Random effects models for complex designs.

Plaid designs are characterised by having one set of treatments applied to rows and another set of treatments applied to columns. In a 2003 publication, Farewell and Herzberg presented an analysis of variance structure for such designs. They presented an example of a study in which medical practitioners, trained in different ways, evaluated a series of videos of patients obtained under a variety of conditions. However, their analysis did not take full account of all error terms. In this paper, a more comprehensive analysis of this study is presented, informed by the recognition that the study can also be regarded as a two-phase design. The development of random effects models is outlined and the potential importance of block-treatment interactions is highlighted. The use of a variety of techniques is shown to lead to a better understanding of the study. Examination of the variance components involved in the expected mean squares is demonstrated to have particular value in identifying appropriate error terms for F-tests derived from an analysis of variance table. A package such as ASReml can also be used provided an appropriate error structure is specified. The methods presented can be applied to the design and analysis of other complex studies in which participants supply multiple measurements under a variety of conditions.

Pulmonary Metastasectomy in Colorectal Cancer: updated analysis of 93 randomized patients - control survival is much better than previously assumed.

AIM: Lung metastases from colorectal cancer are resected in selected patients in the belief that this confers a significant survival advantage. It is generally assumed that the 5-year survival of these patients would be near zero without metastasectomy. We tested the clinical effectiveness of this practice in Pulmonary Metastasectomy in Colorectal Cancer (PulMiCC), a randomized, controlled noninferiority trial. METHOD: Multidisciplinary teams in 14 hospitals recruited patients with resectable lung metastases into a two-arm trial. Randomization was remote and stratified according to site, with minimization for age, sex, primary cancer stage, interval since primary resection, prior liver involvement, number of metastases and carcinoembryonic antigen level. The trial management group was blind to patient allocation until after intention-to-treat analysis. RESULTS: From 2010 to 2016, 93 participants were randomized. These patients were 35-86 years of age and had between one and six lung metastases at a median of 2.7 years after colorectal cancer resection; 29% had prior liver metastasectomy. The patient groups were well matched and the characteristics of these groups were similar to those of observational studies. The median survival after metastasectomy was 3.5 (95% CI: 3.1-6.6) years compared with 3.8 (95% CI: 3.1-4.6) years for controls. The estimated unadjusted hazard ratio for death within 5 years, comparing the metastasectomy group with the control group, was 0.93 (95% CI: 0.56-1.56). Use of chemotherapy or local ablation was infrequent and similar in each group. CONCLUSION: Patients in the control group (who did not undergo lung metastasectomy) have better survival than is assumed. Survival in the metastasectomy group is comparable with the many single-arm follow-up studies. The groups were well matched with features similar to those reported in case series.

Do drugs offering only PFS maintain quality of life sufficiently from a patient's perspective? Results from AVALPROFS (Assessing the 'VALue' to patients of PROgression Free Survival) study.

PURPOSE: Trials of novel drugs used in advanced disease often show only progression-free survival or modest overall survival benefits. Hypothetical studies suggest that stabilisation of metastatic disease and/or symptom burden are worth treatment-related side effects. We examined this premise contemporaneously using qualitative and quantitative methods. METHODS: Patients with metastatic cancers expected to live > 6 months and prescribed drugs aimed at cancer control were interviewed: at baseline, at 6 weeks, at progression, and if treatment was stopped for toxicity. They also completed Functional Assessment of Cancer Therapy (FACT-G) plus Anti-Angiogenesis (AA) subscale questionnaires at baseline then monthly for 6 months. RESULTS: Ninety out of 120 (75%) eligible patients participated: 41 (45%) remained on study for 6 months, 36 progressed or died, 4 had treatment breaks, and 9 withdrew due to toxicity. By 6 weeks, 66/69 (96%) patients were experiencing side effects which impacted their activities. Low QoL scores at baseline did not predict a higher risk of death or dropout. At 6-week interviews, as the side effect severity increased, patients were significantly less inclined to view the benefit of cancer control as worthwhile (X2 = 50.7, P < 0.001). Emotional well-being initially improved from baseline by 10 weeks, then gradually returned to baseline levels. CONCLUSION: Maintaining QoL is vital to most patients with advanced cancer so minimising treatment-related side effects is essential. As side effect severity increased, drugs that controlled cancer for short periods were not viewed as worthwhile. Patients need to have the therapeutic aims of further anti-cancer treatment explained honestly and sensitively.

