CO-OP Trial: Community-based Outreach on Obesity in Pregnancy

Overview

Obesity is becoming more common and disproportionately affects the younger generation, the poor and certain ethnic groups. Nearly 50% of reproductive aged women are obese or overweight. Obesity during pregnancy leads to higher rates of hypertensive disorders, stillbirth, less successful breastfeeding, obesity in their offspring, postpartum depression, and higher weight retention postpartum. Dietary counseling can prevent excessive maternal weight gain and is more effective than activity-based interventions. The two objectives of this study are 1) to use community-based techniques to improve dietary counseling for high risk women and 2) to randomize obese women to dietary counseling at Truman Medical Center and measure their success. The hypothesis is that dietary interventions can successfully be tailored for high risk women and that excessive weight gain during pregnancy can be avoided.

Phase 1 will recruit 8-12 obese women from the target population. Initially, 4-6 women will explore dietary guidance and study interventions and to determine how best to tailor the intervention. A repeat set of structured interviews will be conducted with the community-based modifications to provide an assessment of the perceived efficacy of the intervention. Structured interviews will provide feedback on dietary interventions including specific advice on caloric recommendations during pregnancy, level of reassurance by negative ketonuria on urine specimens (which indicates the diet is supplying the nutrition necessary to maintain a healthy pregnancy), general level of trust/interaction with providers, maintaining food diary, and how obesity issues directly affect their lifestyles and health. Phase 2 will be a prospective randomized trial in which obese women will be recruited in the first trimester. Study participants will be identified by providers in the Truman Medical Center clinic population and includes obese pregnant women less than 16 weeks in gestational age during the pregnancy. Dates will be confirmed by standard practices including using last menstrual period and obstetric ultrasound. Once potential women are identified they will be directed to study coordinators who will determine study eligibility (BMI \>30, 16 weeks or less pregnant, willing to participate, singleton pregnancy, normal 1 hour glucola) and consent patients for the study. Women will be randomized to usual care or dietary intervention using opaque numbered envelopes that are generated by blinded statistician. The dietary intervention will be conducted by a registered dietician once. Patients in the study arm will receive a detailed intake and specific dietary guidelines. The dietary guidelines for study participants will be similar to those for gestational diabetes including 18 to 24 kcal/kg diet with 40% carbohydrates, 30% protein, 30% fat. No participant will be given a diet that is less than 2,000 kcal per day. Recording daily food/beverage intake in a diary has been associated with improved adherence to dietary recommendations and is one of the more effective interventions for behavioral modification. Therefore, study participants will be asked to record a food/beverage diary and to bring it to prenatal visits. Recent evidence suggests that highly dense and acellular processed foods are associated with an inflammatory state that leads to more absorption of lipids and deposition of fat.18,19 Study participants will be counseled to limit intake of highly calorie dense and acellular processed foods and beverages to 1-2 times per week if at all and to increase total percentage of fruits and vegetables to 5-9 items per day. It is expected that some women may lose weight during this clinical trial as 40% of women lost weight in a similar trial8. Although those women lost weight, there were no instances of starvation or ketonuria on routine screening and there was no difference in infant birth weights between the two groups.8 Routine urinalysis is performed during pregnancy and screening for ketonuria can indicate metabolic evidence of starvation. Therefore, to monitor safety during our study, routine urinalysis will be inspected and evidence of starvation ketonuria (2+ on urine dip, or 40mg/dl) will prompt investigation by study staff on the cause and determine whether it is related to the study protocol. No evidence of maternal ketonuria was documented in over 200 women in the previous study.8 Data safety and monitoring will be conducted every 6 months by the medical monitor for the study, Michael Sheehan, MD (neonatologist). The medical monitor is specifically chosen outside of the department of ob/gyn to allow for objectivity. Adverse events which would require immediate suspension of the study for review would include maternal death, maternal ICU admission, or a total of 2 perinatal deaths in the study group. Unanticipated adverse events will be reported immediately to the Human Subjects Committee, and its recommendations for temporary cessation or continuation of the study will be followed. The natural frequency of perinatal death at Truman after 16 weeks in obese women is a maximum of 2 perinatal deaths in the 50 women in the study group. If we assume the intervention is no better than current standard of practice then a maximum of 2 perinatal death cases should serve as the rule for study suspension and review. If more than 4 perinatal deaths occur then the study should be concluded as this is more than the expected deviation. Women will complete an intake assessment including maternal height and weight, food frequency and quality of life questionnaires and these will be repeated in the 3rd trimester and at 12 months postpartum. Missing data will be handled in two separate ways for analysis. Numerical values such as missing weight, height, birth weight, etc. will be handled as follows. Three separate attempts will be made to locate the data. For instance, if a patient delivered at another hospital then the patient and hospital will be contacted on 3 separate occasions to obtain the delivery records. Initially, missing data will be excluded from the analysis. Second, missing data will be imputed from last values reported for study participant (for maternal weight, height if within 2 weeks) or the average value for the group will be imputed as the value (ex: birth weight, maternal weight at delivery) if feasible to determine if missing values alter the analysis. Both methods will be reported.