Natural childbirth is a normal physiological approach of human reproduction, while in China, the cesarean section rate has exceeded the upper limit recommended by the WHO by more than two times. As a nonpharmaceutical intervention facilitating natural childbirth, physical exercise should be valued by relevant departments.

To perform a meta-analysis of studies regarding effects of physical exercise interventions on promoting natural childbirth probability, and to attempt to recommend a better exercise scheme.

Randomized controlled trials (RCTs) about the effect of intrapregancy physical exercise versus routine care (health duration or irregular physical exercise) improving the probability of natural childbirth published between 1990 and 2021 were searched in databases of CNKI, SinoMed, CQVIP, PubMed, EMBase, Web of Science, and Cochrane Library. Meanwhile, the references of systematic reviews on relevant RCTs were also searched and reviewed. Data extraction and methodological quality assessment were conducted for the included studies. RevMan 5.2 was used for meta-analysis. GRADE was used to rate the quality of evidence concerning an outcome.

A total of 30 studies with 3 910 subjects were included. Meta-analysis found that physical exercise interventions were superior to routine care in improving the natural childbirth rate〔RR=1.34, 95%CI (1.28, 1.40) , P<0.000 01〕. Subgroup analyses revealed that natural childbirth rate was improved the most by physical exercise intervention started later than 24 weeks of gestation, followed by that started at 13-24 weeks of gestation, and 12 or less than 12 weeks of gestation. With respect to the type of physical exercise, the natural childbirth rate was improved the most by pelvic floor muscle training, general sports lessons, gymnastics, aerobic exercise, birthing ball movement, and yoga. As for the frequency of physical exercise, the natural childbirth rate was improved the most by at least 12 times per week, followed by 3-5 times per week, 6-8 times per week, and 9-11 times per week. In terms of the intensity of physical exercise, the natural childbirth rate was improved the most by moderate intensity of physical exercise. With regard to the duration of physical exercise, the natural childbirth rate was improved the most by physical exercise lasting for 30-<50 minutes per time, followed by less than 30 minutes per time, and at least 50 minutes per time. In respect to the number of weeks of conducting physical exercise, the natural childbirth rate was improved the most by physical exercise lasting for 8 or less than 8 consecutive weeks, followed by l7-24 weeks, 9-16 weeks, and 25-34 weeks. Funnel plot assessing publication bias of the studies was basically symmetrical in shape, indicating that the publication bias was generally weak. The overall quality of evidence concerning improvements in natural childbirth rate by physical exercise interventions in the studies was rated moderate.

The results show that proper physical exercise during pregnancy favorably affects the probability of natural childbirth. The natural childbirth probability could be positively influenced by some types of moderate-intensity physical exercise interventions, such as pelvic floor muscle training, general sports lessons, gymnastics, aerobic exercise, birthing ball movement, and yoga, if they are performed at least 12 times per week (30-<50 minutes each time) within 8 consecutive weeks after 24 weeks of gestation.