Undiagnosed systemic lupus erythematosus in the community

Abstract

Background: Systemic lupus erythematosus (SLE) produces a wide range of symptoms, some of them mild; therefore it may be underdiagnosed. In a pilot study we tested the ability of a screening questionnaire to identify undiagnosed cases in the community.

Methods: A 10-item questionnaire was mailed to 3500 women aged 18-65 randomly selected from a general practice list in Birmingham, UK. A random sample of 300 non-respondents were sent two further mailings. Women returning the questionnaire with three positive answers were to attend for an antinuclear antibody (ANA) test; and those testing positive were to be referred to a rheumatologist for standard clinical assessment.

Findings: 1153 (30%) of the women responded, and 307 of these gave three or more positive answers on the questionnaire. The survey disclosed three women with previously undiagnosed SLE; and, in addition, it picked up all four of the women whose SLE had been diagnosed previously.

Interpretation: The prevalence of SLE revealed by this survey in Birmingham women is 200/100,000 (95% confidence interval 80 to 412), whereas the prevalence of diagnosed SLE in this age group is 54/100,000 (95% CI 47-62). Thus, a substantial number of undiagnosed cases exist, and further work needs to focus on the prevalence in different ethnic groups and the benefits of early diagnosis and treatment.