Alleged Behavioral Indicators of Sexual Abuse

Ross Legrand, Hollida Wakefield, and Ralph Underwager^*

A frequent trigger for suspicion of possible sexual abuse is one of the so-called behavioral indicators.  Many lists of behaviors believed to be caused by sexual abuse have been widely publicized with the instruction to look for sexual abuse when a behavior on the list is observed.  These behavioral indicators may be observed by a teacher at a preschool, a neighbor, a parent, or by a spouse in the midst of a bitter custody battle.  A single behavior, i.e., aggression toward another child, may result in an adult questioning a child in a way that inadvertently elicits statements confirming abuse.  Also, once the process has begun, the the presence of one of the behavioral indicators may be used as evidence to validate the abuse.

As concern about child sexual abuse mounted in the United States, more and more people have been asked to help detect and prevent it.  Intervention is particularly demanded in the case of very young children, who are not only unable to defend themselves but are also less able to give voice to their problems.  Television programs, workshops, newspapers, pamphlets, and magazines are asking parents, relatives, teachers, physicians, clergy, day care workers, and neighbors to be alert to the physical and behavioral signs of sexual abuse in children and to report their suspicions to medical or legal authorities.

One list of behavioral symptoms was published by the prestigious Journal of the American Medical Association (JAMA, 1985) and has been widely reprinted.  We are advised to look for children who have one or more of the following behaviors:

1. Become withdrawn and daydream excessively
2. Evidence poor peer relationship
3. Experience poor self-esteem
4. Seem frightened or phobic, especially of adults
5. Experience deterioration of body image
6. Express general feelings of shame or guilt
7. Exhibit a sudden deterioration in academic performance
8. Show pseudomature personality development
9. Attempt suicide
10. Exhibit a positive relationship toward the offender
11. Display regressive behavior
12. Display enuresis and/or encopresis
13. Engage in excessive masturbation
14. Engage in highly sexualized play
15. Become sexually promiscuous

Such a list is less than helpful.  How does a deterioration in body image manifest itself?  If a neighbor is unsure whether sexual abuse has occurred next door and who the offender might be, should any sign of a too positive relationship with an adult or its opposite, fear of an adult, be considered suspicious?  Because acting too maturely or too immaturely are both symptoms; can a minister be expected to know the range of maturity a child of a given age should display?

Here is another list of symptoms. This list is offered by Sgroi (1982):

1. Overly compliant behavior
2. Acting-out aggressive behavior
3. Pseudomature behavior
4. Hints about sexual activity
5. Persistent and inappropriate sexual play with peers or toys or with themselves
6. Sexually aggressive behavior with others
7. Detailed and age-inappropriate understanding of sexual behavior
8. Arriving early at school or leaving late with few, if any, absences
9. Poor peer relationships or inability to make friends
10. Lack of trust, particularly with significant others
11. Nonparticipation in school and school activities
12. Inability to concentrate m school
13. Sudden drop in school performance
14. Extraordinary fear of males, of strangers or of being left alone
15. Complaints of fatigue or physical illness, which could mask depression
16. Low self-esteem

There is some agreement between these two lists, such as problems with various fears, peer relationships, and sexual knowledge or activity that seem beyond a young child's years.  There are also symptoms that appear on one list, but not the other, such as daydreaming, depression, and being too compliant.  We might decide that behaviors found on both lists are better indicators of sexual abuse than those found on only one.  Or, if a child shows many symptoms rather than only one, perhaps that fact should strengthen our faith in the diagnosis of abuse.  Unfortunately there is no evidence that either of these strategies will be effective.

The task of detecting abuse is made more difficult because these two lists, long as they are, are not exhaustive.  A survey of many such compilations proposed by various experts (Cohen, 1985) adds the following behavioral signs to those listed above:

1. Loss of appetite
2. Clinging to a parent
3. Tics
4. Hypervigilance
5. Running away
6. Irritability
7. Difficulty with eye contact
8. Hyperactivity
9. Extreme interest in fire
10. Unprovoked crying
11. Taking an excessive number of baths
12. Suspiciousness
13. Sleepwalking
14. Sudden massive weight gain or loss
15. Excessive urination
16. Medical conditions such as pneumonia or mononucleosis
17. Confusion
18. Nightmares
19. A poor mother-daughter relationship
20. Overdependency

At this point it seems that nearly every problem behavior ever detected in children has been offered by someone as a sign of child sexual abuse.  The problem is the high probability that any normal child might at some point in childhood exhibit one or more of these behaviors and thereby risk being perceived as an abuse victim.  To spread these lists of behavioral indicators without appropriate cautions and information about their limitations can generate mistakes, confusion, over-reaction, and over-interpretation.

