A systematic review of school-based alcohol and other drug prevention programs facilitated by...

COMPREHENSIVE REVIEW

A systematic review of school-based alcohol and other drugprevention programs facilitated by computers or the Internet

KATRINA E. CHAMPION, NICOLA C. NEWTON, EMMA L. BARRETT & MAREE TEESSON

National Drug and Alcohol Research Centre, University of New SouthWales, Sydney, Australia

AbstractIssues. The use of alcohol and drugs amongst young people is a serious concern and the need for effective prevention is clear.This paper identifies and describes current school-based alcohol and other drug prevention programs facilitated by computersor the Internet. Approach. The Cochrane Library, PsycINFO and PubMed databases were searched in March 2012.Additional materials were obtained from reference lists of papers. Studies were included if they described an Internet- orcomputer-based prevention program for alcohol or other drugs delivered in schools. Key Findings.Twelve trials of 10 programswere identified. Seven trials evaluated Internet-based programs and five delivered an intervention via CD-ROM. Theinterventions targeted alcohol, cannabis and tobacco. Data to calculate effect size and odds ratios were unavailable for threeprograms. Of the seven programs with available data, six achieved reductions in alcohol, cannabis or tobacco use at postintervention and/or follow up.Two interventions were associated with decreased intentions to use tobacco, and two significantlyincreased alcohol and drug-related knowledge. Conclusion. This is the first study to review the efficacy of school-based drugand alcohol prevention programs delivered online or via computers. Findings indicate that existing computer- and Internet-based prevention programs in schools have the potential to reduce alcohol and other drug use as well as intentions to usesubstances in the future. These findings, together with the implementation advantages and high fidelity associated with newtechnology, suggest that programs facilitated by computers and the Internet offer a promising delivery method for school-basedprevention. [Champion KE, Newton NC, Barrett EL, Teesson M. A systematic review of school-based alcohol andother drug prevention programs facilitated by computers or the Internet. Drug Alcohol Rev 2013;32:115–123]

Key words: prevention, school, Internet, review, alcohol.

Introduction

The use of alcohol and other drugs by young people isa serious public health concern and the burden ofdisease, social costs and harms associated with this useare significant [1–5]. The most recent statistics in Aus-tralia show that approximately one-quarter of Austral-ian teenagers (between the ages of 14–19 years) havetried an illicit drug, two-thirds (65%) have consumed afull serve of alcohol in the past year and almost one-fifth(20%) have consumed alcohol at levels that put them atrisk of injury at least once in the past month (defined asmore than four drinks on a single occasion) [1]. Thesefigures are concerning, given that early initiation to

drug use (i.e. before the age of 18) is a risk factor fordeveloping substance use disorders and comorbidmental health problems in adulthood [6,7]. In light ofthis research the need for effective prevention is clear.

In recent years we have seen a substantial increase inthe development of school-based prevention programsfor alcohol and other drug use. Despite this, the major-ity of these programs have shown limited effects,particularly in terms of impacting on behaviour andreducing or preventing substance use [8,9].This is mostlikely due to the many obstacles that impact on suc-cessful program implementation [10,11].These includelimited resources in terms of teachers, money and timeallocated to deliver drug prevention, as well as the fact

Katrina E. Champion BA-Psych (Hons), Research Assistant, Nicola C. Newton BPsych (Hons), PhD, Senior Research Fellow, Emma L. BarrettBPsych (Hons), MPsych, PhD, Research Associate, MareeTeesson BA (Psych), PhD, Professor. Correspondence to Ms Katrina Champion, 22-32King Street, Randwick, NSW 2052, Australia. Tel: +61 2 9385 0175; Fax: +61 2 9385 0222; E-mail: [email protected]

Received 13 June 2012; accepted for publication 3 September 2012.

bs_bs_banner

R E V I E W

Drug and Alcohol Review (March 2013), 32, 115–123DOI: 10.1111/j.1465-3362.2012.00517.x

© 2012 Australasian Professional Society on Alcohol and other Drugs

that programs are often adapted to the school/classenvironment thereby losing their effective ingredients.A study by Ennett and colleagues [12] found that only14% of programs delivered in schools have the correctcontent and modes of delivery identified in the litera-ture as being effective.

Computer- and Internet-based prevention programshave the potential to overcome these obstacles and offermany advantages over traditional drug preventionmethods. First, since professionals are not required todeliver the programs, they are less restrictive in theiravailability and offer increased feasibility of use in set-tings where professional and teaching time is limited.Even when combined with some direct contact or facili-tation from professionals, the burden on professionals isalleviated as it reduces the amount of time they need todedicate to individuals [13]. Second, once a program isdeveloped there is a reduction in implementation costs,ease of updating materials and increased potential toengage large numbers of individuals and overcome geo-graphical and socio-economic constraints. Third, theycan ensure a high degree of implementation fidelity asconsistent and complete delivery of materials can beguaranteed. Finally, specific to drug education, com-puters and the Internet have the potential to increaseself-disclosure and reduce stigmatisation about druguse, by enhancing perceptions of privacy and anonym-ity [14]. In a school environment however, althoughparticipants are likely to feel less vulnerable disclosinginformation online than in a face-to-face setting [15],some students may still have privacy concerns aboutwhether their personal information will be accessed byteachers or parents.

