Primary Care, Vol. 20, No. 2, June 1993, 317-327.

Child Physical Abuse

Eli H. Newberger, MD


Child abuse is an age-old problem for which we have had documentary records for as long as we have had recorded history, but the problem has been intermittently rediscovered and forgotten. In 1962, when C. Henry Kempe and his colleagues published the provocatively entitled article, “The Battered Child Syndrome,” the modern acknowledgment of child abuse began in the United States. This article, and its medical reformulation (or medicalization), stimulated an editorial outcry in professional and lay media. It prompted action by the United States Children’s Bureau to convene a committee of experts to recommend ways of dealing with the newly rediscovered sociomedical problem. Because of the prevailing belief that child abuse was a relatively rare event, one that would be systematically diagnosed and reported by physicians if they were only to be required to do so, a model child abuse reporting law was proposed. By 1968, such laws were adopted in every state.3,4

Physicians were the principal designated reporters in these laws, and it was expected that they would cooperate in breaking the confidentiality strictures of the doctor-patient relationship in the informing of social welfare agencies and the police about their patients’ problems.

Implicit in the laws, as well, were assumptions about the nature of the service response to such disclosures of child abuse. American society would not tolerate children in jeopardy in their homes, argued the optimistic social welfare protagonists of the 1960s. When the plight of battered children became known and the public conscience was piqued, appropriate treatments and resources would surely follow.

The intervening years have seen the dispelling of numerous myths about child abuse,9 including the myths that all abusing parents are mentally ill, that doctors will make the diagnosis and report suspected abuse to mandated state agencies, and that services appropriate to the number and nature of life-threatening adversities to children will be mustered by governmental agencies. It is now manifestly clear that physician response to child abuse and neglect is variable. We are coming to understand that the social status and race of the child, and the nature of the practice setting (public or private) have as much to do with the actions of medical practice as injury to the child.9 Physician refusal to engage with agencies of the state on behalf of children may also be explained by doctors’ reluctance to expose their patients to the possibly incompetent intrusions of underfunded, capricious, and unpredictable child protection programs. In every jurisdiction, these agencies are now besieged by more case reports than they can competently address.5


Physicians face three principal ethical dilemmas in confronting child abuse. First is the dilemma implied in the Hippocratic axiom, “First of all, to do no harm.” One must, in these situations, weigh both the advantages to a child of abiding by the legal responsibility to make a child-abuse case report and consider its possible downsides for the child and family.

With regard to the doctrine of confidentiality, although the child abuse reporting statutes make clear one’s obligation to divulge private information, this obligation will nearly always come as a surprise to the families of child patients. One’s actions must thus be explained, and the family’s mistaken assumptions about the confidential nature of the medical office must be addressed.

The informed consent doctrine is also challenged by this work. No assurance can be given by a doctor about the possibly hurtful involvements in the families’ lives by social welfare agencies or the police. To acknowledge these ethical tensions means thinking through with care one’s obligations to one’s patients as a responsible practitioner and as a citizen under law.

Nothing substitutes in this area of practice for honest and serious communication with one’s patients and their families. If one is concerned to help, one must say so, and a physician can be an authoritative influence both on behalf of children and parents. A doctor’s status can have a very useful effect in getting protective service workers and agencies, for example, to do the right thing and in acquiring such vital services as child care for a family. I have addressed these issues in relation to the content of current practice in two recent articles.10,11

These ethical dilemmas are not easily reconciled. I do not advocate breaking the law, but I think our assumptions about child abuse, the physician’s role, and the reporting statutes must be seriously and thoughtfully addressed. I think we should also aggressively advocate for our patients to ensure that they receive the protection and help they need.


Child abuse is generally defined to include physical and sexual assaults on children, child neglect, emotional abuse, and deprivation of necessary physical and moral supports for a child’s development. (Child sexual victimization is treated in a separate article in this issue; this paper addresses physical abuse.)

A model for understanding child abuse focuses on stresses in the family, which originate in the child, in the social situation, and in the parent. These operate to produce a triggering situation that culminates in the child’s maltreatment. A schematic diagram is attached as Figure 1.14

A physician’s diagnosis of abuse is based on the interviews of the child and family, on the examination of the child, and on the findings from the laboratory.


The physician’s interview can be seen both as an opportunity to gather data that will guide physical examination, subsequent laboratory study, and as a method to establish a relationship with the parent and child. A sense of trust can be built on this relationship that will enable the family to use helpful services and interventions offered to ensure that whatever may have happened to the child will not happen again.

