Rheumatoid arthritis in a population sample.

Many estimates have been made of the prevalence of rheumatic complaints in various populations. In earlier studies information was collected from general practitioners (Ministry of Health, 1924), and later from insurance certificates (Medical Advisory Committee, Scotland, 1945). Field surveys have also been conducted by house-to-house interviews with questionnaires (Woolsey, 1952; Stocks, 1949; Brooke, 1953) and with questionnaires supplemented by clinical examination of individuals with complaints (Edstrbm, 1944; Kellgren, Lawrence, and Aitken-Swan, 1953; Blecourt, 1953). All these studies have shown that rheumatic complaints as a whole are common, particularly in the older age groups, and that they cause much disability. The prevalence rates for rheumatic complaints as a whole in these different studies are similar if one takes into account the different methods employed for collecting data; but attempts to classify rheumatic complaints into diagnostic categories have given widely divergent results. Thus estimates of the prevalence of rheumatoid arthritis have ranged from 0 4 to 23 per cent., and of osteoarthritis from 0-04 to 6-2 per cent., the lowest rates being those for self-diagnosis in reply to questionnaires published by Stocks (1949), and the highest those for clinical examination in the field obtained by Kellgren and others (1953). Owing to the lack of diagnostic definition and objective criteria of disease, these divergent estimates of prevalence probably reflect differences of method and language amongst observers rather than a differing incidence of disease, and if any useful epidemiological studies are to be carried out in this field of medicine more precise and reproducible diagnostic definitions must be used and more precise objective criteria of disease must be introduced. The diagnostic difficulty is great, since