Participants
Study setting {9}. The study includes two clinical sites. One site is at California Polytechnic State University, San Luis Obispo, California (S. Phelan, PI), and the other site is at Brown University & the Miriam Hospital in Providence, Rhode Island (R. Wing, PI).
Source population. GDM recurrence rates are higher in Hispanic and African-American populations;57 thus, our targeted recruitment plan includes 35% Hispanic, 9% African American, 6% Asian, and 50% non-Hispanic white at each site. The geographical regions and recruitment clinic populations selected for this study have a high prevalence of obesity (>35%) and client diversity (35-45% Hispanic; 9% African American).
Recruitment {15}. To reach the target sample size of 252, both sites recruit participants through direct and indirect methods and via administrative databases. As described in Table 1 direct recruitment methods include clinic staff and research assistants at the recruitment clinics providing information about the study at the time of prenatal or postnatal visits for patients with GDM or prior GDM. Indirect methods include presentations in healthcare settings that interact with mothers (e.g., Ob/Gyn, pediatrician, WIC offices). Administrative databases are also used to identify and offer the program to women with history of GDM who might not regularly engage with the targeted healthcare settings after having a baby.
Eligibility criteria {10}. Table 2 describes the eligibility and exclusion criteria for this trial. Participants must have physician documentation of GDM during any prior pregnancy. Given the diversity of clinically acceptable methods used to diagnose GDM,104-107 several diagnostic methods for prior GDM are eligible. Acceptable documentation to confirm prior GDM are as follows: 1) a 3-hour 100 gram oral glucose tolerance test (OGTT) performed at > 20 weeks’ gestation in which 1 or more values exceeded the Carpenter and Coustan criteria108 (i.e., fasting >95 mg/dL; 1 hour > 180 mg/dL; 2 hour > 155 mg/dL or 3 hour > 140 mg/dL); 2) a 75 gram OGTT performed at > 20 weeks’ gestation and 1 or more values exceeded International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria 109 (i.e., fasting > 92, 1 hour > 180; 2 our >153); 3) a 1-hour 50 gram test performed at any time during pregnancy with a value > 185 mg/dL (if ≥ 130 mg/dL but < 185 mg/dL and clinic did not do a follow-up 100 g OGTT, participant would be ineligible);4) a fasting glucose value prior to 20 weeks’ gestation was > 92 mg/dL and < 125 mg/dL and treatment with medication or insulin; or, 5) an HbA1c conducted anytime during pregnancy with a value > 5.6% and the treatment with medication or insulin.
Women must also report chances of having a baby in the next 1-3years; this is determined based on a response to the question, “On a scale of 0-10, what are the chances you see yourself ever having any more children?” Participants reporting > 1 on this scale and who report plans for pregnancy within the study’s timeframe (1-3 years depending on study enrollment year) are considered eligible. Other eligibility criteria include BMI > 25 kg/m2 ; age > 18 years; and English or Spanish speaking. Breastfeeding women are eligible to enroll, as moderate weight loss does not appear to adversely affect lactation.110-114 At study enrollment, participants are encouraged to use medically proven forms of contraception until completion of the initial 16-weeks of intervention, but this is not an eligibility requirement.
Women with Type 2 or Type 1 diabetes are excluded; lack of diabetes is confirmed prior to randomization with an HbA1c test (> 6.5%). Women with a family history of diabetes or with impaired glucose tolerance (“prediabetes”; HbA1c of 5.7-6.4%) may be at increased risk of GDM115 but are included because weight loss could potentially still modify risk of GDM. Other exclusions include: Age < 18 years; current pregnancy; tubal ligation; semi-permanent form of birth control with no plans for removal (e.g., hormonal progesterone intrauterine device or hormonal contraceptive implant); relocating in the next 2 years, medications that affect weight/diabetes (e.g., oral corticosteroid and metformin), serious current physical disease (e.g., heart disease, cancer, renal disease) for which physician supervision of diet and exercise prescription is needed, orthopedic problems that limit the ability to exercise,116 problems with drug abuse and/or symptoms of an eating disorders,117 history or plans of bariatric surgery, and hospitalization for depression or psychological problems in the past year. Also, women who do not show to orientation visit and fail to reschedule are excluded.
