Medical Considerations in the Diagnosis of Child Sexual Abuse

Felicity Goodyear-Smith^*

ABSTRACT: There are no medical signs in the vast majority of sexual abuse cases.  Many findings promoted as physical indicators of abuse have been shown to be present in nonabused children.  In particular hymenal openings said to measure more than 4 mm, genital rashes and redness, and anal reflex dilatation have been demonstrated to be unreliable medical indicators.  Children can be harmed both by unnecessary invasive investigation (including general anesthesia) and by subsequent interventions if the allegations are false.  Doctors must insure that they have an empirical basis for the interpretation of their findings, and that they do not allow someone else's belief that a child has been abused to color their clinical judgment.  Describing normal findings as "consistent with abuse" is decried.  This practice is likely to mislead a court to erroneously believe that there is physical evidence supportive of abuse.
  

Most cases of sexual abuse are diagnosed on historical and behavioral evidence and not on physical findings (Kivlahan, Kruse, & Furnell, 1992).  However, investigation of sexual abuse will often involve a medical examination to look for physical evidence of abuse.  Such information will be eagerly sought by the prosecution, as physical findings represent relatively "hard" evidence compared to psychological assessment and "disclosure" interview findings.  Also, the courts will generally give considerable weight to a physician testifying in support of an abuse allegation.

The majority of sexual abuse cases involve activities such as genital fondling and not penetration of the vagina or anus, and do not cause any marks or damage to the tissues. A diagnosis of sexual abuse is therefore infrequently made solely on medical findings (Lawton, Goodyear, & Stringer, 1987; Royal College of Physicians, 1991).  Sexual activity can be proved if an underage pregnancy has occurred, or if a sexually transmitted disease is detected, but these are relatively rare events.  Finding semen on the genital area also indicates sexual contact, but this is only possible if the child is examined soon after the alleged event (within 72 hours at the outside) (Gabby, Winkleby, Boyce, Fisher, Lanchester, & Sensabaugh, 1992).  In reality, the vast majority of alleged abuse cases present weeks, months or years later.

If a young child does have the vagina or anus penetrated by fingers or a penis, bruising, tearing and bleeding are likely.  It seems probable that the child would also suffer considerable discomfort for the next couple of days or so, especially when urinating or defecating.  Anal penetration by the penis results in severe exacerbation of pain when the child next attempts to defecate (Paul, 1990).  The perineal region has a good blood supply, and usually heals rapidly.

Whether such injuries cause permanent scars detectable months or years later is currently being researched.  McCann, Voris and Simon (1992) studied three children who had suffered genital lacerations from a single isolated episode of assault, one requiring suturing.  They used a camera and colposcope to record their findings and followed up the appearance of the injuries for up to three years.  They found that in these cases, there was very little scar formation and signs of damage were difficult to detect after a couple of months.

Children who have bee n sexually abused on an ongoing basis may well show more obvious signs of trauma.  The vaginal and anal orifices might remain more open and show signs of scarring, although research evidence in this area is still sparse also.

Children's genital regions have not been routinely examined in medical examinations, and until the last decade virtually nothing was written on what normal vaginas and anuses looked like in childhood (Pokorny, Pokorny, & Kramer, 1992; McCann, Voris, Simon & Wells, 1989; McCann, Wells, Simon, &Voris, 1990).
  

Hymenal Findings

In 1983, Cantwell examined and measured the hymens of nearly 250 girls under 13 years of age who were treated at a Crisis Care Unit in Denver.  She reported that 75% of those with horizontal openings greater than 4mm had been sexually abused.  Four years later she amended this figure to 80% (Cantwell, 1987).  This paper is often quoted by medical experts in court rooms and in the absence of any other studies, a horizontal hymen size greater than 4mm has been considered an indicator of sexual abuse.

Examination of this study reveals it seriously flawed, however.  First, the method of substantiating abuse was not made clear, and appears to include a number of girls who denied that they were victims.

