What do medical sources say about Exercise during pregnancy?

1st Trimester: MotherToBaby and ACOG support moderate activity during uncomplicated pregnancies, and the CDC gives similar guidance. A 2019 review found very-low-quality evidence of no overall increase in miscarriage odds, but a large 2007 cohort linked higher-volume or high-impact exercise early in pregnancy with miscarriage; its authors noted possible recall bias and said public-health conclusions were premature. Another 2019 review found no statistically significant rise in birth defects, but most studies measured activity after 12 weeks, so first-trimester evidence is limited. The 2017 PAMELA trial began exercise at 16 to 20 weeks and found no difference in preterm birth or measured newborn outcomes, so it does not settle early-pregnancy questions. MotherToBaby says people starting exercise can build up gradually, and those with complications may need individual guidance. Ask your provider about your situation. 2nd Trimester: MotherToBaby, ACOG, and CDC support moderate activity for uncomplicated pregnancies. A 2019 review found too little evidence to decide whether exercise while lying flat on the back should be avoided; it reported abnormal fetal heart-rate patterns in some acute sessions but found no low-birth-weight association in three trials. The CDC says to try to avoid this position after the first trimester, MotherToBaby advises avoiding it in the second and third trimesters, and ACOG says to avoid it as much as possible. The 2007 cohort did not find a miscarriage association after 18 weeks; its earlier finding concerned higher-volume or high-impact activity and may reflect recall bias. MotherToBaby also advises adapting activity for overheating, exhaustion, falls, and pregnancy complications. Ask your provider about your situation. 3rd Trimester: MotherToBaby and ACOG support moderate activity for uncomplicated pregnancies, with changes as balance shifts. The CDC says to try to avoid lying flat on the back after the first trimester, MotherToBaby advises avoiding it in the second and third trimesters, and ACOG says to avoid it as much as possible. MotherToBaby reports one study of moderate to high intensity exercise later in pregnancy that found smaller, leaner babies still within the normal size range. The underlying trial involved 75 regular exercisers and tested high exercise volume at moderate intensity, not vigorous activity in general. A 2018 review found exercise-only trials did not increase the odds of growth-restricted, preterm, or low-birth-weight babies; MotherToBaby and ACOG also advise adapting activity for heat, falls, and medical complications. Ask your provider about your situation.

What do medical sources say about Exercise while breastfeeding?

MotherToBaby says moderate exercise is not expected to affect breastfeeding and most women make the same amount of milk whether they exercise or not. A randomized study of 33 lactating women found no difference in milk volume, milk composition, or infant weight gain between exercise and control groups. A 2025 systematic review of postpartum activity likewise found no difference in milk quantity or quality, or infant growth and development. A separate 2025 study found that exercise changed some small molecules in breast milk and that some measures were linked with infant body composition; it did not establish a health effect or cause. The UK Chief Medical Officers say activity had no negative impact on breastfeeding; MotherToBaby advises hydration and says postpartum timing depends on recovery, health, and delivery. Ask your provider about your situation.

Where do sources disagree on Exercise in pregnancy or breastfeeding?

Miscarriage and high-volume or high-impact exercise early in pregnancy MotherToBaby, ACOG and CDC support moderate activity in uncomplicated pregnancies. The 2019 systematic review found very-low-quality evidence of no overall increase in miscarriage odds. The 2007 Danish cohort reported an association between higher weekly exercise, especially high-impact activity, and miscarriage in early pregnancy. The association weakened when limited to prospectively collected exposure reports, and the authors noted recall-bias concerns. The cohort included interviews after miscarriage, had limited direct information about intensity and analyzed a specific early-pregnancy, higher-volume or high-impact exposure. The review pooled different studies and found very-low-quality evidence of no overall increase. Neither finding proves cause or answers every exercise dose. The cohort is a concern specific to higher-volume or high-impact activity early in pregnancy, but it cannot establish causation. The pooled review and current guidance support moderate activity generally; the evidence should not be described as showing that all exercise raises miscarriage risk.

