www.thelancet.com Vol 373 January 17, 2009
Child Maltreatment 3
Interventions to prevent child maltreatment and associated
Harriet L MacMillan, C Nadine Wathen, Jane Barlow, David M Fergusson, John M Leventhal, Heather N Taussig
Although a broad range of programmes for prevention of child maltreatment exist, the eff ectiveness of most of the
programmes is unknown. Two specifi c home-visiting programmes—the Nurse–Family Partnership (best evidence) and
Early Start—have been shown to prevent child maltreatment and associated outcomes such as injuries. One population-
level parenting programme has shown benefi ts, but requires further assessment and replication. Additional in-hospital
and clinic strategies show promise in preventing physical abuse and neglect. However, whether school-based educational
programmes prevent child sexual abuse is unknown, and there are currently no known approaches to prevent emotional
abuse or exposure to intimate-partner violence. A specifi c parent-training programme has shown benefi ts in preventing
recurrence of physical abuse; no intervention has yet been shown to be eff ective in preventing recurrence of neglect. A
few interventions for neglected children and mother–child therapy for families with intimate-partner violence show
promise in improving behavioural outcomes. Cognitive-behavioural therapy for sexually abused children with symptoms
of post-traumatic stress shows the best evidence for reduction in mental-health conditions. For maltreated children,
foster care placement can lead to benefi ts compared with young people who remain at home or those who reunify from
foster care; enhanced foster care shows benefi ts for children. Future research should ensure that interventions are
assessed in controlled trials, using actual outcomes of maltreatment and associated health measures.
The fi rst paper of this Series summarised the nature
and consequences of child maltreatment.1 We review
here what is known about approaches to reduce the fi ve
major subtypes of child maltreatment: physical abuse,
sexual abuse, psychological abuse, neglect, and exposure
to intimate-partner violence, and the impairment
associated with these experiences. The framework we
follow (fi gure 1) addresses interventions aimed at
prevention of maltreatment before it occurs, including
both universal and targeted approaches (panel 1), and
prevention of recurrence and adverse outcomes
associated with maltreatment (panel 2). Eff orts to reduce
child maltreatment by improving the social, economic,
and political environments in which children and
families live is beyond the scope of this article; these
issues are discussed in the fourth paper in this Series.2
We highlight the relevant processes for designing and
evaluating interventions according to the public-health
model and as summarised in the 2006 WHO report3 on
preventing child maltreatment: defi ne and measure the
problem; identify causal, risk, and protective factors;
develop and determine eff ectiveness of interventions;
and implement interventions with ongoing monitoring
of outcomes. Too often, interventions are implemented
before undergoing adequate evaluation—the term
“promising” is sometimes misinterpreted as suffi cient
evidence for widespread dissemination.
When available, we have used good quality syntheses of
the literature on maltreatment prevention, ideally a
systematic review;4 when randomised controlled trials
(RCTs) exist, we have not included information from
cohort or case–control studies. We have provided more
Lancet 2009; 373: 250–66
December 3, 2008
See Comment page 195
This is the third in a Series of
four papers about child
Departments of Psychiatry and
Behavioural Neurosciences and
of Pediatrics, and Off ord Centre
for Child Studies, McMaster
University, Hamilton, ON,
Canada (Prof H L MacMillan MD);
Faculty of Information and
Media Studies, The University
of Western Ontario, London,
ON, Canada (C N Wathen PhD);
Warwick Medical School,
University of Warwick,
(Prof J Barlow DPhil);
Department of Psychological
Medicine, Christchurch School
of Medicine and Health
(Prof D M Fergusson PhD);
Department of Pediatrics, Yale
University School of Medicine,
New Haven, CT, USA
(Prof J M Leventhal MD); and
Kempe Center, Departments of
Pediatrics and Psychiatry,
University of Colorado Denver
School of Medicine, Denver, CO,
USA (H N Taussig PhD)
Prof Harriet L MacMillan, Off ord
Centre for Child Studies,
McMaster University, Patterson
Building, Chedoke Hospital,
Hamilton, ON L8N 3Z5, Canada
• Home-visiting programmes are not uniformly eff ective in
reducing child physical abuse, neglect, and outcomes such
as injuries; those that have shown benefi ts are the Nurse–
Family Partnership (best evidence) and Early Start
• The Triple P—Positive Parenting Program has shown positive
eff ects on maltreatment and associated outcomes, but
further assessment and replication are needed
• Hospital-based educational programmes to prevent
abusive head trauma and enhanced paediatric care for
families of children at risk of physical abuse and neglect
show promise but require further assessment
• School-based educational programmes improve children’s
knowledge and protective behaviours; whether they
prevent sexual abuse is unknown
• Parent–child interaction therapy has shown benefi ts in
preventing recurrence of child physical abuse; no
interventions have been shown eff ective in preventing
recidivism of neglect
• Preventing impairment associated with child
maltreatment requires a thorough assessment of the child
and family. Cognitive-behavioural therapy shows benefi ts
for sexually abused children with post-traumatic stress
symptoms. There is some evidence for child-focused
therapy for neglected children and for mother–child
therapy in families with intimate-partner violence
• For maltreated children, foster care placement can lead to
benefi ts compared with young people who remain at
home or those who reunify from foster care, and
enhanced foster care leads to better mental-health
outcomes for children than does traditional foster care
www.thelancet.com Vol 373 January 17, 2009 251
details of studies showing positive eff ects with higher
levels of evidence, or in areas where debate exists about
the eff ectiveness of an intervention.
Selection of outcomes is a crucial methodological issue.
Offi cial reports are thought to be the most objective
assessment of outcome, but represent only the tip of the
iceberg.5 Conversely, relying
self-reports of behaviour is problematic because of biases
due to social desirability and stigma. There is evidence
for the reliability and validity of children’s self-reports of
victimisation,6 but much maltreatment is experienced by
children too young for self-report. Where possible we
have reported objective measures of child and caregiver
behaviours and experiences of maltreatment, and have
not included studies that rely solely on parental
self-reports of abusive behaviour.
solely on caregiver
Prevention before occurrence of maltreatment
Physical abuse and neglect
Reduction of physical abuse and neglect is a combined
focus in many prevention programmes. We therefore
address them together, although they are distinct
subtypes of maltreatment, and can require diff erent
approaches to prevention. In a systematic overview,
Barlow and colleagues7 identifi ed eight systematic reviews
that examined a broad array of programmes aimed at
prevention of child physical abuse and neglect.
Programme quality varied: for example, less rigorous
reviews were not based on systematic searches or there
was inappropriate combination of results across all
interventions or outcomes. The authors concluded that
there is insuffi cient evidence of the eff ectiveness of
services in improving objective measures of abuse and
neglect, and evidence that some types of intervention (eg,
social support) are ineff ective. Home visitation and
multicomponent interventions were identifi ed as being
the only potentially eff ective interventions, although the
evidence across reviews was not uniform. The most
rigorous study of one home visitation programme
showed positive results8 and has since undergone a much
longer follow-up and two replications.
Home-visiting programmes vary widely in their models
of service delivery, content, and staffi ng.9 Although
universal home visiting for very young children and their
parents has existed for decades in many European
countries,9 much of the research has been done in the
USA on targeted programmes. This section will discuss
the evidence for prevention of physical abuse and neglect
and associated outcomes such as child hospitalisations,
emergency department visits, and injuries.
Despite the promotion of a broad range of early
childhood home-visiting programmes,10,11 most of these
have not been shown to reduce physical abuse and
neglect when assessed using RCTs.12 Some systematic
reviews, especially those including meta-analyses, have
concluded that early childhood home visitation is eff ective
in preventing child abuse and neglect13,14 without taking
into account the variability across programmes.15 Such
general statements obscure important diff erences in
design and methods, including outcomes, across
studies.16 Two programmes, the Nurse–Family Partnership
developed in the USA and the Early Start programme in
New Zealand have, however, shown signifi cant benefi ts.
The Nurse–Family Partnership has undergone the most
rigorous and extensive evaluation of child maltreatment
outcomes.17 It has been tested, with high rates of retention,
in three RCTs across a range of samples and US regions:
Elmira, NY (n=400, semi-rural; 89% white sample;
81% follow-up at 15 years);8,18,19 Memphis, TN (n=1139,
urban, 92% black sample; 75% follow-up at 9 years);20,21
and Denver, CO (n=735, urban, 45% Hispanic sample;
86% follow-up at 4 years).22,23
Home visitation is provided by nurses to low-income
fi rst-time mothers beginning prenatally and during
infancy (panel 3). The fi rst and second Nurse–Family
Partnership trials included an additional treatment
Exposure to intimate-
Figure 1: Framework for prevention of child maltreatment and associated impairment
Search strategy and selection criteria
We aimed to identify, evaluate, and summarise recent, high-quality research evidence for
preventing child maltreatment and interventions to reduce the adverse eff ects of such
exposures. Although we did not do a formal systematic review, our search strategies were
designed to identify recent systematic reviews, meta-analyses, and randomised controlled
trials, where available, with evidence from non-randomised designs included only if no
higher level of evidence was available. We were guided by the US Preventive Services Task
Force4 in assessment of the internal validity of the various study methods. The databases
MEDLINE, EMBASE, CINAHL, PsycINFO, the OVID Evidence-Based Medicine Reviews
database (Cochrane DSR, ACP Journal Club, DARE, CCTR, CMR, HTA, and NHSEED), and the
Campbell Collaboration website were searched for citations up to April, 2008, to identify
key studies and evidence syntheses. Database-specifi c terms used to identify the concepts
of child maltreatment (child abuse, child neglect, child sexual abuse, exposure to intimate-
partner violence, foster care, shaken baby syndrome, etc) were identifi ed and paired with
the controlled vocabulary terms appropriate for prevention and intervention. Database,
hand, and internet searches were done up to October, 2008, on key authors and
programmes in the fi eld to locate emerging information. Full search details of all search
strategies and results are available from the authors.
www.thelancet.com Vol 373 January 17, 2009
condition of prenatal visitation without the intensive
postpartum component. The phrase “nurse-visited”
refers here to the group receiving prenatal and intensive
postnatal intervention, since it showed the most positive
During the second postpartum year, the Elmira trial
showed a 32% reduction in emergency department visits
overall (p<0·01), and a 56% fall in emergency department
visits for injuries and ingestions (p<0·05), among
nurse-visited children compared with the control group. A
subgroup of nurse-visited women at highest risk (single,
low-income, teen mothers) had 80% fewer incidents of
verifi ed child abuse and neglect, although this was not
signi fi cant (p=0·07).8 This trial has been criticised because
of the emphasis on fi ndings from subgroup analyses.
However, by the 15-year follow-up,18 child abuse and
neglect were identifi ed less often in the whole sample of
nurse-visited women than in women in the control group
(0·29 vs 0·54 verifi ed reports, p<0·001). This positive
eff ect was not present in homes where moderate-to-high
levels of intimate-partner violence were reported.24
The rate of verifi ed child abuse and neglect in the
sample of children in Memphis (3–4%) was too low to
serve as a feasible outcome for the second trial;25 the
study therefore concentrated on health-care encounters
for injuries and ingestions. At 2 years of age, children
visited by a nurse had 23% fewer health-care encounters
for injuries and ingestions compared with the
control-group children (p<0·05); they were also
hospitalised with injuries or ingestions for 79% fewer
days (p<0·0003).20 By age 9 years,21 children in the control
group were 4·5 times more likely to have died than were
the nurse-visited children, although this diff erence was
not signifi cant (p=0·08).
