one page summary

Social Science & Medicine 58 (2004) 1367–1384

Effective/efficient mental health programs for school-age children: a synthesis of reviews

Gina Browne a,b,

*, Amiram Gafni a,b,c

, Jacqueline Roberts a,b , Carolyn Byrne

a ,

Basanti Majumdar a,d

a System-Linked Research Unit (SLRU), School of Nursing, McMaster University, Hamilton, Ont., Canada

b Department of Clinical Epidemiology and Biostatistics (CE&B), McMaster University, Hamilton, Ont., Canada

c Centre for Health Economics & Policy Analysis, McMaster University, Hamilton, Ont., Canada

d Primary Health Care for Women of KwaZulu-Natal, South Africa

Abstract

The prevalence of mental health problems, some of which seem to be occurring among younger cohorts, leads

researchers and policy-makers to search for practical solutions to reduce the burden of suffering on children and their

families, and the costs to society both immediate and long term. Numerous programs are in place to reduce or alleviate

problem behaviour or disorders and/or assist positive youth development. Evaluated results are dispersed throughout

the literature. To assess findings and determine common elements of effective children’s services, a literature search was

undertaken for evidence-based evaluations of non-clinical programs for school-age children. Prescriptive comments aim

to inform service-providers, policy-makers and families about best practices for effective services such as: early, long-

term intervention including reinforcement, follow-up and an ecological focus with family and community sector

involvement; consistent adult staffing; and interactive, non-didactic programming adapted to gender, age and cultural

needs. Gaps are identified in our understanding of efficiencies that result from effective programs. Policy implications

include the need to develop strategies for intersectoral interventions, including: new financing arrangements to

encourage (not penalize) interagency cooperation and, to ensure services reach appropriate segments of the population;

replication of best practices; and publicizing information about benefits and cost savings. In many jurisdictions

legislative changes could create incentives for services to collaborate on service delivery. Joint decision-making would

require intersectoral governance, pooling of some funding, and policy changes to retain savings at the local level.

Savings could finance expansion of services for additional youth.

r 2003 Elsevier Ltd. All rights reserved.

Keywords: Children; School-age; Mental health; Effective programs; Efficiency; Review

Introduction

This paper provides an overview, drawn from reviews,

of evidence on the effectiveness and efficiency of mental

health services for school-aged children, and addresses

the policy implications of its findings. Evidence of

effectiveness compares outcomes for children (and

youth) receiving and not receiving services; efficiency,

in this paper, considers the cost of providing services

compared to the cost to society of not providing such

services. The study included both universal population-

based services (provided to all children) and early

intervention population-based services (provided only

to at-risk children).

There is a high prevalence of mental health problems

in children (20–30%) (Stephens, Dulberg, & Joubert,

1999) and many of them have multiple problems (Byrne

et al., 2002), which are inadequately treated or

ARTICLE IN PRESS

*Corresponding author. Faculty of Health Sciences, System-

Linked Research Unit (SLRU), HSC-3N46, McMaster Uni-

versity, 1200 Main St. West, Hamilton, Ont., Canada L8N 3Z5.

Tel.: +905-525-9140x22293; fax: +905-528-5099.

E-mail address: [email protected] (G. Browne).

0277-9536/$- see front matter r 2003 Elsevier Ltd. All rights reserved.

doi:10.1016/S0277-9536(03)00332-0

undetected (Offord et al., 1999). Numerous publicly

funded but uncoordinated agencies in the health,

education, social services, recreation and corrections

sectors provide various types of care for children, yet

many still lack the appropriate care that would be

informed by a more comprehensive view of their

problems. Untreated problems in children is costly in

human and fiscal terms, for themselves, their families

and the wider society (Offord, Boyle, & Racine,

1992) (such as, the costs of lost potential, observed in

school dropout rates, unemployment, welfare rates and

crime).

The following questions, suggested by Knapp (1997),

will be used to address the issues:

1. What outcomes are the interventions or services

trying to achieve, a reduction of problems or

promotion of competencies?

2. How? With what mix of services: prevention,

enhancement of positive factors or treatment of

negative factors?

3. For whom is the service intended, for what age or

other characteristics?

4. Where is it delivered: at school, within the family,

primary care, the community or in some combina-

tion?

5. Why and how is the program expected to help? Is the

strategy specialized, or part of a coordinated or

integrated plan?

6. What results are shown?

This paper sets out to analyse the evidence, focussing

on reviews of the literature and seminal studies which

address the above questions.

Correlates of child health outcomes and conceptual

framework

Considerable research has confirmed associations

between developmental, emotional and behavioural

disorders and a wide array of interrelated influences on

the individual, direct and indirect, biological and

contextual (Greenberg, Domitrovich, & Bumbarger,

2001). Contextual factors include influences on a child

within the family, neighbourhood, school and commu-

nity (Offord & Lipman, 1996). Inherited traits and pre-

dispositions, physical health, cultural norms, parental

education, parenting style, income and family stability,

among other factors—all are potentially positive or

negative influences. The positive relationship between

emotional/behavioural problems and family socioeco-

nomic status, e.g., is well established in population

studies (Marmot, Ryff, Bumpass, Shipley, & Marks,

1997). Risk may include specific biological or environ-

mental insults that produce neurological or psychologi-

cal defects, but, as well, may involve the presence or

absence of resources and opportunities that more subtly

shape developmental pathways.

In as much as they provide a child with resources to

cope with or buffer negative stressors and thrive despite

deficits, internal and external factors are protective of

mental health. Both risk and protective factors interact

to help determine child development (Benson & Saito,

1999). Exposure to accumulating risk factors increases

the likelihood of mental health, developmental or

behavioural problems (Offord et al., 1999), yet protec-

tive factors lessen the effect of risk factors as long as

some degree of balance is maintained (Catalano,

Berglund, Ryan, Lonczak, & Hawkins, 1999).

A strong current has developed in the United States

and elsewhere over the past 20 years for services to

strengthen this mental ‘immune system’ in children,

termed resilience, accompanied by a sizable literature. A

resilience checklist by Grotberg (1998) identifies char-

acteristics of resilience across various ages and culture.