Back to search

Exercise

Everyday exposureHeat & activity

Also called · 7 names▾

physical activity, workout, prenatal exercise, pregnancy workout, walking while pregnant, prenatal workouts, working out during pregnancy

Physical activity

General information · effects depend on amount, timing, and how it is used. About these sources

Pregnancy

A 2007 prospective Danish cohort with retrospective case interviews included 92,671 pregnancies and 3,187 miscarriages. Exercising 1 to 44 minutes per week was not associated with increased miscarriage risk. In analyses using the total dataset, the largest estimate was for more than 419 minutes per week at 11 to 14 weeks (HR 3.7, 95% CI 2.9 to 4.7). The association weakened when exposure reports collected before miscarriage were analyzed separately; high-impact activity for 75 to 269 minutes per week had HR 1.8 (95% CI 1.0 to 3.6) at 11 to 14 weeks. The authors noted possible recall bias, lack of a direct exercise-intensity measure and said the result was too early for public-health conclusions. A 2019 systematic review included 46 studies and 266,778 participants. It found very-low-quality evidence of no increased overall miscarriage odds (23 studies, 7,125 women; OR 0.88, 95% CI 0.63 to 1.21) or perinatal mortality odds. Most included studies examined moderate activity up to 60 minutes; the review could not comment on longer exercise periods. A separate 2019 systematic review included 14 studies and 78,735 participants. It reported no statistically significant increase in congenital-anomaly odds (OR 1.23, 95% CI 0.77 to 1.95), but rated the evidence very low quality and noted most studies measured exercise after 12 weeks. Evidence specific to first-trimester exercise and congenital anomalies remains limited. A 2018 review included 135 studies and 166,094 participants. Exercise-only randomized trials did not increase the odds of growth-restricted, preterm or low-birth-weight babies. Fifteen trials with 3,670 participants found lower odds of birth weight above 4,000 g (OR 0.61, 95% CI 0.41 to 0.92). Clapp's 2002 randomized study included 75 regular exercisers. The high-volume group used moderate-intensity, weight-bearing exercise in mid and late pregnancy; mean newborn weight was 3.39 kg versus 3.81 kg in the group that reduced exercise volume, and newborn body fat was 8.3% versus 12.1%. MotherToBaby reports that the babies remained within the normal size range. The study does not test vigorous exercise as a general category. The 2017 PAMELA trial randomized 639 pregnant participants; 594 were included in the intention-to-treat analysis. Its supervised exercise program began at 16 to 20 weeks and ran through 32 to 36 weeks. It found no significant group differences in preterm birth or measured newborn outcomes; it does not answer first-trimester questions. A 2024 cohort of 2,062 mother-neonate pairs found no significant association between meeting exercise guidelines and SGA or LGA, although some total-activity measures were associated with growth outcomes. A 2025 trial-quality review of 40 exercise-only RCTs found only five with no or negligible deviation from intention-to-treat. It found no evidence of improved pregnancy outcomes or lower gestational weight gain when considering only the most rigorous trials. A 2026 review of 88 RCTs reported lower gestational diabetes, hypertensive disorders, cesarean birth and gestational weight gain, but no significant effect on birth weight or preterm birth; most trials had some concerns about bias. These reviews differ in trial inclusion and quality criteria. A supervised 2026 RCT assessed 400 participants and analyzed 256 in a per-protocol analysis. It reported no increased miscarriage risk, but its size and design do not establish that rare outcomes cannot occur. A 2026 meta-analysis of 34 observational studies found associations between higher activity and lower excessive gestational weight gain and LGA; observational data do not establish cause. A 2026 cohort of 323 reported a sex-specific association with female infant body fat, without an association for male infants; this finding has not been independently replicated.

Breastfeeding

A 1994 randomized trial included 33 sedentary lactating women, assigned to exercise (18) or control (15) beginning 6 to 8 weeks after birth. Supervised aerobic exercise was 45 minutes per day, 5 days per week, for 12 weeks. The study found no significant group differences in milk volume or composition, infant weight gain or maternal prolactin response. A 2025 systematic review and meta-analysis included 46 unique studies and 8,766 participants. It found no difference between exercise and control groups in breast-milk quantity or quality, infant growth or development, or maternal injury. Its literature search ended Jan. 12, 2024. A structured review published in November 2025 searched PubMed and Scopus through December 2024 and concluded that moderate exercise did not adversely affect milk volume, nutrient composition or infant growth. This review predates the later metabolome study. A 2025 study evaluated acute exercise in 15 mother-infant dyads and habitual activity in 84 active and 35 inactive participants within a 348-person cohort. The abstract reports changes in some breast-milk metabolites and correlations with infant adiposity measures. The accessible abstract does not establish that exercise caused a clinical infant effect, and it did not measure milk supply as its main outcome.

