2-3 Page paper on case study Social Work Assignment

A Systematic Review of Effective Interventions for Reducing Multiple Health Risk Behaviors in Adolescence

We systematically searched

9 biomedical and social sci-

ence databases (1980–2012)

for primary and secondary

interventions that prevented

or reduced 2 or more ado-

lescent health risk behaviors

(tobacco use, alcohol use,

illicit drug use, risky sexual

behavior, aggressive acts).

We identified 44 random-

ized controlled trials of

universal or selective in-

terventions and were ef-

fective for multiple health

risk behaviors. Most were

school based, conducted

in the United States, and

effectiveformultipleforms

of substance use. Effects

were small, in line with find-

ings for other universal pre-

vention programs. In some

studies, effects for more

than 1 health risk behavior

only emerged at long-term

follow-up.

Integrated prevention pro-

grams are feasible and ef-

fective and may be more

efficient than discrete pre-

vention strategies. (Am J

Public Health. 2014;104:

e19–e41. doi:10.2105/AJPH.

2014.301874)

Daniel R. Hale, PhD, Natasha Fitzgerald-Yau, MSc, and Russell Mark Viner, PhD

ADOLESCENCE IS ASSOCIATED

with an increased prevalence of health risk behaviors, including substance use, sexual risk, and aggressive behavior.1 The vast majority of substance use is initi- ated in adolescence.2,3 In the United Kingdom, adolescence is associated with higher rates of sexually transmitted disease and abortion relative to other age groups.4,5 The majority of young people will experience bullying or aggression during adolescence.6 In addition, adolescent mortality has increased relative to other age groups, largely because of acci- dents and unintentional injuries.7

Early initiation of health risk be- haviors is associated with negative outcomes throughout adolescence and adulthood, such as addiction and substance abuse; poor sexual, mental, and physical health; and lower occupational and educa- tional attainment.8,9 The social and economic costs associated with adolescent risk behaviors have made them a key focus of public health policy initiatives in- ternationally.10

A growing body of research suggests that health risk behaviors often do not occur in isolation. Smoking, drinking, illicit drug use, sexual risk, and aggressive behav- iors are all mutually predictive.11

For drug use and some forms of sexual risk, co-occurrence with other risk behaviors is essentially normative. Previous research sug- gests that co-occurrence of risk behaviors is driven by shared risk factors such as peer influences or sensation seeking or by state- specific traits such as the direct

effects of substance use or ag- gression on other risk behaviors. Common risk factors can be found in many domains, including social, psychological, family, school, and neighborhood.12---14 Evidence also suggests gateway effects, whereby participation in a given health risk behavior leads to increased risk for others, partially attributable to exposure effects and decreases in perceived danger of such behav- iors.15 For example, adolescent smoking and drinking have been linked with subsequent illicit drug use.16

This typical co-occurrence is often not reflected in the organi- zation of policies and interven- tions to reduce adolescent risk behavior. National policy regard- ing adolescent health risk behav- ior is often organized in nonover- lapping risk-specific policies.10

Some intervention developers recognize that single-risk inter- ventions for adolescents may trig- ger effects on other risk behaviors, particularly on multiple forms of substance use.

For several reasons, targeting multiple health risk behaviors (MHRBs) simultaneously may be more effective and efficient than targeting a single risk behavior. Limited funding for prevention interventions requires that inter- ventions reduce health risks effi- ciently, highlighting the impor- tance of synchronized prevention efforts. Time constraints, for ex- ample in schools, also make co- ordinated intervention for multi- ple risks attractive. Furthermore, it is unclear how discrete interven- tions might interact in cases where

they are not coordinated both theoretically and practically, rais- ing the possibility that uncoordi- nated interventions could be in- effective or cause harm.17

Beyond these logistic concerns, research regarding the mecha- nisms for MHRBs suggests that integrated interventions may be essential for the effective preven- tion of risk behaviors. If common risk factors explain co-occurrence of risk behaviors, then targeting those risk factors should prove effective for MHRBs. Gateway theories offer further support for integrated intervention strategies; if a given risk behavior increases risk for another, effective preven- tion strategies for the latter must also focus on the former. For example, sexual intercourse ac- companied by alcohol or illicit drug use is linked to a lower like- lihood of condom use,18 so target- ing substance misuse may be a feasible approach to reducing unsafe sex.

Although the development of integrated interventions for MHRBs requires an understand- ing of their mechanisms, including common risk factors and gateway effects, the existing literature re- garding effective interventions is also a key source of evidence for the development of interventions. The majority of evaluations report on interventions that target 1 risk behavior. However, identifying interventions that have reduced MHRBs can help inform the de- velopment of future interventions by indicating which combinations of risk behaviors can be targeted in coordinated approaches, what

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e19

contexts and approaches are most successful, and what are the other attributes of coordinated inter- ventions, such as duration and participant age.

Limited data exist on effective intervention programs to prevent MHRBs. To date, we are aware of only 1 published review that assessed the effectiveness of in- terventions on MHRBs in young people.13 That review focused ex- clusively on studies reporting concurrently on substance use and sexual risk outcomes. We ex- panded on this work by reviewing additional combinations of out- comes. We undertook a systematic review designed to identify ran- domized controlled trials that reported significant universal or selective intervention effects for at least 2 health risk behaviors among adolescents.

METHODS

We conducted a systematic lit- erature search and selection of articles in accordance with the Preferred Reporting Items for Systematic Reviews and Meta- Analyses (PRISMA) statement.19

We used a standardized search protocol (Appendix A, available as a supplement to this article at http://www.ajph.org) to identify randomized controlled trials that evaluated interventions that re- duced population-level MHRBs (‡ 2 of the following outcomes: to- bacco, alcohol, or illicit drug use; sexual risk behavior; aggressive behavior). We searched 8 elec- tronic databases (PsycINFO, PubMed, Embase, ERIC, British Education Index, Australian Edu- cation Index, Social Sciences Cita- tion Index, CINAHL Plus); in ad- dition we searched the Cochrane Library for reviews on each of the relevant risk behaviors. We then hand-searched references in

review articles and studies and consulted a recent related system- atic review to identify any addi- tional studies.13

Selection Criteria

We selected studies for ap- praisal in a 2-stage process. First, we scanned titles and abstracts identified from the search strategy and excluded them as appropriate with the program EPPI-Reviewer 4 (EPPI-Centre, Social Science Research Unit, Institute of Educa- tion, University of London, UK). We limited our review to peer- reviewed articles published in English between January 1980 and April 2012. Eligible studies (1) were randomized controlled trials with participants who were aged 10 to 19 years at baseline, (2) reported on universal or se- lective interventions (targeting at-risk subpopulations), and (3) reported statistically significant ef- fects on 2 or more of the follow- ing: tobacco use, alcohol, illicit drug use, sexual risk behavior, and aggressive behavior (e.g., delin- quency, truancy) as either primary or secondary outcomes. We ex- cluded studies that evaluated pre- vention programs offered in col- leges or universities, indicative intervention trials (in which par- ticipants were selected because of a priori involvement in the tar- geted risk behavior), and studies that reported attitudinal rather than behavioral changes.

As illustrated in Figure 1, the initial search generated 6299 empirical studies. To ensure interrater reliability, 2 authors reviewed titles or abstracts to as- sess eligibility of studies identified by the database search. This screening and removal of dupli- cates eliminated 6120 items. Most excluded articles were descriptive reports and not intervention stud- ies or their participants did not

meet our age restrictions. We reviewed articles in full when ab- stracts did not provide enough detail to make a decision. We re- trieved 179 full articles and ap- plied our inclusion and exclusion criteria. We discussed discrep- ancies in selections until we reached consensus. Our final re- view comprised 55 articles.

We carried out quality assess- ment with a validated assessment tool that rates the following crite- ria relevant to public health stud- ies: selection bias, allocation bias, confounding, detection bias, data collection, methods, and attrition bias.20 Reviewers then rated each criterion as weak, moderate, or strong. A final global rating was subsequently determined. The quality assessment tool has dem- onstrated good reliability (Cohen’s j = 0.74) and validity.20 We re- solved discrepancies in the quality ratings by discussion.

Data Extraction and Analysis

We recorded detailed informa- tion about each study to identify characteristics of the intervention and its evaluation. We used a data extraction form to collect infor- mation on project title, author, publication date, intervention ob- jectives, setting of intervention (e.g., school, community center, family home), study population (including control group), inter- vention type, domain of effective- ness (i.e., tobacco, alcohol, or illicit drug use; sexual risk; aggressive behaviors), length of follow-up, and key findings. To systematically de- scribe the scope and components of the interventions, we extracted specific features from each article (description, educational theory, duration of intervention). In all cases, 2 authors assessed the articles and extracted the data, with dis- crepancies resolved by joint review and consensus.

The studies chosen for our re- view differed substantially in the following areas: setting, study population, duration, intensity and comprehensiveness of the inter- vention, timing of outcome as- sessments, and outcome measures. The high degree of heterogeneity in both the studies and the reporting of outcomes precluded a meta-analysis. We therefore composed a narrative report of the findings, with interventions cate- gorized by setting (school, com- munity, or family), outcomes, and methodological quality.

We determined effects on health risk behavior outcomes as effect sizes or odds ratios. We selected Cohen d (difference be- tween posttest means divided by the pooled standard deviation) as the effect size index. Where the relevant descriptive statistics were not available, we estimated effect sizes (unadjusted) from available inferential statistics. Depending on the information provided in each study, we calculated effect size(s) from the following data (in order of preference): means, standard deviations or frequencies, and sample sizes for all groups; test of significance value (e.g., F ratio) and significance level; and sample size. When studies pre- sented data from different sub- groups separately (e.g., data for male and female participants presented independently), we calculated effect sizes for each subgroup.

In line with the Cohen classifi- cation,21 we divided effect sizes into 3 levels: small (> 0.2), me- dium (> 0.5), and large (> 0.8). We calculated odds ratios and 95% confidence intervals for dichotomous outcomes and cate- gorized them as small (£ 2.5) medium (> 2.5--- £ 4), and large (> 4).22 We conducted all analyses with an effect size calculator.23

SYSTEMATIC REVIEW

e20 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

RESULTS

The 55 randomized controlled trials that met our inclusion and quality criteria described 44 dis- crete interventions aimed at changing at least 2 types of ado- lescent health risk behavior.

Study Characteristics

Study populations, type and in- tensity of interventions, and out- come measures varied (Table 1). Forty-five studies (82%) took place in the United States; the remaining 10 (18%) took place in Canada, Namibia, Australia, Hong Kong, and Europe. Forty-three studies (78%) evaluated school- based interventions, 11 (23%) of which included a community or family component. The remaining 12 (22%) were either family, community, or Web based. Of the 44 interventions, 14 targeted problem behaviors or aimed to increase healthy behaviors, 17 targeted general substance use, 4 aimed to reduce at least 1 type of substance use and violence or de- linquent behavior, 1 focused on alcohol use and sexual risk, 5 focused on drug use, and sexual risk, alcohol use, and smoking were each the focus of 1 study (Figure 2). The studies took place in suburban, mixed urban, or rural areas. Several were conducted in places with high levels of eco- nomic deprivation.

The ages of participants in the studies ranged from 10 to 21 years, with the majority of inter- ventions targeting adolescents aged 11 to 13 years. Four studies targeted only adolescent girls. Two studies only found significant effects among adolescent boys.

Intervention providers were usually teachers or peer or health educators who had received spe- cialist training and members of the

Title and abstracts identified

n = 6299

Excluded based on abstracts

n = 6120 Not a trial that aimed to prevent or

reduce HRBs, n = 5009

Not an RCT, n = 77

Sample, n = 542

No effect on 2 or more HRBs, n = 489

Not peer-reviewed, n = 3

Number of RCTs

included in this

review

n = 55

Full copies retrieved and

assessed for eligibility

n = 179

Not a trial that aimed to prevent or

reduce HRBs, n = 6

Not an RCT, n = 23

Did not meet sample criteria, n = 20

No effect on 2 or more HRBs, n = 70

Not peer-reviewed, n = 4

Not published in English, n = 1

Excluded based on full text

n = 124

FIGURE 1—Identification of eligible randomized controlled trials (RCTs) in systematic review of effective

interventions for reducing multiple health risk behaviors (HRBs) in adolescence.

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e21

TA B LE

1 — O ve rv ie w o f In te rv e n ti o n s in

S ys te m a ti c R e vi e w o f E ff e c ti ve

In te rv e n ti o n s fo r R e d u c in g M u lt ip le

H e a lt h R is k B e h a vi o rs

in A d o le sc e n c e

In te rv en tio n

St ud y

Se tt in g, Lo ca tio n

Po pu la tio n Ch ar ac te ris tic s

In te rv en tio n Ai m

In te rv en tio n D es cr ip tio n

In te rv en tio n D ur at io n

W eb -b as ed

m ot he r–

da ug ht er pr og ra m

Fa ng

et al .2 4

Fa m ily ho m e ba se d, se ve ra l

As ia n co m m un iti es

As ia n gi rls

ag ed

11 –1 4 ye ar s an d

th ei r m ot he rs w ho

ha d ac ce ss

to a co m pu te r

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

D es ig ne d to im pr ov e gi rls ’ ps yc ho lo gi ca l st at es ,

st re ng th en

su bs ta nc e us e pr ev en tio n sk ill s,

in cr ea se

m ot he r– da ug ht er in te ra ct io ns , en ha nc e

m at er na l m on ito rin g, an d pr ev en t gi rls ’

su bs ta nc e us e.

9 se ss io ns , 1/ w k, 45

m in /s es si on

Ad ol es ce nt Al co ho l

Pr ev en tio n Tr ia l

Ta yl or et al .2 5

Sc ho ol ba se d, Lo s An ge le s,

CA , ar ea

G ra de

7 st ud en ts , 47 % W hi te ,

28 % H is pa ni c, 16 % As ia n,

an d 2. 5%

Af ric an

Am er ic an

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

Le ss on s ab ou t he al th co ns eq ue nc es

of al co ho l

an d dr ug s (w hi ch

co ns tit ut ed

th e co nt ro l

co nd iti on ) co m bi ne d w ith

le ss on s ab ou t

so ci al no rm s ab ou t su bs ta nc e us e an d so ci al

ac ce pt ab ili ty as

w el l as

re si st an ce

sk ill s tr ai ni ng .

