Human Immunodeficiency Virus Transmission by
Child Sexual Abuse
Laura T. Gutman, MD; Karen K. St Claire, MD; Chris Weedy, MSW; Marcia E. Herman-Giddens, PA, MPH;
Barbara A. Lane, MSN; Jeanne G. Niemeyer, MSW; Ross E. McKinney, Jr, MD
\s=b\ During1987-1989,14(14.6%) of the 96 children who
tested positive for the human immunodeficiency virus (HIV)
and were followed up by the Duke University (Durham, NC)
pediatric acquired immunodeficiency syndrome team were
confirmed to have been sexually abused. Every sexually
abused child was evaluated for each of five modes of HIV
transmission, and in nine children the pathway was iden-
tified. Four of the study children acquired HIV from child
sexual abuse and in six, abuse was a possible source. Trans-
mission by child sexual abuse was the most frequent of the
proven modes of acquisition of HIV in this population. The
other proven modes of acquisition were vertical transmis-
sion (n=3) and HIV\p=n-\contaminatedblood transfusion
(n 2). Twelve males were identified (n 8) or suspected
= =
(n=4) of being perpetrators. Three knew themselves to
have HIV at the time of an assault and eight were aware that
the child had HIV at the time of an assault. There was no
indication from any child that "safe sex" precautions had
been observed. Children with HIV infection had multiple
risk factors for abuse or neglect. The sociological descrip-
tors of the lives of the 14 abused children showed multiple
known risk factors for sexual abuse that also overlapped
with known risk factors for or sequelae of the acquisition
of HIV infection. These included drug abuse and alcoholism
in the home, prostitution of a parent, lack of parenting, pov-
erty, and chronic illness of the child. Prevention efforts
should recognize that children as well as adults are at risk
for sexually transmitted HIV infection.
(AJDC. 1991;145:137\p=n-\141)
he Duke University (Durham, NC) pediatrie acquired
immunodeficiency sydrome (AIDS) team began ac¬
cepting pediatrie patients in 1987 and by December 1989
had a census of 96 patients who showed positive results
to the human immunodeficiency virus (HIV) antibody test
and were known to the AIDS team social worker. These
children were at all stages of HIV-related illness and
ranged in age from newborn to 17 years. All children re¬
ceived integrated multidisciplinary services that delivered
Accepted for publication November 6, 1990.
From the Department of Pediatrics, Duke University Medical Cen-
ter, Durham, NC.
Presented in part at the annual meeting of the Society for Pediatric
Research, Anaheim, Calif, May 11, 1990.
Reprint requests to Box 3971, Duke University Medical Center,
Durham, NC 27710 (Dr Gutman).
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/19/2015
optimal therapy and treatment of the complex problems
that characterize children with AIDS.
During this period, 14 children from the AIDS popu¬
lation were confirmed to have been sexually abused. The
medical and social conditions of the lives of the abused
children were reviewed for known risk factors for the ac¬
quisition of HIV and for child sexual abuse (CSA). This
study describes the results of the evaluation of the sexually
abused children, the circumstances surrounding the abu¬
sive experiences, the perpetrators, and the data regarding
the means by which the children had acquired HIV.
PATIENTS AND METHODS
Evaluation Process
All children seen by the AIDS team who presented with or
developed indications of possible abuse or neglect were referred
to the child protection team for evaluation. Data for this study
were collected at the time of the assessment of each child and
represent cumulative information regarding the child's environ¬
ment from the family, child, child protection team, AIDS team,
social service agencies, schools, and, often, charitable agencies.
Child sexual abuse was defined in 1977 as the involvement of
dependent developmentally immature children and adolescents
in sexual activities that they do not fully comprehend, to which
they are unable to give informed consent, or that violate the social
taboos of family roles.1 Child sexual abuse includes rape, pe¬
dophilia, child prostitution, child pornography, child sex rings,
and incest.
Multiple issues often led to referral to the child protection team.
The instigating events included (1) the suspicion of or disclosure
of CSA volunteered by the child or a caretaker (n 2); (2) ab¬
=
normal genital findings revealed by routine medical examination
(n 5); (3) suspected physical abuse or neglect, or a chaotic family
=
environment (n 3); (4) medical or behavioral histories indicative
=
of CSA ( 5); (5) children older than 6 years at onset of HIV
=
disease, or with no known risk factor for HIV (n 3); and (6) the
=
child was an HIV-positive sibling of a sexually abused child
(n l). All of the children were known to be infected with HIV
=
prior to the diagnosis of CSA.
