Content uploaded by Rena Repetti
Author content
All content in this area was uploaded by Rena Repetti
Content may be subject to copyright.
Risky Families: Family Social Environments and the Mental
and Physical Health of Offspring
Rena L. Repetti, Shelley E. Taylor, and Teresa E. Seeman
University of California, Los Angeles
Risky families are characterized by conflict and aggression and by relationships that are cold, unsup-
portive, and neglectful. These family characteristics create vulnerabilities and/or interact with genetically
based vulnerabilities in offspring that produce disruptions in psychosocial functioning (specifically
emotion processing and social competence), disruptions in stress-responsive biological regulatory sys-
tems, including sympathetic-adrenomedullary and hypothalamic–pituitary–adrenocortical functioning,
and poor health behaviors, especially substance abuse. This integrated biobehavioral profile leads to
consequent accumulating risk for mental health disorders, major chronic diseases, and early mortality.
We conclude that childhood family environments represent vital links for understanding mental and
physical health across the life span.
Good health begins early in life. In the first years of childhood,
the family is charged with responsibilities for the care and devel-
opment of the child. In healthy families, children learn that they
can count on the environment to provide for their emotional
security and their physical safety and well-being, and they acquire
behaviors that will eventually allow them to maintain their own
physical and emotional health independent of caregivers. From this
vantage point, a healthy environment for a child is a safe environ-
ment; it provides for a sense of emotional security and social
integration and it offers certain critical social experiences that lead
to the acquisition of behaviors that will eventually permit the child
to engage in effective self-regulation (Basic Behavioral Science
Task Force of the National Advisory Mental Health Council,
1996). Poor health also begins early in life. Research consistently
suggests that families characterized by certain qualities have dam-
aging outcomes for mental and physical health. These character-
istics include overt family conflict, manifested in recurrent epi-
sodes of anger and aggression, and deficient nurturing, especially
family relationships that are cold, unsupportive, and neglectful.
Families with these characteristics are risky because they leave
their children vulnerable to a wide array of mental and physical
health disorders. In this article, we propose a synthetic model of
these links, focusing on the pathways through which risky families
may not only hinder healthy development in childhood, but influ-
ence physical and mental health into adolescence and adulthood.
As indicated in Figure 1, risky family characteristics create a
cascade of risk, beginning early in life. Families with these char-
acteristics may create vulnerabilities and may exacerbate certain
genetically based vulnerabilities, which not only put children at
immediate risk for adverse outcomes (such as is the case with
abuse), but lay the groundwork for long-term physical and mental
health problems. Specifically, we will show that risky families
create deficits in children’s control of and expression of emotions
and in social competence, and also lead to disturbances in physiologic
and neuroendocrine system regulation that can have cumulative,
long-term, adverse effects. We especially focus on disruptions in
sympathetic-adrenomedullary (SAM) reactivity, hypothalamic–
pituitary–adrenocortical (HPA) reactivity, and serotonergic func-
tioning. Children who grow up in risky families are also especially
likely to exhibit health-threatening behaviors, including smoking,
alcohol abuse, and drug abuse; the risk for promiscuous sexual
activity in these children is also high. These forms of behavioral or
substance abuse may represent a method of compensating for
deficiencies in social and emotional development, as well as a
self-medication process whereby adolescents manage the biologi-
cal dysregulations produced or exacerbated by risky families.
Taken together, these behavioral and biological consequences of
risky family environments represent an integrated risk profile that
is associated with mental health disorders across the lifespan,
including depression and aggressive hostility, major chronic ill-
nesses including hypertension and cardiovascular disease, and
early death. As indicated in Figure 1, the cascade of accumulating
risk is also heavily influenced by the social context in which
children and families live, including factors such as chronic stress,
neighborhood violence, and poverty (Taylor, Repetti, & Seeman,
1997). In the conclusions, we identify critical issues raised by our
analysis, address links in the model for which evidence is currently
Rena L. Repetti and Shelley E. Taylor, Department of Psychology,
University of California, Los Angeles; Teresa E. Seeman, Division of
Geriatrics, School of Medicine, University of California, Los Angeles.
Preparation of this article was supported by National Institute of Mental
Health Grants R29-48593 and MH 056880, National Science Foundation
Grant SBR 9905157, a training grant from the National Institute of Mental
Health (Biobehavioral Issues in Physical and Mental Health, MH 15750
NIMH), Grant NIA AG-17265 from the National Institute on Aging’s
Biodemography Center, and by the MacArthur Foundation’s Network on
SES and Health. The authors wish to express their gratitude to Mary Hsieh
for maintaining a complex bibliography and to Terry Au, Jay Belsky,
Thomas Bradbury, Andrew Christensen, Mark Cummings, Christine
Dunkel-Schetter, Jaana Juvonen, Laura Klein, Karen Matthews, Kathleen
McCartney, and Bruce McEwen for their helpful suggestions on a previous
version of this article.
Correspondence concerning this article should be addressed to Rena L.
Repetti, Department of Psychology, University of California, Los Angeles,
405 Hilgard Avenue, Los Angeles, California 90095-1563. E-mail:
repetti@psych.ucla.edu
Psychological Bulletin Copyright 2002 by the American Psychological Association, Inc.
2002, Vol. 128, No. 2, 330–366 0033-2909/02/$5.00 DOI: 10.1037//0033-2909.128.2.330
330
insufficient, propose a research agenda for the future, and discuss
the potential role of intervention in offsetting the cascade of risk
that is generated by risky families.
Risky Family Environments and Mental and Physical
Health Outcomes
The backbone of the model depicted in Figure 1 is an associa-
tion between growing up in a risky family environment and poor
mental and physical health outcomes, with the most serious con-
sequences manifested in adulthood. In this section, we describe
these adverse family characteristics and present empirical evidence
of their connections to mental and physical health outcomes.
Mental Health
Anger and aggression are highly noxious agents in a family
environment. Conditions ranging from living with irritable and
quarreling parents to being exposed to violence and abuse at home
show associations with mental and physical health problems in
childhood, with lasting effects into the adult years. There is over-
whelming documentation in the research literature that overt con-
flict and aggression in the family are associated both cross-
sectionally and prospectively with an increased risk for a wide
variety of emotional and behavioral problems in children, includ-
ing aggression, conduct disorder, delinquency and antisocial be-
havior, anxiety, depression, and suicide (Emery, 1982, 1988;
Grych & Fincham, 1990; Kaslow, Deering, & Racusia, 1994; R. J.
Reid & Crisafulli, 1990; Wagner, 1997). Empirical efforts to tie
different types of maltreatment and abuse in the home to different
forms of psychopathology reveal only a general association of
family violence and child psychopathology (Emery & Laumann-
Billings, 1998). Parenting that constrains, invalidates, and manip-
ulates children’s psychological and emotional experience and ex-
pression is also related to both internalizing and externalizing
symptoms
1
(Barber, 1996). A meta-analysis of 47 studies found
1
Behavioral problems in childhood are often discussed in terms of two
broad symptom categories that relate to self-control or self-regulation
(although the two types of symptoms tend to co-occur). Externalizing
symptoms involve aggression and hyperactivity and are sometimes referred
to as problems of undercontrol. Internalizing symptoms, which involve
social withdrawal and negative emotions such as anxiety, are sometimes
referred to as problems of overcontrol. Thus, in the case of externalizing
symptoms, the child has difficulty successfully inhibiting socially prohib-
ited behavior and controlling impulses and, in the case of internalizing
symptoms, on the other hand, there often appears to be an extreme level of
behavioral inhibition (Eisenberg, Fabes, Guthrie, et al., 1996).
Figure 1. Risky families model. A time line on the left provides a rough idea of the point in development when
an effect of the family environment is first observed. The social context of the family and genetic factors may
directly and indirectly influence all of the variables in the model: the family environment, the sustaining factors,
and health outcomes.
331
RISKY FAMILIES
that higher levels of coercive control in the family were related to
problems of undercontrol on the part of children, particularly more
aggressive and noncompliant behavior (Rothbaum & Weisz,
1994), a relation that is argued to be bidirectional.
Families characterized by high levels of conflict, aggression,
and hostility are often lacking in acceptance, warmth, and support.
However, there is evidence that inadequate emotional nurturance is
independently associated with poor mental health outcomes. Our
use of the adjectives cold, unsupportive, and neglectful covers a
wide range of family characteristics and measures in the research
literature, including emotional neglect of children; parenting that is
unresponsive or rejecting; a lack of parental availability for, in-
volvement in, and supervision of child activities; a lack of cohe-
siveness, warmth, and support within the family; and an experi-
ence of alienation, detachment, or feelings of lack of acceptance by
children. Research studies that assess these characteristics of fam-
ily life report reliable associations between them and a broad array
of mental health risks, including internalizing symptoms such as
depression, suicidal behavior, and anxiety disorders (Chorpita &
Barlow, 1998; Kaslow et al., 1994), and externalizing symptoms
such as aggressive, hostile, oppositional, and delinquent behavior
(Barber, 1996; Rothbaum & Weisz, 1994; Steinberg, Lamborn,
Darling, Mounts, & Dornbusch, 1994).
Although genetic predispositions appear to account for some of
the observed relations between risky family social environments
and child mental health (Plomin, 1994), evidence also suggests
that parenting practices have both direct and indirect effects. A
longitudinal adoption study found that children judged to be at
genetic risk for behavioral problems were more likely to receive
negative parenting from their adoptive parents than were children
not at risk. However, the genetic risk did not explain the associ-
ation of negative parenting and children’s externalizing behavior,
suggesting that environmentally mediated parenting effects on
children’s behavior was a plausible pathway (O’Connor, Deater-
Deckard, Fulker, Rutter, & Plomin, 1998). This conclusion is
supported by 16–18-year longitudinal data showing a significant
association between maladaptive parental behavior and an in-
creased risk of psychiatric disorder in offspring during late ado-
lescence and early adulthood, even after controlling for parental
psychopathology and earlier offspring characteristics (i.e., psychi-
atric disorders during early adolescence and difficult temperament
in childhood; Johnson, Cohen, Kasen, Smailes, & Brook, 2001). In
addition, genetic predispositions may act to increase vulnerability
to risky family environments. For example, a research review
suggests that families with high levels of conflict may not be able
to provide the extra support needed by a child who is at a biolog-
ical risk for depression (Kaslow et al., 1994). In fact, many of the
family environmental effects represented in Figure 1 may well
assume the form of gene–shared environment interactions. With
very few exceptions, the dominant research design in the behav-
ioral genetics field, twin studies, does not model gene–environ-
ment interactions; any variance due to those interactions is attrib-
uted to genetic effects.
2
The assumption made in twin studies, that
these gene–environment interactions are negligible, and therefore
can be ignored, is unwarranted.
Although they may live in the same home, siblings do not
experience their family environment in precisely the same way.
Most obviously, in families with more than one child, each child
has a unique constellation of siblings. In addition, parenting not
only varies between families but also within families, because the
same parent may use different child-rearing techniques and behav-
iors with different children. Thus, in the review of research that
follows, assessments of parent–child interaction or parenting can-
not necessarily be understood as family-level variables. For exam-
ple, in a study of genetic and environmental determinants of
adolescent depression and antisocial behavior, 60% of the variance
in adolescent antisocial behavior and 37% of the variance in
depressive symptoms could be accounted for by nonshared envi-
ronmental factors, specifically conflictual and negative parental
behavior directed to an adolescent; when harsh parental behavior
was directed at a sibling, it appeared to have protective effects for
an adolescent (a phenomenon known as the sibling barricade;
Reiss et al., 1995).
Whether the unit of analysis is the shared family environment or
the parent–child relationship, comprehensive reviews of the re-
search literature associate family relationships that are marked by
high levels of anger and aggression or that are cold, unsupportive,
or neglectful, with mental health problems in childhood and
adolescence.
Physical Health
There also is growing evidence that offspring of risky families
have increased rates of a wide variety of physical health problems
throughout life. For example, short of permanent disability or
death, a history of abuse is associated with chronic problems
resulting from injuries and, in adulthood, with a broad array of
physical symptoms and medical diagnoses (Walker et al., 1999). A
study of 13,494 adults found a strong, graded relationship between
breadth of exposure to abuse or household dysfunction during
childhood and risk for certain adult diseases, including ischemic
heart disease, cancer, chronic lung disease, skeletal fractures, and
liver disease (Felitti et al., 1998). Because there are no reviews of
this literature, Table 1 summarizes research findings that illustrate
a link between growing up in a risky family environment and
physical health outcomes. The table is restricted to studies that
were based on a longitudinal or follow-up research design,
3
or
studies that used data from different sources, or studies with both
characteristics. Many of the studies included observational or
objective measures to ensure that respondent bias did not inflate
the association between the measure of the family environment
and the health outcome.
Support for a link between family conflict and physical health
was found in a representative sample of the Swedish population in
which reports of “serious dissention” in the family during child-
hood were associated with a variety of self-reported illnesses 13
years later, after controlling for psychological distress and mental
illness (O. Lundberg, 1993). Other research, also summarized in
2
That is because interactions between genetic predispositions and fam
-
ily characteristics that are experienced by all siblings (referred to as shared
family environments in the behavioral genetics literature) act to increase the
similarity of monozygotic twins relative to dizygotic twins and are, thus,
automatically included as part of the heritability estimates in twin studies.
3
We use the term follow-up research design for studies that have
multiple data collection points but that do not assess change over time in
the outcome. We reserve the label longitudinal research design for studies
in which change over time in the outcome variable is assessed.
332
REPETTI, TAYLOR, AND SEEMAN
Table 1
Risky Family Characteristics and Physical Health
Study and design Sample characteristics Family environment measure Physical health outcome Key finding
Conflict and aggression
O. Lundberg, 1993
GC/FLW-UP
SELF
Representative sample of Swedish
population (n ⫽ 4,216). Time 1
(1968) age range: 17–62 years,
Time 2 (1981) age range: 30–
75 years.
Time 1: individual responses in 1968 to
“Was there any severe dissention in
your family during your upbringing?”
(10% reported “yes” or “uncertain”).
Time 2, health status: self-reports in 1981
of physical illness (e.g., aches and pains,
high blood pressure) and distress (e.g.,
tiredness, anxiety).
Severe dissention in the family during
upbringing was associated with an
increased risk, 13 years later, of
self-reported illnesses and other
indicators of distress. The increased
risk of health problems at follow-
up was found even after controlling
for psychological distress and
mental illness at Time 1.
Mechanic & Hansell, 1989
CORR/LONG
SELF
7th, 8th, & 9th graders,
N ⫽ 1,067
Time 1: adolescent report of
quarrelling or fighting in home.
Time 2: adolescent report of common
physical symptoms.
More quarelling and fighting at home
was associated with more physical
symptoms 1 year later (controlling
for the earlier level of symptoms).
Conflict was also associated with
increases in depression and anxiety.
Montgomery et al., 1997
GC/FLW-UP
IND RPRTS
OBJ
Representative sample of
over 6,500 participants in a
national longitudinal study of a
British birth cohort. Time 1
age: 7 years, Time 2 age: 33
years.
Time 1: family conflict and
difficulties (4.5% of cohort) based
on a health visitor’s report of family
difficulties (due to domestic tension,
divorce, separation, desertion).
Time 1 and Time 2: assessment of height. Exposure to family conflict was
associated with less height
attainment at ages 7 and in
adulthood. Those who had been
exposed to family conflict were
more likely to be in the lowest 5th
percentile of height distribution
(31% in childhood and 24% in
adulthood).
Stein et al., 1994
CORR/CS
OBS & OBJ
Mothers with (n ⫽ 34) or without
(n ⫽ 24) eating disorders and
their children. Child age range:
12–14 months.
Mother–child conflict: based on an
at-home observation of mother–infant
interaction during mealtime.
Infant weight. More conflict was associated with
lower infant weight attainment, even
after controlling for birth weight and
maternal height, and for mothers’
concerns about her body shape and
whether she had an eating disorder.
Walker et al., 1999
GC/CS
IND RPRTS
N ⫽ 1,225 women enrolled in a
large HMO. M age ⫽ 42 ⫾ 12
years.
Self-reported maltreatment in
childhood. Women in the nonsexual
maltreatment group reported a
history of one or more of the
following: physical abuse, physical
neglect, emotional neglect, or
emotional abuse. Women in the
sexual maltreatment group reported
a history of sexual abuse.
Physician-coded diagnoses during 18-
month period prior to the study, of
minor infectious diseases (e.g., urinary
tract infections, upper respiratory
infections) and other diseases (e.g.,
hypertension, diabetes). Self-reported
physical symptoms during the
previous 6 months.
Women who reported a history of
either sexual or nonsexual
maltreatment during childhood also
reported more physical symptoms,
and their medical records showed
that they suffered from a greater
number of minor infectious
diseases and other diseases, when
compared to women who did not
report histories of childhood
maltreatment.
Weidner et al., 1992
CORR/CS
OBJ
Children and their parents,
N ⫽ 64. Child age range:
6–18 years.
Family conflict: parent report of the
degree to which open expressions of
anger and aggression and conflictual
interactions characterize the family.
Risk of coronary heart disease based on
sample of child’s blood: ratio of total
plasma cholesterol divided by high
density lipoprotein cholesterol.
A high-conflict family environment
was associated with an unfavorable
plasma lipid profile in sons. There
was no association between family
conflict and daughters’ cholesterol.
(table continues)
333
RISKY FAMILIES
Study and design Sample characteristics Family environment measure Physical health outcome Key finding
Cold, unsupportive, and neglectful homes
Gil et al., 1987
CORR/CS
IND REPORTS
Children being treated for severe
atopic dermatitis (n ⫽ 44).
Child age range: 2–21 years.
M child age ⫽ 6.9 years.
Parent or child (if ⬎ 12 years of age)
report of levels of family
organization and cohesion.
Symptom severity: specialty nurse’s report
of the percentage of body area affected
by atopic dermatitis; records of periods
of remission.
Less family cohesiveness was
associated with more severe
symptoms.
Gottman & Katz, 1989
CORR/CS
OBS
Parents and children (n ⫽ 56
families). Age range of child:
4–5 years.
“Negative” or poor parenting
composite: based on observations of
parent behavior during interactions
with child in the lab, a style that was
unstructured, cold, unresponsive,
angry, etc.
Child illness: based on mothers’ reports on
a health scale.
Children of parents with a negative
parenting style had higher rates of
illness.
Gottman et al., 1996, 1997
CORR/FLW-UP
Parents and children (n ⫽ 53
families). Age range of child
Time 1: 4–5 years, Time
2: 6.8–9.2 years.
Time 1, parent “emotion coaching”
composite: 11 ratings, based on an
interview with the parent, of the
degree to which the parent offers
acceptance and assistance when the
child feels anger and sadness (such
as talking, intervening, comforting,
teaching about emotion expression
and coping strategies).
Time 2, childhood illness: parent report of
the child’s overall health and proneness
to illness.
Children whose mothers were poor at
“emotion coaching” had higher
rates of illness and poor health 3
years later. Fathers’“emotion
coaching” was not related to child
health.
Lissau & Sorensen, 1994
GC/LONG
IND RPRTS
Representative sample of Danish
population (n ⫽ 756). Time 1
(1974) child age range: 9–10
years, Time 2 (1984–1985)
child age range: 20–21 years.
Time 1: parent support based on
teacher rating (7% of parents
described as providing “no support,”
57% described as “harmonious”).
Neglect based on a child hygiene
rating provided by the school
medical service (4% described as
“dirty & neglected,” 21% “well-
groomed,” and 75% “averagely
groomed”).
Time 2: obesity in adulthood indicated by
self-reported weight and height; a body
mass index (weight/height) at or above
the 95th percentile.
Children of nonsupportive parents
were at greater risk of obesity in
early adulthood (18% vs. 3% of
children with “harmonious”
support), even after controlling for
body mass index in childhood and
child sex. Neglected children were
also more likely to be obese: 29%
versus 3–4%.
Martin et al., 1998
CORR/CS
OBS & OBJ
Parent (mostly mother)–child
dyads (child under treatment
for diabetes; N ⫽ 74). Child
age range: 8–18 years, M age
⫽ 13.0 years.
Parenting based on observations of
parent–child interaction: parental
emotional support, dyadic conflict
resolution, parent expressions of
warmth, anger, sadness.
Metabolic control of the diabetes assessed
by glycosylated hemoglobin, a metabolic
measure based on blood tests, assessed
over several months.
Children and adolescents whose
parents were less nurturing had
poorer metabolic control over
diabetes.
Russek & Schwartz, 1997
GC/FLW-UP
Men (N ⫽ 85). Time 1: Harvard
undergraduates from classes of
1952, 1953, & 1954, Time 2:
35 years later.
Time 1: student description of
relationship with each parent coded
as positive, “very close” or “warm
and friendly,” or negative, “tolerant”
or “strained and cold” (12% of
relationships with mothers and 20%
with fathers were “negative”).
Time 2: health status based on in-person
interviews and review of available
medical records. Diagnosed conditions
included cardiovascular disease, duodenal
ulcer, and alcoholism.
Men who described a negative
relationship with either their mother
or their father were more likely to
have a diagnosed disease 35 years
later: 85–91% who had described
negative relationships with a parent
compared with 45–50% of those
who had described positive
relationships.
334
REPETTI, TAYLOR, AND SEEMAN
Table 1 (continued)
Study and design Sample characteristics Family environment measure Physical health outcome Key finding
Cold, unsupportive, and neglectful homes (continued)
Shaffer et al., 1982
GC/FLW-UP
SELF
White male physicians (n ⫽ 913)
who graduated from medical
school between 1948 and 1964
and whose health status was
assessed annually.
Medical students’ descriptions of
family members’ attitudes toward
each other (with respect to both
parent–child and marital
relationships) in positive terms
(warm, close, understanding,
confiding) and negative terms
(detached, dislike, hurt, high-tension).
Physicians’ self-reports of a diagnosed
cancer in annual follow-up
questionnaires and interviews during the
years following graduation (25 of 913
participants had a diagnosed cancer).
Men who described more negative
and less positive family
relationships were at increased risk
of future cancer, even after
controlling for health risk factors
such as age, alcohol use, cigarette
smoking, being overweight, and
serum cholesterol levels.
M. Valenzuela, 1997
CORR/CS
OBS/OBJ
Mother–infant dyads (n ⫽ 85).
Full-term, healthy, normal birth
weight babies. Urban, poor
socioeconomic status mothers
in Santiago, Chile. Child age
range: 17–21 months.
Maternal sensitivity based on
observations during a home visit.
Maternal behavior during a problem-
solving task used to assess support
(ability to elicit and maintain child’s
interest and prevent frustration) and
quality of assistance (ability to help
and guide child by scaffolding
instructions to solve problems).
Infant weight for age (compared with
norms); chronic undernourishment
status based on longitudinal data on
health, height, and weight.
Infants with less sensitive mothers
were not growing as well (lower
weight for age) as other infants.
Infants whose mothers were less
supportive and less able to assist
the child were more likely to be
chronically undernourished.
Wickrama et al., 1997
CORR/LONG
OBS
White adolescents and their
families living in a rural area
(N ⫽ 310). Time 1 child age:
7th grade, Time 5 child age:
11th grade.
Time 1, affective quality of parent
behavior toward the adolescent: (a)
observer ratings of two family
discussions (amount of derogatory,
insulting, contemptuous behavior
minus the amount of affectionate and
supportive behavior and
compliments) (b) adolescent report of
the frequency of negative parent
behavior (anger, criticism, threats,
ignoring) minus the frequency of
positive behavior (care, affection,
appreciation, listening).
Time 1–Time 5: annual assessments of
adolescent health, based on self-reports
of 12 common physical symptoms (e.g.,
headaches, sore throat, muscular aches,
stomach aches, congested nose, skin rash,
allergies).
In a latent growth curve analysis,
more hostile and less supportive
parent behavior during family
interactions, as well as adolescents’
reports of more negative and less
positive parent behavior, predicted
increases in the adolescents’
physical health complaints over the
next 4 years.
Note. The key family and outcome variables in each study are highlighted in bold type. For each study, design (GC ⫽ group comparison; CORR ⫽ correlational), timing of assessments (FLW-UP ⫽
multiple data collection points, but change over time in the outcome is not assessed; LONG ⫽ longitudinal; CS ⫽ cross-sectional), and sources of data for primary variables (SELF ⫽ all self-report
data; IND RPRTS ⫽ assessment of predictor and outcome variables based on data from separate sources; OBJ ⫽ objective measures; OBS ⫽ observational data) are reported.
335
RISKY FAMILIES
Table 1 (continued)
the first half of Table 1, indicates that family conflict and aggres-
sion has adverse effects on health in childhood and adulthood, and
on physical growth and development. Two of these studies used a
longitudinal or follow-up research design (Mechanic & Hansell,
1989; Montgomery, Bartley, & Wilkinson, 1997), and three others
used objective measures of health outcomes (Stein, Woolley, Coo-
per, & Fairburn, 1994; Walker et al., 1999; Weidner, Hutt, Connor,
& Mendell, 1992).
