Graduate Student Internship Program

2009 Catalog of Internship Opportunities

The Associations between Pre-pregnancy Weight and Preterm Birth and Pre-Pregnancy Weight and NICU Admission – Ohio, 2000–2007

Ohio Department of Health, State Epidemiology Office
and Division of Family and Community Health Services, Columbus, OH
Skill Area: Data Analysis and Monitoring

Agency Information

The intern will be housed in the State Epidemiology Office (SEO) at the Ohio Department of Health (ODH). The SEO was established in 2008 and its mission is to “guide epidemiologic priorities and activities for the state, operate in coordination and collaboration with local, state and federal partners, building epidemiologic capacity for the state, and assist with the translation and reporting of epidemiologic findings to public health programs and policies in Ohio.“. It is staffed by the Senior State Epidemiologist, State Infectious Disease Epidemiologist, State Chronic Disease Epidemiologist, State Environmental Health Epidemiologist (vacant) and State Maternal and Child Health Epidemiologist. Fellows such as EIS officers and CSTE fellows are housed in this office and supervised either solely or collaboratively by the State Epidemiologists, who have no direct reports. The SEO works closely with and is physically integrated with the Center for Public Health Statistics and Informatics (CPHSI), which houses epidemiologists and researchers who support areas including maternal and child health, vital statistics, and chronic disease. The Pregnancy Risk Monitoring System (PRAMS) is housed within CPHSI.

The Division of Family and Community Health Services (DFCHS) administers programs and oversees data systems fundamental to maternal and child health (MCH). DFCHS houses six program bureaus, representing the spectrum of MCH and other populations. The secondary supervisor works within DFCHS and the intern will have opportunities to work with relevant bureaus within DFCHS.

Purpose, goals, and objectives of internship

The Ohio Department of Health is engaged in a birth outcomes improvement initiative and infant mortality task force which both aim to improve birth outcomes and reduce disparities. There is a focus on preconception health, or the health of a woman before she becomes pregnant, which impacts the success of a pregnancy. Understanding the contribution of preconception weight status to negative birth outcomes in Ohio would inform these efforts.

Maternal weight status before pregnancy has been shown to impact pregnancy outcomes, both in terms of maternal morbidities and infant outcomes. Pre-pregnancy underweight, overweight and obesity have been linked to poor maternal and infant outcomes, and can be compounded by inadequate or excessive weight gain during pregnancy. In Ohio, 19.8% of new mothers were overweight prepregnancy in 2005 and 21.0% were obese (source, Ohio PRAMS 2005, unpublished).

Globally, preterm birth accounts for more than 25 percent of all neonatal deaths. In Ohio, the percent of preterm births have increased from 10.15% in 2000 to 11.35% in 2005 (Ohio Vital Statistics). In 2005 in Ohio, 16.6% of all infant deaths were attributed to short gestation or low birth weight (Ohio Vital Statistics). To understand infant health in terms of mother’s pre-pregnancy weight, preterm birth is a preferred indicator over low birth weight because of the risk of macrosomia to neonates of high weight mothers. Another measure of infant health status in addition to preterm birth is infant admission to the neonatal intensive care unit. Pre-pregnancy overweight or obesity have been shown to cause high weight for gestational age, but also by some to be associated with shorter length of gestation. Infant admission to the intensive care unit related to maternal pre-pregnancy weight was examined in at least one published study that found it to be significantly increased in overweight and obese mothers with a relatively small effect size.

The purpose of the internship is to provide useful information on the contribution of prepregnancy weight status on birth outcomes among Ohio mothers.

The goals are to provide more detailed information on 1) the relationship between prepregnancy weight status and preterm birth, 2) the relationship between prepregnancy weight status and NICU admission, and 3) if appropriate, the population attributable fraction of maternal obesity to these outcomes.

The objectives are to 1) develop an analysis plan informed by a literature review of the topic, 2) conduct appropriate and valid analyses given the data sources and research questions outlined in the plan, 3) summarize and interpret results of the analyses, and 4) draft a brief epidemiological report and present an oral presentation to ODH staff.

PROPOSED METHODS The data source for this study is the Ohio Pregnancy Risk Assessment Monitoring System (PRAMS) analysis datasets for the years 2000-2007. PRAMS is an on-going population-based mixed method survey conducted by state health departments. Sponsorship of the survey is provided by the Centers for Disease Control and Prevention as part of an initiative to reduce infant mortality and low birth weight. A random sample of mothers receives a survey concerning maternal experiences and attitudes by mail two to six months postpartum. Non-responders are contacted by telephone and asked to provide data via a telephone survey. The Ohio Department of Health began active participation in 1999.

Birth outcomes for the analysis will include preterm births (less than 37 weeks gestation) and neonatal intensive care unit (NICU) admission or transfer. Preterm births may be further broken down into the categories of very preterm (10-31 weeks), and moderately preterm (32-36 wks). The frequencies of these outcomes as well as frequencies of known risk factors for poor birth outcomes will be described and compared. Separate multivariable logistic regression models should be constructed for each of the two primary outcomes.

Body mass index (kg/m2) is calculated from self-reported height and pre-pregnancy weight captured in PRAMS. Women are stratified into four BMI categories based on the Institute of Medicine (IOM) BMI categories for pregnant women, i.e., underweight (BMI<19.8), normal weight (BMI 19.8-25.9), overweight (BMI 26.0-29.0), and obese (BMI>29.0).

Since previous studies have shown that weight gain during pregnancy may modify the association between prepregnancy BMI and preterm delivery, this relationship should be examined within each BMI category. Weight gain rates per trimester may be estimated using a method similar to the study by Dietz et al using data and weight gain categories recommended by the IOM. Potential confounders which may be explored include maternal characteristics and health behaviors (maternal age, race and ethnicity, vitamin use, smoking, alcohol use, pregnancy intention, prenatal care initiation); socio-demographic characteristics (education, marital status, physical abuse by partner, urban/rural residence); and pregnancy complications (e.g., gestational diabetes, small for gestational age, delivery induction, preeclampsia, c-section). Health care factors (insurance status and hospital variables) may be examined as well.

Statistical analyses will be carried out using SAS-callable SUDAAN.

Data or analytic tasks and activities

The project will require development of research questions; a literature review; writing of an analysis plan; identification of variables of interest (outcome variables, primary variables of interest, potential confounders and effect modifiers) within the PRAMS database; recoding of variables as necessary; application of appropriate analysis methods, and interpretation of results, summarizing results in a brief written report; and preparation and presentation of an oral presentation.

Data or analytic skills required

The internship position requires basic SAS data and analytic skills, familiarity with multivariate regression techniques, and writing and communication skills. Familiarity with SUDAAN is preferred.

Supervisors

Primary: Dr. Elizabeth Conrey, State MCH Epidemiologist and Centers for Disease Control and Prevention field assignee to the Ohio Department of Health, State Epidemiology Office.

Secondary: Lori Deacon, supervisor of the Maternal and Child Health (MCH) Unit within the Bureau of Child and Family Health Services of the Ohio Department of Health.

Internship begins

The timing of the internship position is flexible. Summer is preferred.

Housing

Staff will be able to assist the student in locating and securing housing.

Transportation

The ODH is located in downtown Columbus with excellent access to public transportation. Several bus lines pass directly in front of the ODH building. Apartments, sublets and other housing available for short term lease exist nearby.

Back to top