NothingasPracticalasaGoodTheory?TheTheoreticalBasisof ...downloads.hindawi.com/journals/art/2012/345327.pdf1International Centre for Reproductive Health, Faculty of Medicine and Health
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Hindawi Publishing CorporationAIDS Research and TreatmentVolume 2012, Article ID 345327, 18 pagesdoi:10.1155/2012/345327
Review Article
Nothing as Practical as a Good Theory? The Theoretical Basis ofHIV Prevention Interventions for Young People in Sub-SaharanAfrica: A Systematic Review
Kristien Michielsen,1 Matthew Chersich,1, 2 Marleen Temmerman,1
Tessa Dooms,2 and Ronan Van Rossem3
1 International Centre for Reproductive Health, Faculty of Medicine and Health Sciences, Ghent University,De Pintelaan 185 P3, 9000 Ghent, Belgium
2 Centre for Health Policy, School of Public Health, University of the Witwatersrand, Johannesburg 2000, South Africa3 Department of Sociology, Faculty of Political and Social Sciences, Ghent University, 9000 Ghent, Belgium
Correspondence should be addressed to Kristien Michielsen, [email protected]
Received 29 February 2012; Revised 26 April 2012; Accepted 3 May 2012
This paper assesses the extent to which HIV prevention interventions for young people in sub-Saharan Africa are groundedin theory and if theory-based interventions are more effective. Three databases were searched for evaluation studies of HIVprevention interventions for youth. Additional articles were identified on websites of international organisations and throughsearching references. 34 interventions were included; 25 mentioned the use of theory. Social Cognitive Theory was most prominent(n = 13), followed by Health Belief Model (n = 7), and Theory of Reasoned Action/Planned Behaviour (n = 6). These cognitivebehavioural theories assume that cognitions drive sexual behaviour. Reporting on choice and use of theory was low. Only threearticles provided information about why a particular theory was selected. Interventions used theory to inform content (n = 13), forevaluation purposes (n = 4) or both (n = 7). No patterns of differential effectiveness could be detected between studies using andnot using theory, or according to whether a theory informed content, and/or evaluation. We discuss characteristics of the theoriesthat might account for the limited effectiveness observed, including overreliance on cognitions that likely vary according to typeof sexual behaviour and other personal factors, inadequately address interpersonal factors, and failure to account for contextualfactors.
1. Introduction
With an estimated 2.7 million new infections worldwidein 2010, HIV incidence remains at very high levels [1].Sub-Saharan Africa, accounting for 70% of these infections,remains particularly affected. About 40% of new HIVinfections occur in the age group 15 to 24 years [1]. There-fore, targeted prevention programmes for young people areessential in reversing the HIV epidemic [2, 3]. Over thepast decades, a considerable number of HIV preventioninterventions for young people in sub-Saharan Africa havebeen developed, implemented, and evaluated. Nevertheless,even though these interventions seem to increase knowledge
and encourage positive attitudes, radical changes in sexualbehaviour have not occurred [4, 5].
Theory is said to be an essential component of successfulhealth promotion interventions [6, 7]. Behavioural theorycan assist to understand the determinants of risky and safesexual behaviour [8] and hence help to identify underlyingprinciples about how people change their behaviour [9].Further, it aims to explains why and how behaviours occurand allows us to predict future behaviours by establishingrelationships between key variables. Beyond providing con-structs, processes and hypotheses for setting up interven-tions, theories can also provide the basis for testing theeffectiveness of interventions [10]. Furthermore, theories
2 AIDS Research and Treatment
can serve as a framework for accumulating knowledge[11]. Reviews that assessed the theoretical underpinningsof behavioural interventions for young people worldwidegenerally claim that a theoretical foundation contributes toeffectiveness [6, 12–16], although a direct link has not yetbeen established.
In health promotion research, a large number of theoriescoexist that aim to understand health-related behaviour andprovide tools for behaviour change. The Social Learning/Cognitive Theory (SCT), Theory of Reasoned Action/Planned Behaviour (TRA/TPB), and Health Belief Model(HBM) are the most dominant theories, more recently joinedby the Stages of Change (SoC) and Social Ecological Model(SEM) [17–21].
The SCT posits that people acquire and maintain par-ticular behavioural patterns through a constant interactionbetween three factors: environment, personal factors, andbehaviour [22, 23]. Behaviour is not simply the result of theenvironment and the person, just as the environment is notmerely a function of the person and behaviour [17]. TheHBM is based on an understanding that a person will take ahealth-related action if that person believes s/he is susceptibleto the condition (perceived susceptibility), that the conditionhas serious consequences (perceived severity), that takingaction would reduce their susceptibility to the conditionor its severity (perceived benefits), and that these benefitsoutweigh the cost of taking action (perceived barriers).Action is taken more easily if the person is exposed tofactors that prompt action (cues to action) and is confidentin her/his ability to successfully perform an action (self-efficacy) [20, 24–26]. By contrast, the TRA suggests thata person’s behaviour is determined by her/his intentionto perform the behaviour. This intention is predicated bytheir attitude toward the specific behaviour and by beliefsabout whether individuals who are important to the personapprove or disapprove of the behaviour (subjective norm).The TPB includes an additional determinant: the beliefs/he has control over a particular behaviour (perceivedbehavioural control) [20, 27, 28]. SoC theory argues that, inorder to change a behaviour, an individual passes throughfive stages: precontemplation, contemplation, preparation,action, and maintenance [29]. People at different stages havedifferent informational needs and benefit from interventionstailored to their particular stage [20]. The SEM identifiesa number of interacting levels that influence behavior(individual, interpersonal, organizational, community, andpublic policy). According to this model, behaviours areshaped by the social environment [20, 30].
These dominant theories work at various levels andfor different purposes. While the HBM and TRA/TPB areexplanatory theories operating at the individual level, theSCT and SEM include the interpersonal and environmentallevels, respectively. The SoC theory, in turn, is a changetheory, not explaining a particular behaviour, but providinga framework for how people alter their behaviour.
With the overarching objective of improving effectivenessof HIV prevention interventions that target young people’ssexual behaviour in sub-Saharan Africa, this paper examinesthe extent to which these interventions are grounded in
theory, how these theories are applied and assesses if theory-based interventions are more effective in modifying sexualbehaviour than interventions not explicitly grounded intheory.
2. Methods
2.1. Study Eligibility, Literature Search, and Data Extraction.We performed a systematic review to locate evaluated inter-ventions that aim to reduce sexual risk behaviour of youngpeople in sub-Saharan Africa. Studies were considered eligi-ble if they reported on the evaluation of an HIV preventionintervention for young people on the subcontinent, had acontrol group, and were published between January 1990 andMarch 2012. Further, to be included, studies had to report onthe general population of young people (10–25 years) andthe intervention needed to aim to prevent HIV transmissionby reducing sexual risk taking. Searches were performed inthe online databases Medline (PubMed interface), ISI Web ofScience, and EBSCOhost. Additional articles were identifiedon websites of international organisations and throughsearching references of eligible articles. Data extractionwas then done in duplicate by five investigators using apredesigned and pretested extraction sheet. Further detailsof the search terms, study eligibility, and data extraction aredetailed elsewhere [4].
2.2. Study Measures. We extracted data on characteristicsof the interventions and theory use. Firstly, whether anytheory had been used and, if so, which. Secondly, forwhat purpose the theory was used. We extracted full-textdescriptions of how the theory had been used, which waslater recoded into three categories: theory used to inform theintervention (e.g., for curriculum development); theory usedto guide evaluation (e.g., to develop indicators); or both.Thirdly, a binary variable was derived, capturing whetheran explanation was provided about why this theory waschosen. For the studies not reporting the use of a theory,we looked at the topics dealt with in the interventionsand the envisaged interventions’ outcomes. This gives us anindication of the underlying theoretical assumptions used inthese interventions.
Data were also extracted on the behavioural outcomes ofthe interventions: condom use (at last sex; consistency andintention), sexual behaviour (primary abstinence; the pro-portion of sexually active youth; recent sexual intercourse;number of sexual partners and multiple partnerships), andbiological outcomes (HIV/STI incidence).
3. Results
1073 article titles and/or abstract were screened. Afteranalysis of title and abstract, we reviewed 73 full-textpublications. In total, evaluations of 34 studies met theinclusion criteria, reported on in 38 articles. Table 1 sums themain intervention characteristics and study designs.
AIDS Research and Treatment 3
Ta
ble
1:C
har
acte
rist
ics
ofst
udi
esin
clu
ded
insy
stem
atic
revi
ewof
use
ofbe
hav
iou
ralt
heo
ryin
HIV
prev
enti
onin
terv
enti
ons
inyo
uth
insu
b-Sa
har
anA
fric
a.
Au
thor
,yea
rC
oun
try
Year
ofin
terv
enti
onSt
udy
desi
gnSa
mpl
esi
zeat
base
line
(mal
es/f
emal
es)
Mai
nin
terv
enti
onac
tivi
ties
(du
rati
on)
Inte
rven
tion
sett
ing
(urb
an/r
ura
l)
Th
eory
orth
eori
esu
sed
Rol
eof
theo
ryin
the
stu
dy
Exp
lan
atio
npr
ovid
edab
out
why
theo
ryu
sed?