Findings summarized in the second part of Table 1 show that
growing up in a cold, unsupportive, or neglectful home is also
associated with poor physical health and development. Investiga-
tions based on longitudinal and follow-up research designs have
tied a lack of support and deficient nurturing during childhood to
higher rates of illness and physical complaints several years later
(Gottman, Katz, & Hooven, 1996, 1997; Wickrama, Lorenz, &
Conger, 1997), to obesity in early adulthood (Lissau & Sorensen,
1994), and to more serious medical conditions in midlife (Russek
& Schwartz, 1997; Shaffer, Duszynski, & Thomas, 1982). Cross-
sectional studies that included independent assessments of a child
health outcome and the family environment have shown links to
poorer growth during infancy (M. Valenzuela, 1997), poorer gen-
eral health (Gottman & Katz, 1989), and, among children with a
diagnosed medical problem, less control over or more severe
symptoms of the disease (Gil et al., 1987; M. T. Martin, Miller-
Johnson, Kitzmann, & Emery 1998).
Summary
In summary, diverse research literatures consistently point to
adverse developmental effects of the two general characteristics of
a risky family social environment, conflict and aggression and a
cold, unsupportive, or neglectful home. Besides direct effects on
health, such as can be the case with physical abuse, the effects may
be mediated and sustained by disruptions in the child’s ability to
mount a successful physiologic/neuroendocrine and/or behavioral
response to stress, and to acquire appropriate emotional and be-
havioral self-regulatory skills, issues to which we now turn.
Mediating and Sustaining Factors
How are the long-lasting, even permanent, health consequences
of growing up in a risky family environment sustained over the
lifetime? Our model proposes that children from risky families
experience disruptions in their physiologic/neuroendocrine func-
tioning, especially in response to stress, and also develop deficits
in emotion processing, social competence, and behavioral self-
regulation. Although there is no essential mapping across these
systems, the emotional, social, and biological disruptions appear to
be linked to each other in a cascade arrangement. Figure 1 depicts
a succession of developmental processes whose course can be
influenced by risky family environments, so that early disruptions
continue to have an impact on development in future stages. Not
all of the connections in Figure 1 have been convincingly made,
but emerging evidence suggests the value of considering these
self-regulatory and biological dysfunctions as aspects of an inte-
grated profile of risk.
Alterations in Physiological/Neuroendocrine Responses to
Stressful Situations
The bulk of damage done to physical health in risky families
may come from the initiation of biologically dysregulated re-
sponses to stress, the effects of which may be cumulative over the
lifespan. As a result, the trajectories of major causes of morbidity
and mortality in developed countries, namely the chronic diseases
of hypertension, cardiovascular disease, diabetes, and some can-
cers, may begin as early as childhood in these biological dysregu-
lations. The evidence currently suggests that constant or recurrent
exposure to the stressful circumstances created by risky families
may lead to alterations in the SAM system and HPA axis responses
to stress and to disruptions in serotonergic functioning.
4
Deficien-
cies in the ability to mount a parasympathetic nervous system
(PNS) response to stress may be affected as well. Given the
interdependence of different components of the neuroendocrine
system, we would anticipate disruptions in other systems as well,
for example in dopamine regulation, but at present these links are
not documented.
Dysregulation in vital biological regulatory systems may occur
in utero or in infancy, periods that are thought to be critical in the
normal development of biological regulatory systems. Dysregula-
tion may also result from damage due to repeated activation of
these systems. In making this case, we draw on the concept of
allostatic load (McEwen & Stellar, 1993), arguing that repeated
social challenges in a child’s environment can disrupt basic ho-
meostatic processes that are central to the maintenance of health.
The consequences of exposure to these sources of family stress
early in childhood may be cascading, potentially irreversible in-
teractions between genetic predispositions and these environmen-
tal factors that, over time, can lead to large individual differences
in susceptibility to stress, in biological markers of the cumulative
effects of stress, and in stress-related physical and mental disor-
ders. The fact that risky families appear to fuel such a wide range
of mental and physical health disorders lends credence to the role
that such families may play in exacerbating a broad array of
genetic predispositions. Accumulating allostatic load is believed to
lead to accelerated aging, which may include, over the long term,
chronic hypertension, slower cardiovascular recovery from stress,
hippocampal atrophy, and certain cognitive dysfunctions, signs of
dysregulation of the HPA axis (such as flat or prolonged cortisol
response to stress), dysregulation of the PNS, and dysregulation of
serotonergic functioning, among other deleterious biological
changes (see McEwen, 1998; Seeman & Robbins, 1994; Seeman,
Singer, Horwitz, & McEwen, 1997; for reviews). We confine our
coverage here to system dysregulations that have been related both
to the antecedent conditions of risky families and to the deleterious
consequences for mental and physical health outcomes.
SAM Functioning
The chronic stress of a risky family may produce repeated or
chronic SAM activation in children, or both, which in turn leads to
wear and tear on the cardiovascular system that, over time, may
4
A caution needs to be kept in mind that not all indicators of the
functioning of these systems show these effects.
336
REPETTI, TAYLOR, AND SEEMAN
lead to pathogenic changes in sympathetic or parasympathetic
functioning, or both. Such a model has been put forth by several
investigations as the developmental underpinnings of essential
hypertension (e.g., Ewart, 1991) and coronary heart disease (e.g.,
Woodall & Matthews, 1989). A secondary pathway to the same
adverse cardiovascular outcomes argues that the impact of risky
families on the SAM functioning of offspring routes through the
development of a hostile interpersonal style, which itself increases
the frequency of conflictual social interactions, recurrent SAM
activation, and the likely development of risk factors for coronary
heart disease (CHD). There is empirical evidence suggestive of
both models.
Most children show increases in sympathetic arousal in response
to exposure to angry adult interactions (El-Sheikh, Cummings, &
Goetsch, 1989), and children in families marked by conflict are
exposed to this physiological activation on a recurrent basis. Such
repeated exposure may lead to alterations in SAM reactivity to
stress. In support of this reasoning, Woodall and Matthews (1989)
found that boys (but not girls) from families characterized as less
supportive by parents had stronger heart rate responses to a series
of laboratory stressors, compared with boys from more supportive
families. These family characteristics were also associated with
higher anger and hostility, providing suggestive evidence for the
pathway to adverse CHD outcomes via hostility, but this media-
tional hypothesis was not directly tested in the study. Luecken
(1998) similarly found that college students who reported poor
family relationships in childhood had higher blood pressure, both
at resting level and in response to a laboratory challenge. Early
studies relating so-called Type A behaviors in children to height-
ened cardiovascular responses to stress (e.g., Brown & Tanner,
1988; Lawler, Allen, Critcher, & Standard, 1981; U. Lundberg,
1983; Matthews & Jennings, 1984; T. H. Schmidt, Thierse, &
Eschweiler, 1986) are also consistent with this link. Such changes
in reactivity have, in turn, been tied to traditional risk factors for
CHD. For example, cardiovascular reactivity to stress among boys
as young as 8 to 10 years old has been associated with increased
left ventricular mass, a risk factor for CHD (Allen, Matthews, &
Sherman, 1997). A genetic basis for heightened or prolonged
sympathetic reactivity to stress may also be implicated in the
phenotypic expression and exacerbation of these responses. In
support of this reasoning, Ballard, Cummings, and Larkin (1993)
found that children of hypertensive parents showed greater systolic
blood pressure reactivity to interadult anger than children of non-
hypertensive parents. They suggested that familial transmission of
essential hypertension may be mediated, in part, by recurrent
exposure to such hostile episodes (see also Ewart, 1991).
In sum, there is evidence that a risky family environment is
associated with heightened cardiovascular reactivity to stress,
which is linked with cardiovascular disease risk factors in children.
Studies that track all of these links within a single population are
lacking, however, and should be a focus of future work. The
second pathway, namely that risky families increase a propensity
for an angry, hostile interpersonal style, which, in turn, heightens
risk for cardiovascular disease is discussed further in the section on
social competence.
5
Alterations in SAM reactivity in children
exposed to risky family environments may also reflect contribu-
tions from the counterregulatory parasympathetic nervous system.
Negative emotions such as anxiety and hostility have been asso-
ciated with lower heart rate variability, a marker of reduced para-
sympathetic response (Kawachi, Sparrow, Vokonas, & Weiss,
1994; Sloan et al, 1994), and lower heart rate variability has in turn
been linked to increased health risks (Kristal-Boneh, Raifel,
Froom, & Rivak, 1995). These findings suggest that children in
risky families may experience reductions in PNS activity, an
important counterregulatory “brake” on SAM activity, thereby
contributing to dysregulation of the SAM system.
HPA Functioning
Alterations in HPA reactivity have also been tied to risky family
characteristics. In response to stress, the hypothalamus releases
corticotrophin-releasing hormone (CRH), which influences the
pituitary gland to secrete adrenocorticotropic hormone, which, in
turn, stimulates the adrenal cortex to release corticosteroids. This
integrated pattern of HPA activation modulates a wide range of
somatic functions, including energy release, immune activity,
mental activity, growth, and reproductive function. Appropriate
cortisol regulation (an indicator of HPA responses to stress) allows
the body to respond to stress by preparing for short-term demands.
However, persistent activation of the HPA system is associated
with immune deficiencies, inhibited growth, delayed sexual matu-
rity, damage to the hippocampus, cognitive impairment, and cer-
tain forms of psychological problems, such as depression (Chrou-
sos & Gold, 1992). As such, chronically elevated corticosteroids
have potentially deleterious effects on developing competent cog-
nitive and emotional functioning (Gunnar, 1998).
The cortisol response may be adversely affected in the offspring
of families characterized by conflict, anger, and aggression. J.
Hart, Gunnar, and Cicchetti (1996) found that children who had
been physically abused in their families had somewhat elevated
afternoon cortisol concentrations; maltreated children who were
also depressed had lower morning cortisol concentrations com-
pared with nondepressed maltreated children and were more likely
to show a rise, rather than the expected decrease in cortisol, from
morning to afternoon. HPA functioning also appears to be affected
by parental nurturing behavior. Studies of rhesus monkeys have
found that ventral contact between offspring and mother following
a threatening event promotes rapid decreases in HPA activity
(Gunnar, Gonzalez, Goodlin, & Levine, 1981; Mendoza, Smother-
man, Miner, Kaplan, & Levine, 1978). Paralleling the animal
results, a study of 264 infants, children, and adolescents found that
a family environment characterized by few positive affectionate
interactions and a high level of negative interactions, including
irrational punishment and unavailable or erratic attention from
parents, was associated with abnormal cortisol response profiles,
diminished immunity, and frequent illnesses (Flinn & England,
1997). Chorpita and Barlow (1998) noted that in families charac-
terized by low levels of warmth and high levels of social control,
5
Noradrenergic functioning may also be implicated directly in the health
consequences of risky families. For example, a compelling animal model
of stress-induced behavioral symptoms of depression (Weiss 1991; Weiss,
Bailey, Pohorecky, Korzeniowski, & Grillione, 1980) links exposure to
uncontrollable stress to norepinephrine depletion in the locus coeruleus;
this depletion has, in turn, been linked to depressive symptomatology
including declines in motor activity and appetitive behavior, weight gain,
and decreased responding for rewarding brain stimulation. Whether risky
families give rise to this pattern is unknown at present.
337
RISKY FAMILIES
HPA axis functioning may be disrupted in response to stress,
leading to increased CRH and hypercortisolism. Two retrospective
studies suggested that these biological dysregulations can persist
into adulthood. In one, poor family relationships, assessed by
college students’ ratings of their early family environment, were
associated with elevated cortisol responses to a laboratory chal-
lenge (Luecken, 1998). In the other, women who reported having
been abused in childhood showed greater HPA responses to lab-
oratory stress than did women without such histories; these re-
sponses were greater for abused women who also had symptoms of
anxiety and depression (Heim et al., 2000; see also Kaufman,
Plotsky, Nemeroff, & Charney, 2000).
The child’s attachment to parents also affects cortisol re-
sponses.
6
For example, children with disorganized/disoriented at-
tachment patterns exhibit higher cortisol levels following brief
separations from their mothers (Hertsgaard, Gunnar, Erickson, &
Nachmias, 1995). In other research, attachment patterns moderated
cortisol responses of babies in stressful circumstances (Gunnar,
Brodersen, Nachmias, Buss, & Rigatuso, 1996; Nachmias, Gun-
nar, Mangelsdorf, Parritz, & Buss, 1996). The protective effects of
a secure attachment relationship were especially evident for so-
cially fearful or inhibited children—a finding that parallels previ-
ous research by Levine and Wiener (1988). To explain these
results, Gunnar and her associates argued that temperamentally
fearful children are especially vulnerable to stress, and an insecure
attachment is implicated in their elevated cortisol responses to new
situations, serving both as a marker of the parent–infant relation-
ship and as a potential predictor for later anxiety disorders. This
underscores the interactive role that parenting style may play with
other risk factors, especially genetic risks, in the etiology of
biological and psychological stress responses.
Although evidence is currently lacking on the longer term
stability of child patterns of HPA reactivity, animal data suggest
that early experiences can permanently alter adrenocortical activity
in rat pups in response to stress, responses mediated by maternal
attention. Research by Meaney and associates found that rat pups
handled in early infancy were the recipients of more maternal
licking and grooming; in turn, these rat pups had more hippocam-
pal glucocorticoid receptors, lower hypothalamic corticosteroid
releasing factor, and less glucocorticoid secretion during a stressor
and faster recovery to baseline afterward. They also showed more
open field exploration (an indicator of less anxiety) and more
receptors in the brain for benzodiazepines (anxiety-reducing tran-
quilizers; Liu et al., 1997; Meaney, Aitken, van Berkel, Bhatnagar,
& Sapolsky, 1988; see also Caldi et al., 1998). The handled rats
were also less likely to show age-related onset of HPA dyregula-
tion in response to challenge and less likely to exhibit age-related
cognitive deficits. This compelling animal model suggests that
nurturant stimulation by the mother modulates responses of off-
spring to stress in early life in ways that have permanent effects on
the offspring’s HPA response.
Serotonergic Functioning
Dysregulations in the serotonergic system have been tied to
several of the health outcomes related to risky families, including
depression, suicidality, aggression, and as will be seen, substance
abuse. Research on serotonergic function in children remains rel-
atively sparse, in part because of the invasive nature of assess-
ments of central serotonergic functioning, but the existing evi-
dence does suggest that children from risky families experience
serotonergic dysregulation. Kaufman et al. (1998) identified de-
pressed abused and depressed non-abused children from clinic
records, and compared them with normal children on a serotonin
challenge (L-5-HTP) administered intravenously. Prolactin re-
sponses to the challenge, an indicator of serotonergic functioning,
were highest in the abused children, and those responses were
correlated both with clinical ratings of aggressive behavior and
with family history of suicide attempt
7
(see also Petty, Davis,
Kabel, & Kramer, 1996; Risch, 1997).
The dysregulation in serotonergic functioning associated with
abuse may be due both to genetic and to experiential factors.
Suggestive evidence for a genetic link comes from the fact that
first- and second-degree relatives of depressed abused children
have elevated rates of depression, suicidality, and aggressive be-
havior (Kaufman et al., 1998), and serotonin dysregulation is
highest in people with the greatest familial loading for these
disorders (Coccaro, Silverman, Klar, Horvath, & Siever, 1994;
Halperin et al., 1997; Linnoila, DeJong, & Virkkunen, 1989; Pine,
Shaffer, Davies, & Schonfeld, 1997). However, deficient nurturing
may also contribute to these dysfunctions. For example, Pine and
colleagues (Pine, Coplan, et al., 1997; Pine et al., 1996) found that
children from families characterized by deficient nurturing pro-
duced boys at risk for delinquency who also showed evidence of
serotonergic dysfunction. Primate studies provide the most com-
prehensive data linking nurturant family and maternal environ-
ments to biobehavioral profiles characterized by better serotoner-
gic regulation and less behavioral dysregulation. Such studies have
found that monkeys raised in poor early rearing conditions show
long-term alterations in serotonergic functioning (Kraemer &
Clarke, 1990; Rosenblum et al., 1994). Studies by Suomi (1987,
1991, 1997) also have suggested a role of maternal behavior in
moderating genetic risk for serotonergic dysfunction. One study
compared monkeys with two forms of the 5-HTT allele, a gene
related to serotonin transport; the short form of the gene confers
decreased serotonin function, whereas the long form is associated
with normal serotonin functioning (Suomi, 1999). In addition,
some of the monkeys were raised by their mothers and some were
raised by peers. Being raised by peers is a risk factor for the
development of a reactive, impulsive temperament, and other
deficits in social behavior, including more aggressive exchanges
6
According to attachment theory, children use an adult caregiver as a
secure base for exploration and as a haven of safety in times of stress.
Individual differences in secure base behavior are thought to reflect the
extent to which the attachment figure is a source of security for the infant.
On the basis of their emotional and behavioral responses to brief separa-
tions and reunions with the attachment figure, typically the mother, chil-
dren are classified either as secure or into one of several insecure catego-
ries (Waters, Vaughn, Posada, & Kondo-Ikemura, 1995). Evidence
suggests that maternal sensitivity and other dimensions of parenting be-
havior are significant components of infant attachment security (De Wolff
& van IJzendoorn, 1997).
7
Although lower levels of serotonergic activity are associated with
aggressive behavior in adults, in preadolescent samples, the relation is
often the reverse with especially high serotonin levels correlated with
aggression, depression, and other adverse outcomes (Kaufman et al., 1998).
338
REPETTI, TAYLOR, AND SEEMAN
and less grooming.
8
Monkeys with the short 5-HTT allele who
were raised by peers showed lower concentrations of the primary
central serotonin metabolite (5-HIAA) than was true for monkeys
with the long allele. But for monkeys raised by their mothers,
primary serotonin metabolite concentrations were identical for
monkeys with either allele. This pattern clearly suggests a protec-
tive effect of maternal behavior on expression of genetic risk for
low levels of serotonin.
Additional evidence for the genetic–experiential interaction is
provided by Suomi (1987), who randomly assigned rhesus monkey
neonates selectively bred for differences in temperamental reac-
tivity to foster mothers who were either unusually nurturant or
within the normal range of mothering behavior. Infants whose
pedigrees suggested normative patterns of reactivity exhibited
normal patterns of biobehavioral development, independent of the
relative nurturance of the foster mother. In contrast, highly reactive
infants cross-fostered to normal mothers exhibited deficits in early
exploration and exaggerated behavioral and physiological re-
sponses to minor environmental perturbations. In adulthood, they
tended to drop to and remain low in the dominance hierarchy
(Suomi, 1991). Highly reactive infants cross-fostered to exception-
ally nurturant females, in contrast, appeared to be behaviorally
precocious. They left their mothers earlier, explored the environ-
ment more, and displayed less behavioral disturbance during
weaning than both control (low-reactive) infants reared by either
type of foster mother and highly reactive infants cross-fostered to
normal mothers. In addition, when permanently separated from
their foster mothers and moved into larger groups, the highly
reactive animals cross-fostered to nurturant mothers became adept
at recruiting and retaining other group members as allies, and most
achieved high dominant status.
The primate studies by Suomi (1987, 1991, 1997) are significant
for the present argument, not only because they suggest an impor-
tant role of parenting for modifying the expression of genetically
based temperamental differences, but also because they tie sero-
tonergic dysfunction directly to aggression and other deficits in
social behavior that are similar to behavioral problems found in
children from risky families (see for review, Suomi, 1997). Causal
links between serotonin levels and aggressive behaviors are further
indicated by research showing that lowering levels of serotonin
through pharmacologic treatment with fenfluramine hydrocholo-
ride leads to increases in aggression (Raleigh et al, 1986). Peer-
raised adolescent monkeys also show certain propensities for sub-
stance abuse, for example, requiring larger doses of the anesthetic
ketamine to reach a state of sedation and consistently consuming
more alcohol and developing a greater tolerance for alcohol, rel-
ative to mother-raised monkeys, a pattern predicted by their central
nervous system’s serotonin turnover rates (Higley et al., in press).
Difficulty in moderating aggressive impulses, problems in devel-
oping and maintaining social relationships, and risk for substance
abuse are among the outcomes most consistently seen in response
to risky family characteristics of hostility, conflict, and deficient
nurturing.
The potential intergenerational transmission of these behavior
patterns also warrants attention. Female monkeys who are raised
by peers (rather than by their mothers) are significantly more likely
to exhibit neglectful or abusive treatment of their own offspring
(especially firstborns) as compared with their mother-reared coun-
terparts (Champoux, Byrne, Delizio, & Suomi, 1992; Suomi &
Levine, 1998). Exposure to effective parenting reduces these be-
haviors. In the study of females bred for high reactivity mentioned
earlier, when females became mothers they adopted the maternal
style of their foster mothers, independent of their own reactivity
profile (Suomi, 1987). Highly reactive females raised by especially
nurturant foster mothers not only showed lower reactivity them-
selves, but learned effective parenting, which they passed on to the
next generation, thus providing clear evidence of behavioral inter-
generational transmission of nurturance.
Other System Dysregulations
The interaction of biological systems virtually demands that
other system dysregulations also occur as a result of exposure to a
risky family environment. The HPA axis, for example, influences
the norepinephrine system in ways that may affect anxiety and
depression (Rosen & Schulkin, 1998). Serotonin and dopamine
functioning influence each other (Kostrzewa, Reader, & Des-
carries, 1998), and it is possible that serotonin dysregulation may
be related to a heightened risk for substance abuse via its effect on
dopamine regulation.
9
SAM, PNS, HPA, and serotonergic func-
tioning are all interrelated and, in turn, affect the immune system
(Schleifer, Scott, Stern, & Keller, 1986), the metabolic system
(Raikkonen, Keltikangas-Jarvinen, Hautanen, & Adlercreutz,
1997), and other physiologic regulatory systems of the body. As
such, a risky family situation might produce or exacerbate a
seemingly minor dysregulation in one system, which could, in
turn, produce a small dysregulation in another system, and so on.
Over time, small but accumulating perturbations in the body’s
regulatory systems may lead to multiple vulnerabilities, which may
further help to explain the association of risky family characteris-
tics with the wide range of documented mental and physical health
risks.
Summary
In summary, alterations in SAM, HPA, and serotonergic func-
tioning appear to result from exposure to risky family environ-
8
Monkeys raised with peers have difficulty moderating behavioral re
-
sponses to rough-and-tumble play, sometimes escalating those bouts into
full-blown aggressive exchanges. Rearing with peers is also associated
with other social deficits, such as diminished social competence and low
levels of grooming (Coccaro 1992; Coccaro, Kavoussi, Cooper, & Hauger,
1997; Higley, Mehlman, et al., 1996; Higley, Suomi, & Linnoila, 1996a,
1996b; Mehlman et al., 1994; Raleigh & McGuire, 1994; Virkkunen &
Linnoila, 1993).
9
Dopamine pathways are consistently implicated in substance abuse
including opiates, cocaine, amphetamines, nicotine, and alcohol (Koob,
Sanna, & Bloom, 1998), as well as in neuropsychiatric disorders, such as
bipolar disorder (Depue & Iacono, 1989). In animal studies, early social
isolation or maternal deprivation are associated with alterations in dopa-
mine concentrations in the nucleus accumbens and the caudate-putamen
(Lewis et al., 1990; Kehoe, Shoemaker, Arons, Triano, & Suresh, 1998),
with permanent alterations in dopamine function (Hall, Wilkinson, Humby,
& Robbins, 1999) and receptor sensitivity (Lewis et al., 1990), with
ambivalent responses to novelty, cognitive impairment, and altered re-
sponses to stress (Robbins, Jones, & Wilkinson, 1996), and with stereo-
typed behavior sequences (e.g., Lewis et al., 1990). As yet, however,
specific qualities of parenting have not been directly implicated in this
process.
339
RISKY FAMILIES
ments characterized by conflict, anger, and aggression and by
deficient nurturing. The evidence for disruptions in SAM func-
tioning comes primarily from boys at risk for cardiovascular
disease, suggesting a potential exacerbation of preexisting vulner-
abilities in risky families. Offspring from risky families are also
more likely to show abnormal cortisol reactivity; these patterns
appear especially common in children predisposed to a fearful,
inhibited temperament. Evidence for the mediating role of seroto-
nergic functioning in explaining the association between risky
families and multiple outcomes stems largely from studies of
abused children and from studies on offspring (and especially
nonhuman primates) raised in situations characterized by deficient
nurturing. The significance of this link stems from the associations
between serotonin dysregulation and other adverse outcomes as-
sociated with risky parenting, including aggression, depression,
social incompetence, and substance abuse. In light of the growing
evidence of extensive “cross-talk” among biological regulatory
systems, other system dysregulations are expected to emerge, as
research on these issues progresses.