U ns pe ci fie d

Ad ol es ce nt s Tr an si tio n

Pr og ra m

Co nn el l et al .2 6

Sc ho ol an d fa m ily ba se d,

no rt hw es t U ni te d St at es

Si xt h- gr ad e st ud en ts an d th ei r

fa m ili es

fr om

3 m id dl e sc ho ol s

in an

et hn ic al ly di ve rs e

m et ro po lit an

co m m un ity

Ta rg et pr ob le m be ha vi or s,

un iv er sa l

M ul til ev el pr og ra m in co rp or at in g Fa m ily

Ch ec k- U p in te rv en tio n an d SH AP e

cu rr ic ul um , m od el ed

af te r Li fe Sk ill s

Tr ai ni ng

pr og ra m . Th e 6 SH AP e

se ss io ns

fo cu se d on

sc ho ol su cc es s,

he al th de ci si on s, bu ild in g po si tiv e

pe er gr ou ps , th e cy cl e of re sp ec t,

co pi ng

w ith

st re ss an d an ge r, an d

so lv in g pr ob le m s pe ac ef ul ly .

6 se ss io ns

Pr ev en tin g al co ho l us e

am on g ur ba n yo ut h

Sc hi nk e et al .2 7

Co m m un ity -b as ed

af te r- sc ho ol

ag en ci es

(e .g ., re cr ea tio n ce nt er s,

tu to rin g se rv ic es , sp or ts ce nt er s) ,

N ew

Yo rk Ci ty

M os tly

Af ric an

Am er ic an

an d

H is pa ni c ch ild re n, ag ed

10 .8 y

at ba se lin e

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

Tw o in te rv en tio n ar m s: (1 ) CD

in te nd ed

to in cr ea se

kn ow le dg e an d ch an ge

at tit ud es

re ga rd in g su bs ta nc e us e

an d te ac h pr ob le m so lv in g, no rm s,

so ci al in flu en ce s, se lf- ef fic ac y, co pi ng

w ith

pr es su re , as se rt iv en es s, re fu sa l

re sp on se s, st re ss re du ct io n, re la xa tio n,

an d so ci al su pp or ts ; (2 ) pa re nt in te rv en tio n

w ith

pr in te d m at er ia l an d vi de ot ap e te ac hi ng

sk ill s fo r he lp in g yo ut hs

ap pl y pr og ra m co nt en t.

10 45 -m in le ss on s, an nu al

bo os te r se ss io ns

ov er 7 y

Al l St ar s Pr og ra m

M cN ea l et al .2 8

14 se co nd ar y sc ho ol s in Le xi ng to n

an d Lo ui sv ill e, KY

St ud en ts ag ed

11 –1 3 y

Su bs ta nc e ab us e pr ev en tio n,

un iv er sa l

Ai m ed

to re du ce

ad ol es ce nt ris k be ha vi or

by ta rg et in g ke y m ed ia to rs st ro ng ly lin ke d

to ad ol es ce nt ris k be ha vi or : no rm at iv e

be lie fs , lif es ty le in co ng ru en ce , co m m itm

en t,

an d bo nd in g to sc ho ol . D el iv er ed

by 2 gr ou ps :

sp ec ia lis ts hi re d by th e pr oj ec t w ho

w er e

ou ts id er s to th e sc ho ol , an d re gu la r

cl as sr oo m te ac he rs .

22 se ss io ns

ov er 1 y

Bi g Br ot he rs Bi g Si st er s

G ro ss m an

an d

Ti er ne y2 9

Co m m un ity

ag en ci es

in Te xa s, O hi o,

M in ne so ta , Pe nn sy lv an ia , N ew

Yo rk ,

Ka ns as , an d Ar iz on a

Ch ild re n an d ad ol es ce nt s fr om

si ng le -p ar en t ho us eh ol ds ,

ag ed

10 –1 6 y

M en to rin g, se le ct iv e

U nr el at ed

ad ul t vo lu nt ee rs pa ire d

w ith

yo ut hs , m et 2– 4· /m o fo r

‡ 1 y; ty pi ca l m ee tin g la st ed

3– 4 h.

1 y

C on tin ue d

SYSTEMATIC REVIEW

e22 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

TA B LE

1 — C o n ti n u e d

BR AV E

G rif fin

et al .3 0

Sc ho ol ba se d, At la nt a, G A

Ei gh th -g ra de

Af ric an

Am er ic an

st ud en ts

fr om

a w or ki ng -p oo r to m id dl e- cl as s

ne ig hb or ho od

Su bs ta nc e us e an d vi ol en ce

pr ev en tio n, se le ct iv e

Ai m ed

to ad dr es s ec on om ic di sa dv an ta ge s

an d pr ev en t al co ho l an d dr ug

us e an d

vi ol en ce

th ro ug h sk ill -b ui ld in g ex er ci se s

w ith

re in fo rc ed

pr ac tic e ac ro ss so ci al

co nt ex ts . Cl as sr oo m te ac he rs , w ho

w er e

ce rt ifi ed

in st ru ct or s, co nd uc te d he al th

ed uc at io n se ss io ns , in cl ud in g tr ai ni ng

in

H IV /A ID S pr ev en tio n an d pe rs on al hy gi en e.

90 -m in se ss io ns

2– 3· /w k fo r

9 w k du rin g sc ho ol ye ar

Cl as sr oo m -c en te re d an d

fa m ily –s ch oo l

pa rt ne rs hi p in te rv en tio n

Fu rr -H ol de n et al .3 1

Sc ho ol an d fa m ily ba se d,

m id -A tla nt ic U S st at es

Fi rs t- gr ad e st ud en ts fr om

9 ur ba n

pr im ar y sc ho ol s in a si ng le pu bl ic

sc ho ol ca tc hm en t ar ea , 80 %

of th e sa m pl e fo llo w ed

un til

ei gh th gr ad e

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

Ai m ed

to re du ce

ea rly

ris k be ha vi or s in pr im ar y

sc ho ol . Th e cl as sr oo m in te rv en tio n ha d

3 co m po ne nt s: cu rr ic ul ar en ha nc em en ts ,

im pr ov ed

cl as sr oo m be ha vi or m an ag em en t

pr ac tic es , an d su pp le m en ta ry st ra te gi es

fo r

ch ild re n no t pe rf or m in g ad eq ua te ly . Th e

fa m ily in te rv en tio n ai m ed

to en ha nc e

pa re nt –s ch oo l co m m un ic at io n an d pr ov id e

pa re nt s w ith

ef fe ct iv e te ac hi ng

an d ch ild

be ha vi or m an ag em en t st ra te gi es .

D ur in g fir st gr ad e

Cl im at e Sc ho ol s

N ew to n et al .3 2

Sc ho ol ba se d, Sy dn ey , Au st ra lia ,

m et ro po lit an

ar ea

Se co nd ar y sc ho ol st ud en ts ag ed

13 y

Pr ev en tio n of al co ho l an d

ca nn ab is us e, un iv er sa l

Al co ho l an d ca nn ab is co ur se

em be dd ed

in th e sc ho ol he al th cu rr ic ul um , de liv er ed

as ca rt oo ns

vi a th e In te rn et .

12 40 -m in se ss io ns

ov er 6 m o

Co m m un iti es

Th at Ca re

H aw ki ns

et al .3 3 ,3 4

Co m m un ity

ba se d, Co lo ra do ,

Ill in oi s, Ka ns as , M ai ne , O re go n,

U ta h, an d W as hi ng to n

Fi ft h- gr ad e st ud en ts in 24

to w ns

D ru g us e an d de lin qu en cy

pr ev en tio n, un iv er sa l

In te rv en tio n co m m un iti es

se le ct ed

13 di ffe re nt

te st ed

an d ef fe ct iv e pr ev en tio n pr og ra m s to

im pl em en t in th e fir st ye ar , 16

in th e

se co nd

ye ar , an d 14

in th e th ird

ye ar .

Pr og ra m s w er e sc ho ol ba se d (e .g ., Al l-S ta rs ,

Li fe Sk ill s Tr ai ni ng ), co m m un ity

ba se d, yo ut h

fo cu se d (e .g ., Bi g Br ot he rs Bi g Si st er s) , an d

fa m ily fo cu se d (e .g ., St re ng th en in g Fa m ili es ).

3 y, fif th –e ig ht h gr ad es ; fo llo w -u p

st ud y 6 y af te r in st al la tio n of

Co m m un iti es

Th at Ca re an d 1 y

af te r st ud y re so ur ce s en de d

Co m pu te r- de liv er ed ,

pa re nt -in vo lv em en t

su bs ta nc e us e

pr ev en tio n

Sc hi nk e et al .3 5

Fa m ily ba se d, N ew

Yo rk Ci ty

Ad ol es ce nt gi rls

(a ve ra ge

m ea n

ag e = 12 .6 7 y)

Su bs ta nc e us e pr ev en tio n,

un iv er sa l fo r gi rls

So ug ht to re du ce

ris k th ro ug h m ot he r– da ug ht er

in te ra ct io ns . In cr ea se d co m m un ic at io n an d

m on ito rin g of ad ol es ce nt be ha vi or w hi le

bu ild in g ad ol es ce nt se lf- es te em

an d

es ta bl is hi ng

ru le s an d co ns eq ue nc es

fo r su bs ta nc e us e.

9 45 -m in se ss io ns

D AR E- Pl us

Pe rr y et al .3 6

Sc ho ol ba se d, M in ne so ta

Se ve nt h- gr ad e st ud en ts in sc ho ol s

w ith

‡ 20 0 st ud en ts

D ru g pr ev en tio n, un iv er sa l

Po lic e of fic er s ta ug ht sk ill s to re si st in flu en ce s

to us e dr ug s an d en ga ge

in vi ol en ce

an d

bu ild

ch ar ac te r (1 0 se ss io ns ), an d pe er s

le d di sc us si on s on

pe er in flu en ce s an d

so ci al sk ill s (4 se ss io ns ); se ss io ns

fo llo w ed

by a th ea te r pr od uc tio n an d m ai le d

an ti– su bs ta nc e us e po st ca rd s.

14 se ss io ns

C on tin ue d

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e23

TA B LE

1 — C o n ti n u e d

D ru g ab us e pr ev en tio n

pr og ra m

G om ez -F ra gu el a

et al .3 7

Sc ho ol ba se d, Sa nt ia go

de

Co m po st el a, Sp ai n

St ud en ts ag ed

14 –1 6 y in 5 pu bl ic

se co nd ar y sc ho ol s

Su bs ta nc e pr ev en tio n, un iv er sa l

Pr es en te d in fo rm at io n on

co ns eq ue nc es

of

su bs ta nc e us e an d ta rg et ed

se lf- es te em ,

de ci si on -m ak in g, an xi et y, so ci al sk ill s,

an d he al th y us e of le is ur e tim

e. Tw o

in te rv en tio n ve rs io ns

te st ed : 1 te ac he r

le d an d 1 re se ar ch er le d.

16 45 –5 0 m in se ss io ns

in fir st

ye ar , 9 re m in de r se ss io ns

in

se co nd

ye ar

Ec oF IT

St or m sh ak

et al .3 8

Sc ho ol an d fa m ily ba se d, U ni te d

St at es

Si xt h- gr ad e st ud en ts fr om

3 pu bl ic

m id dl e sc ho ol s se rv in g an

at -r is k,

lo w -in co m e et hn ic al ly di ve rs e

po pu la tio n

Pr ob le m be ha vi or s, un iv er sa l

Pr ov id ed

“f am ily re so ur ce

ce nt re s” in

sc ho ol s to pr ov id e in fr as tr uc tu re fo r

co lla bo ra tio n be tw ee n st af f an d

fa m ili es

an d pr om ot e po si tiv e

pa re nt in g pr ac tic es . Th es e pr ov id ed

tr ai ni ng , co ns ul ta tio ns

an d fe ed ba ck

fo r pa re nt s.

Av er ag e of 14 6 m in ov er 3 ye ar s

Es pe ci al ly fo r D au gh te rs

O ’D on ne ll et al .3 9

Fa m ily ba se d, N ew

Yo rk Ci ty

Si xt h- gr ad e gi rls

at ba se lin e

(a ge d 11 –1 3 y) , pr ed om in an tly

La tin o an d Af ric an

Am er ic an

fr om

hi gh -p ov er ty pu bl ic sc ho ol s

Al co ho l an d se xu al ris k, se le ct iv e

Fo ur au di o CD s fo r pa re nt s an d th ei r

da ug ht er s w ith

ro le m od el st or ie s

ab ou t 4 fic tio na l fa m ili es . Ai m ed

to in cr ea se

aw ar en es s of th e ris ks

gi rls

m ay fa ce

an d w ha t pa re nt s

ca n do

to pr ev en t ris k be ha vi or s.

4 se ss io ns

at 6- w k in te rv al s

Fa m ili es

U ni da s

Pa nt in et al .4 0

Fa m ily ba se d, Fl or id a

St ud en ts w ith

m ild

pr ob le m s on

‡ 1

su bs ca le (c on du ct di so rd er ,

so ci al iz ed

ag gr es si on , at te nt io n

pr ob le m s) on

th e Re vi se d Be ha vi or

Pr ob le m Ch ec kl is t. Sa m pl ed

fr om

sc ho ol s w ith

pr im ar ily H is pa ni c

st ud en ts in a lo w -in co m e di st ric t

Pr ob le m be ha vi or , se le ct iv e

In te gr at ed

H is pa ni c- sp ec ifi c

cu ltu ra l co nt en t. Pr ov id ed

pa re nt s w ith

sk ill s an d kn ow le dg e

to ra is e ad ol es ce nt s an d m in im iz e

ad ol es ce nt ris k be ha vi or .

9 2- h gr ou p se ss io ns , 10

1- h

fa m ily vi si ts , 4 bo os te r

se ss io ns

at 10 , 16 , 22 ,

an d 28

m o fo llo w -u p

Fa m ily M at te rs Pr og ra m

Ba um an

et al .4 1

Fa m ily ba se d, U ni te d St at es

Ad ol es ce nt s ag ed

12 –1 4 y an d th ei r

fa m ili es

sa m pl ed

fr om

se ve ra l

co nt ig uo us

st at es

To ba cc o an d al co ho l re du ct io n,

un iv er sa l

Ad ol es ce nt –p ar en t pa irs

re ce iv ed

4 bo ok le ts , w ith

fo llo w -u p te le ph on e

ca lls

to pa re nt s fr om

he al th ed uc at or s.

Ad ol es ce nt s w er e re ac he d th ro ug h fa m ily

m em be rs an d no t co nt ac te d di re ct ly

by he al th ed uc at or s.