In recent years, a large number of reviews ofcomputer-based treatment programs have been con-ducted, with results indicating they can be effective interms of outcome and cost by addressing and reducingalcohol and tobacco use [16–33]. However, none ofthese existing reviews have specifically focussed on ado-lescents or on programs delivered in school-based set-tings. Moreover, these reviews have been limited totreatment and therapy interventions and no reviewshave reported specifically on Internet-based programsdesigned to prevent alcohol and drug use. This studywill address these gaps in the literature by reviewing theevidence on, and establishing whether school-basedprevention programs facilitated by the computer or theInternet have the potential to reduce and prevent theuse of alcohol and other drugs in adolescents.

Methods

Data sources

The Cochrane Library, PsycINFO and PubMed data-bases were searched in March 2012, using the following

keywords: ‘online OR web-based OR internet OR com-puter’, ‘drug OR alcohol OR cannabis OR ecstasy ORmarijuana OR substance OR amphetamine OR psychostimulant’, ‘school OR school-based’ and ‘interventionOR prevention OR program OR education’. The titlesand abstracts of the 2574 articles identified were inde-pendently reviewed and full copies of potentially rel-evant papers were obtained to determine if they met theinclusion criteria. Reference lists of individual paperswere manually searched for further publications. Pro-grams were included if they were an Internet- orcomputer-based prevention program for alcohol orother drugs, and if they were delivered at school. Pro-grams targeting school aged students that were imple-mented in the home or community were excluded, aswere those delivered to university or college students.Figure 1 displays a flow chart of the search strategy andstudy selection process.

Study quality

Study quality was evaluated using a validated tool forassessing the quality of randomised controlled trials[34].This instrument has been designed to measure thequality of trials across a broad range of subject areas[35], and has been used previously in reviews of school-based interventions for anxiety and depression [36] anddrug and alcohol use [37]. Studies were rated againstthree key criteria: randomisation, double-blinding, andwithdrawals and drop-outs, and given an overall scoreranging from 0 to 5. As reported previously [38],school-based interventions rarely receive scores above 3as double-blind conditions and full randomisation areoften not possible.

Outcome measures

The primary outcome evaluated was alcohol and drugconsumption, at immediate post-intervention and laterfollow-up occasions. Differences between control andintervention groups were also reported for a range ofsecondary outcome measures including drug-relatedknowledge, attitudes, harms and intentions to use.

Analysis

Effect sizes are reported for continuous outcomes andodds ratios are reported for dichotomous outcomes.Effect size was estimated using Cohen’s d [39], which iscalculated by subtracting the mean intervention scorefrom the mean control score, and dividing this by thepre-intervention pooled standard deviation. Whereavailable, effect sizes and odds ratios were extractedfrom papers, or were provided by the authors of thepaper on request. Due to the small number of studies

116 K. E. Champion et al.


and the heterogeneity of study quality, outcome meas-ures and follow-up occasions, it was not possible tocombine the results into a meta-analysis. For thisreason we chose to report the results in a systematicreview.

Results

Overall, 12 trials of 10 programs were identified. Allprograms were universal, that is, they were deliveredto all students in a year group regardless of their levelof risk for alcohol or drug use. The programs tendedto be based on principles of social influence theory.The ‘social influence approach’ delineates that youngpeople initiate drug use as a result of external pressurefrom peers, family and the media, and that youngpeople are not adequately equipped to deal with suchpressure [40]. Therefore, the goal of this approach isto teach adolescents to resist drugs by providing themwith information, resistance skills and normative edu-cation [41,42]. Two programs were based on princi-ples of social cognitive theory, in which students aretaught skills to enhance self-efficacy and developnegative expectancies about alcohol and drug use

[43], and two interventions drew on the Transtheo-retical Model of Change, which posits that healthbehaviour change occurs through the progression ofsix stages of change [44].

Of the 12 trials, seven evaluated an Internet-basedprogram and five assessed interventions delivered viaCD-ROM. The majority of trials were conducted inAustralia and the USA. All trials were mixed genderand most targeted students in their first two years ofhigh school (13–15 years of age). Eight trials collecteddata post intervention and the follow-up period in thestudies ranged from 6 to 34 months. Most controlgroups received health education as usual, with theexception of the Drugs 4 Real trial, which included avideo component, the Smoking Zine trial whichincluded of a web evaluation control task, and controlsin the Head On trial, which received the Life SkillsTraining program. Overall, study quality was weak,with no studies scoring above 3 (Table 1).This is com-parable to other prevention reviews [37,38] and canlikely be explained by the difficulties achieving full ran-domisation and double-blind conditions in school-based interventions. Of the identified programs, onlythe CLIMATE Schools: Alcohol Module and the Consider

Records identified through database searching (n = 3085)

Scr

een

ing

Incl

uded

Eli

gib

ility

Iden

tifi

cati

on Additional records identified through other sources

(n = 5)

Records after duplicates removed (n = 2574)

Records screened (n = 2574)

Records excluded (n = 2542)

Full-text articles assessed for eligibility

(n = 32)

Full-text articles excluded (n = 20)

• Not school-based (10),

• Not prevention program (5),

• Not online or computer-based (5)

Studies included in qualitative synthesis

(n = 12)

Figure 1. Flow chart of search strategy and study selection.

Systematic review of Internet-based prevention 117


Tab

le1.

Pri

mar

yan

dse

cond

ary

outc

ome

data

for

iden

tified

tria

ls

Pro

gram

Tri

alS

ubst

ance

Sam

ple

Inte

rven

tion

Ori

enta

tion

Con

trol

Sub

stan

ceus

epo

st-i

nter

vent

ion

ES

/OR

Sub

stan

ceus

efo

llow

-up

ES

/OR

Sec

onda

ryou

tcom

esE

S/O

RQ

ualit

yra

ting

Con

side

rT

his

Bul

ler

etal

.20

08[4

5]T

obac

coA

ustr

alia

,10

–16

yrs,

n=

2077

Onl

ine,

6le

sson

sS

ocia

lco

gnit

ive

prin

cipl

esS

tand

ard

heal

thed

ucat

ion

30-d

aysm

okin

gpr

eval

ence

(who

leci

gare

tte)

,E

S0.