Emotion Processing
In Figure 1, risky family environments are associated with the
way that offspring process emotions, a factor that may also be
implicated in the development of mental and physical health
disorders. By emotion processing we mean the experience, control,
and expression of emotion, particularly in emotionally arousing
situations. Despite recent advances in emotion research, emotion
processing is still poorly understood, and we do not attempt here
to further clarify or specify such a complex phenomenon. Our
more modest goal is to present research on those components of
emotion processing that have been studied for their association
with risky family environment variables. Three aspects of emotion
processing meet this criterion: emotional reactivity in emotionally
arousing situations, emotion-focused coping, and emotion
understanding.
Risky Family Characteristics and Emotion Processing
Empirical findings supporting a link between risky family en-
vironments and the three aspects of emotion processing mentioned
above are summarized in Table 2. The first group of studies listed
in Table 2 are short-term reaction studies, in which a child’s
immediate emotional reaction in an emotion-arousing situation is
observed. There are several methodological advantages to these
studies: The investigator can control the situation in which the
child is observed, the child’s behavior can be rated by independent
observers, and information about the family environment is sepa-
rated from the laboratory assessment of the child’s emotional
reaction. One drawback to this approach to studying emotion
processing is that it is difficult to distinguish processes involved in
controlling an emotional state from an individual’s propensity to
experience intense emotions that are difficult to regulate. Because
of the different methodologies involved, Table 2 distinguishes
these short-term reaction studies from other studies, discussed
below, that assess emotion understanding and coping.
Conflict and aggression. Most of the short-term reaction stud-
ies in Table 2 focus on children as they listen to or observe angry
and conflictual interactions (which either are staged in laboratory
settings or are naturalistic interactions in the home). The emotional
and behavioral responses of children whose home lives are char-
acterized by conflict and aggression are then compared with the
responses of children from homes with less aggression and happier
marriages. The findings indicate that high levels of conflict at
home sensitize children to anger. They react with greater distress,
anger, anxiety, and fear (Ballard et al., 1993; E. M. Cummings,
Zahn-Waxler, & Radke-Yarrow, 1981; Davies & Cummings,
1998; O’Brien, Margolin, John, & Krueger, 1991). Increased re-
activity may result from chronic stress levels in conflictual and
violent homes. According to the allostatic load model, a period of
recovery following physiological arousal is essential for the proper
functioning of homeostatic processes in the body (McEwen, 1998;
McEwen & Stellar, 1993). Periods of relief or respite from aroused
emotional states may also be critical to the proper regulation of
emotion dynamics. Chronic or repeated stressors in the environ-
ment, such as high levels of violence and family conflict, may not
allow for sufficient recovery from heightened emotional arousal.
Sustained states of emotional arousal may, over time, increase
reactivity. This is consistent with a model proposed by Perry and
his colleagues that suggests that chronic stress affects neurobio-
logical development and creates a sensitized stress-responsive
system that influences arousal, emotion regulation, behavioral
reactivity, and cardiovascular regulation (Perry & Pollard, 1998).
Thompson and Calkins (1996) suggested that hypervigilance in
children from aggressive and violent homes may also contribute to
increased reactivity.
Emotion processing is also assessed by tasks that measure
children’s understanding of emotions and by self-reports of meth-
ods used to cope with stressful experiences in the past. These are
listed as emotion understanding and coping studies in Table 2. As
operationalized in research studies, emotion understanding in-
cludes the ability to recognize emotional states (both in self and
others), the skills to express emotions in a culturally acceptable
manner, and the knowledge of causal antecedents of different
emotions, factors that are integral to the processing of emotions in
stressful or arousing situations. Because emotion understanding
shapes social perception, we only included emotion understanding
studies in Table 2 that used information from independent sources
to assess the family. In two investigations of young children, those
who were maltreated or whose homes were marked by high levels
of anger and distress had a less accurate understanding of emo-
tions, compared with their peers (Camras et al., 1988; Dunn &
Brown, 1994). This may be because families with high levels of
negative affect are less likely to engage in conversations about
feelings (Dunn & Brown, 1994), and more talk about feeling states
in the home is associated with better emotion understanding in
children (Dunn, Brown, Slomkowski, Tesla, & Youngblade,
1991).
Although there are few studies of emotional reactivity and
emotion understanding in adolescents, there is a research literature
on coping in adolescence. We focus here on strategies for control-
ling emotional states in stressful situations, which are often con-
ceptualized as emotion-focused coping strategies. Most of these
investigations are correlational studies that rely on self-report
measures. To limit the impact of self-report biases, we restrict our
analysis to studies in which the assessment of coping was made
years after the description of the family had been provided. In
(text continues on page 345)
340
REPETTI, TAYLOR, AND SEEMAN
Table 2
Risky Family Characteristics and Emotion Processing
Study and design Sample characteristics Family environment measure
Child/adolescent emotion
processing variable Key findings
Short-term reaction studies (immediate emotional reactions and coping responses in stressful situations): Conflict and aggression
Ballard et al., 1993
GC/CS
IND RPRTS
OBS
Children and their parents (n ⫽ 48).
Distressed homes (n ⫽ 21).
Nondistressed homes (n ⫽ 27).
M child age ⫽ 12.5 years
(SD ⫽ 1.4 years).
Child was categorized as living
in a “maritally distressed”
home if either parent
reported poor marital
adjustment or the use of
physical conflict tactics
during marital disputes.
Children videotaped while “overhearing”
a staged angry interaction between 2
adult strangers in the next room.
Nonverbal reactions coded (e.g.,
freezing, postural distress, facial
distress, smiling, laughing). Children
also reported emotional responses
(i.e., by choosing an emotion face that
represented how they felt and
indicating how much they felt that
way).
Early adolescents from maritally
distressed homes reported
more fear and showed greater
nonverbal reactivity.
Camras & Rappaport, 1993
GC/CS
IND RPRTS
OBS
Maltreated & nonmaltreated
children were paired by the
experimenter (n ⫽ 18 pairs). The
children in each pair were from
same daycare or social service
center. Age range: 3.33–7.25
years. M age ⫽ 4 years
11 months.
Maltreated children identified
by state department of child
and family services.
During a play session with a peer, child
attempts to negotiate sharing of a
gerbil box (e.g., attempts to obtain
box and resistance responses) were
coded.
Maltreated children appeared
reluctant to either engage or
resist the peer. They smiled
more (signs of placation) and
used fewer negative upper
face expressions, such as
“brow frowns” (signs of
determination), when resisting
a peer’s attempts to obtain the
box. They made fewer
attempts to obtain the box
and, after encountering
resistance, waited longer
before renewing an attempt,
and they used “pulls” less
often to obtain the box.
J. S. Cummings et al., 1989
CORR/CS
IND RPRTS
OBS
Children & their mothers (n ⫽ 48).
Child age range: 2–6 years.
Physical aggression in
parents’ marriage (average
of both parents’ reports on
the Conflict Tactics Scale)
Child’s response to lab simulation in
which the experimenter expressed
anger toward the mother.
Physical conflict in the parents’
marriage correlated with
reactions suggesting greater
social responsibility (e.g.,
children were more solicitous,
took actions to comfort and
protect their mothers).
Social responsibility was indicated by
responses to mother, such as
provisions of physical and/or verbal
comfort, by expressions of concern
(such as asking about feelings), and
by defense of mother.
(table continues)
341
RISKY FAMILIES
Study and design Sample characteristics Family environment measure
Child/adolescent emotion
processing variable Key findings
Short-term reaction studies (immediate emotional reactions and coping responses in stressful situations): Conflict and aggression (continued)
E. M. Cummings et al., 1981
Cross-sequential, daily monitoring
Children from intact, middle-class
families (n ⫽ 24). Child age
range: 10–20 months, at start of
study.
Total number of incidents of
interparent fighting, based
on mothers’ daily reports
over a 9-month period.
Maternal description of child emotional
responses to each episode of
interparent fighting as it occurred.
More interparent fighting
associated with greater
sensitization to individual
episodes of fighting (more
anger, distress, and
affectionate/prosocial
behavior; fewer instances of
an unemotional reaction).
Davies & Cummings, 1998
CORR/CS
IND RPRTS
OBS
Children & mothers from maritally
intact families (n ⫽ 56). Child
age range: 6–9 years. M age
⫽ 7.5 years.
Latent variable based on
maternal report of “marital
discord,” which included
measures of disagreements,
adjustment, hostility, and
conflict in the marriage.
Child emotional reactions to a
simulated conflict between mother and
the experimenter, measured by
observer ratings of child distress and
vigilance (watchful attention and/or
preoccupation with the possibility of
danger), and by self-reported feelings
of anger.
More marital discord associated
with more emotional
reactivity (signs of vigilance
and distress, and self-reported
anger).
Gordis et al., 1997
CORR/CS
IND RPRTS
OBS
Family triads (n ⫽ 90). Child age
range: 9–13 years. M age ⫽ 11.3
years.
Physical marital aggression:
both spouses’ reports of
frequency of violence during
the past year.
Child responses during a family
discussion (about aspects of the
child’s behavior that act as a source
of conflict between the parents). Child
responses coded for amount of
withdrawal, anxiety, and signs of
distraction.
Physical marital aggression
correlated with more attempts
to distract during the
discussion (silly/irrelevant
behavior that draws attention
away from the discussion).
O’Brien et al., 1991
GC/CS
IND RPRTS
Mother–son dyads (n ⫽ 35). Child
age range: 8–11 years.
Dyads classified into groups
based on the amount of
marital conflict reported by
mother and son: (a)
physically aggressive
marriages (n ⫽ 12), (b)
verbally aggressive
marriages (n ⫽ 11), (c) low-
conflict marriages (n ⫽ 12).
Sons’ reports of their thoughts and
feelings as they listened to audiotaped
(simulated) family conflicts.
Self-distraction—unique and
tangential comments (reflecting an
inability to deal directly with the
conflict situation); interference—
reports that they would attempt to
verbally or physically intervene in the
conflict; arousal—reports of increased
heart beat, breathing faster, feeling
sweaty.
Sons of verbally or physically
aggressive parents were more
likely to self-distract. Sons of
physically (as opposed to
verbally) aggressive parents
were more likely to self-
distract, interfere, and report
physiological arousal.
342
REPETTI, TAYLOR, AND SEEMAN
Table 2 (continued)
Study and design Sample characteristics Family environment measure
Child/adolescent emotion
processing variable Key findings
Short-term reaction studies: Cold, unsupportive, and neglectful homes
Nachmias et al., 1996
GC/CS
OBS
Mother–toddler (all White) dyads (n
⫽ 77). Child age ⫽ 18 months
(⫹/⫺ 2 weeks).
Mother–child attachment
security assessed in the
Strange Situation (securely
attached vs. insecurely
attached toddlers).
Child coping with novel arousing
stimuli, such as a live boisterous
clown who invited the child to play
while the mother was present. More
competent coping was indicated by
child efforts to use mother to become
engaged in the situation (such as
through social referencing behaviors),
fewer attempts to escape the situation,
fewer infantile self-soothing behaviors
(such as stroking or sucking on a toy).
Less securely attached toddlers
were less competent copers.
They were less likely to use
their mothers to become
engaged and more likely to
attempt to escape the
situation.
Emotion understanding and coping studies: Conflict and aggression
Camras et al., 1988
GC/CS
IND RPRTS
Abused (n ⫽ 20) & nonabused (n
⫽ 20) child–mother dyads. Age
range: 3.33–7.25 years. M age
⫽ 4 years, 11 months.
Child abuse status identified
by state agency.
Facial emotion expression posing task
(posing of six emotions). Expression
recognition task (several weeks after
the production task): Child chooses
expressions to match emotions
described in stories.
Abused children were less
accurate in recognizing
emotional expressions, and
posed less recognizable
expressions. Mothers’ posing
scores correlated with their
children’s recognition scores;
mothers of abused children
posed less recognizable
expressions.
Dunn & Brown, 1994
CORR/FLW-UP
IND RPRTS
OBS
Mother–child dyads (n ⫽ 50). Child
age: 33 months at Time 1; 40
months at Time 2.
Time 1: coding of negative
affect (anger and distress)
during family observations
at home on two separate
occasions.
Time 2 (7 months later): two emotion
understanding tasks, identifying
emotions in facial expressions and
identifying emotional response in
puppet vignettes.
More expressions of maternal
negative affect at home was
associated with less emotion
understanding.
Johnson & Pandina, 1991
CORR/FLW-UP
SELF
Adolescents (n ⫽ 1,308). Ages: 12,
15, 18 years at Time 1.
Time 1: adolescents’
descriptions of parent
hostility.
Time 2 (3 years later): adolescents’
reports of how they generally cope
with problems. Dysfunctional methods
of coping included use of alcohol and
drugs and emotional outbursts.
Maternal and paternal hostility
was associated with more
dysfunctional coping among
sons (a pattern not
consistently observed in the
girls’ data, except for cross-
sectional analyses).
(table continues)
343
RISKY FAMILIES
Table 2 (continued)
Study and design Sample characteristics Family environment measure
Child/adolescent emotion
processing variable Key findings
Emotion understanding and coping studies: Conflict and aggression (continued)
Valentiner et al., 1994
CORR/FLW-UP
SELF
College students (n ⫽ 175; 124
women, and 51 men). M age at
Time 2 ⫽ 20 years (SD ⫽ 0.4).
Time 1 (fall of freshman year):
composite measure based on
students’ descriptions of
amount of marital conflict
at home and the general
emotional quality of their
relationships with their
parents (e.g., supportiveness
of each parent).
Time 2 (2 years later): self-reported
coping with the most important
problem experienced in the past year.
Coping was scored as the percentage
of all coping responses that
represented “approach” coping (i.e.,
positive reappraisal of the problem
and problem solving attempts).
In LISREL models, there was
an association between a
more negative family
environment and less frequent
use of approach coping
strategies (with controllable
problems). The association
between a negative family
environment and
psychological distress was
mediated by less approach
coping.
Emtion understanding and coping studies: Cold, unsupportive, and neglectful homes
Hardy et al., 1993
CORR/CS
IND RPRTS
Children and their mothers (n ⫽
60). Age range: 9–10 years. M
age ⫽ 10.2 years.
Family support rating based
on maternal descriptions of
family cohesion, her own
nurturance and
responsiveness to child
input, her monitoring or
awareness of her child’s life,
and family adaptability.
Interview-based child descriptions of
responses to, and attempts to cope
with, stressful situations that had been
identified by the mother as having
occurred during the past 2 months.
Avoidant strategies: removing self
from the situation, giving up trying,
use of distraction to avoid thinking
about the situation.
A less supportive family
environment associated with
fewer different coping
strategies and with less use of
avoidant coping strategies in
uncontrollable situations.
Laible & Thompson, 1998
CORR/CS
IND RPRTS
Young children and their mothers
(95% White; n ⫽ 40). Age range:
2.5–6 years. M age ⫽ 50 months.
Security of child’s
attachment assessed by Q-
sort in which mother
described her child’s current
behavior with primary
caregivers.
Emotion understanding composite:
recognition of emotions corresponding
to depicted facial expressions;
identification of emotions portrayed
by a puppet in a vignette; naturalistic
assessment of appraisals of emotion
events as they occurred at school
(identification of emotion expressed
by another child and description of its
causal antecedents).
Children with less secure
attachments to their primary
caregivers demonstrated less
understanding of negatively
valenced emotions, such as
sadness, anger, and fear.
van den Boom, 1994
EXP/FLW-UP
OBS
Mother–infant pairs (n ⫽ 100). All
infants were rated as “irritable.”
Child age, start of treatment: 6
months; end of treatment: 9
months; follow up: 12 months.
Time 1: A parenting skills
training program was
provided during home visits
to half of the pairs over a
3-month period. The
intervention succeeded in
enhancing maternal
responsiveness, stimulation,
visual attentiveness, and
control.
Time 2: postintervention observation of
infant behavior during home visits
(mother–infant interaction in a
family context and in a free play
situation to assess quality of infant
exploration) and a 12-month follow-
up assessment of attachment security.
Infants whose mothers received
the intervention were more
sociable, better able to soothe
themselves, and engaged in
more cognitively sophisticated
exploration (at end of
treatment). At follow-up,
more of the intervention
infants were securely
attached.
Note. The key family and outcome variables in each study are highlighted in bold type. For each study, design (GC ⫽ group comparison; CORR ⫽ correlational; EXP ⫽ experimental), timing of
assessments (CS ⫽ cross-sectional; FLW-UP ⫽ multiple data collection points, but change over time in the outcome is not assessed), and sources of data for primary variables (IND RPRTS ⫽
assessment of predictor and outcome variables based on data from separate sources; OBS ⫽ observational data; SELF ⫽ all self-report data) are reported.
344
REPETTI, TAYLOR, AND SEEMAN
Table 2 (continued)
research summarized in Table 2, relating measures of conflict and
hostility at home to adolescents’ descriptions (provided 2–3 years
later) of how they coped with different kinds of problems and
stressors, the strategies favored by teens from the risky families
emphasized a desire to reduce tension and escape the situation (V.
Johnson & Pandina, 1991; Valentiner, Holahan, & Moos, 1994).
This pattern was also found in the short-term reaction studies
reported in Table 2; in the studies of preteens and teens, those from
high-conflict homes tried to distract their own and others’ attention
away from interpersonal conflict (Gordis, Margolin, & John, 1997;
O’Brien et al., 1991). It is interesting that in the short-term
studies of younger children, those from high-conflict homes
sometimes engaged in solicitous or placating behavior (Camras
& Rappaport, 1993; E. M. Cummings et al., 1981; J. S. Cum-
mings, Pellegrini, Notarius, & Cummings, 1989). It is possible
that after repeatedly failing to change stressful events in the
family, perhaps through behaviors such as appeasement and
placation, children growing up in angry and aggressive homes
gradually abandon efforts to control difficult situations and
focus, instead, on simply trying to escape and recover from
heightened emotional arousal. In short, the legacy of growing
up with high levels of overt anger and aggression at home may
be not only a stronger emotional reaction in situations that
involve conflict, but also a particular set of behaviors for
responding in those situations.
Cold, unsupportive, and neglectful homes. Table 2 also sum-
marizes research findings relating deficient nurturing to emotion
processing, effects that have been observed very early in develop-
ment. Babies begin to regulate their emotional responses soon after
birth, engaging in behaviors such as sucking to soothe themselves
(Campos, 1988). Parental nurturing appears to facilitate the devel-
opment of these primitive coping behaviors. For example, in an
experimental study listed in Table 2, an intervention for mothers of
irritable newborns that improved maternal responsiveness, atten-
tiveness, and control also resulted in an increase in infant self-
soothing behaviors. The irritable infants whose mothers did not
receive the intervention actually showed a slight decrease in self-
soothing from 6 to 9 months, and they were judged as having
less secure attachments to their mothers (van den Boom, 1994).
A short-term reaction study of toddlers (Nachmias et al., 1996)
and two coping studies also indicate that an insecure parent–
child attachment or little cohesion and support in the family
are associated with less adaptive coping across a wide age
range (Hardy, Power, & Jaedicke, 1993), and with deficits in
emotion understanding among preschoolers (Laible & Thompson,
1998).
Summary. The findings summarized in Table 2 indicate that
growing up in a risky family environment interferes with the
development of means for processing emotions. In particular, the
data point to high emotional reactivity, deficits in emotion under-
standing, and a reliance on unsophisticated coping responses to
stressful situations. Across studies using different methodologies
and age groups, findings indicate that children living in risky
family environments are more likely than their peers to focus on
tension reduction, distraction, and escape in stressful situations, a
relation that is also found in cross-sectional studies (e.g., Stern &
Zevon, 1990).
Emotion Processing and Mental and Physical
Health Outcomes
Figure 1 depicts emotion processing as a link from risky family
characteristics to adverse mental and physical health outcomes.
Poor regulation of emotions is implicated in more than one half of
the Diagnostic and Statistical Manual of Mental Disorders (4th
ed.) Axis I and in almost all of the Axis II psychiatric disorders
(American Psychiatric Association, 1994). A small but growing
research literature has tied indicators of emotion regulation to both
internalizing and externalizing symptoms in children and adoles-
cents (Eisenberg, Fabes, & Murphy, 1996; Southam-Gerow &
Kendall, 2002; Zahn-Waxler, Iannotti, Cummings, & Denham,
1990). In addition to acting as a mediator of the link between
negative family environments and mental health (Valentiner et al.,
1994), emotion processing may also moderate a child’s vulnera-
bility to risky family characteristics. In one study, for example, the
link between parents’ marital hostility and subsequent externaliz-
ing behavior in offspring was observed only among children with
inadequate emotion regulation (Katz & Gottman, 1995).
Emotion processing is also implicated in physical health, first,
because of interrelations between the regulation of emotional and
physiological responses to stress. For example, children who are
emotionally reactive in certain situations (such as angry social
interactions) are also more likely to be physiologically reactive
(El-Sheikh et al., 1989). High vagal tone, which is based on an
index of heart rate and heart rate variability and is a marker of
parasympathetic functioning, indicates both better homeostatic
capacity and better regulation of emotional responses to daily
stressors (Fabes & Eisenberg, 1997). In one study, high vagal tone
in 4 to 5 year olds predicted more effective emotion regulation 3
years later, and was cross-sectionally correlated with more parent
“emotion coaching” (i.e., discussions of effective ways to cope
with anger or distress; Gottman et al., 1996). In another study, high
vagal tone buffered the impact of marital conflict on physical
health; there was a weaker association between conflict in the
home and health problems among the children with high vagal
tone (El-Sheikh, Harger, & Whitson, 2001).
One emotion in particular, anger, appears to play a significant
role in the development of coronary artery disease and hyperten-
sion, at least among some individuals (e.g., Dembroski, MacDou-
gall, Williams, Haney, & Blumenthal, 1985; Jorgensen, Johnson,
Kolodziej, & Schreer, 1996; Julkunen, Salonen, Kaplan, Chesney,
& Salonen, 1994; Smith, 1992). Emotion processing may also be
indirectly implicated in the onset and course of certain diseases
through its link with psychopathology, particularly with respect to
mental health problems that involve chronic or recurrent negative
emotional states. Depression and anxiety appear to play a signif-
icant role in numerous health risks, including all-cause mortality
(L. R. Martin et al., 1995). Epidemiological, psychological, and
experimental evidence point to a clear dose–response relation of
anxiety and coronary heart disease (Kubzansky, Kawachi, Weiss,
& Sparrow, 1998). Major depression, depressive symptoms, his-
tory of depression, and anxiety have all been identified as predic-
tors of cardiac events (Frasure-Smith, Lesperance, & Talajic,
1995), and depression is a risk factor for mortality following a
myocardial infarction, independent of cardiac disease severity
(Frasure-Smith et al., 1995). State depression and clinical depres-
345
RISKY FAMILIES
sion also have been related to sustained suppressed immunity
(Herbert & Cohen, 1993).
Emotions are pivotal in our model. As depicted in Figure 1,
disruptions in emotion processing occur early in the cascade and
are directly linked to each of the other factors that mediate the
effects of risky families.
Social Competence
As Figure 1 shows, emotion processing ultimately blends into
social competence, that is, how skilled children are at managing
the often frustrating and challenging experiences they have with
family and peers. For example, in order to negotiate difficult social
interactions, such as peer conflicts, children must learn how to
respond in a socially appropriate manner while feeling frustrated
and angry. The importance of emotion regulation for children’s
social functioning has been extensively studied, and research con-
sistently shows that emotionally intense children who are poor
regulators of their emotions are liked less by their classmates and
viewed as less socially competent by observers (e.g., Cassidy,
Parke, Butkovsky, & Braungard, 1992; Eisenberg et al., 1993;
Eisenberg et al., 1997; Gottman et al., 1996; Krevans & Gibbs,
1996). Popular and socially competent children are better able to
control their angry and excited emotions in arousing situations, and
they tend to display less overt negative emotion than other children
(Hubbard & Cole, 1994).
Risky Family Characteristics and the Development of
Social Competence
Table 3 summarizes findings from investigations that have
related risky family characteristics to indicators of social compe-
tence, in particular the quality of social behavior and relationships
outside of the home. The child participants in these studies ranged
in age from infants to adolescents. Two long-term studies, in
which adults were recontacted 20–30 years after an initial assess-
ment when they were in their 20s, are also included (Graves,
Wang, Mead, Johnson, & Klag, 1998; Klohnen & Bera, 1998).
Each of the 16 studies cited in Table 3 has one or more of the
following characteristics: a longitudinal or follow-up research de-
sign (7 studies) and information obtained from independent
sources or observational data, or both (13 studies), ensuring that
correlations between the family environment and social compe-
tence were not inflated by individual respondent bias.
