Risk factors for maltreatment-related infant hospitalizations in New York City, 1995-2004

Research paper by Susan M. Mason, Patricia G. Schnitzer; Valery A. Danilack; Beth Elston; David A. Savitz

Indexed on: 04 Jun '18Published on: 03 Jun '18Published in: Annals of Epidemiology


Publication date: Available online 2 June 2018 Source:Annals of Epidemiology Author(s): Susan M. Mason, Patricia G. Schnitzer, Valery A. Danilack, Beth Elston, David A. Savitz Background Child maltreatment is a major public health problem, but population-based surveillance and research is hindered by limitations of official child welfare data. The current study used a promising complementary data source, hospital discharge data, to investigate risk factors for infant injuries suggestive of maltreatment. Methods Discharge data from all hospital births to New York City residents from 1995-2004 were linked to birth certificates and to subsequent infant hospitalizations within 12 months after delivery. Probable maltreatment of infants was identified with 33 injury diagnosis codes highly correlated with maltreatment. Modified Poisson regression estimated the association of sociodemographic factors and pregnancy/birth characteristics with subsequent infant admission for probable maltreatment. Results Risk factors for maltreatment included neonatal intensive care unit (NICU) stays of >11 days (adjusted RR=1.8; 95% CI: 1.1-2.8), preterm birth (aRR=1.6; 95% CI: 1.2-2.1), maternal age <20 years (aRR=1.5; 95% CI: 1.2 - 1.9), and public insurance (aRR=1.5; 95% CI: 1.2 - 1.9). Factors associated with reduced maltreatment risk included mother born outside the US (aRR=0.7; 95% CI: 0.6-0.8) and female infant (aRR=0.7; 95% CI: 0.6-0.9).\ Conclusions Sociodemographic factors and challenges at birth (preterm, NICU stays) are important risk factors for subsequent maltreatment-related hospitalization, with potential implications for prevention targeting.