Cen
tral
Afr
ica
Van
Ros
sem
and
Mee
kers
[31]
Cam
eroo
n19
96-1
997
Rep
eat
C/S
,qu
asie
xper
imen
tal
1606
(753
/757
)
Beh
avio
ur
chan
geco
mm
u-
nic
atio
nan
dpr
omot
ion
thro
ugh
pee
rsan
din
med
ia,
con
dom
dist
ribu
tion
,yo
uth
-fri
endl
yse
rvic
es(1
3m
onth
s)
Com
mu
nit
y(u
rban
)H
ealt
hB
elie
fM
odel
Dev
elop
men
tof
inte
rven
tion
and
ques
tion
nai
re/
eval
uat
ion
No
Spei
zer
etal
.[3
2]C
amer
oon
1997
-199
8R
epea
tC
/S,
quas
iexp
erim
enta
l80
2(4
00/4
02)
Th
rou
ghdi
scu
ssio
ngr
oups
,on
e-on
-on
em
eeti
ngs
,an
dh
ealt
han
dsp
orta
ssoc
iati
onga
ther
ings
,pe
ered
uca
tors
info
rmed
thei
rpe
ers
and
refe
rred
them
tose
rvic
es.
Pro
mot
ion
alm
ater
ials
wer
edi
stri
bute
din
sch
ools
and
com
mu
nit
y(1
8m
onth
s)
Sch
ool
+C
om-
mu
nit
y(u
rban
)N
R,f
ocu
son
know
ledg
eN
AN
A
Mee
kers
etal
.[3
3]C
amer
oon
2000
-200
1R
epea
tC
/S,p
repo
st-c
ontr
ollin
gfo
rex
posu
re19
56(1
056/
900)
Med
iaan
din
terp
erso
nal
com
mu
nic
atio
nca
mpa
ign
.Pe
ered
uca
tion
,m
agaz
ine,
radi
odr
ama,
radi
oca
ll-in
show
,m
edia
cam
paig
n,
con
dom
prom
otio
n(1
2m
onth
s)
Com
mu
nit
y(u
rban
)
Hea
lth
Bel
ief
Mod
el,S
ocia
lLe
arn
ing
Th
eory
,Th
eory
ofR
easo
ned
Act
ion
Dev
elop
men
tof
inte
rven
tion
and
ques
tion
nai
re/
eval
uat
ion
No
4 AIDS Research and Treatment
Ta
ble
1:C
onti
nu
ed.
Au
thor
,yea
rC
oun
try
Year
ofin
terv
enti
onSt
udy
desi
gnSa
mpl
esi
zeat
base
line
(mal
es/f
emal
es)
Mai
nin
terv
enti
onac
tivi
ties
(du
rati
on)
Inte
rven
tion
sett
ing
(urb
an/r
ura
l)
Th
eory
orth
eori
esu
sed
Rol
eof
theo
ryin
the
stu
dy
Exp
lan
atio
npr
ovid
edab
out
why
theo
ryu
sed?
Eas
tern
Afr
ica
Kle
ppet
al.
[34,
35]
Tan
zan
ia19
90C
ohor
t,ra
ndo
miz
edsc
hoo
ls10
63(5
02/5
61)
Teac
her
spr
ovid
edin
form
a-ti
on,s
tude
nts
crea
ted
post
ers
and
per
form
edso
ngs
,poe
try,
dram
aan
dro
le-p
lay,
smal
l-gr
oup
disc
uss
ion
sam
ong
stu
-de
nts
.In
terv
iew
san
dpa
nel
disc
uss
ion
sw
ith
pare
nts
and
com
mu
nit
ym
embe
rs(2
-3m
onth
s)
Pri
mar
ysc
hoo
l(u
rban
+ru
ral)
Soci
alLe
arn
ing
Th
eory
and
Th
eory
ofR
easo
ned
Act
ion
Dev
elop
men
tof
inte
rven
tion
and
ques
tion
nai
re/
eval
uat
ion
No
Shu
eyet
al.
[36]
Uga
nda
1994
–199
6R
epea
tC
/S,
quas
iexp
erim
enta
l80
0(3
98/4
02)
Stre
ngt
hen
exis
tin
gsc
hoo
lh
ealt
hcu
rric
ulu
m,
mee
tin
gw
ith
pare
nts
and
com
mu
nit
yle
ader
s,fo
rmat
ion
ofsc
hoo
lh
ealt
hcl
ubs
wit
hp
eer
edu
ca-
tion
,qu
esti
onbo
xes
(2ye
ars)
Pri
mar
ysc
hoo
l(u
rban
+ru
ral)
Soci
alC
ogn
itiv
eT
heo
ryN
RN
o
Kin
sman
etal
.[37
]U
gan
da19
97-1
998
Coh
ort,
quas
iexp
erim
enta
l20
77(9
20/1
157)
Ext
racu
rric
ula
rcl
asse
sby
trai
ned
teac
her
s(1
year
)
Pri
mar
yan
dse
con
dary
sch
ools
(ru
ral)
Beh
avio
ur
Ch
ange
sfo
rIn
terv
enti
ons
Mod
el
Dev
elop
men
tof
inte
rven
tion
and
ques
tion
nai
re/
eval
uat
ion
No
Eru
lkar
etal
.[3
8]K
enya
1998
–200
0R
epea
tC
/S,
quas
iexp
erim
enta
l15
44(7
92/7
52)
Adu
ltco
un
sello
rin
com
mu
-n
ity
edu
cati
ng
you
th,r
efer
ral
toyo
uth
-fri
endl
yse
rvic
esan
den
cou
ragi
ng
pare
nt-
child
com
mu
nic
atio
n(3
year
s)
Com
mu
nit
y(u
rban
+ru
ral)
NR
,foc
us
onva
lues
,kn
owle
dge,
gen
der,
and
empo
wer
men
t
NA
NA
AIDS Research and Treatment 5
Ta
ble
1:C
onti
nu
ed.
Au
thor
,yea
rC
oun
try
Year
ofin
terv
enti
onSt
udy
desi
gnSa
mpl
esi
zeat
base
line
(mal
es/f
emal
es)
Mai
nin
terv
enti
onac
tivi
ties
(du
rati
on)
Inte
rven
tion
sett
ing
(urb
an/r
ura
l)
Th
eory
orth
eori
esu
sed
Rol
eof
theo
ryin
the
stu
dy
Exp
lan
atio
npr
ovid
edab
out
why
theo
ryu
sed?
Ros
set
al.
[39]
,D
oyle
etal
.[4
0]
Tan
zan
ia19
98–2
002
Rep
eat
C/S
,ra
ndo
miz
edco
mm
un
itie
s
Ros
s:92
19(5
103/
4116
)D
oyle
:138
14(7
300/
6514
)
Part
icip
ator
y,te
ach
er-l
ed,
peer
-ass
iste
d,in
-sch
ool
pro-
gram
,yo
uth
-fri
end
lyh
ealt
hse
rvic
es,
con
dom
prom
otio
nan
ddi
stri
buti
on,
and
you
thh
ealt
hda
ysan
dvi
deo
show
sin
com
mu
nit
y(3
year
s)
Sch
ool
+C
om-
mu
nit
y(r
ura
l)So
cial
Lear
nin
gT
heo
ryD
evel
opm
ent
ofin
terv
enti
onN
o
Mat
icka
-Ty
nda
leet
al.
[41]
Ken
ya20
02-2
003
Rep
eat
C/S
,ra
ndo
miz
edsc
hoo
ls73
92(3
636/
3764
)
Peer
edu
cati
onon
leve
lof
teac
her
san
dst
ude
nts
,qu
es-
tion
boxe
s,sc
hoo
lh
ealt
hcl
ubs
,in
form
atio
nco
rner
san
das
sem
blie
s,dr
ama,
mu
sic
and
liter
ary
per
form
-an
ces
(18
mon
ths)
Pri
mar
ysc
hoo
l(u
rban
+ru
ral)
Soci
alLe
arn
ing
Th
eory
and
Scri
ptin
gT
heo
ry
Dev
elop
men
tof
inte
rven
tion
Yes
Rijs
dijk
etal
.[4
2]U
gan
da20
08C
ohor
t,ra
ndo
miz
edsc
hoo
ls19
86(1
096/
889)
low
-tec
h,
com
pute
r-ba
sed,
inte
ract
ive
com
preh
ensi
vese
xed
uca
tion
prog
ram
me,
teac
her
-led
(6m
onth
s)
Seco
nda
rysc
hoo
l(u
rban
+ru
ral)
Th
eory
ofP
lan
ned
Beh
avio
ran
dH
ealt
hB
elie
fM
odel
Dev
elop
men
tof
inte
rven
tion
No
Sou
ther
nA
fric
a
Ku
hn
etal
.[4
3]So
uth
Afr
ica
1990
Rep
eat
C/S
,qu
asie
xper
imen
tal
567
(not
repo
rted
)In
ten
se,h
igh
-pro
file
focu
son
AID
Sin
the
sch
ool
byte
ach
-er
s(2
wee
ks)
Seco
nda
rysc
hoo
l(u
rban
)
NR
,foc
us
onkn
owle
dge
and
atti
tude
sN
AN
A
6 AIDS Research and Treatment
Ta
ble
1:C
onti
nu
ed.
Au
thor
,yea
rC
oun
try
Year
ofin
terv
enti
onSt
udy
desi
gnSa
mpl
esi
zeat
base
line
(mal
es/f
emal
es)
Mai
nin
terv
enti
onac
tivi
ties
(du
rati
on)
Inte
rven
tion
sett
ing
(urb
an/r
ura
l)
Th
eory
orth
eori
esu
sed
Rol
eof
theo
ryin
the
stu
dy
Exp
lan
atio
npr
ovid
edab
out
why
theo
ryu
sed?