Conflict and aggression. The first group of studies in Table 3
addresses the social behavior and social standing of children living
in homes with high levels of conflict and aggression. Those living
with hostile and aggressive parents had fewer of the positive skills
that facilitate successful interactions with peers (Crockenberg &
Lourie, 1996; Pettit, Dodge, & Brown, 1988) or were more likely
to behave in an aggressive or antisocial manner (C. H. Hart,
Nelson, Robinson, Olsen, & McNeilly-Choque, 1998; Schwartz,
Dodge, Pettit, & Bates, 1997). Other studies found that sons from
aggressive families were more likely to be rejected and victimized
by peers (Dishion, 1990; Schwartz et al., 1997), and women who
had grown up in troubled and conflictual homes had more avoidant
attitudes and feelings about closeness and intimacy (Klohnen &
Bera, 1998).
Cold, unsupportive, and neglectful homes. The second group
of studies in Table 3 is part of a growing literature pointing to the
negative effects that a lack of warmth and nurturance can have on
the ability to form and maintain social relationships. Most of these
studies have examined links between quality of the mother–child
bond (focusing in particular on attachment security) and children’s
relationships with peers. The results indicate that children whose
parents were less responsive, warm, and sensitive were less likely
to initiate social interactions and were more aggressive and critical
(Brody & Flor, 1998; C. H. Hart et al., 1998; Kerns, Klepac, &
Cole, 1996; Landry, Smith, Miller-Loncar, & Swank, 1998). In
addition, when parents were cold, unsupportive, or neglectful, their
offspring’s social relationships throughout life were more prob-
lematic and less supportive (Booth, Rose-Krasnor, McKinnon, &
Rubin, 1994; Bost, Vaughn, Washington, Cielinski, & Bradbard,
1998; Graves et al., 1998; Kerns et al., 1996; Larose & Boivin,
1998; MacKinnon-Lewis, Starnes, Volling, & Johnson, 1997).
Overall, the findings summarized in Table 3 suggest that the
development of social competence and supportive relationships
outside of the family are compromised by growing up in a risky
family environment. Next, we argue that risky families have this
effect because they shape the way that offspring come to think
about and behave in relationships.
Social skills. There are several ways that risky families can
hinder the early acquisition of social skills for initiating and
maintaining friendships and for managing difficult interpersonal
situations, such as those involving conflict and anger. First, young
children model the social behavior that they observe in the family.
Empirical evidence points to a close correspondence between
social skills observed in the family and a child’s behavior when
interacting with peers. Children growing up in families in which
complex social skills are rarely demonstrated (e.g., sensitivity to
the child’s feelings or needs) demonstrate fewer conflict manage-
ment skills and are less sensitive and responsive with peers (Her-
rera & Dunn, 1997; Lindsey, Mize, & Pettit, 1997; Putallaz, 1987).
Similarly, children who are the recipients of anger, aggression, and
hostility from siblings and parents are, in turn, described by their
teachers as less socially competent and more aggressive (Carson &
Parke, 1996; Stormshak et al., 1996). In addition to acting as role
models, parents engage in active efforts to shape their children’s
relationships and social skills through discussions of social prob-
lems and advice giving (Laird, Pettit, Mize, Brown, & Lindsey,
1994). Mothers who suggest fewer constructive techniques to
solve social problems have children who themselves have fewer
social skills, engage in more aggressive and less prosocial behav-
ior, and are less likely to generate prosocial solutions to problems
(Eisenberg, Fabes, & Murphy, 1996; Mize & Pettit, 1997; Pettit et
al., 1988).
Social cognition. Social relationships in adolescence and
adulthood are also shaped by aspects of social cognition first
developed in childhood. On the basis of the same risky family
experiences that shape social skills, children may develop and
store in memory “social algorithms,”“relationship schemas,” or
“working models” of self and others in close relationships that are
activated and applied in new situations throughout life (Andersen
& Berk, 1998; Bugental, 2000). For example, evidence suggests
that growing up in a violent household shapes the development of
the basic cognitive structures that guide social behavior and rela-
(text continues on page 351)
346
REPETTI, TAYLOR, AND SEEMAN
Table 3
Risky Family Characteristics and Social Competence
Study and design Sample characteristics Family environment measure
Child/adolescent social
competence variable Key finding
Conflict and aggression
Crockenberg & Lourie, 1996
CORR/FLW-UP
IND RPRTS
OBS
Mothers and their children (n ⫽ 42).
Child age: 2 years at Time 1, 6
years at Time 2.
Time 1, maternal negative control
strategies (based on lab and
home observations): frequency
with which mother used negative,
highly power-assertive, child-
control strategies, including
maternal expressions of anger,
annoyance, and criticism, and use
of threats, punishment, and
physical control.
Time 2 (4 years later): children described
peer conflict-resolution strategies in
actual and hypothetical situations.
Unskilled avoidance of conflict: child
gives up own goals and does nothing
overt to resolve the problem; does not
seek help of an authority figure.
Manipulative behavior: covert power
assertion, such as “finagle,” and
manipulation of friendship (e.g., tells
friends not to play with the other
child).
Maternal negative control strategies
predicted less competent
conflict-resolution strategies:
Daughters described more
unskilled avoidance, sons
described more manipulative
strategies.
Dishion, 1990
CORR/CS
IND RPRTS
OBS
Boys and their families (99% White,
mostly working class and lower
SES, 32% of parents unemployed,
35% single-parent homes; n ⫽
206).
Negative and aggressive parenting
composite: (based on home
observations) nattering—
noncontingent aversiveness with
the child, parent negative verbal
(e.g., commands) and physical
behavior regardless of the child’s
behavior; punishment density—
relative frequency of parent-to-
child verbal and physical
aggressive behavior in relation to
positive parent–child interaction;
observer impressions that parents
were ineffective, unfair, and
inconsistent in discipline
practices.
Sociometric assessment, based on
classmates’ nominations of child (a) as
“a friend you like,” and someone “who
makes friends easily” (peer
acceptance) and, (b) as someone who
doesn’t “get along with most other
kids,” and “you don’t want to be
friends with” (peer rejection).
More negative and aggressive
parenting (particularly discipline
that was unfair and inconsistent)
associated with more peer
rejection and less peer
acceptance, an effect mediated
by aggressive and antisocial
behavior.
Aggressive and antisocial behavior
composite (argues, disobedient, lies,
steals, etc. across multiple settings)
based on parent questionnaire and 6
daily telephone interviews, teacher
questionnaire, and child self-report
(structured interview, and daily
telephone interviews).
Hart et al., 1998
CORR/CS
IND RPRTS
Parent–child dyads (children
attending nursery school in
Russia; n ⫽ 207). Child age
range: 3.5–6.5 years. M age ⫽ 5.1
years.
Maternal coercion: self-reported
slapping, grabbing, yelling; use
of physical punishment and
shouting as discipline. Paternal
responsiveness: self-reported
patience, being easy-going and
relaxed, joking and playing,
being responsive to child
feelings/needs, giving comfort
and understanding when child is
upset.
Teacher ratings of relational aggression:
causing harm through damage to
relationships, such as social exclusion;
overt aggression: causing harm
through damage to physical or
psychological well-being, such as
bullying, threatening, and use of
physical force.
Coercive parenting tactics
associated with more relational
and overt aggression (with
controls for other aspects of
parenting, child sex and age,
parent education). Sons (but not
daughters) who observed more
marital conflict were more
aggressive (both overt and
relational aggression).
Marital conflict: mother’s and
father’s reports of overt conflict
observed by the child.
(table continues)
347
RISKY FAMILIES
Study and design Sample characteristics Family environment measure
Child/adolescent social
competence variable Key finding
Conflict and aggression (continued)
Klohnen & Bera, 1998
GC/FLW-UP
SELF
1958 and 1960 college seniors (all
women; n ⫽ 85). Time 1: senior
year in college (age 21), Time 2:
age 52.
Time 1: retrospective description of
“troubled” home life during
childhood and adolescence (e.g.,
whether there was open conflict,
a lack of family closeness, and
whether the participant had
thought about running away from
home).
Time 2 (32 years later): attachment style
measure—attitudes and feelings about
closeness, intimacy, and
interdependence in relationships.
Responses used to categorize midlife
women as having secure versus
avoidant attachment styles.
Women with secure versus
avoidant attachment styles in
midlife were compared. An
avoidant attachment style in
midlife was predicted by a more
troubled and conflictual home in
childhood.
Pettit et al., 1988
CORR/CS
IND RPRTS
Preschoolers and their mothers (most
Caucasian, lower SES, single
parents; n ⫽ 46). Child age range:
4–5 years.
Maternal hostile/aggressive
attitudes: endorsement of
aggressive solutions to interpersonal
problems. Hostile attributions:
interpretation of negative child
behavior as having a hostile intent
(as assessed by stories describing
an ambiguous provocation
directed toward the mother).
Deficient social problem solving: child
responses, to hypothetical social
situations, that suggest a tendency to
generate fewer solutions (especially
effective, prosocial solutions) to social
problems, and a propensity to respond
with aggression. Social competence
composite: teacher ratings and classmates’
nominations of children who are
cooperative and “nice to play with.”
Mothers’ hostile and aggressive
attitudes associated with less
social competence, an effect
mediated by deficient social
problem solving.
Schwartz et al., 1997
GC/FLW-UP
IND RPRTS
OBS
Boys and their mothers (majority
from lower to middle SES
backgrounds; n ⫽ 198). Time 1:
summer before kindergarten. Time
2: 3rd grade (M age ⫽ 8 years) or
4th grade (M age ⫽ 9 years).
Time 1, maternal hostility:
observers’ ratings, during a home
visit, of physical and verbal
hostility and annoyance directed
toward the child. Derived from
interview with mother: (a)
maternal restrictive discipline—
use of severe, strict, physical
discipline with child, (b) parental
aggression—own and partner’s use
of overtly aggressive strategies
with child during conflict, (c)
marital aggression—own and
partner’s use of overtly aggressive
strategies toward each other during
conflicts, and (d) physical harm—
interviewer rating of “probable” or
“definite” harm to child.
Time 2 (4–5 years later): victim/
aggressor status assessed by
sociometric data from classmates.
Victim score: sum of nominations
received for three items (“gets picked
on,”“gets teased,”“gets hit or
pushed”). Aggressor score: sum of
nominations received for three items
(“starts fights,”“says mean things,”
“gets mad easily”). Aggressive
victims: boys whose victim and
aggressor scores were both well above
the mean (n ⫽ 16).
Boys who were “aggressive
victims” in 3rd/4th grade were
compared with all other boys.
Being both a victim and
aggressor was predicted by a
more hostile and aggressive
family life.
Cold, unsupportive, and neglectful parenting
Booth et al., 1994
CORR/LONG:
OBS
Children and their mothers (87%
White; n ⫽ 79). Child mean age
Time 1: 4.3 (⫹/⫺ .1) years,
Time 2: 8.0 (⫹/⫺ .15) years.
Time 1, attachment security:
coding of child’s behavior upon
reunion with the mother
following a brief
(approximately 15 min)
separation.
Social engagement, Time 1: child behavior
during a play session with an unfamiliar
same-sex peer, coded for proportion of
positive or neutral conversation and play.
Time 2 (4 years later): composite based
on child behavior during a play session
with three unfamiliar same-sex peers,
coded for proportion of positive or neutral
conversation and play, and likeability
ratings made by the three other children.
A less secure maternal attachment
predicted a decline in positive
engagement during play.
348
REPETTI, TAYLOR, AND SEEMAN
Table 3 (continued)
Study and design Sample characteristics Family environment measure
Child/adolescent social
competence variable Key finding
Cold, unsupportive, and neglectful parenting (continued)
Bost et al., 1998
CORR/CS
OBS
Mother–child dyads (n ⫽ 69). Child
age range: 3–4 years.
Attachment security: Q-sort rating
made by two observers, after a
home visit, on the basis of child
behavior with a specific
caregiver. Rating based on
child’s use of caregiver as secure
base for exploration and as a
haven when threatened or
otherwise distressed.
Social competence composite: observers’
Q-sort ratings of child classroom
behavior; frequency of attention from
classmates; sociometric ratings from
classmates (nomination and social
preference). Social network
composite: child report of satisfaction
with social support and the number of
individuals available for recreation,
emotional support, and child care tasks.
Less attachment security associated
with less social competence, an
effect mediated by a less
satisfying and smaller social
network.
Brody & Flor, 1998
CORR/CS
OBS
Mother–child dyads from mostly
poor, rural, Black, single-mother
households (n ⫽ 156). Child age
range: 6–9 years.
Observers’ ratings of “harmony”
in the mother–child
relationship (warm, supportive,
relaxed, positive affect) based on
mother–child interaction during
three tasks (game playing, story
telling, model building).
Social competence composite: teacher
ratings of child social competence,
behavioral conduct problems, anti-
social behavior. Child self-regulation
composite: mother and teacher ratings
of child self-control (plans ahead,
persists, focuses attention).
Less mother–child harmony
correlated with less social
competence; an effect mediated
by deficient self-regulation.
Graves et al., 1998
CORR/FLW-UP
SELF
Time 1: students (93% male)
enrolled in medical school (1948–
1964; n ⫽ 589). M age ⫽ 22
years. Time 2: M age ⫽ 56 years.
Time 1, emotional closeness to
parents: self-report adjective
checklist measure of emotions
characterizing mother–child and
father–child relationships
(warmth, understanding,
detached, etc.) assessed while
respondents were in medical
school (1948–1964).
Time 2 (over 20 years later). Social
connectedness: self-reported number of
close friends and family members that
respondent feels “at ease with, can talk
to about private matters, can call for
help, or would lend you money,” as
well as the number of people for whom
respondent provides support functions
like these.
Less closeness to parents predicted
less social connectedness (with
controls for respondent age, race,
and parent education).
Hart et al., 1998
CORR/CS
IND RPRTS
See above. See above. See above. Less paternal responsiveness
associated with more relational
and overt aggressive child
behavior (with controls for other
aspects of parenting, child sex
and age, parent education).
Kerns et al., 1996
Study 1
CORR/CS
IND REPORTS
Study 2
GC/CS
OBS
Study 1: 5th graders (n ⫽ 76);
Study 2: best friend dyads (5th and
6th graders; n ⫽ 55).
Child–mother relationship
security: child report of maternal
responsiveness, availability,
reliability, interest in
communication with child (secure
attachments ⫽ top 2/3 of the
distribution of scores).
Study 1, sociometric ratings:
classmates’ ratings of how much they
like to play with the child, and of
close friendships. Self-reported
feelings of loneliness.
Study 1: Less secure maternal
attachments associated with being
less well liked by classmates,
fewer reciprocated close
friendships, and feeling lonely.
Study 2: Videotaped discussion between
best friends coded for criticisms voiced
(i.e., negative evaluations of another,
hostile comments, complaints) and
responsiveness (degree to which they
attended to, acknowledged, and
responded to each other’s social cues).
Best friends also rated the degree of
companionship in their relationship.
Study 2: Compared with best friend
dyads in which both children had
secure maternal attachments,
dyads in which one child had an
insecure attachment were more
critical and less responsive to
each other’s social cues and felt
less companionship in their
relationship.
(table continues)
349
RISKY FAMILIES
Table 3 (continued)
Study and design Sample characteristics Family environment measure
Child/adolescent social
competence variable Key finding
Cold, unsupportive, and neglectful parenting (continued)
Landry et al., 1998
CORR/LONG
OBS
Infants and their mothers (n ⫽ 299).
Ages: 6 months, 12 months, 24
months, 40 months.
Warm sensitivity composite:
mother’s sensitivity to her child’s
cues (promptness and
appropriateness of reactions,
acceptance of the child’s
interests, amount of physical
affection, positive affect, and
tone of voice).
Infant social skills: growth, over 34
months, in social behaviors (gestures,
positive affect, eye gaze, vocalizations/
words) observed while at home with
mother during daily activities.
A lack of maternal sensitivity to
infant social cues predicted
slower growth in social skills,
particularly with respect to
initiations of social interactions
and responsiveness.Initiations of social interaction: social
behaviors when the mother had not
interacted with the infant for at least 3 s.
Responsiveness: social behaviors that
followed within3sofamaternal
attention-directing event. Plus, orienting
to the focus of mother’s attention
directing behavior (coded at 6 and 12
months) and compliance to maternal
requests (coded at 24 and 40 months).
Larose & Boivin, 1998
CORR/LONG:
SELF
Adolescents (n ⫽ 285). Age range:
15–20 years. M age: 17 years (at
both data collection points).
Time 1 (end of high school),
security in the parent-
adolescent relationship: degree
of mutual trust, quality of
communication, prevalence of
anger and alienation with mothers
and fathers.
Time 2 (middle of the first semester at
college): expectations of social
support to cope with worrisome events
related to late adolescence and to
college transition; feelings of
loneliness experienced in interpersonal
relationships over the last month.
A less secure parent–adolescent
relationship predicted a decline
in expectations of social support
and greater feelings of loneliness
during the transition to college.
MacKinnon-Lewis et al., 1997
CORR/CS (all variables
assessed over the course
of a single year).
OBS
IND REPORTS
Triads: 3rd–5th grade boys, their
siblings, and their mothers (n ⫽
71). Target child age range: 8–10
years. M ⫽ 9.2 years. Sibling age
range: 4–14 years. M ⫽ 8.75
years.
Maternal rejection composite:
target child report, sibling report,
and mother’s observed negativity
with both children during a play
session (board game).
Sibling aggression: negative behavior by
target child and sibling during a 15 min
play session (board game), includes
verbal and physical behavior and
expression of affect. Aggression with
peers: teacher and classmate ratings.
Social acceptance: sociometric
nominations by classmates.
Maternal rejection was associated
with less social acceptance, an
effect mediated by aggressive
behavior (with sibling and
peers).
Note. The key family and outcome variables in each study are highlighted in bold type. For each study, design (CORR ⫽ correlational; GC ⫽ group comparison), timing of assessments (FLW-UP ⫽
multiple data collection points, but change over time in the outcome is not assessed; CS ⫽ cross-sectional; LONG ⫽ longitudinal), and sources of data for primary variables (IND RPRTS ⫽ assessment
of predictor and outcome variables based on data from separate sources; OBS ⫽ observational data; SELF ⫽ all self-report data) are reported. SES ⫽ socioeconomic status.
350
REPETTI, TAYLOR, AND SEEMAN
Table 3 (continued)
tionships in childhood and adulthood. In a study of college stu-
dents, the negative effect of childhood exposure to physical ag-
gression at home on current relationships was mediated by height-
ened rejection sensitivity (worries about social acceptance; S.
Feldman & Downey, 1994). During childhood, the link between
physical abuse at home and aggressive behavior with peers is
partially mediated by patterns of social information processing,
such as tendencies to attribute hostile motives to others, to pay less
attention to relevant social cues, and to think of fewer effective
behavioral responses to problematic social situations (Dodge,
Bates, & Pettit, 1990).
Hostility. Hostility is an oppositional orientation toward peo-
ple stemming from feelings of insecurity about oneself and nega-
tive feelings toward others (Houston & Vavak, 1991). Early family
environments characterized as unsupportive, unaccepting, and
conflictual contribute to the development of hostility (Houston &
Vavak, 1991; Smith, Pope, Sanders, Allred, & O’Keefe, 1988;
Woodall & Matthews, 1989), a link that has been documented in
longitudinal studies (Matthews, Woodall, Kenon, & Jacob, 1996;
Woodall & Matthews, 1993). In addition to its origins in the family
environment, hostility may also have biological origins, specifi-
cally representing a psychological response to high levels of phys-
iological reactivity (Fukudo et al., 1992; Krantz & Manuck, 1984).
To the extent that hostility has a genetic basis in physiological
reactivity, parents and children who share genes that predispose
them to this reactivity may create and respond to the family
environment in ways that foster, rather than counteract, the devel-
opment of hostility. Probably no single construct better reflects the
complex intertwining of biological regulatory systems, emotion
processing, and social competence that is at the core of our concept
of an integrated risk profile. Biopsychosocial constructs such as
hostility are ideally suited for models with multiple developmental
processes that have cascading effects over time; they are precisely
the type of outcomes our model predicts, manifesting themselves
most frequently later in the cascade of developmental processes.
Summary. Parents and siblings in risky families are poor mod-
els of prosocial behavior, and they do not provide other kinds of
active socialization that would facilitate the early development of
complex social skills. Social experiences in risky families may also
contribute to social information-processing rules and biases, and to
mental representations of self and others, that interfere with pos-
itive social interaction and the maintenance of healthy relation-
ships. Moreover, the early and continuing impact of warped emo-
tion processing, such as increased reactivity to anger and conflict,
place added demands on the social skills of children from risky
families and further impede the development of social competence.
Social Competence and Mental and Physical
Health Outcomes
Social competence is an integral component of mental health at
all ages. Longitudinal studies show that school-age children who
are rejected or neglected by their peers are at an increased risk for
behavioral and emotional problems a few years later (Hymel,
Rubin, Rowden, & Le Mare, 1990; Kupersmidt & Patterson,
1991). A lack of social integration among adults, particularly with
respect to primary ties with supportive significant others, such as
a spouse or children, is associated with an increased risk of
depression (George, 1989). There is also a long-term association
between childhood social competence and adult mental health;
rejected children are at an increased risk for adult psychopathology
(Bagwell, Newcomb, & Bukowski, 1998; Parker & Asher, 1987).
In terms of physical health risks, social competence most clearly
translates into the ability to attract and sustain social support. In
over 100 investigations, social support has been documented to
reduce health risks of all kinds, affecting the likelihood of illness
initially, the course of recovery among people who are already ill,
and mortality risk more generally (House, Umberson, & Landis,
1988; Seeman, 1996; Uchino, Uno, & Holt-Lunstad, 1999). Some
of the social behaviors associated with risky family environments,
such as a hostile interpersonal style, can also generate stress.
Research suggests that conflictual social interactions may contrib-
ute as much to illness and poor health as supportive social contacts
contribute to good health (e.g., Rook, 1984; see Taylor, 1999). For
example, hostility has been tied to high levels of low-density
lipoprotein cholesterol, high levels of triglycerides, and a higher
ratio of total cholesterol to high-density lipoprotein cholesterol in
women (Suarez, Bates, & Harralson, 1998), as well as to the
likelihood of developing coronary heart disease in adulthood
(Dembroski et al., 1985). The association between hostility and
cardiovascular health may be multidetermined, partly mediated by
the elevated physiological reactivity that appears to be a compo-
nent of hostility and partly mediated through the added stress of
interpersonal conflict.
Through their impact on social competence, and the skills and
cognitions it entails, childhood family environments influence the
kind of interpersonal relationships that offspring have throughout
life. It is through this channel, particularly because relationships
can act as sources of social support and social stress, that the
development of social competence in the family has a lasting
impact on mental and physical health. Although no single study
has addressed this entire model—which links the family environ-
ment to social competence, the quality of subsequent relationships,
and ultimately, health—there is empirical support for several of the
steps. Ewart (1991) reviewed evidence that hostile parenting pro-
duces deficits in social competence that foster vulnerability to
emotionally charged negative interpersonal events, which, in turn,
is associated with heightened cardiovascular reactivity and disease
risk. With respect to mental health outcomes, evidence indicates
that chronic interpersonal stress in adulthood is one of the condi-
tions that connects exposure to family violence during childhood
to recurrence of depression in adulthood (Kessler & Magee, 1994).
Substance Abuse and Risky Sexual Behavior
Adolescence brings increasing autonomy to make decisions
about how, and with whom, to spend time. The range of options for
teens in the United States includes behaviors that pose serious
threats to health. Although parents can exert some direct control
over teenagers’ behavior and activities, our model suggests that the
family’s indirect effects are even more significant. As depicted in
Figure 1, the cascade of influences impinging upon the behavior of
the adolescent offspring of risky families includes all of the earlier
disruptions in biological regulation, emotion processing, and social
competence. In this section, we discuss how a risky family envi-
ronment might increase the likelihood that an adolescent will
engage in behaviors that threaten health. We focus on two classes
of behavior—substance use (including alcohol, cigarettes, and
351
RISKY FAMILIES
illicit drugs) and risky sexual behaviors (such as early, promiscu-
ous, and unprotected sexual intercourse)—because of the short-
and long-term risks they pose for mortality, severe illness, or
serious life disruption.
The substance abuse treatment literature indicates that repair of
the family social environment reduces adolescent drug use. For
example, family therapy is more effective than individual or peer
group counseling (Stanton & Shadish, 1997), and improvements in
parenting practices over the course of family therapy are associ-
ated with reductions in drug use (S. E. Schmidt, Liddle, & Dakof,
1996). These findings highlight the important part played by the
family in changing patterns of drug use, although they do not rule
out the possibility that the same characteristics of a child may both
undermine the development of supportive family relationships and
also increase his or her propensity to abuse substances (Wills,
DuHamel, & Vaccaro, 1995).