The Denver trial22 diff ered in that it included a
condition to establish if lack of eff ects in earlier studies
with paraprofessionals (home visitor without professional
training, often selected based on personal attributes)
could be attributed to professional background and
training or the programme models. Because of the
complexity of the health-care delivery system, the use of
child or maternal medical records was not possible, and
rates of verifi ed maltreatment reports were too low to
serve as an outcome. In view of these limitations, the
investigators introduced new measures of parental
caregiving. Eff ects among the nurse-visited children and
mothers were consistent with those achieved in the
earlier trials, whereas the eff ects were roughly half as
large among those visited by paraprofessionals.22,25 On
most outcomes, the children of mothers visited by a
paraprofessional did not diff er signifi cantly from those
in the control group.
Studies are underway in the Netherlands and the UK,
and a feasibility evaluation is in progress in Canada to
establish whether the fi ndings can be replicated in other
Early Start programme
The Early Start programme is an intensive home-visiting
programme targeted to families facing stress and diffi -
culties (panel 4).26 In an RCT (n=443) comparing families
receiving Early Start with control families not receiving the
Panel 1: Interventions to prevent exposure to child maltreatment, by type of abuse
Physical abuse and neglect
• Home-visiting programmes are not uniformly eff ective in reducing child physical
abuse and neglect; any home-visiting programme should not be assumed to reduce
child abuse and neglect (systematic reviews with RCTs)
• Eff ective programmes include:
• Nurse–Family Partnership, which reduced child physical abuse and neglect, as
measured by offi cial child protection reports, and associated outcomes such as
injuries in children of fi rst-time, disadvantaged mothers (RCTs)
• Early Start programme, which reduced associated outcomes such as injuries and
hospital admissions for child abuse and neglect but rates of child protection
reports did not diff er between the intervention and control groups (RCT);
replication is recommended
• Paraprofessional home-visiting interventions (including the Hawaii Healthy Start
Program and Healthy Families America) have not been shown eff ective in reducing
child protection reports; recent RCTs showed confl icting evidence with regard to
maternal self-reported child abuse (RCTs; webappendix)
• Triple P—Positive Parenting Programme showed positive eff ects on substantiated
child maltreatment, out-of-home placements, and reports of injuries, based on a
single study that used an ecological design with a small sample size (RCT); further
assessment and replication are recommended
Abusive head trauma education programmes
• Positive eff ects from one study suggest that hospital-based educational programmes
could reduce abusive head injuries (shaken impact syndrome); (cohort study with
historical control; replications underway)
Enhanced paediatric care for families at risk
• Positive but not statistically signifi cant eff ects suggest that enhancing physicians’
abilities to identify and help families decrease risk factors for child maltreatment
might be eff ective but currently insuffi cient evidence (RCT)
• Unknown if educational programmes reduce occurrence of child sexual abuse; some
evidence that they improve children’s knowledge and protective behaviours but could
have some adverse eff ects (systematic reviews with RCTs)
• Attachment-based interventions might improve insensitive parenting and infant
attachment insecurity, but there is no direct evidence that these interventions prevent
psychological abuse (RCTs)
Exposure to intimate-partner violence
Intimate-partner violence prevention
• No evidence of any existing interventions that prevent intimate-partner violence
against women, and by extension, children (systematic review)
RCT=randomised controlled trial.
www.thelancet.com Vol 373 January 17, 2009 253
service,26,27 88% (391) of those families enrolled were
available for outcome assessment at 36 months. At age
3 years, children in Early Start had sig nifi cantly lower
attendance rates at hospital for childhood injuries than
controls (17·5% vs 26·3%; p<0·05) and fewer admis sions
to hospital for severe abuse and neglect. Early Start children
had about a third of the rate of parent-reported physical
abuse (p<0·01). However, rates of referral to offi cial
agencies for care and protection concerns were similar for
Early Start children and controls. This apparent lack of
diff erence was attributed to the fact that Early Start clients
were under closer surveillance and hence more likely to be
referred to offi cial agencies than controls.
Most of the RCTs that assessed the eff ectiveness of home-
visitation programmes for preventing physical abuse and
neglect have focused on models with service delivery by
paraprofessionals,28 specifi cally the Hawaii Healthy Start
Program and Healthy Families America (webappendix).
Overall, results have been disappointing, and have not
matched the benefi ts of the Nurse–Family Partnership or
Early Start programmes.
Although several parent-training programmes are being
used with the stated goal of preventing child maltreatment,
no clinical trials were identifi ed that used actual child
maltreatment outcomes. One RCT has assessed the eff ect
of a population-based preventive intervention on child
abuse and neglect.29 This study involved the dissemination
of Triple P professional training to the existing workforce
alongside universal media and communication strategies,
across 18 randomly assigned counties in one US
southeastern state (panel 5). Compared with the
services-as-usual control condition, there were positive
eff ects in the Triple P—Positive Parenting Program
counties for rates of substantiated cases of child
maltreatment (d=1·09; p<0·03), child out-of-home
placements (d=1·22; p<0·01), and child maltreatment
injuries (d=1·14; p<0·02; p values are for t tests). These
eff ect sizes describe between-cluster rather than
individual diff erences. For the child maltreatment
outcome, there was a post-intervention increase in both
groups. Of note, the authors did not report standard
deviations for outcomes or for the calculation of
Cohen’s d. A one-sided t test was used in comparing
pre–post diff erence scores but this was not stated in the
manuscript. Overall, the fi ndings are promising, but
some details of the analysis are unclear. Additional
clinical trials of this intervention using child maltreatment
outcomes are warranted, as well as population-based
replications in other communities.
Abusive head trauma education programmes
The most widely adopted prevention strategy in US
hospitals aims to prevent abusive head trauma (shaken
Panel 2: Interventions to prevent re-exposure to and adverse outcomes of child
maltreatment, by type of abuse
Physical abuse and neglect
• Limited evidence to support the use of parent-training programmes to reduce the
recurrence of physical abuse (systematic review of RCTs)
• Parent–child interaction therapy (PCIT) reduced recurrence of child-protection services
reports of physical abuse but not neglect (RCT)
• Some programmes (eg, PCIT and Webster-Stratton Incredible Years Program) might
be eff ective in improving some outcomes associated with physically abusive
• Insuffi cient evidence to conclude that multifaceted in-home programmes reduce
recurrence of physical abuse and neglect (RCTs)
• One programme of intensive nurse home visitation was not eff ective in preventing
recurrence of physical abuse or neglect (RCT)
Neglect-specifi c programmes
• Insuffi cient evidence to conclude that neglect-specifi c interventions reduce recurrence
• Some evidence from small studies that resilient-peer training, imaginative play
training, therapeutic day training, and multisystemic therapy improve child outcomes
(systematic review of controlled studies)
Therapeutic counselling for children and families
• Evidence that cognitive-behavioural therapy can improve specifi c mental-health
outcomes for sexually abused children with post-traumatic stress symptoms,
including post-traumatic stress disorder, anxiety, depression (systematic reviews
• Confl icting evidence for cognitive-behavioural therapy in reducing child behavioural
problems (systematic reviews of RCTs)
Programmes for child molesters (webappendix)
• Surgical castration and chemical treatments might reduce recidivism, but sample bias
is a concern (systematic review of non-randomised and randomised studies)
• Some evidence of effi cacy for psychological treatments but further trials needed
before strong conclusions can be drawn (systematic review of randomised studies)
Therapeutic counselling for parents/families
• Limited evidence of the eff ectiveness of interventions specifi cally designed for parents
or caregivers who emotionally abuse their children
• Group-based cognitive-behavioural therapy might be eff ective with some parents
Exposure to intimate-partner violence
Programmes to prevent recurrence of intimate-partner violence
• Evidence for reducing children’s exposure to intimate partner violence by reducing
violence recurrence against women is limited; one post-shelter advocacy intervention
showed improvement in women’s life quality and initial, but not sustained, reductions
in intimate-partner violence (RCT)
• Restraining orders against abusive partners might prevent recurrent abuse
(prospective cohort), but batterer treatment programme evaluations have mixed, and
generally negative, results (RCTs)
(Continues on next page)
www.thelancet.com Vol 373 January 17, 2009
impact syndrome). Dias and colleagues30 assessed an
educational intervention (leafl et, video, posters) about
the dangers of infant shaking and ways to handle
persistent crying provided to parents in 16 hospitals in
New York State. The incidence of abusive head trauma
was substantially reduced during the 66 months after
introduction of the programme
per 100 000 livebirths) compared with the 66 months
before the study (41·5 cases per 100 000 livebirths).
Currently, a statewide replication of this study (Dias M,
personal communication) and assessment of other
postpartum educational programmes to prevent abusive
head trauma are underway in the USA (Leventhal J,
Enhanced paediatric care for families at risk
Dubowitz and colleagues31 examined the effi cacy of the
Safe Environment for Every Kid (SEEK) model of
paediatric primary care in a continuity clinic in
Baltimore, MD, USA. Clinics were randomised into
routine care provided by the paediatric residents
(250 families) or model care (308 families), in which
residents received special training, systematically
identifi ed family problems, and had a social worker
available. Results showed benefi ts in the model care
group compared with the routine care group, including
apparently fewer child-protection services reports
(13·3% vs 19·2%; p=0·06), fewer instances of medical
neglect (p=0·10), and less harsh punishment reported by
parents (p=0·08). Although this study had modest eff ects
on reports to child-protection services, the results suggest
that enhancing physicians’ abilities to help families
decrease risk factors for child maltreatment could be
The main approach to preventing sexual abuse has been
education programmes provided for children.32,33
Systematic reviews undertaken since 1994 have examined
an increasing number of RCTs of universal school-based
programmes,32–35 which have been widely disseminated
in the USA and Canada. The most recent systematic
review32 assessed data from 15 trials that examined the
eff ectiveness of curricula for children from kindergarten
through high school, mainly in the USA. The programmes
included combinations of fi lm and video, discussion, and
role play; control groups generally consisted of children
on the waiting list or those who received the standard
curriculum. Most of the trials reported signifi cant
improvement in measures of knowledge; a smaller
proportion found signifi cant benefi ts in protective
behaviours under simulated conditions. Disclosures of
past or current sexual abuse were measured in only three
determining whether such disclosures were associated
with the intervention. Negative outcomes such as
increased anxiety were reported in three studies. As
noted by the review authors, many of the studies suff ered
from major methodological weaknesses, including lack
of blinding and analyses that failed to consider cluster
randomisation. Follow-up was generally short—typically
3 months post-intervention. Consistent with previous
systematic reviews,33–35 the authors concluded that
whether increased knowledge and use of protective
behaviours translate into reduced sexual abuse is
unknown; therefore, whether education programmes
aimed at children actually prevent sexual abuse is
Despite increasing awareness about its importance in
children’s lives, psychological abuse is poorly
understood and inadequately researched.36 Evaluation
of the eff ectiveness of interventions in the secondary
prevention of early indicators of psychological abuse
often focuses on maternal insensitivity to infant cues.37
For example, one meta-analysis assessing the
eff ectiveness of attachment-based
ranging from home-visiting programmes to parent–
infant psychotherapy,38 showed improvements in
insensitive parenting (d=0·33) and in infant attachment
insecurity (d=0·20). Increased eff ectiveness was
associated with the use of several sessions and a clear
behavioural focus. Maternal insensitivity is an important
aspect of emotionally harmful parent–child relations,
particularly attachment disorders, and brief, focused
interventions of this nature might have a role in their
prevention (panel 6).39
(Continued from previous page)
Psychological treatment for parents and children
• Some evidence for mother–child therapy in families where children are exposed to
intimate-partner violence in reducing children’s internalising and externalising
behaviour problems and symptoms (RCTs)
• Placement in foster care and not reunifying with biological parents can lead to
benefi ts for maltreated children (observational studies)
• Enhanced foster care can lead to better mental-health outcomes for children than
traditional foster care can (observational studies)
Family preservation programmes
• No evidence that these programmes are eff ective in reducing maltreatment
impairment or recurrence (systematic reviews)
• Confl icting evidence about kinship care compared with traditional foster
care (observational studies)
Interventions for youth in foster care (webappendix)
Multidimensional treatment foster care
• Evidence for reduced behaviour problems and fewer failed placements (RCTs)
RCT=randomised controlled trial.
