Cardiovascular disease (CVD) remains the number one health threat in women [1]. Adverse pregnancy outcomes (APOs), such as hypertensive disorders of pregnancy, gestational diabetes, and preterm birth, affect up to 20% of pregnant women and are recognized as major risk factors for latent and future CVD [2,3]. Though research to disentangle the contributions of prepregnancy risk factors from APOs on future CVD risk is ongoing, interventions that prevent APOs as a target for reducing CVD offer immediate and possibly long-term benefits. Effective preventative therapies, however, remain elusive [4]. Moderate- to vigorous-intensity physical activity reduces APO risk [5,6], but less than 1 in 4 pregnant women achieve guidelines [7]. This likely reflects that pregnant women report unique exercise barriers such as fatigue, pain, medical restriction, and concern for the baby [8].Therefore, more feasible lifestyle strategies to prevent APOs are needed.

One such strategy could be reducing sedentary behavior (SED), which is defined as low-intensity behavior that occurs in a seated, reclined, or lying posture [9]. Considering barriers to more intensive physical activity during pregnancy, replacing SED across the day with standing and light-intensity physical activities may be more achievable. Excessive SED has been related to an increased risk of CVD and mortality in general populations, even after accounting for moderate- to vigorous-intensity physical activity [10]. Though sparse data on SED during pregnancy exist, our laboratory recently completed a small cohort study among 120 pregnant women that objectively measured SED and physical activity during each trimester [11]. We found that, on average, pregnant women had approximately 9.5 hours per day of SED, which is higher than the general population [12]. Further, using trajectory analysis, women with a pattern of very high SED across pregnancy (approximately 11 hours per day) versus low SED (approximately 8 hours per day) had 6.76 (95% CI 1.20-38.14) higher odds of experiencing an APO [11]. Of interest, women in the low and medium SED groups (approximately 8 and 9 hours per day, respectively) had no excess risk of APO, indicated a possible threshold effect. Similar threshold effects were observed for time spent standing and in the number of steps per day such that only women in the very low trajectory groups had excess risk. Importantly, moderate- to vigorous-intensity activity was not associated with APO outcomes, bolstering support for a strategy of reducing SED by increasing standing and daily steps as a novel intervention target to improve pregnancy health. Yet, few randomized clinical trials (RCTs) testing SED reduction interventions are available, especially in pregnant women.

The Sedentary Behavior Reduction in Pregnancy Intervention (SPRING) pilot and feasibility study, funded by the American Heart Association (20TPA3549099), seeks to address these research gaps. SPRING is an RCT that tests the feasibility, acceptability, and preliminary effects of a novel intervention designed to reduce SED and improve all-day activity patterns among pregnant women at high risk for APOs.

The SPRING study’s specific aims are to (1) evaluate the effect of SPRING versus control on objectively measured SED (primary outcome, hypothesized to decrease), standing (hypothesized to increase), and steps per day (hypothesized to increase); (2) evaluate the feasibility and acceptability of SPRING, including recruitment, retention, outcome assessment rates, intervention fidelity, and program evaluation; and (3) estimate the preliminary effects of SPRING versus control on maternal-fetal health outcomes including composite APOs (hypertensive disorders of pregnancy, gestational diabetes, and preterm birth), maternal blood pressure and glucose, psychosocial outcomes, gestational age at birth, and infant birth anthropometrics.

In the absence of established SED-reduction interventions or guidelines in pregnant women, a pilot and feasibility RCT among pregnant women with higher risk for SED and APOs was chosen as the first logical step to determine if the intervention could successfully reduce SED by increasing standing and steps per day across pregnancy. Figure 1 describes the timeline of study assessments and intervention procedures (SPIRIT [Standard Protocol Items: Recommendations for Intervention Trials] checklist [13] available in Multimedia Appendix 1). The SPRING pilot and feasibility trial is registered on ClincialTrials.gov (NCT05093842).

Potential participants complete baseline assessments and begin the 1-week baseline activity monitoring protocol between 10 weeks 0 days (10⁰ weeks) and 12 weeks 6 days (12⁶ weeks) of gestation. Upon receipt of valid activity monitoring data with a completed time-use diary and clearance to participate from the participant’s prenatal care provider, pregnant women are randomized in a 2:1 ratio to either a multicomponent SED-reduction intervention or a usual care control group. Beginning at approximately 14 weeks of gestation, participants randomized to the intervention group begin the active intervention, which lasts through 38 weeks of gestation or until the participant gives birth, whichever occurs first. Participants randomized to the usual care control arm do not receive the intervention but are mailed a Fitbit Luxe after giving birth. Participants in both groups complete assessments in each trimester and are remunerated US $40 for completing the first trimester (baseline) assessment between 10⁰ and 12⁶ weeks, US $50 for completing the second-trimester assessment between 20⁰ and 22⁶ weeks of gestation, and US $60 for completing the third-trimester assessment between 32⁰ and 34⁶ weeks of gestation. After participants give birth, maternal-fetal outcomes are abstracted from medical records by trained clinical staff.