Risky Family Characteristics and Substance Abuse and
Risky Sexual Behavior
The specific characteristics of risky families have been tied to
increased rates of smoking, alcohol abuse, drug use, and risky
sexual behavior in adolescence and adulthood. For example, some
research suggests that the experience of abuse in childhood is a
risk factor for these behaviors (Malinosky-Rummell & Hansen,
1993; Small & Luster, 1994). The evidence for this association is
based primarily on cross-sectional analyses of data provided by a
single respondent, often retrospective descriptions of abuse in
childhood (Anda et al., 1999; R. M. Cunningham, Stiffman, Dore´,
& Earls, 1994; Dietz et al., 1999; Felitti et al., 1998; Harrison,
Hoffmann, & Edwall, 1989). Studies with stronger research de-
signs have found increased rates of alcohol abuse among adult
women who grew up in conflictual or abusive homes. In one study,
interpersonal conflict in the adoptive homes of female adoptees
interacted with a biologic background of alcoholism (i.e., having at
least one alcoholic biological parent) to increase the probability of
alcohol abuse or dependence, although neither factor alone pre-
dicted problems with alcohol (Cutrona, Cadoret, et al., 1994). In
another study, individuals with documented histories of abuse or
neglect, or both, during childhood were interviewed approxi-
mately 20 years later. Women (but not men) who had been abused
or neglected, or both, as children were more likely than were
matched control persons to qualify for clinical diagnoses of sub-
stance abuse/dependence, particularly alcohol abuse; the effects of
a history of abuse/neglect were observed even after controlling for
parent alcohol/drug problems, as well as other background char-
acteristics, such as race/ethnicity, childhood social class and neigh-
borhood of residence, and parental arrest (Widom & White, 1997).
There is a larger body of prospective research supporting an
association between growing up in a cold, unsupportive, or ne-
glectful home and health-risk behaviors. The prospective studies
summarized in Table 4 found increased rates of substance abuse or
risky sexual behaviors among offspring of families lacking in
cohesion or offspring of parents who were neglectful and unsup-
portive. In two long-term investigations, a less structured and
organized home or a parenting style that was observed to be
unsupportive, uninvolved, or less demanding in early childhood
predicted more drug and alcohol use in adolescence (Baumrind,
1991; Shedler & Block, 1990). Similar results were obtained when
children were followed over 6 years (Barnes, Reifman, Farrell, &
Dintcheff, 2000; Fisher & Feldman, 1998). In other longitudinal
studies, a less nurturing family environment predicted future cig-
arette smoking, increased drug and alcohol use, and involvement
in a pregnancy (Doherty & Allen, 1994; Mounts & Steinberg,
1995; Scaramella, Conger, Simons, & Whitbeck, 1998). There is a
large body of corroborating cross-sectional data suggesting that
teens who abuse substances or engage in risky sexual behavior are
more likely to live in homes in which interpersonal relationships,
particularly with parents, tend to be hostile, distant, unsatisfying,
nonsupportive, or lacking in cohesion (Barrera, Chassin, & Rog-
osch, 1993; Campo & Rohner, 1992; Denton & Kampfe, 1994;
DiBlasio & Benda, 1990; Miller, Forehand, & Kotchick, 1999;
Newcomb, Maddahian, & Bentler, 1986; Resnick et al., 1997;
Small & Luster, 1994; Spoth, Redmond, Hockaday, & Yoo, 1996;
Turner, Irwin, Tschann, & Millstein, 1993; Wills & Cleary, 1996).
Direct and Mediated Effects of Risky Families
Why are children who grow up in risky families more likely to
abuse substances and engage in risky sexual behaviors? Figure 1
suggests several avenues through which risky families may di-
rectly and indirectly influence these behaviors. First, research
indicates that some of the most potent ingredients in a neglectful
home are inadequate parental knowledge about and supervision of
adolescents’ activities. In homes with less parental monitoring and
more permissiveness, adolescents engage in more frequent sexual
activity and more risky sexual behavior (Jemmott & Jemmott,
1992; Metzler, Noell, Biglan, Ary, & Smolkowski, 1994; Miller et
al., 1999; Romer et al., 1994; Small & Luster, 1994), and they
smoke more (Biglan, Duncan, Ary, & Smolkowski, 1995). In a
6-year longitudinal study summarized in Table 4, the association
between a lack of support and nurturance at home and adolescents’
increased use of alcohol was mediated by the extent to which teens
told their parents about their whereabouts and activities (Barnes et
al., 2000). This is consistent with the suggestion that most mea-
sures of parental monitoring assess open channels of communica-
tion between parent and child that reflect child disclosure at least
as much as parental efforts to control and manage children (Stattin
& Kerr, 2000).
The direct impact of neglectful parents on opportunities for
alcohol and drug use may be compounded through a concomitant
increase in the influence exerted by peers. For example, longitu-
dinal data suggest that the effect of peer drug use is much weaker
when parenting is authoritative (i.e., when parents are involved,
make demands, and supervise while demonstrating acceptance and
warmth; Mounts & Steinberg, 1995) and that the impact of inad-
equate parental monitoring on problem behaviors, such as sub-
stance abuse and risky sexual behavior, is partially mediated by
increased association with peers who engage in antisocial and
deviant behavior (Ary, Duncan, Biglan, et al., 1999). Perhaps
children in supportive families are motivated to please and to
imitate their parents, which imposes internal barriers to engaging
in behaviors that would disappoint them (Andrews, Hops, &
Duncan, 1997). One study found that support from parents lowered
teen tolerance for deviant behavior, which had both a direct impact
on substance use and an indirect impact by decreasing affiliation
with peers who use substances (Wills & Cleary, 1996).
352
REPETTI, TAYLOR, AND SEEMAN
Table 4
Cold, Unsupportive, and Neglectful Homes and Health-Risk Behaviors
Study and design Sample characteristics Family environment measure Health risk behavior variable Key finding
Barnes et al., (2000)
CORR/LONG
SELF
506 adolescents (58% female; 29% Black).
Wave 1 age: 13–16 years (M age
⫽ 14.5 years). Wave 6 age: 18–22 years
(M age ⫽ 19.9 years).
Parenting variables assessed at
Wave 1. Maternal support
and nurturance: adolescent
report of maternal behavior
that communicates the
adolescent is valued and
loved (including
praise/encouragement,
advice/guidance, affection).
Parental monitoring:
adolescent report of how
often he or she tells parents
where he or she is going.
Alcohol misuse index composed of
(a) ounces of alcohol per day, (b)
times drunk in past year, and (c)
frequency of five or more drinks
at a time in past year. The
outcome variable in this study
was a latent factor representing
increases in alcohol misuse (i.e.,
the slope) over 6 years.
Less maternal support and nurturance
was associated with an increase in
adolescent misuse of alcohol
over 6 years, an effect that was
mediated by poor parental
monitoring.
Baumrind, 1991
GC/FLW-UP
OBS
139 children and their parents. White,
affluent population. Age at Time 1: 4
years old. Age at Time 3: 15 years.
Parenting assessed at Time 1
by a team of observers
during a home visit and a
structured family-interaction
task.
Drug and alcohol use based on a
2 hr interview (at Time 3) with
the adolescent about dosage,
frequency, and context of use of
alcohol, tobacco, marijuana, and
other illicit drugs.
The Time 1 ratings of the homes of
children who went on to become
heavy or dependent users of
alcohol, drugs, or both were
compared with the Time 1 home
ratings of the children who were
nonusers at age 15. The childhood
homes of the heavy or dependent
users had been described as less
structured and organized 11
years earlier. In addition, the
parents were more often absent
from the home and made fewer
maturity demands when the
children were 4 years old.
Doherty & Allen, 1994
CORR/LONG
IND RPRTS
Nonsmoking adolescents at Time 1 (n ⫽
312), plus 304 of their mothers and 286
of their fathers. Age at Time 1: 11–13
years.
Parent reports of family
cohesion at Time 1 (1982).
Self-reported smoker status at
Time 2 (1988) defined as one or
more cigarettes in the past
month.
Lower family cohesion was
associated with an increased
likelihood that the child would
have become a smoker during the
subsequent 6 years, after
controlling for parental smoking
status.
Fisher & Feldman, 1998
CORR/FLW-UP
SELF
N ⫽ 116 adolescents. Age at Time 1: 13–
18 years. Age at Time 2: 19–25 years.
Adolescent rating at Time 1 of
cohesion, orderliness, and
clarity of roles and rules
in the family.
Self-reported risk behaviors at
Time 2 based on a composite
of 11 behaviors (e.g., alcohol
use, sexual behavior, tobacco
use, violence-related behavior).
Greater organized cohesiveness in the
family at Time 1 was associated
with less risk behavior 6 years
later.
(table continues)
353
RISKY FAMILIES
Study and design Sample characteristics Family environment measure Health risk behavior variable Key finding
Mounts & Steinberg, 1995
CORR/LONG
SELF
Two samples of 500 students (all
significant findings were replicated in
both samples). Time 1: 9th, 10th, and
11th graders. Time 2: 1 year later.
Time 1 rating of parental
authoritativeness based on
the adolescent’s report of
parenting practices.
Authoritativeness was
indicated by high ratings on
strictness-supervision and
on parental acceptance and
involvement.
Assessed at Time 1 and at Time 2:
adolescent’s self-reported
frequency of use of alcohol,
marijuana, and other drugs.
Among adolescents whose parents
were not authoritative, higher
levels of peer drug use predicted
an increase in the adolescent’s
own use of drugs and alcohol.
This was not the case in the high-
authoritativeness group, however.
Scaramella et al., 1998
CORR/FLW-UP
OBS
6 waves collected over a 7-year period. N
⫽ 368 adolescents and their parents.
Wave 1: 7th grade. M age ⫽ 12 years, 7
months (all-White rural sample).
Observer ratings made during
Wave 1 of parental
warmth and involvement
directed toward the target
adolescent during a family-
interaction task.
Substance use (assessed in 9th and
10th grades): self-reported
frequency of using various
substances such as alcohol,
tobacco, and illegal drugs (e.g.,
marijuana, crack). Involvement
in a pregnancy (assessed in 9th,
10th, 11th, and 12th grades):
self-reported pregnancy status
(females); for males, belief that
he was the father of a child
conceived during this time
period.
Observer ratings of less parental
warmth and involvement in the 7th
grade was associated with greater
use of substances in the 9th and
10th grades and with an increased
likelihood of being involved in a
pregnancy by the 12th grade.
Shedler & Block, 1990
GC/FLW-UP
OBS
N ⫽ 101 children who were studied both
at age 5 and age 18.
A parent’s manner of
interacting with the child
was observed during
parent–child interaction
tasks when the child was 5
years old and was then
described by the observers
with a Q-sort measure.
Drug use was assessed at age 18
during individual interviews.
Frequent users were teens who
reported using marijuana once
per week or more and who had
tried at least one other substance
(n ⫽ 20).
The mothers of frequent users were
more likely to be described as
cold, unresponsive, and
underprotective, when the
children were 5 years old.
Note. The key family and outcome variables in each study are highlighted in bold type. For each study, design (CORR ⫽ correlational; GC ⫽ group comparison), timing of assessments (LONG ⫽
longitudinal; FLW-UP ⫽ multiple data collection points, but change over time in the outcome is not assessed), and sources of data for primary variables (OBS ⫽ observational data; IND RPRTS ⫽
assessment of predictor and outcome variables based on data from separate sources; SELF ⫽ all self-report data) are reported.
354
REPETTI, TAYLOR, AND SEEMAN
Table 4 (continued)
Perhaps most important, risky sexual behavior and substance
abuse may compensate for deficiencies in the biological, social,
and emotional functioning of adolescents from risky families.
Adolescents who engage in risky sexual behavior are more likely
to smoke cigarettes and abuse substances, and there is an associ-
ation between drinking, smoking, and other drug use (Biglan et al.,
1990; Capaldi, Crosby, & Stoolmiller, 1996; Donovan & Jessor,
1985; Kandel & Yamaguchi, 1993; Millstein & Moscicki, 1995;
Shiffman et al., 1994; Tubman, Windle, & Windle, 1996).
10
A
clustering of behaviors often suggests that they share common
origins or serve common functions. Risky sexual behaviors, smok-
ing, and substance abuse may represent adaptations to some of the
negative consequences of having grown up in a risky childhood
environment. As discussed in previous sections, these offspring
may have more social problems than their peers, they may be more
reactive to stress (particularly interpersonal stress), and they may
have fewer coping strategies and sources of social support on
which to draw. In one study, the association between unsupportive
parents and adolescent substance use was partially mediated by
teens’ use of anger to cope and a lack of self-control (Wills &
Cleary, 1996). Early and promiscuous sexual behavior and sub-
stance use may help adolescents manage negative emotions and
gain social acceptance in the absence of adequate emotion coping
strategies or social skills (e.g., Aloise-Young, Hennigan, & Gra-
ham, 1996; Mayne & Buck, 1997; see Taylor, 1999, for a review).
Some of the risky health behaviors discussed here may also act
as methods of self-medication in response to biological dysregu-
lations, in particular to the serotonergic dysfunctions that may
occur in risky families. There is significant evidence relating
smoking, alcohol abuse, and drug abuse to enhancement of sero-
tonergic activity (Balfour & Fagerstrom, 1996; Jaffe, 1990;
Ribeiro, Bettiker, Bogdanov, & Wurtman, 1993; Stahl, 1996; C. F.
Valenzuela & Harris, 1997). Abuse of substances such as these
may function, in part, to alleviate feelings of depression or hostility
that arise in conjunction with serotonergic dysfunction (e.g., Stahl,
1996). Thus, substance abuse may help offspring with dysregu-
lated serotonin functioning by facilitating maintenance of higher
levels of serotonin through increasing release or impeding uptake
of the neurotransmitter, or both.
Summary
By adolescence, the offspring of risky families must adapt to the
cumulative consequences of years spent in a damaging home
environment. Substance abuse and risky sexual behavior may help
these adolescents compensate for their biological, emotional, and
social deficiencies. These processes may be compounded by par-
ents’ inadequate knowledge about and supervision of teens’ activ-
ities and by the adolescents’ weak internal barriers to certain risky
behaviors and their greater susceptibility to peer influence. Pat-
terns of substance abuse and problems in behavioral self-
regulation during adolescence as a result of growing up in a risky
family are, no doubt, influenced by the dramatic hormonal changes
that occur during puberty. Exactly how these changes might ex-
acerbate behavior problems and substance abuse is not currently
known, but investigation of this issue represents the kind of biobe-
havioral research suggested by our analysis.
Socioeconomic Status, Family Characteristics, and Mental
and Physical Health Risks
A detailed examination of the larger social ecology within
which family environments emerge is beyond the scope of this
article. Yet, this sociocultural context is undeniably important.
Peers, schools, and neighborhoods act on the sustaining factors
that are at the core of our model and thereby have indirect effects,
in addition to direct effects, on children’s health and development
(Gump, Matthews, & Raikkonen, 1999; Leventhal & Brooks-
Gunn, 2000). The present analysis is fundamentally compatible
with a macrolevel analysis of psychological and biological dys-
function that takes into account socioeconomic differences, as well
as other aspects of a family’s social context that may give rise to
risky family social environments and to adverse mental and phys-
ical health outcomes (e.g., Aneshensel & Sucoff, 1996; Brody &
Flor, 1998; Flinn & England, 1997; McLoyd, 1998; van IJzen-
doorn & Bakermans-Kranenburg, 1996).
Socioeconomic status (SES), an important dimension of a fam-
ily’s social ecology, is significantly related to early mortality, and
an extensive, highly consistent literature documents the negative
physical and mental health outcomes that result as one moves
lower on the SES gradient (Adler, Boyce, Chesney, Folkman, &
Syme, 1993; Chen, Matthews, & Boyce, 2002; Williams & Col-
lins, 1995). Children growing up in poverty also show early signs
of allostatic load, including elevated secretion of cortisol and
epinephrine, and higher resting blood pressure. Moreover, the
effects of poverty appear to be mediated through the child’s
exposure to multiple chronic stressors, such as violence and family
turmoil (Evans & English, in press).
Low SES has also been tied to all of the risky family charac-
teristics described here (e.g., Dodge, Pettit, & Bates, 1994), and
loss of SES has been associated with an increase in these family
characteristics. Poor children are at heightened risk for physical
mistreatment or abuse (McLoyd, 1998; J. Reid, Macchetto, &
Foster, 1999) and exposure to family violence (Emery &
Laumann-Billings, 1998; Garbarino & Sherman, 1980; U.S. De-
partment of Justice, 1994) and are also more likely to be in family
relationships lacking in warmth and support (Bradley, Corwyn,
McAdoo, & Coll, 2001; McLeod & Shanahan, 1996). Both sus-
tained poverty and descent into poverty appear to move parenting
in more harsh, punitive, irritable, inconsistent, and coercive direc-
tions (Wahler, 1990). McLoyd (1998) reviewed evidence that
descent into poverty precipitates marital and parent–child conflict
that, in turn, alters parental behavior in a hostile and coercive
direction, leading to the development of internalizing and exter-
nalizing symptoms in young children (e.g., Conger, Ge, Elder,
Lorenz, & Simons, 1994; Duncan, Brooks-Gunn, & Klebanov,
1994; see also Elder, 1974; Elder, Nguyen, & Caspi, 1985). These
parenting characteristics may evolve, in part, as a result of defi-
cient coping strategies for managing the stressors associated with
low SES. Poor families living in high-crime neighborhoods must
accommodate to persistent, multiple, uncontrollable demands that
require constant effort to meet immediate physical and psycholog-
10
These behaviors tend to also coexist with other health-compromising
behaviors, such as poor sleep and eating habits and little physical activity
(Donovan, Jessor, & Costa, 1991), and with academic failure (Ary, Dun-
can, Duncan, & Hops, 1999).
355
RISKY FAMILIES
ical needs. Such heavy chronic burdens favor reactive coping skills
and can exacerbate the negative effects of other family vulnera-
bility factors (Aspinwall & Taylor, 1997; Repetti & Wood, 1997).
Low SES may act not only as a contextual factor for under-
standing the development of risky families, but also as an outcome
of growing up in a risky family environment. Research literatures
in developmental psychology, sociology, and public health relate
the characteristics of risky families to a wide range of adverse
adult outcomes that cluster with low SES, including low school
achievement, low educational attainment (in years), low adult
income, high likelihood of divorce in adulthood, low occupational
status, and poor status on other indicators of life success (Power &
Hertzman, 1997). For example, in a study mentioned above, adults
with documented histories of abuse or neglect or both during
childhood were less likely to have completed high school or to be
employed in managerial or professional occupations; they were
also more likely to have arrest records (even excluding drug-
related crimes). Moreover, the effects of the childhood family
environment were observed after controlling for background char-
acteristics, such as race/ethnicity, childhood social class and neigh-
borhood of residence, and parental arrest and alcohol/drug prob-
lems (Widom & White, 1997). The impact of risky family
environments on these other indicators of social success may
represent additional routes by which risky families adversely affect
the functioning of offspring over the lifetime.
Of course, low SES is not inevitably associated with risky
family environments. Just as parenting characteristics may be an
important mediator of the effects of low SES on children’s mental
and physical health, effective parenting may buffer children from
the adverse effects of low SES. Cowen, Wyman, Work, and Parker
(1990) found that so-called stress resilient children, who had
successfully weathered a variety of chronic problems (including
poverty, family turmoil, illness, and violence) were characterized
by nonseparation from the primary caregiver during infancy, pos-
itive parent–child relations during preschool and elementary years,
a strong sense of parenting efficacy by the parents, and parental
use of reasoned, age-appropriate, consistent disciplinary practices
(see also Masten, Morison, Pelligrini, & Tellegen, 1990; Rutter,
1990; Werner & Smith, 1982).
When SES is measured and appropriate analyses are reported,
risky family characteristics continue to be associated with the
sustaining factors in our model and with mental and physical
health even after adjustments are made for SES differences in the
sample. Nonetheless, the social context within which a family lives
can shape the interpersonal environment at home, and these effects
represent an important pathway through which socioeconomic
factors influence health. In addition, social success in adulthood,
evidenced by indicators such as educational and occupational
attainment, may represent an additional pathway through which
the effects of risky family characteristics are maintained over the
lifetime.
Conclusions, Limitations, and Implications
Risky families are characterized by conflict, anger, and aggres-
sion, by relationships that lack warmth and support, and by neglect
of the needs of offspring. These families are risky in multiple
ways. First, several of these characteristics, most notably physical
abuse and neglect, represent immediate threats to the lives and
safety of children. Second, the fact that children’s developing
physiological and neuroendocrine systems must repeatedly adapt
to the threatening and stressful circumstances created by these
family environments increases the likelihood of biological dys-
regulations that may contribute to a buildup of allostatic load, that
is, the premature physiological aging of the organism that en-
hances vulnerability to chronic disease and to early mortality in
adulthood (McEwen & Stellar, 1993; Seeman et al., 1997). Third,
risky families fail to provide children with important self-
regulatory skills, leaving the child unable to effectively create and
enlist social support or to deal with emotion-engaging interper-
sonal situations (as well as a wide array of other stressful events
that may require effective coping skills). Finally, risky families
increase children’s vulnerability to behavior problems and sub-
stance abuse, including smoking, alcohol, drugs, and promiscuous
sexual activity. These risks are multiple and pervasive, and they
are related to each other through common biological and psycho-
social pathways. Separately and in concert, they place a child not
only at immediate risk, but also at long-term and life-long risk for
adverse mental and physical health outcomes.
Of course, the processes that appear to be maladaptive in our
model may, in fact, reflect a process of adaptation within the
context of a risky family environment. For example, in a violent
home, vigilence, attributions of hostile motives to others, and a
coping style in which escape is prioritized may be highly adaptive.
However, a price is paid for these adaptations; the costs may
include heightened emotional and physiological reactivity, social
problems with peers, deficient problem solving, and a lifestyle that
poses health risks. Questions about the adaptiveness or maladap-
tiveness of a response can be addressed only with respect to the
specific characteristics of the outcome in question, such as type of
outcome (e.g., physical vs. psychological well-being), time frame
(e.g., outcomes observed immediately vs. years later), setting (e.g.,
school vs. home), and unit of analysis (e.g., individual outcome vs.
family or community well being; Repetti & Wood, 1997).
Children who are genetically predisposed to problems (such as
those with overly reactive or overly inhibited temperaments) may
be more adversely affected by risky family characteristics than are
children without these preexisting vulnerabilities. Genes may in-
fluence the cascade of risk at multiple levels. A genetic factor may
have direct effects on physiological reactivity, on emotional pro-
cessing, and on social competence. Genes may also have indirect
influences via the family environment, because of overlap in
genetic risk between parent and child and the effect of the child’s
behavior on the social environment. Phenotypic expression of a
genetic vulnerability to a particular mental or physical disorder,
whether an inhibited temperament, a propensity for hostility, or a
risk factor for a chronic disease, may, in turn, be influenced by the
family environment. In families that possess risky characteristics,
exacerbation of a genetic risk may lead to a more rapid and
debilitating development of a problem. In families that lack these
pernicious characteristics, the genetic expression may not occur or
may be delayed. For example, growing up with parents who
provide consistent and effective guidance about the regulation of
emotional states may counteract the impact that heightened phys-
iological and emotional reactivity would otherwise have on peer
relationships and social competence. As noted above, the fact that
the same risky family characteristics appear to fuel such a diverse
array of adverse physical and mental health outcomes suggests that
356
REPETTI, TAYLOR, AND SEEMAN
their effects may be to exacerbate preexisting risks, as well as (or
instead of) creating risks that would not otherwise exist.
In certain respects, the family environment contribution to risky
profiles may be underestimated because these same family char-
acteristics are also associated with a broad array of adverse edu-
cational and social outcomes that themselves represent risk factors
for lower SES and poor health outcomes in adulthood. For exam-
ple, child maltreatment, nonsupportive parenting practices, and
poor parental monitoring also predict poor performance in school
(Ary, Duncan, Biglan, et al., 1999; Cutrona, Cole, Colangelo,
Assouline, & Russell, 1994; Eckenrode, Rowe, Laird, & Brath-
waite, 1995; Pettit, Bates, & Dodge, 1997; Steinberg, Lamborn,
Dornbusch, & Darling, 1992). Moreover, school failure is itself
associated with other health risks. In a large nationally represen-
tative sample, academic problems in youth were associated with
cigarette smoking, alcohol use, and involvement in weapon related
violence (Blum, Beuhring, & Rinehart, 2000). Some evidence
suggests that poor academic competence mediates the association
between unsupportive parenting and substance abuse (Wills &
Cleary, 1996). In addition, our review has focused primarily on
significant clinical outcomes, both mental and physical, in making
the case for pernicious family environments. Yet, subclinical man-
ifestations of the problems associated with adverse early family
characteristics are also noteworthy (Felitti et al., 1998; Walker et
al., 1999). Low social competence and difficulties with the control
and expression of emotions, for example, are significant dysfunc-
tions and liabilities in their own right (see, e.g., Goleman, 1996;
Mayer, Salovey, Caruso, & Sitarenios, 2001). Thus, regardless of
whether the problems created or exacerbated by risky family
characteristics lead to diagnosable forms of psychopathology or
specific chronic diseases, they can cause significant disruption in
adolescent and adult life.