Note the following list of children's behaviors:

1. Depression
2. Overly dependent behavior
3. Aggression
4. Whining
5. Demanding
6. Lack of affectionate behaviors
7. Feminine (passive) aggressiveness
8. Encopresis
9. Anxiety
10. 1Neurotic problems
11. Anxiety about sexual matters
12. Problems of both over-and undercontrol

These symptoms, however, are not claimed as signs of child sexual abuse but as behaviors indicative of parental conflict in the home that may or may not lead to divorce (Emery, 1982).  Many of these symptoms overlap with the suggested signs for sexual abuse.  Neurotic problems could subsume a variety of fears, depression might manifest itself in withdrawal or other signs of sadness, and overcontrol could look a lot like pseudomaturity.

Moreover, the leading diagnostic manual for psychologists and psychiatrists, Diagnostic and Statistical Manual of Mental Disorders, Third Edition-Revised (DSM-III-R, 1987) describes clinical categories, such as childhood depression, separation anxiety, and disorders of adjustment that share many of the behaviors proposed as symptoms of sexual abuse.  Suggested behavioral indicators from the various lists are found in many different situations, including conflict between parents, divorce, economic stress, wartime separations, absent father, natural disaster, and almost any stressful situation children experience (Emery, 1982; Hughes & Barad, 1983; Jaffe, Wolfe, Wilson, & Zak, 1986; Porter & O'Leary, 1980; Wallerstein & Kelly, 1980; Wolman, 1983).  Possible consequences following an allegation of sexual abuse — a frightening or painful physical examination, separation from one or both parents, removal to a foster home, repeated interviews by different people — are themselves sources of significant stress.

Children who are distressed, whether by bitter arguing between parents, by physical or emotional but nonsexual abuse, or by any number of sad or frightening events, may reflect their distress in a wide variety of ways.  Some will strike out while others will withdraw.  Some will develop fears, others will report physical ailments.  Which signs develop in a particular child under stress will be a complex function of the genetic predispositions and the learning environment of that child.  The problem, therefore, is one of symptom specificity.  There is no behavior or set of behaviors that are specific to victims of child sexual abuse or that can lead with reasonably high accuracy to that diagnosis.  There simply is no empirical evidence supporting the suggestion that behaviors on the lists are causally linked to child sexual abuse.

There is one symptom that may seem to stand out from the others on these lists as a more valid indicator of sexual abuse, and that is age-inappropriate sexual play or knowledge (although sexual anxieties are also listed as stemming from parental conflicts).  In a society in which we think the topic of sexuality typically is only broached with young children in general and vague terms, if at all, the detailed sex knowledge of a child will be considered unusual.  Unfortunately, we lack the necessary information to assess the predictive value of what may appear to be precocious sexuality.  We don't think of the many ways we may expose children to sexuality.  What children normally and naturally do sexually is likely to be much more frequent and involved than most people assume (Gundersen, Melas & Skar, 1981; Martinson, 1981).  Without knowing what a normal level is, we cannot determine what is precocious, greater interest than normal, and what may indicate abuse.

Due to restrictions placed on the study of the sexual lives of children, we have not learned what proportion of abused and nonabused children display sexual behaviors in spontaneous play.  We do not know how many children imitate sexual behaviors modeled by siblings, relatives, or playmates.  We do not know how many children have access to depictions of explicit sexual acts in magazines or cable television or on X-rated video cassettes, and whether exposure to these models leads to imitation.  Therefore, while precocious sexual activities of young children may seem more indicative of sexual abuse than do other behavioral signs, there is too little known to form a conclusion.