Enrollment process and consenting. Figure 1 shows the process of study enrollment - from recruitment to randomization. Women who appear interested in the program are screened by phone. If still eligible after phone screening, patients are asked to attend an orientation and consenting visit followed by their baseline assessment visit.
Who will take informed consent? {26a} Trained study staff collect informed consent.
Additional consent provisions for collection and use of participant data and biological specimens {26b} The process of informed consent includes discussion of an option to allow collection and storing of additional biospecimens for future research that may include analyses on genetic material.
Intervention descriptions {11a}
Explanation for the choice of comparators {6b} Two groups are compared in this study. Group 1 is a standard care plus education control group and was selected because the intervention provides a level of care that is consistent with the typically minimal amount of lifestyle counseling received by women before pregnancy and also is not expected to promote clinically significant weight loss. Group 2 is standard care plus education plus weight loss intervention and was selected in order to isolate the effects of pre-conception weight loss on GDM recurrence and other health outcomes.
Group 1: Standard Care + Education Women in this condition receive usual medical care before and during pregnancy and throughout the trial. Also, these women meet with a study interventionist for 20-minute individual sessions at study entry and again after 16 weeks. The first meeting at study entry encourages women to spend the next 16 weeks improving overall health before pregnancy and reviews nutrition (e.g., consuming multivitamins, folic acid) and physical activity recommendations. The second face-to-face meeting occurs after 16 weeks and focuses on managing stress and also includes a “Pregnancy Primer.” Since all women in the study have expressed a desire to have another pregnancy within 1-3 years, this module provides participants with standard information on methods to track ovulation. Participants also receive information on recommended amount of weight gain118 during pregnancy. Throughout the study, women receive quarterly study newsletters with study updates and general information about preconception health and wellness.
Group 2: Standard Care + Education + Weight Loss Intervention
Overview: This group receives all aspects of Group 1 plus a standard lifestyle modification program implemented to induce >10% weight loss over 16 weeks and promote weight loss maintenance over subsequent months (12-36 months depending on enrollment year) until conception. This comprehensive, individually focused, SCT-based weight control program includes education, behavioral self-regulatory strategies, ongoing contact, feedback, and social support. The intervention is based on the DPP and Look Ahead lifestyle interventions,119, 120 which have been proven effective in promoting significant weight loss and maintenance in multiethnic, English and Spanish speaking individuals across the country.119, 120 The intervention provides guidance and resources for English and Spanish-speaking individuals from a variety of different cultures and backgrounds.
Format and contact: The intervention focuses on ongoing, individual contact with a study interventionist to promote weight loss prior to conception. Visits may be conducted in person, on the phone, or through video conferencing. For the first 16 weeks, participants meet weekly for ~30 minutes. Thereafter, participants meet bi-weekly (or more frequently in the context of weight regain) to maintain weight loss until conception. After conception, intervention contacts are discontinued.
Weight Loss Goals: Participants are given a scale and told to aim for a weight loss of 1-2 lbs per week for the first 16 weeks. Patients desiring to lose more weight during the program are encouraged to do so, provided they maintain reasonable eating and activity patterns and do not reduce below normal weight. After 16 weeks, participants may work on weight loss maintenance or continue to lose weight at a moderate rate (1-2 pounds/week) until confirmed conception.
Dietary Goals: Participants are instructed to follow a standard calorie restriction diet used in lifestyle modification programs. Calorie goals are based on study entry weight with 300 calorie/d adjustments for breastfeeding status, if applicable. Participants with an entry weight < 91 kg are prescribed a 1200 kcal/day, self-selected diet, and those with entry weight > 91 kg are prescribed a 1500 kcal/day. A standard, low fat diet is prescribed (35% fat, 20% protein, 45% CHO), since prior research has suggested that recurrence of GDM was greater in women who consumed more fat between pregnancies.121 After the 16 week program, dietary goals may be adjusted to help women maintain their weight loss or weight maintenance goal until conception.