Limitations
There are several limitations of the current analysis that merit
consideration. First, our review has focused on a wide range of
mental health and physical health disorders, and each of these
disorders may have a unique risk profile and trajectory. On the
basis of the evidence presented in this article, we believe that the
adverse family characteristics identified will be distinctively
present in those diverse trajectories. But exactly how they figure
into each of the disorders in question must be identified by em-
pirical investigation.
Second, we have focused our analysis only on the adverse
characteristics of conflict, deficient nurturing, and neglect. Other
family characteristics, such as sexual abuse, divorce, and parental
psychopathology, may also be implicated in health and mental
health disorders. For example, sexual abuse in childhood has been
related to adult mental health problems, such as antisocial behavior
(Pakiz, Reinherz, & Giaconia, 1997); to health risk behaviors, such
as smoking, alcohol and drug abuse, and risky sexual behavior; to
adult medical problems, such as breast disease, gastrointestinal
disorders, pelvic inflammatory disease, and bladder infections (see
Golding, 1999; Leserman, Toomey, & Drossman, 1995, for re-
views); and, among men, to increased prevalence of HIV infection
(Springs & Friedrich, 1992; Zierler et al., 1991). We have not
systematically addressed sexual abuse independent of physical
abuse. Similarly, a substantial literature relates parental loss, sep-
aration, or divorce to adverse outcomes in offspring. For example,
parental divorce has been associated with physical and mental
health problems in childhood, with factors indicative of low social
status in adulthood or less successful performance of life tasks, or
both, as well as with premature mortality (Gottman, 1998; Tucker
et al., 1997; Wadsworth, MacLean, Kuh, & Rodgers, 1990; Web-
ster, Orbuch, & House, 1995). However, this literature is hetero-
geneous as to the reasons underlying the loss or separation, and so
the impact of separation of child from parent in the etiology of risk
profiles is difficult to calculate.
Parental psychopathology has been linked both to risky family
characteristics and to poor health outcomes in offspring. For ex-
ample, depressed mothers have been found to be more critical,
negative, and irritable, and less responsive to their children’s needs
(Downey & Coyne, 1990; Nolen-Hoeksema, Wolfson, Mumme, &
Guskin, 1995), and children of depressed mothers are at increased
risk for depression and suicidal behavior, as well as a variety of
other psychological problems (Kaslow et al., 1994). We elected to
omit coverage of the research literature on parental psychopathol-
ogy because the elevated risks observed in offspring are due to a
poorly understood combination of genetic factors and parenting
practices. With few exceptions, it is impossible to use these studies
to isolate the specific effects of parenting behaviors that are
associated with conflict, deficient nurturing, or neglect on child
development. Nonetheless, parental psychopathology is an unde-
niably important input to risky family environments.
Because of space limitations, we have been unable to cover
the extensive literature on resilience in children and the factors
that protect against the adverse effects of risky environments
(Haggerty, Sherrod, Garmezy, & Rutter, 1994). Nonetheless,
our model is consistent with findings from this literature. Re-
silience is predicted by close and warm relationships with a
parent or parent figure, high expectations and structure in the
family, high SES, and individual characteristics that may be, at
least in part, based on genetic factors, such as good intellectual
functioning and an appealing, sociable disposition (Masten &
Coatsworth, 1998). A supportive family environment may also
contribute to the development of dispositional resources that
successfully regulate emotional and behavioral functioning
across the lifespan and represent additional routes that protect
mental and physical health. Such resources include optimism,
psychological control, and other personality characteristics (As-
pinwall & Taylor, 1997).
Other gaps in our analysis include the fact that the magnitude of
evidence for different points varies substantially. Some areas, such
as the relation of physical abuse to serotonergic dysfunction and
deficiencies in emotion understanding implicate specific family
behaviors, whereas in other cases, the evidence is more conjec-
tural. There has been little consideration of the role of gender in the
patterns discussed. Studies have focused disproportionately on
mothering (over fathering) and on outcomes in sons (over daugh-
ters). The literature hints that risky families are more likely to
produce internalizing symptoms and depressive symptomatology
in girls and externalizing symptoms and problem behavior in boys,
but the evidence is not yet definitive (e.g., E. M. Cummings,
Ballard, & El-Sheikh, 1991; Davis, Hops, Alpert, & Sheeber,
1998; Steinberg, 1987).
357
RISKY FAMILIES
Implications for Future Research
What research agenda for the future does the risky families
analysis suggest? A chief priority will be longitudinal investiga-
tions, beginning in early childhood, that chronicle how families
influence the physical and mental health of offspring across the
lifespan. Short of such ambitious long-term undertakings, investi-
gations that focus on comorbidities of physical and mental health
outcomes are needed. Past investigations have often conceived of
mental health and behavioral variables, such as hostility, depres-
sion, and substance abuse, as potential moderators of health tra-
jectories, such as course of CHD; the present analysis suggests, in
contrast, that adverse mental health, behavioral, and physical
health consequences of risky families may be comorbid outcomes
of common underlying biological and psychosocial processes. A
focus on these common processes has the potential to further
elucidate the underlying dysregulations produced in risky families,
as well as clarify how these dysregulations influence such a broad
array of outcomes.
Accordingly, a third priority will be focused scientific investi-
gations of the links among the sustaining factors we have identi-
fied, namely the biological pathways, emotion processing, and
social competence. Although all three sets of factors have been
linked to risky family characteristics as antecedents and to adverse
physical and mental health outcomes as consequences, fewer em-
pirical investigations have focused on the links among them. For
example, responses to stress have biological, emotional, cognitive,
and behavioral components; this calls for an integration of the
research literature on emotion with the literatures on coping and
biological responses to stress. Any effort in this direction must
proceed from the assumption that each of the sustaining factors in
our model receives multiple inputs from genes, environments, and
the interactions between them, which result in different outcomes
within each system. Therefore, simple, uniform mappings across
the biological, emotional, and behavioral systems should not be
expected. We suspect that individual differences in the interrela-
tionships among the different sustaining factors help to explain
why such a diverse set of physical and mental health outcomes
have been linked to the same two core risky family characteristics.
Nonetheless, it may be possible to identify constructs, such as
hostility, that define a specific set of biopsychosocial processes
that place some individuals at risk for particular health outcomes.
A fourth priority is made possible by the mapping of the genome
and addresses our suspicion that a chief effect of risky families is
to exacerbate certain genetic risks. As the polymorphisms of
genetic risk factors are identified, it may become possible to see
how parenting practices exacerbate such risk; existing animal
models suggest the usefulness of such an approach (e.g., Higley et
al., 1996a, 1996b). Studies are also needed that assess genetic risk
in conjunction with parenting characteristics, as both human adop-
tion studies and studies of non-human primates cross-fostered to
females who provide different rearing experiences suggest that
either genetic risk, poor parenting, or a combination of the two
predispose offspring to certain of the problems typically associated
with risky families (Cadoret, Yates, Troughton, Woodworth, &
Stewart, 1995; Suomi, 1991).
An overarching conclusion of our analysis is that progress in
understanding the effects of parenting on children’s mental and
physical health will not come unless we integrate psychosocial
investigations of these processes with an understanding of the
underlying biology. As part of this agenda, we recommend several
foci: Research should identify potential critical or sensitive periods
throughout childhood and adolescence during which particular
biological dysregulations and dysfunctional psychological pro-
cesses may be most likely to develop and have lasting effects.
Unaddressed in this article is the critical prenatal period during
which biological dysregulations may also develop, and these, too,
merit exploration. Recent research suggests that the social envi-
ronment of pregnant women is related to prenatal development;
less social support for the mother predicts poorer fetal growth (P. J.
Feldman, Dunkel-Schetter, Sandman, & Wadhwa, 2000). The ad-
verse consequences of inadequate prenatal development may act as
developmental vulnerabilities, particularly when the first few years
of life are spent in a risky family. For example, a mother in the
midst of the stress of poverty or in a nonsupportive or conflictual
family environment may give birth to a low birth-weight baby,
who may then be at greater risk for psychological problems and
biological dysregulation in a risky family environment than a
normal birth-weight baby would be. Parallel to this thrust should
be efforts to see if there is biological or behavioral plasticity, or
both, in the adverse effects of risky families, such that appropriate
interventions might reverse or offset their deleterious effects. Such
efforts will require experimental investigations using either animal
models or comparative intervention studies with children and their
families.
Implications for Interventions
The problems associated with risky families are serious ones.
Chronic diseases, such as coronary heart disease, are among the
chief causes of death in most developed countries, and the origins
of behavioral risk factors associated with them appear to arise, in
part, in risky families. Over the past 30 years, there has been a two
to threefold increase in suicide and homicide rates among children,
outcomes reliably tied to adverse family characteristics
(Malinosky-Rummell & Hansen, 1993; Sedlak & Broadhurst,
1996; Wagner, 1997). Rates of depression, especially among
women, are at high levels, and risky family environments are
implicated here, too. Behavioral regulation skills and social com-
petence are vital to managing situations that arise in high-stress
environments, yet these skills are compromised by risky families.
Risky families are an important piece of the puzzle represented
by these rampant social and public health problems. Moreover,
they may be especially so with respect to intervention. Focusing on
family characteristics that represent risk factors for (the exacerba-
tion of) major physical and mental health disorders can provide the
basis for early interventions that may at least partially offset the
potential for cascading risk that may accumulate over the lifetime.
Even if many of the self-regulatory problems and propensities for
adverse health outcomes reviewed here are subsequently found to
have genetic origins, families are the first environment within
which these predispositions are expressed, and, as our review has
shown, parenting appears to exacerbate or restrain the progression
of these proclivities. For example, family interventions that suc-
ceed in reducing family conflict and anger and increasing cohesion
and warmth have demonstrable beneficial effects for children who
are aggressive or diagnosed with other externalizing problems
(Barkley, Guevremont, Anastopoulos, & Fletcher, 1992; Kazdin,
358
REPETTI, TAYLOR, AND SEEMAN
1994; Sayger, Horne, Walker, Passmore, 1988; Serketich & Du-
mas, 1996).
Interventions that help parents learn the kinds of behaviors that
may shape effective behavioral and self-regulatory skills in chil-
dren, or that are specifically tailored to compensate for preexisting
problems in children, are potentially valuable. A number of inter-
ventions have been developed to reduce violent behavior in fam-
ilies, including individual, group, and couples therapy, parent
training, and home visits. In mild to moderate abuse situations,
such programs have shown modest success in teaching parents
better skills and reducing the likelihood of further abuse; in fam-
ilies marked by serious and chronic abuse, results have been less
successful (Albee & Gullotta, 1997; McLoyd, 1998). Other exam-
ples include social service interventions to teach nurturant parent-
ing behaviors, which have also shown some degree of success
(e.g., increasing parents’ emotionally supportive behavior, provi-
sion of intellectual stimulation, and time spent on parenting;
Besherov & Laumann, 1996; Black, Dubowitz, Hutcheson,
Berenson-Howard, & Starr, 1995). As the health outcomes of
adverse family characteristics become better known, we may see
more such family interventions. For example, efforts to modify
temperamental cardiovascular risk factors, such as cynical hostil-
ity, are usually undertaken after an acute coronary event or other
basis for diagnosing cardiovascular disease (e.g., English & Baker,
1983; Siegman, Dembroski, & Crump, 1992). If longitudinal ev-
idence continues to suggest that these characteristics begin to
develop in early childhood, attempts to offset their development
may be pushed farther back into childhood.
Research, policy, and interventions must consider the social
context within which parenting characteristics are expressed. Par-
ents can learn to become more effective and nurturing, but if they
remain in high-stress environments marked by grinding poverty,
high crime rates, inadequate employment opportunities, and insuf-
ficient support systems, then the success of such efforts will be
compromised (Zaslow & Eldred, 1998). The adverse effects of low
SES on mental and physical health outcomes are as close to a
universal truth as social science has offered. Without attention to
the social contexts in which families develop, and in which risk
and resilience are transmitted from generation to generation, sci-
entists will remain the helpless chroniclers of the outcomes de-
scribed in this article.
References
Adler, N. E., Boyce, W. T., Chesney, M. A., Folkman, S., & Syme, L.
(1993). Socioeconomic inequalities in health: No easy solution. Journal
of the American Medical Association, 269, 3140–3145.
Albee, G. W., & Gullotta, T. P. (Eds.). (1997). Primary prevention works.
Thousand Oaks, CA: Sage.
Allen, M. T., Matthews, K. A., & Sherman, F. S. (1997). Cardiovascular
reactivity to stress and left ventricular mass in youth. Hypertension, 30,
782–787.
Aloise-Young, P. A., Hennigan, K. M., & Graham, J. W. (1996). Role of
the self-image and smoker stereotype in smoking onset during early
adolescence: A longitudinal study. Health Psychology, 15, 494–497.
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual of mental disorders (4th ed.). Washington, DC: Author.
Anda, R. F., Croft, J. B., Felitti, V. J., Nordenberg, D., Giles, W. H.,
Williamsion, D. F., & Giovino, G. A. (1999). Adverse childhood expe-
riences and smoking during adolescence and adulthood. JAMA, 282,
1652–1658.
Andersen, S. M., & Berk, M. S. (1998). The social–cognitive model of
transference: Experiencing past relationships in the present. Current
Directions in Psychological Science, 7, 109–115.
Andrews, J. A., Hops, H., & Duncan, S. C. (1997). Adolescent modeling
of parent substance use: The moderating effect of the relationship with
the parent. Journal of Family Psychology, 11, 259–270.
Aneshensel, C. S., & Sucoff, C. A. (1996). The neighborhood context of
adolescent mental health. Journal of Health and Social Behavior, 37,
293–310.
Ary, D. V., Duncan, T. E., Biglan, A., Metzler, C. W., Noell, J. W., &
Smolkowski, K. (1999). Development of adolescent problem behavior.
Journal of Abnormal Child Psychology, 27, 141–150.
Ary, D. V., Duncan, T. E., Duncan, S. C., & Hops, H. (1999). Adolescent
problem behavior: The influence of parents and peers. Behaviour Re-
search and Therapy, 37, 217–230.
Aspinwall, L. G., & Taylor, S. E. (1997). A stitch in time: Self-regulation
and proactive coping. Psychological Bulletin, 121, 417–436.
Bagwell, C. L., Newcomb, A. F., & Bukowski, W. M. (1998). Preadoles-
cent friendship and peer rejection as predictors of adult adjustment.
Child Development, 69, 140–153.
Balfour, D. J. K., & Fagerstrom, K. O. (1996). Pharmacology of nicotine
and its therapeutic use in smoking cessation and neurodegenerative
disorders. Pharmacological Therapeutics, 71, 51–81.
Ballard, M., Cummings, E. M., & Larkin, K. (1993). Emotional and
cardiovascular responses to adults’ angry behavior and to challenging
tasks in children of hypertensive and normotensive parents. Child De-
velopment, 64, 500–515.
Barber, B. K. (1996). Parental psychological control: Revisiting a ne-
glected construct. Child Development, 67, 3296–3319.
Barkley, R. A., Guevremont, D. C., Anastopoulos, A. D., & Fletcher, K. E.
(1992). A comparison of three family therapy programs for treating
family conflicts in adolescents with attention-deficit hyperactivity dis-
order. Journal of Consulting and Clinical Psychology, 60, 450–462.
Barnes, G. M., Reifman, A. S., Farrell, M. P., & Dintcheff, B. A. (2000).
The effects of parenting on the development of adolescent alcohol
misuse: A six-wave latent growth model. Journal of Marriage and the
Family, 62, 175–186.
Barrera, M., Chassin, L., & Rogosch, F. (1993). Effects of social support
and conflict on adolescent children of alcoholic and nonalcoholic fa-
thers. Journal of Personality and Social Psychology, 64, 602–612.
Basic Behavioral Science Task Force of the National Advisory Mental
Health Council. (1996). Basic behavioral science research for mental
health: Family processes and social networks. American Psycholo-
gist, 51, 622–630.
Baumrind, D. (1991). The influence of parenting style on adolescent
competence and substance use. Journal of Early Adolescence, 11,
56–95.
Besherov, D. J., & Laumann, L. (1996). Child abuse reporting. In I.
Garfinkel, J. L. Hochschild, & S. McLanahan (Eds.), Social policies for
children. Washington, DC: Brookings Institution Press.
Biglan, A., Duncan, T. E., Ary, D., & Smolkowski, K. (1995). Peer and
parental influences on adolescent tobacco use. Journal of Behavioral
Medicine, 18, 315–330.
Biglan, A., Metzler, C. W., Wirt, R., Ary, D., Noell, J., Ochs, L., et al.
(1990). Social and behavioral factors associated with high-risk sexual
behavior among adolescents. Journal of Behavioral Medicine, 13, 245–
261.
Black, M. M., Dubowitz, H., Hutcheson, J., Berenson-Howard, J., & Starr,
R. H. (1995). A randomized clinical trial of home intervention for
children with failure to thrive. Pediatrics, 95, 807–814.
Blum, R. W., Beuhring, T., & Rinehart, P. M. (2000). Protecting teens:
Beyond race, income and family structure. Minneapolis: University of
Minnesota, Center for Adolescent Health.
Booth, C. L., Rose-Krasnor, L., McKinnon, J.-A., & Rubin, K. H. (1994).
359
RISKY FAMILIES
Predicting social adjustment in middle childhood: The role of preschool
attachment security and maternal style. Social Development, 3, 189–
204.
Bost, K. K., Vaughn, B. E., Washington, W. N., Cielinski, K. L., &
Bradbard, M. R. (1998). Social competence, social support, and attach-
ment: Demarcation of construct domains, measurement, and paths of
influence for preschool children attending Head Start. Child Develop-
ment, 69, 192–218.
Bradley, R. H., Corwyn, R. F., McAdoo, H. P., & Coll, C. G. (2001). The
home environments of children in the United States: I. Variations by age,
ethnicity, and poverty status. Child Development, 72, 1844–1867.
Brody, G. H., & Flor, D. L. (1998). Maternal resources, parenting prac-
tices, and child competence in rural, single-parent African-American
families. Child Development, 69, 803–816.
Brown, M. S., & Tanner, C. (1988). Type A behavior and cardiovascular
responsitivity in preschoolers. Nursing Research, 37, 152–191.
Bugental, D. B. (2000). Acquisition of the algorithms of social life: A
domain-based approach. Psychological Bulletin, 126, 187–219.
Cadoret, R. J., Yates, W. R., Troughton, E., Woodworth, G., & Stewart,
M. A. (1995). Genetic–environmental interaction in the genesis of
aggresivity and conduct disorders. Archives of General Psychiatry, 52,
916–924.
Caldi, C., Tannenbaum, B., Sharma, S., Francis, D., Plotsky, P. M., &
Meaney, M. J. (1998). Maternal care during infancy regulates the de-
velopment of neural systems mediating the expression of fearfulness in
the rat. Proceedings of the National Academy of Science of the United
States of America, 95, 5335–5340.
Campo, A. T., & Rohner, R. P. (1992). Relationships between perceived
parental acceptance–rejection, psychological adjustment, and substance
abuse among young adults. Child Abuse and Neglect, 16, 429–440.
Campos, R. (1988). Comfort measures for infant pain. Zero to Three, 9,
6–13.
Camras, L. A., & Rappaport, S. (1993). Conflict behaviors of maltreated
and nonmaltreated children. Child Abuse and Neglect, 17, 455–464.
Camras, L. A., Ribordy, S., Hill, J., Martino, S., Spaccarelli, S., & Stefani,
R. (1988). Recognition and posing of emotional expressions by abused
children and their mothers. Developmental Psychology, 24, 776–781.
Capaldi, D. M., Crosby, L., & Stoolmiller, M. (1996). Predicting the timing
of first sexual intercourse for at-risk adolescent males. Child Develop-
ment, 67, 344–359.
Carson, J. L., & Parke, R. D. (1996). Reciprocal negative affect in parent–
child interactions and children’s peer competency. Child Develop-
ment, 67, 2217–2226.
Cassidy, J., Parke, R. D., Butkovsky, L., & Braungard, J. M. (1992).
Family–peer connections: The roles of emotional expressiveness within
the family and children’s understanding of emotions. Child Develop-
ment, 63, 603–618.
Champoux, M., Byrne, E., Delizio, R. D., & Suomi, S. J. (1992). Moth-
erless mothers revisited: Rhesus maternal behavior and rearing history.
Primates, 33, 251–255.
Chen, E., Matthews, K. A., & Boyce, W. T. (2002). Socioeconomic
differences in children’s health: How and why do these relationships
change with age? Psychological Bulletin, 128, 295–329.
Chorpita, B. F., & Barlow, D. H. (1998). The development of anxiety: The
role of control in the early environment. Psychological Bulletin, 124,
3–21.
Chrousos, G. P., & Gold, P. W. (1992). The concepts of stress and stress
system disorders: Overview of physical and behavioral homeostasis.
Journal of the American Medical Association, 267, 1244–1252.
Coccaro, E. F. (1992). Impulsive aggression and central serotonergic
system function in humans: An example of a dimensional brain-behavior
relationship. International Clinical Psychopharmacology, 7, 3–12.
Coccaro, E. F., Kavoussi, R. J., Cooper, T. B, & Hauger, R. L. (1997).
Central serotonin activity and aggression: Inverse relationship with
prolactin response to D-fenfluramine, but not CSF 5-HIAA concentra-
tion, in human subjects. American Journal of Psychiatry, 154, 1430–
1435.
Coccaro, E. F., Silverman, J., Klar, H., Horvath, T., & Siever, L. (1994).
Familial correlates of reduced central serotonergic system function in
patients with personality disorders. Archives of General Psychiatry, 51,
318–324.
Conger, R., Ge, X., Elder, G., Lorenz, F., & Simons, R. (1994). Economic
stress, coercive family process, and developmental problems of adoles-
cents. Child Development, 65, 541–561.
Cowen, E. L., Wyman, P. A., Work, W. C., & Parker, G. R. (1990). The
Rochester Child Resilience Project: Overview and summary of first year
findings. Development and Psychopathology, 2, 193–212.
Crockenberg, S., & Lourie, A. (1996). Parents’ conflict strategies with
children and children’s conflict strategies with peers. Merrill-Palmer
Quarterly, 42, 495–518.
Cummings, E. M., Ballard, M., & El-Sheikh, M. (1991). Responses of
children and adolescents to interadult anger as a function of gender, age,
and mode of expression. Merrill-Palmer Quarterly, 37, 543–560.
Cummings, E. M., Zahn-Waxler, C., & Radke-Yarrow, M. (1981). Young
children’s responses to expressions of anger and affection by others in
the family. Child Development, 52, 1274–1282.
Cummings, J. S., Pellegrini, D. S., Notarius, C. I., & Cummings, E. M.
(1989). Children’s responses to angry adult behavior as a function of
marital distress and history of interparental hostility. Child Develop-
ment, 60, 1035–1043.
Cunningham, R. M., Stiffman, A. R., Dore´, P., & Earls, F. (1994). The
association of physical and sexual abuse with HIV risk behaviors in
adolescence and young adulthood: Implications for public health. Child
Abuse and Neglect, 18, 233–245.
Cutrona, C. E., Cadoret, R. J., Suhr, J. A., Richards, C. C., Troughton, E.,
Schutte, K., & Woodworth G. (1994). Interpersonal variables in the
prediction of alcoholism among adoptees: Evidence for gene–
environment interactions. Comprehensive Psychiatry, 35, 171–179.
Cutrona, C. E., Cole, V., Colangelo, N., Assouline, S. G., & Russell, D. W.
(1994). Perceived parental social support and academic achievement: An
attachment theory perspective. Journal of Personality and Social Psy-
chology, 66, 369–378.
Davies, P. T., & Cummings, E. M. (1998). Exploring children’s emotional
security as a mediator of the link between marital relations and child
adjustment. Child Development, 69, 124–139.
Davis, B. T., Hops, H., Alpert, A., & Sheeber, L. (1998). Child responses
to parental conflict and their effect on adjustment: A study of triadic
relations. Journal of Family Psychology, 12, 163–177.
Dembroski, T. M., MacDougall, J. M., Williams, R. B., Haney, T. L., &
Blumenthal, J. A. (1985). Components of Type A, hostility, and anger-
in: Relationship to angiographic findings. Psychosomatic Medicine, 47,
219–233.
Denton, R. E., & Kampfe, C. M. (1994). The relationship between family
variables and adolescent substance abuse: A literature review. Adoles-
cence, 29, 475–495.
Depue, R. A., & Iacono, W. G. (1989). Neurobehavioral aspects of affec-
tive disorders. Annual Review of Psychology, 40, 457–492.
De Wolff, M. S., & van IJzendoorn, M. H. (1997). Sensitivity and attach-
ment: A meta-analysis on parental antecedents of infant attachment.