15 m o

Im PA CT , Fo cu s on

Ki ds

St an to n et al .4 2

Sc ho ol ba se d, Ba lti m or e, M D

Af ric an

Am er ic an

st ud en ts ag ed

13 –1 6 y

fr om

lo w -in co m e sc ho ol s

Ri sk be ha vi or s, un iv er sa l

Th re e in te rv en tio ns

em ph as iz in g

de ci si on -m ak in g, go al se tt in g,

an d in fo rm at io n re ga rd in g un sa fe

be ha vi or s; 1 in te rv en tio n gr ou p

re ce iv ed

bo os te r se ss io ns

to

re vi ew

m at er ia l.

1 Im PA CT

se ss io n, 8 Fo cu s

on Ki ds

se ss io ns

ov er

1 y pl us

4 90 -m in bo os te r

se ss io ns

ov er se co nd

y

Im PA CT , Fo cu s on

Ki ds

Te lc h et al .4 3

Sc ho ol ba se d, Ca lif or ni a

Se ve nt h- gr ad e st ud en ts (a ge = 12

y) Sm ok in g, un iv er sa l

Vi de ot ap es

ab ou t co ns eq ue nc es

of sm ok in g an d

ex am pl es

of pr es su re to sm ok e, sm ok in g

ad ve rt is em en ts , an d st ra te gi es to re si st

pr es su re . O ne

in te rv en tio n gr ou p al so

in vo lv ed

pe er le ad er s.

5 se ss io ns

C on tin ue d

SYSTEMATIC REVIEW

e24 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

TA B LE

1 — C o n ti n u e d

Ke ep in ’ it RE AL

Ku lis et al .4 4

Sc ho ol ba se d, Ph oe ni x, AZ

Se ve nt h- gr ad e st ud en ts of M ex ic an

he rit ag e fr om

35 pu bl ic m id dl e

sc ho ol s

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

In te rv en tio n to en ha nc e cu ltu ra l id en tifi ca tio n,

pr om ot e pe rs on al an tid ru g no rm s an d be ha vi or s,

an d de ve lo p de ci si on -m ak in g an d re si st an ce

sk ill s.

Cu rr ic ul um

ad ap te d fo r cu ltu ra l di ffe re nc es

an d va lu es

fo r 3 in te rv en tio n gr ou ps : ve rs io n 1

re fle ct ed

M ex ic an

Am er ic an

an d M ex ic an

va lu es ,

ve rs io n 2 w as

gr ou nd ed

in Eu ro pe an

Am er ic an

an d Af ric an

va lu es , an d ve rs io n 3 w as

m ul tic ul tu ra l,

w ith

ha lf th e le ss on s fr om

ea ch

of fir st 2 ve rs io ns .

10 le ss on s ov er 2 y

Li fe Sk ill s Tr ai ni ng

Bo tv in et al .4 5

Sc ho ol ba se d, N ew

Yo rk St at e

Pr ed om in an tly

W hi te , m id dl e- cl as s

se ve nt h- gr ad e st ud en ts fr om

10 su bu rb an

ju ni or hi gh

sc ho ol s

Eq ui p st ud en ts w ith

co pi ng

sk ill s to hi nd er dr ug

so ci al

in flu en ce s, un iv er sa l

M ul tic om po ne nt su bs ta nc e ab us e

pr ev en tio n cl as sr oo m cu rr ic ul um

fo cu si ng

on m aj or so ci al , ps yc ho lo gi ca l,

co gn iti ve , an d at tit ud in al fa ct or s th at

ap pe ar to pr om ot e th e us e of to ba cc o,

al co ho l, an d m ar iju an a. Tw o in te rv en tio n

gr ou ps : 1 de liv er ed

by sp ec ia lly tr ai ne d

ol de r st ud en ts an d 1 by tr ai ne d

cl as sr oo m te ac he rs .

20 co gn iti ve be ha vi or al se ss io ns

Li fe Sk ill s Tr ai ni ng

Bo tv in et al .4 6 ,4 7

Sc ho ol ba se d, N ew

Yo rk St at e

Pr ed om in an tly

W hi te se ve nt h- gr ad e

st ud en ts fr om

56 su bu rb an

sc ho ol s;

fo llo w -u p w ith

12 th -g ra de

st ud en ts

re pr es en tin g 60 .4 1%

of or ig in al sa m pl e

Su bs ta nc e ab us e pr ev en tio n,

un iv er sa l

St ud en ts le ar ne d co gn iti ve be ha vi or al sk ill s

fo r bu ild in g se lf- es te em , re si st in g ad ve rt is in g

pr es su re , m an ag in g an xi et y, co m m un ic at in g

ef fe ct iv el y, de ve lo pi ng

pe rs on al re la tio ns hi ps ,

an d as se rt in g th ei r rig ht s. Tw o in te rv en tio n

co nd iti on s: (1 ) 1- d te ac he r w or ks ho p an d

im pl em en ta tio n fe ed ba ck

by pr oj ec t st af f,

(2 ) te ac he r tr ai ni ng

pr ov id ed

by vi de ot ap e

an d no

im pl em en ta tio n fe ed ba ck .

12 cu rr ic ul um

un its

ta ug ht in

15 cl as s se ss io ns

in gr ad e 7,

10 bo os te r cl as s se ss io ns

in

gr ad e 8, an d 5 cl as s se ss io ns

in gr ad e 9

Li fe Sk ill s Tr ai ni ng

G rif fin

et al .4 8

Sc ho ol ba se d, N ew

Yo rk St at e

Se ve nt h- gr ad e st ud en ts fr om

56 se co nd ar y

sc ho ol s in m id dl e- cl as s su bu rb an

an d ru ra l ar ea s. Fo llo w ed

up in yo un g

ad ul th oo d (m ea n ag e = 24

y)

D ru g ab us e pr ev en tio n, un iv er sa l

Ta ug ht st ud en ts co gn iti ve be ha vi or al sk ill s fo r

bu ild in g se lf- es te em , re si st in g pe er pr es su re

an d m ed ia in flu en ce s, m an ag in g an xi et y,

co m m un ic at in g ef fe ct iv el y, de ve lo pi ng

pe rs on al re la tio ns hi ps , an d as se rt in g

th ei r rig ht s an d pr ob le m -s pe ci fic

sk ill s

re la te d to al co ho l an d dr ug

us e, su ch

as

w ay s to be

as se rt iv e in si tu at io ns

w he re th ey

ex pe rie nc ed

in te rp er so na l pr es su re fr om

pe er s

to en ga ge

in su bs ta nc e us e.

6 pa re nt gr ou p– in di vi du al

se ss io ns

an d an

av er ag e

of 7 ph on e ca lls fr om

a

pa re nt in te rv en tio ni st

C on tin ue d

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e25

TA B LE

1 — C o n ti n u e d

Li nk in g th e In te re st s

of Fa m ili es

an d Te ac he rs

D eG ar m o et al .4 9

Sc ho ol ba se d, m et ro po lit an

ar ea

in Pa ci fic

N or th w es t

12 pu bl ic el em en ta ry sc ho ol s in

ne ig hb or ho od s w ith

a hi gh er th an

av er ag e po lic e co nt ac ts

Pr ev en tio n of an tis oc ia l be ha vi or s,

un iv er sa l

Ta ug ht st ud en ts co gn iti ve be ha vi or al sk ill s

fo r bu ild in g se lf- es te em , re si st in g pe er

pr es su re an d m ed ia in flu en ce s, m an ag in g

an xi et y, co m m un ic at in g ef fe ct iv el y, de ve lo pi ng

pe rs on al re la tio ns hi ps , an d as se rt in g th ei r

rig ht s an d pr ob le m -s pe ci fic

sk ill s re la te d to

al co ho l an d dr ug

us e, su ch

as w ay s to be

as se rt iv e in si tu at io ns

w he re th ey ex pe rie nc ed

in te rp er so na l pr es su re fr om

pe er s to en ga ge

in su bs ta nc e us e.

6 pa re nt gr ou p– in di vi du al

se ss io ns

an d an

av er ag e

of 7 ph on e ca lls fr om

a pa re nt in te rv en tio ni st

M ic hi ga n M od el fo r

H ea lth

O ’n ei ll et al .5 0

Sc ho ol ba se d, M ic hi ga n an d

In di an a

St ud en ts (a ve ra ge

ag e = 9. 56

y)

in sc ho ol s w ith

an av er ag e of 46 %

of st ud en ts el ig ib le fo r fr ee

m ea ls

H ea lth

ed uc at io n, un iv er sa l

Sk ill s- ba se d pr og ra m fo cu si ng

on em ot io na l

he al th , su bs ta nc e us e, sa fe ty , an d nu tr iti on

an d ex er ci se . Ta rg et ed

co gn iti ve , at tit ud in al ,

an d em ot io na l ris k fa ct or s fo r he al th -

pr om ot in g be ha vi or .

25 20 –5 0 m in se ss io ns

in

gr ad e 4, 28

in gr ad e 5

M y Fu tu re is M y Ch oi ce

St an to n et al .5 1

Sc ho ol ba se d, N am ib ia , So ut h

Af ric a

St ud en ts ag ed

15 –1 8 y fr om

10

se co nd ar y sc ho ol s

H IV ris k re du ct io n, un iv er sa l

Pr og ra m ba se d on

Fo cu s on

Ki ds

w ith

se ss io ns

in

sc ho ol af te r sc ho ol ho ur s. Fo cu se d on

kn ow le dg e

of re pr od uc tiv e bi ol og y, H IV , an d re la te d ris ks ,

su ch

as us e of al co ho l an d re la tio ns hi p vi ol en ce

an d de ve lo pm en t of sk ill s su ch

as co m m un ic at io n

sk ill s an d de ci si on -m ak in g.

14 se ss io ns

O pe ni ng

D oo rs

D ew itt et al .5 2

Sc ho ol an d fa m ily ba se d, O nt ar io ,

Ca na da

N in th -g ra de

st ud en ts ag ed

14 y

at ris k fo r su ch

pr ob le m s as

dr ug

us e, tr ua nc y, be ha vi or al

pr ob le m s at sc ho ol , an d vi ol en t

an d ot he r an tis oc ia l be ha vi or , fr om

21 sc ho ol s 12

bo ar ds

ac ro ss O nt ar io .

Re du ce

al co ho l an d dr ug s us e

an d de vi an t be ha vi or , se le ct iv e

D es ig ne d to ea se

th e tr an si tio n fr om

el em en ta ry

to hi gh

sc ho ol . St ud en t co m po ne nt ta ug ht

so ci al sk ill s an d he al th -e nh an ci ng

be lie fs an d

va lu es . Pa re nt co m po ne nt fo st er ed

ho m e

en vi ro nm en t to re in fo rc e st ud en t co m po ne nt .

17 st ud en t se ss io ns

an d 5

pa re nt se ss io ns

de liv er ed

ov er 10

w k

Pe er pr es su re re si st an ce

tr ai ni ng

H an se n an d

G ra ha m 5 3

Sc ho ol ba se d, La s An ge le s an d

O ra ng e co un tie s, CA

7t h gr ad e st ud en ts fr om

12 ju ni or

hi gh

sc ho ol s

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

N or m at iv e ed uc at io n: le ss on s on

in fo rm at io n

an d co ns er va tiv e no rm s re ga rd in g

su bs ta nc e us e.

9 le ss on s

Pl an

fo r Su cc es s

W er ch

et al .5 4

Sc ho ol ba se d, Fl or id a

St ud en ts in gr ad es

11 –1 2

(a ve ra ge

ag e = 17

y)

H ea lth

be ha vi or s, un iv er sa l

D es ig ne d to el ic it a po si tiv e se lf- im ag e of

su cc es s th at in co rp or at es

he al th y be ha vi or s.

Co nt ro l gr ou p re ce iv ed

a go al su rv ey as ki ng

re sp on de nt s to id en tif y ob st ac le s (in cl ud in g ris k

be ha vi or s) to su cc es s. G ro up

2 al so

si gn ed

a

co nt ra ct w ith

se lf- co nc or da nt go al s. G ro up

3

co m pl et ed

su rv ey as

w el l as

a ca re er co ns ul ta tio n

th at pr ov id ed

fe ed ba ck

ab ou t th ei r go al s an d ho w

to re ac h th em .

20 -m in se ss io n

C on tin ue d

SYSTEMATIC REVIEW

e26 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

TA B LE

1 — C o n ti n u e d

Po si tiv e Ac tio n Pr og ra m

Be et s et al .5 5

Sc ho ol ba se d, H aw ai i

Fi rs t- an d se co nd -g ra de

st ud en ts

fr om

20 el em en ta ry sc ho ol s th at

ha d ‡ 25 % fr ee

m ea l el ig ib ili ty ;

w er e in th e lo w er 3 qu ar til es

of SA T

sc or es

am on g H aw ai ia n sc ho ol s; w er e

O ah u, M au i, or M ol ok ai pu bl ic sc ho ol ;

an d ha d an nu al st ab ili ty ra te > 80 %

Su bs ta nc e us e, vi ol en t be ha vi or ,

an d se xu al ac tiv ity

pr ev en tio n,

un iv er sa l

Cl as sr oo m te ac he r– de liv er ed

m ul tic om po ne nt

so ci al an d ch ar ac te r de ve lo pm en t pr og ra m

gr ou pe d in to 6 un its : se lf- co nc ep t, m in d an d

bo dy

po si tiv e ac tio ns , so ci al an d em ot io na l

ac tio ns , ge tt in g al on g w ith

ot he rs , be in g

ho ne st w ith

se lf, an d se lf- de ve lo pm en t.

14 0 15 –2 0 m in le ss on s/ y, ov er

5 sc ho ol ye ar s

Po si tiv e Ac tio n Pr og ra m

Li et al .5 6

Sc ho ol ba se d, Ch ic ag o, IL

Fi ft h- gr ad e st ud en ts fr om

14 el em en ta ry

sc ho ol s

Im pr ov e ac ad em ic s, be ha vi or ,

an d ch ar ac te r, un iv er sa l

Ta rg et ed

di st al an d pr ox im al in flu en ce s on

m ul tip le he al th be ha vi or s. In te rv en tio n

sc ho ol s re ce iv ed

ki nd er ga rt en

th ro ug h

ei gh th -g ra de

po rt io n of pr og ra m ’s

cl as sr oo m cu rr ic ul um , sc ho ol st af f

tr ai ni ng

fr om

th e pr og ra m de ve lo pe r,

an d ki ts fo r sc ho ol pr ep ar at io n, sc ho ol w id e

cl im at e de ve lo pm en t, co un se lo rs , an d

fa m ily cl as se s.

