The
Association
of Conserisual
Sexual
Intercourse
During
Childhood
With
Adolescent
Health
Risk
and
Behaviors
Michael D. Resmck, PhD* and Robert Wm. Blum, MD, PhD, MPH
ABSTRACT. Objective. The association of childhood
sexual intercourse, excluding sexual abuse, with adoles-cent health and risk behaviors was examined using the
urban component of a statewide study on adolescent
health, risky behaviors, and resiliency factors.
Methodology. A specialized cohort design was used to
derive a two-group sample. Index cases consisted of all
adolescents who indicated that they had first intercourse at
or before age 10 years, and controls were adolescents who either had not yet had intercourse or had done so at age 16 years or older. To avoid confounding with issues of sexual abuse, all adolescents who also indicated a history of sexual abuse on the survey were removed from the analysis, so that the comparison could focus on health and behavioral
correlates of respondents who did not define their
child-hood sexual intercourse as constituting abuse. Comparisons
were conducted separately for girls and boys.
Results. A significantly greater proportion of index cases than controls indicated problem substance use by parents, poor school performance, gang involvement, fre-quent and unprotected sexual intercourse, history of
pregnancy involvement, desire to leave the home, history
of mental health treatment, emotional distress, and sui-cidal involvement. Logistic regression revealed signifi-cant group differences including academic risk, gang involvement, frequency of sexual intercourse, and his-tory of mental health treatment. Correct group
classifica-lion with the logit model was approximately 90% for both girls and boys.
Conclusions. The results underscore the importance of childhood sexual intercourse as an indicator of other health-compromising behaviors and risk factors.
Cmi-cians should be alert to this clustering of risk behaviors
in their psychosocial assessment of young people. Pediatrics 199494:907-913; adolescent health, sexual
behavior, risk behaviors.
Over the past 2 decades, there has been a
down-ward shift in the age at onset of sexual intercourse
among adolescents. Recent reports indicate that since
the mid-1960s, the likelihood of engaging in sexual
intercourse generally has increased, with some
fluc-tuations, among girls at every age between 14 and 18
years.’ In fact, this pattern of earlier onset of
inter-course may have occurred for adolescents over the
course of the last half century2’-; however, the
in-crease in the proportion of sexually active youth has
From the *Division of Health Management and Policy and the Division of
General Pediatrics and Adolescent Health, University of Minnesota,
Minneapolis, Minnesota.
Received for publication Jun 28, 1993; accepted Apr 15, 1994.
Reprint requests to (M.D.R.) University of Minnesota, Box 721 UMHC,
Harvard Street at East River Road, Minneapolis, MN 55455.
PEDIATRICS (ISSN 0031 4005). Copyright © 1994 by the American
Aced-emy of Pediatrics.
been most rapid between 1985 and 1988, the latest
data point available.’ The increasingly early onset of sexual intercourse heightens many concerns,
includ-ing acquisition of sexually transmitted diseases3’4 and
further exacerbation of pregnancy rates and
preg-nancy sequelae including abortion and precocious
childbirth,” which already are disproportionately
high among American teenagers compared with
other industrialized nations.6’7
Except for reports focusing on sexual abuse, little
is known about that segment of the pediatric
popu-lation who report a very early onset of intercourse.
Yet, it has been suggested that sexual intercourse
among prepubertal, urban African-American males
may not be uncommon.4’9 A few investigators have
identified serious problem behaviors among
adoles-cents and young adults who first experienced sexual
intercourse during childhood without the
accompa-nying perception of having been sexually abused.’#{176}’2
Existing research on the correlates or consequences
of early sexual intercourse without reference to sex-ual abuse has focused on girls and boys for whom
first intercourse occurred in middle or early
adoles-cence.5”2#{176} As sexual intercourse becomes
increas-ingly prevalent at younger ages, its correlates and
consequences must be reexamined using younger
samples, including those who experienced first
sex-ual intercourse in late childhood.
