Epidemiology of Stress and GDM among Latina Women

University Of Massachusetts Amherst, Amherst MA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Diabetes and obesity have reached epidemic proportions in the U.S. with rates consistently higher in Latinas than non-Latina whites. Shifting the prevention period for these conditions upstream, to pregnancy, may be an innovative strategy. The major aims of this proposal are to evaluate new and timely hypotheses regarding modifiable determinants of GDM among Latina women. We propose to evaluate the association between external stressors (physical activity), internal stressors (psychosocial stress), and mediators between these variables and risk of GDM. Using a prospective cohort design, 2,300 Latina (predominantly Puerto Rican) prenatal care patients will be recruited from Baystate Medical Center at their first prenatal care visit (8-12 weeks gestation). Bilingual interviewers will obtain detailed information on physical activity patterns via the Pregnancy Physical Activity Questionnaire (PPAQ) (NIH/NICHD R03-39341) developed and validated for use in this population. Psychosocial stress will be assessed through 1) self-report (life events, perceived stress, anxiety, depression, and self-esteem) using scales validated in Latina populations, and 2) salivary cortisol, a physiologic marker of response to stress. Information on pregravid BMI, diet, cigarette smoking, substance abuse, medical and obstetric history, acculturation, and sociodemographic factors will also be queried. Weight and waist circumference will be measured. Two subsequent interviews, conducted at the time of routine ultrasound (18-20 weeks gestation) and GDM screen (24-28 weeks gestation), will update information on the variables of interest and repeat the salivary cortisol collection. A gestational-age stratified sample of 100 women from the cohort will be selected to assess diurnal variation in the cortisol measure. Lab reports will be abstracted for GDM screen results and a 4th interview will be conducted at 32-34 weeks gestation among those with elevated screening glucose. After delivery, medical records will be abstracted to confirm incident GDM and birth outcomes. Logistic regression will be used to assess the relationship between physical activity, psychosocial stress and GDM risk. Marginal structural models will assess the role of diet, body fat distribution, weight gain, and smoking as mediating variables in the relationship between stress and GDM. Cortisol will be analyzed in a nested case-control fashion on 116 GDM cases and a matched 4:1 sample of controls. A secondary analysis will prospectively assess the relationship between stress and adverse birth outcomes in among women with elevated screening glucose. The proposed study builds upon our experience in the Latina GDM Study (American Diabetes Association Career Award) and extends the work of prior investigations through prospective assessment at multiple points in gestation, a multidimensional approach to stress measurement, and detailed evaluation of potential confounding and mediating factors in a large, high-risk population. The multidisciplinary research collaboration utilized in this proposal is a key component of the NIH Roadmap for Medical Research in building "Research Teams of the Future".