In this connection, it is important to emphasize the significance of isolation in the model. Many of these families do not have regular contact with people who can provide them with help in times of distress, and the physician’s interview may be one of the first experiences a parent has had in which someone inquires thoughtfully and sympathetically about their child and her or his situation.

Parents should be interviewed separately, when possible. This measure is both to give each a sense of the respect one holds for them and one’s interest in them as individuals, and importantly, to enable them to talk about sensitive information in their own lives, including the use (and abuse) of alcohol and other substances, and whether they themselves are victimized. Notably, women whose children are abused appear themselves to be victims of abuse by their partners in more than half of the cases.6

The interview can be constructed around the usual pediatric history. Here, the intention is both to gather specific data and to assemble a historical sequence of important events that may be associated with a child’s risk of victimization.

One can start the interview with a short discussion of the reason the child and parents are in the office, which can best be done with sympathetic, open-ended questions. When one turns to the history itself, it is well to begin with neutral and benign questions that have predictably objective answers, for example, with regard to the birth date and birth weight. One can ask where the child was born, and then turn to the circumstances of delivery, and the important question: “How was the pregnancy?”

This question often prompts women to reflect on the relationship with the child’s father, points of distress and discomfort in personal relationships more generally, and her hopes and expectations for the baby. If one asks in a sympathetic way questions such as “What was this time like for you as a person?” or “Was this a happy time for you?” in a caring fashion and listens thoughtfully and attentively to the responses, one will often hear precursors and harbingers of troubles to come.

In one’s interviewing style, it is very often of value to affirm the parent’s emotional reactions. For example, with comments like, “That must have been difficult for you” or “I can see it makes you sad to talk about that” one can offer affirmations that demonstrate that you have the ability to attend to the parent’s feelings. This measure will give you a chance to lay the groundwork for discussion later in the interview of sensitive material that pertains directly to the victimization of the child.

Do not hesitate to ask mothers about their relationships with their partners. In pregnancy, it is well documented that between 8% and 11% of women are abused by their partners.8 One can ask about whether the woman has been hurt by anyone during pregnancy just before posing the usual questions about the use of cigarettes, alcohol, and drugs.

When one addresses the postpartum period, the question, “What was your child like when she or he first came home from the hospital?” is often quite informative. How the parent responds to the child’s crying and her or his ability to deal with the normal provocations of infants may have important implications for the child’s subsequent care. It especially is well to be alert to persistently negative characterizations and to conflicts between adults about how best to care for the child. Not infrequently, a discussion of the child’s early life will lead a parent to reflect on his or her own childhood. One should be attentive to this discussion and ask for information about the parent’s own history of physical or sexual abuse. Parents’ abuse experiences appear to weigh heavily in children’s risk for victimization.

The approach to the possibility of abuse or neglect should be framed with sympathetic, open-ended questions such as, “What happened then?” and “What did she say to you?” If the interviewer is alert to the parent’s emotional responses and indicates a capacity to listen to them and support their expression, avoiding leading questions and telegraphing of a specific anticipated response, a reasonable chance of hearing specifics with regard to what actually happened to the child exists.

The task of the interview, however, emphatically is not to establish who did what, to whom, and with what. Such a detailed level of disclosure should not be sought in a medical interview. One’s task is not to be a criminal investigator.” Although it is well to ask for details about what the parent actually saw and heard, it is important to keep in mind the priority of the relational tasks of the interview. An open-minded, interested, and serious demeanor, with expressions of personal warmth and concern, will often lead to trusting communication. This communication, in turn, will give you important insights and leads about the parent’s giving of care.

It is well to ask if other professionals have been consulted, especially with regard to the seeking of medical services, in view of the frequent pattern of parent’s “shopping” for treatment, perhaps to avoid detection. Also, asking about other interventions that may have been provided, for example, by psychiatric care providers, alcohol treatment programs, or battered women’s services, can give a sense of past history that might not otherwise be offered.

Often it is at the end of the interview, sometimes in response to the question, “Do you have any questions for me?” that the most important information emerges. Therefore, one should always leave time for questions, and to listen for the information contained in a parent’s queries. Let parents know at this time that you are going to be available to them and that you do not want to lose touch. Indicate as well that you will be available to answer any subsequent questions they may have. Remember that in this area of medical communication, just as in informing patients about your diagnosis of a chronic disease or cancer, not everything can be done in one sitting.

Interviewing of children has been the subject of much recent discussion, especially with regard to the diagnosis of child sexual abuse (the reader is referred to the articles by Levitt and Rappley and Speare).1,13 In interviewing in child physical abuse, if a child is old enough to report her or his experience, it is well to have a separate, private child interview in which the child is asked specifically about who gives child care, whether anyone has been hurting them, and how they feel about being at the doctor’s for this visit. It is important here to be sensitive to the child’s own sense of vulnerability or shame, to be gentle and caring in one’s approach, and scrupulously to respect the child’s confidences.

“Editor’s note: There are divergent perspectives on investigating victims for potential physical or sexual abuse, where experts in the area of child sexual abuse may feel that no opportunity should be missed, many experts emphasize that the physician’s role is not to obtain an investigative interview but that this is best left to the child protection agencies and law enforcement personnel.

In fulfilling the reporting mandate it is important to keep in mind that statements from the victim may cause distress or embarrassment later on. Therefore, care should be taken to maintain the child’s sense of personal integrity and capacity to control information they have given concerning their experience.

Victims of physical and sexual abuse nearly always suffer a profound sense of powerlessness. If information derived from their direct disclosure to the physician is transmitted, without editing, to child protection workers or criminal investigators, the expected confidentiality of the physician’s interview may be violated and the child may feel a subsequent sense of betrayal when in the civil or criminal court preceding’s the physician’s report is introduced into evidence. Therefore, it is appropriate always to control the attribution of one’s sources and to leave to the investigative agencies their proper interviewing tasks and functions.


The physical examination of a suspected victim of child physical abuse should be conducted in an unhurried way. Efforts should be made to make the child at ease. To the extent possible, prior to the examination, the child should be protected from hearing conversations, including medical history, in which the origins of injuries, and especially conflicting stories about them, are reviewed. Especially with older children, the examiner should take pains to respect the child’s sense of privacy and, if possible, to protect against embarrassment during the course of the examination. Particularly if there has been a seeming incongruity between the offered explanation for the child’s injuries and the anticipated findings, the child may feel awkward and want to avoid possibly further frustrating or enraging his or her parents.

The room should be well lighted, and if possible, the examination should be conducted in natural light. The purpose of this measure is to discern nuances of skin color and texture and to ensure, to the greatest extent possible, that one is not perceiving artifacts of shadow or lighting.

The examination is conducted in the usual way, generally beginning with the head, eyes, ears, nose, and throat. (Do not simply focus on areas of obvious injury.) Palpitation of the head to identify subcutaneous hematomas and examination of the skin behind each ear are frequently useful maneuvers to identify abusive injuries. The neck should be examined with care as well. Especially in infants, the frenulum and oropharynx are vulnerable to bottles and spoons being jammed into the mouth; older children may suffer direct dental impact from blunt trauma to the face.

Facial injuries should be noted and simple drawings made of the distribution, shape, and color of each observed lesion.

Retinal examinations should be conducted with care in any situation in which head trauma may be suspected. Because the general medical examiner using the indirect ophthalmoscope may not be able to get a full impression of the eye grounds, frequently ophthalmologic consultation is sought. Identification of retinal hemorrhages is often made only by a specialist, and the importance of this perception cannot be underestimated, especially in the making of the diagnosis of shaken infant syndrome.

As one proceeds to examine the chest, abdomen, and genitals, it is important fully to disrobe each portion of the body. Injuries of particular concern are often found on the upper, middle, and lower back and across the buttocks. These should be documented in writing and in a drawing. taking care to identify all nuances of color of bruises and distinct shapes of any identified lesions. The absence of abdominal bruises should emphatically not be taken to signify an absence of underlying organ injury. Only here on the child’s body is the skin so distensible that quite substantial traumatic force can be sustained without the development of overlying skin signs. Palpation must be done with care, and abdominal ultrasound examination and surgical consultation should be considered if history and presenting symptoms suggest intra-abdominal injury.

The genital examination as well should be conducted with the child’s perineum completely unclothed. Note that physical and sexual abuse may often be found together; any bruising, signs of discharge, possible hymeneal irregularities, and other signs should be noted and recorded (refer to the article by Levitt et al).

Examination of the extremities should include careful observation of all surfaces, including palms and soles, with careful descriptions of any lesions.


Many modes of injury are grouped under the concept of abuse, and a recent review summarizes many of the more bizarre ones.12 Keep abuse in mind when a child presents with an unusual symptom pattern. A listing of the categories in this compendium should suffice briefly to indicate the panoply of recently described injuries: fatal aspirations, intentional microwave oven burns, thirsting and hypernatremic dehydration, toxic ingestions, pancreatic injuries, “tin ear” syndrome (unilateral ear bruising, ipsilateral cerebral edema, and hemorrhagic retinopathy), and Munchausen’s syndrome by proxy, in which symptoms are created in children by their parents (in nearly all reported cases, by mothers) by inappropriate use of medication, withholding of needed drugs, or imposing on children toxic substances, injectables, irritants, or artifactual symptoms in an effort to provoke medical diagnostic explorations and unneeded medical treatments.7 These injuries are presumed to bring to the parent the satisfaction of the physician’s attention. Many of these cases culminate in fatalities if intervention is not made; not a few, however, appear to be more benign variants of parental neediness in which a child’s continued symptoms represent an acceptable ticket of entry into the physician’s office so that a parent’s distress can be given vent and attention can be paid to an otherwise isolated parent and child.