Second, measuring hymenal size is not a simple procedure, and different examiners are likely to get different results.  To establish the diameter, a child's legs must be spread at the hips and the vaginal lips gently parted to expose the hymen.  Varying the amount of lateral pressure used to part the lips will distort the shape of the hymen and change the apparent diameter.  In addition, the method used for examination — supine with labial separation, supine with labial traction, or knee-chest — affects the measurement of the hymenal orifice diameter (McCann, Voris, Simon, & Wells, 1990).

Hymenal shape is very variable (Heger, 1985; Hyden, & Gallagher, 1992).  Some have several openings, they may be crescent-shaped, slit-shaped (horizontal or vertical), or very irregular.  The hymen might be thick and fleshy or a very thin membrane.  Not only is measurement impossible with any degree of accuracy, but Dr. Raine Roberts, Manchester, reported in the British Medical Journal in 1989 that "the hymen ... can vary, in the same child, from a pinhole to a centimeter, depending on whether she is relaxed or apprehensive, warm or cold."  A medical finding of a dilated hymenal opening must therefore be interpreted with great caution.

The diameter of an average index or middle finger is about 15 to 20mm.  An erect penis is 25 to 40mm in diameter.  The hymen is not a very elastic tissue, but even allowing for some stretching, the belief that any hymenal diameter greater than 4mm is an indicator of abuse is not commonsense.  The Royal College of Physicians (1991) states that a hymenal diameter of 15mm is supportive of abuse, although it should not be used as the sole basis for a diagnosis.

Unfortunately, the belief that hymenal diameters greater than 4mm indicate sexual abuse has permeated the field.  I have examined a number of medical reports of vaginal examinations where hymenal sizes less than 10mm have been reported by the examining physician as indicating probable abuse.

In one particular case, a woman doctor in Christchurch, New Zealand, examined three sisters and gave the opinion that they had all probably been molested.  She claimed that her examination of the 5-year-old revealed "a transverse vaginal diameter of 5mm, and no evidence of a hymen" which she found "highly suggestive of penetration."  The 9-year-old had a transverse vaginal opening of 3.5mm, with hymenal remnants, which she concluded was "suggestive of some interference to the vagina," and the 10-year-old had a transverse opening of 6mm, with no definite hymen, which she believed was "strongly indicative of vaginal penetration."

The three girls were then subjected to a number of sexual abuse assessments.  In her first interview session, the eldest girl was told that the doctor's examination showed that she had been the victim of "bad touching" and had a "hurt between her legs."  Despite being repeatedly questioned about who had caused the "hurt," she continued to deny any molestation.  Even after two counselors performed a role play with her about a "father who hurts kids between their legs" she was adamant that nothing like that had happened to her.  Sadly she was not believed and all three children were placed in a foster home.  Their father was charged with sexual violation of all his daughters, especially the eldest.  It was a year and a half before his case was heard in court, where he was acquitted on all charges.
  

Other Female Genital Findings

Examining doctors often claim that rashes and redness around the vaginal area are "consistent with sexual abuse."  While this may be technically true, there are so many other common causes of such findings that such a claim is likely to mislead a court into believing these findings mean sexual abuse has probably occurred.  In fact, such genital irritation is also consistent with no sexual abuse.  Scratching, masturbating, inadequate washing, irritating soaps and bubble baths, tight-fitting underpants, threadworm, thrush and other nonsexually transmitted infections can all result in redness and irritation.  So can a number of less common causes such as foreign bodies inserted in the vagina (Emans & Goldstein, 1980).

Genital examination of little girls is generally done with the child in one of two positions: knee-chest, where the child is asked "to lie on her tummy with her bottom in the air," or the frogleg position, where she lies on her back, sometimes propped up with her parent sitting behind her, with her legs spread open.  In my experience, once good communication and rapport has been established, most children will tolerate such an examination with little complaint.

There are a few situations, however, where an examination needs to be carried out under a general anesthetic.  In 1984, in conjunction with an Auckland pediatrician specializing in sexual abuse work, I established guidelines in New Zealand for when examination under anesthetic (EUA) is indicated:

It was emphasized that examination under general anesthetic should be a very rare occurrence.  These guidelines were distributed to all New Zealand registered medical practitioners (Lawton et al, 1987).