Exercise Goals: Participants are instructed to increase their physical activity to at least 150 minutes per week during the initial 16-week program (e.g., 30 minutes per day, 5 days per week). Thereafter, they are advised of higher goals (60 min/day) to promote long-term weight loss maintenance. Brisk walking, “child-friendly,” and inexpensive activities are suggested, taking into consideration potentially unsafe neighborhoods. Participants are provided with a pedometer and encouraged to gradually increase the number of steps they walk per day (with an increase of ~250 steps/day each week) until reaching an ultimate goal of about 10,000 steps per day.122
Behavioral Goals. The major features of behavior modification include self-monitoring, behavior chains, stimulus control, goal-setting, self-reinforcement, problem-solving, social assertion, cognitive strategies.123 Participants complete weekly behavioral assignments, which are reviewed by the interventionist. Participants are given self-monitoring records or encouraged to use apps, if preferred, to facilitate self-monitoring.
Criteria for discontinuing or modifying allocated interventions before conception {11b} Site physicians who are trained in obstetrics/gynecology guide decisions as to whether or not to continue intervention or assessments in women with medical difficulties. If needed, the study physicians contact a participant’s provider to discuss treatment/assessment continuation for participants. The study physicians do not provide medical care during the course of the study but refer and help participants obtain appropriate medical or psychiatric care, if needed. Rate of weight loss and caloric restriction are monitored and if any extreme and overly rapid weight losses occur, healthier practices are encouraged and adherence to these recommendations is monitored. A subset of women may be breastfeeding upon enrollment. Moderate weight loss does not appear to adversely impact breastfeeding,124, 125 but study staff are monitoring such occurrences and provide referrals, as needed. To reduce the risk of muscle soreness, muscle strain, or joint sprain, loss of balance, or trauma by falling, the physical activity (walking) is moderate, progressive, and volitional and goals may be modified, as needed for an individual participant.
Strategies to improve adherence to interventions {11c}. The behavioral weight loss program includes a variety of strategies to improve adherence. These include strategies related to appropriate goal setting, cognitive restructuring, relapse prevention, and problem solving. In addition, the study interventionist provides support for all positive behavioral changes. Weekly supervision meetings with the intervention team are designed to promote treatment fidelity. Participant cases are reviewed and strategies discussed with the intervention team to promote participant adherence to attending treatment sessions, completing food and exercise records, engaging in daily self-weighing, and following calorie and activity goals. Intervention fidelity measures (coded audiotaped sessions) are further reviewed and discussed to guide and promote adherence.
Relevant concomitant care permitted or prohibited during the trial {11d} During the trial, participants are not restricted from receiving concomitant care and/or interventions.
Provisions for post-trial care {30} There are no provisions for post-trial care.
Outcomes {12}
Participant timeline {13} Table 3 shows when participants complete the measures in Gestational Diabetes Prevention/Prevención de la Diabetes Gestacional. Bilingual (English/Spanish speaking) assessors are masked to randomization and conduct all major assessments occurring at baseline, after 16 weeks, at 26 weeks’ gestation, and at 6 weeks postpartum. These assessments occur at the study’s research centers or affiliate locations most proximal to participants. If necessary, assessments are conducted at participants’ homes. After 16 weeks, quarterly brief assessments occur until conception. The brief assessments may be conducted in person or over the phone/video conferencing.
Primary Outcome. The study’s primary outcome is GDM diagnosis in next pregnancy. The 2-step approach of diagnosing GDM is done at 24-28 weeks’ gestation and involves participants receiving as part of standard care a 50 g oral glucose solution followed by a 1-hour venous glucose determination. Participants meeting or exceeding the 1-hour screening criteria (a cutoff for an abnormal 1-hour screen of >130 mg/dL) are then referred to the study for completion of the 100-g, 3-hour diagnostic oral glucose tolerance test (OGTT). Positive diagnosis of GDM is based on the Carpenter and Coustan criteria108 based on 2 abnormal values on the 3-hour OGTT that includes a fasting value >95 mg/dL; 1 hour > 180 mg/dL, 2 hour > 155 mg/dL, 3 hours >140 mg/dL. Results of the study’s 100-g OGTT are immediately shared with participants’ medical providers who interpret and inform patients of GDM screening results. If the study measured, 100-g, 3-hour diagnostic OGTT is not obtained, provider assessments done in standard care that are based on acceptable diagnostic methods (Table 4) are used to diagnose GDM. Final determination of GDM diagnosis is done by independent evaluation of records from the study’s Ob/Gyn physician researchers who are masked to randomized group.
