Winkleby Lab In the Prevention Research Center

Mothers, Places, and Preterm Birth

Background: The birth of premature infants is an increasingly serious maternal and child health problem in industrial countries, caused primarily by preterm births (PTB) and intrauterine growth retardation (IUGR). While past studies have elucidated some risk factors for prematurity (e.g. smoking, infection, chronic stress), the underlying causes are still poorly understood. Virtually all studies to date have been cross-sectional in design, and/or have had small sample-sizes, analytic limitations, and limited information on individual-level demographic and social factors. Furthermore, there have been almost no studies examining neighborhood level influences on PTB and IUGR despite an increasing number of studies that suggest that neighborhood environments have independent influences on birth outcomes.

Aims: This study will use repeated crosssectional multilevel analyses to examine whether adverse birth outcomes from 1973-2005 among Swedish mothers are influenced by neighborhood-level social factors over an extended time period (i.e. contextual) and/or individual-level social factors (i.e., compositional). This cross-institutional project builds on long-term partnerships, with investigators from the Karolinska Institutet and Lund University in Sweden and Stanford University in the U.S., together with consultants from U.C.S.F., Harvard, and London Universities. The primary aims are to examine whether individual-level (e.g. socioeconomic status/position [SES], medical and maternal risk factors, exposure to different levels of neighborhood SES based on residential history from 1960), and neighborhood-level social factors (e.g., neighborhood SES, social disorganization, immigrant composition, social capital, neighborhood "type", goods and services, community resources) influence PTB and IUGR. In addition, mediators and cross-level interactions of the neighborhood effect(s) will be examined.

Design/methods: Two separate databases will be created that link annual census, mortality, hospital, and individual survey data. WomMed consists of the entire Swedish population of 3.5 million mothers, aged 15 to 45, and born between 1928 and 1990, whose geocoded addresses have yielded 9,677 neighborhoods. It will include census data as well as individual-level sociodemographic characteristics and maternal risk factors. WomSALLS will include data from face-to-face interviews with representative samples of approximately 18,000 mothers aged 15-45, and born between 1928 and 1990. This dataset will include additional factors that may mediate the relationship between neighborhood-level characteristics and PTB and IUGR, including health behaviors/risk factors, stress and adaptation, social cohesion, and psychological distress. Information from these datasets will be linked to the Swedish Medical Birth Register that includes maternal and medical data.

Public health relevance: This study will show the importance of allocating resources at both the individual and neighborhood levels to improve the health of mothers and children. Findings will be integrated with previous knowledge and results will be disseminated via scientific publications, student learning laboratories, and long-term partnerships with public health practitioners and the lay community.


Footer Links: