The prevalence of low back pain (LBP) is higher in pregnant women compared to women of the same age in a general population. Pregnancy-related LBP persists 6 years after pregnancy in 16% of women. Consequently, pregnancy represents a specific risk for LBP and persistent LBP. Pregnancy-related LBP is usually studied as a single entity, however, only one subgroup of LBP, i.e. pelvic girdle pain (PGP), seems to be associated with pregnancy. Accordingly, possible differences in subgroups of patients with LBP are unknown. The aims of this thesis were the following: 1) to describe the prevalence of clinically classified subgroups of women with LBP in a cohort (no LBP, lumbar pain, PGP, and combined pain (PGP and lumbar pain)) during pregnancy and postpartum, and 2) to determine if there was a disparity in the course, health-related quality of life (HRQL), pain intensity, disability, depressive symptoms, or muscle function in subgroups of the cohort, and 3) to identify predictors for having persistent pregnancy-related PGP postpartum. Consecutively-enrolled pregnant women were classified into LBP subgroups by mechanical assessment of the lumbar spine, pelvic pain provocation tests, standard history, and pain drawings. All women answered questionnaires (background data, EQ-5D). Women with LBP completed the Oswestry Disability Index and pain measures. The Edinburgh Postnatal Depression Scale was used to evaluate depressive symptoms at 3 months postpartum (cut-off ≥10). Trunk muscle endurance, hip muscle strength, and gait speed were investigated. Multiple logistic regression was used to identify predictors from self-reports and clinical examination. At the 12-18 gestational week evaluation, 118/308 (38%) women had no LBP, 33 (11%) had lumbar pain, 101 (33%) had PGP, and 56 (18%) had combined pain. Three months postpartum, 183/272 (67%) women had no LBP, 29 (11%) had lumbar pain, 46 (17%) had PGP, and 14 (5%) had combined pain. Pregnant women with combined pain were most affected in terms of HRQL, pain intensity, and disability. Depressive symptoms were three times more prevalent in women with LBP (27/87, 31%) than in women without LBP (17/180, 9%). Women with PGP and/or combined pain had lower values for trunk muscle endurance, hip extensor strength and gait speed compared to women without LBP. Postpartum, 16-20% of the women had persistent combined pain or PGP, whereas 1/29 had lumbar pain. Predictors for persistent PGP or combined pain were work dissatisfaction, older age, combined pain in early pregnancy, and low endurance of the back flexors. In conclusion, women with combined pain were identified to be a target group since they had the lowest recovery rate and since the classification of combined pain was found to be a predictor for persistent PGP or combined pain postpartum. The hypothesis of an association between muscle dysfunction and PGP was strengthened. Based on the finding of high comorbidity of postpartum depressive symptoms and LBP, it seems important to screen for and consider treatment strategies for both symptoms.