 0.03, 0.32)]. Conclusion HAQ was the strongest predictor of future major surgery. This supports the argument that HAQ should be included in routine clinical assessment.

Have the 10-year outcomes of patients with early inflammatory arthritis improved in the new millennium compared with the decade before? Results from the Norfolk Arthritis Register

Objective To compare the 10-year outcome (disease activity, disability, mortality) of two cohorts of patients with inflammatory polyarthritis (IP) recruited 10 years apart. Methods Patients with IP were recruited to the Norfolk Arthritis Register from 1990 to 1994 (cohort 1 (C1)) and from 2000 to 2004 (cohort 2 (C2)). Demographic and clinical data were collected at baseline and at years 1, 2, 3, 5, 7 and 10. Longitudinal disease activity (swollen/tender 51 joint counts (SJC51/TJC51)) and disability (Health Assessment Questionnaire (HAQ)) were compared between the cohorts using population-average negative binomial regression and generalised estimating equation analysis, respectively. Risk of 10-year mortality was compared between cohorts using Cox models. Risk of cardiovascular disease (CVD) mortality was compared between cohorts using competing risks analysis. Mortality rate ratios (MRR), adjusted for changes in mortality risk of the general population, were calculated using Poisson regression. Results In total 1653 patients were recruited (C1=1022, C2=631). Patients in C2 had 17% lower SJC51 than C1 over 10 years (95% CI −23% to −10%), whereas TJC51 and HAQ were comparable. C2 patients had reduced risk of all-cause and CVD mortality compared with C1 (all-cause: HR 0.72, 95% CI 0.56 to 0.95; CVD: subhazard ratio 0.58, 95% CI 0.37 to 0.93). After accounting for changes in mortality risk in the general population, the difference in mortality was non-significant (all-cause: MRR 0.78, 95% CI 0.56 to 1.10; CVD: MRR 0.77, 95% CI 0.48 to 1.24). Conclusion Disease activity significantly improved in the new millennium, whereas disability and mortality were unchanged.

Baseline patient reported outcomes are more consistent predictors of long-term functional disability than laboratory, imaging or joint count data in patients with early inflammatory arthritis: A systematic review

Objective To assess baseline predictors of long-term functional disability in patients with inflammatory arthritis (IA). Methods We conducted a systematic review of the literature from 1990 to 2017 using MEDLINE and EMBASE. Studies were included if (i) they were prospective observational studies, (ii) all patients had IA with symptom duration ≤2 years at baseline, (iii) follow-up was at least 5 years, and (iv) baseline predictors of HAQ score at long-term follow-up (i.e., ≥5 years following baseline) were assessed. Information on the included studies and estimates of the association between baseline variables and long-term HAQ scores were extracted from the full manuscripts. Results Of 1037 abstracts identified by the search strategy, 37 met the inclusion/exclusion criteria and were included in the review. Older age at baseline and female gender were reported to be associated with higher long-term HAQ scores in the majority of studies assessing these relationships, as were higher baseline HAQ and greater pain scores (total patients included in analyses reporting significant associations/total number of patients analysed: age 9.8k/10.7k (91.6%); gender 9.9k/11.3k (87.4%); HAQ 4.0k/4.0k (99.0%); pain 2.8k/2.9k (93.6%)). Tender joint count, erythrocyte sedimentation rate (ESR) and DAS28 were also reported to predict long-term HAQ score; other disease activity measures were less consistent (tender joints 2.1k/2.5k (84.5%); erythrocyte sedimentation rate 1.6k/2.2k (72.3%); DAS28 888/1.1k (79.2%); swollen joints 684/2.6k (26.6%); C-reactive protein 279/510 (54.7%)). Rheumatoid factor (RF) and erosions were not useful predictors (RF 546/4.6k (11.9%); erosions 191/2.7k (7.0%)), whereas the results for anti-citrullinated protein antibody positivity were equivocal (ACPA 2.0k/3.8k (52.9%)). Conclusions Baseline age, gender, HAQ and pain scores are associated with long-term disability and knowledge of these may aid the assessment of prognosis.

SAT0110 Trajectories of functional disability in patients with early inflammatory polyarthritis and moderate disease activity: results from the early rheumatoid arthritis network and norfolk arthritis register

Background A large group of patients with inflammatory polyarthritis (IP), and its subset rheumatoid arthritis (RA), have moderate disease activity, despite disease modifying therapy. Identifying patients with moderate disease who are likely to have subsequent high disability may prompt different treatment strategies for these patients. Objectives To identify common trajectories of disability progression in patients with moderate disease in two large prospective observational studies. Methods The Early Rheumatoid Arthritis Network (ERAN) recruited 1236 patients with RA (<36 months symptoms) from 23 centres in England from 2002-11. The Norfolk Arthritis Register (NOAR) recruited 1054 IP patients (<24 months symptoms) from Norfolk, England, from 2000-8. At baseline and subsequent follow-ups, functional disability was assessed using the Health Assessment Questionnaire (HAQ). Included patients scored ≥3.2 and <5.1 on the Disease Activity Score (DAS28) at either baseline, year 1 or year 2, and had previously received csDMARDS (NERAN=605; NNOAR=407). Latent class growth models (LCGMs) were used to identify HAQ trajectories independently in each cohort. Age, sex, fulfilment of ACR RA criteria, symptom duration, DMARDs at baseline, and baseline DAS28 were included as predictors of trajectory group membership using multinomial logistic regression. Results Baseline characteristics of the cohorts were similar (ERAN vs NOAR: mean age = 57 vs 56 years; female = 70% vs 69%; met ACR criteria = 88% vs 77%; mean DAS28 = 4.7 vs 4.2). For both cohorts, LCGM analysis indicated 4 subgroups provided best fit (Bayesian Information Criterion), with similar shaped trajectories (figure 1). Multinomial logistic regression indicated that older age (ERAN & NOAR, p<0.005), female gender (ERAN & NOAR, p<0.01), meeting ACR criteria (NOAR only, p<0.05), use of DMARD (ERAN & NOAR, p<0.01), and baseline DAS28 (ERAN & NOAR, p<0.005) were related to an increased likelihood of being in a subgroup with higher disability (vs. lowest disability subgroup). Conclusions Four disability trajectories were observed in both the ERAN and NOAR cohort of patients with moderate disease activity. Patients on a worse trajectory who may benefit from more intensive treatment could potentially be identified earlier in the disease the group of patients with moderate disease activity. Reference: [1] This research was funded by Arthritis Research UK. Disclosure of Interest: None declared