Har
vey
etal
.[4
4]So
uth
Afr
ica
1993
-199
4C
ohor
t,ra
ndo
miz
edsc
hoo
ls10
80(4
47/6
33)
“Sch
ool
open
day”
wit
hdr
ama,
son
g,da
nce
,poe
try,
and
post
ers
prep
ared
and
pres
ente
dby
stu
den
ts(3
days
)
Seco
nda
rysc
hoo
l(u
rban
+ru
ral)
App
lied
beh
avio
ur
chan
gefr
amew
ork
Dev
elop
men
tof
ques
tion
nai
re/
eval
uat
ion
No
Mee
kers
[45]
Sou
thA
fric
a19
94–1
997
Rep
eat
C/S
,qu
asie
xper
imen
tal
226
(0/2
26)
Mas
sm
edia
cam
paig
n,p
eer
edu
cati
onan
dco
ndo
mpr
o-m
otio
nan
ddi
stri
buti
on(3
5m
onth
s)
Com
mu
nit
y(u
rban
)H
ealt
hB
elie
fM
odel
Dev
elop
men
tof
ques
tion
nai
re/
eval
uat
ion
No
Fitz
gera
ldet
al.;
Stan
ton
etal
.[46
,47]
Nam
ibia
1996
Coh
ort,
ran
dom
ized
part
icip
ants
515
(236
/279
)C
urr
icu
lum
tau
ght
bya
teac
her
and
out-
of-s
choo
lyo
uth
(7w
eeks
)
Seco
nda
rysc
hoo
l(u
rban
+ru
ral)
Soci
alC
ogn
itiv
eT
heo
ry/
Pro
tect
ive
Mot
ivat
ion
alT
heo
ry
Dev
elop
men
tof
inte
rven
tion
and
ques
tion
nai
re/
eval
uat
ion
No
Kim
etal
.[4
8]Z
imba
bwe
1997
-199
8R
epea
tC
/S,
quas
iexp
erim
enta
l14
26(7
13/7
13)
Mas
sm
edia
cam
paig
n,
com
mu
nit
ydr
ama
grou
ps,
peer
edu
cato
rs,
you
th-
frie
ndl
yh
ealt
hse
rvic
es(6
mon
ths)
Sch
ool
+C
om-
mu
nit
y(u
rban
)
Step
sto
Beh
avio
ur
Ch
ange
Fram
ewor
k
Dev
elop
men
tof
inte
rven
tion
No
Jam
eset
al.
[49]
Sou
thA
fric
a19
98C
ohor
t,ra
ndo
miz
edsc
hoo
ls11
68(5
42/6
16)
Rea
din
gof
aco
mic
book
(1h
our)
Seco
nda
rysc
hoo
l(u
rban
+ru
ral)
Th
eory
ofH
ealt
hP
rom
otio
nan
dSo
cial
Lear
nin
g
Dev
elop
men
tof
inte
rven
tion
No
AIDS Research and Treatment 7
Ta
ble
1:C
onti
nu
ed.
Au
thor
,yea
rC
oun
try
Year
ofin
terv
enti
onSt
udy
desi
gnSa
mpl
esi
zeat
base
line
(mal
es/f
emal
es)
Mai
nin
terv
enti
onac
tivi
ties
(du
rati
on)
Inte
rven
tion
sett
ing
(urb
an/r
ura
l)
Th
eory
orth
eori
esu
sed
Rol
eof
theo
ryin
the
stu
dy
Exp
lan
atio
npr
ovid
edab
out
why
theo
ryu
sed?
Vis
ser
[50]
Sou
thA
fric
a19
98–2
000
Rep
eat
C/S
,pre
post
-con
trol
ling
for
expo
sure
873
(410
/463
)
Trai
ned
teac
her
san
dpr
o-fe
ssio
nal
spr
ovid
elif
esk
ills
and
HIV
/AID
Sed
uca
tion
.Pa
ren
tsin
clu
ded
inac
tion
com
mit
tee
(1ye
ar)
Seco
nda
rysc
hoo
l(u
rban
)H
ealt
hB
elie
fM
odel
Dev
elop
men
tof
inte
rven
tion
No
Un
derw
ood
etal
.[51
]Z
ambi
a19
99-2
000
Rep
eat
C/S
,qu
asie
xper
imen
tal
921
(378
/543
)Pa
rtic
ipat
ory
deve
lope
dm
ass
med
iaca
mpa
ign
(7m
onth
s)
Com
mu
nit
y(u
rban
+ru
ral)
Stag
eT
heo
ryof
Beh
avio
ur
Ch
ange
Dev
elop
men
tof
inte
rven
tion
No
Mag
nan
iet
al.[
52]
Sou
thA
fric
a19
99–2
001
Coh
ort,
pre
post
-con
trol
ling
for
expo
sure
3052
(137
5/16
77)
Lif
esk
ills
curr
icu
lum
tau
ght
byte
ach
ers
(2ye
ars)
Seco
nda
rysc
hoo
l(u
rban
)So
cial
Lear
nin
gT
heo
ryD
evel
opm
ent
ofin
terv
enti
onN
o
Agh
a[5
3]Z
ambi
a20
00C
ohor
t,ra
ndo
miz
edsc
hoo
ls48
1(2
68/2
13)
Peer
edu
cato
rsu
sin
gdi
s-cu
ssio
nan
ddr
ama
skit
s(1
hou
r45
min
)
Seco
nda
rysc
hoo
l(u
rban
)
NR
,foc
us
onkn
owle
dge,
nor
mat
ive
belie
fs,a
nd
risk
per
cept
ion
NA
NA
Jam
eset
al.
[54]
Sou
thA
fric
a20
01C
ohor
t,ra
ndo
miz
edsc
hoo
ls93
6(4
56/4
66)
Life
skill
sin
terv
enti
onta
ugh
tby
trai
ned
teac
her
s(2
0w
eeks
)
Seco
nda
rysc
hoo
l(u
rban
+ru
ral)
Soci
alC
ogn
itiv
eT
heo
ryan
dT
heo
ryof
Pla
nn
edB
ehav
iou
r
Dev
elop
men
tof
ques
tion
nai
re/e
valu
atio
nN
o
8 AIDS Research and Treatment
Ta
ble
1:C
onti
nu
ed.
Au
thor
,ye
arC
oun
try
Year
ofin
terv
enti
onSt
udy
desi
gnSa
mpl
esi
zeat
base
line
(mal
es/f
emal
es)
Mai
nin
terv
enti
onac
tivi
ties
(du
rati
on)
Inte
rven
tion
sett
ing
(urb
an/r
ura
l)
Th
eory
orth
eori
esu
sed
Rol
eof
theo
ryin
the
stu
dy
Exp
lan
atio
npr
ovid
edab
out
why
theo
ryu
sed?
Pla
utz
etal
.[55
]M
adag
asca
r20
01-2
002
Coh
ort,
pre
post
-con
trol
ling
for
expo
sure
1785
(100
0/78
5)
You
th-f
rien
dly
serv
ices
,m
ass
med
ia,a
nd
inte
r-pe
rson
alco
m-
mu
nic
atio
nby
pee
red
uca
tors
(23
mon
ths)
Com
mu
nit
y(u
rban
+ru
ral)
Soci
alLe
arn
ing
Th
eory
,Hea
lth
Bel
iefM
odel
,an
dT
heo
ryof
Rea
son
edA
ctio
n
Dev
elop
men
tof
inte
rven
tion
and
ques
tion
nai
re/
eval
uat
ion
No
Kar
nel
let
al.[
56]
Sou
thA
fric
a20
02C
ohor
t,ra
ndo
miz
edsc
hoo
ls66
1(3
24/3
37)
Peer
edu
cato
rsu
sin
gre
cord
edm
onol
ogu
esof
fict
ion
alch
arac
-te
rs,t
each
ersu
ppor
t(8
wee
ks)
Seco
nda
rysc
hoo
l(u
rban
)
Soci
alLe
arn
ing
Th
eory
,Soc
ial
Inoc
ula
tion
,C
ogn
itiv
eB
ehav
iou
rT
heo
ry
Dev
elop
men
tof
inte
rven
tion
and
ques
tion
nai
re/
eval
uat
ion
Yes
Vis
ser
[57]
Sou
thA
fric
a20
02-2
003
Rep
eat
C/S
,qu
asie
xper
imen
tal
1918
(858
/106
0)Pe
ered
uca
tion
(18
mon
ths)
Seco
nda
rysc
hoo
ls(u
rban
)Sy
stem
sT
heo
ryD
evel
opm
ent
ofin
terv
enti
onN
o
Jew
kes
etal
.[58
,59]
Sou
thA
fric
a20
03-2
004
Coh
ort,
ran
dom
ized
com
mu
nit
ies
2776
(136
0/14
16)
Part
icip
ator
yle
arn
ing
ap-
proa
ches
tau
ght
byfa
cilit
ator
s,pe
ergr
oup
mee
tin
g,co
mm
u-
nit
ym
eeti
ng
(6–8
wee
ks)
Com
mu
nit
y(r
ura
l)
Part
icip
ator
yLe
arn
ing
App
roac
han
dA
dult
Edu
cati
onT
heo
ry
Dev
elop
men
tof
inte
rven
tion
No
AIDS Research and Treatment 9
Ta
ble
1:C
onti
nu
ed.