Child Development, 68, 571–591.
DiBlasio, F. A., & Benda, B. B. (1990). Adolescent sexual behavior:
Multivariate analysis of a social learning model. Journal of Adolescent
Research, 5, 449–466.
Dietz, P. M., Spitz, A. M., Anda, R. F., Williamson, D. F., McMahon,
P. M., Santelli, J. S., et al. (1999). Unintended pregnancy among adult
women exposed to abuse or household dysfunction during their child-
hood. JAMA 282, 1359–1364.
360
REPETTI, TAYLOR, AND SEEMAN
Dishion, T. J. (1990). The family ecology of boys peer relations in middle
childhood. Child Development, 61, 874–891.
Dodge, K. A., Bates, J. E., & Pettit, G. S. (1990, December 21). Mecha-
nisms in the cycle of violence. Science, 250, 1678–1683.
Dodge, K. A., Pettit, G. S., & Bates, J. E. (1994). Socialization mediators
of the relation between socioeconomic status and child conduct prob-
lems. Child Development, 65, 649–665.
Doherty, W. J., & Allen, W. (1994). Family functioning and parental
smoking as predictors of adolescent cigarette use: A six-year prospective
study. Journal of Family Psychology, 8, 347–353.
Donovan, J. E., & Jessor, R. (1985). Structure of problem behavior in
adolescence and young adulthood. Journal of Consulting and Clinical
Psychology, 53, 890–904.
Donovan, J. E., Jessor, R., & Costa, F. M. (1991). Adolescent health
behavior and conventionality–unconventionality: An extension of
problem-behavior therapy. Health Psychology, 10, 52–61.
Downey, G., & Coyne, J. (1990). Children of depressed parents: An
integrative review. Psychological Bulletin, 108, 50–76.
Duncan, G., Brooks-Gunn, J., & Klebanov, P. (1994). Economic depriva-
tion and early childhood development. Child Development, 65, 296–
318.
Dunn, J., & Brown, J. (1994). Affect expression in the family, children’s
understanding of emotions, and their interactions with others. Merrill-
Palmer Quarterly, 40, 120–137.
Dunn, J., Brown, J., Slomkowski, C., Tesla, C., & Youngblade, L. (1991).
Young children’s understanding of other people’s feelings and beliefs:
Individual differences and their antecedents. Child Development, 62,
1352–1366.
Eckenrode, J., Rowe, E., Laird, M., & Brathwaite, J. (1995). Mobility as a
mediator of the effects of child maltreatment on academic performance.
Child Development, 66, 1130–1142.
Eisenberg, N., Fabes, R. A., Bernszweig, J., Karbon, M., Poulin, R., &
Hanish, L. (1993). The relations of emotionality and regulation to
preschoolers’ social skills and sociometric status. Child Develop-
ment, 64, 1418–1438.
Eisenberg, N., Fabes, R. A., Guthrie, I. K., Murphy, B. C., Maszk, P.,
Holmgren, R., & Suh, K. (1996). The relations of regulation and emo-
tionality to problem behavior in elementary school children. Develop-
ment and Psychopathology, 8, 141–162.
Eisenberg, N., Fabes, R. A., & Murphy, B. C. (1996). Parents’ reactions to
children’s negative emotions: Relations to children’s social competence
and comforting behavior. Child Development, 67, 2227–2247.
Eisenberg, N., Fabes, R. A., Shepard, S. A., Murphy, B. C., Guthrie, I. K.,
Jones, S., et al. (1997). Contemporaneous and longitudinal prediction of
children’s social functioning from regulation and emotionality. Child
Development, 68, 642–664.
Elder, G. (1974). Children of the Great Depression. Chicago: University of
Chicago Press.
Elder, G., Nguyen, T., & Caspi, A. (1985). Linking family hardship to
children’s lives. Child Development, 56, 361–375.
El-Sheikh, M., Cummings, E. M., & Goetsch, V. L. (1989). Coping with
adults’ angry behavior: Behavioral, physiological, and verbal responses
in preschoolers. Developmental Psychology, 25, 490–498.
El-Sheikh, M., Harger, J., & Whitson, S. M. (2001). Exposure to interpa-
rental conflict and children’s adjustment and physical health: The mod-
erating role of vagal tone. Child Development, 72, 1617–1636.
Emery, R. E. (1982). Interparental conflict and the children of discord and
divorce. Psychological Bulletin, 92, 310–330.
Emery, R. E. (1988). Marriage, divorce, and children’s adjustment. New-
bury Park, CA: Sage.
Emery, R. E., & Laumann-Billings, L. (1998). An overview of the nature,
causes, and consequences of abusive family relationships. American
Psychologist, 53, 121–135.
English, E. H., & Baker, T. B. (1983). Relaxation training and cardiovas-
cular response to experimental stressors. Health Psychology, 2, 239–
259.
Evans, G. W., & English, K. (in press). The environment of poverty:
Multiple stressor exposure, psychophysiological stress, and socioemo-
tional adjustment. Child Development.
Ewart, C. K. (1991). Familial transmission of essential hypertension,
genes, environments, and chronic anger. Annals of Behavioral Medi-
cine, 13, 40–47.
Fabes, R. A., & Eisenberg, N. (1997). Regulatory control and adults’
stress-related responses to daily life events. Journal of Personality and
Social Psychology, 73, 1107–1117.
Feldman, P. J., Dunkel-Schetter, C., Sandman, C. A., & Wadhwa, P. D.
(2000). Maternal social support predicts birth weight and fetal growth in
human pregnancy. Psychosomatic Medicine, 62, 715–725.
Feldman, S., & Downey, G. (1994). Rejection sensitivity as a mediator of
the impact of childhood exposure to family violence on adult attachment
behavior. Development and Psychopathology, 6, 231–247.
Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Apitz, A. M.,
Edwards, et al. (1998). Relationship of childhood abuse and household
dysfunction to many of the leading causes of death in adults. American
Journal of Preventive Medicine, 14, 245–258.
Fisher, L., & Feldman, S. S. (1998). Familial antecedents of young adult
health risk behavior: A longitudinal study. Journal of Family Psychol-
ogy, 12, 66–80.
Flinn, M. V., & England, B. G. (1997). Social economics of childhood
glucocorticoid stress responses and health. American Journal of Physical
Anthropology, 102, 33–53.
Frasure-Smith, N., Lesperance, F., & Talajic, M. (1995). The impact of
negative emotions on prognosis following myocardial infarction: Is it
more than depression? Health Psychology, 14, 388–398.
Fukudo, S., Lane, J. D., Anderson, N. B., Kuhn, C. M., Schanberg, S. M.,
McCown, N., et al. (1992). Accentuated vagal antagonism of beta-
adrenergic effects on ventricular repolarization: Evidence of weaker
antagonism in hostile Type A men. Circulation, 85, 2045–2053.
Garbarino, J., & Sherman, D. (1980). High-risk neighborhoods and high-
risk families: The human ecology of child maltreatment. Child Devel-
opment, 51, 188–198.
George, L. K. (1989). Stress, social support, and depression over the
life-course. In K. Markides & C. Cooper (Eds.), Aging, stress, social
support, and health (pp. 241–267). London: Wiley.
Gil, K. M., Keefe, F. J., Sampson, H. A., McCaskill, C. C., Rodin, J., &
Crisson, J. E. (1987). The relation of stress and family environment to
atopic dermatitis symptoms in children. Journal of Psychosomatic Re-
search, 31, 673–684.
Golding, J. M. (1999). Sexual-assault history and long-term physical health
problems: Evidence from clinical and population epidemiology. Current
Directions in Psychological Science, 8(6), 191.
Goleman, D. (1996). Emotional intelligence. New York: Bantam Books.
Gordis, E. B., Margolin, G., & John, R. S. (1997). Marital aggression,
observed parental hostility, and child behavior during triadic family
interaction. Journal of Family Psychology, 11, 76–89.
Gottman, J. M. (1998). Psychology and the study of marital processes.
Annual Review of Psychology, 49, 169–197.
Gottman, J. M., & Katz, L. F. (1989). Effects of marital discord on young
children’s peer interaction and health. Developmental Psychology, 25,
373–381.
Gottman, J. M., Katz, L. F., & Hooven, C. (1996). Parental meta-emotion
philosophy and the emotional life of families. Theoretical models and
preliminary data. Journal of Family Psychology, 10, 243–268.
Gottman, J. M., Katz, L. F., & Hooven, C. (Eds.). (1997). Meta-emotion:
How families communicate emotionally. Hillsdale, NJ: Erlbaum.
Graves, P. L., Wang, N.-Y., Mead, L. A., Johnson, J. V., & Klag, M. J.
(1998). Youthful precursors of midlife social support. Journal of Per-
sonality and Social Psychology, 74, 1329–1336.
361
RISKY FAMILIES
Grych, J. H., & Fincham, F. D. (1990). Marital conflict and children’s
adjustment: A cognitive–contextual framework. Psychological Bulletin,
108, 267–290.
Gump, B. B., Matthews, K. A., & Raikkonen, K. (1999). Modeling
relationships among socioeconomic status, hostility, cardiovascular re-
activity, and left ventricular mass in African American and White
children. Health Psychology, 18, 140–150.
Gunnar, M. R. (1998). Quality of early care and buffering of neuroendo-
crine stress reactions: Potential effects on the developing human brain.
Preventive Medicine, 27, 208–211.
Gunnar, M. R., Brodersen, L., Nachmias, M., Buss, K., & Rigatuso, J.
(1996). Stress reactivity and attachment security. Developmental Psy-
chobiology, 29, 191–204.
Gunnar, M. R., Gonzalez, C. A., Goodlin, B. L., & Levine, S. (1981).
Behavioral and pituitary–adrenal responses during a prolonged separa-
tion period in rhesus monkeys. Psychoneuroendocrinology, 6, 65–75.
Haggerty, R. J., Sherrod, L. R., Garmezy, N., & Rutter, M. (1994). Stress,
risk, and resilience in children and adolescents: Processes, mechanisms,
and interventions. Cambridge, England: Cambridge University Press.
Hall, F. S., Wilkinson, L. S., Humby, T. & Robbins, T. W. (1999).
Maternal deprivation of neonatal rats produces enduring changes in
dopamine function. Synapse, 32, 37–43.
Halperin, J., Newcorn, J., Kepstein, I., McKay, K., Schwartz, S., Siever L.,
et al. (1997). Serotonin aggression and parental psychopathology in
children with attention-deficit hyperactivity disorder. Journal of the
American Academy of Child and Adolescent Psychiatry, 36, 1391–1398.
Hardy, D. F., Power, T. G., & Jaedicke, S. (1993). Examining the relation
of parenting to children’s coping with everyday stress. Child Develop-
ment, 64, 1829–1841.
Harrison, P. A., Hoffman, N. G., & Edwall, G. E. (1989). Sexual abuse
correlates: Similarities between male and female adolescents in chemical
dependency treatment. Journal of Adolescent Research, 4, 385–399.
Hart, C. H., Nelson, D. A., Robinson, C. C., Olsen, S. F., & McNeilly-
Choque, M. K. (1998). Overt and relational aggression in Russian
nursery-school-age children: Parenting style and marital linkages. De-
velopmental Psychology, 34, 687–697.
Hart, J., Gunnar, M., & Cicchetti, D. (1996). Altered neuroendocrine
activity in maltreated children related to symptoms of depression. De-
velopment and Psychopathology, 8, 201–214.
Heim, C., Newport, D. J., Heit, S., Graham, Y. P., Wilcox, M., Bonsall, R.,
et al. (2000). Pituitary adrenal and autonomic responses to stress in
women after sexual and physical abuse in childhood. JAMA, 284, 592–
597.
Herbert, T. B., & Cohen, S. (1993). Depression and immunity: A meta-
analytic review. Psychological Bulletin, 113, 1–15.
Herrera, C., & Dunn, J. (1997). Early experiences with family conflict:
Implications for arguments with a close friend. Developmental Psychol-
ogy, 33, 869–881.
Hertsgaard, L. G., Gunnar, M. R., Erickson, M. R., & Nachmias, M.
(1995). Adrenocortical responses to the strange situation in infants with
disorganized/disoriented attachment relationships. Child Develop-
ment, 66, 1100–1106.
Higley, J. D., Hommer, D., Lucas, K., Shoaf, S., Shomi, S. J., & Linnoila,
M. (in press). CNS serotonin metabolism rate predicts innate tolerance,
high alcohol consumption, and aggression during intoxication in rhesus
monkeys. Archives of General Psychiatry.
Higley, J. D., Mehlman, P. T., Poland, R. E., Taub, D. M., Vickers, J.,
Suomi, S. J., & Linnoila, M. (1996). CSF testosterone and 5-HIAA
correlate with different types of aggressive behaviors. Biological Psy-
chiatry, 40, 1067–1082.
Higley, J. D., Suomi, S. J., & Linnoila, M. (1996a). A nonhuman primate
model of type II excessive alcohol consumption? Part 1. Low cerebro-
spinal fluid 5-hydroxyindoleacetic acid concentrations and diminished
social competence correlate with excessive alcohol consumption. Alco-
holism: Clinical and Experimental Research, 20, 629–642.
Higley, J. D., Suomi, S. J., & Linnoila, M. (1996b). A nonhuman pri-
mate model of type II alcoholism? Part 2. Diminished social competency
and excessive aggression correlates with low cerebrospinal fluid
5-hydroxyindoleacetic acid concentrations. Alcoholism: Clinical and
Experimental Research, 20, 643–650.
House, J. S., Umberson, D., & Landis, K. R. (1988). Structures and
processes of social support. American Review of Sociology, 14, 293–318.
Houston, B. K., & Vavak, C. R. (1991). Cynical hostility: Developmental
factors, psychological correlates, and health behaviors. Health Psychol-
ogy, 10, 9–17.
Hubbard, J. A., & Cole, J. D. (1994). Emotional correlates of social
competence in children’s peer relationships. Merrill-Palmer Quar-
terly, 40, 1–20.
Hymel, S., Rubin, K. H., Rowden, L., & Le Mare, L. (1990). Children’s
peer relationships: Longitudinal prediction of internalizing and external-
izing problems from middle to late childhood. Child Development, 61,
2004–2021.
Jaffe, J. H. (1990). Drug addiction and drug abuse. In A. G. Gilman, T. W.
Rall, A. S. Nies, & P. Taylor (Eds.), Goodman and Gilman’s: The
pharmacological basis of therapeutics (8th ed., pp. 522–573). New
York: McGraw-Hill.
Jemmott, L. S., & Jemmott III, J. B. (1992). Family structure, parental
strictness, and sexual behavior among inner-city black male adolescents.
Journal of Adolescent Research, 7, 192–207.
Johnson, J. G., Cohen, P., Kasen, S., Smailes, E., & Brook, J. S. (2001).
Association of maladaptive parental behavior with psychiatric disorder
among parents and their offspring. Archives of General Psychiatry, 58,
453–460.
Johnson, V., & Pandina, R. J. (1991). Effects of family environment on
adolescent substance use, delinquency, and coping styles. American
Journal of Drug and Alcohol Abuse, 17, 71–88.
Jorgensen, R. S., Johnson, B. T., Kolodziej, M. E., & Schreer, G. E. (1996).
Elevated blood pressure and personality: A meta-analytic review. Psy-
chological Bulletin, 120, 293–320.
Julkunen, J., Salonen, R., Kaplan, G. A., Chesney, M. A., & Salonen, J. T.
(1994). Hostility and the progression of carotid artherosclerosis. Psy-
chosomatic Medicine, 56, 519–525.
Kandel, D., & Yamaguchi, K. (1993). From beer to crack: Developmental
patterns of drug involvement. American Journal of Public Health, 83,
851–855.
Kaslow, M. H., Deering, C. G., & Racusia, G. R. (1994). Depressed
children and their families. Clinical Psychological Review, 14, 39–59.
Katz, L. F., & Gottman, J. M. (1995). Vagal tone protects children from
marital conflict. Development and Psychopathology, 7, 83–92.
Kaufman, J., Birmaher, B., Perel, J., Stull, S., Brent, D., Trubnick, L., et al.
(1998). Serotonergic functioning in depressed abused children: Clinical
and familial correlates. Biological Psychiatry, 44, 973–981.
Kaufman, J., Plotsky, P. M., Nemeroff, C. B., & Charney, D. S. (2000).
Effects of early adverse experiences on brain structure and function:
Clinical implications. Society of Biological Psychiatry, 48, 778–790.
Kawachi, I., Sparrow, D., Vokonas, P. S., & Weiss, S. T. (1994). Symp-
toms of anxiety and risk of coronary heart disease: The Normative Aging
Study. Circulation, 90, 2225–2229.
Kazdin, A. E. (1994). Psychotherapy for children and adolescents. In A. E.
Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behav-
ior change (4th ed., pp. 543–594). New York: Wiley.
Kehoe, P., Shoemaker, W., Arons, C., Triano, L., & Suresh, G. (1998).
Repeated isolation stress in the neonatal rat: Relation to brain dopamine
systems in the 10-day-old rat. Behavioral Neuroscience, 112, 1466–
1474.
Kerns, K. A., Klepac, L., & Cole, A. K. (1996). Peer relationships and
362
REPETTI, TAYLOR, AND SEEMAN
preadolescents’ perceptions of security in the child–mother relationship.
Developmental Psychology, 32, 457–466.
Kessler, R. C., & Magee, W. J. (1994). Childhood family violence and
adult recurrent depression. Journal of Health and Social Behavior, 35,
13–27.
Klohnen, E. C., & Bera, S. (1998). Behavioral and experiential patterns of
avoidantly and securely attached women across adulthood: A 31-year
longitudinal perspective. Journal of Personality and Social Psychol-
ogy, 74, 211–223.
Koob, G. F., Sanna, P. P., & Bloom, F. E. (1998). Neuroscience of
addiction. Neuron, 21, 467–476.
Kostrzewa, R. M., Reader, T. A., & Descarries, L. (1998). Serotonin neural
adaptations to ontogenetic loss of dopamine neurons in rat brain. Journal
of Neurochemistry, 70, 889–898.
Kraemer, G., & Clarke, A. (1990). The behavioral neurobiology of self
injurious behavior in rhesus monkeys. Progress in Neuro-Psycho-
pharmacology and Biological Psychiatry, 14, 5141–5168.
Krantz, D. S., & Manuck, S. B. (1984). Acute psychophysiologic reactivity
and risk of cardiovascular disease: A review and methodologic critique.
Psychological Bulletin, 96, 435–464.
Krevans, J., & Gibbs, J. C. (1996). Parents’ use of inductive discipline:
Relations to children’s empathy and prosocial behavior. Child Develop-
ment, 67, 3263–3277.
Kristal-Boneh, E., Raifel, M., Froom, P., & Rivak, J. (1995). Heart rate
variability in health and disease. Scandinavian Journal of Work and
Environmental Health, 21, 85–95.
Kubzansky, L. D., Kawachi, I., Weiss, S. T., & Sparrow, D. (1998).
Anxiety and coronary heart disease: A synthesis of epidemiological,
psychological, and experimental evidence. Annals of Behavioral Medi-
cine, 20, 47–58.
Kupersmidt, J. B., & Patterson, C. J. (1991). Childhood peer rejection,
aggression, withdrawal, and perceived competence as predictors of
self-reported behavior problems in preadolescence. Journal of Abnormal
Child Psychology, 19, 427–449.
Laible, D. J., & Thompson, R. A. (1998). Attachment and emotional
understanding in preschool children. Developmental Psychology, 34,
1038–1045.
Laird, R. D., Pettit, G. S., Mize, J., Brown, E. G., & Lindsey, E. (1994).
Mother–child conversations about peers contributions to competence.
Family Relations, 43, 425–432.
Landry, S. H., Smith, K. E., Miller-Loncar, C. L., & Swank, P. R. (1998).
The relation of change in maternal interactive styles to the developing
social competence of full-term and preterm children. Child Develop-
ment, 69, 105–123.
Larose, S., & Boivin, M. (1998). Attachment to parents, social support
expectations, and socioemotional adjustment during the high school–
college transition. Journal of Research on Adolescence, 8, 1–27.
Lawler, K. A., Allen, M. T., Critcher, E. C., & Standard, B. A. (1981). The
relationship of physiological responses to the coronary-prone behavior
pattern in children. Journal of Behavioral Medicine, 4, 203–216.
Leserman, J., Toomey, T. C., & Drossman, D. A. (1995). Medical conse-
quences of sexual and physical abuse in women. Humane Medicine, 11,
23–28.
Leventhal, T., & Brooks-Gunn, J. (2000). The neighborhoods they live in:
The effects of neighborhood residence on child and adolescent out-
comes. Psychological Bulletin, 126, 309–337.
Levine, S., & Wiener, S. G. (1988). Psychoendocrine aspects of mother–
infant relationships in nonhuman primates. Psychoneuroimmunol-
ogy, 13, 143–154.
Lewis, M. H., Gluck, J. P., Beauchamp, A. J., Keresztury, M. F., Michael,
F., et al. (1990). Long-term effects of early social isolation in Macaca
mulatta: Changes in dopamine receptor function following apomorphine
challenge. Brain Research, 513, 67–73.
Lindsey, E. W., Mize, J., & Pettit, G. S. (1997). Mutuality in parent–child
play: Consequences for children’s peer competence. Journal of Social
and Personal Relationships, 14, 523–538.
Linnoila, M., DeJong, J., & Virkkunen, M. (1989). Family history of
alcoholism in violent offenders and impulsive fire setters. Archives of
General Psychiatry, 46, 613–616.
Lissau, I., & Sorensen, T. I. A. (1994). Parental neglect during childhood
and increased risk of obesity in young adulthood. Lancet, 343, 324–327.
Liu, D., Dorio, J., Tannenbaum, B., Caldji, C., Francis, D., Freedman, A.,
et al. (1997). Maternal care, hippocampal glucocorticoid receptors, and
hypothalamic–pituitary–adrenal responses to stress. Science, 277, 1659–
1662.
Luecken, L. J. (1998). Childhood attachment and loss experiences affect
adult cardiovascular and cortisol function. Psychosomatic Medicine, 60,
765–772.
Lundberg, O. (1993). The impact of childhood living conditions on illness
and mortality in adulthood. Social Science Medicine, 36, 1047–1052.
Lundberg, U. (1983). Note on Type A behavior and cardiovascular re-
sponse to challenge in 3- to 6-year-old children. Journal of Psychoso-
matic Research, 27, 39–42.
MacKinnon-Lewis, C., Starnes, R., Volling, B., & Johnson, S. (1997).
Perceptions of parenting as predictors of boys’ sibling and peer relations.
Developmental Psychology, 33, 1024–1031.
Malinosky-Rummell, R., & Hansen, D. J. (1993). Long-term consequences
of childhood physical abuse. Psychological Bulletin, 114, 68–79.
Martin, L. R., Friedman, H. S., Tucker, J. S., Schwartz, J. E., Criqui, M. H.,
Wingard, D. L., & Tomlinson-Keasy, C. (1995). An archival prospective
study of mental health and longevity. Health Psychology, 14, 381–387.
Martin, M. T., Miller-Johnson, S., Kitzmann, K. M., & Emery, R. E.
(1998). Parent–child relationships and insulin-dependent diabetes mel-
litus: Observational ratings of clinically relevant dimensions. Journal of
Family Psychology, 12, 102–111.
Masten, A. S., & Coatsworth, J. D. (1998). The development of compe-
tence in favorable and unfavorable environments: Lessons from research
on successful children. American Psychologist, 53, 205–220.
Masten, A., Morison, P., Pelligrini, D., & Tellegen, A. (1990). Competence
under stress: Risk and protective factors. In J. Rolf, A. S. Masten, D.
Cicchetti, K. Neuchterlein, & S. Weintraub (Eds.), Risk and protective
factors in the development of psychopathology (pp. 236–256). Cam-
bridge, England: Cambridge University Press.
Matthews, K. A., & Jennings, J. R. (1984). Cardiovascular responses of
boys exhibiting the Type A behavior pattern. Psychosomatic Medi-
cine, 46, 484–496.
Matthews, K. A., Woodall, K. L., Kenon, K., & Jacob, T. (1996). Negative
family environment as a predictor of boys’ future status on measures of
hostile attitudes, interview behavior, and anger expression. Health Psy-
chology, 15, 30–37.
Mayer, J. D., Salovey, P., Caruso, D. L., & Sitarenios, G. (2001). Emo-
tional intelligence as a standard intelligence. Emotion, 1, 232–242.
Mayne, T. J., & Buck, R. W. (1997). Sex, emotion, and the triune brain.
The Health Psychologist, 19, 8–23.
McEwen, B. S. (1998). Protective and damaging effects of stress media-
tors. New England Journal of Medicine, 338, 171–179.
McEwen, B. S., & Stellar, E. (1993). Stress and the individual: Mecha-
nisms leading to disease. Archives of Internal Medicine, 153, 2093–
2101.
