The aims of this study were to compare prevalence of back pain in parous, nulliparous, pregnant and non-pregnant women and to determine whether there is an association between incontinence and back pain in pregnant women. Associations between back pain, pregnancy, parity and incontinence were assessed in 14,779 younger and 14,099 mid-age women using chi-squared analysis. The odds of back pain were modelled with multinomial logistic regression. Back pain was more frequent in parous than nulliparous (p < 0.001) and pregnant than non-pregnant (p < 0.001) younger women. However, no associations were seen for mid-aged women. Pregnant women who had incontinence had increased odds ratios for 'often' and 'rarely or sometimes' having back pain (8.5 and 3.8, respectively). This study suggests that pregnancy may lead to earlier development of back pain, without affecting long-term prevalence. Incontinence and back pain may be related because of contribution of the trunk muscles to continence and lumbopelvic control.