Miscarriage and high-volume or high-impact exercise early in pregnancy

Source A says: MotherToBaby, ACOG and CDC support moderate activity in uncomplicated pregnancies. The 2019 systematic review found very-low-quality evidence of no overall increase in miscarriage odds. (MotherToBaby: Exercise fact sheet, ACOG Exercise During Pregnancy FAQ, CDC: Pregnant and Postpartum Activity, Prenatal exercise and fetal mortality: systematic review and meta-analysis)
Source B says: The 2007 Danish cohort reported an association between higher weekly exercise, especially high-impact activity, and miscarriage in early pregnancy. The association weakened when limited to prospectively collected exposure reports, and the authors noted recall-bias concerns. (Madsen et al.: Leisure time exercise and miscarriage in the Danish National Birth Cohort)
Why they differ: The cohort included interviews after miscarriage, had limited direct information about intensity and analyzed a specific early-pregnancy, higher-volume or high-impact exposure. The review pooled different studies and found very-low-quality evidence of no overall increase. Neither finding proves cause or answers every exercise dose.
Other sources: A 2026 supervised moderate-intensity RCT reported no increased miscarriage risk, but it analyzed 256 participants and used a per-protocol design. (Supervised multicomponent exercise in pregnancy: randomized trial)
What the evidence shows overall: The cohort is a concern specific to higher-volume or high-impact activity early in pregnancy, but it cannot establish causation. The pooled review and current guidance support moderate activity generally; the evidence should not be described as showing that all exercise raises miscarriage risk.

Exercise while lying flat on the back

Source A says: The CDC says to try to avoid exercise lying flat on the back after the first trimester; MotherToBaby advises avoiding it in the second and third trimesters; ACOG says to avoid it as much as possible. (MotherToBaby: Exercise fact sheet, ACOG Exercise During Pregnancy FAQ, CDC: Pregnant and Postpartum Activity)
Source B says: The 2019 systematic review found insufficient evidence to determine whether supine exercise should be avoided. It reported abnormal fetal-heart-rate patterns during some acute sessions, while three trials found no association with low birth weight. (Supine exercise and maternal/fetal outcomes: systematic review)
Why they differ: The guidelines use a precaution related to possible blood-flow changes. The review found too little evidence on clinical outcomes to settle the question; short-term fetal-heart-rate changes are not the same outcome as newborn illness or growth restriction.
Other sources: The UK CMO guidance supports activity during pregnancy overall and says activity choices should reflect pre-pregnancy activity levels; it does not resolve the specific supine-exercise evidence question. (UK Chief Medical Officers: Physical activity guidelines)
What the evidence shows overall: The position advice is a precaution specific to exercise lying flat on the back. The review does not show that exercise overall causes harm, and guidance continues to support moderate activity in uncomplicated pregnancies.

How strongly prenatal exercise interventions improve clinical outcomes

Source A says: ACOG, CDC and the UK CMO describe potential maternal health benefits. A 2026 review of 88 RCTs reported lower gestational diabetes, hypertensive disorders, cesarean birth and gestational weight gain. (ACOG Committee Opinion No. 804, CDC: Pregnant and Postpartum Activity, UK Chief Medical Officers: Physical activity guidelines, Prenatal exercise interventions and pregnancy outcomes: 88-RCT review)
Source B says: A 2025 review focused on trial quality found no evidence of improved pregnancy outcomes or lower gestational weight gain when restricted to the most rigorous exercise-only trials. (Antenatal activity interventions and trial quality: systematic review)
Why they differ: The reviews used different search cutoffs, eligibility rules and approaches to trial bias. The 2025 analysis prioritized intention-to-treat fidelity; the 2026 review included a broader RCT set and reported that most trials had some concerns about bias. Guidelines may also draw on observational evidence and outcomes beyond those pooled in one review.
What the evidence shows overall: Exercise guidance is well established, but the size and certainty of specific clinical benefits depend on study design and outcome. Describe potential benefits without promising that exercise will prevent a complication.

Last reviewed 2026-10-16

Independent check 2026-10-16

Not yet clinician-reviewed

Ask your provider about your situation.

Hot Tubs & Saunas

Exercise

Exercise — Breastfeeding

What the research shows

What's in it / what was tested

Sources disagree

Options sources describe

Beyond pregnancy

All sources

Related