Unfortunately, these guidelines are not always adhered to, and it appears that children undergo general anesthetics for what I believe are unjustifiable reasons.  In one particular case, a 3-year-old girl in my care was the subject of a custody dispute.  Her father was seeking to return to his home in Europe with his daughter, and alleged that her new stepfather had sexually abused her.  A Family Court hearing did not uphold the allegation, and joint custody was awarded.  The father was upset at the decision, and continued to present the child at several Auckland agencies dealing with sexual abuse.  Often these centers were unaware of previous proceedings, and therefore this child had continuing assessments and interventions for alleged abuse, despite the court decision.

On one occasion when the child was staying with him, her father noticed a small warty lesion on her mons pubis, in the area just above and to the right of the clitoris.  Without informing her mother, he took her immediately to the Child Protection Team at the Children's Hospital.  The Team rapidly decided that this child had almost certainly been sexually abused, and that the wart should be removed under a general anesthetic to be sent to the virology lab for typing (to see if it was of a sexually transmitted type).  They were not going to tell the child's mother, as she was considered to be a "colluder" with her current husband, the alleged offender.  Fortunately, she was informed by a social worker who knew her, and she was able to be present while her daughter underwent this procedure.  I discussed the case with the hospital doctors, who informed me that if the wart was of a sexually transmitted type, the child would undergo a diagnostic interview by the hospital social worker.  The stepfather was also a patient of mine.  He had no history of genital warts, and a genital examination confirmed that none were now present.  Even if the wart was thought to be a sexually transmitted type, therefore, he could not be suspected of abuse.

Previous discussions with the hospital virologist had assured me that even if a wart was identified as a sexually transmittable type, there are nonsexual ways by which it can be transmitted.  Its presence is therefore not definitive evidence of sexual abuse.  Review of recent literature in fact demonstrates that only a minority of children with anal-genital warts have been sexually abused (Derksen, 1991; Gutman, Herman-Giddens, & Phelps, 1992).  I believed that a general anesthetic was not warranted in this child's case.

Despite my expressed concerns, the little girl duly underwent a general anesthetic during which her wart was cut off, her hymen measured and swabs taken for STD testing.  No abnormalities were detected, and typing of the wart did not suggest sexual transmission.  The hospital informed me that they still wished to refer her for further sexual abuse assessment.  I made some of the workers aware of how much intervention this child had already undergone, and eventually no further action was taken.

A month later the little girl's mother brought her to my surgery.  Her warty lesion had regrown and was now larger than the original (about 2 mm diameter).  It had the appearance to me of a benign wart-like infection called molluscum contagiosum.  The child attended a preschool where this condition was common.  Naked play between the children made this a very likely source of the infection.  I treated the lesion with a brief application of liquid nitrogen, which caused the little girl slight discomfort but which she tolerated well without any anesthetic.  A few days later the wart dropped off and has not recurred.  I also inspected her hymen at that time with no discomfort to her.  She would have easily tolerated my taking of swabs if this had been required.

I believe this child was subjected to an unnecessary general anesthetic, which was not required for either diagnosis of sexual abuse or for treatment of her wart.  She underwent a potentially life-threatening procedure as well as suffering the distress of hospital intervention.  It seems likely that too many children's lives are being put at risk from undergoing general anesthetics for which there are inadequate indications.
  

Anal Findings

The other area of contention in medical examination is the physical signs of anal abuse. In particular, the argument centers around a phenomenon of anal gaping called reflex anal dilatation (RAD).  Briefly, this involves gently parting the buttocks and observing the anus for half a minute.  Usually, the sphincter on the outside of the anus will contract and then dilate, as pressure is maintained.  Sometimes the inside sphincter will then also relax giving a view right into the rectum.  It is this response that has been named RAD.

British pediatricians Drs Hobbs and Wynne reported that RAD was present in 42% of anally abused children they examined, and claimed that it was an important indicator of abuse (Hobbs & Wynne, 1986, 1989).  They stated that they had not witnessed RAD in nonabused children.  They also claimed that splits or fissures around the anus are very rare in the nonabused child.