05*

(low

erin

INT

than

CO

)

—F

utur

esm

okin

gin

tent

ions

PI,

OR

0.01

2

Tob

acco

US

A10

–14

yrs

n=

1234

Onl

ine,

6le

sson

sS

ocia

lco

gnit

ive

prin

cipl

esS

tand

ard

heal

thed

ucat

ion

30-d

aysm

okin

gpr

eval

ence

(who

leci

gare

tte)

,E

S0.

23

—F

utur

esm

okin

gin

tent

ions

PI,

OR

0.13

*(I

NT

low

erth

anC

O)

2

Sm

okin

gZ

ine

Nor

man

etal

.20

08[4

6]T

obac

coC

anad

a14

–16

yrs

n=

1402

Onl

ine,

5st

ages

Soc

ial

cogn

itiv

epr

inci

ples

,L

AI,

TT

M

Web

eval

uati

onta

sk—

Cig

aret

teus

e,O

R1.

27;

Cig

aret

teus

eam

ong

non-

smok

ers,

OR

0.79

*(l

ower

inIN

Tth

anC

O)

Res

ista

nce

(who

lesa

mpl

e),

OR

1.03

and

resi

stan

ceam

ong

base

line

smok

ers,

OR

1.22

*(h

ighe

rin

INT

than

CO

);B

ehav

iour

alin

tent

ions

tosm

oke,

OR

1.04

and

beha

viou

ral

inte

ntio

nsam

ong

base

line

smok

ers,

OR

0.82

*(l

ower

inIN

Tth

anC

O)

3

AS

PIR

EP

rokh

orov

etal

.20

08[4

7]T

obac

coU

SA

15–1

6yr

sn

=15

74

CD

-RO

M,

5le

sson

s+

boos

ter

TT

MS

tand

ard

heal

thed

ucat

ion

—S

mok

ing

init

iati

onat

18-m

onth

F/U

,O

R2.

87*

(low

erin

INT

than

CO

);C

igar

ette

smok

ing

beha

viou

rat

18-m

onth

F/U

,E

S0.

12*

(low

erin

INT

than

CO

)

Dec

isio

nal

bala

nce,

ES

0.25

*(I

NT

high

erth

anC

O);

Tem

ptat

ion

tosm

oke,

ES

0.20

*(I

NT

low

erth

anC

O);

Sel

f-ef

ficac

y,E

S0.

02;

Res

ista

nce

skill

s,E

S0.

26,

all

at18

-mon

thF

/U

2

CL

IMA

TE

Sch

ools

:Alc

ohol

Vog

let

al.

2009

[48]

Alc

ohol

Aus

tral

ia13

yrs

n=

1466

CD

-RO

M,

6le

sson

sH

arm

-min

imis

atio

n;so

cial

lear

ning

prin

cipl

es

Sta

ndar

dhe

alth

educ

atio

nA

vera

geal

coho

lco

nsum

ptio

n,E

S0.

25;

Bin

gedr

inki

ng,

ES

0.11

Ave

rage

alco

hol

cons

umpt

ion

at6-

mon

thF

/U,

ES

0.24

*an

d12

-mon

thF

/U,

ES

0.23

*(l

ower

inIN

Tth

anC

Ofo

rfe

mal

eson

ly);

Bin

gedr

inki

ngat

6-m

onth

F/U

,E

S0.

20*

and

12-m

onth

F/U

,E

S0.

20*

(low

erin

INT

than

CO

for

fem

ales

only

).

Alc

ohol

know

ledg

eat

6-m

onth

F/U

,E

S0.

73an

d12

-mon

thF

/U,

ES

0.52

;Alc

ohol

harm

sat

6-m

onth

F/U

,E

S0.

08an

d12

-mon

thF

/U,

ES

0.20

*(I

NT

low

erth

anC

Ofo

rfe

mal

eson

ly);

Alc

ohol

expe

ctan

cies

at6-

mon

thF

/U,

ES

0.27

and

12-m

onth

F/U

,fe

mal

es,

ES

0.41

*,an

dm

ales

,E

S0.

30*,

(IN

Tlo

wer

than

CO

).

3

New

ton

etal

.20

09[4

9]A

lcoh

olA

ustr

alia

13yr

sn

=76

4

Onl

ine,

6le

sson

sH

arm

-min

imis

atio

n;so

cial

lear

ning

prin

cipl

es

Sta

ndar

dhe

alth

educ

atio

nA

vera

geal

coho

lco

nsum

ptio

n,E

S0.

09*

(low

erin

INT

than

CO

);B

inge

drin

king

,E

S0.

06

Ave

rage

alco

hol

cons

umpt

ion

at6-

mon

thF

/U,

0.09

;B

inge

drin

king

at6-

mon

thF

/U,

ES

0.05

Alc

ohol

know

ledg

e,E

S0.

69*

(IN

Thi

gher

than

CO

);A

lcoh

olha

rms,

ES

0.08

;A

lcoh

olex

pect

anci

es,

ES

0.20

,al

lat

6-m

onth

F/U

2



CL

IMA

TE

Sch

ools

:Alc

ohol

&C

anna

bis

New

ton

etal

.20

09[5

0]N

ewto

net

al.

2010

[51]

Alc

ohol

,ca

nnab

isA

ustr

alia

13yr

sn

=76

4

Onl

ine,

12le

sson

s(i

ncl.