Evaluation for CSA
The evaluation of a child for CSA began with a family interview
regarding the family constellation, patterns of care for the child,
interactions of family members, behavioral characteristics of the
child, and concerns or knowledge that the caretaker may have
had regarding possibly abusive events. The following five groups
of data needed for adequate evaluation of possible CSA were then
obtained: (1) a review of the behavioral history of the child; (2)
a review of the medical history of the child; (3) diagnostic in¬
terview^) with the child; (4) results of physical examination; and
(5) assessment for other sexually transmitted diseases.2"4
The diagnostic interviews followed a standard format. Draw-
 ings and anatomically correct dolls were available for use when
appropriate.5 In the present study, "disclosure" refers to the dis¬
closure by the child or witness of a specific sexually abusive act.
Perpetrators were "identified" if the child or a caretaker had spe¬
cifically identified the person as an assailant. Perpetrators were
considered "suspected" if the interview or historical information
indicated sexually inappropriate interactions with the study child
or another child and unsupervised access to the study child.
Each child received a complete physical examination and ex¬
ternal genital examination using standard techniques.6,7 For the
girls, genital examinations included identification of hymenal
type and contour, measurement of the horizontal diameter of the
hymenal opening, and anal examination.8,9 Samples of vaginal
secretions were examined for indicators of bacterial vaginosis and
Trichomonas. Cultures of the introitus or vaginal canal of girls and
of the anus of boys for Neisseria gonorrheae and Chlamydia trachom-
atis were taken unless the child was receiving antimicrobial ther¬
apy. Cultures of other sites and assessments for other infectious
diseases were made when indicated.
Sexual abuse was confirmed if there was a detailed account
from the child or a witness, a sexually transmitted disease in
addition to HIV was found, the genital examination showed ab¬
normal findings clearly indicative of abuse as defined at the 1985
National Child Sexual Abuse Summit meeting10 and subse¬
quently updated,21112 serial genital examinations showed sig¬
nificant changes, or the abuse was acknowledged by the per¬
petrator.
Mode of Acquisition of
HIV by Study Children
The five possible routes of transmission of HIV that were sep¬
arately assessed for each child were vertical transmission, trans¬
fusion, clotting factor concentrates, intravenous drugs, and sex¬
ual abuse (including child prostitution).
Data regarding vertical transmission included date(s) and re¬
sults) of maternal HIV serologie tests relative to the birth of the
child or HIV-related illness of the child. Data regarding
transfusion-transmitted HIV included dates of transfusion to the
child and the results of retrospective blood-bank HIV surveil¬
lance. None of the children in the study had received clotting
factor concentrate therapy or had engaged in intravenous drug
abuse. Data regarding transmission by sexual abuse included
available information on the HIV status of the perpetrator(s) and
the type of abuse.
Since many of the abused children had multiple possible modes
of acquisition of HIV, the influence of each on the transmission
of the HIV infection was stratified into categories of "proven,"
"possible," "disproven," and "unknown."
The following criteria were used to define the categories for
vertical transmission: proven, the mother was HIV positive prior
to or within 6 months of giving birth, or was first shown to be
HIV positive 6 or more months after delivery and the child had
an opportunistic infection or lymphocytic interstitial pneumonia
by the age of 2 years; possible, the mother was first known to be
HIV positive 6 or more months after delivery; disproven, the
mother was HIV negative 6 or more months after delivery; and
unknown, the mother's HIV seroactivity was unknown.
Criteria to define the categories for transmission by transfusion
were as follows: proven, the mother was HIV negative at or after
delivery and the child was known to have received an HIV-
contaminated transfusion; possible, the mother was first known
to be HIV positive 6 or more months after delivery and the child
received an HIV positive transfusion; disproven, the child either
received no transfusion, or it was HIV negative; and unknown,
the child received a transfusion and retrospective surveillance
was not able to trace the unit.
The criteria to define the categories for transmission of HIV
through CSA were as follows: proven, either all other risk factors
were disproven or the child was older than 12 years when HIV
was first diagnosed, sexual contact had involved high risk for
HIV, and no other factor had been proven; possible, the per¬
petrator was either HIV positive or unknown, and no other risk
factor was proven; disproven, identified perpetrator was HIV
negative; and unknown: an additional risk factor was proven.