Two-Part and Related Regression Models for Longitudinal Data.

Statistical models that involve a two-part mixture distribution are applicable in a variety of situations. Frequently, the two parts are a model for the binary response variable and a model for the outcome variable that is conditioned on the binary response. Two common examples are zero-inflated or hurdle models for count data and two-part models for semicontinuous data. Recently, there has been particular interest in the use of these models for the analysis of repeated measures of an outcome variable over time. The aim of this review is to consider motivations for the use of such models in this context and to highlight the central issues that arise with their use. We examine two-part models for semicontinuous and zero-heavy count data, and we also consider models for count data with a two-part random effects distribution.

Meta-analysis of colorectal cancer follow-up after potentially curative resection.

BACKGROUND: After potentially curative resection of primary colorectal cancer, patients may be monitored by measurement of carcinoembryonic antigen and/or CT to detect asymptomatic metastatic disease earlier. METHODS: A systematic review and meta-analysis was conducted to find evidence for the clinical effectiveness of monitoring in advancing the diagnosis of recurrence and its effect on survival. MEDLINE (Ovid), Embase, the Cochrane Library, Web of Science and other databases were searched for randomized comparisons of increased intensity monitoring compared with a contemporary standard policy after resection of primary colorectal cancer. RESULTS: There were 16 randomized comparisons, 11 with published survival data. More intensive monitoring advanced the diagnosis of recurrence by a median of 10 (i.q.r. 5-24) months. In ten of 11 studies the authors reported no demonstrable difference in overall survival. Seven RCTs, published from 1995 to 2016, randomly assigned 3325 patients to a monitoring protocol made more intensive by introducing new methods or increasing the frequency of existing follow-up protocols versus less invasive monitoring. No detectable difference in overall survival was associated with more intensive monitoring protocols (hazard ratio 0·98, 95 per cent c.i. 0·87 to 1·11). CONCLUSION: Based on pooled data from randomized trials published from 1995 to 2016, the anticipated survival benefit from surgical treatment resulting from earlier detection of metastases has not been achieved.

Psychological morbidity associated with ovarian cancer screening: results from more than 23,000 women in the randomised trial of ovarian cancer screening (UKCTOCS).

OBJECTIVE: To examine the psychological sequelae associated with abnormal screening in the United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). DESIGN: Prospective, longitudinal randomised control trial. SETTING: Sixteen UKCTOCS centres. SAMPLE: Women aged 50-70 years randomised to annual multimodal screening, ultrasound screening or control groups. METHODS: Two groups were followed for 7 years: (1) a random sample (n = 1339), taken from all three study groups; and (2) an events sample (n = 22,035) of women with abnormal screens resulting in the need for repeat testing of either low or higher level intensity. MAIN OUTCOME MEASURES: Patient-reported measures of anxiety (scores ranging from 20 to 80) and psychological morbidity. RESULTS: In the random sample the mean difference between anxiety scores after a repeat screening and those following an annual screening was 0.4 (95% CI -0.46, 1.27), and in the events sample it was 0.37 (95% CI 0.23, 0.51). The risk of psychological morbidity was only increased in the event sample for women requiring higher level repeat screening (OR 1.28; 95% CI 1.18, 1.39). The risk of psychological morbidity in women with ovarian cancer was higher at both 6 weeks (OR 16.2; 95% CI 9.19, 28.54) and 6 months (OR 3.32; 95% CI 1.91, 5.77) following surgery. CONCLUSIONS: Screening does not appear to raise anxiety but psychological morbidity is elevated by more intense repeat testing following abnormal annual screens, and in women after surgical treatment for ovarian cancer.