Resilience theory suggests that all children can benefit

from preparation to help them respond to adversity with

effective, healthy strategies and coping mechanisms

(Catalano et al., 1999). Being risk free is not the same

as being prepared (Greenberg et al., 2001). One is

prepared, despite risks, when one can say, to paraphrase

Grotberg, that one has caring people for support and

guidance, confidence in one’s own self-worth, and good

coping skills. Recreational, educational or social pro-

grams may aid healthy child development through risk

factor reduction or positive youth development. Com-

petence, engagement, support, identity and efficacy are

frequently included as mechanisms. Programs may also

address specific behaviours (e.g., substance abuse) or

treat children’s mental health disorders and symptoms

(e.g., attention-deficit hyperactivity). Whereas the for-

mer are likely universal or early intervention, the latter

are more likely targeted or clinical interventions (i.e.,

limited to children exhibiting symptoms or with a

diagnosed disorder).

Since more attention has been given to pre-school

research (Zoritch, Roberts, & Oakley, 1998), we

focussed primarily on reviews of universal and early

intervention services for older children. Programs

providing clinical services were excluded.

Methodology

A search of published and unpublished English-

language scientific literature focussed on evaluations of

universal and early intervention health promotion

initiatives for children at risk related to health and

social welfare, recreation and culture, occupation,

remedial education, housing and corrections. Databases

included: Medline (1990–2000); PubMed; OVID; Social

Sciences Index (WebSPIRS) and Ideas. Website searches

ARTICLE IN PRESS G. Browne et al. / Social Science & Medicine 58 (2004) 1367–13841368

A R T IC LE

IN P R E S S

Table 1 Quality of relevant reviews

a

Review Search strategy

stated

Comprehensive

search

Relevance

criteria

described for

primary studies

Quality of

primary studies

assessments

Comprehensive

quality

assessment

(minimum 3

components)

Findings

integrated

Adequate

data to

support

conclusion

Strength

of the

review

Bauman, Drotar, Leventhal, Perrin,

and Pless (1997)

X X X X X X X Strong

Bennett and Offord (1998) X X X X X X X Strong

Breton et al. (1998) X X X X X Moderate

Catalano et al. (1999) X X X X X Moderate

DiCenso et al. (1999) X X X X X X X Strong

Dowswell et al. (1996) X X X X X X X Strong

Durlak and Wells (1997) X X X X X X X Strong

Emshoff and Price (1999) X Weak

Greenberg et al. (1999) X X X X X X X Strong

Greenwood et al. (2000) X X X X X X X Strong

Heneghan et al. (1996) X X X X X X X Strong

Hodgson, Abbasi and Clarkson

(1996)

X X X X X Moderate

Kalfus (1984) X X X X Moderate

Kirmayer et al. (1999) X X Weak

Lister-Sharp et al. (1999) X X X X X X X Strong

Marcotte (1997) X X X X X X X Strong

Mathur and Rutherford (1991) X X X X X X X Strong

Odom and Strain (1984) X X Weak

Ploeg et al. (2000) X X X X X X X Strong

Rispens et al. (1997) X X X X X X X Strong

Thomas et al. (1999) X X X X X X X Strong

Tilford et al. (1998) X X X X X X X Strong

Yamada et al. (1999) X X X X X X X Strong

a 6–7 criteria Met=Strong; 4–5 criteria Met=Moderate. o3 criteria Met=Weak.

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 6 9

included: the Internet search engine-Copernic; Co-

chrane; Centre for Reviews and Dissemination, UK;

Universit!e de Qu!ebec !a Montr!eal (UQAM); Health

Canada (2000), international health research organiza-

tions, and related links. Restricting reviews to English-

only articles may, unfortunately, have omitted some

useful international findings. Reference Manager (data

management software) provided detailed lists from 1714

studies, which produced additional sources from refer-

ences. Research centres, funding agencies, government

departments, and health service agencies recommended

unpublished material.

Content and quality were the primary selection

criteria for investigators. Evaluation methods were

critically appraised using parameters suggested in

several references (Oxman, 1994). Economic evaluations

had to address program goals, breadth of focus, timing

and intensity, venues, audience, evaluation rigour and

connections to other programs and the community.

Investigators concentrated their attention on ‘multi-

disciplinary’ programs that involved several service

sectors and/or professions, where possible, and used

only reviews of randomized controlled trials or quasi-

experimental comparison groups to increase reliability.

Where articles included a mix of experimental and non-

experimental studies, those meeting the criteria were

assessed first to assemble key conclusions. Non-experi-

mental studies such as descriptive narrative literature

and informed opinion were later considered for context.

The quality of review articles was assessed using

suggestions from relevant literature (Oxman, Cook, &

Guyatt, 1994; Guyatt et al., 2000). A review had to:

address a focussed question; have effective, appropriate

selection methods for relevant articles; appraise study

validity; give sufficient methodology to reproduce

assessments; provide consistent, complete and precise

results; and consider results in terms of importance,

applicability, benefits and limitations. Table 1 rates the

quality of the 23 reviews summarized.

Results

Tables 2–7 provide a study-by-study summary of

findings. Here we discuss a number of patterns and

characteristics common to the early intervention pro-

grams, and to both early intervention and universal

programs.

Reviews discussed efforts to reduce deficiencies

related to depression, anxiety, externalizing/internalizing

or other psychological/social problems (Table 2),

reductions in risky behaviours (Table 3), outcomes to

increase competence and resilience through various

protective strategies (Table 4) or programs with a

combination of both outcome strategies (Table 5). Some

reviews contained school-based programs to promote

positive behaviours and prevent psychosocial problems

(Table 6); others contained community-based programs

with similar aims (Table 7).

Although universal or early intervention programs to

develop protective factors (generally by increasing

competence or skills), are more effective (Tables 4

and 5) than programs to reduce existing negative behavi-

ours (Tables 2 and 3) (Greenberg Domitrovich, &

Bumbarger, 1999); nevertheless, program effectiveness

can vary by age, gender and ethnicity of children.

Younger children, either pre-school age or in early

grades, benefit more than older children (Zoritch et al.,

1998) but programs for some older children are also

effective (Ploeg, Ciliska, & Brunton, 2000). Programs

to address a specific problem or problems, which are

sensitive to cultural or gender-based differences

(Thomas et al., 1999), have greater effect than broad,

unfocussed interventions. For example, because, adoles-

cent boys and girls have responded differently to suicide

prevention programs, gender-focussed programs are

advisable. Similarly, programs for aboriginal children

have more positive results when they use traditional

knowledge and modes, are based on community

initiatives, and involve both family and community

(Kirmayer, Boothroyd, Laliberte, & Simpson, 1999).

