Child Characteristics

Age of child: Physical abuse is more common among older children than younger children. ^, The NIS-3 found that the rate of physical abuse among children ages 12 to 14 years was significantly higher than the incidence among children ages 0 to 2 years. This may be due to a lack of identification among the younger children. Children ages 0 to 2 years may have less exposure to people in the community than older children and be less likely identified as abused by community sentinels. NCANDS also reports a higher percentage of physical abuse cases among older children, but is subject to the same biases as NIS. The association with age has been inconsistent with population-based surveys.

Sex of child: Sex is an inconsistent risk factor for physical abuse. One study determined that male children were more likely to be physically abused. However, another recent study found that girls were at slightly higher risk for physical abuse.

Race: The victim’s race is sometimes found to be a risk factor for physical abuse. NCANDS data revealed different rates of physical abuse by race; 14.6% of physical abuse victims were Asian, 12.9% were African-American, and 9.8% were white. Other studies such as the NIS-3 have not found racial differences in rates of physical abuse. Certain races may be more likely to come to the attention of CPS than others, which would bias the NCANDS results.

Caregiver Characteristics

Age of mother: Mothers younger than 26 years are more likely to physically abuse their children than older mothers. A longitudinal study of 644 families determined that younger mothers were 2.37 times as likely to physically abuse their children.

Mental health of mother: Children who have caregivers with depression ^, or substance abuse ^, are more at risk of being physically abused. Additionally, general maternal sociopathy has also been identified as a risk factor for physical abuse.

Marital status: Caregiver marital status has also been found to be associated with physical abuse. Children who live with only one parent are more likely to experience physical abuse than children who live with both parents. Additionally, the NIS-3 determined that children who live with only their fathers are at a marginally higher risk of being physically abused than children who live with only their mothers.

Family Characteristics

Poverty: Poverty has been found to be a significant predictor of experiencing physical abuse. In the NIS, as income increases, the rate of physical abuse decreases through all income categories.

Number in household: The number of individuals in the household has been found to be a risk factor for physical abuse. One study demonstrated that abused children lived in larger households (average 4.1 members) than children who were not abused (average 3.6 members). ^, Another study determined that children who live with four or five children are more likely to be physically abused than children who live in smaller or larger households.

Domestic violence: Most studies examining the relationship between domestic violence and child abuse have shown that children who live in families where there is domestic violence are at an increased risk of experiencing physical abuse. However, one report using a population-based survey found little relationship between domestic violence and physical abuse, but very strong relationships with other forms of child maltreatment. Another study demonstrated that poor marital quality was associated with physical abuse. Parental conflict and domestic violence might also be associated with physical abuse.

Corporal punishment: Parents who spank have been shown to be more likely to be physically abusive. Two studies have shown that most physically abusive acts are either a result of escalated discipline or in response to a specific child misbehavior. ^, One study using cross-sectional data demonstrated that the risk of physical abuse increases with increasing frequency of spanking and the use of an object (e.g., belt or switch) on the buttocks.

Neighborhood Characteristics

Children who live in impoverished neighborhoods are more likely to be physically abused than children who do not live in poor areas. ^, One study showed that decreasing neighborhood cohesion was associated with increasing rates of all types of maltreatment. A more recent study examined the relationship of social capital to subtypes of abuse. Decreases in social capital were shown to be associated with neglect and psychological abuse but there was no observed association between social capital and physical abuse. It has also been determined that the percentage of female headed households in a neighborhood and the concentration of alcohol vendors is positively associated with rates of physical abuse.

Head (Excluding Brain and Skull) and Neck

Face: Facial injuries are common among child physical abuse victims. One case series of 390 abused children seen as outpatients demonstrated that 59% of these children had orofacial injuries, most commonly bruising or abrasions of the face (95%). A similar study of hospitalized children showed that 41% had facial injuries, with the cheek as the most common site (30% of facial injuries). The eyes (25%), forehead (22%), nose (13%), and ears (10%) were also commonly involved. Injuries to the face involve lacerations, burns, and welts. Most of the children with facial injuries in these case series were under 5 years. ^, In both large cohort studies, the perpetrator was most often male—usually the father of the child or the mother’s boyfriend. ^,

Oropharynx: The mouth is a less frequent but important site of trauma from physical abuse. In some of the same studies cited above and in other retrospective cohort studies, the oropharynx was involved as a site of trauma in 1% to 11% of cases of physical abuse. Tooth fractures, avulsions, labial lacerations, frenulum lacerations, mucosal injury, palatal injury, and fracture of the mandible or maxilla have all been reported. ^,

Injury of the labial frenulum has received somewhat more attention as an injury suggestive of abuse. A systematic review found 19 studies meeting inclusion criteria. These included 30 cases of labial frenulum laceration. Most children suffered fatal abuse (27/30) and most were less than 5 years old (22/30). They identified two cases of frenulum laceration resulting from intubation, complicating the study of frenulum injury. Of the 30 cases, only two had an identified mechanism (direct blow to the face). It has been suggested that forced feeding or pulling of the lips may cause frenulum injury, but there were no documented cases in this review. The most serious limitation to this literature is the absence of cross-sectional or case-control data to understand the specificity or predictive value for frenulum injury and abuse.

Neck: Skin injury to the neck, mostly bruising and abrasions, were included in several studies of facial trauma. In one cohort study, bruising of the neck was identified in 12% and abrasion in 7% of physical abuse victims. One study found that of children hospitalized for abuse, 6% are found to have neck injuries. A series of pediatric cervical spine injuries at a trauma center found 3 of 103 injuries were due to abuse. Abuse injuries were all classified as spinal cord injury without radiographic abnormality (SCIWORA), underscoring the challenge of diagnosing cervical spine injuries in abuse victims. All three of these patients were infants, two suffered head injuries, and the third massive injuries of the chest, abdomen, and bones.

The epidemiology of abusive head trauma is discussed in Chapter 6 .

Visceral Injuries

Several case series have demonstrated a wide range of abdominal injuries from child abuse, including liver laceration, splenic laceration, renal contusion, and hollow viscus injury. Several retrospective cohort studies have examined all admissions to large hospitals for children with abdominal injuries. The rates of abuse among these series range from 11% to 19%. Children with inflicted abdominal injuries are more likely to have higher injury severity scores. Additionally they are more likely to have hollow viscus injury and extraabdominal injuries (such as bruises and rib fractures). Of all children coming to an emergency department with injuries, abdominal injury from abuse is extremely uncommon (<1%); however, of all abdominal injuries seen in the emergency department, 4% were classified as abuse.

Skeletal Injury

Estimates of fractures in physically abused children vary widely by setting (11%-31%). One study found that unsuspected fractures were identified by skeletal survey in 26% of children admitted to a children’s hospital for physical abuse. Unsuspected fractures were most common in children with suspected fractures and head injuries, but uncommon in children admitted with burns. Most unsuspected fractures were found in children less than 1 year (80%).

Rib fractures: Rib fractures are a common skeletal manifestation of abuse. Most abusive rib fractures occur in children less than 2 years old. One cohort study included only infants and found 82% of 39 infants with rib fractures were caused by abuse. Of the remaining seven cases not due to abuse, three were clearly due to unintentional injuries (one motor vehicle, one fall down stairs, and one crush injury), one to birth trauma, and three to bone fragility. A second study included 78 children with a total of 336 rib fractures. Sixty-two children were aged 3 years or younger and 82% were determined to have inflicted rib fractures by the child abuse team. The remaining 11 children (nonabused group) had postoperative rib fractures (5), skeletal dysplasias (3), osteoporosis of prematurity (2), or been in a motor vehicle crash (1). The positive predictive value of a rib fracture for abuse in children less than 3 years old is 95%, and if clinical and historical information are used to exclude children with other causes, the positive predictive value is 100%.