These findings became the basis of a belief in some circles that RAD is proof of anal abuse.  More recent studies and observations, however, refute Hobbs and Wynne's findings.  One study observed the phenomenon in nearly half of the nonabused children they examined (McCann et al, 1989).  It also appears to be more common when a child is constipated and has feces sitting higher in the bowel (Sunderland, 1987; Royal College of Physicians, 1991).  Many doctors also report that they have commonly seen anal fissures in nonabused children (Freeman, 1989; Kean, 1989; Royal College of Physicians, 1991).

Unfortunately, Hobbs and Wynne's theories regarding the relevance of anal reflex dilatation was taken to be established fact by a number of doctors examining children.  In some centers it became policy for all children to undergo genital and anal examinations, no matter what medical problem they had come with.

This practice resulted in the false epidemic of sexual abuse cases in Cleveland, England, which was to receive worldwide attention from the media.  In 1987, two pediatricians working at the Middlesbrough General Hospital in Cleveland, Drs. Marietta Higgs and Geoffrey Wyatt, diagnosed 121 cases of alleged sexual abuse of children in the space of five months.  Their diagnoses were made largely from medical examination findings of reported hymenal irregularities and RAD.  Many of these children had come to the hospital for treatment of complaints such as asthma, and there was no other evidence suggesting that they had been abused (Bernard, 1988; Woods, 1988). Dr.

Higgs held the view that one in ten children are sexually abused, and sincerely believed that her findings proved the abuse.

Despite denials from bewildered and distraught parents, the children were immediately taken from their homes, initially to a hospital and later into care by the social services.  As the numbers escalated, distressed parents sought media and political support.  Eventually a public inquiry was called.  The Cleveland inquiry, headed by Lord Justice Butler-Sloss, found that most of the allegations were unfounded, and the children were returned to their families (Butler-Sloss, 1988).  The process was, however, very traumatic to all concerned, and the children and their parents did not emerge unscathed by the experience.

Despite the findings of the inquiry, and the evidence coming forward from a number of reputable medical sources regarding the unreliability of relying on medical signs, such as RAD, to diagnose abuse, many agencies still maintain their use is valid.  Dr. Higgs still has many supporters within the field who believe her diagnoses were justified.  Some books written about the Cleveland affair (Campbell, 1988; La Fontaine, 1990) present the view that the allegations in the Cleveland case were founded, despite overwhelming evidence that abuse was not substantiated in the vast majority of the cases.
  

Conclusions

There are no physical signs of abuse to be found in the vast majority of sexual abuse cases.  Medical findings supporting or proving abuse are not as clear cut as may be expected.  Many of the medical indicators advocated are frequently found in non-abused children.  The ubiquitous practice of describing completely normal examination findings as being "consistent with abuse" is likely to be misunderstood in a courtroom as evidence supporting an allegation.  Lay people serving as jurors are particularly apt to be misled by medical experts giving such testimony.

Physicians examining a child for possible sexual abuse are likely to have been briefed by other workers who have already decided that the child has been sexually abused.  Many social workers and psychologists believe that false allegations are extremely rare and that "children never lie about abuse," and see their role as a "validator" that the abuse has occurred.  Once a belief that sexual abuse has taken place has become entrenched, very little can be done to sway the believers otherwise.  To even suggest the possibility of a false allegation is often to invite an emotional outburst and accusations of condoning or even colluding with abuse.  Actions and decisions may subsequently be made without scientific substantiation of the allegations.

Doctors called upon to perform forensic sexual abuse examinations should have up-to-date information on the range of normal for nonabused children.  They should be very cautious on how they interpret their findings, and insure that they have an empirical basis for their claims.  Children can be seriously harmed both by invasive investigative practices and by subsequent interventions when the allegations are unfounded.  Physicians must always have in mind the Hippocratic vow, primum non nocere: first do no harm.
  

References

Bernard, V. (1988). Implications of the Cleveland child inquiry: Child sexual abuse demands cooperation. British Medical Journal, 297, 151-152.