The purpose of this study was to compare two
groups of adolescents: those who reported having
sexual intercourse at or before age 10 years and a
control group of adolescents who either had not had
sexual intercourse at all, or had done so at age 16
years or older. Index and control groups of
adoles-cents were compared on a variety of health and
risk-behavior indices, with the hypothesis that
youths who had had sexual intercourse during
child-hood are at greater risk for a variety of adverse outcomes including stress, depression and suicidal
involvement, unprotected sexual intercourse, and
other health-jeopardizing behaviors. We also
hy-pothesized that adolescents representing the index group would report more family stress and dysftmc-lion than controls.
Sampling Design
METHODS
The sample was derived from the Minnesota Adolescent Health Survey, an anonymous self-administered questionnaire given to
36 284 public school students in grades 7 through 12 during the
1986 to 1987 school year. The questionnaire is a comprehensive survey of adolescent health, risk behaviors, and resiliency factors at Viet Nam:AAP Sponsored on September 1, 2020
www.aappublications.org/news
that has been administered to a total of 70 000 students throughout
the United States, including the Minnesota study. Questionnaire development and content, psychometric properties of scales and
indices, and deaning and editing procedures for the self-report
data are described in other publications.2126 Schools were selected
through a multi-stage cluster sampling design, stratified by school
district size, with random sampling occurring within each stratum
for each grade level.
The index and control groups were drawn from the urban component of the Minnesota data set, using the following defini-tions. Adolescents in the index group were those who had had
sexual intercourse at or before age 10 years. Control group
mem-bers had never had sexual intercourse or had done so at age 16
years or older. Among males, the controls included 88.3% who
reported never having sexual intercourse; and among those who
had, 71.3% reported first intercourse at age 16, 26.3% at age 17, and
2.5% at age 18. Among female controls, 88.5% had not had
inter-course, and of those who had, 84% had had intercourse at age 16
and 16% at age 17.
The control groups for male and female index group members
were selected using a “specialized cohorts design,” as described by Leventhal27 and Widom. This does not require the a priori
selection of matching variables by the investigator, thereby
pre-cluding their use as independent variables in the analysis as
occurs with classic case-control matching designs. Rather, the
design assumes that index and control cases approximate each
other except for the index characteristic. In this way, statistically
significant differences signal substantive differences between the
groups, which are otherwise assumed to be drawn from a similar
social milieu. This concept was put into operation by selecting the
case in the data file proximal to each index case, as long as the
proximal case did not have the index characteristic (childhood sexual intercourse). The number of control cases typically does not
match the number of index cases using this selection method,
because only one control is identified by this procedure when two
proximal index cases are encountered in the data file. Sample Modifications
Investigators have discussed the confounding of early sexual
intercourse with self-reported sexual abuse, which is problematic
in studies of adolescent sexual behavior, particularly of younger
adolescents.6 Adolescents who indicate an abuse history have
been shown to be at far higher risk for a variety of health-com-promising behaviors compared with their counterparts without a
self-reported history of abuse?3’ So as not to confound
self-reported sexual abuse and early sexual intercourse, we excluded
those reporting a history of sexual abuse from the subsequent
comparative group analysis. Statistical Analysis
Bivariate comparisons of the index and control groups initially were conducted for each dependent variable using to identify
group differences. Because of the large sample size, a criterion value of P .001 was used to reduce type I error. Each significant
variable was then entered into a logistic regression to assess the ability of those variables to differentiate the index and control groups. All bivariate and multivariate assessments were
con-ducted separately by gender. Sample Characteristics
The adjusted sample (excluding adolescents with a
self-re-ported history of sexual abuse) included 1497 males (813 index
cases and 684 controls) and 369 females (151 index cases and 218 controls). The specialized cohorts design produced no significant
differences between the index and control groups in regard to
respondents’ age or size of household.