> 14 0 15 -m in le ss on s/ gr ad e

de liv er ed

4 d/ w k ov er

2 sc ho ol ye ar s

Pr ep ar in g fo r th e D ru g

Fr ee

Ye ar s

M as on

et al .5 7

Fa m ily ba se d, M id w es t U ni te d

St at es

Si xt h- gr ad e st ud en ts an d th ei r fa m ili es

fr om

22 ru ra l sc ho ol s in 19

co nt ig uo us

st at es

D ru g us e an d pr ob le m be ha vi or

pr ev en tio n, un iv er sa l

D es ig ne d to re du ce

ad ol es ce nt dr ug

us e

an d be ha vi or pr ob le m s w ith

sk ill s- ba se d

cu rr ic ul um

to he lp pa re nt s ad dr es s ris ks

th at ca n co nt rib ut e to dr ug

ab us e w hi le

st re ng th en in g fa m ily bo nd in g by bu ild in g

pr ot ec tiv e fa ct or s.

5 ; 2- h w ee kl y pa re nt in g se ss io ns

Sk ill s- ba se d CD -R O M

in te rv en tio n

Sc hw in n an d

Sc hi nk e5 8

Sc ho ol an d fa m ily ba se d, N ew

Yo rk Ci ty

St ud en ts ag ed

11 y at ba se lin e, m aj or ity

Af ric an

Am er ic an , fr om

sc ho ol s in

im po ve ris he d ar ea s

Al co ho l re du ct io n, se le ct iv e

CD -R O M ta ug ht go al se tt in g, pe er pr es su re ,

re fu sa l sk ill s, an d su bs ta nc e us e no rm s.

O ne

in te rv en tio n co nd iti on

al so

in cl ud ed

pa re nt in te rv en tio n w ith

30 -m in vi de ot ap e

an d pr in t m at er ia ls th at in tr od uc ed

pa re nt s to pr og ra m an d its

go al s an d

ho w pa re nt s co ul d he lp ch ild re n av oi d

su bs ta nc e us e.

10 se ss io ns

+ 3 an nu al bo os te r

se ss io ns

Pr ev en tio n of dr ug

an d

al co ho l ab us e in

N at iv e Am er ic an

yo ut hs

Sc hi nk e et al .5 9

Sc ho ol an d co m m un ity

ba se d,

U S re se rv at io ns

N at iv e Am er ic an

(m ea n ag e = 10 .2 8 y)

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

Co nv en tio na l lif e sk ill s tr ai ni ng

(f or su bs ta nc e

ris k si tu at io ns , pe er in flu en ce s, an d he al th y

lif es ty le s) ta ilo re d to N at iv e Am er ic an

cu ltu re .

A co m m un ity

in vo lv em en t in te rv en tio n ar m

al so

pa rt ic ip at ed

in ac tiv iti es

to ra is e

aw ar en es s of th e su bs ta nc e ab us e pr ev en tio n

m es sa ge

th ro ug h po st er s, fly er s, an d

in fo rm at io na l m ee tin gs .

15 50 -m in se ss io ns

+ se m ia nn ua l

bo os te r se ss io ns

Pr ev en tio n of dr ug

an d

al co ho l ab us e

in N at iv e Am er ic an

yo ut hs

Sc hi nk e et al .6 0

Sc ho ol ba se d, W as hi ng to n St at e

N at iv e Am er ic an s st ud en ts ag ed

11 .8 y

at ba se lin e

D ru g an d al co ho l ab us e

pr ev en tio n, un iv er sa l

Ta ug ht co m m un ic at io n an d co pi ng

sk ill s as

w el l

as sk ill s to an tic ip at e te m pt at io n an d ex pl or e

he al th y al te rn at iv es

to su bs ta nc e us e.

10 se ss io ns

C on tin ue d

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e27

TA B LE

1 — C o n ti n u e d

Pr oj ec t AL ER T

El lic ks on

et al .6 1 ,6 2

Sc ho ol ba se d, So ut h D ak ot a

Se ve nt h- gr ad e st ud en ts fr om

55 m id dl e

sc ho ol s in ru ra l, sm al l to w n, an d

ur ba n ar ea s; fo llo w -u p to ag e 21

y

of st ud en ts w ho

w er e se xu al ly ac tiv e

bu t no t m ar rie d

D ru g us e pr ev en tio n, un iv er sa l

Fo cu se d on

kn ow le dg e an d co ns eq ue nc es

of

dr ug

us e, re du ci ng

ba rr ie rs to dr ug

re si st an ce ,

bu ild in g so ci al no rm s ag ai ns t dr ug

us e, an d

sk ill s fo r re si st in g pr o- dr ug

pr es su re s an d

lin ka ge

to ot he r ris ky be ha vi or s.

14 cl as s le ss on s ov er gr ad es

7 an d 8

Pr oj ec t AL ER T

O rla nd o et al .6 3

Sc ho ol ba se d, So ut h D ak ot a

Se ve nt h- gr ad e st ud en ts

Sm ok in g an d al co ho l us e,

un iv er sa l

In te ra ct iv e te ac hi ng

m et ho ds

fo cu se d on

sm ok in g ce ss at io n an d al co ho l us e w ith

so ci al no rm s ap pr oa ch

to bu ild

se lf- ef fic ac y

an d pr ov id e ro le m od el s.

11 le ss on s in gr ad e 7 an d

3 le ss on s in gr ad e 8

Pr oj ec t Ch ar lie

H ur ry et al .6 4

Sc ho ol ba se d, H ac kn ey , Lo nd on ,

U K

St ud en ts fr om

2 pr im ar y sc ho ol s

D ru g ed uc at io n, un iv er sa l

D ru g pr ev en tio n pa ck ag e ba se d on

th e

lif e sk ill s m od el , ai m in g to de ve lo p

ch ild re n’ s se lf- es te em

an d th ei r ab ili ty

to ex pr es s th ei r fe el in gs an d to re si st pe er

an d so ci al pr es su re an d to in fo rm

th em

of

bo th po si tiv e an d ne ga tiv e ef fe ct s of dr ug s

(m ed ic in es , to ba cc o, an d al co ho l).

13 -m in se ss io ns

w ee kl y fo r

1 or 2 y

Pr oj ec t PA TH S

Sh ek

an d Yu 6 5

Sc ho ol ba se d, H on g Ko ng , Ch in a

St ud en ts ag ed

12 y fr om

se co nd ar y sc ho ol s

Ri sk be ha vi or s, un iv er sa l

Fo cu se d on

de ve lo pm en ta l co nc er ns

(d ru gs ,

se xu al in te rc ou rs e, fin an ce s, re sp on si bi lit y,

lif e m ea ni ng ) an d de ve lo pi ng

st re ng th s

(c on ce rn fo r so ci et y, in fo rm at io n te ch no lo gy

sk ill s) ; ad di tio na l su pp or t gi ve n to th os e

id en tifi ed

as at in cr ea se d ris k (;

20 %

of st ud en ts ).

20 h/ y ov er 3 y

Pr oj ec t SM AR T

G ra ha m et al .6 6

Sc ho ol ba se d, Ca lif or ni a

Se ve nt h- gr ad e st ud en ts , 3 co ho rt s:

19 82 –1 98 3, 19 83 –1 98 4, 19 84 –1 98 5

sc ho ol ye ar s; fo llo w -u p m ea su re d 70 %

in ei gh th gr ad e

D ru g us e pr ev en tio n, un iv er sa l

Th e so ci al sk ill s pr og ra m (S O CI AL ) ta ug ht

st ud en ts so ci al sk ill s fo r re si st in g dr ug

of fe rs . Th e af fe ct m an ag em en t pr og ra m

(A FF EC T) co nt ai ne d no

so ci al sk ill s se ss io ns

fo r 1 co ho rt an d so m e fo r 2 co ho rt s bu t

fo cu se d on

pe rs on al de ci si on -m ak in g, va lu es

cl ar ifi ca tio n, an d st re ss m an ag em en t te ch ni qu es .

12 se ss io ns

of ei th er SO CI AL

or

AF FE CT

pr og ra m ov er 1 y

Pr oj ec t SP O RT

W er ch

et al .6 7

Sc ho ol ba se d, Fl or id a

St ud en ts in gr ad es

9 an d 11

(m ea n

ag e = 15 .2 4 y)

H ea lth

be ha vi or , un iv er sa l

O ne -o n- on e co ns ul ta tio n fo r he al th be ha vi or

sc re en , fit ne ss pr es cr ip tio n, an d in fo rm at io n

on he al th y be ha vi or . D es ig ne d to pr om ot e

po si tiv e se lf- im ag e an d he al th y ac tiv iti es

an d pr es en t ne ga tiv e co ns eq ue nc es

of

su bs ta nc e us e.

12 -m in se ss io n + ta ke -h om e

m at er ia ls

Pr oj ec t To w ar d N o

D ru g Ab us e

D en t et al .6 8

Sc ho ol ba se d, Lo s An gl es , CA

St ud en ts ag ed

14 –1 7 y (9 th –1 1t h gr ad es )

en ro lle d at 3 pu bl ic hi gh

sc ho ol s

D ru g us e pr ev en tio n, un iv er sa l

Cl as sr oo m se ss io ns

ta ug ht sk ill s, su ch

as

he al th y co pi ng

an d se lf- co nt ro l; ed uc at ed

st ud en ts ab ou t m yt hs

an d m is le ad in g

in fo rm at io n th at en co ur ag e su bs ta nc e us e

an d w ar ne d of ch em ic al de pe nd en cy an d

ot he r ne ga tiv e co ns eq ue nc es .

3 50 -m in se ss io ns /w k fo r

3 co ns ec ut iv e w k

C on tin ue d

SYSTEMATIC REVIEW

e28 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

TA B LE

1 — C o n ti n u e d

Ra is in g H ea lth y

Ch ild re n

Br ow n et al .6 9

Sc ho ol an d fa m ily ba se d,

Se at tle , W A

Fi rs t- an d se co nd -g ra de

st ud en ts an d th ei r

fa m ili es

fr om

10 su bu rb an

pu bl ic

el em en ta ry sc ho ol s

Ta rg et de ve lo pm en ta lly

ap pr op ria te ris k an d

pr ot ec tiv e fa ct or s,

un iv er sa l

So ci al de ve lo pm en ta l pr og ra m in co rp or at in g

sc ho ol , fa m ily , an d in di vi du al st ra te gi es . Sc ho ol

in te rv en tio ns

de si gn ed

to en ha nc e le ar ni ng ,

pr ob le m sk ill s, sc ho ol co nn ec te dn es s, an d

ac ad em ic pe rf or m an ce . In di vi du al st ra te gi es

fo cu se d on

ac ad em ic ac hi ev em en t, sc ho ol

co nn ec te dn es s, re fu sa l sk ill s, an d pr os oc ia l

be lie fs ab ou t he al th y be ha vi or s. Fa m ily st ra te gi es

fo cu se d on

pa re nt al sk ill s, ed uc at io na l su pp or t,

de cr ea si ng

fa m ily co nfl ic t, pe er re si st an ce

sk ill s,

an d cl ar ify in g fa m ily st an da rd s an d ru le s ab ou t

st ud en t be ha vi or s.

Te ac he rs in gr ad es

1– 7 re ce iv ed

‡ 6 st af f de ve lo pm en t

w or ks ho p se ss io ns ;

fa m ily in te rv en tio n de liv er ed

du rin g gr ad es

1– 8;

st ud en t in te rv en tio n de liv er ed

in gr ad es 4– 6

Re al Te en

Sc hw in n et al .7 0

W eb

ba se d, 42

U S st at es

an d

4 Ca na di an

pr ov in ce s

G irl s ag ed

14 y at ba se lin e, re cr ui te d

th ro ug h ad ol es ce nt -o rie nt ed

W eb

si te

D ru g ab us e pr ev en tio n, un iv er sa l

fo r gi rls

W eb

si te pr ov id ed

ne w s fe ed , ho ro sc op es , fo ru m ,

an d tr ai ni ng

in se lf- ef fic ac y, co m m un ic at io n,

as se rt iv en es s, go al se tt in g, dr ug

fa ct s, an d

de al in g w ith

si tu at io ns

th at in vo lv ed

dr ug s.

12 on lin e tr ai ni ng

se ss io ns

H ea lth

de ve lo pm en t

pr og ra m

H om el et al .7 1

Sc ho ol ba se d, Sy dn ey , Au st ra lia

St ud en ts fr om

1 se co nd ar y an d 2 in fa nt

an d pr im ar y sc ho ol s

H ea lth

de ve lo pm en t, un iv er sa l

Cl as sr oo m te ac he rs pl an ne d an d w ro te a

he al th /p er so na l de ve lo pm en t cu rr ic ul um

co or di na te d ac ro ss th e sc ho ol ye ar s

(k in de rg ar te n to ye ar 12 ) th at ai m ed

to

br in g ab ou t po si tiv e ch an ge s in he al th

kn ow le dg e, at tit ud es , an d be ha vi or s of ch ild re n.

2 y

Sk ill s en ha nc em en t

pr og ra m

G ilc hr is t et al .7 2

Co m m un ity

ba se d, Pa ci fic

N or th w es t

N at iv e Am er ic an

yo ut hs

(m ea n ag e =

11 .3 4 y)

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

In te rv en tio n si te s re ce iv ed

cu ltu ra lly ta ilo re d

sk ill s en ha nc em en t tr ai ni ng

se ss io ns

de liv er ed

in cl as sr oo m s an d tr ib al ce nt er s. Sk ill s ta ug ht

in cl ud ed

se lf- pr ai se , co m m un ic at io n, an d

id en tif yi ng

pr ec ip ita nt s of al co ho l an d dr ug

us e.

10 60 -m in se ss io ns

St re ng th en in g Fa m ili es

Pr og ra m

Sp ot h et al .7 3 ,7 4

Sc ho ol an d fa m ily ba se d, Io w a

St ud en ts re cr ui te d fr om

33 ru ra l sc ho ol s in

co m m un iti es

w ith

< 85 00

po pu la tio n an d

‡ 15 % el ig ib le fo r fr ee

m ea ls ; ag e 11

at

pr et es t fo llo w -u p af te r 6 y

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

Ta rg et ed

po or di sc ip lin in g sk ill s an d pa re nt –c hi ld

re la tio ns hi ps

in fa m ili es

an d in cr ea se d re si lie nc e

in ad ol es ce nt s by en co ur ag in g em pa th y,

co m m un ic at io n sk ill s, an d re si st an ce

sk ill s.

