Variation in the recall of socially rewarding information and depressive symptom severity: a prospective cohort study.

OBJECTIVE: To test the association between recall for socially rewarding (positive) and/or socially critical (negative) information and depressive symptoms. METHOD: Cohort study of people who had visited UK primary care in the past year reporting depressive symptoms (N = 558, 69% female). Positive and negative recall was assessed at three time-points, 2 weeks apart, using a computerised task. Depressive symptoms were assessed at four time-points using the Beck Depression Inventory (BDI). Analyses were conducted using multilevel models. RESULTS: Concurrently we found evidence that, for every increase in two positive words recalled, depressive symptoms reduced by 0.6 (95% CI -1.0 to -0.2) BDI points. This association was not affected by adjustment for confounders. There was no evidence of an association between negative recall and depressive symptoms (-0.1, 95% CI -0.5 to 0.3). Longitudinally, we found more evidence that positive recall was associated with reduced depressive symptoms than vice versa. CONCLUSION: People with more severe depressive symptoms recall less positive information, even if their recall of negative information is unaltered. Clinicians could put more emphasis on encouraging patients to recall positive, socially rewarding information, rather than trying to change negative interpretations of events that have already occurred.

Minimal clinically important difference on the Beck Depression Inventory--II according to the patient's perspective.

BACKGROUND: The Beck Depression Inventory, 2nd edition (BDI-II) is widely used in research on depression. However, the minimal clinically important difference (MCID) is unknown. MCID can be estimated in several ways. Here we take a patient-centred approach, anchoring the change on the BDI-II to the patient's global report of improvement. METHOD: We used data collected (n = 1039) from three randomized controlled trials for the management of depression. Improvement on a 'global rating of change' question was compared with changes in BDI-II scores using general linear modelling to explore baseline dependency, assessing whether MCID is best measured in absolute terms (i.e. difference) or as percent reduction in scores from baseline (i.e. ratio), and receiver operator characteristics (ROC) to estimate MCID according to the optimal threshold above which individuals report feeling 'better'. RESULTS: Improvement in BDI-II scores associated with reporting feeling 'better' depended on initial depression severity, and statistical modelling indicated that MCID is best measured on a ratio scale as a percentage reduction of score. We estimated a MCID of a 17.5% reduction in scores from baseline from ROC analyses. The corresponding estimate for individuals with longer duration depression who had not responded to antidepressants was higher at 32%. CONCLUSIONS: MCID on the BDI-II is dependent on baseline severity, is best measured on a ratio scale, and the MCID for treatment-resistant depression is larger than that for more typical depression. This has important implications for clinical trials and practice.

Allowing for non-adherence to treatment in a randomized controlled trial of two antidepressants (citalopram versus reboxetine): an example from the GENPOD trial.

BACKGROUND: Meta-analyses suggest that reboxetine may be less effective than other antidepressants. Such comparisons may be biased by lower adherence to reboxetine and subsequent handling of missing outcome data. This study illustrates how to adjust for differential non-adherence and hence derive an unbiased estimate of the efficacy of reboxetine compared with citalopram in primary care patients with depression. METHOD: A structural mean modelling (SMM) approach was used to generate adherence-adjusted estimates of the efficacy of reboxetine compared with citalopram using GENetic and clinical Predictors Of treatment response in Depression (GENPOD) trial data. Intention-to-treat (ITT) analyses were performed to compare estimates of effectiveness with results from previous meta-analyses. RESULTS: At 6 weeks, 92% of those randomized to citalopram were still taking their medication, compared with 72% of those randomized to reboxetine. In ITT analysis, there was only weak evidence that those on reboxetine had a slightly worse outcome than those on citalopram [adjusted difference in mean Beck Depression Inventory (BDI) scores: 1.19, 95% confidence interval (CI) -0.52 to 2.90, p = 0.17]. There was no evidence of a difference in efficacy when differential non-adherence was accounted for using the SMM approach for mean BDI (-0.29, 95% CI -3.04 to 2.46, p = 0.84) or the other mental health outcomes. CONCLUSIONS: There was no evidence of a difference in the efficacy of reboxetine and citalopram when these drugs are taken and tolerated by depressed patients. The SMM approach can be implemented in standard statistical software to adjust for differential non-adherence and generate unbiased estimates of treatment efficacy for comparisons of two (or more) active interventions.

A pragmatic randomised controlled trial to evaluate the cost-effectiveness of a physical activity intervention as a treatment for depression: the treating depression with physical activity (TREAD) trial.