2bo

oste

rle

sson

s)

Har

m-m

inim

isat

ion;

soci

alle

arni

ngpr

inci

ples

Sta

ndar

dhe

alth

educ

atio

nA

vera

geal

coho

lco

nsum

ptio

n,E

S0.

18;

Bin

gedr

inki

ng,

ES

0.90

;F

requ

ency

ofca

nnab

isus

e,E

S0.

18

Ave

rage

alco

hol

cons

umpt

ion

at6-

mon

thF

/U,

ES

0.16

*an

d12

-mon

thF

/U,

ES

0.38

*(l

ower

inIN

Tth

anC

O);

Bin

gedr

inki

ngat

6-m

onth

F/U

,E

S0.

05an

d12

-mon

thF

/U,

ES

0.17

*(l

ower

inIN

Tth

anC

O);

Fre

quen

cyof

cann

abis

use

at6-

mon

thF

/U,

ES

0.19

*lo

wer

inIN

Tth

anC

O)

and

12-m

onth

F/U

,E

S0.

31

Alc

ohol

know

ledg

eat

6-m

onth

F/U

,E

S0.

75*

and

12-m

onth

F/U

,E

S0.

76*

(IN

Thi

gher

than

CO

);C

anna

bis

know

ledg

eat

6-m

onth

F/U

,E

S0.

56*

and

12-m

onth

F/U

,E

S0.

61*

(IN

Thi

gher

than

CO

);A

lcoh

olha

rms

at6-

mon

thF

/U,

ES

0.04

and

12-m

onth

F/U

,E

S0.

26;

Can

nabi

sha

rms

at6-

mon

thF

/U,

ES

0.04

and

12-m

onth

F/U

,E

S0.

12;A

lcoh

olex

pect

anci

esat

6-m

onth

F/U

,E

S0.

16an

d12

-mon

thF

/U,

ES

0.3;

Can

nabi

sat

titu

des

at6-

mon

thF

/U,

ES

0.03

and

12-m

onth

F/U

,E

S0.

21.

3

Com

bine

dPA

SIn

terv

enti

onK

onin

get

al.

2009

[52]

Kon

ing

etal

.20

11[4

3]

Alc

ohol

Net

herl

ands

12–1

3yr

sn

=33

68

Onl

ine,

4le

sson

san

d/or

pare

nted

ucat

ion

+bo

oste

r

Soc

ial

lear

ning

prin

cipl

esS

tand

ard

heal

thed

ucat

ion

—O

nset

ofhe

avy

wee

kly

alco

hol

use

at10

-mon

thF

/U,

OR

0.36

*,22

-mon

thF

/U,

OR

0.80

and

34-m

onth

F/U

,O

R0.

69*

(les

sus

ein

INT

than

CO

);O

nset

ofw

eekl

yal

coho

lus

eat

10-m

onth

F/U

,O

R,

0.67

*,22

-mon

thF

/UO

R0.

71*

and

34-m

onth

F/U

,O

R0.

69*

(les

sus

ein

INT

than

CO

);F

requ

ency

ofm

onth

lydr

inki

ngat

10an

d22

-mon

thF

/Ua

—3

TT

MIn

terv

enti

onA

veya

rdet

al.

2001

[53]

Tob

acco

UK

13–1

4yr

sn

=83

52

CD

-RO

M,

3le

sson

sT

TM

AO

—W

eekl

ysm

okin

gat

24-m

onth

F/U

,O

R1.

06;

Pos

itiv

ech

ange

ofst

age

at24

-mon

thF

/U,

OR

1.25

—2

Hea

dO

nM

arsc

het

al.

2007

[54]

Tob

acco

,al

coho

l,ca

nnab

is

US

A12

yrs

n=

272

CD

-RO

M,

15le

sson

sS

ocia

lle

arni

ngpr

inci

ples

Lif

eS

kills

Tra

inin

gpr

ogra

m

Fre

quen

cyof

smok

ing*

a

(IN

Thi

gher

than

CO

);F

requ

ency

ofdr

inki

nga ;

Fre

quen

cyof

mar

ijuan

aus

ea

—D

rug-

rela

ted

know

ledg

e*a

(IN

Thi

gher

than

CO

);In

tent

ions

tous

eal

coho

l,ci

gare

ttes

and

mar

ijuan

aa ;A

ttit

udes

tow

ards

drug

usea ;

Lik

elih

ood

ofre

fusa

la

0

Ref

use

toU

seD

unca

net

al.

2000

[55]

Can

nabi

sU

SA

,15

yrs,

n=

65

CD

-RO

M,

1le

sson

Soc

ial

lear

ning

prin

cipl

esS

tand

ard

Hea

lth

Edu

cati

on

——

Effi

cacy

tore

fuse

mar

ijuan

a*a

(IN

Thi

gher

than

CO

);In

tent

ion

tore

fuse

*a(I

NT

high

erth

anC

O)

3

Dru

gs4

Rea

lL

ord

&D

’Am

ante

[56]

Alc

ohol

,C

anna

bis,

Tob

acco

US

A,

12–1

4yr

s,n

=29

5

Onl

ine,

6vi

sits

Soc

ial

lear

ning

prin

cipl

esA

O;V

ideo

cont

rol

Dru

g-re

late

dkn

owle

dge*

a

(IN

Thi

gher

than

CO

);In

tent

ions

tous

eal

coho

l,ci

gare

ttes

and

mar

ijuan

aa ;A

ttit

udes

tow

ards

drug

usea ;

Lik

elih

ood

ofre

fusa

la

——

1

*Sig

nific

ant

diff

eren

ceat

P<

0.05

betw

een

inte

rven

tion

and

cont

rol

grou

ps.

a Aut

hors

wer

eco

ntac

ted,

but

effe

ctsi

zew

asun

able

tobe

calc

ulat

ed.