7 w ee kl y 2- h se ss io ns

St re ng th en in g Fa m ili es

Pr og ra m co m bi ne d

w ith

Li fe Sk ill s Tr ai ni ng

Sp ot h et al .7 5

Sc ho ol an d fa m ily ba se d,

m id w es te rn U S st at e

Se ve nt h- gr ad e st ud en ts fr om

36 ru ra l

sc ho ol s w ith

20 % of fa m ili es

be lo w

or cl os e to po ve rt y le ve l

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

Fa m ily in te rv en tio n de liv er ed

to pa re nt s an d

st ud en ts co nc ur re nt ly in th e ev en in g. Li fe Sk ill s

Tr ai ni ng

de liv er ed

in sc ho ol to pr om ot e sk ill s an d

de ve lo p se lf- m an ag em en t, re si st an ce

sk ill s, an d

ot he r so ci al sk ill s.

Fa m ily in te rv en tio n, 7 se ss io ns

+

4 bo os te r se ss io ns ; Li fe Sk ill s

Tr ai ni ng , 15

se ss io ns

+ 5

bo os te r se ss io ns

So ci al de ve lo pm en t

cu rr ic ul um

an d

sc ho ol /c om m un ity

in te rv en tio n

Fl ay et al .7 6

Sc ho ol an d co m m un ity

ba se d,

Ch ic ag o, IL

Fi ft h- gr ad e st ud en ts fr om

a hi gh -r is k sa m pl e

of 12

po or Af ric an

Am er ic an

in ne r- ci ty an d

su bu rb an

sc ho ol s

Ta rg et ris k be ha vi or s of vi ol en ce ,

pr ov ok in g be ha vi or , su bs ta nc e

us e, sc ho ol de lin qu en cy , an d

se xu al pr ac tic es , se le ct iv e

So ci al de ve lo pm en t cu rr ic ul um

fo cu se d on

so ci al

co m pe te nc e sk ill s ne ce ss ar y to m an ag e si tu at io ns

in w hi ch

hi gh -r is k be ha vi or s oc cu r. Sc ho ol /

co m m un ity

in te rv en tio n ha d so ci al de ve lo pm en t

cu rr ic ul um , sc ho ol w id e cl im at e, pa re nt , an d

co m m un ity

co m po ne nt s.

16 –2 1 le ss on s/ y in gr ad es

5– 8

C on tin ue d

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e29

local community. In 3 cases, the intervention was computer based and used no facilitators. Table 1 also shows the amount of curricu- lum time devoted to each program and whether the program pro- vided booster sessions to reinforce program messages. The interven- tion intensity varied from 4 to 140 sessions, and the duration ranged from 10 weeks to 8 years. Seven studies included booster sessions. The majority of studies incorpo- rated a follow-up measurement of 6 months or more. Studies reported on a variety of substance use, sexual risk, and aggressive behavior measures. All studies relied on self-reported substance use with no biochemical veri- fication, although 1 study also conducted a saliva test to en- courage honest reporting. In the majority of cases, self-reported marijuana use was the drug use outcome measure, although 10 studies (18%) measured other drug use (e.g., amphetamines, tranquilizers).

Overall, 28 studies (51%) were methodologically strong. Twenty- three (89%) of these reported on interventions based in schools, 2 (7%) that were family based, and 3 (11%) that were community based. All 44 studies applied in- tention-to-treat analyses. The ma- jority had a follow-up of 6 months or longer.

Effectiveness

Most effect sizes were small, although several studies reported medium effect sizes. The findings and quality assessment of each study are presented in Tables 2 and 3. School-based interventions. Forty-

four studies evaluated 32 school- based interventions, of which 24 took place exclusively in the school setting. The other 8 school-based interventions

included family or community components, such as homework assignments with parents, parental skills training, or incorporation of prevention skills training into existing community events. Eigh- teen interventions showed a sig- nificant effect for 2 substances (smoking, alcohol use, or illicit drug use). Nine had a positive outcome for all 3 substances. All 9 of these interventions were multi- component and aimed to increase resilience by enhancing adoles- cents’ refusal skills. This was achieved through developing stu- dents’ basic life skills, such as problem-solving skills, personal decision-making, and stress man- agement. Only 1 intervention fo- cused on the health consequences of tobacco use; however, it also incorporated strategies to resist peer pressure. Three interven- tions included a family compo- nent designed to support positive parenting practices and help parents reinforce their child’s re- fusal skills.

The majority of interventions focused on multiple substance use, but 5 were effective for both sub- stance use and aggression and 2 for substance use and sexual risk behavior. Four interventions reported significant effects in all 5 domains. Some interventions reported significant effects for other health risk behaviors several years after program completion. For instance, Project ALERT was effective for alcohol, tobacco, and marijuana use up to 18 months,61

but a later evaluation identified protective effects against sexual risk behavior in young adult- hood.62 The 32 interventions shared characteristics associated with recommendations for effec- tive treatment of adolescent health risk behaviors.79 All studies used empirically validated interven- tion strategies relevant to the

developmental needs of adoles- cents. They also focused on tar- geting the specific risks and pro- tective factors associated with the initiation and maintenance of substance use. The majority of programs recognized the impor- tant influence of peers in risky behavior (Table 2). Family-based interventions. Six

studies evaluated 5 family-based interventions, 2 of which were rated strong. The family-based in- terventions comprised parenting skills, training in groups, home- work tasks requiring parental participation, mailed booklets, home visits, and a mixture of these approaches. Most were based on family interaction theory or social or behavioral learning models and aimed to improve student---parent communication, reinforce refusal skills, teach effective parenting skills, and develop problem- solving approaches. All 5 inter- ventions were effective for 2 health risk behaviors, and 1 pro- duced positive results for 4 health risk behaviors. Two of the inter- ventions had significant effects on both substance use and sexual risk, including an increase in con- dom use. One intervention tar- geted both substance use and ag- gression. All interventions demonstrated that health risk be- havior change was maintained at follow-up (Table 3). Community-based interventions.

We identified 5 studies that eval- uated 4 community-based inter- ventions. They consisted of a skills enhancement program, a youth program with parental reinforce- ment, a multicomponent inter- vention, and a counseling supportive-listening approach. We identified 3 interventions that were effective for 2 health risk behaviors and 1 that was effective for 3 (tobacco and alcohol use and delinquent behavior). One study

TA B LE

1 — C o n ti n u e d

U np lu gg ed

Fa gg ia no

et al .7 7 ,7 8

Sc ho ol ba se d, Au st ria , Be lg iu m ,

G er m an y, G re ec e, It al y, Sp ai n,

an d Sw ed en

St ud en ts ag ed

12 –1 4 y;

fo llo w -u p af te r 18

m o

Su bs ta nc e us e pr ev en tio n,

un iv er sa l

In te rv en tio n ta rg et ed

ex pe rim en ta l an d re gu la r us e of

al co ho l, to ba cc o, an d ill ic it dr ug s wi th cu rri cu lu m

ba se d on

co m pr eh en si ve so ci al in flu en ce ap pr oa ch ,

in co rp or at in g co m po ne nt s of lif e sk ill s in to a co gn iti ve

so ci al in flu en ce m od el . Th re e in te rv en tio n ar m s: gr ou p 1,

ba si c cu rri cu lu m ; gr ou p 2, ba si c cu rri cu lu m wi th pe er

in vo lve m en t; gr ou p 3, ba si c cu rri cu lu m wi th pa re nt

in vo lve m en t.

12 1- h w ee kl y un its

ov er 1

sc ho ol ye ar

SYSTEMATIC REVIEW

e30 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

reported evidence of a medium effect of a skill enhancement pro- gram for Native American youths on decreasing alcohol, marijuana, and inhalant use. One program had a medium effect for smokeless tobacco initiation. For the other outcomes, effect sizes and odds ratios were relatively small. A study that evaluated the All Stars program reported significant ef- fects for sexual risk behaviors 7 years after the end of the program. We did not identify any commu- nity interventions that had a sig- nificant effect for both substance use and sexual risk behavior (Table 3). Web-based interventions. We

found evidence from 1 random- ized controlled trial that a Web- based intervention program can produce a long-term decrease in recent (past 30 days) alcohol use, binge drinking, and tobacco use. However, effect sizes were small for all behaviors.

DISCUSSION

Our systematic review of effec- tive interventions for MHRBs identified 55 studies, describing 44 interventions. These studies

varied considerably in quality, methodology, intervention tech- niques, and results, making cohesive data synthesis difficult. Effect sizes ranged from small to medium. In general, the meth- odological quality of included studies was strong to moderate. The majority of studies took place in the United States and examined school-based inter- ventions that focused on the reduction or prevention of multiple-substance use.

We categorized the majority of effects as small; however, the Cohen categorization system was not specifically devised to assess universal prevention, for which effects are generally smaller than, for example, indicated interven- tion.80 Effect sizes in the reviewed studies were generally in pro- portion with those reported for universal interventions on ado- lescent risk behavior.13,76,81

This is important because it suggests that intervention effects for additional risk behaviors do not cause a dilution of effect sizes.

A large proportion of the in- terventions identified themselves as targeting substance use. This

partially explains why the majority were effective for multiple- substance use. Different forms of substance use appear to be con- ceptually similar, and intervention developers acknowledge that overlapping skills and attributes are necessary to prevent all forms of substance use or misuse. Argu- ably, the risk factors for sexual risk and aggressive acts and for smok- ing, drinking, and drug use are as comparable as the shared risk factors among substance use be- haviors.14 Furthermore, we found no clear differences in the extent to which any of these health risk behaviors are associated with one another.11 Our review suggests that multirisk interventions tar- geting multiple-substance use can also be effective for other health risk behaviors.

The majority of the interven- tions were specifically designed to target MHRBs. However, several were designed to target a single health risk behavior, usually drug use, with intervention outcomes for other health risks character- ized as secondary effects. Al- though we were unable to identify the mechanisms for these second- ary effects, it is likely they relate to

targeting risk and preventive factors common to various risk behaviors or preventing gate- way effects. Interventionists, researchers, and policymakers should be aware of the far- reaching potential of well- designed interventions—even those not focused on MHRBs— and efforts to monitor secondary effects may be warranted.

The wider literature on univer- sal prevention indicates that in- tervention effects are typically strongest immediately after the intervention, and they often de- crease or disappear by long-term follow-ups.82 The general pattern for the interventions identified in our review differed from this norm. Often effect sizes were larger at later follow-ups, and in many cases, significant effects appeared for no or only 1 risk behavior at the first postinterven- tion test, with further significant effects identified at long-term fol- low-up. This is likely related to the mechanisms for intervention ef- fects. If, as theorized, these pro- grams are targeting more distal factors, such as common risk fac- tors, or are preventing gateway effects, it may take longer for effects to emerge, and they may prove more pervasive. For exam- ple, nearly all interventions we reviewed targeted individual at- tributes and skills, such as self- efficacy, and social competen- cies, such as refusal skills and strengthening peer relationships and connectedness. It may take time for effects to trickle down to risk behaviors or for partici- pants to internalize and apply learned skills or attitudes. It was relatively rare for the programs to emphasize risk- specific knowledge. This fits the pattern of results we ob- served, because substance- specific knowledge would be

14.6%

8.3%

18.8%

25%

8.3%

16.7%

Substance use, sexual risk, and aggression

Interventions, %

0 5 10 15 20 25 30

Substance use and aggression

Substance use and sexual risk

Tobacco, alcohol, and drug use

Alcohol and illicit drug use

Tobacco and illicit drug use

Tobacco and alcohol use

8.3%

FIGURE 2—The proportion of interventions (school, family, or community based) targeting tobacco,

alcohol, and drug use; sexual risk; and aggression in systematic review of effective interventions for

reducing multiple health risk behaviors in adolescence.

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e31

TABLE 2—Health Risk Behavior Outcomes for School-Based Prevention Programs in Systematic Review of Effective Interventions for Reducing

Multiple Health Risk Behaviors in Adolescence

Domains for Effectiveness/Intervention Study Quality Assessment Effect Size, Cohen d or OR a (95% CI)

Tobacco and alcohol use

Life Skills Training Botvin et al. 47

Strong 3-y follow-up (adjusted)

Intervention 1: training workshop and implementation feedback

Smoked in past mo, OR = 1.33 (1.11, 1.59), small

Smoked in past wk, OR = 1.23 (1.02, 1.49), small

Frequency of getting drunk, OR = 1.29 (1.09, 1.54), small

Intervention 2: training video, no feedback

Smoked in past mo, OR = 1.40 (1.18, 1.67), small

Smoked in past wk, OR = 1.39 (1.15, 1.67), small

Pack-a-day smoker, OR = 1.37 (1.06, 1.79), small

Frequency of getting drunk, OR = 1.35 (1.15, 1.59), small

Strengthening Families Program Spoth et al. 74

Strong Annual up to 6-y follow-up

Reduced growth rates for initiation of alcohol use without

parental permission

Reduced growth rates for lifetime cigarette use

Reduced growth rates for incidence of drunkenness

Drug abuse prevention program Gomez-Fraguela et al.37 Moderate One-y follow-up (unadjusted)

Teacher led

Monthly frequency of beer, 0.23 (0.06, 0.40), small

Monthly frequency of tobacco, 0.29 (0.11, 0.46), small

Researcher led

Monthly frequency of spirits, 0.24 (0.05, 0.42), small

Health development program Homel et al.71 Moderate 2-y follow-up

Not smoking (boys only), 0.13 (0.02, 0.25), small

Daily smoking reduced (boys only), 0.14 (0.02, 0.25), small

Not drinking (boys only), 0.18 (0.05 0.30), small

Daily drinking rates (boys only), 0.35 (0.23, 0.47), small

Daily drinking rates (girls only), 0.13 (0.01, 0.25), small

Adolescent Alcohol Prevention Trial Taylor et al. 25

Weak Annual until 4-y follow-up

Reduced growth for recent alcohol use, lifetime alcohol use,

lifetime drunkenness, recent cigarette use, and lifetime

cigarette use

Project SPORT Werch et al.67 Weak 3-mo follow-up (unadjusted)

