Article

Maternal and Child Health Journal

, Volume 13, Issue 3, pp 343-348

State Infant Mortality: An Ecologic Study to Determine Modifiable Risks and Adjusted Infant Mortality Rates

  • David A. PaulAffiliated withPediatrics and Neonatology, Christiana Care Health Services, Christiana HospitalDepartment of Pediatrics, Thomas Jefferson University Email author 
  • , Amy MackleyAffiliated withPediatrics and Neonatology, Christiana Care Health Services, Christiana Hospital
  • , Robert G. LockeAffiliated withPediatrics and Neonatology, Christiana Care Health Services, Christiana HospitalDepartment of Pediatrics, Thomas Jefferson University
  • , John L. StefanoAffiliated withPediatrics and Neonatology, Christiana Care Health Services, Christiana HospitalDepartment of Pediatrics, Thomas Jefferson University
  • , Charlan KroelingerAffiliated withCenter for Excellence in Maternal Child Health and Epidemiology, Division of Public Health, Delaware Health and Social Services

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Objective To determine factors contributing to state infant mortality rates (IMR) and develop an adjusted IMR in the United States for 2001 and 2002. Design/Methods Ecologic study of factors contributing to state IMR. State IMR for 2001 and 2002 were obtained from the United States linked death and birth certificate data from the National Center for Health Statistics. Factors investigated using multivariable linear regression included state racial demographics, ethnicity, state population, median income, education, teen birth rate, proportion of obesity, smoking during pregnancy, diabetes, hypertension, cesarean delivery, prenatal care, health insurance, self-report of mental illness, and number of in-vitro fertilization procedures. Final risk adjusted IMR’s were standardized and states were compared with the United States adjusted rates. Results Models for IMR in individual states in 2001 (r 2 = 0.66, P < 0.01) and 2002 (r 2 = 0.81, P < 0.01) were tested. African-American race, teen birth rate, and smoking during pregnancy remained independently associated with state infant mortality rates for 2001 and 2002. Ninety five percent confidence intervals (CI) were calculated around the regression lines to model the expected IMR. After adjustment, some states maintained a consistent IMR; for instance, Vermont and New Hampshire remained low, while Delaware and Louisiana remained high. However, other states such as Mississippi, which have traditionally high infant mortality rates, remained within the expected 95% CI for IMR after adjustment indicating confounding affected the initial unadjusted rates. Conclusions Non-modifiable demographic variables, including the percentage of non-Hispanic African-American and Hispanic populations of the state are major factors contributing to individual variation in state IMR. Race and ethnicity may confound or modify the IMR in states that shifted inside or outside the 95% CI following adjustment. Other factors including smoking during pregnancy and teen birth rate, which are potentially modifiable, significantly contributed to differences in state IMR. State risk adjusted IMR indicate that other factors impact infant mortality after adjustment by race/ethnicity and other risk factors.

Keywords

Infant mortality Racial disparities Adjusted infant mortality rate Ecologic study Teenage pregnancy Smoking