BACKGROUND:Current knowledge of the natural outcomes of untreated women in preterm labour is both incomplete and outcomes vary significantly between the available studies. The aim of this study was to systematically review outcomes of preterm labour without tocolysis and determine if outcome variation could be accounted for by differences in study populations. Such data could potentially assist in the interpretation of intervention trials that do not include a no-treatment arm. METHODS:Included studies reported outcomes of women in clinically diagnosed preterm labour without tocolytic treatment between 1950 and 2011. Studies that were limited to preterm prelabour rupture of membranes, recurrent preterm labour or in which the women without tocolysis represented a potentially biased subgroup, or were not tocolysed because of contraindications were excluded. Study quality, design, and population characteristics were abstracted. Outcomes included pregnancy prolongation and the proportion of women undelivered at 48-72 h, 7 days, and term. Study characteristics associated with differing odds of preterm birth were explored through logistic regression. RESULTS:Three hundred and eighty-five citations were initially identified, of which 26 were included encompassing 1383 women. The percentage of patients who were undelivered at 48-72 hours was 62.8%, at 7 days 53.4% and 40.4% delivered at term, though the range was very wide. Characteristics associated with decreased odds of delivery were not consistent among reported outcome measures. CONCLUSIONS:Most women without tocolysis do not deliver within 7 days, although the range is very wide. The majority of this variation is unrelated to reported differences in study design or reported population characteristics.