Nineteen studies (6,339 participants, range 43 to 1,578) were included in the review. Two studies were at low risk of bias, three were at moderate risk of bias and 14 high risk of bias.
Overall results indicated a significant increase in physical activity in women who received the intervention (SMD 0.16, 95% CI 0.04 to 0.29; 19 studies) but there was significant statistical heterogeneity (Ι²=66%). Subgroup analyses indicated that intervention delivery mode explained some of the heterogeneity: group delivery (SMD 0.36, 95% CI 0.17 to 0.54; 11 studies; Ι²=59%), individual delivery (SMD -0.02, 95% CI -0.35 to 0.31; five studies; Ι²=63%) and community delivery (SMD -0.02, 95% CI -0.10 to 0.05; three studies; Ι²=0%).
Sensitivity analyses did not significantly alter the results. A funnel plot showed some evidence of publication bias.