HbA1c tests. Diabetes is an exclusion criterion (based on HbA1c at screening >6.5%). However, after screening, annual HbA1c tests are performed until conception and test results shared with participants’ providers. This is expected to minimize early clinical screenings for pre-existing DM; however, any participant receiving early (< 24 weeks gestation) clinical diagnosis of GDM is included in analysis.
Maternal insulin resistance/ physiologic parameters Maternal fasting glucose, insulin, Homeostatic Model Assessment of Insulin Resistance (HOMA-IR), leptin, TNF-alpha, C-reactive protein, adiponectin, and lipids are measured by trained staff at each research site, following establish protocols. Blood draws are scheduled 12-24 hours after the most recent bout of exercise and after an overnight fast. HOMA-IR is used to estimate insulin resistance. Systolic and diastolic blood pressure are measured using a standard mercury manometer and appropriate size cuffs with participants in the sitting position with both feet on the ground. After resting 5 minutes, the average of two measurements are recorded, with a 1-2 minute interval between measures.
Maternal anthropometrics Weight is measured to the nearest 0.1 kg using calibrated standard digital scales. Two measures are completed with participants measured in light clothing (without shoes). Scale calibration is checked weekly with known weights. Standing height is measured twice in patients without shoes in millimeters with a wall-mounted Harpenden stadiometers. Given emerging evidence of a relationship between abdominal fat and GDM,126 waist circumference is measured over bare skin or underwear using a tape measure, and following standardized protocols.
Lifestyle behaviors are measured to examine treatment effects and relationships with maternal physiology and GDM recurrence.Physical Activity is measured for 7 days using the Actigraph accelerometer (MTI, Inc) which provides minutes and time spent in light, moderate, and vigorous activity over a period of days or weeks.127-129 TV and sedentary behavior are assessed by pre-established questionnaires.130, 131 Dietary intake is measured using 24-hour recalls on 2 random days over a week132-136 and completed in an interview format using the NCI Automated Self-Administered 24 hour recall (ASA24 http://riskfactor.cancer.gov/tools/instruments/asa24.html). The primary variables of interest are: calories, protein, carbohydrates, and fat; consumption of sugar-sweetened beverages, and fast food. Fast food consumption is also assessed based on validated self-report questions.137 Weight Control Practices are assessed using the validated Weight Control Strategies scale.138 A supplemental brief assessment 139 is administered to assess frequency of self-weighing, self-monitoring, and meal patterns. Given the association between perceptions of risk and adoption of lifestyle changes, perceived risk of GDM recurrence is measured at baseline and after 4 months using the Risk Perception Survey modified for GDM.140 The Center for Epidemiologic Studies Depression (CES-D) screener 141 is used to examine levels of depressive symptoms; the 14-item Perceived Stress Scale is used to measure levels of stress, which is a predictor of GDM142, and the Eating Inventory143 assesses three dimensions of dietary restraint, including cognitive restraint, disinhibition, and hunger. The General Sleep Disturbance Scale (GSDS) is used for a subjective measure of sleep disturbance has been related to risk of GDM.144
Maternal/infant consequences of GDM. Chart abstractions are conducted by trained research staff to determine whether the intervention results in fewer adverse maternal and neonatal health outcomes. Consistent with prior research, rates of inadequate and excessive GWG are computed based on the National Academy of Medicine guidelines,118 using measured pre-pregnancy and last clinic visit weights. Other maternal adverse outcomes clinically defined based on chart abstractions include gestational hypertension, preeclampsia, cesarean delivery, labor induction, and delivery < 37, 0 weeks’ gestation. Adverse outcomes among infants include birth weight greater than 4000 g, large size for gestational age (defined as birth weight above the 90th percentile), and small size for gestational age (birth weight below the 10th percentile).145 A composite measure of serious perinatal complications is defined as one or more of the following: death (stillbirth or neonatal death), hypoglycemia, hyperbilirubinemia, neonatal hyperinsulinemia, shoulder dystocia/birth trauma (brachial plexus palsy or clavicular, humeral, or skull fracture), admission to the neonatal intensive care unit (NICU), and respiratory distress syndrome.146, 147 An additional composite morbidity outcome is computed based on prior research in women with obesity that includes at least one of the following: Cesarean delivery, hypertensive disorders of pregnancy (HDP), birth weight ≥ 4000 g, birth weight < 2500 g, or NICU admission.148 In women diagnosed with GDM, prescribed treatments are examined to explore whether the intervention impacted intensity of treatment/severity of GDM.