For

each

tria

l,E

Ss

and

OR

sar

ere

port

edat

post

-int

erve

ntio

nan

d/or

each

follo

w-u

poc

casi

on.

For

the

Nor

man

etal

.tri

al[4

6],t

heO

Rs

repo

rted

byth

eau

thor

sw

ere

aver

aged

acro

sspo

st-i

nter

vent

ion,

3-m

onth

and

6-m

onth

scor

es.I

nth

eK

onin

get

al.s

tudy

[43,

52],

only

the

com

bine

dpa

rent

/stu

dent

inte

rven

tion

was

sign

ifica

ntly

diff

eren

tfr

omth

eco

ntro

lgr

oup.

For

the

TT

Min

terv

enti

on[5

3],p

osit

ive

chan

geof

stag

ew

asde

fined

asa

mov

emen

tto

ast

age

whe

reac

quis

itio

nof

smok

ing

was

less

likel

y,or

cess

atio

nm

ore

likel

y.A

O,a

sses

smen

ton

ly;C

O,c

ontr

olgr

oup;

ES

,eff

ect

size

;F/U

,fol

low

up;I

NT

,in

terv

enti

ongr

oup;

LA

I,L

ikel

ihoo

dof

Act

ion

Inde

x;O

R,

odds

rati

o;P

I,po

stin

terv

enti

on;T

TM

,Tra

nsth

eore

tica

lM

odel

ofC

hang

e.



This program had been evaluated more than once.Effect size or odds ratios were unable to be calculatedfor three trials [54–56], and therefore are not includedin the results below (but are described in Table 1).

Primary outcomes

Table 1 presents outcome data relating to alcohol ordrug use for each trial. Five trials targeted tobacco andthree were associated with some reduction in smoking.In one trial [45], there was only a small effect at postintervention and in another [46] the intervention wasonly effective at reducing cigarette use among non-smokers at baseline. A third trial [47] was associatedwith a medium effect at the 18-month follow up. Allfour trials that measured alcohol consumption wereassociated with some reduction in alcohol use at postintervention and/or follow up. Effect size (ES) wassmall at post intervention (ES 0.09) and similarlymodest at follow up (ES 0.16–0.38 and odds ratio0.36–0.71). Two trials [48,51] were associated withpositive outcomes relating to the frequency of bingedrinking. Of the seven programs, only one targetedcannabis. This program was associated with a signifi-cant reduction in the frequency of cannabis use at6-month follow up with a small effect size (0.19).

Secondary outcomes

Table 1 displays data for the secondary outcomes meas-ured in each trial. Three trials measured future inten-tions to use tobacco. Of these, one was associated witha small reduction in intentions to smoke post interven-tion, and another found a larger effect, but only amongsmokers at baseline. All three trials that assessed drug-related knowledge demonstrated a significant increasein knowledge in the intervention groups compared withcontrols. Effect size for positive results ranged frommodest to large (0.69–1.33). Of the three trials thatassessed attitudes towards cannabis and alcohol, onefound a reduction in positive expectancies and atti-tudes, with the strongest effects occurring at 12-monthfollow up (ES 0.4 females, ES 0.3 males). One trial wasassociated with a reduction in alcohol-related harms,however only for females and only at 12-month followup [48], and another was associated with an increase inresistance skills, but only among baseline smokers [46].Finally, one study found a small, yet significant increasein decisional balance relating to tobacco use, as well asa reduction in temptations to smoke.

Discussion

Overall, we identified 10 computer- or Internet-basedprograms that have been trialled for the prevention of

alcohol and drug use in schools, and obtained effect sizeand/or odds ratios for seven of these programs. Six ofthe seven programs achieved a reduction in alcohol ordrug use at post intervention and/or follow up, two wereassociated with decreased intentions to smoke in thefuture and two programs significantly increasedalcohol- and drug-related knowledge.The results of thepresent review indicate that existing computer- andInternet-based programs in schools are a potentiallyefficacious method of delivering drug and alcohol pre-vention to adolescents.

Effect size and odds ratios for drug and alcohol usewere small. Of the 6 trials that assessed drug andalcohol consumption at follow up, 5 showed lastingeffects, ranging from 6 months to 34 months. Althougheffects were modest, this is comparable to effect sizereported in a recent review of Internet-based interven-tions for the treatment of substance use in young adults[20].The effect sizes were also similar to those reportedfor Internet-based interventions for anxiety and depres-sion in adolescents, with the exception of those thatinvolved an additional motivational interviewing orinformational component, which showed greater effects[57]. Results from the present review also comparefavourably with traditional, non-computerised pro-grams such as those reported in a recent review ofAustralian school-based prevention programs [37].This suggests that computer- and Internet-based inter-ventions can be as effective, if not more effective, asschool-based programs delivered without computers.

Overall, effect size and odds ratios for secondaryoutcomes were similarly modest. The greatest effectswere achieved in relation to drug- and alcohol-relatedknowledge, with effectiveness persisting at 6- and12-month follow ups for three trials. Previous system-atic reviews of school-based programs and Internet-based interventions have tended to exclude secondaryoutcomes such as knowledge, intentions to use andattitudes.Therefore, a clear strength of this review is theinclusion of a wide range of secondary outcomes, ena-bling a more comprehensive review, and providing amore complete picture of the impact of school-basedonline prevention programs.