See Online for webappendix
www.thelancet.com Vol 373 January 17, 2009 255
Exposure to intimate-partner violence
The most direct way to prevent children’s exposure to
intimate-partner violence is through preventing or ending
the violence itself, but there are few high quality, empirical
studies of interventions.40 Two systematic reviews
highlight the lack of evidence for eff ective interventions
to prevent the initiation of intimate-partner violence,40,41
and therefore to prevent children’s exposure to it.
Prevention of recurrence and impairment
This category of intervention is sometimes referred to as
treatment, but we prefer to conceptualise it as outlined in
the fi gure, because maltreatment is an exposure, not a
symptom or a disorder. The two related but distinct goals
of prevention of recurrence and impairment are not
necessarily achieved with the same type of intervention.
Child-protection services have typically focused on
preventing recurrence, whereas prevention of impairment
has generally been the purview of the mental-health
Prevention of recurrence lends itself to classifi cation by
type of maltreatment, since the emphasis of such
interventions is on reducing specifi c abusive or neglectful
behaviours of adults, often within the context of
parenting. Importantly, a family assessment to identify
risk and protective factors that can be altered (eg,
substance misuse, mental-health conditions, support
systems), and to assess the appropriateness of existing
services should be done to identify approaches to prevent
recurrence. We review here those approaches specifi cally
directed at reduction of maltreatment recurrence and
associated impairment; the discussion of general
interventions, such as substance misuse treatment
programmes, is beyond the scope of our article.
Out-of-home care is a broad category of intervention
discussed separately, since it is used as an approach to
reduce recurrence of all types of maltreatment, based on
the principle that the child is removed from the care of an
individual who is abusive or neglectful, or who is failing to
protect the child from such behaviours in others. It has also
been assessed as an approach to prevent impairment.
There has been increasing recognition that the broad
range of physical and mental-health conditions associated
with maltreatment show little specifi city by type of
exposure. Furthermore, many children are exposed to
multiple types of maltreatment. However, not all children
exposed to one or more types of maltreatment experience
impairment. For these reasons, an essential aspect of the
response to maltreatment is a thorough assessment to
establish whether children have symptoms or disorders
that would benefi t from intervention, and then to ensure
they receive the best available interventions for the
conditions identifi ed.
One meta-analysis42 examined the eff ectiveness of
maltreatment. The authors concluded that there was an
overall positive eff ect (d=0·54), although this was
for all categories of
reduced (d=0·21) when self-report and parental reports
of child outcomes were excluded. We considered the
interventions too heterogeneous to draw meaningful
conclusions from this meta-analysis. Similarly, three
recent systematic reviews assessed the eff ectiveness of
interventions in reducing psychological harm in
children and adolescents exposed to trauma;43–45 their
defi nition of trauma was very broad, and included
community violence and natural disasters as well as
child maltreatment. Information about interventions
for specifi c types of maltreatment that could be extracted
from reviews is discussed below. These studies mainly
included samples of sexually abused children or
adolescents, with a few focused on physical abuse and
Physical abuse and neglect
Parent-training programmes have been included in
several reviews of interventions for physically abusive
parents,14,46 but only one had focused explicitly on the
eff ectiveness of training programmes for physically
abusive and neglectful parents.47 Seven RCTs were
included that had targeted parents with a history of child
physical abuse (fi ve studies), physical abuse and neglect
(one study), or unspecifi ed abuse (one study); of the
seven, three used a control group and four used an
alternative treatment group. Only three of the studies
Panel 3: Interventions for preventing child physical abuse and neglect: Nurse–Family
• Home-visiting programme based on theories of human ecology, self-effi cacy, and
• Nurses develop a trusting relationship with the mother and other family members to
promote sensitive, empathic care of their children; assist mothers to review their own
childrearing histories and decide how they want to parent their children
• Improving pregnancy outcomes by assisting women to improve their prenatal
• Improving children’s postnatal health and development by helping parents provide
responsible and competent child care
• Improving parents’ economic self-suffi ciency by assisting them to plan for their
future, including subsequent pregnancies and employment
• Home visiting by nurses with bachelor’s degree in nursing (RN in the Elmira trial); they
underwent 4 weeks of training before the programme
• Women who were pregnant for the fi rst time and of low socioeconomic status were
recruited from prenatal clinics before 29 weeks’ gestation (before 25 weeks in the
• Nurses follow detailed visit-by-visit guidelines and a standardised protocol of visits
• Frequency of home visits changed with stages of pregnancy and adapted to parents’
needs; in the three trials, nurses completed an average of 6·5–9 visits prenatally and
21–26 visits from birth to the child’s second birthday; visits lasted around 75–90 min
www.thelancet.com Vol 373 January 17, 2009
examined the eff ect of parent training on objective
measures of recurrence, including reports by child-
protection workers,48 number of injuries,49 or offi cial
re-reports of physical abuse and neglect.50 The most
recent compared the effi cacy of parent–child interaction
therapy (PCIT) and PCIT plus individualised enhanced
services (EPCIT) with a standard community-group
panel 7).50 At follow-up (median of 850 days), 19% of
parents assigned to PCIT had a re-report for physical
abuse, compared with 36% of EPCIT parents and 49% of
parents in the community group (p=0·02). The only
signifi cant predictor of physical abuse re-referral was the
PCIT condition (p=0·03). There was no diff erence in
re-reports of neglect.
The second study compared the use of cognitive-
behavioural therapy in modifying risk factors associated
with child physical abuse with a family therapy
programme focused on family interaction (n=38).49 There
were signifi cantly fewer child reports (27% vs 59%;
p<0·007) and parental reports (9% vs 53%; p<0·04) of
physical discipline or force in the cognitive-behavioural
therapy group compared with the family therapy group.
The small number of injuries observed precluded
The third study assessed a group-based parenting
programme of child management techniques and
problem-solving; however, the small sample size (n=16)
precluded conclusions about eff ectiveness.48
post-intervention assessment only.47 The review by Barlow
and colleagues47 reported little evidence to support the
use of parent-training programmes to reduce the
recurrence of physical abuse. The most eff ective type of
programme seems to be PCIT. There is also evidence to
suggest that some types of parenting programmes (eg,
Webster-Stratton Incredible Years Program51) could be
eff ective in improving some outcomes that are associated
with physically abusive parenting including, for example,
child reports of parental anger.49
published studies provided immediate
Home-visitation and in-home programmes
A Canadian RCT assessed a programme of home visiting
by nurses provided to families involved with child-
protection services.52 Families with at least one child
who had experienced physical abuse or neglect were
randomly assigned to a 2-year programme of nurse
home visiting in addition to child-protection services, or
intervention included family support, referral to other
services, and education about parenting, tailored to the
needs of the family. Although based on similar
principles, it diff ered substantially in sample, focus,
and content from programmes aimed at preventing
maltreatment before it occurs. At 3-year follow-up, there
was no diff erence between groups in incidents of
physical abuse or neglect; nor was there any reduction
in associated outcomes such as injuries. A post-hoc
subgroup analysis showed that nurse-visited families
involved with child-protection services for fewer than
3 months had a signifi cant reduction in physical abuse,
but not neglect (p<0·05).
Project SafeCare, an in-home treatment programme
for families where physical abuse or neglect has
occurred, is based on Project-12-Ways, an earlier,
multifaceted, in-home programme streamlined to a
24-week intervention with three main components:
child health care, home safety and injury prevention,
and parent–child interaction.53 Project SafeCare has
been reported to reduce the recurrence of physical abuse
and neglect when compared with a family preservation
programme;53,54 however, major limitations in study
design and methodological weaknesses currently
preclude any conclusions about its eff ectiveness in
reducing recidivism—this is being assessed in a current
services alone. The
Programmes focused specifi cally on neglect
In a systematic review of controlled studies evaluating
interventions for children exposed to neglect or for their
caregivers, Allin and colleagues56 concluded that few
evidence-based treatments are available. Resilient peer
treatment57 was noted in one trial to improve social
interactions and reduce behaviour problems, although the
sample size was small (n= 46) and follow-up was only
2 months. A larger RCT of resilient peer treatment (n=82)
Panel 4: Interventions for preventing child physical abuse and neglect: Early Start26,27
• Home-visiting service for families based on a social learning-model approach
• Crucial elements include: assessment of family needs and resources; development of a
positive partnership between client and family support worker; collaborative
problem-solving; and provision of support, advice, and mentoring to mobilise
families’ strengths and resources
• Improve child health
• Reduce risk of child abuse
• Improve parenting skills
• Encourage family socioeconomic and material wellbeing
• Encourage stable partnerships
• Home-visiting by nurses or social workers, bachelor’s level-prepared; they were given a
5-week training programme
• Nurses referred any families with two or more risk factors on an 11-point screening
measure that included parent and family functioning, plus those where nurses had
concerns about a client’s ability to care for the child
• 1-month period to assess family needs; those that scored above a cutoff point
indicating problems in family functioning were off ered the full programme
• Services were tailored to meet the needs of the family
• Families were seen on average over 50 times in the fi rst year; services can be provided
for up to 5 years; visits last around 60–90 min
www.thelancet.com Vol 373 January 17, 2009 257
published after the review58 confi rmed earlier positive
eff ects of this programme when integrated into Head Start
classrooms.58 A programme of imaginative play training59
led to improved peer interactions, positive aff ect, and
better cooperation; again the sample size was small (n=34)
and the follow-up was only a month. Multisystemic
therapy, when compared with a parent-training
programme, showed improved parent–child interactions;
the sample size was small (n=33), groups were not
equivalent on some characteristics, and the follow-up was
only 1 week post-treatment.60 A specifi c therapeutic day
treatment programme assessed in a non-randomised
controlled study (n=34) showed some eff ect in increasing
neglected children’s self-concept.61
Programmes for children and families
Various psychological treatments aimed at reducing
impairment associated with sexual abuse62,63 (or trauma
including sexual abuse)43–45 have been systematically
reviewed. Outcomes included
externalising symptoms or disorders, and sexualised
behaviour. The children participating have ranged in age
from 2 to 17 years, and some interventions have included
parents in the treatment. Ramchandani and Jones62
reviewed 12 RCTs published before December, 2002; nine
from the USA, one from Australia, and two from the UK.
Three studies looked at group cognitive-behavioural
therapy, six were of individual cognitive-behavioural
therapy, one assessed the addition of group therapy to a
family therapy programme, and two compared individual
and group therapy. Comparisons generally involved either
a wait-list control group or a group receiving some type of
supportive therapy. The authors concluded that the best
evidence was for cognitive-behavioural therapy, particularly
for children who had symptoms of post-traumatic stress
disorder; they also noted that those studies with a positive
eff ect involved a parent or caregiver in the treatment.
There was also improvement in behavioural problems,
including sexualised behaviour. The authors described the
overall methodological quality of the studies as low, often
because of inadequate description of the methods. They
also emphasised that although most of the children and
families improved, some became worse. The evidence
regarding eff ectiveness of individual versus group therapy
was deemed too inconsistent to reach a conclusion.