Butler-Sloss, E. (1988, June 6). Report of the inquiry into child abuse in Cleveland, 1987. Presented to Parliament by the Secretary of State for Social Services by Command of Her Majesty. London. England: Her Majesty's Stationery Office.

Campbell. B. (1988). Unofficial Secrets. Child Sexual Abuse: The Cleveland Case (). London: Virago Press.

Cantwell, H. (1983). Vaginal inspection as it relates to child sexual abuse in girls under thirteen. Child Abuse & Neglect, 7, 171-176.

Cantwell, H. (1987). Update on vaginal inspection as it relates to child sexual abuse in girls under thirteen. Child Abuse & Neglect, 11, 545.

Derksen. D. J. (1992). Children with condylomata acuminata. The Journal of Family Practice, 34, 419-423.

Emans, J., & Goldstein D. (1980). The gynecologic examination of the prepubertal child with vulvovaginitis: Use of the knee-chest position. Pediatrics, 65, 758-760.

Freeman. N. (1987. October 31). Child sexual abuse (letter to the editor). The Lancet, p. 1017.

Gabby. T., Winkleby. M., Boyce. T., Fisher, D., Lanchester, A., & Sensabaugh, G. (1992). Sexual abuse of children: The detection of semen on skin. American Journal of Diseases in Children. 146, 70~703.

Gutman, L., Herman-Giddens, M., & Phelps. W. (1992). Transmission of human genital papillomavirus disease: comparison of data from adults and children. Pediatrics, 91, 31-38.

Heger. A. (1985). Child sexual abuse: A medical view. Los Angeles: United Way. Inc., pp. 2-3.

Hobbs, C., & Wynne, J. (1986, October 4). Buggery in childhood. The Lancet, pp. 792-796.

Hobbs, C., & Wynne J. (1987, October 10). Child sexual abuse: An increasing rate of diagnosis. The Lancet, pp. 837-841.

Hobbs, C., & Wynne, J. (1989). Sexual abuse of English boys and girls: The importance of anal examination. Child Abuse & Neglect, 13, 195-210.

Hyden, P., & Gallagher, T. (1992). Child abuse intervention in the emergency room. Pediatric Clinics of North America, 39, 1053-1081.

Kean, H. (1987, October 31). Child sexual abuse (Letter to the editor). The Lancet, p. 1018.

Kivlahan, C., Kruse, R., & Furnell, D. (1992). Sexual assault examinations in children: The role of a statewide network of health care providers. American Journal of Diseases in Childhood, 146, 1365-1370.

La Fontaine, J. (1990). Child sexual abuse (). England: Polity Press.

Lawton, M., Goodyear, F., & Stringer, P. (1987). Sexual Assault Examinations — A Guide for Medical Practitioners. Wellington: DSIR

McCann, J., Voris, J., & Simon, M. (1992). Genital injuries resulting from sexual abuse: A longitudinal study. Pediatrics, 89, 307-317.

McCann, J., Voris J., Simon, M., & Wells, R. (1989). Perianal findings in prepubertal children selected for nonabuse: A descriptive study. Child Abuse & Neglect, 13, 179-193.

McCann, J., Voris, J., Simon, M., & Wells, R~ (1990). Comparison of genital examination techniques in prepubertal girls. Pediatrics, 85, 182-187.

McCann, J., Wells, R., Simon, & Voris, J. (1990). Genital findings in prepubertal girls selected for nonabuse: A descriptive study. Pediatrics, 86, 428-439.

Paul, D. M. (1990). The pitfalls which may be encountered during an examination for signs of sexual abuse. Medical Science and the Law, 30(1), 3-11.

Pokorny, S., Pokorny, W., & Kramer, W. (1992). Acute genital injury in the prepubertal girl. American Journal of Obstetrics and Gynecology, 166, 1461-1466.

Royal College of Physicians of London (1991). Physical Signs of Sexual Abuse in Children. Salisbury, Wilts: Cathedral Press Ltd.

Sunderland, R. (1987, October 31). Child sexual abuse (Letter to the editor). The Lancet, p.1018.

Woods, M. (1988). Child abuse — Fact and fantasy. Family, 8-9.