There were expected significant group differences in regard to
respondent’s race and socioeconomic status, including a signifi-cantly greater proportion of African-American students in the
index versus control group (33.3% versus 6.7%) and more white
controls than index cases (85.0% versus 59.4%, P< .001). There was
little difference in the distributions of Hispanic, American Indian,
or Asian/Pacific Islander students. Socioeconomic status was computed as a combination of parental educational level and
employment status. Here, more index cases came from low- or
middle- than upper-class families (x = 29.87, df = 2,
P
< .001).Another measure of socioeconomic status is eligibility for free or
low-cost school lunch, and significantly more of the index group
than controls accessed subsidized lunches (37.1% versus 22.9%,
= 47.38, df = 2,
P
< .001). Parental marital status also differedmarkedly between the index and control groups; a significantly
greater proportion of the index group indicated that their parents
were divorced, separated, or never married. Initial group
differ-ences in racial composition, social class status, and parental
mar-ital status as well as age were adjusted in the subsequent
multi-variate analyses. (Additional demographic details are available
upon request.)
Bivariate Relations
RESULTS
Perceived Parental Substance Use
Among males, the index group was over one and a
half times more likely to indicate that one or both
parents had adrug or drinking problem within the past
5 years (21.3% versus 13.0%, = 15.87, df= 1, P < .001).
A significantly greater proportion of index males and
females reported use of marijuana by either parent,
including a substantially greater likelihood of parental
use monthly or more often (Table 1).
Academic Risk and Gang Activity
A greater proportion of the index group was
aca-demically at risk, with over twice as many index
males and females as controls indicating
below-av-erage school performance. This contrast was
signffi-cant among males (P < .001) and nearly significant
among females (P = .003, Table 2). Male and female
index cases also were significantly more likely than
controls to indicate gang involvement, including
per-sonal participation in gangs or having friends in
gangs (Table 3).
Sexual Behaviors
Among students who had ever had sexual
inter-course, a greater proportion of male index cases than
controls reported having intercourse several times a
week (16.8% versus 9.3%) or approximately every
day (12.4% versus 2.7%), although these differences
did not quite attain significance
(x
= 12.19, df = 3, P= .007). Among females, there were varied findings:
one and a half times as many index as control cases
reported that they rarely had sexual intercourse
(56.3% versus 37.5%), whereas over four times as
many index cases indicated that they had intercourse
nearly every day (18.5% versus 4.2%). These
differ-ences again did not quite attain significance
(x
=14.82, df = 3, P = .002).
Respondents were asked to indicate what kind of
birth control they and/or their partner used most
often, including withdrawal, foam, condoms,
rhythm, diaphragm, birth control pills, sponges,
douches, intrauterine device or none. Response
cat-egories were combined and classified as noted in
Table 4. Male index cases were over twice as likely as
controls to indicate that they and their partner did
not use any form of contraception (31.5% versus
13.3%), were more likely to use ineffective methods
(13.2% versus 9.3%), and were less likely to have a
partner using oral contraceptives (8.3% versus
20.0%) or barrier methods (47.0% versus 57.3%).
TABLE 1. Perceived Use of Substances by Parents
Male Respondents Female Respondents
Index Group Controls Totals Index Group Controls Totals
Parents with problem use
Yes 158 83 241 29 35 64
21.3% 13.0% 17.5% 21.3% 17.2% 18.8%
No 584 556 1140 107 169 276
78.7% 87.0% 82.5% 78.7% 82.8% 81.2%