The effi cacy of cognitive-behavioural therapy for sexually
abused children was assessed in a review of randomised
or quasi-randomised studies before November, 2005.63
The review included the cognitive-behavioural therapy
studies listed above and two additional US trials. Sample
sizes typically ranged between 25 and 100 participants,
with the largest including 229 children.64 The interventions
varied in programme content and frequency (six to
20 sessions), but generally included the following themes
for the child sessions: safety education, coping skills,
cognitive processing of the abusive experience,
identifi cation of inappropriate behaviours, relaxation
techniques, dealing with problems related to the abuse,
and graduated exposure in reducing avoidance behaviour.63
Parent or joint sessions focused on parent–child
communication, psycho-education, cognitive reframing,
and parent-management training.63 Results of the meta-
analyses indicated decreases in depressive (p=0·06), post-
traumatic stress disorder (p=0·004), and other anxiety
(p=0·09) symptoms at 1-year follow-up, but no eff ect, on
average, on sexualised behaviour or externalising
symptoms. Methodological aspects of the individual
studies were poorly reported. Macdonald and colleagues63
commented that those studies in which symptoms of
post-traumatic stress disorder were an inclusion criterion
showed a positive eff ect on this outcome.
There was consensus across the two systematic reviews
specifi c for sexual abuse that cognitive-behavioural
therapy should be considered as the fi rst-line treatment
for sexually abused children and their families, but the
evidence for benefi ts is not as broad or as compelling as
other authors suggest. Ramchandani and Jones62
emphasised the following treatment considerations:
ensuring the child’s safety from further abuse; taking
into account the context, including other adversities for
Panel 5: Interventions for preventing child physical abuse and neglect: Triple P—
Positive Parenting Program29
• Public-health population-based approach to child maltreatment
• Comprehensive population-level system of parenting and family support
• Multiple levels of social learning based programme to meet the needs of diff erent
groups of parents
• Address the diffi culties of restricted access of population to evidence-based parenting
• Enhance parental competence, and prevent or alter dysfunctional parenting practices
• Multilevel system including fi ve intervention levels of increasing intensity and
narrowing population reach and delivered by a range of specially trained practitioners
• Universal Triple P (level 1): use of media and informational strategies including radio,
local newspapers, newsletters at schools, mass mailing to family households, presence
at community events, and website information
• Selected Triple P (level 2): consists of brief and fl exible consultations with individual
parents (1–2 consultations of 20 min each), parenting seminars with large groups of
parents, or both
• Primary care Triple P (level 3): consists of four brief consultations (20 min)
incorporating active skills training and use of parenting tip sheets
• Standard and group Triple P (level 4): a ten-session programme (90 min per session)
with individual families using active skills training, home visits, or clinic observation
sessions, or an eight-session group-administered programme (fi ve 2-h group
sessions) using observation, discussion, practice, and feedback plus three 15–30 min
telephone follow-up sessions
• Enhanced Triple P (level 5): is an augmented version of level 4—eg, optional modules
on partner communication, mood management, and stress coping skills
www.thelancet.com Vol 373 January 17, 2009
the child and family; recognising comorbid psychiatric
conditions; and understanding the need for outreach, in
view of the high attrition in many of the treatment
studies. These issues are applicable to the assessment of
children exposed to any type of maltreatment.
Although one review of interventions to reduce
psychological harm associated with traumatic events
concluded that there was
cognitive-behavioural therapy,43 others were more cautious.
Stallard44 noted that attrition rates were often not adequately
reported, and intention-to-treat analyses were rarely used.
Although post-treatment positive eff ects seemed to be
maintained, few studies had follow-up periods extending
beyond 12 months. A substantial proportion of children
with post-traumatic stress who received cognitive-
behavioural therapy (16–40%) still met the diagnostic
criteria for the disorder at the end of treatment.44 Silverman
strong evidence for
and colleagues45 concluded that only trauma-focused
cognitive-behavioural therapy (panel 8)64,65 met the
well-established criteria of Chambless and Hollon.66 Eff ect
sizes for sexual abuse treatments ranged from 0·10 to 0·46
(0·46 was the eff ect size for post-traumatic stress
symptoms). Although Silverman and colleagues45 regarded
the evidence for trauma-focused cognitive-behavioural
therapy to be more robust than previous authors,44,62,63 they
emphasise that the studies are limited in power, length of
follow-up, and lack of intention-to-treat analyses. Treatment
programme approaches for child molesters are reviewed
in the webappendix.
No single approach has been used to address psychological
abuse, possibly because it is such a wide-ranging topic and
potentially includes activities that do not promote the
child’s social adaptation
missocialisation, in which children are exposed to harmful
environments such as intimate-partner violence and drug
misuse.37 There is a paucity of high-quality studies
evaluating the eff ectiveness of interventions specifi cally
designed for parents or caregivers who psychologically
abuse their children.37 The available evidence includes a
RCT comparing two group-based versions of cognitive-
behavioural therapy (standard and enhanced versions of
the Triple P programme) directed at psychologically
abusive parents.67 The standard programme aimed to
teach parents child-management strategies designed to
promote children’s competence and development and to
help parents manage misbehaviour; the enhanced
programme included additional components to change
parental misattributions and anger. Both treatment groups
made substantial gains in a range of outcomes; however,
this study did not include a control group, and many
parents had self-referred. Parents who are severely abusive
might be less inclined to self-refer or to recognise the
eff ects of their own behaviour on children’s externalising
One RCT, comparing
psychotherapy programme with a psycho-educational
home visiting programme and a community standard
intervention group, seemed to favour a psychotherapeutic
intervention in terms
representations of their mother and of themselves, and
also children’s expectations of the mother–child
relationship. However, the measurement of this particular
construct was more likely to favour the psychotherapy
programme than psycho-educational home visiting, and
no other outcomes were included.69 These fi ndings
suggest several approaches to reducing psychological
abuse, but further research is necessary.
of children’s negative
Exposure to intimate-partner violence
Systematic reviews have highlighted the lack of evidence
for the eff ectiveness of screening women to reduce
subsequent exposure to intimate-partner violence.40,41,70
Panel 6: Interventions for preventing psychological abuse:
improving maternal sensitivity38
The following provides one39 of several possible methods of
working with parents to prevent psychological abuse by
improving maternal sensitivity:
• Home-based video feedback with optional attachment
• Improve maternal sensitivity using written information
about sensitive parenting and video feedback
• Improve infant–mother attachment
• Participants consisted of a screened group of insecurely
attached mothers with a fi rstborn, 4-month-old child
• Four 1·5–3-h home visits every 3–4 weeks delivered by
two of the study authors plus third intervener; session
videotaped for use in subsequent session
• Session 1: baby’s contact seeking and exploration
behaviour; use of baby diary to note behaviour and
parental activities for 3 consecutive days
• Session 2: “speaking for the baby” technique to draw
mother’s attention to subtle signals and expressions;
used videotape to identify baby’s and mother’s
feelings; provided brochure outlining baby’s need to
feel understood and secure
• Session 3: adequate and prompt reactions to baby’s
cues; used videotape to identify baby’s signal,
response from mother, and baby’s reaction; brochure
provided on sensitive play with young children
• Session 4: sharing emotions and aff ective attunement
using videotape to focus on the child’s emotions and
• A second intervention group included additional discussions
focused on the mother’s past attachment experiences and
their possible infl uences on her parenting style
www.thelancet.com Vol 373 January 17, 2009 259
The most promising intervention to date is a post-shelter
counselling intervention tested with women in a
RCT (n=284) by Sullivan and Bybee (panel 9).71,72 This
programme of advocacy services compared with no
additional services signifi cantly reduced repeat violence
and improved women’s quality of life at 2 years’
follow-up.71 However, the eff ect on violence reduction was
lost by 3 years’ follow-up.72 The generalisability of these
results to non-shelter samples is unknown.
No other published studies that we know of provide
high-quality evidence for interventions to reduce
exposure to intimate-partner violence. Although there is
some evidence that approaches such as restraining orders
against abusive partners might prevent recurrent
violence,73 batterer treatment programmes have had
mixed, but generally negative, results.74
RCTs of interventions for children exposed to intimate-
partner violence have shown positive outcomes.
Lieberman and colleagues75 did a RCT (n=75) to assess the
eff ectiveness of child–parent
mother–preschooler dyads where the mother was a victim
of intimate-partner violence and had confi rmed that the
child (aged 3–5 years) had exposure (panel 10). The
child–parent psychotherapy group showed a signifi cant
improvement over time compared with controls,
including fewer children meeting the diagnostic criteria
for traumatic stress disorder. These eff ects persisted at
6 months’ follow-up.76 Although this was a rigorous RCT,
the sample was fairly small. However, these results,
alongside other effi cacy trials of child-only compared with
child-mother therapy and with controls,77 indicate that
these forms of mother–child therapy in families where
children are exposed to intimate-partner violence warrant
further evaluation in larger and more diverse samples.
Out-of-home care interventions
This section and the webappendix discuss outcomes
associated with social services’ placement of maltreated
children in out-of-home care (including foster care,
kinship care, residential treatment, group homes, and
shelter care; panel 11). In the summary below, we use
those terms that appear in the individual studies.
Assessing the relative merits of out-of-home care as an
intervention is diffi cult because of the lack of randomised
studies. Quasi-experimental studies have compared:
abused and neglected children who are placed in
out-of-home care to those who remain at home; and foster
children who reunify with their biological families to
children who remain in foster care. Two studies that
compared maltreated children placed in care with those
who remain at home reported that they did not diff er on
delinquency and adult criminal outcomes.78,79 One study,
with a very small sample, noted that children who were
placed in foster care after kindergarten compared with
those who remained at home had more behaviour
problems as assessed by their teachers.80 However, children
placed in foster care are likely to have experienced more
serious and chronic maltreatment and are more likely to
have parents who are unable to handle child-rearing
responsibilities than children who remain at home.79,81,82
Despite the potential increased risk for children
removed from their homes, several other studies have
reported that children placed in care actually fared better
Panel 7: Interventions for reducing recurrence of physical abuse: parent–child
interaction therapy (PCIT)50
• PCIT involving the treatment of parents alongside children
• Behaviourally defi ned approach to skills training
• To increase parental motivation and enhance skills
• To improve parent–child interaction through use of direct coaching and practice of
skills in dyadic parent–child sessions
• Three modules delivered by PCIT trainers who ranged in experience from graduate
students to experts with years of training in PCIT
• Parent–child dyads referred as they entered the child-protection system for a new
confi rmed physical abuse report
• Module 1: six-session orientation group aimed at increasing motivation by fostering
an understanding of the negative consequences of severe physical discipline and
development of self-motivational cognitions and self-effi cacy expectations
• Module 2: 12–14-session course of PCIT consisting of clinic-based, individual
parent–child dyad sessions in two phases. Phase I (child directed interaction) focuses
on teaching relationship-enhancement skills and establishing a daily positive
interaction; phase II (parent directed) focuses on teaching command-giving skills and
a behavioural discipline protocol to promote the child’s compliance
• Module 3: four-session follow-up group programme to address any implementation
problems; children attend a concurrent social-skills programme
Panel 8: Interventions for preventing impairment after sexual abuse: trauma-focused
cognitive-behavioural therapy 43–45,62,65
• Psychotherapeutic intervention based on cognitive and social learning theories
• To alleviate symptoms of post-traumatic stress disorder and related diffi culties
experienced by sexually abused children
• Sessions provided by trained mental health professionals with diverse backgrounds
(eg, social workers, psychologists) who underwent 3 days of training
• Children and their families recruited from outpatient clinical programmes where
referrals made from a broad range of providers (eg, police, child-protection workers)
• Specifi c elements include skills in expressing feelings, coping, recognising links
between feelings and behaviours; gradual exposure through developing a child’s
narrative; reprocessing the abuse; psychoeducation about child sexual abuse and
safety; parent management skills. In a multisite trial64,65 (n=229), treatment was
provided in 12-weekly individual sessions to parents and children by one therapist
with 45 min for each individual session; three sessions included a joint parent–child
session for 30 min (total of weekly sessions 90 min)
www.thelancet.com Vol 373 January 17, 2009
than maltreated children who remained at home did in
the following domains: antisocial behaviour,83 sexual
activity,84 school attendance and academic achievement,85
social behaviour, and quality of life.86 A few other studies
suggest that foster care could provide a benefi t for
vulnerable youth. In an innovative study, abandoned,
institutionalised Romanian children were randomly
assigned to either stay in the institution or to live with a
foster family.87 Those who went to live in foster care,
especially the young abandoned children, had improved
cognitive outcomes relative to those who remained in the
institution. A large US study noted that enhanced foster
care (which included better trained caseworkers and
greater access to services, and supports for youth and
foster families) led to fewer mental and physical health
problems for foster care alumni than did traditional foster
care.88 Other uncontrolled studies have reported that
young children’s adaptive behaviour improved after
placement in foster care89 and that placement in foster
care reduced children’s lead exposure.90
Family preservation programmes—intensive, short-
term services to keep maltreated children at home—have
been widely implemented in the USA. Most experimental
studies have not shown a reduction in placements for the
treatment group.91–95 Design weaknesses include: few
RCTs, poorly developed evaluation plans, small samples
and diff erential attrition, inconsistent programme goals,
diverse services provided, failure to identify families who
could benefi t, and lack of fi delity in implementation.93–95
Once children have been placed in out-of-home care,
there is often an assumption that reunifi cation is the
optimum outcome.96–99 Although 50–75% of children
placed in out-of-home care eventually reunify, between
20–40% of those reunifi ed subsequently re-enter foster
care.100–108 Studies have recorded better outcomes for
children who were not reunifi ed with their families of
origin than those who were, including gains in intelligence
scores,109 greater overall wellbeing,110 and less criminal
recidivism.111 These studies, however, did not control for
behavioural functioning at entry to foster care.
