The aim of the present study was to describe, on the basis of specific classification criteria and for a period of two years after delivery, the prognosis for women suffering from pregnancy-related pelvic joint pain, and to describe the characteristics influencing the prognosis. One thousand seven hundred and eighty-nine pregnant women who were booked for delivery at Odense University Hospital formed a cohort to investigate the prognosis. Women whose reported daily pain from pelvic joints could be objectively confirmed were divided, according to symptoms, into five subgroups (n=405) - four classification groups (pelvic girdle syndrome, symphysiolysis, one-sided sacroiliac syndrome and double-sided sacroiliac syndrome) and one miscellaneous. The women in the five subgroups were re-examined at regular intervals for two years after delivery or until disappearance of symptoms (whichever was less). Thre hundred and forty-one women from the 5 subgroups participated in the postpartum follow-up. The majority (62.5%) of women in the four classification groups experienced disappearance of pain within a month after delivery. Two years after parturition 8.6% were still suffering from pelvic joint pain (determined subjectively and objectively). Persistence of pain was found to vary significantly from one classification group to another. None of those initially classified as suffering from symphysiolysis had pain 6 months after delivery in comparison to the 21 percent of those with pelvic girdle syndrome who continued to have pain at the two-year mark. This study shows that pregnancy-related pelvic joint pain had an excellent postpartum prognosis (in general) in three out of four classification groups. The women with pelvic girdle syndrome (pain in all 3 pelvic joints) had a markedly worse prognosis than the women in the other three classification groups. High number of positive test and a low mobility index were identified as giving the highest relative risk for long term pain.