Totals 742 639 1381 136 204 340
53.7% 46.3% 100% 40.0% 60.0% 100%
(x2= 15.87, df = 1,
P
= .000)(x2
0.674, df 1,P
.412)Father’s marijuana
use
Never 533 567 1100 103 182 285
83.4% 95.8% 89.4% 83.1% 96.8% 91.3%
<Monthly 31 12 43 7 2 9
4.9% 2.0% 3.5% 5.6% 1.1% 2.9%
Monthly or more 75 13 88 14 4 18
11.7% 2.2% 7.1% 11.3% 2.1% 5.8%
Totals 639 592 1231 124 188 312
51.9% 48.1% 100% 39.7% 60.3% 100%
(x2= 51.41, df = 2,
P
= .000)(x’
17.85, df 2,P
.000)Mother’s marijuana use
Never 636 621 1257 117 202 319
86.9% %.4% 91.4% 82.4% 96.7% 90.9%
<Monthly 35 17 52 8 4 12
4.8% 2.6% 3.8% 5.6% 1.9% 3.4%
Monthly or more 61 6 67 17 3 20
8.3% 0.9% 4.9% 12.0% 1.4% 5.7%
Totals 732 644 1376 142 209 351
53.2% 46.8% 100% 40.5% 59.5% 100%
(x2= 46.12, df = 2, P = .000)
(x2
= 21.79, df = 2,P
= .000)TABLE 2. Self-Assessed School Performance
School Performance Index Group Controls Totals
Male respondents
Below average 106 37 143
13.2% 5.4% 9.6%
Average 436 307 743
54.2% 45.0% 50.0%
Above average 262 338 600
32.6% 49.6% 40.4%
Totals 804 682 1486
54.1% 45.9% 100%
(x2= 55.68, df = 2,
P
= .000)Female respondents
Below average 14 9 23
9.4% 4.1% 6.3%
Average 88 104 192
59.1% 47.7% 52.3%
Above average 47 105 152
31.5% 48.2% 41.4%
Totals 149 218 367
40.6% 59.4% 100%
(x2= 12.00, df = 2,
P
= .003)females, almost four times as many index cases as
controls indicated that they and their partner did not
use any form of contraception (47.9% versus 12.0%);
they were less than half as likely as controls to use
oral contraceptives (1 1 .4% versus 28.0%), but also
less than half as likely to use ineffective methods
of contraception (12.9% versus 28.0%). These
differ-ences did not quite attain the threshold level of
sig-nificance (P = .003, Table 4). Index cases were almost
15 times more likely than controls to indicate that
they had ever been involved in a pregnancy, a highly
significant difference for both males and females
(Table 5).
Family Relationships
There were no group differences in feelings of
being cared for and connected to parents, family, and
other adults, although there was a trend toward more index cases indicating that parents and other
adults did not care about them, and that their
families did not understand them. A significantly
TABLE 3. Involvement With Gangs
Gang Involvement Index Group Controls Totals
Male respondents
Not involved 205
36.4%
332 537
72.6% 52.6%
Friends involved 293
52.0%
110 403
24.1% 39.5%
Personally involved 65
11.5%
15 80
3.3% 7.8%
Totals 563
55.2%
457 1020
44.8% 100%
(x2= 134.83, df = 2, P = .000)
Female respondents
Not involved 41
35.7%
87 128
65.9% 51.8%
Friends involved 65
56.5%
38 103
28.8% 41.7%
Personally involved 9
7.8%
7 16
5.3% 6.5%
Totals 115
46.6%
132 247
53.4% 100%
(x2= 22.80, df = 2, P = .000)
TABLE 4. Use of Contraception
Contraceptive Type* Index Group Controls Totals
Male respondents
Barrier 342
47.0%
43 385
57.3% 48.0%
Birth control pill 60
8.3%
15 75
20.0% 9.4%
Ineffective methods 96
13.2%
7 103
9.3% 12.8%
Neither partner uses 229
31.5%
10 239
13.3% 29.8%
Totals 727
90.6%
75 802
9.4% 100%
(x2= 19.85, df = 3, P = .000) Female respondents
Barrier 39
27.9%
8 47
32.0% 28.5%
Birth control pill 16
11.4%
7 23
28.0% 13.9%
Ineffective methods 18
12.9%
7 25
28.0% 15.2%
Neither partner uses 67
47.9%
3 70
12.0% 42.4%
Totals 140
84.8%
25 165
15.2% 100%
(x2= 13.95, df = 3, P = .003)
* Barrier methods include: condoms, diaphragm, sponges, intrauterine device. Ineffective methods were so classified according to user
failure rates, including withdrawal, foam, rhythm, and douches.
thought about leaving home “a lot of the time”
(25.3% versus 10.4%, P < .001, for males; 31.4%
versus 12.9%, P < .001, for females). A significantly
greater proportion of female index-group members
also reported that they rarely or never had fun with
their families (29.8% versus 14.8%, P < .001) and that
their families paid little or no attention to them
(32.1% versus 15.8%, P < .001).
