Quantitative activity levels and gestational age at delivery: a prospective cohort study among nulliparous women

Whitney A. Booker, Etoroabasi E. Ekpe, Roxane C. Handal-Orefice, Yuan Zhang, Ananth Cande, Cynthia Gyamfi-Bannerman, Vanessa Nieto

Research output: Contribution to journalLetterpeer-review

1 Scopus citations


OBJECTIVE: Despite the knowledge that bed rest does not reduce the risk of preterm birth (PTB), it continues to be recommended by many providers worldwide. This is because there are no quantitative data assessing the relationship between PTB and physical activity in pregnancy.1–3 We designed a prospective cohort study using a Fitbit activity tracker to quantitatively explore the association between baseline physical activity in pregnancy in steps/day and the risk of PTB (<37 weeks). STUDY DESIGN: This was a prospective cohort study assessing the association between the risk of PTB and physical activity in healthy nulliparous women from 10 to 24 weeks to delivery. The physical activity (San Francisco, California) was measured from the time of entry into the study until the day before admission for delivery using the Fitbit Flex 2. The participants wore the faceless device 24/7 without modifying their activity. The primary exposure was steps/day in low- (<5000 steps/d) and high-level (≥5000 steps/d) activity groups. The primary outcome was the rate of PTB (<37 weeks). An additional unplanned secondary analysis was performed using a 3500 steps/d cutoff. The secondary outcomes included peripartum complications and median steps/day in term vs preterm groups. Adjusted analyses were performed to account for possible confounders. RESULTS: A total of 134 women were enrolled, of which 25 (19%) and 109 (81%) were in the low- and high-level activity groups, respectively. Overall, 11 (8.2%) women delivered preterm. The high-level activity group was older, partnered, employed, and had a higher education level. The PTB did not differ between the groups (adjusted risk ratio, 0.99; 95% confidence interval [CI], 0.99–1.00) (Table). There was no difference in the median steps/d between preterm and term deliveries (7767 interquartile range, [5188–10,387] vs 6986 [5412–8528]); percentile difference, −442; (95% CI, −2233 to 1507) steps. Using a 3500 steps/d cutoff, there was a 75% reduction in the PTB risk (29% vs 7%, respectively; risk ratio, 0.25; 95% CI, 0.05–2.35) (Table). CONCLUSION: This prospective study of nulliparous women showed no difference in the risk of PTB between low- vs high-activity groups using a cutoff of 5000 steps/d. The gestational age at delivery was similar between the groups. No significant difference in the number of steps/d was observed between women who delivered preterm compared with term. The women who were prescribed activity restriction (AR) had a marked reduction in their median number of steps/d after AR was prescribed. However, their median number of steps per day (>5000) reflected that they remained active despite this instruction. An additional analysis using 3500 steps/d as a cutoff for exposure groups showed a significantly increased risk of PTB in the <3500 steps/d group than the ≥3500 steps/d group. It is therefore plausible that activity levels <3500 steps/d are associated with an increased risk of PTB.

Original languageEnglish
Article number100503
JournalAmerican Journal of Obstetrics and Gynecology MFM
Issue number1
StatePublished - Jan 2022
Externally publishedYes


Dive into the research topics of 'Quantitative activity levels and gestational age at delivery: a prospective cohort study among nulliparous women'. Together they form a unique fingerprint.

Cite this