Programming that has multiple, integrated elements

involving more than the single domain of family, school

or community, is more likely to have positive results

than single focus, single domain interventions (Tables 4

and 5). This characteristic was shared by initiatives to

create competence by skills acquisition (Catalano et al.,

1999), to address clustered risky behaviours (Dowswell,

Towner, Simpson, & Jarvis, 1996), to reduce risk and, to

some extent, to change established behaviour (Lister-

Sharp, Chapman, Stewart-Brown, & Sowden, 1999).

Theoretical bases of programming seem effective

when appropriate for the type of intervention (Con-

tento, Balch, Bronner, & Lythe, 1995). For example,

positive outcomes associated with skill acquisition were

enhanced by interventions using interactive learning

based on social learning theory, developmental social

norms, social influence and social reinforcement (Lister-

Sharp et al., 1999), on social pressure modelling and on

skill rehearsal (Tables 4 and 5) (DiCenso, Guyatt, &

Willan, 1999). Effect sizes decreased over time for

knowledge and skills acquisition (Rispens, Aleman, &

Goudena, 1997) and behaviour reduction (Thomas et al.,

1999), suggesting the need for periodic follow-up and

reinforcement of positive interventions. An exception,

Marcotte (1997), found an increased effect size over time

(Table 2) among programs to treat depression in

adolescents by cognitive behavioural change. This may

derive from the intervention or from unrelated factors

such as the natural progression of milder disorders.

Certain methods of program delivery (Table 6) are

associated with lower effectiveness. Fear-inducing

ARTICLE IN PRESS G. Browne et al. / Social Science & Medicine 58 (2004) 1367–13841370

A R T IC LE

IN P R E S S

Table 2 Reductions in deficiencies: psychosocial problems, injury, abuse, hyperactivity

Topic/author Studies included Goals: program orientation Intervention strategies Outcomes Results

Treating depression in

adolescence (Marcotte,

1997)

N ¼ 7 Decrease depressive symptoms

Role-play Depression, self-esteem,

anxiety, conflict resolution,

irrational beliefs

Effect size:

Social skills Self concept Reductions in children’s

negative behaviour:

Self-modeling Cognitive distortions At post-test from 0.41 to

1.70 (small to large)

Rational-emotional therapy At follow-up from 0.60 to

1.69 (medium to large)

Cognitive behaviour Treatment more effective

with parental involvement

Preventing unintentional

injuries in children and

young adolescents

(Dowswell et al., 1996)

N=not stated Prevent unintentional injury Cycle helmets Injury rates Educational programs alone

have little effect

Car seats Community programs with

broad range of strategies/

participation more effective

Road safety

Crossing patrollers

Redistribute traffic safety

Home devices:

Smoke detectors

Child-resistant containers

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 7 1

A R T IC LE

IN P R E S S

Table 3 Reductions in risky behaviours: teen pregnancy, sexually transmitted diseases (STDs), crime, family breakdown, suicide

Topic/author Studies included Goals: program orientation

Intervention strategies Outcomes Results

Effectiveness of school-based interventions in reducing adolescent risk behaviour: a review of reviews (Thomas et al., 1999)

N ¼ 18 Knowledge Lectures Reductions in behaviour:

Drug prevention and sexual risk reduction programs more comprehensively evaluated than emotional/behavioural problem prevention programs:

Social influence Class series Smoking Didactic, knowledge-based programs have no effect on behaviour

Social norms Peer led Alcohol Interactive programs more effective in changing behaviour than non-interactive

Reasoned action Teachers led: Drug use Intervention success decreases with time Social learning Discussion group Sexual risk Interactive programs based on social learning

theory, including developmental social norms and social reinforcement are most effective

Health belief Role playing Behaviour and emotional problem

Overall effective programs result in modest changes

Skills practice Pregnancy rates Gender differences STD rates

Attitude change

A systematic review of the effectiveness of adolescent pregnancy primary prevention programs (DiCenso et al., 1999)

N ¼ 20 RCTs (none strong methodologically)

Pregnancy prevention School, community and clinic-based interventions by trained adult or peer leaders

Sexual activity Focus on sexuality does not increase sexual activity

Pregnancies Effective programs substantial in duration, focussed on behaviours; theory-based; engaged participants; shared facts, focussed on social pressures, modelling and skill rehearsal; included trained adult or peer leaders

A systematic review of the effectiveness of primary prevention programs to prevent STDs (Yamada et al., 1999)

N ¼ 24 randomized or controlled clinical trails

Reduction in STDs Primary prevention of STDs by trained peer/ professional/ paraprofessional educational sessions

Condom use 4 ‘moderate’ studies had a positive impact on at least one outcome

(None strong methodologically; 4 moderately strong)

For low-income African- American and Hispanic adolescents

No. of sexual partners

Effective programs theory-based, include interpersonal skills training

Frequency of intercourse; protected/ unprotected oral/ anal/vaginal intercourse

Minimum 8h with trained facilitators

Diagnosed STDs

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 7 2

A R T IC LE

IN P R E S S

Evaluating intensive family preservation programs: a methodological review (Heneghan et al., 1996)

N ¼ 5 RCTs Support the family and prevent out-of-home placements

Family workers provide home-based intensive services:

Out-of-home placement rate

5 RCT placement rates:

N ¼ 5 Quasi- experimental design

Case management Costs Treatment group 24–43%

Family counselling Family functioning Control group 20–57% Concrete services

(financial, transportation) Recurrent abuse Methodologically flawed studies show no

benefit of family prevention services in reducing out-of-home placements

Effectiveness of school-based curriculum suicide prevention program for adolescents (Ploeg et al, 2000)

N ¼ 7 Suicide prevention Psychological education (cognitive-behavioural principles)

Suicide-related knowledge

Insufficient evidence to support school-based suicide prevention curriculum

Stress-inoculation; coping skills acquisition, rehearsal

Attitudes: mental health indicators

Beneficial and harmful effects

Perceived stress Programs may need modification for at-risk and girls vs. boys

Anger Comprehensive, multistrategy programs to address adolescent clustered Risk behaviours

Self-esteem

Suicide prevention and mental health promotion in first nations and inuit communities (Kirmayer et al., 1999)

N ¼ 5 Suicide prevention Community social development programs:

Attempts at self-injury

Effective programs are:

School-based skills Community-initiated Band councils Partnership with band councils or/aboriginal

organizations Competency Draw from traditional knowledge/wisdom of

elders Continuum of services:

prevention, early intervention, crisis psychotherapy, after care

Community consultation

Aimed at biological, psychological and spiritual dimensions

Broad focus

Suicide intervention and prevention programs in Canada (Breton et al., 1998)