Mental Health
Twice as many index males as controls reported a
history of mental health treatment, either during the
past year or over I year ago (P = .001). Group
differ-ences for females were not significant (Table 6).
Twice as many index males and females reported
that they had ever had a “nervous breakdown,”
again a significant difference for boys. Similar
differ-ences were evident for previous suicide attempts,
with three to four times as many index males and
females reporting attempts within the past year as
compared to controls (Table 6).
Multivariate Analyses
Significant variables next were entered into a
lo-gistic regression to identify the salience of items in
differentiating index and control cases. Net of
demo-graphic factors, up to three significant variables
emerged in the models. For males, these were
aca-demic risk, gang involvement, and frequency of
sex-ual intercourse. For females, significant variables
were gang involvement, history of mental health
treatment, and frequency of sexual intercourse. All
together, the model variables correctly classified
TABLE 5. Frequency of Causing Pregnancy/Being Pregnant
Frequency Index Group Controls Totals
Male respondents
Never 453 444 897
81.3% 97.8% 88.7%
Do not know 22 5 27
3.9% 1.1% 2.7%
One time 47 4 51
8.4% 0.9% 5.0%
Two or more 35 1 36
times
6.3% 0.2% 3.6%
Totals 557 454 1011
55.1% 44.9% 100%
(=68.39,df=3,P=.000)
Female respondents
None 101 172 273
84.2% 98.9% 92.9%
One or more 19 2 21
15.8% 1.1% 7.1%
Totals 120 174 294
40.8% 59.2% 100%
(x2= 20.93, df = 1, P = .000)
TABLE 6. History of Mental Health Indices
Male Respondents Female Respondents
Index Group Controls Totals Index Group Controls Totals
Mental health
treatment
During past year 30 12 42 2 5 7
4.7% 2.0% 3.4% 1.5% 2.6% 2.2%
Over I year ago 41 21 62 6 7 13
6.5% 3.5% 5.0% 4.6% 3.7% 4.0%
Never 561 563 1124 123 178 301
88.8% 94.5% 91.5% 93.9% 93.7% 93.8%
Totals 632 596 1228 131 190 321
51.5% 48.5% 100.0% 40.8% 59.2% 100.0%
(x2 = 13.13, df = 2, P = .001) (x = 7.47, df = 1, P = .05)
Nervous
breakdowns
Yes 77 36 113 17 10 27
12.1% 6.0% 9.1% 13.1% 5.2% 8.4%
No 559 564 1123 113 181 294
87.9% 94.0% 90.9% 86.9% 94.8% 91.6%
Totals 636 600 1236 636 191 321
51.5% 48.5% 100% 51.5% 59.5% 100%
(x2= 14.50, df = 1, P = .000) (x 7.47, df 1, P .05)
Suicide attempts
Yes, during past 44 11 55 11 4 15
year
5.4% 1.6% 3.7% 7.3% 1.8% 4.1%
Yes, >1 year ago 50 23 73 13 18 31
6.2% 3.4% 4.9% 8.6% 8.3% 8.4%
Never 719 650 1369 127 196 323
88.4% 95.0% 91.4% 84.1% 89.9% 87.5%
Totals 813 684 1497 151 218 369
54.3% 45.7% 100% 40.9% 59.1% 100%
(x2= 51.59, df = 2, P = .000)
(x2
7.00, df 2, P .05)DISCUSSION
The results of this statewide, school-based sample
of adolescents indicated that 3.4% of youths reported
having first sexual intercourse at or before age 10
years. Childhood sexual intercourse was reported by
a disproportionately high percentage of males,
Afri-can Americans, and students from families of lower
socioeconomic status or marital breakup, as
com-pared with controls. These demographic differences
were consistent with the findings of several other
investigators.4’8’#{176}”5”6 Although trends over the
past few decades indicate that changing social norms
have made it increasingly normative for adolescents
to have sexual intercourse at younger ages,32 first
intercourse during childhood is still atypical for most
young people.