The functional MICA-129 polymorphism is associated with skin but not joint manifestations of psoriatic disease independently of HLA-B and HLA-C.

A methionine/valine polymorphism at amino acid 129 of the major histocompatibility complex class I chain-related gene A (MICA-129) categorizes alleles into strong and weak binders of the natural killer (NK) and T-cell receptor NKG2D. We investigated whether MICA-129 is differentially associated with skin and joint manifestations of psoriatic disease (PsD) independently of human leukocyte antigen (HLA)-C and HLA-B in patients and controls from Toronto and St. John's. The MICA-129 methionine (Met) allele, particularly Met/Met homozygosity, was strongly associated with both cutaneous psoriasis (PsC) and psoriatic arthritis (PsA) independently of HLA-B and HLA-C in Toronto patients, and was also associated with PsA in St. John's patients, but with no additional effect of Met/Met homozygosity. No association remained after adjustment for HLA alleles in St. John's patients. MICA-129 was not associated with PsA when compared with PsC. We conclude that MICA-129 is a marker of skin manifestations of PsD that is independent of HLA class I in Toronto patients.

Drivers and barriers to patient participation in RCTs.

BACKGROUND: Recruitment of patients into randomised clinical trials (RCTs) is essential for treatment evaluation. Appreciation of the barriers and drivers towards participation is important for trial design, communication and information provision. METHOD: As part of an intervention to facilitate effective multidisciplinary team communication about RCTs, cancer patients completed two study-specific questionnaires following trial discussions. One questionnaire examined reasons why patients accepted or declined trial entry, the other perceptions about their health-care professionals' (HCPs) information giving. RESULTS: Questionnaires were completed by 74% (358/486) of patients approached; of these 81% (291/358) had joined an RCT, 16% (56/358) had declined and 3% (11/358) were undecided. Trial participation status of the 128 patients not returning questionnaires is unknown. Trial acceptance was not dependent on disease stage, tumour type, sex or age. Satisfaction with trial information and HCPs' communication was generally very good, irrespective of participation decisions. The primary reason given for trial acceptance was altruism (40%; 110/275), and for declining, trust in the doctor (28%; 12/43). Decliners preferred doctors to choose their treatment rather than be randomised (54% vs 39%; P<0.027). Acceptors were more likely to perceive doctors as wanting them to join trials (54% vs 30%; P<0.001). Trial type, that is, standard treatment vs novel or different durations of treatment, also influenced acceptance rates. CONCLUSION: The drivers and barriers to trial participation are partly related to trial design. Unease about randomisation and impact of duration on treatment efficacy are barriers for some. Altruism and HCPs' perceived attitudes are powerful influencing factors.

Teams Talking Trials: results of an RCT to improve the communication of cancer teams about treatment trials.

BACKGROUND: Previous research has shown that communication between members of multidisciplinary teams (MDTs) is often suboptimal and communication about trials between MDTs and their patients is difficult. Educational interventions can help dyadic exchanges with different aspects of trial recruitment but less work has focussed on team interventions. METHODS: 22 multidisciplinary cancer teams in the UK participated in an RCT of a novel Teams Talking Trials (TTT) Workshop aimed at improving the following: awareness, involvement, communication and recruitment to cancer trials. MDTs were randomised following either 6 or 12 months of audits, which were repeated after the intervention. Audits included numbers approached about trials, team members' attitudes, involvement and awareness of their teams' trial portfolios. RESULTS: There was no significant difference in the rate of approaching patients about trials post workshop (estimated improvement 22% higher regression coefficient of 0.2, exp. (0.2)=1.22). There was improvement in team members' involvement in trials in 4 areas (p≤0.04): the pressure to enter patients into RCTs, the likelihood of a start-up meeting to discuss a newly accepted trial, the informational role played by individuals and recognition of this HCP's role by other team members. Also, confidence in communication about RCTS increased and awareness of different aspects of trial management improved on all 14 aspects (p=0.001). CONCLUSION: Attendance by teams at focussed workshops designed to enhance communication and trial recruitment improved several aspects of team functioning, but a significant impact on the number of patients approached could not be demonstrated.

