Connecting concentrated disadvantage and birth outcomes to enhance program targeting

Background: Identifying target communities for public health programs is challenging when local-level health data is unavailable or unreliable.  Concentrated disadvantage (CD), one of 59 AMCHP-recommended life course indicators, measures local community economic wellbeing.  This study examined the relationship between CD and birth outcomes to determine whether CD is a reasonable proxy to inform geographical targeting of maternal and child health programs.

Methods: County-level Census (2010) and American Community Survey (2008-2012) data were obtained for five variables: % households in poverty, % female-headed households, % unemployment, % receiving public assistance, and % <18 years.  Z-scores were computed for each variable, averaged into a composite CD z-score, and then classified into CD quartiles.  Illinois death certificate (2009-2011) and birth certificate (2010) data were used to compute the rates of: low birth weight (LBW), very low birth weight (VLBW), infant mortality (IM), less than adequate prenatal care (PNC), and teen birth (for 15-19 year olds) in each CD quartile.  Binomial regression was used to assess the differences between CD quartiles.

Results: All five outcomes yielded a dose-response association with CD quartile.  Counties in the highest CD quartile had significantly higher rates of poor outcomes than counties in the lowest CD quartile (LBW RR=1.25 [1.18–1.32]; VLBW  RR=1.45 [1.26–1.67]; IM RR=1.37 [1.11–1.69]; <Adequate PNC RR= 1.54 [1.49–1.59]; Teen birth RR=2.10 [1.97–2.22]).

Conclusions: County-level CD is associated with a wide array of birth outcomes and may be a useful metric to inform geographic targeting of MCH programs.