All research procedures were approved by the University of Pittsburgh Human Research Protection Office (STUDY20110193).

Eligibility criteria for the SPRING Study are listed in Textbox 1. We recruit participants in their first trimester of pregnancy who are at risk for high levels of SED and who have at least 1 risk factor for APO. These criteria reflect an early gestational age when we can intervene meaningfully across pregnancy and higher SED appropriate for modification (ie, reduction) to reach the levels that we have found to be associated with better pregnancy outcomes [11]. The criteria of having at least 1 risk factor for APO relate to our long-term goal to test whether our intervention can reduce APO risk; therefore, we sought to evaluate feasibility and preliminary effects in this specific population. Lastly, to allow for our exploratory outcome assessment using prenatal records of maternal-fetal outcomes, participants are required to consent to abstraction either through our affiliation with the University of Pittsburgh Medical Center or through separate medical record release if they are receiving prenatal care elsewhere.

Participants younger than 18 years or older than 45 years are excluded to avoid an inability to provide consent and more complicated pregnancies with advanced age, respectively. Those with preexisting hypertension or diabetes are excluded because existing management of these conditions could affect our study outcomes (eg, prenatal blood pressure and glucose test results). Any other contraindications to exercise during pregnancy or severe mobility limitation are also exclusionary since the intervention asks participants to increase standing and daily steps. We also require participants to receive clearance for participation in the intervention from their prenatal care provider as an additional safety precaution. Lastly, those participating in other interventional research that could confound our study outcomes are excluded to reduce bias or contamination.

A randomization ratio of 2:1 (2 intervention to 1 control) was chosen to provide more robust data to evaluate the feasibility and acceptability of our intervention and assessment protocols [16]. To help achieve balance over time, blocks of 6 were selected, and 9 blocks of computer-generated, randomly ordered sets were placed sequentially into sealed, numbered envelopes by a blinded study personnel. These envelopes were then provided to the unblinded study personnel (principal investigator and interventionist) and kept in a secure location.

Participant randomization begins with the blinded study coordinator notifying the principal investigator that a participant is eligible and has completed all baseline assessment procedures. The study investigator then opens the next sequential envelope and records the randomization assignment and date in the randomization log. The principal investigator then contacts the participant to inform her regarding her randomization assignment, describe next steps, ensure that the participant comprehends that the intervention and assessment study procedures are separate, and reinforce the importance of continued participation in either group.

Study outcome assessments are conducted by trained, blinded study personnel and occur via web-based videoconference once per trimester of pregnancy (Table 1). Any necessary assessment materials are mailed to participants prior to study visits.

SED and physical activity are measured in each trimester. Objective assessment of activity patterns is conducted using a thigh-mounted activPAL3 monitor (PAL Technologies, Glasgow) worn for 24 hours over 8 days. This protocol is the best-practice methodology for field monitoring of SED and standing [17], and we have shown it to be a valid measure of moderate- to vigorous-intensity physical activity in pregnant women [18]. Participants wear the monitor on the anterior upper thigh for 8 days, with removal only for swimming activities. Concurrently, participants are asked to record periods of work, nonwork, sleep, and nonwear in a diary for accurate data processing. Participants are verbally instructed regarding procedures for correct monitor wear at each virtual assessment visit, including visual verification of correct placement. Detailed written instructions on monitor wear and diary completion are also provided. The activPAL3 24-hour event data are downloaded and cleaned by removing periods of sleep and nonwear; remaining valid wear time is used to calculate average daily durations of time spent in SED (total and in prolonged bouts of at least 30 and 60 minutes), standing, and stepping. Daily frequency of steps and sit-stand transitions are also averaged across days. Lastly, to estimate moderate- to vigorous-intensity physical activity, time with a stepping frequency of at least 75 steps per minute and 100 steps per minute are averaged across valid wear days from daily summary data. Averages are calculated overall and during working times (if applicable). Data are considered valid with at least 5 days of at least 10 hours of waking wear time recorded [19].