30-d alcohol frequency, 0.32 (0.16, 0.49), small

30-d alcohol quantity, 0.32 (0.16, 0.49), small

30-d heavy use of alcohol (‡ 5 drinks in a row), 0.27 (0.11, 0.44), small

Length of time using alcohol, 0.29 (0.13, 0.46), small

Stage of alcohol initiation (from “never will try” to “have

started using”), 0.35 (0.19, 0.52), small

30-d cigarette frequency, 0.19 (0.00, 0.35), small

1-y follow-up

Length of time using alcohol, 0.20 (0.03, 0.37), small

30-d cigarette frequency, 0.28 (0.10, 0.45), small

Stage of cigarette initiation, 0.33 (0.16, 0.50), small

Continued

SYSTEMATIC REVIEW

e32 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

TABLE 2—Continued

Tobacco and illicit drug use

Classroom component and family–school partnership Furr-Holden et al.31 Strong Followed up each y first–eighth grade (adjusted)

Classroom component

Smoking initiation, OR = 1.22 (0.52, 7.33), small

Illicit drug use, OR = 2.44 (1.11, 6.69), small

Family–school partnership

Smoking initiation, OR = 1.63 (0.64, 49), small

Plan for Success Werch et al.54 Strong 1-mo follow-up

Reduction in length of time using alcohol

Reduction in length of time using marijuana

Project Charlie Hurry et al.64 Moderate 4-y follow-up

Ever smoked for subset 1, 0.90 (0.16, 1.63), large

Ever smoked for subset 1/2, 0.28 (0.06, 0.50), small

Ever tried an illegal drug, 0.29 (0.07, 0.51), small

Alcohol and illicit drug use

Raising Healthy Children Brown et al.69 Strong Posttest (adjusted)

Less growth in frequency of alcohol use

Less growth in frequency of marijuana use

Project Toward No Drug Abuse Dent et al.68 Strong 1-y follow up

Reduction in frequency of hard drug use (30 d)

Reduction in frequency of alcohol use (30 d)

Opening Doors Dewitt et al. 52

Strong Posttest (adjusted)

‡ 5 drinks on 1 occasion, 0.35 (0.05, 0.66), small Frequency of marijuana use (monthly), 0.40 (0.10, 0.71), small

BRAVE Griffin et al.30 Strong 1-y follow-up (after baseline; adjusted)

Frequency of alcohol use (past 30 d), 0.60, medium

Frequency of marijuana use (past 30 d), 0.41, small

Climate Schools Newton et al. 32

Strong 6-mo follow-up (unadjusted change scores from pretest)

Average weekly alcohol consumption, 0.20 (0.04, 0.36), small

Frequency of marijuana use (past 3 mo), 0.19 (0.03, 0.34), small

Michigan Model for Health O’neill et al.50 Strong Posttest (unadjusted)

Ever consumed alcohol, OR = 1.51 (1.11, 2.04), small

Drank in past 30 d, OR = 1.73 (1.12, 2.66), small

Ever smoked cigarettes, OR = 1.54 (1.05, 2.27), small

Smoked in past 30 d, OR = 3.17 (1.67, 6.01), medium

Keepin’ it REAL Kulis et al. 44

Strong 14-mo follow-up

Multicultural version

Recent substance use, 0.05, small

Recent alcohol use, 0.04, small

Recent marijuana use, 0.04, small

Project Alert Orlando et al.63 Moderate Posttest (adjusted)

Past-mo smoking, 0.10 (0.04, 0.17), small

Alcohol misuse (including weekly use, binging, and negative

consequences of alcohol), 0.06 (0.00, 0.12), small

Strengthening Families Program and Life Skills Training Spoth et al. 75

Moderate 1-y follow-up (unadjusted)

Rate of lifetime alcohol use, 0.14 (0.01, 0.28), small

Rate of lifetime marijuana use, 0.15 (0.02, 0.28), small

Continued

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e33

TABLE 2—Continued

Tobacco, alcohol, and Illicit drug use

Life Skills Training Botvin et al.45 Strong Posttest (adjusted)

Peer led

Tobacco use (monthly), 0.11 (0.00, 0.23), small

Marijuana use (monthly), 0.13 (0.01, 0.25), small

Marijuana use (weekly), 0.15 (0.02, 0.26), small

Frequency of drunkenness, 0.14 (0.01, 0.27), small

Amount of alcohol consumed, 0.15 (0.02, 0.29), small

Life Skills Training Botvin et al.46 Strong Posttest

Condition 1

Reduced tobacco use

Reduced marijuana use

Condition 2

Reduced tobacco use

Reduced marijuana use

Reduced frequency of getting drunk

Adolescents Transition Program Connell et al.26 Strong Posttest (age 11–17 y)

Less growth in tobacco use

Less growth in alcohol

Less growth in marijuana use

Unplugged Faggiano et al. 77

Strong 3-mo follow-up

Cigarette smoking (daily), OR = 1.43 (1.06, 1.92)

1 episode of drunkenness (30 d), OR = 1.39 (1.11, 1.72)

‡ 3 episodes of drunkenness (30 d), OR = 1.45 (1.01, 2.08) Marijuana use (30 d), OR = 1.30 (1.00, 1.67)

Unplugged Faggiano et al.78 Strong 18-mo follow-up

Any episode of drunkenness, OR = 1.25 (1.03, 1.49)

Frequent episodes of drunkenness, OR = 1.61 (1.23, 2.13)

Marijuana use (past 30 d), OR = 1.35 (1.00, 1.89)

Strengthening Families Program Spoth et al. 73

Strong 48-mo follow-up (adjusted)

Ever drank alcohol, OR = 2.13 (1.28, 3.57), small

Ever drank without parental permission, OR = 2.17 (1.35, 3.45), small

Ever been drunk, OR = 2.27 (1.37, 3.70), small

Ever smoked, OR = 2.04 (1.25, 3.33), small

Ever used marijuana, OR = 2.70 (1.28, 5.88), medium

Past-mo drinking, 0.26 (0.03, 0.49), small

Past-mo cigarette use, 0.31 (0.08, 0.54), small

Peer pressure resistance training Hansen and Graham 53

Strong Posttest

Alcohol, 0.14 (0.06, 0.22), small

Marijuana, 0.11 (0.03, 0.19), small

Tobacco use, 0.09 (0.01, 0.17), small

Project SMART Graham et al.66 Moderate 1 y follow-up

Cigarette use

Marijuana use

Alcohol use

Skills-based CD-ROM intervention Schwinn and Schinke 58

Moderate 6-mo follow-up (adjusted)

Past-mo use of alcohol, 0.29 (0.02, 0.55), small

Past-mo use of marijuana, 0.36 (0.10, 0.63), small

Continued

SYSTEMATIC REVIEW

e34 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

TABLE 2—Continued

Prevention of drug and alcohol abuse

in Native American youths

Schinke et al.60 Weak Posttest

Smokeless tobacco use in past 2 wk

Alcohol use in past 2 wk

Marijuana use in past 2 wk

Nonmedical drug use in past 2 wk

6-mo follow-up

Smokeless tobacco use in past 2 wk

Alcohol use in past 2 wk

Marijuana use in past 2 wk

Inhalant use in past 2 wk

Smoking use in past 2 wk

Prevention of drug and alcohol abuse

in Native American youths

Schinke et al. 59

Weak 30-mo follow-up (unadjusted)

‡ 7 uses of smokeless tobacco in past wk, OR = 1.61 (1.08, 2.38), small ‡ 4 drinks in past wk, OR = 1.25 (0.93, 1.67), small 42-mo follow-up

‡ 4 uses of marijuana in past wk, OR = 2.33 (1.56, 3.34), small ‡ 7 uses of smokeless tobacco in past wk, OR = 1.89 (1.35, 2.63), small ‡ 4 drinks in past wk, OR = 1.45 (1.12, 1.89), small ‡ 4 uses of marijuana in past wk, OR = 2.33 (1.56, 3.34), small

ImPACT Focus on Kids Stanton et al. 42

Weak 2-y follow-up (adjusted)

Both interventions (combined) compared with control group in past 6 mo

Mean number of school suspensions, 0.14 (0.00, 0.28), small

Carried a bat as a weapon, OR = 2.50 (1.39, 4.35), medium

Smoked cigarettes, OR = 2.04 (1.41, 2.94), small

Used illicit drugs other than marijuana, OR = 4.17 (1.72, 10.00), large

Asked sexual partner if he or she always used a condom at past intercourse,

OR = 1.91 (1.40, 2.61), small

ImPACT Focus on Kids Telch et al. 43

Weak Posttest

Peer leader

Transition from nonsmoking to experimental smoking

Transition from nonsmoking to regular smoking

Transition from experimental to regular smoking

Adoption rates for alcohol

Adoption rates for marijuana

Video only

Transition from nonsmoking to regular smoking

Transition from experimental to regular smoking

Adoption rates for alcohol

Substance use and aggression

Linking the Interests of Families and Teachers DeGarmo et al.49 Strong Grades 5–12 (adjusted)

Reduced rates of growth in use of tobacco and illicit drugs for girls

Lower average levels of use for tobacco, alcohol, and illicit drugs for all youths

Tobacco initiation, 10% reduced risk

Alcohol initiation, 9% reduced risk

Reductions in playground aggression during fifth grade

Positive Action Program Li et al. 56

Moderate 3-y follow-up

Reduction in substance use index

Reduction in serious violent behaviors

Continued

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e35

TABLE 2—Continued

DARE-plus Perry et al.36 Moderate 6-mo and 18-mo follow-up (difference in growth rate, unadjusted)

For boys only (no significant results for girls)

Alcohol behavior and intentions, 0.07 (0.01, 0.15), small

Past-y drinking, 0.07 (0.01, 0.15), small

Past-mo drinking, 0.07 (0.01, 0.15), small

Tobacco behaviors and intentions, 0.07 (0.01, 0.15), small

Current smoking, 0.07 (0.01, 0.15), small

Drug behavior and intentions, 0.07 (0.01, 0.15), small

Physical victimization, 0.08 (0.00, 0.16), small

My Future is My Choice Stanton et al.51 Moderate Posttest

Condom use among baseline virgins, OR = 7.14 (1.15, 50.00), large

6-mo follow-up (unadjusted)

Discussing partner’s history with new sexual partner, OR = 1.59

(1.03, 2.45), small

Past 6-mo alcohol use, OR = 1.69 (1.05, 2.70), small

12-mo follow-up

Abstinence among baseline virgins, OR = 2.07 (1.15, 3.73), small

EcoFIT Stormshak et al.38 Weak Annual follow-up for 3 y

Antisocial behavior in past mo (including stealing, carrying a weapon,

and physical aggression)

30-d cigarette use

30-d alcohol use

30-d marijuana use

Substance use and sexual risk

Project ALERT Ellickson et al.61 Strong 18-mo follow-up

Reduced cigarette initiation

Reduced marijuana initiation

Reduced alcohol misuse

Project ALERT Ellickson et al. 62

Strong 5/7-y follow-up

Unprotected sexual intercourse because of drug use (14% reduction)

Sexual intercourse with multiple partners (12.5% reduction)

All Stars Program McNeal et al.28 Weak Teacher led

Alcohol use, 0.06, small

Cigarette use, 0.06, small

Smokeless tobacco use, 0.04, small

Inhalant use, 0.07, small

Substance use, sexual risk, and aggressive behaviors

Positive Action Program Beets et al. 55

Strong Posttest

Substance use (lifetime), OR = 1.45 (0.33, 1.94), small

Violent behaviors, OR = 1.39 (0.32, 2.70), small

Sexual activity, OR = 3.13 (0.09, 1.95), medium

Project PATHS Shek and Yu65 Strong Semiannual until 3-y follow-up

Delinquency in past 6 mo (included stealing, truancy, damaging

property, assault)

6-mo ketamine use

6-mo psychotropic drug use

Sexual intercourse in past 6 mo

Trespassing

Continued

SYSTEMATIC REVIEW

e36 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

less likely to influence multiple risk behaviors simultaneously and would also be more likely to disappear over time.

Several effective interventions made use of long-term booster sessions, delivered months or years after delivery of the main portion of the intervention. Nei- ther the wider literature83 nor our review provide much evidence that the absolute length of inter- vention programs is related to effectiveness. However, the use of booster sessions has been clearly linked to an increase in magnitude and longevity for intervention effects.84,85 This may explain why intervention effects for many studies persisted over time.

The majority of identified in- terventions took place in schools. Schools offer a useful context (and a captive audience) for the wide- spread dissemination of universal adolescent prevention programs. Systematic reviews in adolescent

prevention in several domains suggest that school-based inter- ventions are common.82,86,87

However, in the prevention of MHRBs, targeting schools may not only be practical, but also sub- stantially contribute to effective- ness. This is because of the im- portance of school and peer effects for many risk behaviors. School climate, including student partici- pation and engagement and teacher---student relationships, is associated with several health risk behaviors.88,89 Also, peer effects such as social mimicry,90 peer pressure, and social norms18,91

contribute to an increase in likeli- hood of risk behaviors, and these can be perpetuated in the school context. Targeting these common risk factors has been associated with reduced risk behavior in several domains.92 School-based interventions provide a platform for effectively targeting common school and peer risk factors for MHRBs. However, it is important

to note that similar reasoning can be applied to family-based interventions, and our review affirms their effectiveness, both individually and in com- bination with school-based interventions.

Limitations

The identified studies varied considerably in quality; although we found most to be of adequate quality, all suffered from some limitations that compromised re- liability and validity (e.g., study dropout, weak outcome measures, selection bias, confounding). All risk behavior measures were self-reported. Although this is the norm in intervention studies, self-report is subject to bias from both over- and underre- porting of behaviors.93 Many studies reported analyses of a large number of behavioral outcomes, with few reporting adjustment for multiple hypothe- sis testing. Some positive findings

may therefore have been attrib- utable to chance.