This is also the first review to focus specifically oncomputer- and Internet-based programs for the preven-tion of alcohol and drugs in schools. Other trials haveevaluated Internet-based programs among universityand college students, and others have assessed the effi-cacy of implementing computer- and Internet-basedprograms in the home or community, and among olderpopulations. However, given the link between the earlyonset of drug use and later substance use disorders inadulthood [7], it is important to implement and evaluateonline programs that are delivered to adolescents whilethey are still at high school, before they initiate drug use.



Although the number of trials identified in thisreview is small, the results have implications for thedelivery of alcohol and drug prevention in schools.Theresults of this review support the use of the Internet asa potentially efficacious means of overcoming theobstacles associated with the implementation of tradi-tional prevention programs. Specifically, Internet- andcomputer-based programs offer increased accessibilityand feasibility of use and high implementation fidelity.When considering these advantages in conjunction withthe present results, Internet-based program deliveryappears to be a promising framework for the provisionof school-based education and prevention in the future.

As well as establishing whether Internet-based pro-grams are efficacious in preventing alcohol and druguse, it is critical to gauge why these programs might behaving an effect. One factor that may be associated withprogram success is the number of sessions included inthe intervention. All programs in this review that pro-duced significant results were comprised of between 4and 12 lessons. In comparison, the TranstheoreticalModel of Change Intervention [53], which only con-sisted of three lessons, failed to produce effects. Anadditional factor that has been cited previously as con-tributing to program success, is the inclusion of boostersessions [41]. In the present review, all three programsthat included booster lessons [43,47,50–52] showedsignificant effects. Program orientation may also havehad an impact on the efficacy of trials in terms ofalcohol and drug use. Five of the six programs thatfound significant reductions in substance use wereorientated around social learning or social cognitiveprinciples. This suggests that some of the effectiveingredients in Internet-based prevention programs arenormative education, resistance skills training andreducing positive expectancies [41,42]. However, it isimportant to note there may be reasons why youngpeople use alcohol and drugs, other than peer pressureand poor resistance skills. For example, it is possiblethat teenagers may have realistic positive expectanciesabout alcohol or drug use, and may actively desire toalter their conscious state [58]. Therefore, future pre-vention programs that address these potential motiva-tions, in addition to social influence factors, mayproduce larger intervention effects for adolescents. Afinal factor possibly associated with program efficacy inthe present review is the inclusion of a parenting com-ponent. Although only one program included a parentintervention [43,52], program effects for this trial per-sisted at 34-month follow up. In recent years, researchhas suggested that adding a parental component touniversal prevention programs can strengthen programeffects [53,59]. Therefore, future trials may benefitfrom adding a parental component to existing Internet-based programs for drug and alcohol prevention.

A potential limitation of the present study is that thetrials included in the review relied solely on studentself-report. However, studies have found the self-reportof behaviours such as substance use among adolescentsis highly consistent with behavioural observations, aslong as confidentiality and anonymity is assured [60]. Afurther limitation is the small number of studiesincluded in the review, differences in outcome measuresassessed and the unavailability of data to calculate effectsizes for three of the identified programs. Additionally,only two of the 10 programs had been evaluated morethan once. This highlights a clear need for cross-validation and replication studies of these existing pro-grams, to provide further support for the effectivenessof Internet-based prevention for alcohol and otherdrugs, delivered in schools. Finally, of the 12 trialsincluded in this review, only two [48,55] analysedresults separately for males and females, and only oneof these [48] had available data to calculate effect size.The importance of distinguishing between males andfemales has been noted in the literature [58], and issupported by the differential effect these interventionshad by gender.Therefore, where sample size and powerare adequate to do so, future evaluations shouldattempt to consider results for males and females sepa-rately, especially in countries where the recommendeddrinking guidelines differ for males and females.

Conclusion

Despite the significant harms associated with alcoholand other drug use and the need for effective and prac-tical prevention programs, there are relatively few trialsof school-based alcohol and other drug prevention pro-grams facilitated by computers or the Internet. Amongthose that do exist, it appears that the use of computersand the Internet can be effective in overcoming tradi-tional obstacles to implementation and have the poten-tial to reduce the uptake and use of alcohol and druguse in adolescents. These promising results, togetherwith the numerous implementation advantages andhigh fidelity associated with new technology, suggestthat Internet-facilitated programs offer a promisingdelivery method for school-based prevention.

Acknowledgements

Funding for this paper was supported by an NHMRCproject grant APP1004744. Dr Nicola Newton is aUNSW Vice Chancellor Post-Doctoral ResearchFellow and Professor Maree Teesson is funded on Aus-tralia National Health and Medical Research CouncilResearch Fellowship.



Competing interests

Dr Nicola Newton and Professor MareeTeesson led thedevelopment and evaluation of the Climate Schoolsprograms. They derive no financial income from theClimate Schools website.

References

[1] Australian Institute of Health and Welfare. 2010 NationalDrug Strategy Household Survey report, in drug statisticsseries no. 25. Cat. No. PHE 145. Canberra: AIHW, 2011.

[2] Hibell B, Guttormsson U, Ahlström S, et al. The 2007ESPAD report: substance use among students in 35 Euro-pean countries. Stockholm: The European School SurveyProject on Alcohol and Other Drugs, 2007.

[3] Begg S, Vos T, Barker B, Stevenson C, Stanley L, LopezAD. The burden of disease and injury in Australia 2003,PHE 82. Canberra: AIHW, 2007.

[4] Collins DJ, Lapsley HM. The costs of tobacco, alcohol andillicit drug abuse to Australian society in 2004/05. Can-berra: Commonwealth of Australia, 2008.