McLeod, J. D., & Shanahan, M. J. (1996). Trajectories of poverty and
children’s mental health. Journal of Health and Social Behavior, 37,
207–220.
McLoyd, V. C. (1998). Socioeconomic disadvantage and child develop-
ment. American Psychologist, 53, 185–204.
Meaney, M. J., Aitken, D. H., van Berkel, C. H., Bhatnagar, S., &
Sapolsky, R. M. (1988, February 12). Effect of neonatal handling on
age-related impairments associated with the hippocampus. Science, 239,
766–768.
363
RISKY FAMILIES
Mechanic, D., & Hansell, S. (1989). Divorce, family conflict, and adoles-
cents’ well-being. Journal of Health and Social Behavior, 30, 105–116.
Mehlman, P. T., Higley, J. D., Faucher, I., Lilly, A. A., Taub, D. M.,
Vickers, J., et al. (1994). Low CSF 5-HIAA concentrations and severe
aggressive and impaired impulse control in nonhuman primates. Amer-
ican Journal of Psychiatry, 151, 1485–1491.
Mendoza, S. P., Smotherman, W. P., Miner, M., Kaplan, J., & Levine, S.
(1978). Pituitary–adrenal response to separation in mother and infant
squirrel monkeys. Developmental Psychobiology, 11, 169–175.
Metzler, C. W., Noell, J., Biglan, A., Ary, D., & Smolkowski, K. (1994).
The social context for risky sexual behavior among adolescents. Journal
of Behavioral Medicine, 17, 419–438.
Miller, K. S., Forehand, R., & Kotchick, B. A. (1999). Adolescent sexual
behavior in two ethnic minority samples: The role of family variables.
Journal of Marriage and the Family, 61, 85–98.
Millstein, S. G., & Moscicki, A.-B. (1995). Sexually-transmitted disease in
female adolescents: Effects of psychosocial factors and high risk behav-
iors. Journal of Adolescent Health, 17, 83–90.
Mize, J., & Pettit, G. S. (1997). Mothers’ social coaching, mother–child
relationship style, and children’s peer competence: Is the medium the
message? Child Development, 68, 312–332.
Montgomery, S. M., Bartley, M. J., & Wilkinson, R. G. (1997). Family
conflict and slow growth. Archives of Disease in Childhood, 77, 326–
330.
Mounts, N. S., & Steinberg, L. (1995). An ecological analysis of peer
influence on adolescent grade point average and drug use. Developmen-
tal Psychology, 31, 915–922.
Nachmias, M., Gunnar, M., Mangelsdorf, S., Parritz, R. H., & Buss, K.
(1996). Behavioral inhibition and stress reactivity: The moderating role
of attachment security. Child Development, 67, 508–522.
Newcomb, M. D., Maddahian, E., & Bentler, P. M. (1986). Risk factors for
drug use among adolescents: Concurrent and longitudinal analyses.
American Journal of Public Health, 76, 525–531.
Nolen-Hoeksema, S., Wolfson, A., Mumme, D., & Guskin, K. (1995).
Helplessness in children of depressed and nondepressed mothers. De-
velopmental Psychology, 31, 377–387.
O’Brien, M., Margolin, G., John, R. S., & Krueger, L. (1991). Mothers’
and sons’ cognitive and emotional reactions to simulated marital and
family conflict. Journal of Consulting and Clinical Psychology, 59,
692–703.
O’Connor, T. G., Deater-Deckard, K., Fulker, D., Rutter, M., & Plomin, R.
(1998). Genotype–environment correlations in late childhood and early
adolescence: Antisocial behavioral problems and coercive parenting.
Developmental Psychology, 34, 970–981.
Pakiz, B., Reinherz, H. Z., & Giaconia, R. M. (1997). Early risk factors for
serious antisocial behavior at age 21: A longitudinal community study.
American Journal of Orthopsychiatry, 67, 92–110.
Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal
adjustment: Are low-accepted children at risk? Psychological Bulletin,
102, 335–389.
Perry, B. D., & Pollard, R. (1998). Homeostasis, stress, trauma, and
adaptation. Stress in Children, 7, 33–51.
Pettit, G. S., Bates, J. E., & Dodge, K. A. (1997). Supportive parenting,
ecological context, and children’s adjustment: A seven-year longitudinal
study. Child Development, 68, 908–923.
Pettit, G. S., Dodge, K. A., & Brown, M. M. (1988). Early family
experience, social problem solving patterns, and children’s social com-
petence. Child Development, 59, 107–120.
Petty, F., Davis, L. L., Kabel, D., & Kramer, G. L. (1996). Serotonin
dysfunction disorders: A behavioral neurochemistry perspective. Jour-
nal of Clinical Psychiatry, 57, (Suppl. 8), 11–17.
Pine, D. S., Coplan, J. D., Wasserman, G., Miller, L. S., Fried, J. A.,
Davies, M., et al. (1997). Neuroendocrine response to d,1-fenfluramine
challenge in boys: Associations with aggressive behavior and adverse
rearing. Archives of General Psychiatry, 54, 839–846.
Pine, D. S., Shaffer, D., Davies, M., & Schonfeld, I. S. (1997). Minor
physical anomalies: Moderators of environmental risk factors for psy-
chopathology? Journal of the American Academy of Child and Adoles-
cent Psychiatry, 36, 395–403.
Pine, D. S., Wasserman, G., Coplan, J., Fried, J. A., Huang, Y., Kassir, S.,
et al. (1996). Associations between platelet serotonin-2A receptor char-
acteristics and parenting factors for boys at risk for delinquency. Amer-
ican Journal of Psychiatry, 153, 538–544.
Plomin, R. (1994). Genetics and experience. Newbury Park, CA: Sage.
Power, C., & Hertzman, C. (1997). Social and biological pathways linking
early life and adult disease. British Medical Bulletin, 53, 210–221.
Putallaz, M. (1987). Maternal behavior and children’s sociometric status.
Child Development, 58, 324–340.
Raikkonen, K., Keltikangas-Jarvinen, L., Hautanen, A., & Adlercreutz, H.
(1997). Neuroendocrine mechanisms in chronic perceived stress: Asso-
ciations with the metabolic syndrome. Endocrinology and Metabo-
lism, 4, 247–254.
Raleigh, M. J., Brammer, G. L., Ritvo, E. R., Geller, E., McGuire, M. T.,
& Yuwiler, A. (1986). Effects of chronic fenfluramine on blood sero-
tonin, cerbrospinal fluid metabolites, and behavior in monkeys. Psycho-
pharmacology, 90, 503–508.
Raleigh, M. J., & McGuire, M. T. (1994). Serotonin, aggression and
violence in vervet monkeys. In R. D. Masters, & M. D. McGuire (Eds.),
The neurotransmitter revolution (pp. 129–145). Carbondale: Southern
Illinois University Press.
Reid, J., Macchetto, P., & Foster, S. (1999). No safe haven: Children of
substance-abusing parents. New York: Center on Addiction and Sub-
stance Abuse.
Reid, R. J., & Crisafulli, A. (1990). Marital discord and child behavior
problems: A meta-analysis. Journal of Abnormal Child Psychology, 18,
105–117.
Reiss, D., Hetherington, M., Plomin, R., Howe, G. W., Simmens, S. J.,
Henderson, S. H., et al. (1995). Genetic questions for environmental
studies: Differential parenting and psychopathology in adolescence.
Archives of General Psychiatry, 52, 925–936.
Repetti, R. L., & Wood, J. (1997). Families accommodating to chronic
stress: Unintended and unnoticed processes. In B. H. Gottlieb (Ed.),
Coping with chronic stress (pp. 191–220). New York: Plenum.
Resnick, M. D., Bearman, P. S., Blum, R. W., Bauman, K. E., Harris,
K. M., Jones, J., et al. (1997). Protecting adolescents from harm: Find-
ings from the National Longitudinal Study on Adolescent Health. JAMA,
278, 823–832.
Ribeiro, E. B., Bettiker, R. L., Bogdanov, M., & Wurtman, R. J. (1993).
Effects of systemic nicotine on serotonin release in rat brain. Brain
Research, 621, 311–318.
Risch, S. C. (1997). Recent advances in depression research: From stress to
molecular biology and brain imaging. Journal of Clinical Psychiatry, 58
(Suppl. 5), 3–6.
Robbins, T. W., Jones, G. H., & Wilkinson, L. S. (1996). Behavioral and
neurochemical effects of early social deprivation in the rat. Journal of
Psychopharmacology, 10, 39–47.
Romer, D., Black, M., Ricardo, I., Feigelman, S., Kaljee, L., Galbraith, J.,
et al. (1994). Social influences on the sexual behavior of youth at risk for
HIV exposure. American Journal of Public Health, 84, 977–985.
Rook, K. S. (1984). The negative side of social interaction: Impact of
psychological well-being. Journal of Personality and Social Psychol-
ogy, 46, 1097–1108.
Rosen, J. B. & Schulkin, J. (1998). From normal fear to pathological
anxiety. Psychological Review, 105, 325–360.
Rosenblum, L., Coplan, J., Friedman, S., Bassoff, T., Gorman, J., &
Andrews, M. (1994). Adverse early experiences affect noradrenergic and
364
REPETTI, TAYLOR, AND SEEMAN
serotonergic functioning in adult primates. Biological Psychiatry, 35,
221–227.
Rothbaum, F., & Weisz, J. R. (1994). Parental caregiving and child
externalizing behavior in nonclinical samples: A meta-analysis. Psycho-
logical Bulletin, 116, 55–74.
Russek, L. G., & Schwartz, G. E. (1997). Feelings of parental caring can
predict health status in mid-life: A 35-year follow-up of the Harvard
Mastery of Stress study. Journal of Behavioral Medicine, 20, 1–13.
Rutter, M. (1990). Psychosocial resilience and protective mechanisms. In J.
Rolf, A. S. Masten, D. Cicchetti, K. Neuchterlein, & S. Weintraub
(Eds.), Risk and protective factors in the development of psychopathol-
ogy (pp. 181–215). Cambridge, England: Cambridge University Press.
Sayger, T. V., Horne, A. M., Walker, J. M., & Passmore, J. L. (1988).
Social learning family therapy with aggressive children: Treatment
outcome and maintenance. Journal of Family Psychology, 1, 261–285.
Scaramella, L. V., Conger, R. D., Simons, R. L., & Whitbeck, L. B. (1998).
Predicting risk for pregnancy by late adolescence: A social contextual
perspective. Developmental Psychology, 34, 1233–1245.
Schleifer, S. J., Scott, B., Stern, M., & Keller, S. E. (1986). Behavioral and
developmental aspects of immunity. Journal of the American Academy
of Child Psychiatry, 16, 751–763.
Schmidt, S. E., Liddle, H. A., & Dakof, G. A. (1996). Changes in parenting
practices and adolescent drug abuse during multidimensional family
therapy. Journal of Family Psychology, 10, 12–27.
Schmidt, T. H., Thierse, H., & Eschweiler, J. (1986). Behavioral correlates
of cardiovascular reactivity in school children. In T. H. Schmidt, T. M.
Dembroski, & G. Blumchen (Eds.), Biological and psychological factors
in cardiovascular disease (pp. 187–227). New York: Springer-Verlag.
Schwartz, D., Dodge, K. A., Pettit, G. S., & Bates, J. E. (1997). The early
socialization of aggressive victims of bullying. Child Development, 68,
665–675.
Sedlak, A. J., & Broadhurst, D. D. (1996). Third national incidence study
on child abuse and neglect. Washington, DC: U.S. Department of Health
and Human Services.
Seeman, T. E. (1996). Social ties and health. Annals of Epidemiology, 6,
442–451.
Seeman, T. E., & Robbins, R. J. (1994). Aging and hypothalamic–
pituitary–adrenal response to challenge in humans. Endocrinology Re-
view, 15, 233–260.
Seeman, T. E., Singer, B., Horwitz, R., & McEwen, B. S. (1997). The price
of adaptation–allostatic load and its health consequences: MacArthur
studies of successful aging. Archives of Internal Medicine, 157, 2259–
2268.
Serketich, W. J., & Dumas, J. E. (1996). The effectiveness of behavioral
parent training to modify antisocial behavior in children: A meta-
analysis. Behavior Therapy, 27, 171–186.
Shaffer, J. W., Duszynski, K. R., & Thomas, C. B. (1982). Family attitudes
in youth as a possible precursor of cancer among physicians: A search
for explanatory mechanisms. Journal of Behavioral Medicine, 5, 143–
163.
Shedler, J., & Block, J. (1990). Adolescent drug use and psychological
health: A longitudinal inquiry. American Psychologist, 45, 612–630.
Shiffman, S., Fischer, L. A., Paty, J. A., Gnys, M., Hickcox, M., & Kassel,
J. D. (1994). Drinking and smoking: A field study of their association.
Annals of Behavioral Medicine, 16, 203–209.
Siegman, A. W., Dembroski, T. M., & Crump, D. (1992). Speech rate,
loudness, and cardiovascular reactivity. Journal of Behavioral Medi-
cine, 15, 519–532.
Sloan, R. P., Shapiro, P. A., Bigger, T., Jr., Bagiella, E., Steinman, R. C.,
& Gorman, J. M. (1994). Cardiac autonomic control and hostility in
health subjects. American Journal of Cardiology, 74, 298–300.
Small, S. A., & Luster, T. (1994). Adolescent sexual activity: An ecolog-
ical, risk-factor approach. Journal of Marriage and the Family, 56,
181–192.
Smith, T. W. (1992). Hostility and health: Current status of a psychoso-
matic hypothesis. Journal of Personality and Social Psychology, 48,
813–838.
Smith, T. W., Pope, M. K., Sanders, J. D., Allred, K. D., & O’Keefe, J. L.
(1988). Cynical hostility at home and work: Psychosocial vulnerability
across domains. Journal of Research in Personality, 22, 525–548.
Southam-Gerow, M. A., & Kendall, P. C. (2002). Emotion regulation and
understanding: Implications for child psychopathology and therapy.
Clinical Psychology Review, 22, 189–222.
Spoth, R., Redmond, C., Hockaday, C., & Yoo, S. (1996). Protective
factors and young adolescent tendency to abstain from alcohol use: A
model using two waves of intervention study data. American Journal of
Community Psychology, 24, 749–770.
Springs, F. E., & Friedrich, W. N. (1992). Health risk behaviors and
medical sequelae of childhood sexual abuse. Mayo Clinic Proceed-
ings, 67, 527–532.
Stahl, S. M. (1996). Essential psychopharmacology: Neuroscientific basis
and practical applications. Cambridge, MA: Cambridge University
Press.
Stanton, M. D., & Shadish, W. R. (1997). Outcome, attrition, and family–
couples treatment for drug abuse: A meta-analysis and review of the
controlled, comparative studies. Psychological Bulletin, 122, 170–191.
Stattin, H., & Kerr, M. (2000). Parental monitoring: A reinterpretation.
Child Development, 71, 1072–1085.
Stein, A., Woolley, H., Cooper, S. D., & Fairburn, C. G. (1994). An
observational study of mothers with eating disorders and their infants.
Journal of Child Psychology and Psychiatry, 35, 733–748.
Steinberg, L. (1987). The impact of puberty on family relations: Effects of
pubertal status and pubertal timing. Developmental Psychology, 23,
451–460.
Steinberg, L., Lamborn, S. D., Darling, N., Mounts, N. S., & Dornbusch,
S. M. (1994). Over-time changes in adjustment and competence among
adolescents from authoritative, authoritarian, indulgent, and neglectful
families. Child Development, 65, 754–770.
Steinberg, L., Lamborn, S. D., Dornbusch, S. M., & Darling, N. (1992).
Impact of parenting practices on adolescent achievement: Authoritative
parenting, school involvement, and encouragement to succeed. Child
Development, 63, 1266–1281.
Stern, M., & Zevon, M. A. (1990). Stress, coping, and family environment:
The adolescent’s response to naturally occurring stressors. Journal of
Adolescent Research, 5, 290–305.
Stormshak, E. A., Bellanti, C. J., Bierman, K. L., Coie, J. D., Dodge, K. A.,
Greenberg, M. T., et al. (1996). The quality of sibling relationships and
the development of social competence and behavioral control in aggres-
sive children. Developmental Psychology, 32, 79–89.
Suarez, E. C., Bates, M. P., & Harralson, T. L. (1998). The relation of
hostility to lipids and lipoproteins in women: Evidence for the role of
antagonistic hostility. Annals of Behavioral Medicine, 20, 59–63.
Suomi, S. J. (1987). Genetic and maternal contributions to individual
differences in rhesus monkey biobehavioral development. In N. A.
Krasnagor, E. M. Blass, M. A. Hofer, & W. P. Smotherman (Eds.),
Perinatal development: A psychobiological perspective (pp. 397–420).
New York: Academic Press.
Suomi, S. J. (1991). Up-tight and laid-back monkeys: Individual differ-
ences in the response to social challenges. In S. Brauth, W. Hall, & R.
Dooling (Eds.), Plasticity of development (pp. 27–56). Cambridge, MA:
MIT Press.
Suomi, S. J. (1999). Attachment in rhesus monkeys. In J. Cassidy & P.
Shaver (Eds.), Handbook of attachment: Theory, research, and clinical
applications (pp. 181–197). New York: Guilford Press.
Suomi, S. J. (1997). Early determinants of behaviour: Evidence from
primate studies. British Medical Bulletin, 53, 170–184.
Suomi, S. J., & Levine, S. (1998). Psychobiology of intergenerational
effects of trauma: Evidence from animal studies. In Y. Danieli (Ed.),
365
RISKY FAMILIES
Intergenerational handbook of multigenerational legacies (pp. 623–
637). New York: Plenum Press.
Taylor, S. E. (1999). Health psychology (4th ed.). New York: McGraw-
Hill.
Taylor, S. E., Repetti, R. L., & Seeman, T. E. (1997). Health psychology:
What is an unhealthy environment and how does it get under the skin?
Annual Review of Psychology, 48, 411–447.
Thompson, R. A., & Calkins, S. D. (1996). The double-edged sword:
Emotional regulation for children at risk. Development and Psychopa-
thology, 8, 163–182.
Tubman, J. G., Windle, M., & Windle, R. C. (1996). The onset and
cross-temporal patterning of sexual intercourse in middle adolescence:
Prospective relations with behavioral and emotional problems. Child
Development, 67, 327–343.
Tucker, J. S., Friedman, H. S., Schwartz, J. E., Criqui, M. H., Tomlinson-
Keasey, C., Wingard, D. L., et al. (1997). Parental divorce: Effects on
individual behavior and longevity. Journal of Personality and Social
Psychology, 73, 381–391.
Turner, R. A., Irwin Jr., C. E., Tschann, J. M., & Millstein, S. G. (1993).
Autonomy, relatedness, and the initiation of health risk behaviors in
early adolescence. Health Psychology, 12, 200–208.
Uchino, B. N., Uno, D., & Holt-Lunstad, J. (1999). Social support, phys-
iological processes and health. Current Directions in Psychological
Science, 8, 145–148.
U.S. Department of Justice. (1994). Violence against women: A national
crime victimization survey report (Bureau of Justice Statistics Selected
Findings, NCJ-145325). Washington, DC: U.S. Government Printing
Office.
Valentiner, D. P., Holahan, C. J., & Moos, R. H. (1994). Social support,
appraisals of event controllability, and coping: An integrative model.
Journal of Personality and Social Psychology, 66, 1094–1102.
Valenzuela, C. F., & Harris, R. A. (1997). Alcohol: Neurobiology. In J. H.
Lowinson, P. Ruiz, R. B. Millman, & J. G. Langrod (Eds.), Substance
abuse: A comprehensive textbook (7th ed., pp. 119–142). Baltimore:
Williams & Wilkins.
Valenzuela, M. (1997). Maternal sensitivity in a developing society: The
context of urban poverty and infant chronic undernutrition. Develop-
mental Psychology, 33, 845–855.
van den Boom, D. C. (1994). The influence of temperament and mothering
on attachment and exploration: An experimental manipulation of sensi-
tive responsiveness among lower-class mothers with irritable infants.
Child Development, 65, 1457–1477.
van IJzendoorn, M. H., & Bakermans-Kranenburg, M. J. (1996). Attach-
ment representations in mothers, fathers, adolescents, and clinical
groups: A meta-analytic search for normative data. Journal of Consult-
ing and Clinical Psychology, 64, 8–21.
Virkkunen, M., & Linnoila, M. (1993). Brain serotonin, type II alcoholism
and impulsive violence. Journal of Studies in Alcohol, 11, 163–169.
Wadsworth, M. E. J., MacLean, M., Kuh, D., & Rodgers, B. (1990).
Children of divorced and separated parents: Summary and review of
findings from a long-term follow-up study in the U.K. Family Prac-
tice, 7, 104–109.
Wagner, B. M. (1997). Family risk factors for child and adolescent suicidal
behavior. Psychological Bulletin, 121, 246–298.
Wahler, R. G. (1990). Some perceptual functions of social networks in
coercive mother–child interactions. Journal of Social and Clinical Psy-
chology, 9, 43–53.
Walker, E. A., Gelfand, A., Katon, W. J., Koss, M. P., Von Korff, M.,
Bernstein, D., & Russo, J. (1999). Adult health status of women with
histories of childhood abuse and neglect. The American Journal of
Medicine, 107, 332–339.
Waters, E., Vaughn, B. E., Posada, G., & Kondo-Ikemura, K. (1995).
Caregiving, cultural, and cognitive perspectives on secure-base behavior
and working models: New growing points of attachment theory and
research. Monographs of the Society for Research in Child Develop-
ment, 60, (2–3, Serial No. 244).
Webster, P. S., Orbuch, T. L., & House, J. S. (1995). Effects of childhood
family background on adult marital quality and perceived stability.
American Journal of Sociology, 101, 404–432.
Weidner, G., Hutt, J., Connor, S. L., & Mendell, N. R. (1992). Family
stress and coronary risk in children. Psychosomatic Medicine, 54, 471–
479.
Weiss, J. M. (1991). Stress-induced depression: Critical neurochemical and
electrophysiological changes. In J. Madden IV (Ed.), Neurobiology of
learning, emotion and affect (pp. 123–154). New York: Raven Press.
Weiss, J. M., Bailey, W. H., Pohorecky, L. A., Korzeniowski, D., &
Grillione, G. (1980). Stress-induced depression of motor activity corre-
lates with regional changes in brain norepinephrine but not in dopamine.
Neurochemical Research, 5, 9–22.
Werner, E., & Smith, R. (1982). Vulnerable but invincible: A study of
resilient children. New York: McGraw-Hill.
Wickrama, K. A. S., Lorenz, F. O., & Conger, R. D. (1997). Parental
support and adolescent physical health status: A latent growth-curve
analysis. Journal of Health and Social Behavior, 38, 149–163.
Widom, C. S., & White, H. R. (1997). Problem behaviours in abused and
neglected children grown up: Prevalence and co-occurrence of substance
abuse, crime, and violence. Criminal Behaviour and Mental Health, 7,
287–310.
Williams, D. R., & Collins, C. (1995). U.S. socioeconomic and racial
differences in health: Patterns and explanations. Annual Review of So-
ciology, 21, 349–386.
Wills, T. A., & Cleary, S. D. (1996). How are social support effects
mediated? A test with parental support and adolescent substance use.
Journal of Personality and Social Psychology, 71, 937–952.
Wills, T. A., DuHamel, K., & Vaccaro, D. (1995). Activity and mood
temperament as predictors of adolescent substance use: Test of a self-
regulation mediational model. Journal of Personality and Social Psy-
chology, 62, 901–916.
Woodall, K. L., & Matthews, K. A. (1989). Familial environment associ-
ated with Type A behaviors and psychophysiological responses to stress
in children. Health Psychology, 8, 403–426.
Woodall, K. L., & Matthews, K. A. (1993). Changes in and stability of
hostile characteristics: Results from a 4-year longitudinal study of chil-
dren. Journal of Personality and Social Psychology, 64, 491–499.
Zahn-Waxler, C., Iannotti, F. J., Cummings, E. M., & Denham, S. (1990).
Antecedents of problem behaviors in children of depressed mothers.
Development and Psychopathology, 2, 271–291.
Zaslow, M. J., & Eldred, C. A. (Eds.). (1998). Parenting behavior in a
sample of young mothers in poverty: Results of the New Chance Obser-
vational Study. New York: Manpower Demonstration Research.
Zierler, S., Feingold, L., Laufer, D. Velentgas, P., Kantrowitz-Gordon, I.,
& Mayer, K. (1991). Adult survivors of childhood sexual abuse and
subsequent risk of HIV infection. American Journal of Public
Health, 81, 572–575.
Received August 12, 1999
Revision received June 12, 2001
Accepted July 2, 2001 䡲
366
REPETTI, TAYLOR, AND SEEMAN