Studies varied substantially in outcome measures, analytic methods, and adjustment for con- founders, thus making collating or comparing findings difficult. A similar problem applies to the in- terventions themselves: they varied in methods, theoretical underpinning, context, and par- ticipants, making it difficult to draw general conclusions about effective interventions. The ma- jority of studies were conducted in the United States, so caution is warranted in generalizing find- ings to other countries. Further- more, we included only ran- domized controlled trials, so interventions that did not lend themselves to evaluation by that method but that may have been effective in reducing MHRBs would not be represented in our results. Such interventions might involve changing legal frameworks, law enforcement

TABLE 2—Continued

Social development curriculum

and school/community intervention

Flay et al.76 Moderate Posttest

Social development

Violent behavior, 0.31, small

Substance use, 0.42, small

School/community intervention

Violent behavior, 0.41, small

Provoking behavior, 0.41, small

School delinquency, 0.61, medium

Substance use, 0.45, small

Recent sexual intercourse, 0.65, medium

Condom use, 0.66, medium

Life Skills Training Griffin et al. 48

Weak 10-y follow-up

Reduced growth in alcohol

Reduced growth in marijuana intoxication

HIV risk index score, OR = 1.43 (1.04, 1.96), small

Note. CI = confidence interval; OR = odds ratio. All odds ratios < 1 were converted to > 1 for ease of interpretation. Only intervention conditions with significant program effects are included. Effect sizes are presented for all studies in which effect sizes are presented in text or sufficient information is available to calculate them. Significant effects were always in favor of the intervention program. For effect sizes noted as adjusted, the study authors adjusted for key characteristics such as gender, ethnicity, socioeconomic status, or preintervention substance use. a The effect sizes were reported as Cohen d except where indicated to be odds ratios (OR). All odds ratios above 1 indicate favourable outcomes in the intervention group.

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e37

TABLE 3—Health Risk Behavior Outcomes for Community-, Family-, and Web-Based Prevention Programs in Systematic Review of Effective

Interventions for Reducing Multiple Health Risk Behaviors in Adolescence

Domains for Effectiveness/Intervention Study Quality Assessment Effect Size, Cohen d or OR a (95% CI)

Family based

Tobacco and alcohol use: Family Matters Program Bauman et al. 41

Strong 3- and 12-mo follow-up (adjusted)

Smoking, OR = 1.36 (1.02 [lower bound]), b small

Drinking alcohol, OR = 1.34 (1.06 [lower bound]), b small

Tobacco and illicit drug use: computer-delivered,

parent-involvement substance use prevention

Schinke et al. 35

Moderate 1-y follow-up (unadjusted)

30-d alcohol use, 0.26 (0.13, 0.40), small

30-d marijuana use, 0.14 (0.01, 0.28), small

30-d illicit prescription drug use, 0.14 (0.01, 0.28), small

30-d inhalant use, 0.08 (0.05, 0.21), small

2-y follow-up

30-d alcohol use, 0.30 (0.16, 0.43), small

30-d marijuana use, 0.20 (0.06, 0.34), small

30-d illicit prescription drug use, 0.13 (0.01, 0.26), small

30-d inhalant use, 0.06 (0.07, 0.20), small

Alcohol and illicit drug use: Web-based

mother–daughter program

Fang et al.24 Moderate 6-mo follow-up (posttest)

Alcohol use (30 d), 0.08, small

Marijuana use (30 d), 0.07, small

Prescription drugs for nonmedical purposes (30 d), 0.04, small

Substance use and aggression: Preparing

for the Drug Free Years

Mason et al. 57

Moderate 5 waves of data

Slower rate of linear increase in polysubstance use

Slower rate of linear increase in delinquency

Substance use and sexual risk

Especially For Daughters O’Donnell et al.39 Strong 3-mo follow-up (adjusted)

Used alcohol or been drunk, OR = 2.63 (1.03, 6.67), medium

Sexual risk, OR = 2.56 (1.14, 5.88), medium

Familias Unidas Pantin et al. 40

Moderate 6-mo, 18-mo, and 30-mo follow-up (unadjusted)

Growth of 30-d substance use (smoking, drinking, and illicit drug use), 0.25, small

Growth for condom use, 0.30, small

Community based

Tobacco and alcohol use: preventing alcohol

use among urban youth

Schinke et al.27 Moderate 7-y follow-up (unadjusted), both intervention arms compared with control group

30-d alcohol consumption, 0.18 (0.03, 0.38), small

30-d binge drinking, 0.16 (0.04, 0.37), small

30-d cigarette use, 0.21 (0.00, 0.41), small

Alcohol and illicit drug use: skills

enhancement program

Gilchrist et al. 72

Strong 6-mo follow-up from pretest

Alcohol use, 0.70 (0.29, 1.12), medium

Marijuana use, 0.54 (0.13, 0.96), medium

Big Brothers Big Sisters Grossman and Tierney 29

Moderate Inhalant use, 0.54 (0.13, 0.96), medium

18-mo follow-up

Significantly less likely to have started using illegal drugs or alcohol

Continued

SYSTEMATIC REVIEW

e38 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

strategies, social services, or pub- lic health guidelines.

We included in our review only studies in which the intervention was effective for 2 or more risk behaviors. We did not include all studies that assessed or reported 2 or more health risk behavior out- comes, effective or not. Our rea- sons were pragmatic. We believe that reporting bias, which restricts reporting of results in abstracts largely to positive findings, partic- ularly for secondary outcomes, would make attempts to include the latter set of studies accurately essentially impossible. In addition, the sheer scale of identifying all trials that assessed 2 or more risk behavior outcomes in adolescents would make this infeasible. Be- cause our aim was to identify effective interventions in a devel- oping field rather than to assess the effectiveness of a particular

intervention, we chose not to at- tempt to include studies that were not effective across 2 or more behaviors.

It is possible that our review missed some trials that were ef- fective for more than 1 risk be- havior but did not report this in the abstract. Because our findings suggest that even interventions designed to target a single risk can have beneficial effects on other behaviors, some programs might not have been identified as effec- tive for multiple behaviors if other risk behaviors were not measured. Furthermore, interventions might have been excluded from the re- view if data were split into multi- ple publications, each focusing on different outcomes. More important, we could not ascertain which characteristics of effective interventions differentiated them from ineffective ones.

Although it is important to iden- tify which programs are effica- cious for multiple health risk behaviors, further research is needed to determine what factors are associated with suc- cessful (and unsuccessful) pre- vention efforts.

Conclusions

Integrated risk prevention pro- grams can be effective across a range of health risk behaviors in adolescence, with effect sizes that are generally small but compara- ble to those of interventions that target single risk factors. The evi- dence is strongest for various forms of substance use and for school-based interventions. These interventions appear to be suc- cessfully targeting common risk factors for a range of health be- haviors, contributing to both the breadth and the longevity of their

effectiveness. Evidence for inter- ventions outside the United States is very limited, however, and a sub- stantial proportion of studies in- volved high-risk ethnic minority groups in the United States. Further work is needed to assess the gen- eralizability of these findings out- side North America.

Our review serves as a compre- hensive survey of effective inter- ventions for MHRBs in adolescence that can be used by practitioners and policymakers to guide further development of intervention strate- gies in preventing MHRBs. j

About the Authors The authors are with the General and Adolescent Paediatrics Unit, Institute of Child Health, University College London, UK. Correspondence should be sent to Daniel R.

Hale, General and Adolescent Paediatrics Unit, Institute of Child Health, University College London, 30 Guilford St, London, UK, WC1N 1EH (e-mail: [email protected]).

TABLE 3—Continued

Substance use and aggression

Communities That Care Hawkins et al.33 Strong Grade 5–8 (adjusted)

Alcohol use initiation, OR = 1.60 (1.05, 2.44), small

Cigarette initiation, OR = 1.79 (1.09, 2.92), small

Smokeless tobacco initiation, OR = 2.34 (1.34, 4.09), small

Delinquent behavior initiation, OR = 1.41 (1.05, 1.89), small

Grade 8 (adjusted)

Alcohol use (past 30 d), OR = 1.25 (1.04, 1.52), small

Communities That Care Hawkins et al.34 Strong Smokeless tobacco use (past 30 d), OR = 1.79 (1.23, 2.62), small

Binge drinking (past 2 wk), OR = 1.40 (1.07, 1.84), small

Delinquent behaviors (past y), OR = 1.34 (1.20, 1.49), small

Grade 10 (adjusted)

Tobacco use (past 30 d), OR = 1.27 (1.01, 1.56), small

Any delinquency (past y), OR = 1.20 (1.01, 1.45), small

Any violence (past y), OR = 1.33 (1.03, 1.72), small

Web based

Tobacco and alcohol use: RealTeen Schwinn et al.70 Weak 6-y follow-up (unadjusted)

Past mo alcohol use, 0.29 (0.08, 0.49), small

Past mo heavy drinking, 0.20 (0.00, 0.41), small

Past mo cigarette use, 0.23 (0.03, 0.44), small

Note. CI = confidence interval; OR = odds ratio. All odds ratios < 1 were converted to > 1 for ease of interpretation. Only intervention conditions with significant program effects are included. Effect sizes are presented for all studies in which effect sizes are presented in text or sufficient information is available to calculate them. Significant effects were always in favor of the intervention program. For effect sizes noted as adjusted, the study authors adjusted for key characteristics such as gender, ethnicity, socioeconomic status, or preintervention substance use. a The effect sizes were reported as Cohen d except where indicated to be odds ratios (OR). All odds ratios above 1 indicate favourable outcomes in the intervention group. b Only the lower bound of the CI was reported in this article.

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e39

Reprints can be ordered at http://www.ajph. org by clicking the “Reprints” link.

This article was accepted January 2, 2014.

Contributors D. R. Hale established the eligibility criteria and search strategy. D. R. Hale and N. Fitzgerald-Yau conducted database searches, quality assessment, and data extraction; applied eligibility criteria to identified studies; calculated effect sizes; and prepared the article. R. M. Viner was the project leader; contributed to study design, including search strategy and eligibility criteria; and supervised article preparation.

Acknowledgments The Policy Research Unit in the Health of Children, Young People and Fami- lies (CPRU) is funded by the Department of Health Policy Research Program.

We thank members of the CPRU: Terence Stephenson, Catherine Law, Becky Fauth, Ruth Gilbert, Miranda Wolpert, Amanda Edwards, Steve Morris, Helen Roberts, and Catherine Shaw.

Note. The views expressed in this in- dependent report are not necessarily those of the Department of Health.

Human Participant Protection No protocol approval was required because only publicly available data were used.

References 1. World Health Statistics 2012. Ge- neva, Switzerland: World Health Organi- zation; 2012.

2. Degenhardt L, Chiu W-T, Sampson N, et al. Toward a global view of alcohol, tobacco, cannabis, and cocaine use: find- ings from the WHO World Mental Health Surveys. PLoS Med. 2008;5(7):e141.

3. Oh DL, Heck JE, Dresler C, et al. Determinants of smoking initiation among women in five European coun- tries: a cross-sectional survey. BMC Public Health. 2010;10(1):74.

4. Health Protection Agency. STI an- nual data tables. Available at: http:// www.hpa.org.uk/stiannualdatatables. Accessed November 10, 2012.

5. Abortion Statistics, England and Wales: 2010. London, UK: Department of Health; 2011.

6. Rigby K, Smith P. Is school bullying really on the rise? Soc Psychol Educ. 2011;14(4):441---455.

7. Viner RM, Coffey C, Mathers C, et al. 50-year mortality trends in children and young people: a study of 50 low-income,

middle-income, and high-income countries. Lancet. 2011;377(9772):1162---1174.

8. Mirza KAH, Mirza S. Adolescent substance misuse. Psychiatry. 2008; 7(8):357---362.

9. Flory K, Lynam D, Milich R, Leukefeld C, Clayton R. Early adolescent through young adult alcohol and mari- juana use trajectories: early predictors, young adult outcomes, and predictive utility. Dev Psychopathol. 2004;16(1): 193---213.

10. Hale DR, Viner RM. Policy re- sponses to multiple risk behaviours in adolescents. J Public Health (Oxf). 2012;34(suppl 1):i11---i19.

11. Guilamo-Ramos V, Litardo HA, Jaccard J. Prevention programs for re- ducing adolescent problem behaviors: implications of the co-occurrence of problem behaviors in adolescence. J Adolesc Health. 2005;36(1):82---86.

12. Bridges S, Gill V, Omole T, Sutton R, Wright V. Smoking, Drinking and Drug Use Among Young People in England in 2010. London, UK: National Centre for Social Research and National Foundation for Educational Research; 2011.

13. Jackson C, Geddes R, Haw S, Frank J. Interventions to prevent substance use and risky sexual behaviour in young people: a systematic review. Addiction. 2012;107(4):733---747.

14. Institute of Medicine, Committee on the Science of Adolescence. The Science of Adolescent Risk-Taking: Workshop Sum- mary. Washington, DC: National Acade- mies Press; 2010.

15. Pudney S. The road to ruin? Se- quences of initiation to drug use and crime in Britain. Econ J. 2003;113(486): C182---C98.

16. Wagner FA, Anthony JC. Into the world of illegal drug use: exposure op- portunity and other mechanisms linking the use of alcohol, tobacco, marijuana, and cocaine. Am J Epidemiol. 2002;155 (10):918---925.

17. Domitrovich CE, Bradshaw CP, Greenberg MT, Embry D, Poduska JM, Ialongo NS. Integrated models of school- based prevention: logic and theory. Psy- chol Sch. 2010;47(1):71---88.

18. Parkes A, Wight D, Henderson M, Hart G. Explaining associations between adolescent substance use and condom use. J Adolesc Health. 2007;40(2):180. e1---e18.

19. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535.

20. Thomas H. Quality Assessment Tool for Quantitative Studies. Hamilton, Ontario, Canada: Effective Public Health

Practice Project; 2003. Available at: http://www.ephpp.ca/tools.html. Accessed August 15, 2012.

21. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hill- sdale, NJ: Lawrence Erlbaum; 1988.

22. Rosenthal JA. Qualitative descriptors of strength of association and effect size. J Soc Serv Res. 1996;21(4):37---59.

23. Campbell Collaboration. Practical meta-analysis effect size calculator. Avail- able at: http://www.campbellcollaboration. org/resources/effect_size_input.php. Accessed August 20, 2012.

24. Fang L, Schinke SP, Cole KA. Pre- venting substance use among early Asian-American adolescent girls: initial evaluation of a web-based, mother- daughter program. J Adolesc Health. 2010;47(5):529---532.

25. Taylor BJ, Graham JW, Cumsille P, Hansen WB. Modeling prevention program effects on growth in substance use: analysis of five years of data from the adolescent alcohol prevention trial. Prev Sci. 2000;1(4):183---197.

26. Connell AM, Dishion TJ, Yasui M, Kavanagh K. An adaptive approach to family intervention: linking engagement in family-centered intervention to reduc- tions in adolescent problem behavior. J Consult Clin Psychol. 2007;75(4):568--- 579.

27. Schinke SP, Schwinn TM, Fang L. Longitudinal outcomes of an alcohol abuse prevention program for urban ad- olescents. J Adolesc Health. 2010;46(5): 451---457.