N ¼ 15 Suicide prevention School based prevention education

Process vs. outcome evaluations

Prevention strategies poorly defined

Quasi- experimental

Gate keepers as interventions

Intervention strategies better defined but insufficient information on screening procedures

6 prevention School based programs should include family and community domains

5 intervention 4 mixed

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 7 3

A R T IC LE

IN P R E S S

Table 4 Increasing child/youth competence/resilience: positive youth development, nutrition, health promotion

Topic/author Studies included Goals: program orientation

Intervention strategies Outcomes Results

Positive youth development programs

N ¼ 25 Positive youth development

Skill development: Competence Positive changes in:

(Catalano et al., 1999) School (N ¼ 6) Social Self-efficiency Youth behaviour Community (N ¼ 2) Cognitive Pro-social norms Interpersonal skills School/family (N ¼ 7) Decision making Pro-social

Involvement Quality of adult/peer relationships

School/community (N ¼ 1) Coping Recognition for positive behaviour

Self control

School/family/community (N ¼ 9)

Refusal/resistance Bonding Problem solving

Positive identity Self efficacy Environmental, organizational change strategies influencing:

Self determination Academic achievement

Teachers Belief in future Best practice interventions: Peer norms Resiliency Include more than one domain Peer perceptions Spirituality Address 5 youth outcomes (minimum)

Improving community relations

9 months or more

Careful attention to implementation and outcome evaluation

Effectiveness of nutrition education: a review

N ¼ 217 Nutrition education and intervention

Behavioural change Eating behaviour measured by dietary recalls, records

Effective programs behaviourally focussed and based on appropriate theory of behaviour change

N ¼ 43 re. school-aged Communication and educational strategies for enhancing awareness

Impact on knowledge, attitudes, skills, behaviours, health outcomes

Most effective programs:

(Contento et al., 1995) Environmental interventions Effect of behaviour interventions— parental role

Actively involve participants, surrounding school, community and environment Involve self-assessment and

feedback Require active participation

Tailor messages to motives of target groups Educate intermediaries Address short-term cognitive behaviour

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 7 4

tactics such as ‘shock incarceration’ programs (Table 5)

seem ineffective (Greenwood, Model, Rydell, & Chiesa,

2000). Programs that deliver information only, and in a

didactic mode, appear to be less effective (Hertzman &

Wiens, 1996) than interactive activities impacting both

school and family. Long-term programming, from

several months to years, is shown to be more effective

than short, intensive initiatives (Heneghan, Horwitz, &

Leventhal, 1996). Early interventions for children at risk

(Durlak & Wells, 1997) or in the early stages of

disordered behaviour can also be effective (Greenwood

et al., 2000). Certain behaviours and attitudes proved

more resistant to change (e.g., substance misuse, unsafe

sex and oral hygiene (Thomas et al., 1999).

The continuing presence of appropriate adult staff

(Tilford, Delaney, & Vogels, 1998), and mentoring or a

stable relationship with a successful adult, were im-

portant aspects of program delivery. The latter pro-

motes positive social/emotional development, academic

achievement, and reduces disordered behaviour (Gross-

man & Tierney, 1998). Peer mentoring effectively

promotes favourable academic and social behaviour in

early intervention programs (Kalfus, 1984) and social

skills in children with behaviour disorders (Mathur &

Rutherford, 1991), but is less reliable for general

competencies and skill maintenance (Odom & Strain,

1984).

Almost every review dealt with services that were all

or, in part, within a school venue. Easily accessible on-

site, school-based services encourage continuing partici-

pation, an important element of an intervention, yet risk

breaches of confidentiality and labelling of participants.

Programs operated out of community centres can

provide confidentiality and serve a larger catchment

area, but reach a smaller proportion of area children

than school-based programming. When children are

already exhibiting symptoms, the inclusion of families in

community-centre-based interventions is an important

factor for success (Greenwood et al., 2000). A compre-

hensive solution would include services in both venues.

Educational and fiscal policies that limit the use of

schools for non-curricular activities are current chal-

lenges to such a solution.

Though a lack of data about cost–benefits in the

reviews renders economic evaluation difficult, other

findings in the policy literature (Browne et al., 1999;

Browne, Byrne, Roberts, Gafni, & Whittaker, 2001)

identify cost savings from preventive children’s health

initiatives.

Discussion

The findings have numerous implications for further

research and policy direction in the child mental health

field. Reviewers noted some inconsistent methodology

and deficiencies in study design, intervention strategies

and reporting, necessitating caution toward some results

(Breton et al., 1998). However, the number of common

findings from so many differing samples and interven-

tions lends credence to their reliability.

The benefits of creating programs around an ecolo-

gical approach to children’s services are echoed in the

broader literature (US Public Health Service, 2000). It

seems clear that effective services for school-aged

children should address their individual needs and

involve the multiple domains and support systems in

their lives. The evidence calls for universal services to

bolster protective factors and for tailored, long-term,

timely interventions for high-risk children, an approach

consistent with other recent findings (Board on Chil-

dren, Youth and Families, 2002; Offord et al., 1999). An

underlying thread is that effective children’s services,

and agencies, should address the whole child rather than

focussing only on a single problem behaviour, since

children often have a cluster of emotional/behavioural

problems, interrelated with one another and with

external factors. Research is still needed to prove the

benefits of specific innovative, intersectoral combina-

tions of health, social, educational and recreational

programs to promote competence in the face of

deficiencies and risks, assist behavioural change, and

affect the prognosis of child/youth behaviour problems.

We need to determine the accessibility and population

coverage of effective, universal and early intervention

strategies by age, gender and culture and evaluate

policies to encourage their adoption and support.

Comprehensiveness of research into interventions can

be assessed using the framework proposed in Fig. 1.

Furthermore, we need to understand what organiza-

tional and financial barriers impede the implementation

of ecological, community-wide, universal and early

intervention strategies, study best practices and normal-

ize inclusion of cost effectiveness as a part of evaluation.

Such evidence would then inform policy changes to

facilitate intersectoral cooperation and appropriate

long-term funding. Transparent public decision-making

would rely on the dissemination of evidence about

effective interventions and upon mechanisms to encou-

rage replication of proven effective services.