Longitudinal studies that examined the eff ect of
reunifi cation, controlling for functioning assessed
pre-reunifi cation, have reported that reunifi ed youth
showed worse outcomes in internalising and externalising
problems, risky behaviours, competencies, grades, school
dropout, involvement in the criminal justice system,
adverse life events, and witnessing physical violence.112–115
One of these studies also reported that reunifi ed youth
were more likely to experience physical and psychological
violence when disciplined and were less likely to receive
mental-health treatment even after controlling for baseline
levels of internalising symptoms.113 Although a smaller
eff ect, one study reported that reunifi ed children had
lower perceived social isolation than non-reunifi ed
youth.112 Finally, in another study, children who were
formerly in foster care were 1·5-times more likely to die
from a violent death than were children who remained in
foster care, and three times more likely to die from violent
causes than were children in the general population.116
Placement of children in kinship care is a common
child welfare practice in developed countries. Research
has shown that salient risk and protective factors diff er
between kinship and foster caregivers. On average,
kinship caregivers are older, less well educated, less likely
to be married, report more problematic parenting
attitudes, receive fewer non-child welfare services, and
have less caseworker oversight.117–122 However, research
Panel 9: Interventions for preventing exposure by reducing intimate-partner
violence: Post-Shelter Advocacy Programme71,72
• Paraprofessional counselling and advocacy
• To reduce re-exposure to violence and improve quality of life
• To ensure the safety of women and advocates
• Advocacy services provided by female undergraduate students in a community
psychology course who attended two orientation sessions and one semester of
• Women recruited while in shelters for abused women
• The advocates focused on: devising safety plans with women; and using a fi ve-stage
process of assessment, implementation, monitoring, secondary implementation, and
termination to access and mobilise community resources including housing,
employment, transportation, child care, and legal assistance services provided after
leaving shelter for 4–6 h per week through twice-weekly visits for 10 weeks
Panel 10: Interventions for preventing impairment from exposure to intimate-
partner violence: Child–Parent Psychotherapy (CPP)75,76
• Focus on the mother–child relationship
• Based on theories of attachment, parenting and traumatic stress, including social
learning and cognitive-behavioural theories, and the intergenerational transmission
• To reduce children’s emotional and behavioural problems and post-traumatic stress
• To reduce maladaptive behaviours and support developmentally appropriate interactions
• To assist the mother and child in creating a narrative of the traumatic events while
moving towards resolution
• Clinicians had Masters and PhD-level training in clinical psychology and were trained
using a CPP manual developed for this purpose
• CPP provided to mother–preschooler (aged 3–5 years) dyads where the mother was a
victim of intimate-partner violence and the child had been exposed to intimate-
• The mother was actively involved in setting the treatment plan and received
individual counselling as required
• Weekly 60-min CPP sessions for 50 weeks including child’s free play with appropriate
toys to elicit trauma play and social interaction
www.thelancet.com Vol 373 January 17, 2009 261
has shown that children in kinship care are less likely to
be maltreated and have fewer placement changes (relative
to children in foster care), both of which are associated
with better behavioural outcomes.103,123–126
Studies comparing kinship to non-relative foster care
have shown mixed results, with some studies indicating
few or no diff erences on indices of behavioural, cognitive,
educational, medical, and interpersonal functioning.117,127–131
Other studies have found that children in kinship care
seem to fare better in terms of behavioural, educational,
mental health, and social functioning.117,130,132–134 Finally,
two studies have shown more negative outcomes for
children in kinship care, in terms of delinquent
behaviour135 and IQ.136 A major issue that aff ects the
interpretability of these fi ndings is the lack of control for
baseline functioning, since there has been some
suggestion that children who are placed in kinship care
come from less dysfunctional families than do those in
foster care.137 Others have suggested that children with
fewer behavioural or emotional problems are more likely
to be placed in kinship care homes.127,132,138 Those few
interventions shown to be effi cacious with out-of-home
care samples are discussed in the webappendix.
Despite the lack of evidence for eff ective interventions in
the area of child maltreatment compared with other
paediatric public-health problems,139 there have been some
important gains over the past 30 years in approaches to
prevention of maltreatment and its associated impair-
ment. The programme with the best evidence for
preventing child physical abuse and neglect is the
Nurse–Family Partnership, which has shown reductions
in objective measures of child maltreatment or associated
outcomes when administered to high-risk families
prenatally and in the fi rst 2 years of a child’s life; however,
most home-visiting programmes have failed to show such
benefi ts.15,25,140 Similarly, the Early Start programme has
shown positive eff ects in one trial but requires evaluation
in other sites. Three common features of Nurse–Family
Partnership and Early Start could explain their success:
they were developed as research programmes rather than
as service provision methods; both use workers with
tertiary level qualifi cations; and they have made substantial
investments in ensuring the fi delity of programme
delivery. In theory, programmes that share common
features with Nurse–Family Partnership and Early Start
should be eff ective in preventing child maltreatment;
however, the weight of evidence140,141 suggests that most
interventions of this type are ineff ective. The eff ectiveness
of other home-visitation programmes should be assessed
in randomised trials before dissemination.
The Triple P—Positive Parenting Program showed
positive eff ects on substantiated reports of child
maltreatment and associated outcomes in one population-
based trial; however, eff ects arise from a single study
using an ecological design (allocation of intact units)
with a small sample size and some details of the analysis
are unclear.29 Furthermore, replication of these fi ndings
in another setting is important. Preliminary fi ndings
suggest that some prevention programmes for abusive
head trauma could be eff ective in reducing infl icted head
injury,30 and a programme of enhanced paediatric care
for families might show benefi ts in reducing physical
abuse and neglect in children, but further research is
Much less is known about approaches for preventing
sexual abuse, psychological abuse, and children’s
exposure to intimate-partner violence. Sexual abuse
education programmes improve
protective behaviours under simulated conditions; their
eff ect on preventing occurrences of sexual abuse remains
unknown. The history of sexual abuse prevention
programmes highlights the problem in disseminating an
intervention before it has undergone adequate evaluation.
When these programmes were fi rst developed, there was
the opportunity to undertake a trial with outcomes that
included incidents of sexual abuse—both disclosures
and reports from child-protection services—measured
over a reasonable follow-up period. Such programmes
are now widespread, so a RCT with an appropriate
follow-up is unlikely to be undertaken, although
comparison with a usual care group is still possible. In
the prevention of psychological abuse, there is some
preliminary evidence that attachment-based interventions
can reduce maternal insensitivity, an early form of
emotionally harmful parenting, but whether such
programmes prevent the later occurrence of psychological
abuse is unknown.
Preventing the recurrence of maltreatment is
particularly important when a caregiver living with the
child is the identifi ed perpetrator; this occurs less often
with sexual abuse compared with the other types of
maltreatment. A broad range of parent-training
programmes and in-home interventions are provided to
families to prevent recurrence, but there is little
evidence for their eff ectiveness.47 PCIT has shown
Panel 11: Defi nitions for out-of-home care interventions*
Used to denote substitute parental care in a family household by non-relative adults who
receive compensation to be caregivers for children who have been removed from their
biological parents’ care by social services
Used to denote substitute parental care of children by relatives or any adult who has a
kinship bond with a child; this could include family friends or godparents. In this review,
we are referring to children placed with kin by social services because of child
maltreatment, although there are many circumstances when children live with kin
without social services’ involvement. In some jurisdictions, kinship caregivers can become
licensed or certifi ed (sometimes referred to as kinship foster care) and then could be
entitled to compensation
*Defi nitions continued in webpanel.
www.thelancet.com Vol 373 January 17, 2009
benefi ts as an intervention to reduce recidivism of
physical abuse but not neglect.50 Home visitation by a
nurse52 did not reduce recurrence of either neglect or
physical abuse, although such programmes might be
benefi cial in reducing physical abuse but not neglect,
in families newly involved with child-protection
services. The negative results from these two RCTs50,52
in reducing neglect underscore the substantial
challenges in preventing its recurrence. Project
SafeCare55 is promoted as reducing recidivism of
physical abuse and neglect; although ongoing RCTs
might answer this question, current studies provide
insuffi cient evidence of eff ectiveness.