Response to: Immortal Time Bias: Comment on the Article by Gwinnutt et al

In a study of patients with rheumatoid arthritis, Gwinnutt et al reported a trend toward reduced mortality risk in the early treatment group and a significant reduction in mortality in the late treatment group compared to the never treatment group (1). In attempting to address an important question in rheumatology care regarding the “window of opportunity,” whereby early treatment initiation may lead to improved outcomes, the analyses used in this study unfortunately fell prey to immortal time bias (2). Immortal time bias refers to a time period during which patients could not have died, by study design. The bias was created by looking through the entire 20-year follow-up period to determine whether patients belonged in the early, late, or never treatment groups, and then treating these groupings as baseline variables. This creates a period of time between study inclusion and treatment initiation in both the early and late treatment groups during which patients cannot die and are therefore “immortal.” An instructive example of this fallacy was a study of the effect on mortality of the amount of chemotherapy received (3). Those who lived the longest received the most chemotherapy, but the analysis wrongly concluded that those with longer treatment had a longer life. Similarly, in the study by Gwinnutt et al, those with the longest follow-up had more opportunity for intervention, so a longer period of observation caused the treatment status and not the other way around. While this bias was recognized in the 1970s in epidemiologic cohort studies of the survival benefit of heart transplantation, it is still not familiar to many investigators and may have recently become more prevalent due to the increased availability of long-term computerized databases of observational data (4,5). This bias can be avoided by counting the time from study inclusion to treatment initiation in the unexposed (i.e., not yet treated) group. When this immortal time is erroneously attributed to the treated group, it lowers the event rate among those who are treated and also falsely decreases the denominator in the unexposed group, leading to increased event rates. Together these biases lead to a conclusion that the treatment is efficacious, as was found in this study. While the authors are to be commended for their careful attention to the bias of confounding by indication (i.e., sicker patients are more likely to receive treatment) and survivorship bias (i.e., dropout for repeated measures analyses is not random), immortal time bias is equally important and should be addressed to understand how it might have affected the conclusions.

THU0112 Stress and depression in patients with early inflammatory polyarthritis: natural history and associations with disease activity, disability and pain over five years

Background Stress and depression are common in patients with inflammatory polyarthritis (IP). There is little research on long term patterns of depression and stress or how these variables relate to disease activity, disability and pain. Objectives To describe the natural history of stress and depression over five years and to assess the association of baseline, one year prior and current disease activity, disability and pain with longitudinal stress and depression. Methods Patients recruited to the Norfolk Arthritis Register (NOAR) (inclusion criteria: ≥2 swollen joints for ≥4 weeks) from 2005–2008 were included in this analysis. Demographics, medication use, 51 swollen/tender joint counts (SJC51/TJC51), pain visual analogue scale, HAQ, comorbidities and the Arthritis Impact Measurement Scales 2 (AIMS2) depression and stress subscales (range 0–10; high score = worse health status) were collected at baseline and years 1, 2, 3 and 5. Rheumatoid factor (RF), anti-cyclic citrullinated peptide antibodies (anti-CCP2) and C-reactive protein (CRP) were measured in baseline blood samples. ACR/EULAR 2010 RA criteria were applied to baseline characteristics. Depression and stress over five years were described using descriptive statistics. Multivariate random effects models were applied to assess the association of baseline disease activity (SJC51/TJC51), pain and disability with depression and stress over time adjusting for baseline age, gender, RF, anti-CCP2, CRP, sDMARD use, comorbidities, depression and stress. Similar methods were used to assess one year prior and current disease activity, pain and disabilitys association with stress and depression. Missing data were imputed using multiple imputation. Results 509 patients were included (median (IQR) age: 57 (45, 68) years; 321 (63.1%) female; 305 (59.9%) ACR/EULAR RA). Baseline median (IQR) depression and stress were 2.5 (1.5, 4.5) and 4.0 (2.5, 5.5) respectively and remained constant over five years. Baseline SJC51, TJC51, pain and HAQ were not independently associated with depression or stress over five years. Current HAQ and pain, but not SJC51/TJC51, were independently associated with depression and stress (per unit increase in HAQ: depression β 0.83, 95% CI 0.69, 0.97; stress β 0.76, 95% CI 0.61, 0.90; per 1cm increase in pain: depression β 0.09, 95% CI 0.06, 0.12; stress β 0.09, 95% CI 0.05, 0.12). Higher HAQ was independently associated with increased depression and stress one assessment later (per unit increase in HAQ: depression β 0.21, 95% CI 0.09, 0.32; stress β 0.21, 95% CI 0.10, 0.33) but not pain, SJC51 or TJC51. Conclusions There were no associations between measures of disease activity and depression or stress. Prospectively higher HAQ scores were associated with worse psychological health a year later. This may have implications for holistic management of IP. Disclosure of Interest None declared