With some notable exceptions (Knapp, 1997), re-

search overlooks organizational and financial govern-

ance and policy mechanisms needed to foster integration

within and across differently financed services. Few

studies evaluate whether universal and early intervention

initiatives can save the public sector money. However,

that aspect is receiving more recent attention (Browne

et al., 2001).

Current programs developed to influence children’s

development and mental health are generally uncon-

nected and certainly unintegrated. Many programs

initiated locally to meet perceptions of community need

ARTICLE IN PRESS G. Browne et al. / Social Science & Medicine 58 (2004) 1367–1384 1375

A R T IC LE

IN P R E S S

Table 5 Outcomes for combination of risk reduction/enhanced competence

Topic/author Studies included Goals: program orientation

Intervention strategies Outcomes Results

Health promotion in schools: a systematic review

N ¼ 32 of 200 reviews Health promotion Class room changes in school ethics and environment, community and family involvement

Most affected: Ecological multidomain approaches more effective than single domain

(Lister-Sharp et al., 1999) Healthy eating Fitness Most effective programs based on

social learning and social influence Injury

prevention and abuse Mental health

Least affected: Substance

misuse Safe sex Oral hygiene

Primary prevention mental health programs for children and adults—a meta analytic review

N ¼ 177 Primary prevention of behavioural and social problems in preschool, primary, secondary school children

Primary prevention with mental health focus

Externalizing/ internalizing behaviours

Average participant surpasses performance of control group average (50–82%)

(Durlak and Wells, 1997) 150 published Academic achievements

Outcomes reflected 8–46% difference favouring prevention

27 unpublished Environment-centered, aimed at school/home environments

Socioeconomic status

Most interventions reduced problems and increased competencies

or Cognitive processes

Need studies with longer follow-up and more details of interventions

Person centered Psychosocial skills

Schools mental health and life quality (Bennett and Offord, 1998)

N ¼ 4 Mental health School characteristics controlling for student, classroom, and local socioeconomic characteristics

Cognitive behaviour emotional outcomes

Wide variation in student cognitive and behaviour outcomes

Prospective Quality of life Examination process

School-to-school variations not explained by student entry characteristics teacher–pupil ratios, instructional resources, physical facilities

Cohort School attendance School and classroom processes (working conditions, teacher, self- efficiency, morale, commitment; ability grouping, disciplinary climate,

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 7 6

A R T IC LE

IN P R E S S

parent–school relations) related to student outcome

Analytic Classroom behaviour

Relationship between school processes and student outcomes not well understood

Prevention of mental health problems (Greenberg et al., 1999)

Criteria Violence prevention

Curriculum-based teaching Psychopathology: Best practice

N ¼ 34 of 130 programs Social/cognitive skill building

Conflict resolution Aggression Stressing protective factors (competence and skills) more effective than targeting disordered or risk behaviour

Changing school ecology

Anger management Depression Youth participation more effective than lecture

20 Targeted Multicomponent Empathy skills Anxiety Multiple, coordinated, ecological approaches more effective in creating competence but not in reducing risky behaviour

14 Universal Multidomain Team building Multiyear programs have more enduring effectiveness

Role playing Risky behaviour: Interaction Impulsiveness Future studies Linking families and children

Antisocial More rigorous designs

Deficiencies in cognitive skill

Longer follow-up

Cognitive and social skills competence

Aim more at internalizing (mood problems)

Address who most benefits from which approaches? Measure multiple outcomes

A review of psychosocial interventions for children with chronic health problems

RCTs N ¼ 11 Psycho-social health in face of physical illness

Structured intervention manual

Self esteem 11 studies demonstrate positive outcome in at least one psychosocial variable

(Bauman et al., 1997) Non RCTs N ¼ 4 Self-efficiency Useful points about lack of methodologically sound studies

Asthma N ¼ 7 Focus on control Cancer N ¼ 3 Family functioning Epilepsy N ¼ 2 Mixed diagnosis N ¼ 3

Prevention of child sexual abuse victimization

N ¼ 16 studies Assess effects of child sex abuse prevention programs

Instructional concepts Knowledge of sex abuse concepts

Longer duration and skills emphasis most effective

a meta analysis of school programs

Behavioural: Acquisition of self- protection skills

Post test effect size was 0.71 (moderate)

(Rispens et al., 1997) Protection skills Follow-up 0.62 effect size Film Colouring book

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 7 7

A R T IC LE

IN P R E S S

Diverting children from a lifetime of crime

N ¼ 493 Crime prevention Supportive early childhood intervention (4 years) for children at risk for later antisocial behaviours (N ¼ 7)

Trouble with law/ probation

Early:

(Greenwood et al., 2000) Competency Teacher ratings 6% referred to probation compared to 22% of matched controls

Development: Interventions for families with children acting out N ¼ 6

School achievements

Reductions in child abuse 4% vs. 19%

Target high risk 4 years of school-based interventions, e.g., incentives to graduate

One-half the arrests compared to controls at 27 years follow up

Address substance abuse, anger

Better grades

Cognitive behavioural skills

Teacher ratings: More motivation

Interventions early in delinquency (Andrews, et al., 1990) N ¼ 80)

Acting out behaviour

More employment at age 19

Lipsey N ¼ 400 (1992) Cognitive scores Decreased acting out Educational

attainment Reduction in aggression, externalizing behaviour school failure Reductions in recidivism by 30–50% Better school achievement, less delinquency Less destructive Graduation incentives increase high school completion and college enrollment 30% of the arrests of control students Decreased troublesome youths Some programs reduce recidivism equally by as much as 50% ‘Shock incarceration’ and ‘Scared straight’ techniques more harmful than beneficial

Health promotion in schools: a systematic review.