OBJECTIVE: The TREAting Depression with physical activity (TREAD) study investigated the cost-effectiveness of a physical activity intervention, in addition to usual general practitioner care, as a treatment for people with depression. DESIGN: An individually randomised, pragmatic, multicentre randomised controlled trial with follow-up at 4, 8 and 12 months. A subset of participants took part in a qualitative study that investigated the acceptability and perceived benefits of the intervention. SETTING: General practices in the Bristol and Exeter areas. PARTICIPANTS: Aged 18-69 years with an International Statistical Classification of Diseases and Related Health Problems, 10th Edition (ICD-10) diagnosis of depression and scoring ≥ 14 on the Beck Depression Inventory (BDI). Those who were unable to complete self-administered questionnaires in English, with medical contraindications to physical activity or with psychosis, bipolar disorder or serious drug abuse were excluded. INTERVENTIONS: We devised an intervention designed to encourage choice and autonomy in the adoption of physical activity. It consisted of up to three face-to-face and ten telephone contacts delivered by a trained physical activity facilitator over an 8-month period. MAIN OUTCOME MEASURES: The primary outcome was the BDI score measured at 4 months. Secondary outcomes included depressive symptoms over the 12 months and quality of life, antidepressant use and level of physical activity. RESULTS: The study recruited 361 patients, with 182 randomised to the intervention arm and 179 to the usual care arm; there was 80% retention at the 4-month follow-up. The intervention group had a slightly lower BDI score at 4 months [-0.54, 95% confidence interval (CI) -3.06 to 1.99] but there was no evidence that the intervention improved outcome for depression. Neither was there any evidence to suggest a difference in the prescription of or self-reported use of antidepressants. However, the amount of physical activity undertaken by those who had received the intervention was increased (odds ratio 2.3, 95% CI 1.3 to 3.9) and was sustained beyond the end of the intervention. From a health-care perspective, the intervention group was more costly than the usual care group, with the cost of the intervention £220 per person on average. It is therefore extremely unlikely that the intervention is cost-effective as a treatment for depression using current willingness-to-pay thresholds. CONCLUSIONS: This physical activity intervention is very unlikely to lead to any clinical benefit in terms of depressive symptoms or to be a cost-effective treatment for depression. Previous research has reported some benefit and there are three possible reasons for this discrepancy: first, even though the intervention increased self-reported physical activity, the increase in activity was not sufficiently large to lead to a measurable influence; second, only more vigorous activity might be of benefit; and third, previous studies had recruited individuals with a pre-existing commitment to physical activity. Future research is needed to identify and explain the mechanisms by which depression might be effectively treated, including, in particular, specific guidance on the optimum type, intensity and duration of physical activity required to produce a therapeutic effect. TRIAL REGISTRATION: Current Controlled Trials ISRCTN16900744. FUNDING: This project was funded by the NIHR Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 16, No. 10. See the HTA programme website for further project information.

Childhood diet and behavioural problems: results from the ALSPAC cohort.

OBJECTIVE: To investigate whether a 'junk food' diet at 81 months of age is associated with the development of behavioural problems over the following 16 months. SUBJECTS/METHODS: The study used data from the Avon Longitudinal Study of Parents and Children (ALSPAC) and 12,942 children were included. The main outcome measure was behavioural problems, measured using the Strengths and Difficulties Questionnaire (SDQ). SDQ scores were available at 81 and 97 months of age. Child-based dietary data were collected at 81 months by food frequency questionnaire; from this a 'junk food' score was derived, and mean weekly non-milk extrinsic sugar (NMES) intake estimated. Statistical analyses examined the associations between dietary exposures at 81 months and SDQ outcomes at 97 months. Children with SDQ scores suggesting behavioural problems at baseline were excluded in order to identify new cases. Adjustments were made for potential confounders such as socioeconomic status. RESULTS: Unadjusted analyses suggested associations between the 'junk food' score at 81 months and both total difficulties and pro-social behaviour at 97 months. However, adjustment for baseline SDQ scores attenuated these associations, with confidence intervals including the null for both total difficulties (OR (95% CI): 1.05 (0.92, 1.21); P=0.45) and pro-social behaviour (1.13 (1.00, 1.26); P=0.04). Adjustment for other potential confounders further attenuated the effects. Adjustment for confounders similarly attenuated modest associations between NMES intake and behavioural problems. CONCLUSIONS: There was no evidence to support an association between a 'junk food' diet at 81 months of age and behavioural problems after 16 months.

