More comprehensive programmes did not produce better results at group level. Also short and less costly interventions based on educational and cognitive principles were valuable for persons with longstanding fibromyalgia. More attention must be given to evaluating the clinical effect of programmes.

Arthritis of the sacro-iliac joints is a fairly common condition occurring either in isolation or as a part or pelvospondylitis. Because of the difficulty of establishing a diagnosis before X-ray changes have appeared, various radionuclide scanning techniques have been developed to make possible an early diagnosis. Opinions have differed as to the reliability of radionuclide methods using "bone seeking" radiopharmaceuticals. Patients with clinically and/or radiologically proven sacro-iliitis were imaged by means of 99m-Tc-EHDP. An S.I. index was calculated as the ratio of the maximum activity of the sacro-iliac joints to the minimum activity of the iliac wings. The accuracy of the method was very good, and it is recommended as a supplement to clinical examination and X-ray studies.

A total of 222 female patients with seronegative rheumatic diseases were investigated by sacroiliac joint X-ray and HLA-A, B and Ctyped. The frequency of the B27 antigen was significantly increased in the following groups: definite pelvospondylitis (26 patients and relative risk (RR) = 99), possible pelvospondylitis (20 patients, RR=4.5), definite sacroiliitis (14 patients, RR-20.1) and uroarthritis (8 patients). When 24 B27 negative patients from these B27 associated diseases were analysed, significantly increased frequencies of the BW22 (RR=16.7) and CW1 (RR=14.4) antigens were found. There were no significantly deviating HLA frequencies in the following diagnostic groups: clinical sacroiliitis (20 patients), polyarthritis with clinically silent sacroiliitis (12 patients), polyarthritis without sacroiliitis (47 patients), arthralgia preceded by beta-streptococcal infection (17 patients), other arthralgia (33 patients), osteo-arthritis (13 patients) and other arthritis (12 patients).

Over the period 1976-83 the clinical records were studied of 150 patients with the diagnosis of confirmed or possible sacro-iliitis. All these patients had pathologic sacro-iliac indices on quantitative bone scintigraphy. The most recent radiographs of the sacro-iliac joints were examined by two radiologists independently of each other. 103 (68%) patients with a mean duration of symptoms of 6.7 years, had normal radiographs. The frequency of HLA B27 in this group was only 27%, contrary to the expected 90-100% in an ankylosing spondylitis (AS) population. In a follow-up study, 30 patients, who in 1976-79 had normal sacro-iliac joint X-rays, were further investigated. The mean duration of symptoms was 11.3 years. 21 patients (70%) still had normal radiographs of their sacro-iliac joints and the HLA B27 frequency in this group was 28%. Normal radiographs of the sacro-iliac joints, in spite of the long duration of symptoms together with a low frequency of HLA B27, makes the diagnosis of AS most unlikely. This indicates a low specificity for quantitative sacro-iliac scintigraphy in the early diagnosis of AS. In the follow-up study, Calin's screening test for AS was included and was also found to have a low specificity.