Much more progress has been made in developing
interventions to reduce impairment. The strongest
evidence for reducing psychological symptoms in
children who have experienced sexual abuse is for
cognitive-behavioural therapy; outcomes are improved
when the treatment is targeted to children with symptoms
of post-traumatic stress and a non-off ending parent is
involved in treatment.62,63
For neglected children, there is some preliminary
evidence for resilient peer treatment,57,58 an imaginative
play programme,59 multisystemic therapy,60 and a day
treatment intervention.61 Although recognition of
exposure to intimate-partner violence as a specifi c type of
child maltreatment has occurred only recently,
child–parent psychotherapy shows positive outcomes as
an intervention for children with such experiences.75,76
Out-of-home care is one of the most widely used
interventions for maltreated children, yet there are few
rigorous studies examining its eff ects. There is increasing
evidence from observational studies that placement and
remaining in foster care can lead to benefi ts for maltreated
children compared with reunifi cation; promising
interventions include multidimensional treatment foster
care and adaptations of this model. Studies of training
programmes for foster parents show mixed results
Clearly, the fi eld of maltreatment needs rigorous
designs applied to the assessment of programmes across
the range of interventions. Although the reluctance to
use RCTs seems to be decreasing, there are still few
controlled trials of programmes to reduce the recurrence
of maltreatment.7,142 In those areas where controlled trials
have been done, such as reduction of impairment
associated with child sexual abuse, there are several
common limitations:7,42,44,62,63 poor reporting of methods
including sample size determination, randomisation
procedure, and loss to follow-up; inadequate attention to
reasons for attrition; short-term follow-ups; inappropriate
analyses, including lack of intention-to-treat approaches;
insuffi cient replication studies in determining external
validity; and problems with outcome assessment. In
planning future studies (panel 12), many of these issues
can be addressed by careful adherence to the CONSORT
The selection of outcomes across the range of
interventions is of prime importance. We agree with
Skowron and Reinemann,42 who recommend a so-called
multimethod and multisource approach to the assessment
of maltreatment, but would also add that there needs to
be clear a-priori identifi cation of primary and secondary
outcomes. The potential for bias in selection of any
outcomes needs to be addressed; there has been
over-reliance on use of parental self-reports and reports
of child behaviours44,143 in interventions aimed at reducing
abusive or neglectful behaviours in parents. Use of child-
protection services reports is often not possible,
particularly in assessment of programmes aimed at
Panel 12: Evidence gaps
Prevention of exposure to child maltreatment
• Physical abuse—need further clinical and population-based trials of parent training to
establish eff ectiveness of existing programmes. Studies evaluating interventions to
prevent abusive head trauma (shaken impact syndrome) require replication;
important to consider whether large-scale RCT could be done, in view of low base rate
of abusive head trauma
• Neglect—need to determine essential features of eff ective home-visiting programmes
for prevention of physical abuse and neglect. Additional strategies needed to prevent
neglect; home visitation will not be the only answer
• Sexual abuse—where such programmes do not yet exist, there is the opportunity to do
a RCT that includes outcomes of incidence of sexual abuse as well as proxy outcomes
of knowledge and behaviour; adverse outcomes need to be measured
• Psychological abuse—interventions are required and studies need to include
well validated measures of psychological abuse
• Exposure to intimate-partner violence—interventions assisting women to prevent
intimate-partner violence need to consider prevention of intimate-partner violence
exposure in children
Trials are underway to establish if community-level interventions prevent one or more of
the fi ve types of child maltreatment*
Prevention of recurrent abuse or adverse outcomes associated with child
• Physical abuse—further studies of parent–child interaction therapy required; other
parent-training studies should include direct measures of physical abuse
• Neglect—Project Safecare trials currently underway should establish if this programme
is eff ective in reducing recurrence of neglect
• Sexual abuse—trials of cognitive-behavioural therapy need better methods with
longer follow-ups and consistency of outcome assessment across trials
• Psychological abuse—larger-scale studies of treatment for parents of emotionally
abused children plus development of treatments for children, with both assessed
using direct outcomes of such abuse
• Exposure to intimate-partner violence—further evaluation, in larger and more diverse
samples, of mother–child therapy in families where children are exposed to intimate-
• Out-of-home care—replication of high-quality observational studies determining
eff ectiveness of foster care in improving outcomes for children; further evaluation of
multidimensional treatment, foster care treatment, and adaptations of this
*All programmes above need to be evaluated with randomised trials where possible, and use objective outcome measures, clear
specifi cation of primary and secondary outcomes, without sole reliance on self-report measures.
www.thelancet.com Vol 373 January 17, 2009 263
preventing maltreatment, because of low base rates and
system challenges in accessing such reports.12 Some
argue that surveillance bias precludes the use of child-
protection records in assessing outcome144 but at least
one study refutes this concern.145 Furthermore, systematic
approaches to reviewing child-protection services records
taking into account source of report and use of sensitivity
analyses52 can address this issue, especially in assessment
of programmes for families involved with child-protection
services. Also, trials need to include objective measures
of child health, such as injuries and encounters with the
health-care system, in addition to direct observations of
parenting. Measuring only the risk factors thought to
lead to abuse and neglect is not suffi cient—programmes
must assess actual outcomes of maltreatment and related
This review is limited by its focus on interventions
aimed at the individual (child or caregiver) or family,
because of the emphasis on describing those programmes
that have undergone the most rigorous evaluation,
although one population-based
reviewed.29 Increasingly, interventions at the community
level are being considered in the prevention of child
maltreatment. For example, some communities are
implementing preventive systems of care—strategies to
bring together community agencies into a coordinated
system with the goal of reducing child maltreatment.146
Dodge and colleagues146 suggest that lack of coordination
among social-service agencies could prevent some
families from receiving the fi nancial support or health
services that could lead to better parenting skills. In
Durham, NC, USA, a preventive system has been
implemented based on principles of a system of care,
defi ned as a comprehensive range of mental-health
resources and other support services organised into a
network to meet the needs of children and families.
Researchers are proposing to use offi cial rates of child
maltreatment, with other indices, including visits to
hospital emergency departments,
anonymous surveys of parents about parenting practices.
A second example of a community-based intervention,
Strong Communities for Children, is being assessed by
Melton and colleagues in two South Carolina counties.147
This approach involves a comprehensive strategy of
engaging all sectors of everyday life; it relies on volunteers
and organisations to increase the support for families of
young children. Community-based initiatives are
attractive as a public-health approach to reducing child
maltreatment, but such programmes must be evaluated.
Whether such approaches reduce maltreatment is
unclear, despite their promising theoretical foundation.
In addition to improved assessment of existing services,
additional approaches to reducing maltreatment should
be considered. Bugental148 recommends, for example,
that greater attention should be given to programmes
aimed at preventing men from physically abusing
children. Increasingly, there is recognition of the overlap
of diff erent types of maltreatment exposure, and the need
to take this into consideration in developing prevention
programmes. Other studies underscore the high rates of
comorbidity between exposure to intimate-partner
violence and other types of child maltreatment, and
associated impairment.149,150 In reducing impairment,
Cohen and colleagues151 recommend that treatment
models should target symptom clusters, rather than
focusing on abuse and neglect exposures.
Important advances have been made over the past
30 years in developing interventions to reduce child
maltreatment; a broad range of disciplines are now
involved, such as public health, social work, psychology,
nursing, paediatrics, and psychiatry. A commitment
across disciplines to apply evidence-based principles and
link science with policy is essential.
Confl ict of interest statement
We declare that we have no confl ict of interest. The corresponding author
had full access to all the papers used in the study and had fi nal
responsibility for the decision to submit for publication.
We thank Ellen Jamieson for her help in editing the manuscript,
David Finkelhor for his comments, and the steering group, including
Ruth Gilbert, Danya Glaser, Pat Hamilton, Rosalyn Proops,
Richard Reading, and June Thoburn for their suggestions on this
manuscript. HMacM receives support from the David R (Dan) Off ord
Chair in Child Studies. NW is supported by a Canadian Institutes of
Health Research—Ontario Women’s Health Council New Investigator
Award. HT would like to acknowledge the support of the National
Institute of Mental Health Grant (R01 MH076919).
1 Gilbert R, Spatz Widom C, Browne K, Fergusson D, Webb E,
Janson S. Burden and consequences of child maltreatment in
high-income countries. Lancet 2008; published online Dec 3.
2 Reading R, Bissell S, Goldhagen J, et al. Promotion of children’s
rights and prevention of child maltreatment. Lancet 2008; published
online Dec 3. DOI:10.1016/S0140-6736(08)61709-2.
3 Butchart A, Harvey AP, Mian M, Fürniss T. Preventing child
maltreatment. A guide to taking action and generating evidence.
Geneva: World Health Organization, 2006.
4 Harris RP, Helfand M, Woolf SH, et al. Current methods of the
US Preventive Services Task Force: a review of the process.
Am J Prev Med 2001; 20 (3 suppl): 21–35.
5 Theodore AD, Chang JJ, Runyan DK, et al. Epidemiologic features
of the physical and sexual maltreatment of children in the
Carolinas. Pediatrics 2005; 115: e331–37.
6 Hamby SL, Finkelhor D. The victimization of children:
recommendations for assessment and instrument development.
J Am Acad Child Adolesc Psychiatry 2000; 39: 829–40.
7 Barlow J, Simpkiss D, Stewart-Brown S. Interventions to prevent or
ameliorate child physical abuse and neglect: fi ndings from a
systematic review. J Children’s Services 2006; 1: 6–28.
8 Olds DL, Henderson CR Jr, Chamberlin R, Tatelbaum R. Preventing
child abuse and neglect: a randomized trial of nurse home
visitation. Pediatrics 1986; 78: 65–78.
9 Kamerman SB, Kahn AJ. Home health visiting in Europe.
Future Child 1993; 3: 39–52.
10 American Academy of Pediatrics, Council on Child and
Adolescent Health. The role of home-visitation programs in
improving health outcomes for children and families. Pediatrics
1998; 101: 486–89.
11 Bull J, McCormick G, Swann C, Mulvihill C. Ante- and post- natal
home-visiting programmes: a review of reviews: evidence briefi ng.
UK: NHS Health Development Agency, 2004. https://www.nice.org.
Nov 7, 2008).
www.thelancet.com Vol 373 January 17, 2009
12 Olds DL, Sadler L, Kitzman H. Programs for parents of infants and
toddlers: recent evidence from randomized trials.
J Child Psychol Psychiatry 2007; 48: 355–91.
13 Bilukha O, Hahn RA, Crosby A, et al, for the Task Force on
Community Preventive Services. The eff ectiveness of early
childhood home visitation in preventing violence: a systematic
review. Am J Prev Med 2005; 28 (2 suppl 1): 11–39.
14 MacLeod J, Nelson G. Programs for the promotion of family
wellness and the prevention of child maltreatment: a meta-analytic
review. Child Abuse Negl 2000; 24: 1127–49.
15 Gomby DS. The promise and limitations of home visiting:
implementing eff ective programs. Child Abuse Negl 2007; 8: 793–99.
16 Sweet MA, Appelbaum MI. Is home visiting an eff ective strategy?
A meta-analytic review of home visiting programs for families with
young children. Child Dev 2004; 75: 1435–56.
17 Krugman SD, Lane WG, Walsh CM. Update on child abuse
prevention. Curr Opin Pediatr 2007; 19: 711–18.
18 Olds DL, Eckenrode J, Henderson CR Jr, et al. Long-term eff ects of
home visitation on maternal life course and child abuse and neglect.
Fifteen-year follow-up of a randomized trial. JAMA 1997; 278: 637–43.
19 Olds D, Henderson CR Jr, Cole R, et al. Long-term eff ects of nurse
home visitation on children’s criminal and antisocial behavior:
15-year follow-up of a randomized controlled trial. JAMA 1998;
20 Kitzman H, Olds DL, Henderson CR Jr, et al. Eff ect of prenatal and
infancy home visitation by nurses on pregnancy outcomes,
childhood injuries, and repeated childbearing. A randomized
controlled trial. JAMA 1997; 278: 644–52.
21 Olds DL, Kitzman H, Hanks C, et al. Eff ects of nurse home visiting
on maternal and child functioning: age-9 follow-up of a randomized
trial. Pediatrics 2007; 120: e832–45.
22 Olds DL, Robinson J, O’Brien R, et al. Home visiting by
paraprofessionals and by nurses: a randomized, controlled trial.
Pediatrics 2002; 110: 486–96.
23 Olds DL, Robinson J, Pettitt L, et al. Eff ects of home visits by
paraprofessionals and by nurses: age 4 follow-up results of a
randomized trial. Pediatrics 2004; 114: 1560–68.
24 Eckenrode J, Ganzel B, Henderson CR Jr, et al. Preventing child
abuse and neglect with a program of nurse home visitation: the
limiting eff ects of domestic violence. JAMA 2000; 284: 1385–91.