N ¼ 32 of 200 studies met criteria

Health promotion Class room; changes in school ethics and environment

Most affected Ecological multidomain approaches more effective than single domain

(Lister-Sharp et al., 1999) Community and family involvement

Healthy eating Most effective programs based on social learning and social influence

Fitness Injury

Table 5(Continued)

Topic/author Studies included Goals: program orientation

Intervention strategies Outcomes Results

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 7 8

A R T IC LE

IN P R E S S

prevention and abuse Mental health

Least affected: Substance

misuse Safe sex Oral hygiene

Effectiveness of mental health promotion intervention—a review

N ¼not stated studies: 1980– 1995

Mental health promotion for children, young people, adults, elderly and high risk

Re: youth: Self concept Appropriate staff necessary for self-concept programs

(Tilford et al., 1998) Health education Mental health School programs effective Outward bound program Minority groups need tailored,

separate self-concept activities School curriculum, coping

skill development Outdoor activities a good means of developing self-concept

Exercise for pregnant teens

Effective mental health promotion: a review

N ¼ 6 RCTs school-aged Intervention focussed on:

School-based groups Coping with negative feelings

Better conflict resolution

(Hodgson et al., 1996) Coping skills High risk groups Social skills Less shyness Social

relationships Peer relationships Fewer learning problems

Healthy environments

Attitudes to school More socially competent

Meaningful activities

Less depressive conduct disorders

Social policy Better academic achievements Reduction in life

stresses Less substance use

Successful programs: Aim to influence a combination of risk/protective factors Involve group’s social network, e.g., teachers, parents, family Intervene at different times not once only Combine interventions, e.g., social support and coping skills

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 7 9

A R T IC LE

IN P R E S S

Table 6 Where and how to provide universal and early intervention services: school-based programs: peer mediation, day-care, health education, positive behaviour promotion, psychosocial

problem prevention

Topic/author Studies included Goals: program orientation

Intervention strategies Outcomes Results

Schools mental health and life quality

N ¼ 4 Mental health School characteristics (controlling for student, classroom, and local socioeconomic characteristics)

Cognitive behaviour emotional outcomes

Wide variation in cognitive and behaviour outcomes

(Bennett and Offord, 1998) Prospective Quality of life Examination process School-to-school variations not explained by student entry characteristics, teacher– pupil ratios, instructional resources, physical facilities

Cohort School attendance Student outcomes related to school and classroom processes (working conditions, teacher, self-efficiency, morale, commitment; ability grouping, disciplinary climate, parent–school relations)

Analysis Classroom behaviour

Peer mediated intervention: a critical review

N ¼ 39 peers as tutors Peers’ positive influence on the behaviour of target children

Peers as tutors Academic accomplishments (spelling, reading, arithmetic)

Peers effective in promoting favourable academic and social behaviour outcome

(Kalfus, 1984) N ¼ 6 peers as facilitators Peers as reinforcing agents Classroom behaviour, articulation, social behaviour

Peer value as facilitators of generalized and maintenance of competencies less clear

N ¼ 20 peers as reinforcers Peers as facilitators of generalizations

Peer mediated interventions promoting social skills of children and youth with behaviour disorders

N ¼ 21 Promoting social skills of children with a behaviour disorder

Peers as mediators Social skills Peer-mediated approaches produce immediate positive treatment effects on promoting social skills

(Mathur and Rutherford Jr., 1991)

Social competence Typologies of peer-mediation identified

Prevention and intervention strategies with children of alcoholics

N-not stated Reduction in substance use

Short-term small group format emphasizing:

Knowledge Increased information

(Emshoff and Price, 1999) Information Social support Skills-building in coping and social competence

Problem/emotion focussed Coping skills Social support Coping skills Emotional function Effective programs have outlet for safe

expression of feelings Social/emotional support

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 8 0

A R T IC LE

IN P R E S S

Positive youth development programs

N ¼ 25 School (N ¼ 6Þ Skill development: Competence Best practices:

Community (N ¼ 2) Social Self-efficiency Include more than one domain (Catalano et al., 1999) School/family

(N ¼ 7) Cognitive Pro-social norms Address minimum 5 youth outcomes for 9

months or more School/community ðN ¼ 1Þ

Decision making Pro-social involvement

Careful attention to implementation and outcome evaluation

School/family/ community ðN ¼ 9)

Coping Recognition for positive behaviour

Refusal, resistance Bonding Positive identity

Environmental: organizational change strategies:

Self determination

influencing teachers, peer norms, peer perceptions

Belief in future

Improving relations with the community

Resiliency

Spirituality

Prevention of mental health problems

Criteria Violence prevention Curriculum-based teaching Psychopathology: Best practice:

(Greenberg et al., 1999) N ¼ 34 of 130 programs Social/cognitive skill building

Conflict resolution Aggression Stressing protective factors (competence and skills) more effective than targeting disordered or risk behaviour

Changing school ecology

Anger management Depression Youth participation more effective than lecture

20 targeted Multicomponent Empathy skills Anxiety Multiple, coordinated, ecological approaches more effective in creating competence, not in reducing risky behaviour

14 universal Multidomain Team building Multiyear programs more enduring effectiveness

Role playing Risky behaviour: Interaction Impulsiveness Future: Linking families and children Antisocial More rigorous designs

Deficiencies in cognitive skill

Longer-term follow-up

Cognitive and social skills competence

Aim more at internalizing (mood problems)

Who (with what characteristics) most benefits from which approaches? Measure multiple outcomes

G .

B ro

w n

e e t

a l.

/ S

o c ia

l S

c ie

n c e

& M

e d

ic in

e 5

8 (

2 0

0 4

) 1

3 6

7 –

1 3

8 4

1 3 8 1

compete for funding and community support. Deter-

mining program effectiveness is only a beginning.

Given service delivery problems, some children are

undoubtedly not receiving potentially beneficial pro-

grams. Jones and colleagues make the point that

‘universally available’ is not synonymous with ‘equal

access’ or ‘equal participation’ (Jones, 1992). Policy

initiatives need to be developed to ensure sufficient

funding and promote delivery of effective programs to

appropriate children. Program planning should also

consider meeting intersectoral needs, providing trans-

portation if necessary, or other tangible aid to families,

to reduce program attrition or bolster intervention

effectiveness.

ARTICLE IN PRESS

Table 7 Where and how to provide universal and early intervention services: Community-based programs: mentoring

Topic/author Studies included Goals: program

orientations

Intervention

strategies

Outcomes Results

Peer-mediated

approaches to

promoting

children’s social

interaction: a review

(Odom and Strain,

1984)

N ¼ 4 Promote positive social behaviour of

targeted youth

Peer-mediated

interventions using:

Play, social

behaviour

‘Prompt and

reinforce’ methods

and peer/initiation

methods more

effective than

proximity re.

positive social

behaviour

N ¼ 5 Proximity (4) Results mixed as to whether gains

generalized to other

settings

N ¼ 6 Prompt and reinforce (5)

Generalization

appears to be

related to socially

responsive peers

Peer/initiation (5)

Research Overlay

Child Care

Social Housing

Corrections

Labour

Education

Recreation

Social Services

Health

CLINICAL/ REMEDIAL

EARLY INTERVENTION

UNIVERSALGOAL/FOCI

Scope of Human Services (Browne, et. al., 2002)

PUBLIC PRIVATE

NON- PROFIT

CONTINUITY FINANCINGSECTORS

Fig. 1. Integration model.