In assessing a behavioral indicator, we must take into account the likelihood of whether a particular symptom or set of symptoms could be the result of sexual abuse or some other cause.  Estimates of the prevalence of child sexual abuse range widely.  Taken as a whole, various studies suggest that approximately 20% of women report having had some type of sexual contact with an adult during childhood.  Females are believed to have had such sexual contact at twice the rate of males.  One problem with such estimates is that the type of sexual contact reported may range from a single act of exhibitionism through subtle fondling to repeated attempts at penetration.  We do not know how severe or how frequent such sexual contacts must be before they results in emotional trauma and noticeable behavioral symptoms, or do we know how the nature of the contacts is related to the number and severity of symptoms (Wakefield & Underwager, 1988).

Let us suppose that roughly 20% of females and 10% of males in the United States have sexual contact with an adult sometime before the age of 18.  This averages to 15% of the total population.  We are particularly concerned here with children so young that they have difficulty understanding and communicating the abuse to an adult.  About 80% of sexual abuse occurs with children over the age of eight, according to the National Center on Child Abuse and Neglect (NCCAN, 1981b) as they begin to develop more sexually mature features, and only 20% with children eight or under.  If we use the 15% estimate of abuse for all children and limit our estimate to children age eight or younger, then in the general population of young children, perhaps 3% have been subjected to some type of sexual contact with an adult.

How does this estimate of the prevalence of sexual abuse compare with other childhood difficulties that might produce the same behavioral symptoms?  Estimates of physical abuse and emotional abuse run several times that of sexual abuse (NCCAN, 1981a).  Over 40% of all children in the United States will live in a one-parent household for at least part of their childhood, primarily due to divorce (Emery, 1982).  Many more will endure parental conflict that does not lead to divorce.  Add to this the relatively small number of children who will develop genuine childhood mental illnesses.  In sum, the probability that a young child who is showing problem behaviors is the victim of sexual abuse is far less than the probability of some other cause.  The teacher, minister, or other person, who, with all good intentions, jumps to the conclusion that a young child showing one or more of the long list of suggested behavioral indicators has been molested is likely to be making a grave error that can cause harm to the child.

It is an error in diagnosis to use non-discriminating signs to make a diagnosis.  If a sign can be caused by different variables, it cannot to be used to select a single one.  Most of the proposed behavioral indicators of abuse result from stress in general and are not specific to the stress of sexual abuse.

The base rates of the presence of many such behaviors in normal children, in troubled children, in nonabused children, and as part of the developmental process for all children, is so high that any attempt to use them as indicating abuse will result in a high rate of error and damage to children.  Douglas Besharov, the first head of the NCCAN, states that the only time that behavioral indicators are useful is when there is an unexplained physical injury (1985).
  

A History of Behavioral Indicators

In the late 19th and early 20th centuries there was a great deal of public attention given to the pernicious and destructive habit of masturbation by children.  This campaign to stamp out masturbation was part of a movement to increase healthy life styles in the populace.  Masturbation was said to produce blindness, dementia, all manner of physical illness, and thus destroy children.  The anti-masturbation campaign caused some extreme responses.  Female children had clitorectomies (surgical removal of the clitoris) performed.  Male children were kept in hand restraints for years.  Today there is general consensus that this anti-masturbation campaign was built on foolishness and error.

Among others, J. Kellogg, M.D., originator of corn flakes, produced several manuals for parents to help them stamp out the evil of masturbation.  In his books, he listed behavioral signs for parents to be alert for in order to determine whether their child was masturbating (Money, 1985).  These behavioral signs for masturbation included the following (current suggested behavioral indicators for sexual abuse are in italics): 

1. General debility, including exhaustion (Complaints of fatigue or physical illness which could mask depression)
2. Sudden change in disposition (Display regressive behavior)
3. Lassitude, dislike for play and lifelessness (Become withdrawn and daydream excessively)
4. Sleeplessness (Nightmares; Sleepwalking)
5. Failure of mental capacity (Sudden deterioration in academic performance; Inability to concentrate in school; Sudden drop in school performance)
6. Untrustworthiness (Poor peer relationships or inability to make friends; Acting-out aggressive behavior; Lack of trust, particularly with significant others)
7. Love of solitude (Become withdrawn and daydream excessively)
8. Bashfulness (Seems frightened or phobic, especially of adults)
9. Unnatural boldness (Acting-out aggressive behavior; Persistent and inappropriate sexual play with peers or toys or with themselves; Become sexually promiscuous)
10. Easily frightened (Seems frightened or phobic, especially of adults)
11. Confusion of ideas (including vulgar joking) (Confusion; Hints about sexual activity)
12. Capricious appetite (Sudden massive weight gain or loss)
13. Unnatural paleness (Experience deterioration of body image; Complaints of fatigue or physical illness which could mask depression)
14. Wetting the bed (Display enuresis and/or encopresis; Excessive urination)
15. Unchastity of speech, including fondness for obscene stories (Hints about sexual activity; Engage in highly sexualized play)
16. Early symptoms of consumption, or what are supposed to be such, including cough, short breathing, and soreness of the lungs (Medical conditions such as pneumonia or mononucleosis)