28. McNeal RB, Hansen WB, Harrington NG, Giles SM. How All Stars works: an examination of program effects on medi- ating variables. Health Educ Behav. 2004;31(2):165---178.

29. Grossman JB, Tierney JP. Does mentoring work? An impact study of the Big Brothers Big Sisters program. Eval Rev. 1998;22(3):403---426.

30. Griffin JP Jr, Holliday RC, Frazier E, Braithwaite RL. The BRAVE (Building Resiliency and Vocational Excellence) Program: evaluation findings for a career- oriented substance abuse and violence preventive intervention. J Health Care Poor Underserved. 2009;20(3): 798---816.

31. Furr-Holden CDM, Ialongo NS, Anthony JC, Petras H, Kellam SG. De- velopmentally inspired drug prevention: middle school outcomes in a school-based randomized prevention trial. Drug Alcohol Depend. 2004;73(2):149---158.

32. Newton NC, Andrews G, Teesson M, Vogl LE. Delivering prevention for alco- hol and cannabis using the internet: a cluster randomised controlled trial. Prev Med. 2009;48(6):579---584.

33. Hawkins JD, Oesterle S, Brown EC, et al. Results of a type 2 translational research trial to prevent adolescent drug use and delinquency: a test of Communi- ties That Care. Arch Pediatr Adolesc Med. 2009;163(9):789---798.

34. Hawkins JD, Oesterle S, Brown EC, et al. Sustained decreases in risk exposure and youth problem behaviours after in- stallation of the Communities That Care prevention system in a randomized trial. Arch Pediatr Adolesc Med. 2012;166(2): 141---148.

35. Schinke SP, Fang L, Cole KC. Computer-delivered, parent-involvement intervention to prevent substance use among adolescent girls. Prev Med. 2009; 49(5):429---435.

36. Perry CL, Komro KA, Veblen- Mortenson S, et al. A randomized con- trolled trial of the middle and junior high school D.A.R.E. and D.A.R.E. Plus programs. Arch Pediatr Adolesc Med. 2003;157(2):178---184.

37. Gomez-Fraguela JA, Luengo MA, Romero E. Drug-abuse prevention in the school: four-year follow-up of a pro- gramme. Psychol Spain. 2003;7(1):29---38.

38. Stormshak EA, Connell AM, Véronneau MH, et al. An ecological ap- proach to promoting early adolescent mental health and social adaptation: family-centered intervention in public middle schools. Child Dev. 2011;82 (1):209---225.

39. O’Donnell L, Myint-U A, Duran R, Stueve A. Especially for daughters: parent education to address alcohol and sex- related risk taking among urban young adolescent girls. Health Promot Pract. 2010;11(3 suppl):70S---78S.

40. Pantin H, Prado G, Lopez B, et al. A randomized controlled trial of Familias Unidas for Hispanic adolescents with behavior problems. Psychosom Med. 2009;71(9):987---995.

41. Bauman KE, Ennett ST, Foshee VA, Pemberton M, King TS, Koch GG. Influ- ence of a family program on adolescent smoking and drinking prevalence. Prev Sci. 2002;3(1):35---42.

42. Stanton B, Cole M, Galbraith J, et al. Randomized trial of a parent intervention: parents can make a difference in long- term adolescent risk behaviors, percep- tions, and knowledge. Arch Pediatr Ado- lesc Med. 2004;158(10):947---955.

43. Telch MJ, Miller LM, Killen JD, Cooke S, MacCoby N. Social influences approach to smoking prevention: the ef- fects of videotape delivery with and without same-age peer leader participa- tion. Addict Behav. 1990;15(1):21---28.

44. Kulis S, Marsiglia F, Elek E, Dustman P, Wagstaff DA, Hecht ML. Mexican/ Mexican American adolescents and

SYSTEMATIC REVIEW

e40 | Systematic Review | Peer Reviewed | Hale et al. American Journal of Public Health | May 2014, Vol 104, No. 5

keepin’ it REAL: an evidence-based sub- stance use prevention program. Child Sch. 2005;27(3):133---145.

45. Botvin GJ, Baker E, Renick NL, Filazzola AD, Botvin EM. A cognitive- behavioral approach to substance abuse prevention. Addict Behav. 1984;9:137---147.

46. Botvin GJ, Baker E, Filazzola A, Botvin EM. A cognitive-behavioral ap- proach to substance abuse prevention: a one-year follow-up. Addict Behav. 1990;15(1):47---63.

47. Botvin GJ, Schinke SP, Epstein JA, Diaz T, Botvin EM. Effectiveness of cul- turally focused and generic skills training approaches to alcohol and drug abuse prevention among minority adolescents: two-year follow-up results. Psychol Addict Behav. 1995;9(3):183---194.

48. Griffin KW, Botvin GJ, Nichols TR. Effects of a school-based drug abuse pre- vention program for adolescents on HIV risk behaviors in young adulthood. Prev Sci. 2006;7(1):103---112.

49. DeGarmo DS, Eddy JM, Reid JB, Fetrow RA. Evaluating mediators of the impact of the Linking the Interests of Families and Teachers (LIFT) multimodal preventive intervention on substance use initiation and growth across adolescence. Prev Sci. 2009;10(3):208---220.

50. O’neil JM, Clark JK, Jones JA. Pro- moting mental health and preventing substance abuse and violence in elemen- tary students: a randomised control trial of the Michigan Model for Health. J Sch Health. 2011;81(6):320---330.

51. Stanton BF, Li X, Kahihuata J, et al. Increased protected sex and abstinence among Namibian youth following a HIV risk-reduction intervention: a random- ized, longitudinal study. AIDS. 1998; 12(18):2473---2480.

52. DeWitt DJ, Steep B, Silverman G, et al. Evaluating an in-school drug pre- vention program for at-risk youth. Alberta J Educ Res. 2000;46(2):117---133.

53. Hansen WB, Graham JW. Preventing alcohol, marijuana, and cigarette use among adolescents: peer pressure resistance train- ing versus establishing conservative norms. Prev Med. 1991;20(3):414---430.

54. Werch CE, Bian H, Moore MJ, et al. Brief multiple behavior health interven- tions for older adolescents. Am J Health Promot. 2008;23(2):92---96.

55. Beets MW, Flay BR, Vuchinich S, et al. Use of a social and character de- velopment program to prevent substance use, violent behaviors, and sexual activity among elementary-school students in Hawaii. Am J Public Health. 2009;99 (8):1438---1445.

56. Li KK, Washburn I, DuBois DL, et al. Effects of the Positive Action programme in problem behaviours in elementary

school students: a matched-pair rando- mised control trial in Chicago. Psychol Health. 2011;26(2):187---204.

57. Mason WA, Kisterman R, Hawkins JD, Haggerty KP, Spoth RL. Reducing adolescents’ growth in substance use and delinquency: randomized trial effects of a parent-training prevention intervention. Prev Sci. 2003;4(3):203---212.

58. Schwinn TM, Schinke SP. Preventing alcohol use among late adolescent urban youth: 6-year results from a computer- based intervention. J Stud Alcohol Drugs. 2010;71(4):535---538.

59. Schinke SP, Tepavac L, Cole KC. Preventing substance use among Native American youth: three- year results. Ad- dict Behav. 2000;25(3):387---397.

60. Schinke SP, Botvin GL, Trimble JE, Orlandi M, Gilchrist LD, Locklear VS. Preventing substance abuse among American-Indian adolescents: a bicultural competence skills approach. J Couns Psy- chol. 1988;35(1):87---90.

61. Ellickson PL, McCaffrey DF, Ghosh- Dastidar B, Longshore DL. New inroads in preventing adolescent drug use: results from a large-scale trial of Project ALERT in middle schools. Am J Public Health. 2003;93(11):1830---1836.

62. Ellickson PL, McCaffrey DF, Klein DJ. Long-term effects of drug prevention on risky sexual behavior among young adults. J Adolesc Health. 2009;45(2):111---117.

63. Orlando M, Ellickson PL, McCaffrey DF, Longshore DL. Mediation analysis of a school-based drug prevention program: effects of Project ALERT. Prev Sci. 2005; 6(1):35---46.

64. Hurry J, Lloyd C, McGurk H. Long- term effects of drugs education in primary schools. Addict Res. 2000;8(2):183---202.

65. Shek DTL, Yu L. Prevention of adolescent problem behavior: longitudi- nal impact of the project P.A.T.H.S. in Hong Kong. ScientificWorldJournal. 2011;11:546---567.

66. Graham JW, Johnson CA, Hansen WB, Flay BR, Gee M. Drug use prevention pro- grams, gender, and ethnicity: evaluation of three seventh-grade Project SMART cohorts. Prev Med. 1990;19(3):305---313.

67. Werch CE, Moore MM, DiClemente CC, Owen DM, Carlson JM, Jobli E. Single vs. multiple drug prevention: is more always better?: a pilot study. Subst Use Misuse. 2005;40(8):1085---1101.

68. Dent CW, Sussman S, Stacy AW. Project Towards No Drug Abuse: gener- alizability to a general high school sample. Prev Med. 2001;32(6):514---520.

69. Brown EC, Catalan RF, Fleming CB, Haggerty KP, Abbott RD. Adolescent substance use outcomes in the Raising Healthy Children Project: a two-part

latent growth curve analysis. J Consult Clin Psychol. 2005;73(4):699---710.

70. Schwinn TM, Schinke SP, Di Noia J. Preventing drug abuse among adolescent girls: outcome data from an internet-based intervention. Prev Sci. 2010;11(1):24---32.

71. Homel PJ, Daniels P, Reid TR, Lawson JS. Results of an experimental school-based health development pro- gramme in Australia. Int J Health Educ. 1981;24(4):263---270.

72. Gilchrist LD, Schinke SP, Trimble JE, Cvetkovich G. Skills enhancement to pre- vent substance abuse among American Indian adolescents. Int J Addict. 1987; 22(9):869---879.

73. Spoth RL, Redmond C, Shin C. Randomized trial of brief family inter- ventions for general populations: adoles- cent substance use outcomes 4 years following baseline. J Consult Clin Psychol. 2001;69(4):627---642.

74. Spoth R, Redmond C, Shin C, Azecedo K. Brief family intervention effects on adolescent substance initiation: school-level growth curve analyses 6 years following baseline. J Consult Clin Psychol. 2004;72(3):535---542.

75. Spoth RL, Redmond C, Trudeau L, Shin C. Longitudinal substance initiation outcomes for a universal preventive in- tervention combining family and school programs. Psychol Addict Behav. 2002;16 (2):129---134.

76. Flay BR, Graumlich S, Segawa E, Burns JL, Holliday MY. Effects of 2 pre- vention programs on high-risk behaviors among African American youth: a ran- domized trial. Arch Pediatr Adolesc Med. 2004;158(4):377---384.

77. Faggiano F, Galanti MR, Bohrn K, et al. The effectiveness of a school-based substance abuse prevention program: EU-Dap cluster randomised controlled trial. Prev Med. 2008;47(5):537---543.

78. Faggiano F, Vigna-Taglianti F, Burkhart G, et al. The effectiveness of a school-based substance abuse prevention program: 18-month follow-up of the EU-Dap cluster randomized controlled trial. Drug Alcohol Depend. 2010;108(1---2):56---64.

79. Terzian MA, Andrews KM, Moore KA. Preventing multiple risky behaviors among adolescents: seven strategies. 2011. Child Trends Research-to-Results Brief. Available at: http://www.childtrends. org/wp-content/uploads/2011/09/Child_ Trends-2011_10_01_RB_RiskyBehaviors. pdf. Accessed January 22, 2014.

80. Gottfredson DC, Wilson DB. Char- acteristics of effective school-based sub- stance abuse prevention. Prev Sci. 2003; 4(1):27---38.

81. Bond L, Patton G, Glover S, et al. The Gatehouse Project: can a multilevel school intervention affect emotional wellbeing

and health risk behaviours? J Epidemiol Commun Health. 2004;58(12):997--- 1003.

82. Durlak JA, Weissberg RP, Dymnicki AB, Taylor RD, Schellinger KB. The im- pact of enhancing students’ social and emotional learning: a meta-analysis of school-based universal interventions. Child Dev. 2011;82(1):405---432.

83. Kirby DB, Laris BA, Rolleri LA. Sex and HIV education programs: their im- pact on sexual behaviors of young people throughout the world. J Adolesc Health. 2007;40(3):206---217.

84. Cuijpers P. Effective ingredients of school-based drug prevention programs: a systematic review. Addict Behav. 2002;27(6):1009---1023.

85. Bry BH, Krinsley KE. Booster ses- sions and long-term effects of behavioral family therapy on adolescent substance use and school performance. J Behav Ther Exp Psy. 1992;23(3):183---189.

86. Shepherd J, Kavanagh J, Picot J, et al. The effectiveness and cost-effectiveness of behavioural interventions for the pre- vention of sexually transmitted infections in young people aged 13---19: a system- atic review and economic evaluation. Health Technol Assess. 2010;14(7): 1---206, iii---iv.

87. Blank L, Guillaume L. Systematic Review of the Effectiveness of Universal Interventions Which Aim to Promote Emo- tional and Social Wellbeing in Secondary Schools. London, UK: National Institute for Health and Clinical Excellence; 2009.

88. Fletcher A, Bonell C, Hargreaves J. School effects on young people’s drug use: a systematic review of intervention and observational studies. J Adolesc Health. 2008;42(3):209---220.

89. Gendron BP, Williams KR, Guerra NG. An analysis of bullying among stu- dents within schools: estimating the ef- fects of individual normative beliefs, self- esteem, and school climate. J Sch Violence. 2011;10(2):150---164.

90. Moffitt TE. Adolescence-limited and life-course-persistant antisocial behavior: a developmental taxonomy. Psychol Rev. 1993;100(4):674---701.

91. Gardner M, Steinberg L. Peer in- fluence on risk taking, risk preference, and risky decision making in adolescence and adulthood: an experimental study. Dev Psychol. 2005;41(4):625---635.

92. Peters LWH, Kok G, Ten Dam GTM, Buijs GJ, Paulussen TGWM. Effective elements of school health promotion across behavioral domains: a systematic review of reviews. BMC Public Health. 2009;9:182.

93. Williams RJ, Nowatzki N. Validity of adolescent self-report of substance use. Subst Use Misuse. 2005;40(3):299---311.

SYSTEMATIC REVIEW

May 2014, Vol 104, No. 5 | American Journal of Public Health Hale et al. | Peer Reviewed | Systematic Review | e41

Copyright of American Journal of Public Health is the property of American Public Health Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.