The present study suggests that childhood sexual
intercourse, not perceived in adolescence as sexual
abuse, is associated with several potentially serious
indicators of distress and health-compromising
be-haviors among young people. It was associated with
TABLE 7. Significant Logistic Regression Results: Net of Demographic Factors*
Variable B Standard Error B Wald df Significance
Males
Academic performance
Below average -1.47 0.69 4.58 1 .030
Gang involvement
Friends involved -0.98 0.37 6.89 1 .009
Personally involved -1.61 0.79 4.12 1 .040
Frequency of intercourse
Several times/week -5.17 0.59 76.93 1 .000
About every day -5.54 0.64 74.74 1 .000
Females
Gang involvement
Friends involved -2.48 0.81 9.48 1 .002
Mental health treatment
>1 year ago 4.88 2.00 5.98 1 .015
Frequency of intercourse
Rarely (few times/year) -9.01 2.17 17.22 1 .001
About every day -4.59 1.19 14.79 1 .000
* Correct group classification with 15-variable model: males 90.4%, model x = 487.09, df = 26, P = .000; females 89.5%, j = 145.60, df =
26, P= .000. Variables in the model include: age, race, free/low-cost school lunch, parental marital status, parental problem substance use,
father’s use of marijuana, mother’s use of marijuana, school performance, gang involvement, frequency of sexual intercourse, wanting to leave home, history of nervous breakdown, history of mental health treatment, history of suicidal involvement, and pregnancy involvement.
unprotected and more frequent sexual intercourse,
pregnancy, emotional distress, and suicidal
involve-ment. It is important to note that the above
differ-ences were evident between the index and control
groups even when using a specialized cohort design
that maximized similarities between the groups.
Fi-nally, significant differences in risk behavior
per-sisted when controlling for the effects of background
demographic factors including age, race, family
resources, and parental marital status.
It should be noted that the use of a school-based
sample precludes generalizability of findings to
out-of-school youth. The sample size also prevented
con-sideration of all combinations of birth control use,
which would have been useful in assessing the risk
of acquiring sexually transmitted diseases.
Our results are consistent with those predicted by
problem-behavior theory, which suggests that early
sexual behavior is associated with involvement in
multiple other problem behaviors, as opposed to
conventional behaviors.33 Stanton et al9
demon-strated that in the case of resource-deprived,
inner-city African American adolescents, there is in fact
little covariation between early adolescent sexual
intercourse and other problem behaviors. However,
our study represents a downward extension of
stud-ies pertaining to intercourse in early adolescence,
rather than in childhood. These data show that it is
false to assume that all sexual intercourse in the first
decade of life is nonconsensual. Likewise, they raise
a warning flag that childhood sexual
intercourse-even when perceived as consensual and
nonabu-sive-is associated with a range of severe social and
emotional correlates.
As numerous reports and national panels have
suggested,4’ we cannot wait until the young
per-son reaches puberty before asking questions
explor-ing sexual vulnerability. For the clinician, certain
ethnic groups appear to be at higher risk for
child-hood sexual intercourse, as is the child with poor
school performance, with a parental history of
sub-stance abuse, or with a mental health history. To say
that our factor is causal goes beyond the available
data; however, when one sees the constellation of
findings associated with childhood sexual
inter-course, it is incumbent upon us as clinicians to raise
questions regarding several behaviors, for the
likeli-hood of pregnancy is exceptionally high. Conversely,
when one becomes aware of early sexual behavior,
one must consider not only abuse, but also the severe
emotional and behavioral correlates that tend to be
associated with early sexuality so as to provide
appropriate assessment and counseling.
ACKNOWLEDGMENT
This analysis was supported in part by MCH Grant HRSA/
MCH-273A-03-110.