The behavioral indicators parents could use then to know if their children were masturbating are the same behavioral indicators now said to suggest that a child has been sexually abused. John Money (1985) states,

Kellogg's listing of suspicious signs has been given a new lease on life currently by the professional detectives of sexual child-abuse.  Here is an example of those who have not learned from history being condemned to repeat it, replete with all its dreadful consequences (p. 97).

The cautions about suggested behavioral signs do not mean that adults should not try to identify and aid children who show signs of distress.  A sensitive and caring adult who notices problem behaviors by a child will want to try to find out what is wrong.  But the adult must keep an open mind about what might be troubling the child and must be careful about the nature of the questions asked.  A rush to judgment and premature closure on sexual abuse as a cause of the problems behaviors should be avoided.  The odds are against this diagnosis.
  

References

AMA diagnostic and treatment guidelines concerning child abuse and neglect (1985). Journal of the American Medical Association, 254, 796-800.

American Psychiatric Association (1987). Diagnostic and Statistical Manual of Mental Disorders (3rd Edition-Revised) ()(). Washington DC: Author.

Besharov, D. J. (1985, November). Paper given at the VOCAL (Victims of Child Abuse Laws) National Convention, Minneapolis, Minnesota.

Cohen, A. (1985). The unreliability of expert testimony on the typical characteristics of sexual abuse victims. Georgetown Law Journal, 74, 429-456.

Emery, R. E. (1982). Interparental conflict and the children of discord and divorce. Psychological Bulletin, 92, 310-330.

Gundersen, B. H., Melas, P.S., & Skar, J. E. (1981). Sexual behavior in preschool children: Teachers' observations. In L. L. Constantine & F. M. Martinson (Eds.), Children and Sex: New Findings, New Perspectives () (pp.45-51). Boston: Little, Brown & Company.

Hughes, H. M. & Barad, S. J. (1983). Psychological functioning of children in a battered woman's shelter: A preliminary investigation. American Journal of Orthopsychiatry, 53, 525-531.

Jaffe, P., Wolfe, D., Wilson, S., & Zak, L. (1986). Similarities in behavioral and social maladjustment among child victims and witnesses to family violence. American Journal of Orthopsychiatry, 56, 142-146.

Martinson, F. M. (1981). Eroticism in infancy and childhood. In L. L. Constantine & F. M. Martinson (Eds.), Children and Sex: New Findings, New Perspectives () (pp. 23-35). Boston: Little, Brown & Company.

Money, J. (1985). Destroying Angels (). Buffalo, New York: Prometheus Books.

NCCAN (National Center on Child Abuse and Neglect) (1981a). Executive summary: National study of the incidence and severity of child abuse and neglect, National Center on Child Abuse and Neglect. (DDHS Publication No.81-30329). Washington, DC: U.S. Government Printing Office.

NCCAN (National Center on Child Abuse and Neglect) (1981b). Study findings: National study of the incidence and severity of child abuse and neglect, National Center on Child Abuse and Neglect. (DDHS Publication No.81-30325). Washington, DC: U.S. Government Printing Office.

Porter, B. & O'Leary, D. (1980). Marital discord and child-hood behavior problems. Journal of Abnormal Psychology, 8, 287-195.

Sgroi, S. M. (1982). Handbook of Clinical Intervention in Child Sexual Abuse (). Lexington, MA: Lexington Books.

Wakefield, H., and Underwager, R. (1988). Accusations of Child Sexual Abuse ()(). Springfield, IL: Charles C. Thomas.

Wallerstein, J. S., & Kelly, J. B. (1980). Surviving the Breakup: How Children and Parents Cope with Divorce (). New York: Basic Books.

Wolman, B. (Ed.) (1983). Handbook of Developmental Psychology (). Englewood Cliffs, NJ: Prentice- Hall.