Ilse Larson

Date of Award


Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)

First Advisor

John Leventhal


The purpose of this study was to determine the patterns of fractures in children less than three years of age that are distinctive of abuse, and to examine changes in the frequency of abusive fractures in young children evaluated at a major pediatric hospital over three time periods: 1979-1983 (early), 1991-1994 (middle), and 1999-2002 (late). All children 0 to 36 months of age who were treated for bone fractures at a major medical center from 1/99 to 12/02 were selected. Medical records were abstracted for sociodemographic and clinical characteristics, and radiographs were examined. Using specific criteria, each case was rated by 2 clinicians and 2 pediatric radiologists on a 7-point scale ranging from definite abuse to definite accident. Cases were rated independently by each reviewer; when disagreements occurred, a consensus rating was reached. Cases rated as definite, likely, or questionable abuse were considered abuse. Demographics of the abused children were compared to those with either accidental fractures or fractures of unknown etiology using chi-square statistics. The proportion of children rated as abuse in the late sample was compared to the proportions previously identified in the early and middle time periods using adjusted odds ratios controlling for race and physician type (clinic vs. private). Several fracture types were highly associated with abuse; 100% of rib fractures, 29.2% of femur fractures, 19.5% of humerus fractures, and 12.8% of tibia/fibula fractures were rated as abuse. Abused children were more likely than those with accidental or unknown fractures to present with vague or missing histories to explain the fracture (60.0% vs. 11.6%), and were more likely to be less than 12 months of age (68.0% vs. 26.6%), insured by Medicaid or to be self-pay patients (68.0% vs. 41.1%), and of minority race (56.0% vs. 29.9%). For the late time period, 10.8% of 232 cases were classified as abuse; in the middle group, 10.0% of 240 cases, and in the early group, 22.5% of 200 cases. Children in the early group had two and a half times the odds of an abusive fracture when compared with the late group (adjusted OR 2.58, 95% CI=1.43, 4.65). The odds of abuse did not differ significantly between the middle and late groups (adjusted OR 0.86, 95% CI=0.46, 1.63). Fractures of the ribs, femur, humerus, and tibia/fibula were most highly associated with abuse. Abused children were more likely to present with vague or missing histories to explain the fracture, and were more likely to be less than 12 months of age, of minority race, and either self-pay patients or insured by Medicaid. The rate of fractures due to abuse has decreased dramatically over the past three decades at one major pediatric center.


This is an Open Access Thesis.

Open Access

This Article is Open Access