The history and physical examination can guide the approach to laboratory studies to exclude the many competing diagnostic possibilities that may masquerade as child abuse. Table 1 summarizes in schematic form the principal diagnostic entities, clinical findings, and laboratory, radiographic, and specialized studies.


The physician’s main responsibility in child abuse management is diagnostic. In some areas, however, the physician may indeed have to play a continued intervention role, which should be tailored appropriately depending on available resources.

The principal intervention approaches to addressing child abuse are summarized subsequently:

  1. Social casework is the core service offeted in child protection agencies. Here, social workers assess the needs of the child and family with a focus on the protection of the child and offer and coordinate counseling and other services designed to improve parental functioning and parent-child relationships.


Clinical Findings

Differential Diagnosis

Differential Tests

Cutaneous lesions


Von Willebrand’s disease
Anaphylactoid purpura
Purpura fulminans
Ehlers-Danlos syndrome

Prothrombin time,
partial thromboplastin time
Bleeding time
Rule out sepsis
Rule out sepsis

Local erythema or bullae

Staphylococcal impetigo
Bacterial cellulitis
Pyoderma gangrenosum
Photosensitivity and phototoxicity reactions
Herpes, zoster or simplex
Epidermolysis bullosa
Contact dermatitis, allergic or irritant

Culture, Gram stain
Culture, Gram stain
Culture, Gram stain
History of sensitizing agent, oral or topical

Clinical history and characteristics
Skin biopsy
Clinical characteristics

Ocular findings
Retinal hemorrhage

Shaking or other trauma
Bleeding disorder

Coagulation studies


Conjunctival hemorrhage

Bacterial or viral conjunctivitis
Severe coughing

Culture. Gram stain

Orbital swelling

Orbital or periorbital cellulitis
Metastatic disease
Epidural hematoma

Complete blood count, culture, sinus radiographs Radiograph, CT scan; CNS examination
Radiograph. CT scan; CNS examination


Urinary tract infection
Acute or chronic forms of glomerular injury
(e.g., glomerulonephritis)
Hereditary or familial renal disorders
(e.g., familial benign recurrent hematuria)
Other (e.g., vasculitis, thrombosis, neoplasm, anomalies, stones, bacteremia, exercise)

Rule out other disease
Renal function tests, biopsy


History, cultures. Intravenous pyelogram

Acute abdomen

Intrinsic gastrointestinal disease
(e.g., peritonitis, obstruction, inflammatory bowel disease, Meckel’s diverticulum)
Intrinsic urinary tract disease (infection, stone)
Genital problem (e.g., torsion of spermatic cord, ovarian cyst)
Vascuiar accident, as in sickle cell crisis
Other (e.g., mesenteric adenitis. strangulated hernia, anaphylactoid purpura, pulmonary disease, pancreatitis, lead poisoning, diabetes)

Rule out other disease
Radiographs, stool tests, and others

Culture, intravenous pyelogram
History, physical examination. radiograph laparascopy(?)
Angiography, sickle preparation
As appropriate

Osseous lesions
Fractures (multiple or in various stages of healing)

Osteogenesis imperfecta
Birth trauma


Status after osteomyelitis or septic arthritis Neurogenic sensory deficit

Radiograph and blue sclerae
Nutritional history
Birth history
Decreased alkaline phosphatase
Complete blood count, bone marrow
Bone marrow, biopsy

Physical examination

Metaphysea lesions, epiphyseal lesions, or both


Menkes syndrome

“Little League” elbow
Birth trauma

Radiograph and nutritional history
Decreased copper, decreased ceruloplasmin

Subperiosteal ossification

Osteogenic malignancy
Infantile cortical, hyperostosis
Osteoid osteoma

Radiograph and biopsy
Serology tests
No metaphyseal irregularity
Response to aspirin
Nutritional history

Sudden infant death syndrome

Asphyxia (aspiration, nasal obstruction, laryngospasm, sleep apnea)
Infection (botulism?)
Cardiac arrhythmia?
Metabolic abnormality
(calcium? magnesium?)
Hypersensitivity to cow’s milk protein

“Near-miss” history

Cultures, bacterial and viral Immunoglobulins
Ca++, Mg++

From Bittner S, Newberger EH: Pediatric understanding of child abuse and neglect. Pediatr Rev 2:197-208, 1981;with permission.