Au
thor
,yea
rC
oun
try
Year
ofin
terv
enti
onSt
udy
desi
gnSa
mpl
esi
zeat
base
line
(mal
es/f
emal
es)
Mai
nin
terv
enti
onac
tivi
ties
(du
rati
on)
Inte
rven
tion
sett
ing
(urb
an/r
ura
l)
Th
eory
orth
eori
esu
sed
Rol
eof
theo
ryin
the
stu
dy
Exp
lan
atio
npr
ovid
edab
out
why
theo
ryu
sed?
Tib
bits
etal
.[6
0]So
uth
Afr
ica
2004
-200
5C
ohor
t,ra
ndo
miz
edsc
hoo
ls40
40(2
020/
2020
)C
ompr
ehen
sive
,ri
sk-r
edu
ctio
nlif
esk
ills
curr
icu
lum
for
adol
es-
cen
ts,t
each
er-l
ed(2
4m
onth
s)
Seco
nda
rysc
hoo
l,u
rban
Sele
ctiv
eop
tim
izat
ion
wit
hco
mp
ensa
tion
,Se
lf-
Det
erm
inat
ion
Th
eory
,an
dSo
cial
Cog
nit
ive
Th
eory
Dev
elop
men
tof
inte
rven
tion
Yes
Mas
on-J
ones
etal
.[61
]So
uth
Afr
ica
2007
-200
8C
ohor
t,qu
asie
xper
imen
tal
3934
(166
1/22
11)
Gov
ern
men
t-le
dpe
ered
uca
tion
proj
ect,
incl
ass
stan
dard
cur-
ricu
lum
,co
nver
sati
ons
outs
ide
clas
s,re
ferr
al(1
8m
onth
s)
Seco
nda
rysc
hoo
l(u
rban
+ru
ral)
NR
,kn
owle
dge
and
psyc
hos
ocia
lch
arac
teri
stic
s
NA
NA
Bai
rdet
al.
[62]
Mal
awi
2008
-200
9C
ohor
t,ra
ndo
miz
edsc
hoo
ls37
96(0
/379
6)M
onth
lyca
shtr
ansf
erpr
o-gr
amm
eto
redu
ceth
eri
skof
STI
infe
ctio
n(2
4m
onth
s)
Sch
ool
+co
m-
mu
nit
y(u
rban
+ru
ral)
NR
,foc
us
onst
ruct
ura
lfac
tor
(pov
erty
and
edu
cati
on)
and
know
ledg
e
NA
NA
Bu
rnet
tet
al.
[63]
Swaz
ilan
dN
RC
ohor
t,ra
ndo
miz
edyo
uth
204
(101
/103
)
Teac
her
-led
life-
skill
sH
IVpr
even
tion
edu
cati
onpr
ogra
m,
curr
icu
lum
,in
tera
ctiv
ete
ch-
niq
ues
,ro
lepl
ayin
g,an
dgr
oup
disc
uss
ion
s(1
3w
eeks
)
Seco
nda
rysc
hoo
l(u
rban
)
Self
-effi
cacy
theo
ryan
dP
rote
ctio
nM
otiv
atio
nT
heo
ry
Dev
elop
men
tof
inte
rven
tion
No
10 AIDS Research and Treatment
Ta
ble
1:C
onti
nu
ed.
Au
thor
,yea
rC
oun
try
Year
ofin
terv
enti
onSt
udy
desi
gnSa
mpl
esi
zeat
base
line
(mal
es/f
emal
es)
Mai
nin
terv
enti
onac
tivi
ties
(du
rati
on)
Inte
rven
tion
sett
ing
(urb
an/r
ura
l)
Th
eory
orth
eori
esu
sed
Rol
eof
theo
ryin
the
stu
dy
Exp
lan
atio
npr
ovid
edab
out
why
theo
ryu
sed?
Wes
tern
Afr
ica
Bri
eger
etal
.[6
4]N
iger
iaan
dG
han
a19
94–1
997
Rep
eat
C/S
,qu
asie
xper
imen
tal
1784
(not
repo
rted
)
Peer
edu
cato
rs,
prom
otio
nof
com
mu
nit
y-le
vel
net
-w
orks
,re
ferr
alto
serv
ices
(30
mon
ths)
Sch
ool
+C
om-
mu
nit
y(u
rban
)
NR
,foc
us
onkn
owle
dge
and
atti
tude
sN
AN
A
Faw
ole
etal
.[6
5]N
iger
ia19
96C
ohor
t,pr
epo
st-c
ontr
ollin
gfo
rex
posu
re45
0(2
04/2
46)
Edu
cati
onse
ssio
ns
byco
m-
mu
nit
yph
ysic
ian
sw
ith
hel
pof
teac
her
s(1
mon
th)
Seco
nda
rysc
hoo
l(u
rban
)
NR
,foc
us
onkn
owle
dge
and
atti
tude
sN
AN
A
Oko
nof
ua
etal
.[66
]N
iger
ia19
97-1
998
Rep
eat
C/S
,ra
ndo
miz
edsc
hoo
ls18
96(8
77/1
008)
Est
ablis
hm
ent
ofre
pro-
duct
ive
hea
lth
clu
bin
sch
ool,
hea
lth
awar
enes
sca
mpa
ign
sby
prof
essi
onal
s,di
stri
buti
onof
prin
tm
ate-
rial
,p
eer
edu
cati
on,
you
th-
frie
ndl
yse
rvic
es(1
1m
onth
s)
Seco
nda
rysc
hoo
l(u
rban
)
NR
,foc
us
onkn
owle
dge
and
barr
iers
NA
NA
Van
Ros
sem
and
Mee
kers
[67]
Gu
inea
1997
-199
8C
ohor
t,qu
asie
xper
imen
tal
2016
(925
/109
1)
Peer
edu
cato
rs(d
iscu
ssio
nan
dth
eatr
e),
con
dom
pro-
mot
ion
,bi
llboa
rds,
you
th-
frie
ndl
yse
rvic
esan
dco
n-
trac
epti
ondi
stri
buti
on(8
mon
ths)
Com
mu
nit
y(u
rban
)H
ealt
hB
elie
fM
odel
Dev
elop
men
tof
ques
tion
nai
re/
eval
uat
ion
No
Atw
ood
etal
.[6
8]Li
beri
a20
07-2
008
Coh
ort,
ran
dom
ized
sch
ools
812
(455
/357
)C
urr
icu
lum
-bas
edpr
ogra
mby
hea
lth
edu
cato
rs(8
wee
ks)
Pri
mar
ysc
hoo
l(u
rban
)
Soci
alC
ogn
itiv
eT
heo
ryan
dT
heo
ryof
Rea
son
edA
ctio
n
Dev
elop
men
tof
inte
rven
tion
No
C/S
:Rep
eate
dcr
oss-
sect
ion
alde
sign
.N
R:N
oth
eory
isex
plic
itly
repo
rted
,dom
inan
tco
nst
ruct
su
sed
inth
ein
terv
enti
on.
NA
:Not
appl
icab
le.
AIDS Research and Treatment 11
3.1. Theoretical Basis of the Interventions. About three quar-ters of the studies—25 of 34—mentioned having used at leastone theory. In total, 19 different theories were mentioned42 times. Several stated that they had applied two or moretheories, with three papers reporting that the interventiondesign drew on three theories.
Of all the theories mentioned, the SCT was mostprominent (n = 13). Other theories that were mentionedmore than once are the HBM (n = 7) and the TRA/TPB(n = 6). Four studies mentioned using a behaviourchange framework: behaviour changes for interventionsmodel [37], applied behaviour change framework [44], stepsto behaviour change framework [48] and stage theory ofbehaviour change [51]. Assessment of the concepts usedin the interventions not explicitly mentioning the useof a theory indicated that they also operated from anassumption that knowledge, attitudes, beliefs, and/or rolemodels determine sexual behaviour. Hence, it seems thatmost interventions are implicitly or explicitly guided bycognitive behavioural frameworks. The one exception isBaird (2012); this intervention uses an indirect pathway totry to influence HIV incidence, namely, through encouraginggirls’ school attendance.
Description of the main activities indicates that mostinterventions use one or a combination of participatorylearning techniques, such as drama plays, poetry, songs, clubformation, peer education (role modelling), and discussionsand debates. This suggests that the learning strategies ofmost interventions were based on participatory learningapproaches.
A small, but considerable proportion of interventions[32, 34–36, 38–40, 58, 62, 64] go beyond focusing on theindividual young person and facilitate community involve-ment in the interventions. Here, the implicit theoreticalassumption is that in order to change the participants’ sexualbehaviour, the community needs to be involved (cf. SEM).
There is no clear evolution detectable over time in thefrequency of use of different theories; of the 20 studies whichbegan in the decade 1990–1999, 14 reported theory use, while11 of the 14 beginning after 2000 used theory.
3.2. Use of Theory in the Research Projects. Of 25 inter-ventions that mentioned a theory, 7 said that the theorywas used to both inform the content of the intervention(e.g., the curriculum) and to inform the evaluation orquestionnaire design. In 13 studies, theory was reportedlyused only to inform the intervention content, and in 4only for designing the evaluation or questionnaire. Onestudy mentioning theory use did not specify how this wasapplied [36]. The SCT was almost exclusively used to informthe intervention. The HBM was mostly used for evaluationpurposes, predominately in studies from Population ServicesInternational [31, 33, 45, 55, 67], as was the TRA/TPB.
