Recorded presentation

Janet T. Eyster, PhD and Claudia Holzman, DVM, MPH, PhD. Department of Epidemiology, Michigan State University, College of Human Medicine, 4660 South Hagadorn Rd., Ste. 660, E. Lansing, MI 48895, (517) 353-8623, eysterj@msu.edu

Preterm delivery (PTD) rates are often assessed in conjunction with maternal characteristics, health status and health behaviors recorded on birth certificates. Community contextual data can be added to the analysis by geocoding addresses from the birth certificates and using multi-level modeling to incorporate small area census data and community information. However, small area community health status/health behavior data are difficult to obtain. One approach is to aggregate individual health status/health behavior data collected on birth certificates. In Michigan, multi-level analyses of birth certificate data from 16 urban communities was used to explore the relationships between PTD and maternal age, education, individual maternal smoking status, and ‘census tract smoking rates among women with live births’ (aggregating from birth certificates). Among both white non-Hispanic and black non-Hispanic women, the increase in PTD rates in women age 30 years and above became more pronounced as the ‘census tract smoking rates in women with live births’ increased. This interaction remained after adjusting for individual smoking behavior. An additional example using aggregated medical conditions data from the birth certificate will be presented, and the Michigan results will be compared with results from areas in North Carolina, Maryland and Philadelphia. The advantages and limitations of developing contextual data by aggregating smoking and medical conditions data recorded on birth certificates will be discussed.

Learning Objectives:

• To discuss the use of aggregating individual characterists to obtain neighborhood statistics for multi-level data analysis and present an example of the technique.

Keywords: Statistics, Maternal and Child Health

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

Recorded presentation