Infant measures. At 6 weeks, infant length, weight, and skinfold thickness measurements are performed by trained staff using standardized procedures. BMI z-scores are calculated using the WHO Child Growth Standards for age and sex.149 A z score of > 1 will be used to define at risk for obesity.
Demographics & medical/reproductive history At baseline, participants complete a demographic questionnaire assessing age, race, ethnicity, and weight history (e.g.,inter-pregnancy weight changes). Given prior relationships with GDM recurrence,150 extensive pregnancy history information (maternal and neonatal) are collected.151, 152 At follow-ups, changes in smoking, prescription medications, unsafe dieting practices,117 job status, and participation in other weight loss programs, and changes in medical history are assessed. Pregnancy urine tests are used to assess pregnancy status at quarterly visits until conception. Pregnancy confirmation is documented through clinical ultrasound results.
Process measures. Recruitment, eligibility, refusal rates/reasons, and retention rates/reasons are tracked, as well as the number of women who conceive during the trial and the average duration until conception. Intervention acceptability is measured based on participants’ ratings of various aspects of the program, the interventionist, content, and overall impression. To measure intervention fidelity, all intervention sessions are audiotaped and a randomly selected subset (10%) are coded for content by a trained study staff (not involved in any assessment data collection). Adherence to the intervention is measured via attendance at treatment sessions, number of self-monitoring records returned, and the activity, eating, and behavioral measures. To assess the safety of the intervention, levels of hunger, depressive symptoms, injuries due to physical activity, changes in medical status, and unintended reduction in milk supply (in breastfeeding women) are assessed.
Sample size {14} Power calculations assume a 60% GDM recurrence rate in women with overweight or obesity.19, 57, 58 With a target sample size of 252 participants and assumed >70% pregnancy rate,153, 154>176 pregnant participants (>88 in each group) would provide adequate statistical power (>81.98%) to detect intervention effects on the proportions developing recurrent GDM,83, 155 taking into account estimations of site-specific clustering effects and effect modifiers (i.e., weight status, ethnicity, parity) of GDM recurrence.24, 57 Under this scenario, the minimum detectable effect size (odds ratio) would be 0.43 and proportions with GDM in educational control and intervention groups respectively would be 60% (n =53/88) and 38% (n = 33/88). For secondary aims, 88 pregnancies in each group would yield > 80% power to detect effects equal to or smaller than those reported in prior work testing effects of lifestyle interventions on reductions in glucose, triglycerides, CRP,156 insulin, leptin 157 adiponectin,158 and blood pressure,159 taking into account estimations of site-specific clustering effects and effect modifiers (i.e., weight status, ethnicity, parity) of insulin resistance,160 and CVD risk factors.161, 162 For secondary aims examining effects of the intervention on prenatal and perinatal complications, the study would have >80% power to detect moderate effect sizes.163 For the fourth aim examining effects of the intervention on pre-pregnancy weight losses and improvements in eating (calories, macronutrient balance, fast food) and activity, 88 participants in each group would yield > 90% power to detect effects reported in prior work testing effects of lifestyle interventions on reductions in weight159 and behavioral variables.164, 165 For exploratory mediator analyses, >80% power is achieved to detect >21% increase in the mediated odds of GDM per kg difference in prepregnancy weight loss (equivalent to a mediated slope in a logistic model of > 0.189).
