Related Experiment Video
Updated: Jun 20, 2026
Training Persons with Spinal Cord Injury to Ambulate Using a Powered Exoskeleton
Published on: June 16, 2016
Development of Composite Measures of Neighborhood Economic Factors for Use in Spinal Cord Injury Outcomes Studies: A
Amanda L Botticello1, Lauren Murphy1, Susan Charlifue2
1Center for Outcomes and Assessment Research, Kessler Foundation, West Orange, NJ; Department of Physical Medicine and Rehabilitation, Rutgers New Jersey Medical School, NJ.
Objective:
To develop composite measures of neighborhood economic factors for use with the national Spinal Cord Injury Model Systems (SCIMSs) database in cross-sectional and longitudinal investigations of the social determinants of health.
Design:
Secondary data analysis of administrative data from the 2009, 2014, and 2019 American Community Survey (ACS) 5-year estimates and survey data collected for the SCIMS database.
Setting:
Community.
Participants:
The validity of the neighborhood economic measures developed from the ACS data was tested with a sample of SCIMS participants who completed a follow-up interview between 2017 and 2021 (N=8,130). The predictive validity of the neighborhood measures was assessed with a subsample of cases with complete data on the outcome and covariate measures (N=6,457).
Interventions:
Not applicable.
Main Outcome Measures:
A binary measure of self-rated health status (1=poor/fair health; 0=good/very good/excellent).
Results:
A combination of panel review and data reduction techniques yielded 2 distinct measuring neighborhood socioeconomic status (SES) and neighborhood socioeconomic disadvantage that were validated using 3 waves of ACS data and the SCIMS data. The odds of reporting poor health were lower among people living in moderate- and high-SES neighborhoods and highest among people living in moderately and highly disadvantaged neighborhoods. The negative association between neighborhood SES and poor health was fully attenuated by differences in participants' individual demographic and economic characteristics whereas the positive association between neighborhood disadvantage and poor health persisted after adjusting for individual differences.
Conclusions:
The two composite measures of neighborhood economic factors developed by this study are robust in samples from different periods of time and valid for use with the SCIMS database. Future investigations conducting surveillance of the needs of the SCI population using this resource may consider using these measures to assess the effect of the social determinants of health in outcomes after SCI.