Many children with suspected abuse will undergo cardiopulmonary resuscitation (CPR) prior to first radiographic study. This raises the question of chest compressions as a potential cause of the rib fracture. A recent systematic review of rib fractures caused by CPR reviewed 427 studies, but only six met inclusion criteria. Of the 923 children who underwent CPR, only three had rib fractures, all anterior. Rib fractures from CPR are rare and only anterior fractures have been associated with CPR. A recent study, however, did identify subtle rib fractures at autopsy in 11% of resuscitated infants after the parietal pleura was stripped from the ribs. The fractures were anterior and lateral in location rather than posterior, and most were not visible before the pleura was removed.

Limb fractures: Among all children, femur fractures are rarely due to abuse. However, several studies have demonstrated that femur fractures among young children, especially preambulatory children, are more likely due to abuse. One study identified 139 children less than 4 years old with femur fractures. The overall rate of fractures due to abuse was 9% with an average age of 1.1 years in the abuse group and 2.3 years in the unintentional injury group. Children who are not yet walking were more likely to be victims of abuse. A case-control study of fractures from abuse and unintentional injuries found that 93% of abuse injuries were in children aged less than 1 year. This study found no differentiating characteristics either in fracture type or radiographic appearance. A study of a referral center’s trauma registry found that abusive injuries accounted for 67% of lower extremity fractures for children under 18 months old compared with 1% for children 18 months or older. Of children hospitalized with an abusive lower extremity fracture, 68% had femur fractures and 56% had tibia fractures. A study of a national administrative database found that 15% of all femur fractures in children under 2 years were coded as caused by abuse and almost no abusive fractures were reported among older children. It is clear from these studies that abuse should be considered as a potential cause of long bone fractures in young children, especially those who are nonambulatory.

Fractures outside of the axial skeleton and lower extremities can be due to abuse, but such injuries have been the subject of less research. In a national U.S. study of a probability sample from administrative data, 1053 children were hospitalized for abuse with 1794 fractures. The axial skeleton was the site of 50% of these fractures (59% skull, 37% rib, 3% vertebrae, 1% pelvis). Only 14% of fractures were to the upper extremity (45% humerus, 34% radius/ulna, 17% scapula/clavicle, 4% carpal/metacarpal). The lower extremity fractures accounted for 18% of these injuries (59% femur, 37% tibia/ fibula/ankle, 2% tarsal/metatarsal).

Skin Injury

Bruises: Skin injury is one of the most common presentations for physical abuse, with bruises by far the most common injury. However, bruises are an extremely common injury in all children. A large prospective study of children seen for nontrauma reasons found that 76.6% had recent skin injuries, mostly bruises, and 17% had five or more injuries. Epidemiological studies have been invaluable in characterizing normal versus abnormal bruising. Large case series of abused children, nonabused children, and case-control studies have been used to characterize normal and potentially abusive bruising.

Nonabused children rarely have bruises before starting to transition to independent mobility (<1%). The most common sites for nonabusive bruises are over the legs, bony prominences, and the head for infants and toddlers. Child abuse victims commonly have bruises (28%-98%). Bruises due to abuse tend to be greater in number, to be present with older injuries (i.e., scar or healing abrasion), and to be defensive in location (outer arm). Abusive bruises can carry the imprint of an implement such as a cord. Bruises that are high in number (studies suggest 10-15), unusual in location or pattern, or occurring in young children not yet walking should be considered for abuse or bleeding disorder.

Burns: Most epidemiological studies of burns compare cases of inflicted pediatric burns with unintentional burns. In series of hospitalized pediatric burn patients, the rates of abuse and/or neglect range from 4% to 16%. These studies often combine abuse and neglect. A burn registry has recently allowed epidemiological study of nearly all serious pediatric burns in the United States. This study found that 6% of children aged 12 years or younger admitted to burn units were suspected victims of abuse. The use of registries comes at the cost of detail, and the assessment of abuse is less clear and perhaps less standardized than a single center’s approach.

Inflicted burns are most often due to liquid scald (78% of inflicted burns versus 59% of unintentional burns). Abusive burns tend to be larger, involve younger children, have higher risk of mortality, and longer hospital stays. ^, They tend to be deeper and more often require grafting. ^, They more often involve both hands or both feet. Social stress is a prominent risk factor in these injuries. Victims of abusive burns are more often from unstable families, ^, from single parent families, live in poverty, ^, and have had prior involvement with protective services.