[5] Hall W, Degenhardt L. Adverse health effects of non-medical cannabis use. Lancet 2009;374:1383–91.

[6] Behrendt S, Wittchen HU, Höfler M, Lieb R, Beesdo K.Transitions from first substance use to substance use disor-ders in adolescence: is early onset associated with a rapidescalation? Drug Alcohol Depend 2009;99:68–78.

[7] Chen CY, Sorr CL, Anthony JC. Early-onset drug use andrisk for drug dependence problems. Addict Behav2009;34:319–22.

[8] Botvin GJ, Griffin KW. School-based programmes toprevent alcohol, tobacco and other drug use. Int Rev Psy-chiatry 2007;19:607–15.

[9] Foxcroft DR, Tsertsvadze A. Universal school-based pre-vention programs for alcohol misuse in young people.Cochrane Database Syst Rev 2011;(5)CD009113.

[10] Tobler NS. Lessons learned. J Prim Prev 2000;20:261–74.

[11] Tobler NS, Roona MR, Ochshorn P, Marshall DG, StrekeAV, Stackpole KM. School-based adolescent drug preven-tion programs: 1998 meta-analysis. J Prim Prev 2000;20:275–336.

[12] Ennett ST, Ringwalt CL, Thorne J, et al. A comparison ofcurrent practice in school based substance use programswith meta-analysis findings. Prev Sci 2003;4:1–14.

[13] Titov N. Status of computerized cognitive behaviouraltherapy for adults. Aust N Z J Psychiatry 2007;41:95–114.

[14] Joinson AN, Paine CB. Self-disclosure, privacy and theInternet. In: Joinson AN, McKenna K, Postmes T, ReipsetU-D, eds. Oxford handbook of internet psychology. Oxford:Oxford University Press, 2007:237–52.

[15] Ben Ze’ev A. Privacy, emotional closeness, and openness inCyberspace. Comput Human Behav 2003;19:451–67.

[16] Bock BC, Graham AL, Whiteley JA, Stoddard JL. A reviewof web-assisted tobacco interventions (WATIs). J MedInternet Res 2008;10:e39.

[17] Rooke S, Thorsteinsson E, Karpin A, Copeland J, Allsop D.Computer-delivered interventions for alcohol and tobaccouse: a meta-analysis. Addiction 2010;105:1381–90.

[18] Myung SK, McDonnell DD, Kazinets G, Seo HG,Moskowitz JM. Effects of Web- and computer-basedsmoking cessation programs: meta-analysis of randomizedcontrolled trials. Arch Intern Med 2009;169:929–37.

[19] Riper H, van Straten A, Keuken M, Smit F, Schippers G,Cuijpers P. Curbing problem drinking with personalized-feedback interventions: a meta-analysis. Am J Prev Med2009;36:247–55.

[20] Tait RJ, Christensen H. Internet-based interventions foryoung people with problematic substance use: a systematicreview. Med J Aust 2010;192 (11 Suppl.):S15–21.

[21] Saitz R. Alcohol screening and brief intervention in primarycare: absence of evidence for efficacy in people withdependence or very heavy drinking. Drug Alcohol Rev2010;29:631–40.

[22] Moreira MT, Smith LA, Foxcroft D. Social norms inter-ventions to reduce alcohol misuse in university or collegestudents. Cochrane Database Syst Rev 2009;(3)CD006748.

[23] Bewick BM, Trusler K, Barkham M, Hill AJ, Cahill J,Mulhern B. The effectiveness of web-based interventionsdesigned to decrease alcohol consumption – a systematicreview. Prev Med 2008;47:17–26.

[24] Carey KB, Scott-Sheldon LA, Elliott JC, Bolles JR, CareyMP. Computer-delivered interventions to reduce collegestudent drinking: a meta-analysis. Addiction 2009;104:1807–19.

[25] Gainsbury S, Blaszczynski A. A systematic review ofInternet-based therapy for the treatment of addictions. ClinPsychol Rev 2011;31:490–8.

[26] Gremeaux V, Coudeyre E. The Internet and the therapeuticeducation of patients: a systematic review of the literature.Ann Phys Rehabil Med 2010;53:669–92.

[27] Khadjesari Z, Murray E, Hewitt C, Hartley S, Godfrey C.Can stand-alone computer-based interventions reducealcohol consumption? A systematic review. Addiction2011;106:267–82.

[28] Lehto T, Oinas-Kukkonen H. Persuasive features inweb-based alcohol and smoking interventions: a systematicreview of the literature. J Med Internet Res 2011;13:e46.

[29] Moore BA, Fazzino T, Garnet B, Cutter CJ, Barry DT.Computer-based interventions for drug use disorders: asystematic review. J Subst Abuse Treat 2011;40:215–23.

[30] Portnoy DB, Scott-Sheldon LA, Johnson BT, Carey MP.Computer-delivered interventions for health promotion andbehavioral risk reduction: a meta-analysis of 75 randomizedcontrolled trials, 1988–2007. Prev Med 2008;47:3–16.

[31] Riper H, Spek V, Boon B, et al. Effectiveness of E-self-helpinterventions for curbing adult problem drinking: a meta-analysis. J Med Internet Res 2011;13:e42.

[32] Vernon ML. A review of computer-based alcohol problemservices designed for the general public. J Subst AbuseTreat2010;38:203–11.

[33] Brouwer W, Kroeze W, Crutzen R, et al. Which interven-tion characteristics are related to more exposure to internet-delivered healthy lifestyle promotion interventions? Asystematic review. J Med Internet Res 2011;13:e2.