REFERENCES
1. Newcomer S. Baldwin W. Demographics of adolescent sexual behavior,
contraception, pregnancy and STD’s. ISch Health. 1992;62:265-270
2. Dryer PH. Sexuality during adolescence. In: Handbook of Developmental Psychology. Englewood Cliffs, NJ: Prentice-Hall; 1982:559-601
3. Bingham CR, Miller BC, Adams GR. Correlates of age at first sexual
intercourse in a national sample of young women. IAdolesc Res. 1990;
5:18-33
4. Alexander CS, Ensminger ME, Kim YJ, Smith BJ,Johnson KE, Dolan U.
Early sexual activity among adolescents in small towns and rural areas:
race and gender patterns. Fam Plann Perspect. 1989;21:261-266
5. Resnick MD. Adolescent pregnancy options. I Sch Health. 1992;62:
294-303
6. Hofferth SU, Kahn JR. Baldwin W. Premarital sexual activity among US
teenage women over the past three decades. Fam Plann Perspect. 1987;
19:46-53
7. Department of International Economic and Social Affairs. Adolescent Reproductive Behavior: Evidence from Developed Countries. Vol. 1. New
York: United Nations Population Studies; 1988:109
8. Hogan D, Kitagawa E. The impact of social status, family structure and neighborhood on the fertility of Black adolescents. Am ISociol. 1985;90: 825-834
9. Stanton B, Romer D, Ricardo I, Black M, Feigelman 5, Galbraith J. Early
initiation of sex and its lack of association with risk behaviors among adolescent African-Americans. Pediatrics. 1993;92:13-19
10. Weber FF, Elfenbein DS, Richards NL, Davis AB, ThomasJ. Early sexual
activity of delinquent adolescents. IAdolesc Health Care. 1989;10:398-403
adolescence and sexual behavior and sexual adjustment in young
adulthood. Arch Sex Behav. 1989;18:299-313
12. Gillmore MR, HawkinsJD, Catalano RFJr, Day UE, Moore M.Structure
of problem behaviors in preadolescence. JConsult Clin Psychol. 199L59: 499-506
13. Faulkenberry JR. Vincent M, James A, Johnson W. Coital behaviors,
attitudes and knowledge of students who experience early coitus.
Adolescence. 1987;22:321-332
14. Udry JR. Billy JOG. Initiation of coitus in early adolescence. Am Sociol Rev. 198752:541-855
15. Handler A. The correlates of the initiation of sexual intercourse among young urban Black females. IYouth Adolesc. 1990;2:19-170
16. Rosenbaum E, Kandel DB. Early onset of adolescent sexual behavior
and drug involvement. JMan’ Fam. 1990:783-798
1 7. Scott-Jones D, White AB. Correlates of sexual activity in early adoles-cence. JEarly Adolesc. 1990;10:221-238
18. Wyatt GE. Reexamining factors predicting Afro-American and White
American women’s age at first coitus. Arch Sex Behav. 1989;18:271-298
19. Jessor R, Costa F, Jessor U, Donovan JE. Time of first intercourse: a prospective study. IPers Soc Psychol. 1983;44:608-626
20. Ketterlinus RD. Uamb ME, Nitz K, Elster AB. Adolescent nonsexual and
sex-related problem behaviors. JAdolesc Res. 1992;7:431-456
21. Blum RW, Resnick MD, Geer U,Hutton U, Rosenwinkel K. The
Minne-sota Adolescent Health Survey: implications for physicians. Minn Med. 1988;71 :143-146
22. Blum RW, Harmon B, Harris U, Geer U, Resnick MD. American Indian-Alaskan Native youth health. JAMA. 1992267:1637-1644
23. Resnick MD, Blum RW, Geer U, Hutton U, Rosenwinkel K. Pediatrics
and adolescent health: implications from the Minnesota Adolescent
Health Survey. Minn Ped. 1988:3-5
24. Remafedi G, Resnick MD, Blum RW, Bergeisen U. The demography of
sexual orientation in adolescents. Pediatrics. 1991$9:714-721
25. Story M, Rosenwinkel K, Himes JH, Resnick MD, Blum RW.
Demo-graphic and risk factors associated with chronic dieting in adolescents. Am JDis Child. 1991;145:994-998
26. Resnick MD, Chambliss SA, Blum RW. Health and risk behaviors of
urban adolescent males involved in pregnancy. Fams Soc. 1993;74:
366-374
27. Ueventhal JM. Research strategies and methodologic standards in stud-ies of risk factors for child abuse. Child Abuse Neglect. 1982;6:113-123
28. Widom CS. Child abuse, neglect, and adult behavior: research design and findings on criminality, violence and child abuse. Am J Orthopsy-chiatry. 198959:355-367
29. Rasch MA, Wagner EE. Initial psychological effects of sexual abuse on
female children as reflected in the Hand test. JPen Assess. 1989;53:
761-769
30. Saunders BE, Villeponteaux UA, Uipovsky JA. Child sexual assault as a
risk factor for mental disorders among women: a community survey.
