This is a systematic review on the effect of maternal position during the first stage of labor. The main question addressed here is: Does encouraging women to adopt an upright position or to ambulate during the first stage of labor reduce the duration of this stage? All randomized controlled trials comparing upright position or ambulation with any other position were taken into consideration. We defined upright position as the adoption of a not supine position during labor, (i.e. walking, sitting, standing, kneeling, and squatting). The following endpoints were evaluated: need for labor augmentation, mode of delivery, use of analgesia, neonatal condition at birth and maternal comfort and satisfaction.
Studies were identified by performing a search of the following electronic databases: MEDLINE, Popline, the Scientific Electronic Library On-line (SciELO) and the Latin American and Caribbean Health Science Information (LILACS). These databases were searched using the following strategy and keywords: ("labor") AND ("first stage" OR "position" OR "mobility" OR "up right" OR "upright" OR "active phase" OR "latent phase" OR "maximum slope" OR "recumbent" OR "lateral" OR "sitting" OR "standing" OR "ambulation" OR "kneeling" OR "squatting"). This search was not restricted by date or language. In addition, the proceedings of several scientific meetings were hand-checked and reference lists of retrieved publications were screened.
Possible eligibility was assessed independently by two reviewers. Initially, the citations identified were evaluated on the basis of their titles and/or abstracts and full text was retrieved if found to be eligible. All citations considered to be clearly irrelevant were excluded. If the information provided by titles or abstracts was considered insufficient to decide on inclusion or exclusion of the publication, the full-text article was retrieved and evaluated. Studies were assessed independently by two reviewers and decision upon inclusion was based on consensus between the two. Methodological quality assessment took into account the adequacy of allocation concealment. The concealment of allocation was considered adequate (and scored as 'A') when the study under consideration adopted a process in which the person deciding about the inclusion of a participant into a randomized controlled trial did not know the comparison group into which that individual was being allocated. In case allocation concealment was unclear, the score was 'B' (table 1). The occurrence of attrition, performance and detection bias was screened and the sampling method and the mode of presentation of results were also evaluated.
Data extraction and statistical analysis were performed according to the guidelines recommended in the Cochrane Handbook . The decision whether to perform data pooling was based on the clinical similarity of studies. Whenever an alternative measure of variability was applied, an approximation or a direct algebraic relationship was used to obtain the standard deviation. After data pooling, statistical heterogeneity was identified and evaluated as moderate or high (i.e. more than 40% using the I2 statistic) . If heterogeneity was low, a fixed effects model was used for statistical analysis, and if heterogeneity was moderate or high the random effects model was applied. Depending on the heterogeneity level, the standardized mean difference (SMD, low heterogeneity) or the weighted mean difference (WMD, moderate to high heterogeneity) was used. Odds ratios (OR) with 95% confidence intervals (CI's) were calculated. Subgroup analyses according to parity and other characteristics of labor (maximum slope phase and protracted labor) were performed. Analysis was carried out using the Revman software package, version 4.2.8 (The Nordic Cochrane Centre, Rigshospitalet 2003).