Inclusion in this report required that sexual abuse of the child had
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/19/2015
Table 1.—Forms of Child Sexual Abuse
Female Children Male Children
(n = 11) (n 3)
=
Genital-vaginal 7 Not applicable
Digital-vaginal 2 Not applicable
Genital-rectal 4 1
Genital-oral 3 0
Unknown
4 2
been confirmed. A conservative definition of proven transmis¬
sion of HIV by CSA gave priority to other known modes of trans¬
mission.
Sociological Setting
The social settings of the subjects were evaluated for five so¬
ciological risk factors for CSA identified from the 1987 National
Survey of Children.13 In that study, a significantly greater inci¬
dence of CSA was experienced by girls who had lived apart from
their biological parents before age 16 years; had been raised in
poverty; had an emotional, physical, or mental handicap; and
had family members who were alcoholic or drug abusers.
Other aspects of the social setting that were recorded in the
present study included prostitution by adults in the home; the
presence of multiple, unrelated, and frequently changing live-in
visitors to the household; significant personality disorder of a
caretaker; and AIDS encephalopathy or other AIDS-related dis¬
ability of a caretaker.
RESULTS
Fourteen HIV-positive children were confirmed to have
been sexually abused. The children included three boys
and 11 girls (79% female) whose ages at the time of the
diagnosis of CSA ranged from 3.5 to 13 years and whose
mean age was 6.2 years. Thirteen of the 14 children were
classified as P-2 based on Centers for Disease Control
(CDC) criteria at the time they were diagnosed to be HIV
positive. Eleven children were black, two were white, and
one was Native American.
Diagnosis of CSA
Each of the 14 children with confirmed CSA either had
disclosed sexual abuse, had genital or anal findings
strongly indicative of CSA, or both. All of the 11 girls had
abnormal examinations of the introitai area and/or hymen.
Two of the three boys had abnormal anal examination re¬
sults. The genital findings indicative of abuse were scars
or healed lesions of the posterior fourchette, ( 4); scars,
=
tears, notches, or significant distortions of the hymen
(n 5); recurrent or persistent vaginal discharge and odor
=
or vaginal bleeding (n 6); hymenal opening size greater
=
than or equal to 8 mm in a prepubertal child or a significant
change in the genital examination results on serial eval¬
uations (n 8); -perianal scars (n 2); significant and rapid
= =
anal dilation (n 4); and perianal lesions (n l). Neither
= =
gonorrheae nor C trachomatis was isolated from the va¬
gina, throat, or rectum of any study child. Three children
had a sexually associated disease other than HIV; two had
bacterial vaginosis14; and one had condylomata acumina¬
timi.15
Eight children were able to disclose their abuse. Two
made partial disclosure and four were unable to disclose.
The forms of sexual abuse described by the eight children
who were able to disclose are listed in Table 1.
All identified or suspected perpetrators were male. The
number of known or suspected assailants and their rela¬
tionship to the child when information was available are
 Table 2.—Relation of Identified and Suspected
Perpetrator(s) to an Assaulted Child
No.
Relation Identified
Brother 1*
Father 1 3
Uncle 2 1*
Grandfather 2*
Foster father 1*
Foster brother 1*
Nonrelated assailant of
child prostitute 7*
*One of multiple identified or suspected assailants of a single child.
Table 3.—Assessed Mode of Acquisition of
Human Immunodeficiency Virus
in Sexually Abused Children (n 14)* =
Mode of
Acquisition Proven Possible Disproven
Child sexual abuse 4 6
...
Vertical transmission 3 5 5 1
Transfusion 2 ... 12
'Acquisition of human immunodeficiency virus by clotting factor
concentrate and by intravenous drug use was disproven in all cases.
shown in Table 2. For four children, the number and iden¬
tity of perpetrator(s) was unknown. A single perpetrator
was identified for three children, and for three children
multiple perpetrators were identified. For two children a
single perpetrator was suspected. For one child, one per¬
petrator was identified and additional perpetrators were
also suspected. For another child, one perpetrator was
suspected while the multiple perpetrators known to have
abused the child were unidentified.
Acquisition of HIV
Using the criteria stated in the "Patients and Methods"
section, an assessment was made of the mode of trans¬
mission of HIV when all data were considered. The results
are shown in Table 3. For four (29%) of the 14 children,
CSA was the only means of transmission of HIV to the
child. A brief synopsis of the histories of these children
follows. To protect the identities of the children, gender
identification is not provided.