Self-reported SED and physical activity are assessed using the Pregnancy Physical Activity Questionnaire [20], with questions regarding proportion of time spent in SED during work and nonworking times added to improve the accuracy of self-reported SED assessment [21].

Blood pressure is assessed virtually during study assessment visits in each trimester using a validated oscillometric monitor [22]. Either a UA-611 monitor (A&D Medical) that was provided as part of prenatal care at the University of Pittsburgh Magee-Womens Hospital or a study-provided BP7250 monitor (Omron Healthcare Inc) is used. Participants use the same monitor and cuff for all visits. During the virtual assessment, participants first provide verbal confirmation of abstention from caffeine and nicotine for 1 hour prior. Then, participants place the cuff on the left upper arm and complete a supervised 5-minute quiet rest with their arm supported at chest level, back supported against a chair, and feet flat on the floor or other support with ankles and knees uncrossed. After the rest, the participant initiates the measurement by pressing the start button with the blood pressure monitor facing the video camera and visible only to the assessor (to reduce participant reactivity). The assessor leads the participant through 3 measurements, taken with a 1-minute rest in between, and records blood pressures and heart rates in the database prior to informing the participant of all values. The average of the second and third measurements is used for analysis.

Demographics, medical history, reproductive history, and medication use are collected at baseline using a standardized form. Anthropometry is assessed at baseline via self-reported height and prepregnancy weight, which are subsequently verified through medical record abstraction (described later). To capture changes that occurred following randomization, updates to demographics (eg, new employment) are reported in a questionnaire and changes to medical history or medication use (eg, a new diagnosis or starting, stopping, or changing dosage of medications) are assessed during an interval medical history review at each follow-up visit.

Dietary intake is assessed at each visit using the Diet Screening Questionnaire, a semiquantitative questionnaire that measures typical dietary habits [23]. Psychosocial behavioral targets and outcomes are measured at baseline and each follow-up visit via web-based self-report questionnaires. Validated questionnaires assessing behavioral targets of the intervention include the Sitting during Pregnancy Barriers and Outcome Expectations Questionnaire [24], Expected Outcomes and Barriers for Habitual Physical Activity [25], the Multidimensional Scale of Perceived Social Support [26], and Social Support for Exercise [27]. Validated questionnaires assessing possible psychosocial effects of the intervention include the Profile of Mood States [28], the Centers for Epidemiologic Studies Depression Scale (CES-D) [29], the Perceived Stress Scale [30], Health-Related Quality of Life for Nausea and Vomiting during Pregnancy [31], and the Pittsburgh Sleep Quality Index [32].

Adverse events are collected systematically at follow-up assessment visits by blinded study personnel. Details of new or worsening medical conditions, including pregnancy-related adverse events, are recorded and classified regarding severity and relationship. In addition, ongoing assessment of adverse events in the intervention group is conducted by intervention staff. All adverse events are reviewed by the principal investigator and, if needed, the study physician for accuracy, possible relationship to the intervention, and to make any necessary modifications or terminate participation in the intervention.

Pregnancy health and maternal-fetal outcomes are abstracted from the participant’s medical record of prenatal visits and labor and delivery notes. Maternal outcomes extracted include gestational weight gain, clinic-measured blood pressures, pregnancy complications including APOs, glucose screening values, and tolerance tests (if available), and labor and delivery details. Fetal outcomes include gestational age at birth, anthropometrics, and complications. All abstractions are reviewed by a second reviewer and discrepancies are resolved. The study’s maternal-fetal medicine physician coinvestigator (AH) confirms all APOs and a random sample of healthy pregnancies for accuracy.

Feasibility and acceptability are also assessed through a variety of methods across the study period (Table 2). Screening outcomes are recorded prospectively including recruitment sources and yield, eligibility rates and ineligibility reasons, and rates of enrollment. We also administer a study-developed questionnaire at baseline (prior to randomization) to assess participant expectations, motivations, and experiences when enrolling in the SPRING RCT. Rates of valid assessment completion are also prospectively recorded. Intervention fidelity with respect to delivery, receipt, and enactment [33] is assessed at each intervention contact (see Intervention Content and Schedule section). A final web-based program evaluation assesses acceptability and includes 22 short-answer, ranking, and multiple-choice questions. The evaluation is distributed after the last intervention lesson to intervention participants only and aims to evaluate participants’ perceptions regarding their health, usefulness of the intervention components, general satisfaction and acceptability of the intervention, and suggestions for improvement.