Autoantibodies as biomarkers for the prediction of neuropsychiatric events in systemic lupus erythematosus.

OBJECTIVE: Neuropsychiatric events occur unpredictably in systemic lupus erythematosus (SLE) and most biomarker associations remain to be prospectively validated. This study examined a disease inception cohort of 1047 SLE patients to determine which autoantibodies at enrolment predicted subsequent neuropsychiatric events. METHODS: Patients with a recent SLE diagnosis were assessed prospectively for up to 10 years for neuropsychiatric events using the American College of Rheumatology case definitions. Decision rules of graded stringency determined whether neuropsychiatric events were attributable to SLE. Associations between the first neuropsychiatric event and baseline autoantibodies (lupus anticoagulant (LA), anticardiolipin, anti-ß(2) glycoprotein-I, anti-ribosomal P and anti-NR2 glutamate receptor) were tested by Cox proportional hazards regression. RESULTS: Disease duration at enrolment was 5.4 ± 4.2 months, follow-up was 3.6 ± 2.6 years. Patients were 89.1% female with mean (±SD) age 35.2 ± 13.7 years. 495/1047 (47.3%) developed one or more neuropsychiatric event (total 917 events). Neuropsychiatric events attributed to SLE were 15.4% (model A) and 28.2% (model B). At enrolment 21.9% of patients had LA, 13.4% anticardiolipin, 15.1% anti-ß(2) glycoprotein-I, 9.2% anti-ribosomal P and 13.7% anti-NR2 antibodies. LA at baseline was associated with subsequent intracranial thrombosis (total n=22) attributed to SLE (model B) (HR 2.54, 95% CI 1.08 to 5.94). Anti-ribosomal P antibody was associated with subsequent psychosis (total n=14) attributed to SLE (model B) (HR 3.92, 95% CI 1.23 to 12.5, p=0.02). Other autoantibodies did not predict neuropsychiatric events. CONCLUSION: In a prospective study of 1047 recently diagnosed SLE patients, LA and anti-ribosomal P antibodies are associated with an increased future risk of intracranial thrombosis and lupus psychosis, respectively.

Differential major histocompatibility complex class I chain-related A allele associations with skin and joint manifestations of psoriatic disease.

About 30% of patients with psoriasis have psoriatic arthritis (PsA), an inflammatory arthritis that can affect both axial and peripheral joints. Major histocompatibility complex class I chain-related A (MICA) alleles have previously been shown to be associated with PsA; however it is unclear whether there is a differential association of MICA alleles with skin and joint manifestations of PsA. Here, we describe a case-control study that aims to validate previously reported MICA allele associations with PsA and determine whether MICA alleles differentiate patients with PsA from those with psoriasis without PsA. Two hundred forty-nine unrelated Caucasian PsA patients, 243 psoriasis patients without arthritis, and 248 healthy controls were genotyped for 55 MICA alleles using PCR-SSP, and for human leucocyte antigen (HLA)-B and HLA-C alleles by PCR-SSO reverse line blot. Allele frequencies were calculated and logistic regressions were performed, adjusting for HLA-B and HLA-C alleles previously shown to be associated with psoriasis and/or PsA. Several MICA alleles were associated with psoriatic disease, PsA, and psoriasis compared with controls, and PsA compared with psoriasis in univariate analyses. Haplotype analysis showed evidence of strong linkage disequilibrium (LD) between PsA and psoriasis risk alleles of HLA-C, HLA-B, and MICA. After adjusting for significant HLA-B and HLA-C alleles in multivariate analyses, MICA*016 remained significantly associated with psoriasis [odds ratio (OR) = 5.5, P = 0.008]. MICA*00801 homozygosity was associated with susceptibility to PsA when compared with patients with psoriasis alone (OR = 2.26, P = 0.009). We conclude that most MICA allele associations with psoriasis and PsA are dependent on LD with HLA-B and HLA-C risk alleles. Independent of HLA, only MICA*016 influences the risk of developing psoriasis without arthritis, and homozygosity for MICA*00801 increases the risk of developing PsA in patients with psoriasis.