[34] Jadad AR, Moore RA, Carroll D, et al. Assessing the qualityof reports of randomized clinical trials: is blinding neces-sary? Control Clin Trials 1996;17:1–12.

[35] Moher D, Jadad AR, Tugwell P. Assessing the quality ofrandomized controlled trials. Current issues and futuredirections. Int J Technol Assess Health Care 1996;12:195–208.

[36] Neil AL, Christensen H. Australian school-based preven-tion and early intervention programs for anxiety anddepression: a systematic review. Med J Aust 2007;186:305–8.



[37] Teesson M, Newton NC, Barrett EL. Australian school-based prevention programs for alcohol and other drugs: asystematic review. Drug Alcohol Rev 2012;31:731–6.

[38] Neil AL, Christensen H. Efficacy and effectiveness ofschool-based prevention and early intervention programsfor anxiety. Clin Psychol Rev 2009;29:208–15.

[39] Cohen JD. Statistical power analysis for the behaviouralsciences, 2nd edn. Hillsdale, NJ: Lawrence Earlbaum Asso-ciates, 1988.

[40] Evans RI. Smoking in children: developing a social psycho-logical strategy of deterrence. Prev Med 1976;5:122–7.

[41] Botvin GJ. Preventing drug abuse in schools: social andcompetence enhancement approaches targeting individual-level etiologic factors. Addict Behav 2000;25:887–97.

[42] Cuijpers P. Effective ingredients of school-based drugprevention programs. A systematic review. Addict Behav2002;27:1009–23.

[43] Koning IM, van den Eijnden RJ, Verdurmen JE, Engels RC,Vollebergh WA. Long-term effects of a parent and studentintervention on alcohol use in adolescents: a clusterrandomized controlled trial. Am J Prev Med 2011;40:541–7.

[44] Prochaska JO, Velicer WF. The transtheoretical model ofhealth behavior change. Am J Health Promot 1997;12:38–48.

[45] Buller DB, Borland R, Woodall WG, et al. Randomizedtrials on consider this, a tailored, internet-delivered smokingprevention program for adolescents. Health Educ Behav2008;35:260–81.

[46] Norman CD, Maley O, Li X, Skinner HA. Using the inter-net to assist smoking prevention and cessation in schools: arandomized, controlled trial. Health Psychol 2008;27:799–810.

[47] Prokhorov AV, Kelder SH, Shegog R, et al. Impact of asmoking prevention interactive experience (ASPIRE), aninteractive, multimedia smoking prevention and cessationcurriculum for culturally diverse high-school students.Nicotine Tob Res 2008;10:1477–85.

[48] Vogl L, Teesson M, Andrews G, Bird K, Steadman B,Dillon P. A computerized harm minimization preventionprogram for alcohol misuse and related harms: randomizedcontrolled trial. Addiction 2009;104:564–75.

[49] Newton NC, Vogl LE, Teesson M, Andrews G. CLIMATESchools: alcohol module: cross-validation of a school-based

prevention programme for alcohol misuse. Aust N Z J Psy-chiatry 2009;43:201–7.

[50] Newton NC, Andrews G, Teesson M, Vogl LE. Deliveringprevention for alcohol and cannabis using the Internet: acluster randomised controlled trial. Prev Med 2009;48:579–84.

[51] Newton NC, Teesson M, Vogl LE, Andrews G. Internet-based prevention for alcohol and cannabis use:final results ofthe Climate Schools course. Addiction 2010;105:749–59.

[52] Koning IM, Vollebergh WA, Smit F, et al. Preventing heavyalcohol use in adolescents (PAS): cluster randomized trial ofa parent and student intervention offered separately andsimultaneously. Addiction 2009;104:1669–78.

[53] Aveyard P, Sherratt E, Almond J, et al. The change-in-stageand updated smoking status results from a cluster-randomized trial of smoking prevention and cessation usingthe transtheoretical model among British adolescents. PrevMed 2001;33:313–24.

[54] Marsch LA, Bickel WK, Grabinski MJ. Application of inter-active, computer technology to adolescent substance abuseprevention and treatment. Adolesc Med State Art Rev2007;18:342–56. xii.

[55] Duncan TE, Duncan SC, Beauchamp N, Wells J, Ary DV.Development and evaluation of an interactive CD-ROMrefusal skills program to prevent youth substance use:‘refuse to use. J Behav Med 2000;23:59–72.

[56] Lord SE, D’Amante D. Efficacy of online alcohol and otherdrug prevention for early adolescents. J Adolesc Health2007;40:S4.

[57] Calear AL, Christensen H. Review of internet-based pre-vention and treatment programs for anxiety and depressionin children and adolescents. Med J Aust 2010;192 (11Suppl.):S12–4.

[58] Skager R. Having fun and defying adults: speculations onwhy most young people ignore negative information on thedangers of drinking alcohol. Addiction 2009;104:576–7.

[59] Yap M, Jorm A, Bazley R, Kelly C, Ryan S, Lubman D.Web-based parenting program to prevent adolescentalcohol misuse: rationale and development. Australas Psy-chiatry 2011;19:339–44.

[60] Teesson M, Memedovic S, Mewton L, Slade T, Baillie A.Detection, evaluation, and diagnosis of alcohol use disor-ders. In: Saunders JB, Rey JM, eds. Young people andalcohol: impact, policy, prevention, treatment. Chichester:Wiley-Blackwell, 2011:212–28.



A systematic review of school-based alcohol and other drug prevention programs facilitated by...

Documents

Transcript of A systematic review of school-based alcohol and other drug prevention programs facilitated by...