Assignment of interventions: allocation
Sequence generation {16a} Eligible participants are randomized in a 1:1 ratio into the intervention or control group based on a computer-generated (R 4.0.4 for Windows) random sequence. Randomization is stratified by site, pre-diabetes status (HbA1c <5.7 vs > 5.7) and prior method of GDM diagnosis (one-step, 2 hour test vs other methods) to ensure a balance of the two interventions within each stratum.
Concealment mechanism {16b} Allocation sequence is implemented via sequentially numbered, opaque, sealed envelopes that are concealed until the interventions are assigned to a participant.
Implementation {16c} The study statistician generates the allocation sequence. Study interventionists enroll participants and, based on opening the envelope, assigns participants to interventions.
Assignment of interventions: Masking
Who will be masked {17a} Research Assistants are masked to randomization, and participants do not know assigned group until after baseline measures are completed.
Procedure for unmasking if needed {17b} Unmasking is not needed.
Data collection and management
Plans for assessment and collection of outcomes {18a} All staff involved in data collection must demonstrate competence in administering all measures. The Research Assistants collecting data are masked to the participants’ intervention assignment. The Research Assistants review all assessment data for accuracy and completion. Participants are immediately re-contacted to provide missing data or to clarify responses. Loss to follow up and missing and incomplete data are monitored closely to solve potential issues of missing data before there is a substantial impact on the results.
Plans to promote participant retention and complete follow-up {18b} To minimize loss-to -follow up, at each data collection visit participants are scheduled by phone, sent written reminders, and called the day before. Missed visits are rescheduled and followed up. Costs for transportation and childcare are provided or, alternatively, home visits are arranged for participants with repeatedly missed assessments. Honoraria are provided to promote retention: $25 for visits at study entry, 16 weeks, and 6 weeks postpartum visit, $15 per quarterly visit until conception, and $50 for the primary outcome assessment at 26 weeks’ gestation. As a retention tool, women in both groups also receive quarterly newsletters with basic information about preconception health and wellness.
Data management {19} Research Electronic Data Capture (REDCap) is used for storing study outcome measurement data. A customized internal study tracking system is used to track enrollment and scheduling of visits. Both systems require a login identification and password in order to gain access to the data. Range checks are built into the data collection procedures to alert staff to data that should be clarified. Error checking and preliminary analyses of all data are done to ensure accuracy. Electronic data files are backed up; a copy is stored offsite at both locations to protect against loss or damage. The destruction of any paper files will be at least 7 years from the termination of the study and will be authorized by the PI.
Confidentiality {27}. Participant identification numbers are used to track questionnaires and data collection documents. A password-protected file is maintained that associates the participant name with the participant’s study identification number. Access to electronic data is password protected and restricted to the study team. Paper data are stored in a locked file cabinet. Paper data may be removed for the purpose of coding, data entry, or auditing only. When taking participant files to intervention visits and assessments, files are transported in a locked box. Upon reaching the destination, these boxes are brought into the building or residence with the interventionists. Also, interventionists’ files identify participants by first name and last initial only. Participant home addresses are not included in the files. The study’s research coordinators in CA and RI work closely with the statistician and data manager to ensure the secure exchange and storage of all project databases and questionnaires. Data exchanged between study sites are de- identified, encrypted and, password protected.
Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis in this trial/future use {33} As noted above, participants are asked to give explicit consent for the DNA and RNA collection and use, and future research of data and samples. For DNA and RNA, the collected samples sit in the collection tube at room temperature for 2 hours then are placed in a -20°C freezer for the first 24 hours before moving to -80°C freezer.