Only three articles provided information on why aparticular theory was selected [41, 56, 60]. Nine authorslimited themselves to a brief explanation of the theory itself[31, 37, 42, 45, 48, 50, 51, 63, 67]. The remainder did notprovide any information on theory selection.
3.3. Theory Use and Intervention Effectiveness. Overall, thebehavioural outcomes of the 34 studies were markedlyheterogeneous, with little reduction in heterogeneity afterstratifying by theory use (Table 2). It was not possible todiscern any patterns in differential effectiveness between therole of theory in a study, or between studies reporting or notreporting theory use. Nor did we find particular differencesin intervention design by theory use.
Four studies reported biological measures of interventioneffectiveness [39, 40, 58, 59, 62]. Jewkes succeeded inreducing HSV-2 incidence. Baird’s study, not explicitly basedon theory, reported a reduced HIV incidence and HSV-2 incidence in the intervention group as compared to thecontrol group, but these data were not controlled for baselineprevalence and should be treated with some caution [62].Since the three other studies reporting biological measuresall based their intervention on a theory, it is not possible tocompare the effectiveness of theory- and non-theory-basedinterventions in changing these outcomes.
3.4. Evaluation of the Theory. Four studies refer to theirtheoretical basis in their conclusions, criticizing the the-ory, specifically “the theoretical approaches underlying theprogram have built in shortcomings which could resultin the program not having significant impact on thestudents’ behavioural intentions” [69]; “the discrepancies inthe findings may be substantiated by the lack of system-atic information that was available on the empirical andtheoretical underpinnings upon which the KwaZulu-NatalDepartment of Education’s program was based—a findingsimilar to reports of those educational programs that werenot grounded in a theoretical understanding of adolescentsexual behaviour [. . .]” [54]; “These findings present mixedevidence regarding the relationship between self-efficacyand outcome expectations and HIV protective behavioursamong adolescents in Swaziland.” [63]; “TPB has receivedconsiderably more support from research for its predictivepower of safe sex behaviour than the HBM.” [42].
4. Discussion
The review found that the majority of HIV preventioninterventions targeted at youth in sub-Saharan Africa usetheory-based approaches. A wide range of theories have beenemployed, but three behavioural theories predominate: SCT,HBM, and TRA/TPB. No one theory emerged dominant, asreporting on the choice, use, and specific evaluation of theorywas low.
4.1. Comparison with Other Reviews. Broadly, the results areconsistent with reviews of HIV risk-reduction interventionselsewhere, though some variation in use of theory can benoted across these reviews. Pedlow and Carey [70] reviewed23 randomized controlled trials of HIV risk-reductioninterventions for adolescents in the United States and foundan explicit theoretical rationale in all but one study. Similarto our review, SCT was most common (18/23). Three othertheories were used in four or more studies (TRA, HBM,
12 AIDS Research and Treatment
Ta
ble
2:D
escr
ipti
onof
stu
dyou
tcom
esst
rati
fied
byro
leof
theo
ryu
sein
each
stu
dy.
Con
dom
use
atla
stse
xEv
er/c
onsi
sten
tly
use
dco
ndo
m
Sexu
alde
but,
prop
orti
onof
sexu
ally
acti
veyo
uth
Sexu
alin
terc
ours
ein
past
mon
ths
Nu
mbe
rof
sexu
alpa
rtn
ers
HIV
inci
den
ceH
SV-2
Oth
erST
IsO
verv
iew
Th
eory
use
dfo
rde
velo
pmen
tof in
terv
enti
on
Vis
ser,
2005
◦R
oss,
2007
++
Vis
ser,
2005
−−M
atic
ka-T
ynda
le,
2007
◦V
isse
r,20
05◦
Ros
s,20
07◦
Ros
s,20
07◦
Ros
s,20
07◦
44ou
tcom
es
Mag
nan
i,20
05+
+U
nde
rwoo
d,20
06+
+M
agn
ani,
2005
−−V
isse
r,20
07+
+M
agn
ani,
2005
++
Jew
kes,
2008
◦Je
wke
s,20
08+
+Je
wke
s,20
08◦
++
7
Mat
icka
-Tyn
dale
,20
07+
Kim
,200
1◦M
atic
ka-T
ynda
le,
2007
+K
im,2
001+
Ros
s,20
07+
Doy
le,2
010◦
Doy
le,2
010◦
Doy
le,2
010◦
+10
Ros
s,20
07+
Mag
nan
i,20
05+
+R
oss,
2007
◦V
isse
r,20
07−−
◦ 22
Jew
kes,
2008
◦A
twoo
d20
12◦
Kle
pp,1
997+
Kim
,200
1+− 1
Un
derw
ood,
2006
◦K
im,2
001+
Atw
ood,
2012
−−−−
4
Vis
ser,
2007
◦U
nde
rwoo
d,20
06−
Doy
le,2
010+
Doy
le,2
010◦
Atw
ood,
2012
+
Tib
bits
,201
1◦B
urn
et,2
011◦
Doy
le,2
010◦
Tib
bits
,201
1◦
Jam
es,2
006+
Har
vey,
2000
++
Har
vey,
2000
◦Ja
mes
,200
6+H
arve
y,20
00◦
12ou
tcom
es
Mee
kers
,199
8◦M
eeke
rs,1
998◦
Van
Ros
sem
,199
9+V
anR
osse
m,
1999
◦+
+1
Th
eory
use
dfo
rde
velo
pmen
tof
eval
uat
ion
or ques
tion
nai
re
Van
Ros
sem
,199
9+V
anR
osse
m,1
999+
Mee
kers
,199
8◦+
5 ◦ 6 − 0 −−0
AIDS Research and Treatment 13
Ta
ble
2:C
onti
nu
ed.
Con
dom
use
atla
stse
xEv
er/c
onsi
sten
tly
use
dco
ndo
m
Sexu
alde
but,
prop
orti
onof
sexu
ally
acti
veyo
uth
Sexu
alin
terc
ours
ein
past
mon
ths
Nu
mbe
rof
sexu
alpa
rtn
ers
HIV
inci
den
ceH
SV-2
Oth
erST
IsO
verv
iew
Van
Ros
sem
,200
0◦V
anR
osse
m,2
000+
+Fi
tzge
rald
,199
9◦Fi
tzge
rald
,199
9◦Fi
tzge
rald
,19
99◦
15ou
tcom
es
Fitz
gera
ld,1
999◦
Mee
kers
,200
5+V
anR
osse
m,2
000◦
Van
Ros
sem
,20
00+
++
1
Th
eory
use
dfo
rde
velo
pmen
tof in
terv
enti
onan
dev
alu
atio
nor
ques
tion
nai
re
Mee
kers
,200
5+Fi
tzge
rald
,199
9◦P
lau
tz,2
003−
+3
Pla
utz
,200
3◦P
lau
tz,2
003◦
◦ 10
Kar
nel
l,20
05◦
− 1 −−0
Th
eory
use
d,bu
tu
nce
rtai
nin
wh
ich
phas
eof
stu
dy
Shu
ey+
+1
outc
ome
++
1
Spei
zer+
Agh
a,20
02◦
Spei
zer+
+Sp
eize
r−−
Agh
a,20
02+
+B
aird
,201
2++
Bai
rd,2
012+
+B
aird
,201
2◦18
outc
omes
Agh
a,20
02◦
Oko
nof
ua+
+B
rieg
er,2
001−
−A
gha,
2002
++
Faw
ole,
1999
◦+
+6
Inte
rven
tion
sn
otex
plic
itly
base
don
theo
ry
Eru
lkar
,200
4+K
uh
n◦
Faw
ole,
1999
◦E
rulk
ar,2
004◦
Eru
lkar
,200
4++
4
Faw
ole,
1999
◦Fa
wol
e,19
99◦
Eru
lkar
,200
4+◦ 1
2M
ason
-Jon
es,
2011
◦B
aird
,201
2◦B
aird
,201
2◦− 0
Mas
on-J
ones
,20
11−−
−−3
++
sign
ifica
nt
posi
tive
inte
rven
tion
effec
ton
outc
ome
vari
able
for
the
wh
ole
stu
dypo
pula
tion
.+
sign
ifica
nt
posi
tive
inte
rven
tion
impa
cton
outc
ome
vari
able
for
asu
bgro
up
ofth
eta
rget
pop
ula
tion
,an
dn
osi
gnifi
can
tim
pact
onth
ew
hol
est
udy
pop
ula
tion
orw
hol
ep
opu
lati
onim
pact
not
rep
orte
d.◦ n
osi
gnifi
can
tin
terv
enti
onim
pact
onth
eou
tcom
eva
riab
le.
− sig
nifi
can
tn
egat
ive
inte
rven
tion
effec
ton
outc
ome
vari
able
ina
sub-
grou
pof
the
targ
etp
opu
lati
on,a
nd
no
sign
ifica
nt
impa
cton
the
wh
ole
stu
dyp
opu
lati
onor
wh
ole
pop
ula
tion
impa
ctn
otre
port
ed.
−−si
gnifi
can
tn
egat
ive
inte
rven
tion
effec
ton
the
outc
ome
vari
able
for
the
wh
ole
stu
dyp
opu
lati
on.
