235950 Developing Composite Area-Level Indicators of Socioeconomic Position Using Multilevel Factor Analysis

Tuesday, November 1, 2011

Donna Almario Doebler, DrPH, MS, MPH , Graduate School of Public Health, Department of Biostatistics, University of Pittsburgh, Pittsburgh, PA
Roslyn A. Stone, PhD , Department of Biostatistics, University of Pittsburgh Graduate School, Pittsburgh, PA
Kevin H. Kim, PhD , Department of Psychology in Education, University of Pittsburgh, School of Education, Pittsburgh, PA
Ravi K. Sharma, PhD , Department of Behavioral and Community Health Sciences, University of Pittsburgh, Graduate School of Public Health, Pittsburg, PA
Stephen B. Thomas, PhD , Department of Health Services Administration, Center for Health Equity, University of Maryland, School of Public Health, College Park, MD
BACKGROUND: "Neighborhood" and socioeconomic position (SEP) measures are inconsistently defined in the literature. This research describes an innovative statistical approach to develop composite SEP indicators at two area levels in Pittsburgh, Pennsylvania. METHODS: Twelve SEP measures were derived from U.S. Census 2000. Exploratory multilevel factor analysis (MFA) was used to develop composite SEP indicators at block group (BG) and neighborhood (NB) levels. Validity was assessed by correlating SEP indicators and low birth weight (LBW) percentage; maps depicted SEP indicators and LBW percentage at BG and NB levels. RESULTS: Different SEP indicators emerged at each level. Two BG-level indicators were (i) “material and economic deprivation” (MEDij), representing percentage not owning a car, renting their residence, in poverty, receiving public assistance, and earning low income, and (ii) “concentrated disadvantage” (CDij), representing percentage of Blacks, single-headed families, having family members under 18 years old, and receiving public assistance. At NB-level, all 12 SEP measures were included in one indicator, “overall neighborhood deprivation” (ONDj). Indicators were positively correlated with LBW percentage: NB-level correlation between ONDj and LBW percentage was r=0.45 (n=89); BG-level correlations between MEDij and CDij and LBW percentage were r=0.25 and r=0.31, respectively (n=341). Geographic distributions of SEP indicators and LBW percentage were similar at each level. CONCLUSION: The study outlines a novel approach to creating SEP indicators. Differences in operationalization of SEP at BG and NB levels could facilitate the development of targeted interventions within and between neighborhoods in improving birth outcomes as well as other public health concerns.

Learning Areas:
Biostatistics, economics
Environmental health sciences

Learning Objectives:
Describe a novel approach to creating socioeconomic position indicators at the neighborhood and block group levels Demonstrate the association between socioeconomic position indicators and low birth weight at the neighborhood and block group levels Describe geographic distribution of socioeconomic position indicators and low birth weight using maps.

Keywords: Statistics, Community Health

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am qualified to be an abstract author given my graduate work on creating socioeconomic indicators at the neighborhood and block group levels.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.

See more of: Poster Session: Student Papers
See more of: Statistics