Clinicians' attitudes towards clinical trials of cancer therapy.

BACKGROUND: Patient accrual into cancer clinical trials remains at low levels. This survey elicited attitudes and practices of cancer clinicians towards clinical trials. METHOD: The 43-item Clinicians Attitudes to Clinical Trials Questionnaire was completed by participants in an intervention study aimed at improving multi-disciplinary involvement in randomised trials. Responses from 13 items were summed to form a research-orientation score. RESULTS: Eighty-seven clinicians (78%) returned questionnaires. Physicians, more often than surgeons, chose to prioritise prolonging a patient's life, recruited ≥50% of patients into trials and attended more research-focussed conferences. Clinicians at specialist centres were more positive about trials with no-treatment arms than those at district general hospitals, more likely to believe clinician, rather than patient reluctance to participate was the greater obstacle to trial accrual, and preferred national and international to local recognition. Clinicians belonging to breast and colorectal teams were less disappointed about not enrolling patients in trials and more accepting of no-treatment arm trials. Research orientation was higher in physicians than surgeons and higher in specialist centres than district hospitals. CONCLUSIONS: This study provides greater understanding of clinicians' attitudes to trials. Results have been used to inform training interventions for clinicians targeting the problem of low and selective accrual.

SF-36 summary and subscale scores are reliable outcomes of neuropsychiatric events in systemic lupus erythematosus.

OBJECTIVE: To examine change in health-related quality of life in association with clinical outcomes of neuropsychiatric events in systemic lupus erythematosus (SLE). METHODS: An international study evaluated newly diagnosed SLE patients for neuropsychiatric events attributed to SLE and non-SLE causes. The outcome of events was determined by a physician-completed seven-point scale and compared with patient-completed Short Form 36 (SF-36) health survey questionnaires. Statistical analysis used linear mixed-effects regression models with patient-specific random effects. RESULTS: 274 patients (92% female; 68% Caucasian), from a cohort of 1400, had one or more neuropsychiatric event in which the interval between assessments was 12.3 ± 2 months. The overall difference in change between visits in mental component summary (MCS) scores of the SF-36 was significant (p<0.0001) following adjustments for gender, ethnicity, centre and previous score. A consistent improvement in neuropsychiatric status (N=295) was associated with an increase in the mean (SD) adjusted MCS score of 3.66 (0.89) in SF-36 scores. Between paired visits when the neuropsychiatric status consistently deteriorated (N=30), the adjusted MCS score decreased by 4.00 (1.96). For the physical component summary scores the corresponding changes were +1.73 (0.71) and -0.62 (1.58) (p<0.05), respectively. Changes in SF-36 subscales were in the same direction (p<0.05; with the exception of role physical). Sensitivity analyses confirmed these findings. Adjustment for age, education, medications, SLE disease activity, organ damage, disease duration, attribution and characteristics of neuropsychiatric events did not substantially alter the results. CONCLUSION: Changes in SF-36 summary and subscale scores, in particular those related to mental health, are strongly associated with the clinical outcome of neuropsychiatric events in SLE patients.

An assessment of disease flare in patients with systemic lupus erythematosus: a comparison of BILAG 2004 and the flare version of SELENA.