'Junk food' diet and childhood behavioural problems: results from the ALSPAC cohort.

BACKGROUND/OBJECTIVES: To determine whether a 'junk food' diet at age 4(1/2) is associated with behavioural problems at age 7. SUBJECTS/METHODS: Data on approximately 4000 children participating in the Avon Longitudinal Study of Parents and Children, a birth cohort recruited in Avon, UK in 1991/92 were used. Behavioural problems were measured at age 7 using the Strengths and Difficulties Questionnaire (SDQ; maternal completion). Total difficulties and scores for the five sub-scales (hyperactivity, conduct and peer problems, emotional symptoms and pro-social behaviour) were calculated. Principal components analysis of dietary data (frequency of consumption of 57 foods/drinks) collected at age 4(1/2) by maternal report was used to generate a 'junk food' factor. Data on confounders were available from questionnaires. RESULTS: A one standard deviation increase in 'junk food' intake at age 4(1/2) years was associated with increased hyperactivity at age 7 (odds ratio: 1.19; 95% confidence interval: 1.10, 1.29). This persisted after adjustment for confounders including intelligence quotient score (odds ratio: 1.13; 95% confidence interval: 1.01, 1.15). There was little evidence to support an association between 'junk food' intake and overall behavioural difficulties or other sub-scales of the SDQ. CONCLUSIONS: Children eating a diet high in 'junk food' in early childhood were more likely to be in the top 33% on the SDQ hyperactivity sub-scale at age 7. This may reflect a long-term nutritional imbalance, or differences in parenting style. This finding requires replication before it can provide an avenue for intervention.

Alcohol consumption as a risk factor for non-recovery from common mental disorder: results from the longitudinal follow-up of the National Psychiatric Morbidity Survey.

BACKGROUND: Alcohol is commonly considered to be associated with persistence of common mental disorder (CMD; anxiety/depression). However no community-based longitudinal studies have investigated the direction of causality. METHOD: We examined the association between alcohol consumption and recovery from CMD using data on 706 community-based subjects with CMD who were followed for 18 months. Alcohol consumption at baseline was defined as hazardous drinking [Alcohol Use Disorders Identification Test (AUDIT) 8], binge drinking (defined as six or more units of alcohol on one occasion, approximately two to three pints of commercially sold beer) and dependence. RESULTS: When compared with a non-binge-drinking group, non-recovery at follow-up was associated with binge drinking on at least a monthly basis at baseline, although the confidence interval (CI) included unity [adjusted odds ratio (OR) 1.47, 95% CI 0.89-2.45]. There was also weak evidence that alcohol dependence was associated with non-recovery (adjusted OR 1.37, 95% CI 0.67-2.81). There was little evidence to support hazardous drinking as a risk factor for non-recovery (adjusted OR 1.12, 95% CI 0.67-1.88). CONCLUSIONS: Binge drinking may be a potential risk factor for non-recovery from CMD, although the possibility of no effect cannot be excluded. Larger studies are required to refute or confirm this finding.

GPs' experiences of primary care mental health research: a qualitative study of the barriers to recruitment.

OBJECTIVE: To investigate the perceived barriers among GPs towards introducing participation in randomized controlled trials (RCTs) to patients presenting with depression during consultations. METHODS: Qualitative study using semi-structured interviews. Interviews were recorded using a digital voice recorder, transcribed verbatim and analysed using the Framework Approach. The participants were 41 GPs from five primary care trusts in the South West who were collaborating with the University of Bristol on an RCT recruiting patients with depression. RESULTS: Three themes were identified: (i) concern about protecting the vulnerable patient and the impact on the doctor-patient relationship; (ii) the perceived lack of skill and confidence of GPs to introduce a request for research participation within a potentially sensitive consultation; and (iii) the priority given to clinical and administrative issues over research participation. These themes were underpinned by GPs' observations that consultations with people about depression differed in content, style and perceived difficulty compared to other types of consultations. CONCLUSION: Depressed patients were often viewed as vulnerable and in need of protection and it was seen as difficult and intrusive to introduce research. Patients were not always given the choice to participate in research in the same way that they are encouraged to participate in treatment decision making. A lack of skills in introducing research could be addressed with training through the new Primary Care Research Network. A more radical change in clinician attitudes and policy may be needed in order to give research a higher priority within primary care.