JInterpers Viol. 1992;7:189-204
31. Mey BJV. The sexual victimization of male children: areview of
previ-ous research. Child Abuse Neglect. 1988;12:61-72
32. Udry JR. Age at menarche, at first intercourse, and at first pregnancy. I
Biosoc Sci. 1979;11:433-441
33. Jessor SU, Jessor R. Transition from virginity to nonvirginity among
youth: a social-psychological study over time. Dev Psychol. 1975;11:
473-484
34. Office ofTechnology Assessment, US. Congress. Adolescent Health: Vol.
II:Background and Effectiveness of Selected Prevention and Treatment Ser-vices. Washington, DC: United States Government Printing Office; 1991,
Chapter 10
35. Hayes C. Risking the Future: Adolescent Sexuality, Pregnancy and Child-bearing. Vol. 1. Washington, DC: National Academy Press; 1987 36. National Commission on Children. Beyond Rhetoric: A New American
Agenda for Children and Families. Washington, DC: United States Government Printing Office; 1991
37. National Guidelines Task Force. Guidelines for Comprehensive Sexuality
Education. New York: Sex Information and Education Council of the United States; 1992
38. Public Health Service, Health Resources and Services Administration.
Healthy Children 2000: National Health Promotion and Disease Prevention Objectives Related to Mothers, Infants, Children, Adolescents and Youth.
Washington, DC: United States Department of Health and Human
Services, publication HRSA-MCH-91-2; 1991
39. Healthy Kids for the Year 2000: An Action Plan for the Schools. Arlington,
VA: American Association of School Administrators; 1990 40. Kirby D. School-based programs to reduce sexual risk-taking behaviors.
ISch Health. 199262:280-287
41. Brindis C. Adolescent pregnancy prevention for Hispanic youth: the
role of schools, families and communities. I Sch Health. 1992;62:345-351
NATIONAL MEDICAL AGREES TO SETTLEMENT OF FRAUD CHARGES
TOTALING $380 MILLION
“National Medical Enterprises, Inc detailed a broad settlement of fraud charges
with the United States and 28 states involving payments of a record $380 million
and federal guilty pleas on eight criminal counts by two of its units.”
“The settlement total is more than three times the previous largest
health-care-fraud payment, $111.4 million by National Health Laboratories, Inc in 1992. It is
higher even than the nearly $300 million the government has recovered from ten
defense contractors in its Ill Wind probe, the huge Pentagon fraud investigation
that ended early this year.”
Harris Jr, R. The Wall Street Journal, June 29, 1994.
Noted by J.F.L., MD
at Viet Nam:AAP Sponsored on September 1, 2020
www.aappublications.org/news
1994;94;907
Pediatrics
Michael D. Resnick and Robert Wm. Blum
Health Risk and Behaviors
The Association of Consensual Sexual Intercourse During Childhood With Adolescent
Services
Updated Information &
http://pediatrics.aappublications.org/content/94/6/907
including high resolution figures, can be found at:
Permissions & Licensing
http://www.aappublications.org/site/misc/Permissions.xhtml
entirety can be found online at:
Information about reproducing this article in parts (figures, tables) or in its
Reprints
http://www.aappublications.org/site/misc/reprints.xhtml
1994;94;907
Pediatrics
Michael D. Resnick and Robert Wm. Blum
Health Risk and Behaviors
The Association of Consensual Sexual Intercourse During Childhood With Adolescent
http://pediatrics.aappublications.org/content/94/6/907
the World Wide Web at:
The online version of this article, along with updated information and services, is located on
American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.