G. Browne et al. / Social Science & Medicine 58 (2004) 1367–13841382

A strategy of risk factor reduction entails long-term

initiatives in education and a rebalance of societal

resources to address core risk factors such as socio-

economic inequity. Protective factor enhancement and

promoting of competencies may be more readily

achievable with ‘relatively’ short-term comprehensive,

early, multimodal and multidisciplinary initiatives.

Though further research is needed, the findings from

this and similar reviews could enhance current services

and inform development of effective intersectoral

services for youth.

This evidence compels us to examine policy changes to

foster integration of separately financed and governed

children’s services at a local level. For many jurisdic-

tions, legislative changes could create incentives for such

services to collaborate on service delivery. Collaboration

rather than consolidation avoids creating a new bureau-

cracy, and preserves both agency autonomy and peer

checks and balances necessary for productivity. Joint

decision-making would require intersectoral governance,

pooling of some funding, and policy changes to retain

savings at the local level. Savings could finance

expansion of services for additional youth. An inter-

sectoral governance structure could include all the

sectors identified in Fig. 1, representatives of the

continuum of services and public, private, not-for-profit

funding sources.

References

Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J., Genedreau,

P., & Cullen, F. T. (1990). Does correction treatment work?

A clinically-relevant and psychologically-informed meta-

analysis. Criminology, 28(3), 369–404.

Bauman, L. J., Drotar, D., Leventhal, J. M., Perrin, E. C., &

Pless, I. B. (1997). A review of psychosocial interventions

for children with chronic health conditions. Pediatrics,

100(2), 244–251.

Bennett, K. J., & Offord, D. R. (1998). Schools, mental health

and life quality. In National Forum on Health (Eds.),

Determinants of health: Settings and issues, Vol. 3: Setting

and issues (pp. 47–86). Sainte-Foy, Quebec: !editions Multi-

Mondes.

Benson, P. L., & Saito, R. N. (1999). The scientific foundations

of youth development. Minneapolis, MN: Search Institute,

Available on the World Wide Web, May 2002 at

www.ppv.org/pdffiles/ydv/ydv 4.pdf.

Board on Children, Youth and Families, National Research

Council. (2002). Community programs to promote youth

development. Committee on Community-Level Programs

for Youth, Available from http://www.bocyf.org/Recent

Publications.html.

Breton, J. J., Boyer, R., Bilodeau, H., Raymond, S., Joubert,

N., & Nantel, M. A. (1998). Review of evaluative research on

suicide intervention and prevention programs for young people

in Canada: Theoretical context and results. Health Canada.

Browne, G., Byrne, C., Roberts, J., Gafni, A., & Whittaker, S.

(2001). When the bough breaks: Provider-initiated compre-

hensive care is more effective and less expensive for sole-

support parents on social assistance. Social Science &

Medicine, 53(12), 1697–1710.

Browne, G., Roberts, J., Gafni, A., Byrne, C., Weir, R.,

Majumdar, B., & Watt, S. (1999). Economic evaluations of

community-based care: Lessons from 12 studies in Ontario.

Journal of Evaluation of Clinical Practice, 5(4), 367–385.

Byrne, C., Browne, G., Roberts, J., Bell, B., Chalklin, L., Mills,

M., Kramer, J., Wallik, D., & Mills, A. (2002). Adolescent

emotional/behavioural problems and risk behaviour in Ontario

primary care: Comorbidities and costs. System-Linked

Research Unit, McMaster University, Hamilton, Ont.,

Working Paper Series #02-01.

Catalano, R. F., Berglund, M. L., Ryan, J. A. M., Lonczak,

H. S., & Hawkins, J. D. (1999). Positive youth development

in the United States. Research findings on evaluations of the

Positive Youth Development Programs. Report to the US

Department of Health and Human Services, Office of the

Assistant Secretary for Planning and Evaluation and

National Institute for Child Health and Human Develop-

ment, Available on the World Wide Web at http://

aspe.hhs.gov/hsp/PositiveYouthDev99/index.htm.

Contento, I., Balch, G. I., Bronner, Y. L., & Lythe, L. A.

(1995). The effectiveness of nutrition education and

implications for nutrition education policy, programs and

research—A review of research. Journal of Nutrition

Education, 27, 277–418.

DiCenso, A., Guyatt, G., & Willan, A. (1999). A systematic

review of the effectiveness of adolescent pregnancy primary

prevention programs. Public Health Research Education and

Development (PHRED).

Dowswell, T., Towner, E. M., Simpson, G., & Jarvis, S. N.

(1996). Preventing childhood unintentional injuries–

what works? A literature review. Injury Prevention, 2(2),

140–149.

Durlak, J. A., & Wells, A. M. (1997). Primary prevention

mental health programs for children and adolescents: A

meta-analytic review. American Journal of Community

Psychology, 25, 115–152.

Emshoff, J. G., & Price, A. W. (1999). Prevention and

intervention strategies with children of alcoholics. Pedia-

trics, 103(5), 1112–1121.

Greenberg, M. T., Domitrovich, C., & Bumbarger, B. (1999).

Preventing mental disorders in school-age children: A review

of the effectiveness of prevention programs. Available on the

World Wide Web at www.prevention.psu.edu/CMHS.html.

Greenberg, M. T., Domitrovich, C., & Bumbarger, B. (2001).

The prevention of mental disorders in school-aged children:

current state of the field. Prevention & Treatment 4, article 1,

posted March 30, 2001, American Psychological Associa-

tion, Available on the World Wide Web at http://journals.

apa.org/prevention/volume4/pre0040001a.html.

Greenwood, P. W., Model, K. E., Rydell, C. P., & Chiesa, J. R.

(2000). Diverting children from a life of crime: measuring

costs and benefits. Available on the World Wide Web at

www.rand.org/publications/MR/MR699/.

Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work?

An impact study of the Big Brothers, Big Sisters program.

Evaluation Review, 22(3), 403–426.

ARTICLE IN PRESS G. Browne et al. / Social Science & Medicine 58 (2004) 1367–1384 1383

Grotberg, E. (1998). I am, I have, I can: What families

worldwide taught us about resilience. Reaching Today’s

Youth: The Community Circle of Caring, 1(3), 36–39.