Predictors of persistent neck pain after whiplash injury.

OBJECTIVE: To establish the aetiological influences of persistent neck pain following a motor vehicle collision and to construct a model for use in the emergency department for identifying patients at high risk of persistent symptoms. DESIGN: Prospective cohort study. Patients recruited from hospital emergency departments were sent a questionnaire to gather information on various exposures. They were followed up at 1, 3, and 12 months to identify those with persistent symptoms. MAIN OUTCOME MEASURE: Persistent neck pain (pain at 1, 3, and 12 months after collision). RESULTS: The baseline survey included 765 patients. Subsequently, 480 completed a questionnaire at each follow up time point, of whom 128 (27%) reported neck pain on each occasion. Few collision specific factors predicted persistent neck pain. In contrast, a high level of general psychological distress, pre-collision history of widespread body pain, type of vehicle, whiplash associated symptoms, and initial neck disability best predicted the persistence of symptoms. Furthermore, these factors, in combination, accounted for more than a fivefold increase in the risk of persistent neck pain. CONCLUSION: The greatest predictors of persistent neck pain following a motor vehicle collision relate to psychological distress and aspects of pre-collision health rather than to various attributes of the collision itself. With these factors, and those relating to initial injury severity, it is possible to identify a subgroup of patients presenting with neck pain with the highest risk of persistent symptoms. Thus, it is possible to identify whiplash patients with a poor prognosis and to provide closer follow up and specific attention to management in these individuals.

The NOAR Damaged Joint Count (NOAR-DJC): a clinical measure for assessing articular damage in patients with early inflammatory polyarthritis including rheumatoid arthritis.

OBJECTIVES: To evaluate the reliability and validity of the Norfolk Arthritis Register Damaged Joint Count (NOAR-DJC) in patients with early inflammatory polyarthritis (IP). METHODS: The NOAR-DJC examines deformity in 51 joints. Deformity is defined as inability to adopt the anatomical position, reduction in range of movement by at least one-third, and/or surgical alteration of the joint. Reliability was investigated by assessing intra- and inter-observer agreement in 40 and 32 patients, respectively. Validity was assessed by correlating the NOAR-DJC with the eroded joint count (criterion validity), the Health Assessment Questionnaire (HAQ) (convergent construct validity) and tender and swollen joint counts (divergent construct validity) and by discriminating between those who did and did not satisfy criteria for rheumatoid arthritis (discriminant validity). RESULTS: The intraclass correlation coefficient for the intra- and inter-rater studies were 0.88 [95% confidence interval (CI) 0.79, 0.94, P<0.00001] and 0.74 (95% CI 0.53, 0.86, P<0.00001), respectively. Correlations with eroded joint counts and HAQ scores after 5 yr follow-up were r(s) = 0.42 (95% CI 0.35, 0.49, P<0.01) and r(s) = 0.45 (95% CI 0.4, 0.5, P<0.01), respectively. Correlations with tender and swollen joint counts were weak (r(s) = 0.28 and r(s) = 0.33). CONCLUSION: The NOAR-DJC is a quick, reliable and valid tool for assessing articular damage in patients with early IP.

Influence of disease-modifying therapy on radiographic outcome in inflammatory polyarthritis at five years: results from a large observational inception study.

OBJECTIVE: To determine the effect of early treatment with disease-modifying antirheumatic drugs (DMARDs) in reducing radiographic progression over a 5-year period in patients with new-onset inflammatory polyarthritis. METHODS: Three hundred thirty-five consecutive patients with paired radiographs obtained 1 year and 5 years after enrollment in a population-based arthritis register were studied. Logistic regression was used to model differences in baseline factors associated with the start of DMARDs. The time from symptom onset to first use of DMARDs was stratified to represent 4 groups: no DMARD use, <6 months, 6-12 months, and >12 months. Radiographs of the hands and feet were scored using the Larsen method. Progression in the Larsen score was evaluated as a 5-year score adjusted for the first film score. Negative binomial regression was used to compare Larsen score progression for each of the 3 treatment groups with that for patients not receiving DMARDs. Results were then adjusted for severity, based on propensity modeling. RESULTS: Patients who received treatment had more radiographic progression than did patients who were untreated. Coefficients (95% confidence intervals), expressed as a multiple of the Larsen score in DMARD-treated patients compared with untreated patients, were as follows: 1.6 (1.1-2.3) for <6 months, 2.4 (1.5-3.6) for 6-12 months, and 2.0 (1.4-2.8) for >12 months. As expected, patients receiving treatment had more severe disease at baseline. Using the propensity score as a method of adjusting for disease severity, the influence of treatment on outcome became attenuated as follows: 1.1 (0.8-1.7) for <6 months, 1.6 (1.0-2.6) for 6-12 months, and 1.5 (1.0-2.2) for >12 months. This effect was also seen in the crude Larsen score at year 5. CONCLUSION: In this observational study, DMARD treatment was a marker not only of worse disease at presentation but also of the radiographic state and radiographic progression at 5 years. After adjustments were made for baseline disease severity, earlier therapy was shown to have a beneficial effect on outcome.