14 AIDS Research and Treatment
and Information-Motivation-Behavioural Skills Model). Areview on the impact of HIV and sex education programs onyouth throughout the world [6] found that more than fourfifths of the 83 interventions identified one or more theory.SCT formed the basis for more than half (54%) of theseinterventions. TRA (19%), HBM (12%), TPB (10%), and theInformation-Motivation-Behavioural Skills Model (10%)were also commonly mentioned. Two other reviews coveringHIV, STD, or pregnancy risk-reduction interventions amongadolescents in the United States had comparable results, witha similar distribution of theories used [12, 71]. While severalnew theories or integrated models have been developed sincethe outbreak of HIV focussing specifically on sexual healthbehaviours like condom use [8, 72], they are not used in HIVinterventions for young people in sub-Saharan Africa.
4.2. Gaps in (the Use of) Theory. By focussing on cognitiveconstructs of behaviour, the interventions explicitly orimplicitly start from the assumption that cognitions influ-ence the person’s thinking and decision making, and thusdrive sexual behaviour [73]. In the remaining discussion,we will focus on the utility of grounding HIV preventioninterventions for young people on a cognitive behaviouralframework. We attempt to identify critical areas for attentionand improvement on different levels.
Firstly, cognitive behavioural models aim to explain aparticular behaviour. The theoretical constructs that influ-ence behavioural decisions may vary, depending on thebehaviour in question. This poses marked challenges forHIV prevention interventions, since they generally attemptto influence a wide range of behaviours—for example,increasing condoms use, reducing the number of sexualpartners, minimising sexual activity, delaying the onset ofsexual debut—which are influenced by different factors.To further complicate matters, sexual decisions may varydepending on the reasons for sexual intercourse (rangingbetween, e.g., intimacy or desire, external factors, and affectmanagement). These, in turn, are further influenced bygender and psychological characteristics (e.g., depression,self-esteem, and impulsiveness) [74]. Thus, sexual behaviouritself is far from a uniform behaviour, but rather a collectionof several relatively distinct behaviours, that can be shapedby different factors in different contexts. While the use of atheoretical framework provides grip in structuring an HIVprevention intervention, the interventionist needs to be veryclear about what behaviour they aim to alter and whichfactors determine this behaviour.
Second, the applicability of cognitive behavioural modelsto youth sexual behaviours may vary between developmentstages. For instance, applying cognitive theory to youngpeople with no or limited sexual experience may be difficult.This group may not yet have well-anchored ideas, and conse-quently their attitudes, norms, and beliefs about safe sexualbehaviour may be less clear and stable than for their adultcounterparts [70, 73]. Theories used in HIV interventionstargeted at youth could be strengthened by accounting forthe extent to which individual decision making is supportedby one’s age, gender, or other personal characteristics.
Third, these theories seem to ignore the fact that sexualintercourse takes place between two persons, within a rela-tionship. Sexual decisions do not depend on the individual,but also on the sexual partner and the type of relationship.Young people might have specific types of partners that mayinfluence sexual decision making. For example, relationshipswith someone who is much older are risky because itexposes the younger person (mostly girls) to a partner whois more likely to be sexually experienced and hence morelikely to be HIV-positive [75, 76]. Often, these age-disparaterelationships are transactional in nature, with money or giftsgiven in exchange for sexual intercourse [75, 77–81]. Also,young people in same-age relationships might have differenttypes of relationships than adults. They tend to be in whatis called by Bastard et al. [82] the “courtship-seduction”phases of relationships, in which the predominant concernsare to “present the best image, win trust, and avoid sourcesof conflict. These concerns take precedence over that ofprotecting oneself from the risk of AIDS.”
Fourth, while some interventions recognize the impor-tance of involving the community, only the SCT explicitlystresses the influence of contextual and structural factorson an individual’s behaviour. Even though the theory statesthat the social environment is an important determinantof the behaviour, many interventions based on SCT didnot attempt to include or influence environmental factors.Most interventions are limited to providing information andteaching skills. TRA/TPB implicitly includes this level bystating that personal attitudes, and norms are influencedby behavioural and normative beliefs in the society, whichis useful for tracking varying modes of sexual socialisation.However, this is an indirect effect of the environment onindividual behaviour, while still ignoring the broader struc-tural factors that shape sexual behaviour. Many recent studieshave demonstrated the contribution of structural factorsto young people’s vulnerability for HIV [75, 83–85]. Theseenvironmental aspects include both distal influences—suchas taboos on adolescent sexuality, norms and values, policies,poverty, education as well as more proximate influences.These include families’ opinions about adolescent relation-ships or teachers refusing to talk about condoms. Increasedefforts in future studies to account for structural factors ata theoretical level may improve the design of interventionsand assist in their evaluation, by understanding the possiblebarriers between motivation and actual behaviour change.
According to Gielen and Sleet [86], behavioural interven-tions can be subdivided into three categories, those aimed atintrapersonal factors (e.g., knowledge, skills, and intentions);interpersonal factors (including relational motivations andsocial desirability); community factors (e.g., culture, genderinequalities, poverty, and violence). We have already arguedthat the most common theories used in HIV interventionsdirected to youth do not adequately address interpersonalfactors, the failure to account for contextual factors furthercompounds the difficulty of evaluating interventions andunderstanding the possible barriers between motivation andactual behaviour change.
Finally, while cognitive behavioural theories of changemight be successful in altering cognitions and behavioural
AIDS Research and Treatment 15
intentions, they provide insufficient directives on translatingthis into actual behaviour change. Thus interventions couldbe regarded as successful in having altered motivationsand intentions, even though behavioural change may notresult. Similarly, interventions based on the HBM, mightincrease the perceived severity and susceptibility of a person,and relieve barriers to behaviour change, but in itself,might be insufficient to alter the sexual behaviour. Clearly,motivations or beliefs about behaviour change on a cognitiveor rational level need to be accompanied by a clear strategyfor introducing a new behaviour [87].
5. Conclusion
In the end, it boils down to two key questions: what deter-mines sexual behaviour of young people? And what frame-works are most useful for making sense of and impactingpositively on determinants of youth sexual behaviour? Rec-ognizing the complexity and heterogeneity of this particularbehaviour, theory can provide help in generalizing keydeterminants and making them operational. Theories aimto describe determinants and processes that account foror guide behaviour (change) through the rationalizationof individual decisions. This aids in understanding humanbehaviour, and when used appropriately, can provide a solidgrounding for program development and evaluation. Thestrength of theory is to generalize and simplify complex situ-ations. However, in the case of HIV prevention interventionsfor young people, the dominant theories might oversimplifysexual behaviour. While such cognitive behavioural modelscan explain the links between intention and behaviour,particularly at an intrapersonal level, they are less able toaccount for interpersonal and contextual factors related tothe complexity of sex, the experience of youth and disparitiesin social, cultural, and economic realities of youth in sub-Saharan Africa.
Acknowledgments
K. Michielsen acknowledges the Research Foundation Flan-ders (FWO) for financial support. The authors acknowledgeStanley Luchters and Petra De Koker for data extraction.
References
[1] UNAIDS: UNAIDS World AIDS Day Report 2011, How to getto zero: Faster. Smarter. Better, Geneva, Switzerland, 2011.
[2] T. J. Coates, L. Richter, and C. Caceres, “Behavioural strategiesto reduce HIV transmission: how to make them work better,”The Lancet, vol. 372, no. 9639, pp. 669–684, 2008.
[3] C. Marston and E. King, “Factors that shape young people’ssexual behaviour: a systematic review,” The Lancet, vol. 368,no. 9547, pp. 1581–1586, 2006.
[4] K. Michielsen, M. F. Chersich, S. Luchters, P. de Koker, R.van Rossem, and M. Temmerman, “Effectiveness of HIVprevention for youth in sub-Saharan Africa: systematic reviewand meta-analysis of randomized and nonrandomized trials,”AIDS, vol. 24, no. 8, pp. 1193–1202, 2010.
[5] S. M. N. Mavedzenge, A. M. Doyle, and D. A. Ross, “HIV pre-vention in young people in sub-Saharan Africa: a systematicreview,” Journal of Adolescent Health, vol. 49, no. 6, pp. 568–586, 2011.
[6] D. B. Kirby, B. A. Laris, and L. A. Rolleri, “Sex and HIVeducation programs: their impact on sexual behaviors ofyoung people throughout the world,” Journal of AdolescentHealth, vol. 40, no. 3, pp. 206–217, 2007.
[7] J. Green, “The role of theory in evidence-based healthpromotion practice,” Health Education Research, vol. 15, no.2, pp. 125–129, 2000.
[8] M. Fishbein, “The role of theory in HIV prevention,” AIDSCare, vol. 12, no. 3, pp. 273–278, 2000.
[9] Family Health International, Behavior Change—A Summary ofFour Major Theories, Family Health International, Arlington,Tex, USA, 2002.
[10] D. R. Rutter and L. Quine, Changing Health Behaviour.Intervention and Research with Social Cognitive Models, OpenUniversity Press, Philadelphia, Pa, USA, 2002.
[11] D. C. Des Jarlais, C. Lyles, N. Crepaz, and Group T, “Improvingthe reporting quality of nonrandomized evaluations of behav-ioral and public health interventions: the TREND statement,”American Journal of Public Health, vol. 94, no. 3, pp. 361–366,2004.
[12] J. B. Jemmott and L. S. Jemmott, “HIV risk reduction behav-ioral interventions with heterosexual adolescents,” AIDS, vol.14, supplement 2, pp. S40–S52, 2000.