AIMS: To compare the British Isles Lupus Assessment Group (BILAG) 2004, the Safety of Estrogens in Lupus Erythematosus National Assessment (SELENA) flare index (SFI) and physician's global assessment (PGA) in assessing flares of disease activity in patients with systemic lupus erythematosus (SLE). METHODS: Sixteen patients with active SLE were assessed by a panel of 16 rheumatologists. The order in which the patients were seen was randomised using a 4×4 Latin square design. Each patient's flare status was determined at each assessment using the BILAG 2004 activity index; the SFI and a PGA. A group of five specialists designated each patient into severe, moderate, mild or no flare categories. RESULTS: The rate of complete agreement (95% CI) of the four individual examining physicians for any flare versus no flare was 81% (55% to 94%), 75% (49% to 90%) and 75% (49% to 90%) for the BILAG 2004 index, SELENA flare instrument and PGA, respectively. The overall agreement between flare defined by BILAG 2004 and the SFI was 81% and when type of flare was considered was 52%. Intraclass correlation coefficients (95% CI), as a measure of internal reliability, were 0.54 (0.32 to 0.78) for BILAG 2004 flare compared with 0.21 (0.08 to 0.48) for SELENA flare and 0.18 (0.06 to 0.45) for PGA. Severe flare was associated with good agreement between the indices but mild/moderate flare was much less consistent. CONCLUSIONS: The assessment of flare in patients with SLE is challenging. No flare and severe flare are identifiable but further work is needed to optimise the accurate 'capture' of mild and moderate flares.

The attitudes of 1066 patients with cancer towards participation in randomised clinical trials.

BACKGROUND: barriers to randomised clinical trial (RCT) recruitment include failure to identify eligible patients, reluctance of staff to approach them and attitudes of some health-care professionals and patients. As part of a larger UK prospective study examining the communication and involvement in RCTs of 22 multidisciplinary teams in Wales, we also assessed the attitudes of patients they treat towards trials. METHODS: out of 1146 patients attending outpatient departments who were approached, 1146 (93%) completed the seven-item Attitudes to Randomised Trials Questionnaire (ARTQ), probing their general attitudes towards medical research and likely participation in a hypothetical two-arm RCT. RESULTS: randomisation initially deterred many patients from endorsing a willingness to participate. However, if information about the trial logic, voluntary nature and rights to withdraw were provided, together with further treatment details, 83% (886 out of 1066) would potentially participate. Other variables associated with a positive inclination towards participation included previous trial experience (P<0.01), male gender (P<0.01) and younger age, with patients > or =70 years less likely to consider trial entry (P<0.01). CONCLUSION: the majority of patients were receptive to RCT participation. Many of those initially disinclined because of randomisation would consider joining if given further details that form part of standard GCP consent guidelines. These data show the importance and need for clear communication and information to encourage RCT participation. Evidence-based training courses are available to assist with this.

One-stage parametric meta-analysis of time-to-event outcomes.

Methodology for the meta-analysis of individual patient data with survival end-points is proposed. Motivated by questions about the reliance on hazard ratios as summary measures of treatment effects, a parametric approach is considered and percentile ratios are introduced as an alternative to hazard ratios. The generalized log-gamma model, which includes many common time-to-event distributions as special cases, is discussed in detail. Likelihood inference for percentile ratios is outlined. The proposed methodology is used for a meta-analysis of glioma data that was one of the studies which motivated this work. A simulation study exploring the validity of the proposed methodology is available electronically.

Awareness of ovarian cancer risk factors, beliefs and attitudes towards screening: baseline survey of 21,715 women participating in the UK Collaborative Trial of Ovarian Cancer Screening.