Benchmarking: the five year outcome of rheumatoid arthritis assessed using a pain score, the Health Assessment Questionnaire, and the Short Form-36 (SF-36) in a community and a clinic based sample.

BACKGROUND: Treatment, and therefore outcome, of rheumatoid arthritis (RA) will improve in the next few years. However, improvement in outcome can only be judged against the probability of certain outcomes with current conventional treatment. AIM: To document the five year outcome of RA in the late 1990s. SETTING: Norfolk Arthritis Register (NOAR). DESIGN: Longitudinal observational cohort study. METHODS: 318 patients with recent onset inflammatory polyarthritis recruited by NOAR in 1990-91 completed five years of follow up. Four groups were assessed: the whole cohort, all those referred to hospital, those who satisfied criteria for RA at baseline, and those referred to hospital who satisfied criteria for RA at baseline. Outcome was assessed with a visual analogue scale for pain, the Health Assessment Questionnaire (HAQ), and the Short Form-36 (SF-36). RESULTS: Of the RA hospital attenders, 50% had a visual analogue scale pain score of 5 cm or less and an HAQ score of 1.125 or less. SF-36 scores were reduced in all domains. Results are presented as cumulative percentages. CONCLUSIONS: These results can be used for comparison and to set targets for improvement.

Reduced disability at five years with early treatment of inflammatory polyarthritis: results from a large observational cohort, using propensity models to adjust for disease severity.

OBJECTIVE: To determine the effect of treatment with disease-modifying antirheumatic drugs (DMARDs) and/or steroids on 5-year disability outcome in patients with inflammatory polyarthritis. METHODS: Three hundred eighty-four patients registered by the Norfolk Arthritis Register (a primary care-based observational cohort) were followed up for 5 years. Treatment details and Health Assessment Questionnaire (HAQ) scores were recorded annually. Logistic regression was used to model differences in baseline factors associated with the start of DMARDs and/or steroids within 12 months of baseline. Based on this model, each subject was given a probability of starting treatment ("propensity score"). A second model compared the odds of disability (HAQ score > or =1.00) in treated and untreated patients, adjusting for differences in disease severity using the propensity score. RESULTS: Unadjusted analysis suggested that patients who received treatment had an increased odds of a worse outcome compared with those who did not receive treatment. When adjusted for differences in disease severity, using the propensity score, early treatment (within 6 months of symptom onset) was associated with a similar odds of disability at 5 years compared with those not treated (odds ratio 0.71; 95% confidence interval 0.34, 1.44). In contrast, starting treatment later (> or =6 months) was associated with a 2-fold increased odds of having a HAQ score > or =1.00 at 5 years. CONCLUSION: The propensity score was a useful method of adjusting for "confounding by indication" in observational studies. Furthermore, this study showed that early treatment with DMARDs/steroids (within 6 months of symptom onset) reduced the odds of disability 5 years later to a level comparable with that of patients judged clinically as not requiring treatment.

The association of cigarette smoking with disease outcome in patients with early inflammatory polyarthritis.