[13] B. T. Johnson, M. P. Carey, K. L. Marsh, K. D. Levin, and L. A.J. Scott-Sheldon, “Interventions to reduce sexual risk for thehuman immunodeficiency virus in adolescents, 1985–2000:a research synthesis,” Archives of Pediatrics and AdolescentMedicine, vol. 157, no. 4, pp. 381–388, 2003.
[14] J. D. Fisher and W. A. Fisher, “Changing AIDS-risk behavior,”Psychological Bulletin, vol. 111, no. 3, pp. 455–474, 1992.
[15] J. A. Kelly and S. C. Kalichman, “Behavioral researchin HIV/AIDS primary and secondary prevention: recentadvances and future directions,” Journal of Consulting andClinical Psychology, vol. 70, no. 3, pp. 626–639, 2002.
[16] J. P. Moatti and Y. Souteyrand, “Editorial: HIV/AIDS socialand behavioural research: past advances and thoughts aboutthe future,” Social Science and Medicine, vol. 50, no. 11, pp.1519–1532, 2000.
[17] K. Glanz, B. K. Rimer, and F. M. Lewis, Health Behavior andHealth Education. Theory, Research and Practice, John Wiley &Sons, San Francisco, Calif, USA, 2002.
[18] K. Glanz and D. B. Bishop, “The role of behavioral sciencetheory in development and implementation of public healthinterventions,” Annual Review of Public Health, vol. 31, pp.399–418, 2010.
[19] J. E. Painter, C. P. C. Borba, M. Hynes, D. Mays, and K. Glanz,“The use of theory in health behavior research from 2000 to2005: a systematic review,” Annals of Behavioral Medicine, vol.35, no. 3, pp. 358–362, 2008.
[20] National Cancer Institute, Theory at Glance. A Guide for HealthPromotion Practice, National Cancer Institute, Washington,DC, USA, 2005.
[21] K. Glanz, F. M. Lewis, and B. K. Rimers, Health Behavior andHealth Education: Theory, Research, and Practice, Jossey-Bass,San Francisco, Calif, USA, 1990.
[22] A. Bandura, “Self-efficacy: toward a unifying theory ofbehavioral change,” Psychological Review, vol. 84, no. 2, pp.191–215, 1977.
[23] A. Bandura, Self-Efficacy: The Exercise of Control, Freeman,New York, NY, USA, 1997.
16 AIDS Research and Treatment
[24] I. M. Rosenstock, V. J. Strecher, and M. H. Becker, “Sociallearning theory and the Health Belief Model,” Health Educa-tion Quarterly, vol. 15, no. 2, pp. 175–183, 1988.
[25] I. Rosenstock, “Historical origins of the health belief model,”Health Education Quarterly, vol. 2, no. 4, pp. 328–335, 1974.
[26] M. H. Becker, “THe health belief model and personal healthbehavior,” Health Education Quarterly, vol. 2, no. 4, pp. 324–508, 1974.
[27] I. Ajzen and B. L. Driver, “Prediction of leisure participationfrom behavioral, normative, and control beliefs: an applica-tion of the theory of planned behavior,” Leisure Sciences, vol.13, no. 3, pp. 185–204, 1991.
[28] I. Ajzen and M. Fishbein, Understanding Attitudes and Pre-dicting Social Behavior, Prentice Hall, Upper Saddle River, NJ,USA, 1980.
[29] J. O. Prochaska and C. C. DiClemente, “Stages and processesof self-change of smoking: toward an integrative model ofchange,” Journal of Consulting and Clinical Psychology, vol. 51,no. 3, pp. 390–395, 1983.
[30] U. Bronfenbrenner, The Ecology of Human Development:Experiments by Nature and Design, Harvard University Press,Cambridge, Mass, USA, 1979.
[31] R. van Rossem and D. Meekers, “An evaluation of the effec-tiveness of targeted social marketing to promote adolescentand young adult reproductive health in Cameroon,” AIDSEducation and Prevention, vol. 12, no. 5, pp. 383–404, 2000.
[32] I. S. Speizer, B. O. Tambashe, and S. P. Tegang, “An evaluationof the “Entre Nous Jeunes” peer-educator program foradolescents in Cameroon,” Studies in Family Planning, vol. 32,no. 4, pp. 339–351, 2001.
[33] D. Meekers, S. Agha, and M. Klein, “The impact on condomuse of the “100% Jeune” social marketing program inCameroon,” Journal of Adolescent Health, vol. 36, no. 6, article530, 2005.
[34] K. I. Klepp, S. S. Ndeki, M. T. Leshabari, P. J. Hannan, andB. A. Lyimo, “AIDS education in Tanzania: promoting riskreduction among primary school children,” American Journalof Public Health, vol. 87, no. 12, pp. 1931–1936, 1997.
[35] K. I. Klepp, S. S. Ndeki, A. M. Seha et al., “AIDS educationfor primary school children in Tanzania: an evaluation study,”AIDS, vol. 8, no. 8, pp. 1157–1162, 1994.
[36] D. A. Shuey, B. B. Babishangire, S. Omiat, and H. Bagarukayo,“Increased sexual abstinence among in-school adolescents asa result of school health education in Soroti district, Uganda,”Health Education Research, vol. 14, no. 3, pp. 411–419, 1999.
[37] J. Kinsman, J. Nakiyingi, A. Kamali et al., “Evaluation of acomprehensive school-based aids education programme inrural Masaka, Uganda,” Health Education Research, vol. 16, no.1, pp. 85–100, 2001.
[38] A. S. Erulkar, L. I. A. Ettyang, C. Onoka, F. K. Nyagah,and A. Muyonga, “Behavior change evaluation of a culturallyconsistent reproductive health program for young Kenyans,”International Family Planning Perspectives, vol. 30, no. 2, pp.58–67, 2004.
[39] D. A. Ross, J. Changalucha, A. I. Obasi et al., “Biological andbehavioural impact of an adolescent sexual health interventionin Tanzania: a community-randomized trial,” AIDS, vol. 21,no. 14, pp. 1943–1955, 2007.
[40] A. M. Doyle, D. A. Ross, K. Maganja et al., “Long-termbiological and behavioural impact of an adolescent sexualhealth intervention in tanzania: follow-up survey of thecommunity-based mema kwa vijana trial,” PLoS Medicine, vol.7, no. 6, Article ID e1000287, 2010.
[41] E. Maticka-Tyndale, J. Wildish, and M. Gichuru, “Quasi-experimental evaluation of a national primary school HIVintervention in Kenya,” Evaluation and Program Planning, vol.30, no. 2, pp. 172–186, 2007.
[42] L. E. Rijsdijk, A. E. Bos, R. A. Ruiter, J. N. Leerlooijer,and B. de Haas, “The world starts with me: a multilevelevaluation of a comprehensive sex education programmetargeting adolescents in Uganda,” BMC Public Health, vol. 11,article 334, 2011.
[43] L. Kuhn, M. Sternberg, and C. Mathews, “Participation of theschool community in AIDS education: an evaluation of a highschool programme in South Africa,” AIDS Care, vol. 6, no. 2,pp. 161–171, 1994.
[44] B. Harvey, J. Stuart, and T. Swan, “Evaluation of a drama-in-education programme to increase AIDS awareness in SouthAfrican high schools: a randomized community interventiontrial,” International Journal of STD and AIDS, vol. 11, no. 2, pp.105–111, 2000.
[45] D. Meekers, The Effectiveness of Targeted Social Marketing toPromote Adolescent Reproductive Health: the Case of Soweto,South Africa, Population Services International WorkingPaper, no. 16, 1998.
[46] A. M. Fitzgerald, B. F. Stanton, N. Terreri et al., “Use ofWestern-based HIV risk-reduction interventions targetingadolescents in an african setting,” Journal of Adolescent Health,vol. 25, no. 1, pp. 52–61, 1999.
[47] B. F. Stanton, X. Li, J. Kahihuata et al., “Increased protected sexand abstinence among Namibian youth following a HIV risk-reduction intervention: a randomized, longitudinal study,”AIDS, vol. 12, no. 18, pp. 2473–2480, 1998.
[48] Y. M. Kim, A. Kols, R. Nyakauru, C. Marangwanda, andP. Chibatamoto, “Promoting sexual responsibility amongyoung people in Zimbabwe,” International Family PlanningPerspectives, vol. 27, no. 1, pp. 11–19, 2001.
[49] S. James, P. S. Reddy, R. A. C. Ruiter et al., “The effectsof a systematically developed photo-novella on knowledge,attitudes, communication and behavioural intentions withrespect to sexually transmitted infections among secondaryschool learners in South Africa,” Health Promotion Interna-tional, vol. 20, no. 2, pp. 157–165, 2005.
[50] M. J. Visser, “Life skills training as HIV/AIDS preven-tive strategy in secondary schools: evaluation of a large-scale implementation process,” Journal of Social Aspects ofHIV/AIDS Research Alliance, vol. 2, no. 1, pp. 203–216, 2005.
[51] C. Underwood, H. Hachonda, E. Serlemitsos, and U. Bharath-Kumar, “Reducing the risk of HIV transmission amongadolescents in Zambia: psychosocial and behavioral correlatesof viewing a risk-reduction media campaign,” Journal ofAdolescent Health, vol. 38, no. 1, pp. 55.e1–55.e13, 2006.
[52] R. Magnani, K. MacIntyre, A. M. Karim et al., “The impactof life skills education on adolescent sexual risk behaviors inKwaZulu-Natal, South Africa,” Journal of Adolescent Health,vol. 36, no. 4, pp. 289–304, 2005.