Statistical methods
Statistical methods for primary and secondary outcomes {20a} and methods for additional analyses (e.g. subgroup analyses) {20b} A multiple logistic regression analysis will be used to examine the effect of treatment group on the proportion of women who develop GDM. The model will include site and covariates to adjust for pre-randomization variables that may relate to the outcome, including parity, age, education, income, smoking, race/ethnicity, BMI, and time since last pregnancy. The effects of the prepregnancy weight-loss intervention on weight, eating, activity, and physiologic outcomes will be examined using a linear mixed model with fixed effects for treatment condition(the between-groups factor), time (baseline, 16 weeks, 26 weeks gestation), site and baseline covariates. Linear regression and logistic regression analyses will be performed to address the possible effects of the intervention on cases of excessive GWG, gestational hypertension, cesarean delivery, and large for gestational age at birth and 6 weeks, with site and baseline covariates entered in the models. A multiple linear regression analysis will also be used to examine the effect of treatment group on composite scores of adverse maternal/neonatal outcomes, and logistic regressions will be used to examine separate effects on offspring obesity and odds of exceeding National Academy of Science guidelines, including the same covariates described above. Generalized logistic models will be used to examine relationships among pre-pregnancy changes in weight, eating, activity, and physiology and GDM recurrence; we will follow approaches outlined by Kraemer et al166 to explore potential mediators of treatment outcome.
Interim analyses {21b} The trial has no interim analyses or stopping rules.
Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c} Under intention-to-treat principles, all participants with confirmed pregnancy will be included in the primary analysis. If study measured OGTT results are not available, provider assessments will be used (based on chart abstraction). Missing data related to outcomes will be evaluated to assess whether the missing mechanism may be ignorable or non-ignorable.167-169 If the missing data mechanism is judged to be ignorable, where appropriate, analyses involving mixed models may be used such that all existing values are analyzed, and no observations are deleted due to missing values. Alternatively, multiple imputation may be carried out to create several complete data sets. For each complete data set, overall tests of interest for the outcome will be conducted and the results of each combined to create a single test result. For completeness, a pattern-missing analysis will be conducted to investigate non-ignorable missingness. If the missing data mechanism is likely to be non-ignorable, multiple imputations can be conducted using a version of the approximate Bayesian bootstrap based on distance-based selection criteria.170 Sensitivity analyses under various assumptions regarding the missing data will be conducted to confirm the robustness of the results.
Plans to give access to the full protocol, participant level-data {31c}. Upon publication of study’s pre-specified outcomes, a de-identified version of the database will be made available upon reasonable request to the PI.
Oversight and monitoring
Composition of the coordinating center and trial steering committee {5d}
The trial does not include a coordinating center or steering committee.
Composition of the data monitoring committee, its role and reporting structure {21a} The trial includes two external safety officers with expertise in clinical trial weight control research and maternal/fetal health. Twice per year, safety officers review reports of recruitment, retention, fidelity, and safety information on all participants, including number of pregnancies before the 4-month intervention is over, number of injuries due to physical activity, number of miscarriages, and other serious adverse events. Weekly internal investigator meetings also occur to review recruitment, attendance, retention, and safety data on an ongoing basis.
Adverse event reporting and harms {22} Adverse events (AE) include any event that causes or increases risk of harm to the participant or others. Serious adverse events (SAE) include any event that results in death, inpatient hospitalization or prolongation of existing hospitalization, a persistent or significant disability or incapacity, or a congenital anomaly or birth defect. A fatality, including fetal, are reported within 24 hrs. AEs reported at core assessment visits or informally at any time are evaluated by the research team and the investigators to determine if they are unanticipated problems involving risk to subjects and others or not. The participant's situation is also assessed with regard to study and/or intervention continuation. Any SAEs are recorded by the research coordinator, reported to the PI and investigative team, the Safety Officers, and the IRB.
Frequency and plans for auditing trial conduct {23} The trial includes close monitoring by the PI/Co-Is, IRBs, and two external safety officers. Annual progress reports are provided to the sponsor and IRBs. Sponsor or other external site visitor audits are not planned.
Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees) {25}. Any changes to eligibility criteria, outcomes, or analyses are reviewed by the IRB and updated in Clinicaltrials.gov.
Dissemination plans {31a}. Results of the trial will be presented at professional conferences and local community events and shared via clinicaltrials.gov and formal publications and furthermore to the general public through social media outlets. A summary of primary outcome findings will be created in English and Spanish and shared with study participants.