CASE 1.—The child was an emotionally abused, phys¬
ically healthy child who lived in fear of the stepfather. At
age 13 years, the child ran away from home and lived as
a child prostitute for 3 months, during which time sexual
practices included unprotected anal-receptive intercourse
with multiple high-risk adults. Intravenous drugs were
never used. The child had never had a transfusion and was
not hémophilie. The mother was not tested but had no risk
factors for HIV. Atage 15years, the child requested an HIV
assay because of the high-risk history. Results of enzyme-
linked immunosorbent assay (ELISA) and Western blot
were positive, and the disease was CDC class P-l, subclass
B. It was concluded that the child had acquired HIV as a
child prostitute.
CASE 2.—At age 3 years 9 months, this child was HIV
positive based on ELISA and Western blot test results; the
disease was CDC class P-2, subclass A. The child's mother
was HIV negative. The child had not received a transfu-
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/19/2015
sion or blood products and was not hémophilie. Genital
examination results were abnormal and included condy-
loma acuminatum, which had first been noted at age 3
No. years and 3 months. After partial disclosure it was con¬
Suspected cluded that the abuse had occurred during a chaotic family
episode when numerous caretakers had had access to the
child, and that the child had acquired HIV during the
abuse.
CASE 3.—This child was aged 2 years 2 months when
diagnosed to be HIV positive by ELISA and confirmed by
Western blot; the disease was CDC class P-2, subclass F.
The mother was HIV negative by ELISA, Western blot,
and HIV culture. Although the child had received a trans¬
fusion, the donor was HIV negative. The child was not
hémophilie. Genital examination was positive for trau¬
matic abuse. Although verbal disclosure was obtained,
identification of the assailant was not made, and a sus¬
pected assailant refused HIV testing. Because of the dis¬
closure and the fact that another child had been sexually
abused in the same home, the child was removed from the
household and restricted from contact with the suspected
perpetrator. Further abuse was subsequently documented
by the development of new physical signs and symptoms,
Unknown but specific identification of the perpetrator again could
4
not be made. It was concluded that the child had acquired
HIV during abuse by multiple perpetrators.
CASE 4.—This child was diagnosed at age 6 years and
2 months to be H1TV positive by ELISA and Western blot,
and classified as CDC P-2 subclass C. At the time of the
diagnosis, the mother was HIV-negative, the child had
never had a transfusion, and the child was not hémophilie.
The genital examination revealed definitive evidence of trau¬
matic abuse, as did the medical history. Two assailants were
identified, one of whom was known to be HIV positive. It
was concluded that the child had acquired HIV infection dur¬
ing child sexual abuse by an HIV-positive assailant.
Knowledge by the Assailant of the Possibility
of HIV Transmission to or From the Child
Three of the assailants who were identified or suspected
were HIV positive and knew of their HIV status at the time
they assaulted the child. There was no indication from the
children that "safe sex" precautions had been taken.
Eight identified or suspected assailants were aware that
the child was HIV positive at the time of the assault. Five
of these eight were themselves either HIV negative or their
status was unknown when the child was assaulted. Again,
there was no indication from the children that "safe sex"
precautions had been taken.
COMMENT
In a recent review of the data from the 1987 National
Survey of Children, social settings were identified that in¬
creased the risk of CSA.13 All but one child in the present
study had two or more of these risk factors, and the pro¬
portion of children who hada given risk factor ranged from
43% to 71%, as shown below.
Social Conditions No. (%) of Children
Lived apart from
both biological parents 6 (43)
Raised in poverty* 10 (71)
Child was handicappedt 9 (64)
Alcoholic family member* 7 (50)
Drug abusing family member* 10 (71)
Prostitution at home 4 (29)
Transient adults living at home 8 (57)
Mentally ill caretaker 5 (36)
AIDS-related disability of caretaker 2 (14)
*Risk factors for CSA identified by National Survey of Children.13
tHandicap was probably apparent before onset of CSA.
 In the National Survey of Children, 6% of girls with no
riskfactor, 9% of girls with one risk factor, 26% of girls with
two risk factors, and 68% of girls with three or more risk
factors had been sexually abused as children. In the
present study, five of these identified factors were as¬
sessed, and 11 (78%) of 14 children had three or more of
these risk factors each, demonstrating that the living cir¬
cumstances of the abused children in the present study
included previously described indicators for increased risk
of sexual abuse.
