State-level analysis of variation in infant mortality rate: A socio-ecological approach

Nicole M Sobiech, Purdue University


Background: Infant mortality has been used as a predictor of population health, and was higher in the United States compared to other developed countries. The rate of infant mortality varies greatly at the State level, with infants born in southern States having an increase risk. Methods: This paper investigates influential variables in all five dimensions of the socio-ecological model on State infant mortality in the U.S. Multiple regression analyses were conducted to determine the amount of variance that could be explained between and at the State level. Results: The socio-ecological model explained 73% variation of State infant mortality in the U.S. The fifteen States with the highest infant mortality rates were located in the South East half of the country. The percentage of the population classified as African American, educational attainment, the percent of children in poverty, the percent of the population classified as obese and smoking and the percentage of births to unmarried women remained were statistically significant. Institutional and community variables explained more variance than policy and interpersonal and individual variables. States with higher infant mortality rates had higher average percentages of the population classified as African American, children uninsured, and children born to unmarried women. They also had a lower average of primary care physicians per 100,000 populations. Conclusion: The results suggest that interventions to reduce infant mortality should focus on variables within the institutional and community dimensions of the socio-ecological model. Access to primary care physicians and improved overall population health in a State could result in a lowering of State infant mortality rates.




Hyner, Purdue University.

Subject Area

Public health

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server