25 Olds DL, Eckenrode J, Henderson CR, et al. Preventing child abuse
and neglect with home visiting by nurses. In: Dodge KA, Coleman
D, eds. Community based prevention of child maltreatment. New
York: Guilford (in press).
26 Fergusson DM, Grant H, Horwood LJ, Ridder EM. Randomized
trial of the Early Start program of home visitation. Pediatrics 2005;
27 Fergusson DM, Horwood LJ, Grant H, Ridder E. Early Start
evaluation report. Christchurch: Early Start Project Ltd. http://www.
Nov 7, 2008).
28 Geeraert L, Van den Noortgate W, Grietens H, Onghena P. The
eff ects of early prevention programs for families with young
children at risk for physical child abuse and neglect:
a meta-analysis. Child Maltreat 2004; 9: 277–91.
29 Prinz RJ, Sanders MR, Shapiro CJ, Whitaker DJ, Lutzker JR.
Population–based prevention of child maltreatment: the US
Triple P system population trial. Prev Sci (in press).
30 Dias MS, Smith K, DeGuehery K, et al. Preventing abusive head
trauma among infants and young children: a hospital-based, parent
education program. Pediatrics 2005; 115: e470–77.
31 Dubowitz H, Feigelman S, Lane W, Kim J. Pediatric primary care to
help prevent child maltreatment: the Safe Environment for Every
Kid (SEEK) model. Pediatrics (in press).
32 Zwi KJ, Woolfenden SR, Wheeler DM, O’Brien TA, Tait P,
Williams KW. School-based education programmes for the
prevention of child sexual abuse. Cochrane Database Syst Rev 2007;
33 Davis MK, Gidycz CA. Child sexual abuse prevention programs:
a meta-analysis. J Clin Child Psychol 2000; 29: 257–65.
34 Rispens J, Aleman A, Goudena PP. Prevention of child sexual abuse
victimization: a meta-analysis of school programs.
Child Abuse Negl 1997; 21: 975–87.
35 MacMillan HL, MacMillan JH, Off ord DR, Griffi th L, MacMillan A.
Primary prevention of child sexual abuse: a critical review. Part II.
J Child Psychol Psychiatry 1994; 35: 857–76.
36 Glaser D, Prior V, Lynch MA. Emotional abuse and emotional
neglect: antecedents, operational defi nitions and consequences.
York: BASPCAN, 2001.
37 Barlow J, Schrader-MacMillan A, Carter Y, Paul M, Sidebotham P,
Stewart-Brown S. Systematic review of the eff ectiveness of
interventions in the treatment of emotional abuse. London:
Department of Health (in press).
38 Bakermans-Kranenburg MJ, van IJzendoorn MH, Juff er F. Less is
more: meta-analyses of sensitivity and attachment interventions in
early childhood. Psychol Bull 2003; 129: 195–215.
39 Bakermans-Kranenburg MJ, Juff er F, van IJzendoorn MH.
Interventions with video feedback and attachment discussion: does
type of maternal insecurity make a diff erence? Infant Ment Health J
1998; 19: 202–19.
40 Nelson HD, Nygren P, McInerney Y, Klein J,
US Preventive Services Task Force. Screening women and elderly
adults for family and intimate-partner violence: a review of the
evidence for the US Preventive Services Task Force. Ann Intern Med
2004; 140: 387–96.
41 Wathen CN, MacMillan HL. Interventions for violence against
women: scientifi c review. JAMA 2003; 289: 589–600.
42 Skowron E, Reinemann DHS. Eff ectiveness of psychological
interventions for child maltreatment: a meta-analysis.
Psychother Theor Res Pract Train 2005; 42: 52–71.
43 Wethington HR, Hahn RA, Fuqua-Whitley DS, et al, Task Force on
Community Preventive Services. The eff ectiveness of interventions
to reduce psychological harm from traumatic events among
children and adolescents: a systematic review. Am J Prev Med 2008;
44 Stallard P. Psychological interventions for post-traumatic reactions
in children and young people: a review of randomised controlled
trials. Clin Psychol Rev 2006; 26: 895–911.
45 Silverman WK, Ortiz CD, Viswesvaran C, et al. Evidence-based
psychosocial treatments for children and adolescents exposed to
traumatic events. J Clin Child Adolesc Psychol 2008; 37: 156–83.
46 Edgeworth J, Carr A. Child abuse. In: Carr A, ed. What works with
children and adolescents? A critical review of psychological
interventions with children, adolescents and their families. London:
Routledge, 2000: 17–48.
47 Barlow J, Johnston I, Kendrick D, Polnay L, Stewart-Brown S.
Individual and group-based parenting programmes for the
treatment of physical child abuse and neglect.
Cochrane Database Syst Rev 2006; 3: CD005463.
48 Wolfe DA, Sandler J, Kaufman K. A competency-based parent
training program for child abusers. J Consult Clin Psychol 1981;
49 Kolko DJ. Clinical monitoring of treatment course in child physical
abuse: psychometric characteristics and treatment comparisons.
Child Abuse Negl 1996; 20: 23–43.
50 Chaffi n M, Silovsky JF, Funderburk B, et al. Parent-child
interaction therapy with physically abusive parents: effi cacy for
reducing future abuse reports. J Consult Clin Psychol 2004;
51 Hughes JR, Gottlieb LN. The eff ects of the Webster-Stratton
parenting program on maltreating families: fostering strengths.
Child Abuse Negl 2004; 28: 1081–97.
52 MacMillan HL, Thomas BH, Jamieson E, et al. Eff ectiveness of
home visitation by public-health nurses in prevention of the
recurrence of child physical abuse and neglect: a randomised
controlled trial. Lancet 2005; 365: 1786–93.
53 Gershater-Molko RM, Lutzker JR, Wesch D. Using recidivism data
to evaluate Project Safecare: teaching bonding, safety, and health
care skills to parents. Child Maltreat 2002; 7: 277–85.
54 Gershater-Molko RM, Lutzker JR, Wesch D. Project SafeCare:
improving health, safety, and parenting skills in families reported
for, and at-risk for child maltreatment. J Fam Violence 2003;
55 National Center for Injury Prevention and Control. Using
evidence-based parenting programs to advance CDC eff orts in child
maltreatment prevention: research activities. Atlanta: Centers for
Disease Control and Prevention, 2004.
www.thelancet.com Vol 373 January 17, 2009 265
56 Allin H, Wathen CN, MacMillan H. Treatment of child neglect:
a systematic review. Can J Psychiatry 2005; 50: 497–504.
57 Fantuzzo J, Sutton-Smith B, Atkins M, et al. Community-based
resilient peer treatment of withdrawn maltreated preschool
children. J Consult Clin Psychol 1996; 64: 1377–86.
58 Fantuzzo J, Manz P, Atkins M, Meyers R. Peer-mediated treatment
of socially withdrawn maltreated preschool children: cultivating
natural community resources. J Clin Child Adolesc Psychol 2005;
59 Udwin O. Imaginative play training as an intervention method with
institutionalised preschool children. Br J Educ Psychol 1983;
60 Brunk M, Henggeler SW, Whelan JP. Comparison of multisystemic
therapy and parent training in the brief treatment of child abuse
and neglect. J Consult Clin Psychol 1987; 55: 171–78.
61 Culp RE, Little V, Letts D, Lawrence H. Maltreated children’s
self-concept: eff ects of a comprehensive treatment program.
Am J Orthopsychiatry 1991; 61: 114–21.
62 Ramchandani P, Jones DP. Treating psychological symptoms in
sexually abused children: from research fi ndings to service
provision. Br J Psychiatry 2003; 183: 484–90.
63 Macdonald GM, Higgins JP, Ramchandani P. Cognitive-behavioural
interventions for children who have been sexually abused.
Cochrane Database Syst Rev 2006; 4: CD001930.
64 Cohen JA, Deblinger E, Mannarino AP, Steer RA. A multisite,
randomized controlled trial for children with sexual abuse-related
PTSD symptoms. J Am Acad Child Adolesc Psychiatry 2004;
65 Deblinger E, Mannarino AP, Cohen JA, Steer RA. A follow-up study
of a multisite, randomized, controlled trial for children with sexual
abuse-related PTSD symptoms. J Am Acad Child Adolesc Psychiatry
2006; 45: 1474–84.
66 Chambless DL, Hollon SD. Defi ning empirically supported
therapies. J Consult Clin Psychol 1998; 66: 7–18.
67 Sanders MR, Pidgeon AM, Gravestock F, Connors MD, Brown S,
Young RW. Does parental attributional retraining and anger
management enhance the eff ects of the Triple P—Positive
Parenting Program with parents at risk of child maltreatment?
Behav Ther 2004; 35: 513–35
68 Boulton S, Hindle D. Emotional abuse: the work of a
multidisciplinary consultation group in a child psychiatric service.
Clin Child Psychol Psychiatry 2000; 5: 439–52.
69 Toth SL, Maughan A, Manly JT, Spagnola M, Cicchetti D. The
relative effi cacy of two interventions in altering maltreated
preschool children’s representational models: implications for
attachment theory. Dev Psychopathol 2002; 14: 877–908.
70 Ramsay J, Richardson J, Carter YH, Davidson LL, Feder G. Should
health professionals screen women for domestic violence?
Systematic review. BMJ 2002; 325: 314.
71 Sullivan CM, Bybee DI. Reducing violence using community-based
advocacy for women with abusive partners. J Consult Clin Psychol
1999; 67: 43–53.
72 Bybee D, Sullivan CM. Predicting re-victimization of battered
women 3 years after exiting a shelter program.
Am J Community Psychol 2005; 36: 85–96.
73 Holt VL, Kernic MA, Lumley T, Wolf ME, Rivara FP. Civil protection
orders and risk of subsequent police-reported violence. JAMA 2002;
74 Babcock JC, Green CE, Robie C. Does batterers’ treatment work?
A meta-analytic review of domestic violence treatment.
Clin Psychol Review 2004; 23: 1023–53.
75 Lieberman AF, Van Horn P, Ippen CG. Toward evidence-based
treatment: child-parent psychotherapy with preschoolers exposed to
marital violence. J Am Acad Child Adolesc Psychiatry 2005;
76 Lieberman AF, Ghosh Ippen C, Van Horn P. Child-parent
psychotherapy: 6-month follow-up of a randomized controlled trial.
J Am Acad Child Adolesc Psychiatry 2006; 45: 913–18.
77 Graham-Bermann SA, Lynch S, Banyard V, DeVoe ER, Halabu H.
Community-based intervention for children exposed to intimate-
partner violence: an effi cacy trial. J Consult Clin Psychol 2007;
78 Runyan DK, Gould CL. Foster care for child maltreatment: impact
on delinquent behavior. Pediatrics 1985; 75: 562–68.
79 Widom CS. The role of placement experiences in mediating the
criminal consequences of early childhood victimization.
Am J Orthopsychiatry 1991; 61: 195–209.
80 Lawrence CR, Carlson EA, Egeland B. The impact of foster care on
development. Dev Psychopathol 2006; 18: 57–76.
81 Zuravin S, DePanfi lis D. Factors aff ecting foster care placement of
children receiving child protective services. Soc Work Res 1997;
82 Zuravin S, DePanfi lis D. Predictors of child protective service
intake decisions: case closure, referral to continuing services, or
foster care placement. In: Curtis PA, Dale Jr G, Kendall JC, eds.
The foster care crisis: translating research into practice and policy.
Nebraska: University of Nebraska Press, 1999: 63–83.
83 Colton M, Aldgate J, Heath AF. Behavioural problems of children in
and out of care. Soc Work Soc Sci Rev 1991; 2: 177–91.