Many of the circumstances surrounding the sexual
abuse of these children are also previously identified risk
factors for adult HIV transmission. First, promiscuous sex¬
ual activity with multiple partners is a well-defined risk
factor for adult HIV infection.16 In the present series, seven
(50% ) of 14 children were known or suspected to have been
assaulted by multiple perpetrators, eight children (57%)
lived in homes in which casual adult acquaintances fre¬
quently slept in the home, thereby creating opportunities
to abuse a vulnerable child, and four of these eight homes
were also the sites for prostitution by adult caretakers. One
of the children in the study had turned to child prosti¬
tution, itself a form of CSA. A second risk factor for adult
HIV infection is sexual contact that is physically traumatic
or involves impaired mucosal barriers. Both anal-receptive
and oral-receptive sex have been associated with increased
rates of transmission of HIV in adults, and were acts that
these children described.18"20 Furthermore, the children in
the study had an unusually high incidence of physical
signs of genital injury compared with usual groups of sex¬
ually abused children. A third risk factor for adult HIV
infection is sexual intercourse without barrier protection.
Assailants of children in this study practiced high-risk and
unprotected sex even when the assailant knew himself to
be HIV infected or when he knew the child to be HIV in¬
fected. A fourth risk factor for adult HIV infection, which
was also found in most children in the study, is genital
mucosal lesions as evidenced by vaginal bleeding, dis¬
charge, infections, and scars, perianal infections, lesions,
and scars, and other sexually transmitted diseases.21 Fi¬
nally, the adult population with whom many children
lived were at increased risk of HIV from the use of illicit
drugs. Use of drugs and alcohol by caretakers may also
have diminished their ability to protect the child and low¬
ered their threshold for sexual aggression.22,23 Conse¬
quently, these children were subjected to forms of sexual
intercourse that were known from adult studies to be es¬
pecially hazardous regarding transmission of HIV, and the
children lived in family and social settings in which many
adults who were their caretakers were at increased risk of
being HIV infected.
In spite of the multiple risk factors for CSA that char¬
acterize the lives of many children with HIV, acquisition
of HIV by children through abusive sexual assault has
been infrequently considered or reported in the medical
literature. Although individual case reports have provided
evidence that this route of transmission should be exam¬
ined for children with HIV,24'25 few medical reviews of the
routes of transmission of HIV or of unsolved epidemio¬
logie problems with HIV allude to transmission through
CSA. ,26"29 For example, the exposure categories for re-
cases of childhood HIV infection by the Centers
Í)orting
or Disease Control do not include the category of expo¬
sure through sexual contact. The lack of data and the need
for policies on the testing of abused children for HIV has
been the subject of a recent review.30 The present study
indicates that at least four (4.2%) of 96 children with HIV
and four (29%) of 14 sexually abused children with HIV
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/19/2015
acquired the infection through CSA. These represent the
minimal percentages of pediatrie HIV disease that can be
attributed to CSA, since the abused children were iden¬
tified during standard, nondirected pediatrie interactions
rather than by specific screening for abuse. In addition,
children who also had other "possible" modes of trans¬
mission, such as maternal HIV, could not be proven to
have acquired their disease through CSA. It was notable
that four of the five children for whom vertical transmis¬
sion of HIV was assessed to have been "possible" because
the mother was HIV positive were older than 3 years when
AIDS-like illness first began. This would be a long incu¬
bation period for perinatally transmitted HIV,31 and some
of these children may instead have acquired their disease
through CSA. This situation is exemplified by one of the
study cases, in which the HIV-positive man who was iden¬
tified as the perpetrator of a child's abuse was also the
source of the mother's HIV infection.
The diagnosis of CSA in children who have AIDS or who
are in a high-risk environment is of particular importance
because some immediate and delayed behavioral sequelae
of CSA may put the adolescent and adult survivor at in¬
creased risk of exposure to HIV or transmission of HIV to
others if they are infected. First, high-risk sexual behaviors
that may characterize female survivors of CSA include
early entry into sexual activities, sexual promiscuity, and
a particular vulnerability to further abuse and sexual ex¬
ploitation, including prostitution and unintended preg¬
nancy at an early age.32"40 Male victims of CSA may de¬
velop a cycle of sexual behavior in which the child victim
becomes a sexual aggressor of other children, either im¬
mediately or in adulthood.41 Multiple victims may be in¬
volved. The consequences of the intersection of child
sexual abuse with immediate and delayed risk factors for
acquisition of HIV are dramatically highlighted in the
"street" children of New York, NY. Many of these chil¬
dren were sexually abused in earlier childhood, presum¬
ably providing their motivation for leaving their homes at
a highly vulnerable, adolescent age.
