11
The use of data from early arthritis clinics for clinical research

https://doi.org/10.1016/j.berh.2008.11.008Get rights and content

Many early arthritis clinics have been started in the past decade. A major objective of these clinics is to improve our understanding of early arthritis in its undifferentiated form and to help provide guidance, recommendations, or criteria for diagnostic and therapeutic decision-making in patients with such presentation. Increasingly, they will allow aspects of pathogenesis – including autoantibodies and potential genetic markers – to be included in the set of clinical predictors that a rheumatologist is presented with. From an analytical perspective, usually logistic regression modelling is used to identify the best predictors of potentially long-lasting and/or erosive disease. Classification tree analysis might be another way to analyse data, and has the advantage that the results are easier to interpret than statistical parameters. In the past, many such projects have been published, none of which has achieved widespread use. Currently, the American College of Rheumatology and the European League Against Rheumatism are in the process of defining new criteria for rheumatoid arthritis that will allow earlier diagnosis and treatment of patients and definition of patients with early disease for inclusion in clinical trials.

Section snippets

Typical profiles of patients enrolled in early arthritis cohorts

The typical patient in an early arthritis cohort is one with a recent-onset joint swelling that has sought medical help because of considerable pain. Signs of inflammation are not necessarily present in this patient. The types of arthritis detected and enrolled in the cohorts clearly depend on how the clinics are situated on the basis of a supporting general practitioner (GP) network, or how the healthy population is directly made aware of the problem. If, for example, patients present directly

Where can early arthritis data help us?

Before we can discuss what early arthritis cohorts can tell us, we need to be aware that in early arthritis some immunological abnormalities occur years before symptom onset [10]. This has led to the idea that patients can never be seen before all steps in disease development occur. Thus in order to study as many sequelae in disease pathogenesis as possible, it is beneficial to get potential patients at the earliest possible time points.

But what are the delays to treatment? It is only partly

Logistic regression analysis

The traditional statistical approach to studies aiming at identifying predictors of a dichotomous outcome is the logistic regression model. In these models, the probability of occurrence of a dichotomous event is predicted by fitting independent variables to a logistic curve. The event is traditionally coded as 0/1. Logistic regression models are part of the family of generalized linear models, which can handle independent variables that are continuous or categorical. Predictors can be

Circularity of models

In using data from patients with early disease to predict rheumatoid arthritis, one central limitation is inherent to any model, namely the problem of circularity. Put simply, circularity in this context is present as soon as the status of the dependent variable (i.e. the presence or absence of the outcome) is inherently influenced by the levels of the baseline variables. A typical example is the use of RA as the dependent variable by defining it using the traditional 1987 classification

The need for ‘official’ criteria and the value of joint forces

In the past there have been several reports from individual centres providing algorithms, rules, and criteria to diagnose RA early in its course *[14], *[15], [16], *[17], *[18]. Each individual in these reports has particular merits and helps to better understand signs of a potentially persistent and erosive disease. So far, all these predictive models have suffered from the same problem of not getting widespread acceptance or application in an international community, and, with some

Pathogenetic considerations

Another major achievement from the use of data from early arthritis clinics is the growing insight into the pathogenesis of disease. A nice example is that in various early arthritis registries the association between cigarette smoking and RA has been reported [22]. Possible links between smoking and the development of RA have been demonstrated, especially in seropositive RA [23], [24]. Subsequently it has been demonstrated that the association of smoking was specific for patients with certain

Future prospects

Early arthritis clinics have helped considerably to advance the field of arthritis research at a time when it was recognized that following the pyramid approach of watchful waiting without risking therapeutic harm was no longer a respectable strategy. Nevertheless, the process of starting an early arthritis clinic is cumbersome and has taken several years for many investigators until the childhood diseases of these cohorts were subsisted. Although until now several publications have come from

References (29)

  • H. Kallberg et al.

    Epidemiological Investigation of Rheumatoid Arthritis study group.Gene-gene and gene-environment interactions involving HLA-DRB1, PTPN22, and smoking in two subsets of rheumatoid arthritis

    Am J Hum Genet

    (2007 May)
  • F. Wolfe

    The natural history of rheumatoid arthritis

    J Rheumatol

    (1996)
  • T. Pincus et al.

    Severe functional declines, work disability, and increased mortality in seventy-five rheumatoid arthritis patients studied over nine years

    Arthritis Rheum

    (1984)
  • J.S. Smolen et al.

    Patients with rheumatoid arthritis in clinical care

    Ann Rheum Dis

    (2004)
  • K.R. Wilske et al.

    Challenging the therapeutic pyramid: a new look at treatment strategies for rheumatoid arthritis

    J Rheumatol

    (1990)
  • J.F. Fries et al.

    The relative toxicity of disease-modifying antirheumatic drugs

    Arthritis Rheum

    (1993)
  • J.F. Fries et al.

    The relative toxicity of nonsteroidal antiinflammatory drugs

    Arthritis Rheum

    (1991)
  • F.C. Arnett et al.

    The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis

    Arthritis Rheum

    (1988)
  • A.J. Silman et al.

    Selection of study population in the development of rheumatic disease criteria: comment on the article by the American College of Rheumatology Diagnostic and Therapeutic Criteria Committee

    Arthritis Rheum

    (1995)
  • J. van Aken et al.

    The Leiden Early Arthritis Clinic

    Clin Exp Rheumatol

    (2003 Sep–Oct)
  • M.M. Nielen et al.

    Specific autoantibodies precede the symptoms of rheumatoid arthritis: a study of serial measurements in blood donors

    Arthritis Rheum

    (2004)
  • P. Emery et al.

    Early referral recommendation for newly diagnosed rheumatoid arthritis: evidence based development of a clinical guide

    Ann Rheum Dis

    (2002)
  • H. van Dongen et al.

    Efficacy of methotrexate treatment in patients with probable rheumatoid arthritis: a double-blind, randomized, placebo-controlled trial

    Arthritis Rheum

    (2007 May)
  • Y.P. Goekoop-Ruiterman et al.

    Comparison of treatment strategies in early rheumatoid arthritis: a randomized trial

    Ann Intern Med

    (2007 Mar 20)
  • Cited by (0)

    View full text