Guyatt, G. H., Haynes, R. B., Jaeschke, R. Z., Cook, D. J.,

Green, L., Naylor, C. D., Wilson, M. C., & Richardson, W.

S. (2000). Users’ guides to the medical literature XXV:

Evidence-based medicine: Principles for applying the users’

guides to patient care. Journal of American Medical

Association, 284(10), 1290–1296.

Health Canada. (2000). Celebrating success: a self-regulating

service delivery system for children and youth—A discussion

paper. Available on the World Wide Web at http://www.

hc-sc.gc.ca/hppb/childhood-youth/cyfh/pdf/Celebrating.pdf.

Heneghan, A. M., Horwitz, S. M., & Leventhal, J. M. (1996).

Evaluating intensive family preservation programs: A

methodological review. Pediatrics, 97, 535–542.

Hertzman, C., & Wiens, M. (1996). Child development and

long-term outcomes: A population health perspective and

summary of successful interventions. Social Science &

Medicine, 43, 1083–1095.

Hodgson, R., Abbasi, T., & Clarkson, J. (1996). Effective

mental health promotion: A literature review. Health

Education Journal, 55, 55–74.

Jones, M. B. (1992). The benefits of beneficence. Social Service

Review, June, pp. 183–217 (Cited in Offord, D. R., Hanna,

E. M., Hoult, L. A., Recreation and the development of

children and youth: A discussion paper. Prepared for the

Ministry of Tourism and Recreation, Ontario).

Kalfus, G. R. (1984). Peer mediated intervention: A critical

review. Child & Family Behaviour Therapy, 6(1), 17–43.

Kirmayer, L. J., Boothroyd, L. J., Laliberte, A., & Simpson, B.

L. (1999). Suicide prevention and mental health promotion in

First Nations and Inuit communities (Culture and Mental

Health Research Unit Report #9). Institute of Community

and Family Psychiatry, Jewish General Hospital, Montreal.

Knapp, M. (1997). Economic evaluations and interventions for

children and adolescents with mental health problems.

Journal of Child Psychology & Psychiatry, 38(1), 3–25.

Lipsey, M. W. (1992). Juvenile delinquency treatment: a meta-

analytic inquiry into the variability of effects. Thomas Cook

et al., (Eds.), Meta-Analysis for Explanation (pp. 83–126).

New York: Russell Sage Foundation.

Lister-Sharp, D., Chapman, S., Stewart-Brown, S., & Sowden,

A. (1999). Health Promotion in Schools: Two systematic

reviews. Health Technology Assessment, 3(22), 1–207.

Marcotte, D. (1997). Treating depression in adolescence: A

review of the effectiveness of cognitive-behavioural treat-

ments. Journal of Youth and Adolescence, 26(3), 273–283.

Marmot, M., Ryff, C. D., Bumpass, L. L., Shipley, M., &

Marks, N. F. (1997). Social inequalities in health: Next

questions and converging evidence. Social Science &

Medicine, 44(6), 901–910 (Quoted in Stephens, Dulberg, &

Joubert, 1999).

Mathur, S. R., & Rutherford Jr, R. B. (1991). Peer-mediated

interventions promoting social skills of children and youth

with behavioural disorders. Education and Treatment of

Children, 14(3), 227–242.

Odom, S. L., & Strain, P. S. (1984). Peer mediated approaches

to promoting children’s social interaction: A review.

American Journal of Orthopsychiatry, 54, 544–557.

Offord, D. R., Boyle, M. H., & Racine, Y. A. (1992). Outcome,

prognosis and risk in a longitudinal follow-up study.

Journal of the American Academy of Child and Adolescent

Psychiatry, 31, 916–923.

Offord, D. R., Kraemer, H. C., Kazdin, A. D., Jensen, P. S.,

Harrington, R., & Gardner, J. S. (1999). Lowering the

burden of suffering: monitoring the benefits of clinical,

targeted, and universal approaches. In D. P. Keating, & C.

Hertzman (Eds.), Developmental health and the wealth of

nations: Social, biological and educational dynamics

(pp. 293–310). New York: The Guilford Press.

Offord, D. R., & Lipman, E. (1996). Emotional and behavioural

problems. Prevalence and correlates (pp. 119–126). Human

Resources Development Canada and Statistics Canada.

Oxman, A. D. (1994). Checklists for review articles. British

Medical Journal, 309(6955), 648–651.

Oxman, A. D., Cook, D. J., & Guyatt, G. H. (1994). Users’

guide to the medical literature. VI: How to use an

overview. Journal of American Medical Association,

17(272), 1367–1371.

Ploeg, J., Ciliska, D., & Brunton, G. (2000). Effectiveness of

school-based curriculum suicide prevention program for

adolescents. Public Health Research Education and Devel-

opment (PHRED).

Rispens, J., Aleman, A., & Goudena, P. (1997). Prevention of

child sexual abuse victimization: A meta-analysis of school

programs. Child Abuse & Neglect, 21, 975–987.

Stephens, T., Dulberg, C., & Joubert, N. (1999). Mental health

of the Canadian population: A comprehensive analysis.

Chronic Diseases in Canada, 20(3), 118–126.

Thomas, H., Siracusa, L., Gross, G., Beath, L., Hanna, L.,

Michaud, M., Moore, P., Partington, B., Tober, J.,

Voorberg, N., & Brunton, G. (1999). Effectiveness of

school-based interventions in reducing adolescent risk beha-

viour: A systematic review of reviews. Public Health

Research Education and Development (PHRED).

Tilford, S., Delaney, F., & Vogels, M. (1998). Effectiveness of

mental health promotion interventions—A review. Health

Education Authority, Available on the World Wide Web at

www.york.ac.uk/inst/crd/ehc33.pdf.

US Public Health Service. (2000). Report of the Surgeon

General’s Conference on Children’s Mental Health: A

National Action Agenda. Washington, DC: Department of

Health and Human Services.

Yamada, J., DiCenso, A., Feldman, L., Cormillott, P., Wade,

K., Wignall, R., & Thomas, H. (1999). A systematic review

of the effectiveness PF primary prevention programs to

prevent sexually transmitted diseases (STDs) in adolescents.

Public Health Research Education and Development

(PHRED).

Zoritch, B., Roberts, I., & Oakley, A. (1998). The health and

welfare effects of day-care: A systematic review of

randomized controlled trials. Social Science & Medicine,

47(3), 317–327.

ARTICLE IN PRESS G. Browne et al. / Social Science & Medicine 58 (2004) 1367–13841384