In conclusion, this study has demonstrated that child
sexual abuse was the proven mode of transmission in at
least 4% of all study children with HIV followed up by the
pediatrie AIDS team, and may have been the mode of
transmission for a considerably larger proportion of cases.
The abused children lived in circumstances that put them
at high risk for both CSA and HIV infection, and these risk
factors often overlapped. Assailants were known to have
abused children in spite of knowing themselves or the
child to be HIV positive. Prevention of HIV transmission
in populations of children and adolescents cannot be suc¬
cessful without the development of policies and resources
to identify and eliminate the underlying sexual abuse to
which these children are exposed. Sexual abuse and its
consequences provide a major mechanism for the intro¬
duction of HIV to children and adolescents.
We are grateful for funding from the National Institutes of Allergy
and Infectious Disease for the Duke University Pediatrie AIDS Treat¬
ment and Evaluation Unit, Durham, NC.
References
1. Kempe HC. Sexual abuse, another hidden pediatric prob-
lem: the 1977 C. Anderson Aldrich Lecture. Pediatrics.
1978;62:382-389.
2. Herbert CP. Expert medical assessment in determining
probability alleged child sexual abuse. Child Abuse Negl.
of
1987;11:213-221.
3. Sargent DA, Blumberg M, Broughton D, et al. AMA diag-
nostic and treatment guidelines concerning child abuse and ne-
glect. JAMA. 1985;254:796-800.
 4. Berson N, Herman-Giddens M, eds. Duke University Med-
ical Center Child Protection Team Manual 1990. Durham, NC:
Duke University Medical Center Child Protection Team; 1986.
5. Berkowitz CD. Sexual abuse of children and adolescents.
Adv Pediatr. 1987;34:275-312.
6. Finkel MA. The medical evaluation of child sexual abuse.
In: Schetky D, Green A, eds. Child Sexual Abuse. New York,
NY: Brunner-Mazel Inc; 1988:82-103.
7. McCann J, Voris J, Simon M, Wells R. Comparison of gen-
ital examination techniques in prepubertal girls. Pediatrics.
1990;85:182-187.
8. Goff CW, Burke KR, Rickenback C, Buebendorf DP. Vag-
inal opening measurement in prepubertal girls. AJDC.
1989;143:1366-1368.
9. Herman-Giddens ME, Frothingham TE. Prepubertal female
genitalia: examination for evidence of sexual abuse. Pediatrics.
1987;80:203-208.
10. Tipton AC. Child sexual abuse: physical examination
techniques and interpretation of findings. Adolesc Pediatr Gy-
necol. 1989;2:10-25.
11. Emans SJ, Woods E, Flagg M, Freeman A. Genital findings
in sexually abused symptomatic and asymptomatic girls. Pedi-
atrics. 1987;79:778-785.
12. White ST, Ingram DL, Lyner PR. Vaginal introital diameter
in the evaluation of sexual abuse. Child Abuse Negl.
1989;13:217-224.
13. Moore KA, Nord CW, Peterson JL. Nonvoluntary sexual
activity among adolescents. Fam Plann Perspect. 1989;21:110\x=req-\
114.
14. Hammerschlag MR, Cummings M, Doraiswamy B, Cox P,
McCormack WM. Nonspecific vaginitis following sexual abuse
in children. Pediatrics. 1985;75:1028-1031.
15. Herman-Giddens ME, Gutman LT, Berson NL, Duke Child
Protection Team. Association of coexisting sexually transmitted
disease and multiple abusers in female children with genital
warts. Sex Transm Dis. 1988;15:63-67.
Pape JW, Liautaud B, Thomas F, et al. The acquired im-
16.
munodeficiency syndrome in Haiti. Ann Intern Med.
1985;103:674-678.
17. Cohen Ml. Effective low enforcement strategies for han-
dling juvenile prostitution.Nat Sheriff. 1988;40:49-52.
18. Fischl MA, Dickinson GM, Scott GB, Klimas N, Fletcher
MA, Parks W. Evaluation of heterosexual partners, children, and
household contacts of adults with AIDS. JAMA. 1987;257:640\x=req-\
644.