[53] S. Agha, “An evaluation of the effectiveness of a peer sexualhealth intervention among secondary-school students inZambia,” AIDS Education and Prevention, vol. 14, no. 4, pp.269–281, 2002.
[54] S. James, P. Reddy, R. A. C. Ruiter, A. McCauley, and B. van denBorne, “The impact of an HIV and AIDS life skills programon secondary school students in Kwazulu-Natal, South Africa,”AIDS Education and Prevention, vol. 18, no. 4, pp. 281–294,2006.
[55] A. Plautz, D. Meekers, and J. Neukom, The Impact ofthe Madagascar TOP R,seau Social Marketing Program on
AIDS Research and Treatment 17
Sexual Behavior and Use of Reproductive Health Services, PSIResearch Division Working Paper no. 57, 2003.
[56] A. P. Karnell, P. K. Cupp, R. S. Zimmerman, S. Feist-Price,and T. Bennie, “Efficacy of an American alcohol and HIVprevention curriculum adapted for use in South Africa: resultsof a pilot study in five township schools,” AIDS Education andPrevention, vol. 18, no. 4, pp. 295–310, 2006.
[57] M. J. Visser, “HIV/AIDS prevention through peer educationand support in secondary schools in South Africa,” Journal ofSocial Aspects of HIV/AIDS Research Alliance, vol. 4, no. 3, pp.678–694, 2007.
[58] R. Jewkes, M. Nduna, J. Levin et al., “Impact of stepping stoneson incidence of HIV and HSV-2 and sexual behaviour in ruralSouth Africa: cluster randomised controlled trial,” The BritishMedical Journal, vol. 337, no. 7666, pp. 391–395, 2008.
[59] R. Jewkes, M. Nduna, J. Levin et al., “A cluster randomized-controlled trial to determine the effectiveness of steppingstones in preventing HIV infections and promoting safersexual behaviour amongst youth in the rural Eastern Cape,South Africa: trial design, methods and baseline findings,”Tropical Medicine and International Health, vol. 11, no. 1, pp.3–16, 2006.
[60] M. K. Tibbits, E. A. Smith, L. L. Caldwell, and A. J. Flisher,“Impact of HealthWise South Africa on polydrug use andhigh-risk sexual behavior,” Health Education Research, vol. 26,no. 4, pp. 653–663, 2011.
[61] A. J. Mason-Jones, C. Mathews, and A. J. Flisher, “Can peereducation make a difference? Evaluation of a South Africanadolescent peer education program to promote sexual andreproductive health,” AIDS and Behavior, vol. 15, no. 8, pp.1605–1611, 2011.
[62] S. J. Baird, R. S. Garfein, C. T. McIntosh, and B. Ozler, “Effectof a cash transfer programme for schooling on prevalenceof HIV and herpes simplex type 2 in Malawi: a clusterrandomised trial,” The Lancet, vol. 379, no. 9823, pp. 1320–1329, 2012.
[63] S. M. Burnett, M. R. Weaver, P. N. Mody-Pan, L. A. ReynoldsThomas, and C. M. Mar, “Evaluation of an intervention toincrease human immunodeficiency virus testing among youthin Manzini, Swaziland: a randomized control trial,” Journal ofAdolescent Health, vol. 48, no. 5, pp. 507–513, 2011.
[64] W. R. Brieger, G. E. Delano, C. G. Lane, O. Oladepo, andK. A. Oyediran, “West African youth initiative: outcome of areproductive health education program,” Journal of AdolescentHealth, vol. 29, no. 6, pp. 436–446, 2001.
[65] I. O. Fawole, M. C. Asuzu, S. O. Oduntan, and W. R. Brieger,“A school-based AIDS education programme for secondaryschool students in Nigeria: a review of effectiveness,” HealthEducation Research, vol. 14, no. 5, pp. 675–683, 1999.
[66] F. E. Okonofua, P. Coplan, S. Collins et al., “Impact ofan intervention to improve treatment-seeking behavior andprevent sexually transmitted diseases among Nigerian youths,”International Journal of Infectious Diseases, vol. 7, no. 1, pp. 61–73, 2003.
[67] R. van Rossem and D. Meekers, An evaluation of theEffectiveness of Targeted Social Marketing to Promote Ado-lescent Reproductive Health in Guinea, PSI Research DivisionWorking Paper no. 23, 1999.
[68] K. A. Atwood, S. B. Kennedy, S. Shamblen et al., “Impactof school-based HIV prevention program in post-conflictLiberia,” AIDS Education and Prevention, vol. 24, no. 1, pp. 68–77, 2012.
[69] M. Visser, “Evaluation of the first AIDS kit, the AIDS andlifestyle education programme for teenagers,” South AfricanJournal of Psychology, vol. 26, no. 2, pp. 103–113, 1996.
[70] C. T. Pedlow and M. P. Carey, “HIV sexual risk-reductioninterventions for youth: a review and methodological critiqueof randomized controlled trials,” Behavior Modification, vol.27, no. 2, pp. 135–190, 2003.
[71] L. Robin, P. Dittus, D. Whitaker et al., “Behavioral interven-tions to reduce incidence of HIV, STD, and pregnancy amongadolescents: a decade in review,” Journal of Adolescent Health,vol. 34, no. 1, pp. 3–26, 2004.
[72] D. Kasprzyk, D. E. Montano, and M. Fishbein, “Applicationof an integrated behavioral model to predict condom use: aprospective study among high HIV risk groups,” Journal ofApplied Social Psychology, vol. 28, no. 17, pp. 1557–1583, 1998.
[73] J. de Wit, L. Breeman, and L. Woertman, “Hoe beredeneerd isseksueel gedrag van jongeren?” Tijdschrift voor Sociologie, vol.29, no. 3, pp. 125–131, 2005.
[74] L. H. Dawson, M. C. Shih, C. de Moor, and L. Shrier, “Reasonswhy adolescents and young adults have sex: associations withpsychological characteristics and sexual behavior,” Journal ofSex Research, vol. 45, no. 3, pp. 225–232, 2008.
[75] C. Underwood, J. Skinner, N. Osman, and H. Schwandt,“Structural determinants of adolescent girls’ vulnerability toHIV: views from community members in Botswana, Malawi,and Mozambique,” Social Science and Medicine, vol. 73, no. 2,pp. 343–350, 2011.
[76] S. Leclerc-Madlala, “Age-disparate and intergenerational sexin southern Africa: the dynamics of hypervulnerability,” AIDS,vol. 22, supplement 4, pp. S17–S25, 2008.
[77] K. Hawkins, N. Price, and F. Mussa, “Milking the cow: youngwomen’s construction of identity and risk in age-disparatetransactional sexual relationships in Maputo, Mozambique,”Global Public Health, vol. 4, no. 2, pp. 169–182, 2009.
[78] M. Hunter, “The materiality of everyday sex: thinking beyond‘prostitution’,” African Studies, vol. 61, no. 1, pp. 99–120, 2002.
[79] M. Silberschmidt and V. Rasch, “Adolescent girls, illegalabortions and “sugar-daddies” in Dar es Salaam: vulnerablevictims and active social agents,” Social Science and Medicine,vol. 52, no. 12, pp. 1815–1826, 2001.
[80] J. Wamoyi, A. Fenwick, M. Urassa, B. Zaba, and W. Stones,“‘Women’s bodies are shops’: beliefs about transactional exand implications for understanding gender power and HIVprevention in Tanzania,” Archives of Sexual Behavior, vol. 40,no. 1, pp. 5–15, 2011.
[81] J. Wamoyi, D. Wight, M. Plummer, G. H. Mshana, and D.Ross, “Transactional sex amongst young people in rural north-ern Tanzania: an ethnography of young women’s motivationsand negotiation,” Reproductive Health, vol. 7, no. 1, article 2,2010.
[82] B. Bastard, L. Cardia-Voneche, D. Peto, and L. van Campen-houdt, “Relationships between sexual partners and ways ofadapting to the risk of AIDS: landmarks for a relationship-oriented conceptual framework,” in Sexual Interactions andHIV Risk New Conceptual Perspectives in European Research, L.van Campenhoudt, M. Cohen, G. Guizzardi, and D. Hausser,Eds., Taylor & Francis, London, UK, 1997.
[83] E. Sumartojo, “Structural factors in HIV prevention: concepts,examples, and implications for research,” AIDS, vol. 14,supplement 1, pp. S3–S10, 2000.
[84] G. R. Gupta, J. O. Parkhurst, J. A. Ogden, P. Aggleton, andA. Mahal, “Structural approaches to HIV prevention,” TheLancet, vol. 372, no. 9640, pp. 764–775, 2008.
18 AIDS Research and Treatment
[85] A. Harrison, M. L. Newell, J. Imrie, and G. Hoddinott,“HIV prevention for South African youth: which interventionswork? A systematic review of current evidence,” BMC PublicHealth, vol. 10, article 102, 2010.
[86] A. C. Gielen and D. Sleet, “Application of behavior-changetheories and methods to injury prevention,” EpidemiologicReviews, vol. 25, pp. 65–76, 2003.
[87] A. Baban and C. Crciun, “Changing health-risk behaviors: areview of theory and evidence-based interventions in healthpsychology,” Journal of Cognitive and Behavioral Psychothera-pies, vol. 7, no. 1, pp. 45–67, 2007.