OBJECTIVE: Cigarette smoking is known to increase rheumatoid factor (RF) and nodule formation in patients with rheumatoid arthritis (RA). In this study, we examined the influence of smoking on disease outcome at 3 years among patients newly presenting with inflammatory polyarthritis (IP). METHODS: We studied 486 patients with IP who were referred to the Norfolk Arthritis Register, of whom 323 (67%) satisfied the American College of Rheumatology 1987 criteria for RA. Smoking status was assessed at baseline. Disease outcome was assessed at 3 years, using measures of joint inflammation, functional disability, and radiologic damage. The influence of smoking on disease outcome was explored using logistic regression techniques, with patients who had never smoked as the referent group. Results are expressed as odds ratios (ORs), with their 95% confidence intervals (95% CIs). RESULTS: Current smokers were significantly more likely to be RF positive at baseline (47%) than were ex-smokers (34%) and never smokers (31%). After 3 years, rheumatoid nodules were significantly more common in smokers (13%) compared with ex-smokers/never smokers (4%), a relationship which persisted after adjusting for age and sex (OR 4.07, 95% CI 1.38-12). In contrast, after adjusting for age and sex, current smokers had significantly fewer swollen joints (OR 0.61, 95% CI 0.37-0.98). However, smoking status had no influence on the development of erosions or functional disability. CONCLUSION: Despite smokers being more likely to develop nodules and to be RF positive, current smokers did not have higher levels of radiologic damage, and had fewer swollen joints. We hypothesize that this could be due to either the effect of cigarette smoking on the inflammatory response or other factors (e.g., reduced physical activity in smokers) which may limit joint inflammation and damage.

One year followup variables predict disability 5 years after presentation with inflammatory polyarthritis with greater accuracy than at baseline.

OBJECTIVE: To compare the accuracy of simple demographic and clinical variables recorded at baseline with those recorded after one year followup, in predicting self-reported functional disability recorded 5 years after initial assessment in patients with early inflammatory polyarthritis (IP). METHODS: We followed annually for 5 years 528 patients registered by the Norfolk Arthritis Register (a primary care based cohort of patients with early IP) using the Health Assessment Questionnaire (HAQ). Backward stepwise logistic regression was used to determine the clinical and demographic variables, collected at either baseline or first followup, that were associated with disability (HAQ > or = 1.00/> or =1.50) at 5 years. RESULTS: At the 5th anniversary assessment, the prevalence of moderate disability (HAQ > or = 1.00) was 47%. Twenty-nine percent reported more severe disability (HAQ > or = 1.50). Variables recorded at first anniversary assessment were better able to predict patients at risk of developing a poor outcome than baseline variables. Multivariate methods identified age at symptom onset, HAQ score, presence of nodules, and a statistically derived factor describing joint tenderness recorded at first year as important predictors of both moderate disability (HAQ > or =1.00) and a higher level of disability (HAQ > or = 1.50). When tested in an independent validation sample, the accuracy of the models generated from data recorded at the first year was 76% (HAQ > or = 1.00) and 83% (HAQ > or = 1.50). CONCLUSION: It was possible to predict disability at 5 years with high accuracy using simple clinical variables and demographic data collected 4 or 5 years previously. First year HAQ score was the strongest predictor of future disability. HAQ score at 5 years could be predicted more accurately using data collected at first anniversary visit than using data recorded at baseline.

The impact of rheumatoid arthritis on employment status in the early years of disease: a UK community-based study.

OBJECTIVE: To establish the prevalence of work disability and predictors of change in employment status in patients with early rheumatoid arthritis (RA). SETTING: The Norfolk Arthritis Register (NOAR), a primary-care based inception cohort of patients with recent-onset inflammatory arthritis. METHODS: Two cohorts of patients notified to NOAR, who satisfied the 1987 ACR criteria for RA at the time of notification (baseline) and who were economically active at the time of RA symptom onset, were identified. Cohort 1 consisted of 160 patients with an onset of RA between 1989 and 1992, and was followed for a mean of 8.6 yr from symptom onset. For 110 of these cases, a control group, matched for age, gender and employment status at baseline, was identified from the local population. Their employment histories were compared in 1995. Cohort 2 consisted of 134 patients with an onset of RA between 1994 and 1997, and was followed for a mean of 4.1 yr from symptom onset. RESULTS: One-third of RA cohort 1 had stopped working on the grounds of ill health by 1995. The baseline health assessment questionnaire (HAQ) score was the most important predictor of work disability. These patients were 32 times more likely to stop work on health grounds than the matched controls. The rates for work disability for the RA cases 1, 2, 5 and 10 yr after symptom onset were 14, 26, 33 and 39% respectively. For cohort 2, the rates for work disability 1 and 2 yr from onset were 23 and 33% respectively. CONCLUSION: Work disability is an important outcome in RA patients of working age. Many people stop working very early in the disease process, often before they are referred to hospital or started on disease-modifying anti-rheumatic drugs. Although the peak rates for work disability are in the early years, people with RA continue to leave the work force several years after onset. Thus, the recent move to earlier, more aggressive treatment has had no effect on the rates of work disability.