19. Padian N, Marquis L, Francis DP, et al. Male-to-female
transmission of human immunodeficiency virus. JAMA.
1987;258:788-790.
20. Frederick W, Olopoenia L, Delapenha R, Barnes S, Sax-
inger C, Greaves W. Sexual practices associated with HIV trans-
mission among female sexual partners of HIV seropositive men.
Presented at the 29th International Conference on Antimicro-
bial Agents and Chemotherapy; September 18,1989; Houston,
Tex.
21. Greenblatt RM, Lukehart SA, Plummer FA, et al. Genital
ulceration as a risk factor for human immunodeficiency virus
infection. AIDS. 1988;2:47-50.
22. Browning DH, Boatman B. Incest: children at risk. Am J
Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/19/2015
Psychiatry. 1977;134:69-72.
23. Famularo R, Stone K, Barnum R, Wharton R. Alcoholism
and severe child mistreatment. Am J Orthopsychiatry.
1986;56:481-485.
24. Leiderman IZ, Grimm KT. A child with HIV infection.
JAMA. 1986;256:3094.
25. Gellert GA, Durfee MJ. HIV infection and child abuse. N
Engl J Med. 1989;321:685.
26. Gellert GA, Mascola L. Rape and AIDS. Pediatrics.
1989;83(suppl):644-645.
27. Fuller AK, Bartucci RJ. HIV transmission and childhood
sexual abuse. JAMA. 1988;259:2235-2236.
28. Osterholm MT, MacDonald KL. Facing the complex
issues of pediatric AIDS: a public health perspective. JAMA.
1987;258:2736-2737.
29. Novick BE, Rubinstein A. AIDS: the pediatric perspective.
AIDS. 1987;1:3-7.
30. Gellert GA, Durfee MJ, Berkowitz CD. Developing guide-
lines for HIV antibody testing among victims of pediatric sexual
abuse. Child Abuse Negl. 1990;14:9-17.
31. Johnson JP, Nair P, Hines SE, et al. Natural history and
serologic diagnosis of infants born to human immunodeficiency
virus\p=n-\infectedwomen. AJDC. 1989;143:1147-1153.
32. Browne A, Finkelhor D. Impact of child sexual abuse: a
review of the research. Psychol Bull. 1986;99:66-77.
33. Briere J, Runtz M. Post sexual abuse trauma: data and
implications for clinical practice. J Interpers Violence.
1987;2:367-379.
34. Miller J, Moeller D, Kaufman A, Divasto P, Pathak D,
Christy J. Recidivism among sexual assault victims. Am J Psy-
chiatry. 1978;135:1103-1104.
35. Sedney MA, Brooks B. Factors associated with a history
of childhood sexual experience in a nonclinical female popu-
lation. J Am Acad Child Psychiatry. 1984;23:215-218.
36. Finkelhor D, Browne A. The traumatic impact of child sex-
ual abuse: a conceptualization. Am J Orthopsychiatry.
1985;55:530-541.
37. Bagley C, McDonald M. Adult mental health sequels of
child sexual abuse, physical abuse and neglect in maternally
separated children. Can J Commun Ment Health. 1984;3:15-26.
38. Silbert MH, Pines AM. Sexual child abuse as an anteced-
ent to prostitution. Child Abuse Negl. 1981;5:407-411.
39. James J, Meyerding J. Early sexual experience and pros-
titution. Am J Psychiatry. 1977;134:1381-1385.
40. Gershenson HP, Musick JS, Ruch-Ross HS, Magee V, Ru-
bino KK, Rosenberg D. The prevalence of coercive sexual ex-
perience among teenage mothers. J Interpers Violence.
1989;4:204-219.
41. Longo R. Sexual learning and experience among adoles-
cent sexual offenders. Int J Offender Ther Comp Crimonol.
1982;26:235-241.
42. Abel GG, Becker JV, Mittelman M, Cunningham-Rathner
J, Rouleau JL, Murphy WD. Self-reported sex crimes of non-
incarcerated paraphiliacs. J Interpers Violence. 1987;2:3-25.
43. Stricof R, Novick LF, Kennedy J, Weisfuse I. HIV sero-
prevalence of adolescents at Convenant House/under 21, New
York City. Presented at the American Public Health Association
Conference; November 1988; Boston, Mass.