BACKGROUND: Women's awareness of ovarian cancer (OC) risks, their attitudes towards and beliefs about screening, together with misunderstandings or gaps in knowledge, may influence screening uptake. METHODS: In total, 21 715 post-menopausal women completed questionnaires before randomisation into the UK Collaborative Trial of Ovarian Cancer Screening. RESULTS: In all, 42.3% correctly identified their lifetime risk of OC; 87.1% knew that a family history of OC increased risk, but only 26.7% appreciated the association with a family history of breast cancer. Although 38.2% acknowledged increased risk post-menopause, only 8.8% were aware that OC diagnoses are highest in women over 65 years. Few (13.7%) recognised the association between pregnancy and reduced OC risk or protective effects of breastfeeding (6.2%). There were common misconceptions; 37.2% thought that an abnormal cervical smear and 26.4% that oral contraception increased the likelihood of OC. Although 84.4% recognised that most ovarian masses are benign, 20.2% thought having had a benign cyst increased OC risk. Most (99.4%) believed that a high uptake of OC screening would reduce mortality and (96.2%) that screen-detected cancers would have an improved prognosis. CONCLUSIONS: The results show a need for improved public understanding about OC risks and provide important information for GPs and health educationalists about initiatives needed for future awareness, prevention and screening programmes.

Multi-state Markov models for disease progression in the presence of informative examination times: an application to hepatitis C.

In many chronic diseases it is important to understand the rate at which patients progress from infection through a series of defined disease states to a clinical outcome, e.g. cirrhosis in hepatitis C virus (HCV)-infected individuals or AIDS in HIV-infected individuals. Typically data are obtained from longitudinal studies, which often are observational in nature, and where disease state is observed only at selected examinations throughout follow-up. Transition times between disease states are therefore interval censored. Multi-state Markov models are commonly used to analyze such data, but rely on the assumption that the examination times are non-informative, and hence the examination process is ignorable in a likelihood-based analysis. In this paper we develop a Markov model that relaxes this assumption through the premise that the examination process is ignorable only after conditioning on a more regularly observed auxiliary variable. This situation arises in a study of HCV disease progression, where liver biopsies (the examinations) are sparse, irregular, and potentially informative with respect to the transition times. We use additional information on liver function tests (LFTs), commonly collected throughout follow-up, to inform current disease state and to assume an ignorable examination process. The model developed has a similar structure to a hidden Markov model and accommodates both the series of LFT measurements and the partially latent series of disease states. We show through simulation how this model compares with the commonly used ignorable Markov model, and a Markov model that assumes the examination process is non-ignorable.

Prospective analysis of neuropsychiatric events in an international disease inception cohort of patients with systemic lupus erythematosus.

OBJECTIVES: To determine the frequency, accrual, attribution and outcome of neuropsychiatric (NP) events and impact on quality of life over 3 years in a large inception cohort of patients with systemic lupus erythematosus (SLE). METHODS: The study was conducted by the Systemic Lupus International Collaborating Clinics. Patients were enrolled within 15 months of SLE diagnosis. NP events were identified using the American College of Rheumatology case definitions, and decision rules were derived to determine the proportion of NP disease attributable to SLE. The outcome of NP events was recorded and patient-perceived impact determined by the SF-36. RESULTS: 1206 patients (89.6% female) with a mean (+/-SD) age of 34.5+/-13.2 years were included in the study. The mean disease duration at enrollment was 5.4+/-4.2 months. Over a mean follow-up of 1.9+/-1.2 years, 486/1206 (40.3%) patients had > or =1 NP events, which were attributed to SLE in 13.0-23.6% of patients using two a priori decision rules. The frequency of individual NP events varied from 47.1% (headache) to 0% (myasthenia gravis). The outcome was significantly better for those NP events attributed to SLE, especially if they occurred within 1.5 years of the diagnosis of SLE. Patients with NP events, regardless of attribution, had significantly lower summary scores for both mental and physical health over the study. CONCLUSIONS: NP events in patients with SLE are of variable frequency, most commonly present early in the disease course and adversely impact patients' quality of life over time. Events attributed to non-SLE causes are more common than those due to SLE, although the latter have a more favourable outcome.