84 Polit DF, Morton TD, White CM. Sex, contraception and pregnancy
among adolescents in foster care. Fam Plann Perspect 1989;
85 Wald MS, Carlsmith JM, Leiderman PH. Protecting abused and
neglected children. Stanford, CA: Stanford University
86 Davidson-Arad B, Englechin-Segal D, Wozner Y. Short-term
follow-up of children at risk: comparison of the quality of life of
children removed from the home and children remaining at home.
Child Abuse Negl 2003; 27: 733–50.
87 Nelson CA 3rd, Zeanah CH, Fox NA, Marshall PJ, Smyke AT,
Guthrie D. Cognitive recovery in socially deprived young children:
the Bucharest Early Intervention Project. Science 2007; 318: 1937–40.
88 Kessler RC, Pecora PJ, Williams J, et al. Eff ects of enhanced foster
care on the long-term physical and mental health of foster care
alumni. Arch Gen Psychiatry 2008; 65: 625–33.
89 Horwitz SM, Balestracci KM, Simms MD. Foster care placement
improves children’s functioning. Arch Pediatr Adolesc Med 2001;
90 Chung EK, Webb D, Clampet-Lundquist S, Campbell C.
A comparison of elevated blood lead levels among children living in
foster care, their siblings and the general population. Pediatrics
2001; 107: E81.
91 Dagenais C, Bégin J, Bouchard C, Fortin D. Impact of intensive
family support programs: a synthesis of evaluation studies.
Child Youth Serv Rev 2004; 26: 249–63.
92 Heneghan AM, Horwitz SM, Leventhal JM. Evaluating intensive
family preservation programs: a methodological review. Pediatrics
1996; 97: 535–42.
93 Jacobs F. What to make of family preservation services evaluations.
Discussion Paper CS-70. Chicago: The Chapin Hall, 2001.
94 Littell J, Schuerman J. A synthesis of research on family
preservation and family reunifi cation programs. Westat, Inc. in
association with James Bell Associates and the Chapin Hall Center
for Children at the University of Chicago, 1995.
95 Schuerman J. Best interests and family preservation in America.
Chicago: Chapin Hall Center for Children at the University of
96 Maluccio AN, Abramczyk LW, Thomlison B. Family reunifi cation of
children in out-of-home care: research perspectives.
Child Youth Serv Rev 1996; 18: 287–305.
97 Maluccio AN, Pine BA, Warsh R. Protecting children by preserving
their families. Child Youth Serv Rev 1994; 16: 295–307.
98 Berliner L. Is family preservation in the best interest of children?
J Interpers Violence 1993; 8: 556–57.
99 Gelles RJ. Family reunifi cation/family preservation: are children
really being protected? J Interpers Violence 1993; 8: 557–62.
100 Barth RP, Berry M. Implications of research on the welfare of
children under permanency planning. In: Barth R, Berrick JD,
Gilbert N, eds. Child welfare research review: Vol 1. New York:
Columbia University Press, 1994: 323–43.
101 Berrick JD, Needell B, Barth RP, Jonson-Reid M. The tender years:
toward developmentally sensitive child welfare services for very
young children. New York: Oxford University Press, 1998.
102 Courtney ME. Factors associated with the reunifi cation of foster
children with their families. Soc Serv Rev 1994; 68: 81–108.
103 Courtney ME. Reentry to foster care of children returned to their
families. Soc Serv Rev 1995; 69: 226–41.
Series Download full-text
www.thelancet.com Vol 373 January 17, 2009
104 Fein E, Maluccio AN. Permanency planning: another remedy in
jeopardy? Soc Serv Rev 1992; 66: 335–48.
105 Festinger T. Returning to care: discharge and re-entry in foster care.
Washington, DC: Child Welfare League of America, 1994.
106 Fraser MW, Walton E, Lewis RE, Pecora PJ, Walton WK.
An experiment in family reunifi cation: correlates of outcomes at
one-year follow-up. Child Youth Serv Rev 1996; 18: 335–61.
107 Terling T. The effi cacy of family reunifi cation practices: reentry rates
and correlates of reentry for abused and neglected children reunited
with their families. Child Abuse Negl 1999; 23: 1359–70.
108 Wulczyn F. Family reunifi cation. Future Child 2004; 14: 94–113.
109 Fanshel D, Shinn EB. Children in foster care: a longitudinal
investigation. New York: Columbia University Press, 1978.
110 Lahti J. A follow-up study of foster children in permanent
placements. Soc Serv Rev 1982; December: 556–71.
111 Jonson-Reid M, Barth RP. From placement to prison: the path to
adolescent incarceration from child welfare supervised foster or
group care. Child Youth Serv Rev 2000; 22: 493–516.
112 Lau AS, Litrownik AJ, Newton RR, Landsverk J. Going home: the
complex eff ects of reunifi cation on internalizing problems among
children in foster care. J Abnorm Child Psychol 2003; 31: 345–58.
113 Litrownik AJ, Newton R, Mitchell BE, Richardson KK. Long-term
follow-up of young children placed in foster care: subsequent
placements and exposure to family violence. J Fam Violence 2003;
114 Sinclair I, Baker C, Wilson K, Gibbs I. Foster children: where they
go and how they get on. London: Jessica Kingsley, 2005.
115 Taussig HN, Clyman RB, Landsverk J. Children who return home
from foster care: a 6-year prospective study of behavioral health
outcomes in adolescence. Pediatrics 2001; 108: e10.
116 Barth RP, Blackwell DL. Death rates among California’s foster care
and former foster care populations. Child Youth Serv Rev 1998;
117 Berrick JD, Barth RP, Needell B. A comparison of kinship
foster homes and foster family homes: implications for kinship
foster care as family preservation. Child Youth Serv Rev 1994;
118 Leslie LK, Landsverk J, Ezzet-Lofstrom R, Tschann JM, Slymen
DJ, Garland AF. Children in foster care: factors infl uencing
outpatient mental health service use. Child Abuse Negl 2000;
119 Brooks D, Barth RP. Characteristics and outcomes of drug-exposed
and non drug-exposed children in kinship and non-relative foster
care. Child Youth Serv Rev 1998; 20: 475–501.
120 Ehrle J, Geen R. Kin and non-kin foster care-fi ndings from
a national survey. Child Youth Serv Rev 2002; 24: 15–35.
121 Gebel TJ. Kinship care and non-relative family foster care:
a comparison of caregiver attributes and attitudes. Child Welfare
1996; 75: 5–18.
122 Jones HB, Clyman RB, Kriebel DK, Lyons ME. Kith and kin care:
parental attitudes and resources of foster and relative caregivers.
Child Youth Serv Rev 2004; 26: 657–71.
123 National Research Council. Understanding child abuse and neglect.
Washington, DC: National Academy Press, 1993.
124 Courtney ME, Needell B. Outcomes of kinship care: lessons from
California. In: Barth RP, Berrick JD, Gilbert N, eds. Child
welfare research review, vol II. New York, NY: Columbia University
Press, 1997: 130–49.
125 Newton RR, Litrownik AJ, Lansverk JA. Children and youth in
foster care: disentangling the relationship between problem
behaviors and number of placements. Child Abuse Negl 2000;
126 Taussig HN, Culhane SE. Contextual infl uences on mental health
and behavioral outcomes of children in foster care. Poster Session
Presented at the 15th International Conference on Mental Health
Services Research, National Institute of Mental Health,
Washington, DC, 2002.
127 Benedict MI, Zuravin S, Stallings RY. Adult functioning of children
who lived in kin versus nonrelative family foster homes.
Child Welfare 1996; 75: 529–49.
128 Carpenter SC, Clyman RB. The long-term emotional and physical
wellbeing of women who have lived in kinship care.
Child Youth Serv Rev 2004; 26: 673–86.
129 Carpenter SC, Clyman RB, Davidson AJ, Steiner JF. The association
of foster care or kinship care with adolescent sexual behavior and
fi rst pregnancy. Pediatrics 2001; 108: E46.
130 Dubowitz H, Feigelman S, Harrington D, Starr R Jr, Zuravin S,
Sawyer R. Children in kinship care: how do they fare?
Child Youth Serv Rev 1994; 16: 85–106.
131 De Robertis MT, Litrownik AJ. The experience of foster care:
relationship between foster parent disciplinary approaches and
aggression in a sample of young foster children. Child Maltreat
2004; 9: 92–102.
132 Iglehart AP. Kinship foster care: placement, service, and outcome
issues. Child Youth Serv Rev 1994; 16: 107–22.
133 Keller TE, Wetherbee K, LeProhn NS, et al. Competencies and
problem behaviors of children in family foster care: variations by
kinship placement status and race. Child Youth Serv Rev 2001;
134 Timmer SG, Sedlar G, Urquiza AJ. Challenging children in kin
versus nonkin foster care: perceived costs and benefi ts to caregivers.
Child Maltreat 2004; 9: 251–62.
135 Shore N, Sim KE, LeProhn NS, Keller TE. Foster parent and teacher
assessments of youth in kinship and non-kinship foster care
placements: are behaviors perceived diff erently across settings?
Child Youth Serv Rev 2002; 24: 109–34.
136 Singer LT, Minnes S, Short E, et al. Cognitive outcomes of
preschool children with prenatal cocaine exposure. JAMA 2004;
137 Dubowitz H, Zuravin S, Starr RH Jr, Feigelman S, Harrington D.
Behavior problems of children in kinship care. J Dev Behav Pediatr
1993; 14: 386–93.
138 Beeman SK, Kim H, Bullerdick SK. Factors aff ecting placement of
children in kinship and nonkinship foster care. Child Youth Serv Rev
2000; 22: 37–54.
139 The Lancet. The neglect of child neglect. Lancet 2003; 361: 443.
140 Gomby DS, Culross PL, Behrman RE. Home visiting: recent
program evaluations—analysis and recommendations. Future Child
1999; 9: 4–26.
141 Chaffi n M. Is it time to rethink Healthy Start/Healthy Families?
Child Abuse Negl 2004; 28: 589–95.
142 Chaffi n M, Friedrich B. Evidence-based treatments in child abuse
and neglect. Child Youth Serv Rev 2004; 26: 1097–1113.
143 Fantuzzo JW, Twentyman CT. Child-abuse and
psychotherapy-research—merging social concerns and
empirical investigation. Prof Psychol Res Pr 1986; 17: 375–80.
144 Roberts I, Kramer MS, Suissa S. Does home visiting prevent
childhood injury? A systematic review of randomised controlled
trials. BMJ 1996; 312: 29–33.
145 Chaffi n M, Bard D. Impact of intervention surveillance bias on
analyses of child welfare report outcomes. Child Maltreat 2006;
146 Dodge KA, Berlin LJ, Epstein M, et al. The Durham Family
Initiative: a preventive system of care. Child Welfare 2004;
147 Melton GB, Holaday BJ, Kimbrough-Melton RJ. Community life,
public health, and children’s safety. Fam Community Health 2008;
148 Bugental DB. Finding ways to reduce the prevalence of child
maltreatment among fathers: a comment on the alternative
approaches. Clin Psychol Sci Prac 2004; 11: 112–15.
149 Evans SE, Davies C, DiLillo D. Exposure to domestic violence:
a meta-analysis of child and adolescent outcomes.
Aggress Violent Behav 2008; 13: 131–40.
150 Kitzmann KM, Gaylord NK, Holt AR, Kenny ED. Child witnesses to
domestic violence: a meta-analytic review. J Consult Clin Psychol
2003; 71: 339–52.
151 Cohen JA, Mannarino AP, Murray LK, Igelman R. Psychosocial
interventions for maltreated and violence-exposed children.
J Soc Issues 2006; 62: 737–66.