  1. Psychotherapy is often provided to individuals, couples, and families, and it should be considered for children in all cases of abuse. This recommendation is made with a view to addressing the important child developmental and psychiatric burdens associated with the victimization experience.
  2. Health services include pediatric continuing care, public health nursing, and adult medical and dental care. Victims of child abuse appear to suffer more than the usual impacts of acute and chronic illness, and their medical needs are often neglected, both by parents and by social welfare agencies, particularly if they may be placed in foster home care. Systematic efforts should always be made to ensure that a child has her or his health needs documented and met.
  3. Family-focused interventions include homemaker services, parent aide programs, and homebuilders. These are key ingredients in what are coming to be characterized as “family preservation services,” which are offered to prevent family breakup, often at the threshold of children’s being separated from their parents’ care. Other interventions of proven value are child care services and alcohol and substance abuse treatment for the families who need them. Substitute care, which includes foster home and institutional care, may be given on a voluntary or court-mandated basis. It should be considered as a service, although too often it is proposed as a punishment. Parents Anonymous, a self-help voluntary movement with chapters and hotlines throughout the United States, offers opportunities for parents who can benefit from talking through their problems with others.
  4. Battered women’s services are often essential in providing protection to children, and it is necessary to be informed about the network in one’s community.
  5. The criminal justice system can also be of tremendous value in protecting children, especially in the face of criminally deviant behavior by caregivers whose actions cry out for stringent social controls. The deterrent value of criminal sanctions in preventing child abuse in society more generally is a subject of some debate. A disproportionate share of governmental resources appears to go into investigation and prosecution, as opposed to the provision of help.5


The diagnosis and treatment of child abuse is one of the more personally and professionally challenging aspects of primary care medicine. In this area of practice, as elsewhere in the profession, success is often connected to the clinician’s informed intelligence, mature attitude, and professional demeanor. When one maintains a thoughtful and skeptical approach to information; displays warmth, respect, and calm to children and parents; and when one’s findings and impressions are shared with care and accuracy, good information will be forthcoming and, hopefully, excellent management will follow.


  1. Bentovim A, Elton A, Hildebrand J, et al: Child Sexual Abuse Within the Family: Assessment and Treatment. London, Wright, 1988
  2. Bittner S, Newberger EH: Pediatric understanding of child abuse and neglect. Pediatr Rev 2:’197-208,1981
  3. Gershenson CP: Child maltreatment and the federal role. In Gil D (ed): Child Abuse and Violence. New York. AMS Press, 1979. pp 18-36
  4. Kempe CH. Silverman FN. Steele BF, et al: The battered child syndrome. JAMA 181:17-24, 1962
  5. Krugman R: Child abuse and neglect: critical first steps in response to national emergency. Am J Dis Child 5:513-515. 1991
  6. McKibbon L. DeVos E, Newberger EH: Victimization of mothers of abused children: A controlled study. Pediatrics 84:531-535, 1989
  7. Meadow R: Munchausen syndrome by proxy. The hinterland of child abuse. Lancet 2:343, 1937
  8. Newberger EH, Barkan SE, Lieberman ES, et al: Abuse of pregnant women and adverse birth outcome: Current knowledge and implications for practice. JAMA 267:2370-2372, 1992.
  9. Newberger EH: Child abuse. In Rosenberg ML. Fenley MA (eds): Violence: A Public Health Approach. New York. Oxford University Press, 1991, pp 49-78
  10. Newberger EH: Intervention in child abuse. In Schetky DH, Benedek EP (eds): Clinical Handbook of Child Psychiatry and the Law. Baltimore, Williams & Wilkins, 1992. pp 145-161
  11. Newberger EH: Pediatric interview assessment of child abuse: Challenges and opportunities. Pediatr Clin North Am 37:943-954, 1990
  12. Reece RM: Unusual manifestations of child abuse. Pediatr Clin North Am 37:4, 1990
  13. Walker CE, Bonner BL, Kallfman KL: The Physically and Sexually Abused Child: Evaluation and Treatment. New York, Pergamon, 1988
  14. White KM, Snyder JC. Bourne R, et al: Treating Child Abuse and Family Violence in Hospitals. Lexington, MA, Lexington Books, 1989

From the Family Development Program. Children’s Hospital, Boston, Massachusetts