Rheumatoid factor positivity in the general populationBMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e5841 (Published 06 September 2012) Cite this as: BMJ 2012;345:e5841
- Julia F Simard, assistant professor,
- Marie Holmqvist, research fellow
- 1Department of Medicine, Clinical Epidemiology Unit, Karolinska Institutet, 171 76 Stockholm, Sweden
Rheumatoid factor is an autoantibody used in the classification of rheumatoid arthritis. It can also be present in other inflammatory conditions and is seen in people without known inflammatory conditions. Some studies suggest that the prevalence of rheumatoid factor positivity in the general population increases with age and smoking status.1 2 As a consequence, merely being rheumatoid factor positive, without any symptoms, rarely prompts follow-up aimed at early identification of rheumatoid arthritis today.
In a linked research paper (doi:10.1136/bmj.e5244), Nielsen and colleagues investigated the long term risk of rheumatoid arthritis in the Danish general population according to titre of rheumatoid factor. They used national hospital registers to detect rheumatoid arthritis in a sample of the general population in whom plasma samples had been collected and frozen in 1981-83.3 Around 4.3% of the cohort were seropositive—a proportion generally comparable to other studies in non-rheumatoid arthritis populations.1 2 4 The authors reported that the risk of being given a discharge diagnosis of rheumatoid arthritis during long term (up to 28 years) follow-up increased as the baseline rheumatoid factor titre increased, and they estimated 10 year risks by age, sex, smoking, and rheumatoid factor titre at start of follow-up. On the basis of their findings, they suggest that people with an incidentally high rheumatoid factor titre should be followed up by a rheumatologist or early arthritis clinic.
Investigators have previously shown that rheumatoid factor positivity antedates the clinical onset of rheumatoid arthritis.5 6 The current study looked at the association prospectively in a sample of the general population that was selected neither for exceptional health nor some underlying disease. The study did not assess the usefulness of rheumatoid factor in patients diagnosed as having rheumatoid arthritis but essentially monitored people with an incidental rheumatoid factor positive result. By estimating the risk of rheumatoid arthritis in this way the authors challenge how “normal” it is for the average person to be rheumatoid factor seropositive and whether such a person is at high risk of the disease and would benefit from clinical monitoring. The study’s findings suggest that the assessment of rheumatoid factor titre, sex, age, and smoking status could identify subgroups of patients who would benefit by being followed up more closely. The authors estimated that women aged 50-69 years with a history of smoking who were in the highest category of rheumatoid factor titre had the highest 10 year risk of rheumatoid arthritis (32% risk). Without information on whether smoking was current or past, however, we cannot know whether the dose, time since quitting, or duration of smoking matters.
Although these findings are interesting epidemiologically they may not be that useful in clinical practice. Seropositivity is a common feature of the diagnosis of rheumatoid arthritis and part of the classification criteria for the disease.7 8 Nielsen and colleagues suggest that incidentally detected rheumatoid factor should prompt increased clinical vigilance and perhaps a referral to a rheumatologist. However, rheumatoid factor testing tends to follow a clinical suspicion so it is unclear how often such incidental findings are likely to occur in clinical practice. The authors do not explicitly suggest screening, so individuals are likely to be detected as seropositive only rarely.
Although useful, register based data on hospital admissions often cannot provide true estimates of the start date of the illness, either because of lack of data from primary care or because care was received before the registers’ start date. Nielsen and colleagues report that the time from baseline to “incidence” may be inversely associated with rheumatoid factor titre, such that those with higher titres at sampling developed rheumatoid arthritis sooner than those with lower titres. It is not clear whether this finding can refute the previously held belief that rheumatoid factor titre increases with age, as the authors suggest it does, or whether it is subject to some misclassification or other bias. This might be the focus of future research that investigates the link between rheumatoid factor and rheumatoid arthritis.
Clinicians have now shifted towards using anti-citrullinated protein antibody, another autoantibody that is more specific than rheumatoid factor, in the diagnosis of rheumatoid arthritis. The role of this autoantibody in the development of rheumatoid arthritis is better understood, so it would be useful to estimate the risks of rheumatoid arthritis in the general population according to anti-citrullinated protein antibody titre, as the authors mention.
Cite this as: BMJ 2012;345:e5841
Competing interests: Both authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work .
Provenance and peer review: Commissioned; not externally peer reviewed.