Culture in Nursing, Nursing Leadership, Writing and rhetorical, Introduction to recreational therapy, Inclusive recreation services and Psychopharmacology
Edero2103
Part5-2.pdf
www.thelancet.com/public-health Vol 8 February 2023 e141
Review
Lancet Public Health 2023; 8: e141–54
Published Online November 2, 2022 https://doi.org/10.1016/ S2468-2667(22)00252-3
Department of Primary Care, Population Sciences and Medical Education, University of Southampton, Southampton, UK (L Johnson MBBS); Collaborative Centre for Inclusion Health, Department of Epidemiology and Public Health (L Johnson, M Bradbury BMBS, G Sumner BA, L Wills MSc, Prof A C Hayward MD, A Story PhD, B Sultan FRCP, S A Luchenski FFPH) and Centre for Public Health Data Science, Institute of Health Informatics (Prof R W Aldridge PhD, N Pathak MBBS), University College London, London, UK; Centre for Academic Primary Care, Department of Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK (L C Potter MBChB, Prof G Feder MD); Oxford University Hospitals NHS Foundation Trust, Oxford, UK (H Beeching MBBCh); Department of Psychology, The American University of Paris, Paris, France (B Matos PGDip); Find & Treat, University College London Hospital, London, UK (A Story, B Sultan); Centre for Primary Care and Public Health, Queen Mary University, London, UK (K Worthing MPH); Guy’s and St Thomas’ NHS Foundation Trust, London, UK (N Pathak); Faculty of Public Health and Policy, London School of Hygiene & Tropical Medicine, London, UK (Prof L Platt PhD)
Correspondence to: Dr Luke Johnson, Department of Primary Care, Population Sciences and Medical Education, University of Southampton, Southampton SO17 1BJ, UK [email protected]
Interventions to improve health and the determinants of health among sex workers in high-income countries: a systematic review Luke Johnson, Lucy C Potter, Harriet Beeching, Molly Bradbury, Bella Matos, Grace Sumner, Lorna Wills, Kitty Worthing, Robert W Aldridge, Gene Feder, Andrew C Hayward, Neha Pathak, Lucy Platt, Al Story, Binta Sultan, Serena A Luchenski
Many sex worker populations face high morbidity and mortality, but data are scarce on interventions to improve their health. We did a systematic review of health and social interventions to improve the health and wider determinants of health among adult sex workers in high-income countries. We searched MEDLINE, Embase, PsycINFO, CINAHL, the Cochrane Library, Web of Science, EthOS, OpenGrey, and Social Care Online, as well as the Global Network of Sex Work Projects and the Sex Work Research Hub for studies published between Jan 1, 2005 and Dec 16, 2021 (PROSPERO CRD42019158674). Quantitative studies reporting disaggregated data for sex workers were included and no comparators were specified. We assessed rigour using the Quality Assessment Tool for Quantitative Studies. We summarised studies using vote counting and a narrative synthesis. 20 studies were included. Most reported findings exclusively for female sex workers (n=17) and street-based sex workers (n=11). Intervention components were divided into education and empowerment (n=14), drug treatment (n=4), sexual and reproductive health care (n=7), other health care (n=5), and welfare (n=5). Interventions affected a range of mental health, physical health, and health behaviour outcomes. Multicomponent interventions and interventions that were focused on education and empowerment were of benefit. Interventions that used peer design and peer delivery were effective. An outreach or drop-in component might be beneficial in some contexts. Sex workers who were new to working in an area faced greater challenges accessing services. Data were scarce for male, transgender, and indoor-based sex workers. Co-designed and co-delivered interventions that are either multicomponent or focus on education and empowerment are likely to be effective. Policy makers and health-care providers should improve access to services for all genders of sex workers and those new to an area. Future research should develop interventions for a greater diversity of sex worker populations and for wider health and social needs.
Introduction Sex work spans a wide range of activities, but is defined in this Review as the provision of sexual services in exchange for money or goods. Sex workers are a heterogeneous population—there is extensive variability in the structural, economic, social, and legal context in which they work and in their health and social needs.1
Stigma and the hidden—often transient—nature of sex work restrict the availability of accurate data.1 There are an estimated 1 million sex workers in the USA and 70 000 in the UK.2,3 There are large research gaps in the understanding of their health needs in different settings. Street-based sex workers are highly marginalised and face disproportionate health inequities and harms related to alcohol and drug use, and sometimes HIV and sexually transmitted infections (STIs), hepatitis B, and hepatitis C.4–8 Sex workers can encounter high rates of physical, verbal, and sexual violence from intimate partners, perpetrators posing as clients, and the police.7,9–13 They frequently have poor mental health, with increased rates of anxiety, depression, loneliness, post- traumatic stress disorder, self-harm, and suicide.5,7,14–16 There can be severe, complex social needs and structural determinants underlying these health issues, including homelessness or insecure housing, unemployment, adverse childhood experiences, gender and racial inequality, poverty, sex work criminalisation, and the setting of sex work.5,7,10,14,15,17–21 However, many sex workers
do not face this severe marginalisation or these adverse health outcomes and remain largely unrepre sented in academic literature. The legal context in which sex work occurs varies substantially between countries and can either exaggerate or mitigate these harms, with repressive policing practices and criminalisation worsening health outcomes.22
Many sex workers face large barriers to accessing health and social care.23 There are few specialist services for this community,24 and mainstream services are often unaware of sex working and not tailored to sex workers’ needs.17,25 Sex workers are often unaware of available services,26 and might fear legal implications from being identified as a sex worker.5 Additionally, past experiences of judgement and stigmatisation while using services could deter them from seeking care again.5,17
WHO guidelines state the importance of high-quality, integrated services to meet the health needs of sex workers.27 However, there is little published evidence on effective health and social care interventions for sex workers in high-income countries.28 There have been three previous systematic reviews, which have focused on psychological interventions for all sex workers,29 HIV and STI behaviour change interventions for female sex workers in the USA,10 and interventions for illicit drug use in street-working female sex workers.30 A com prehensive understanding of interventions tailored to sex workers is
e142 www.thelancet.com/public-health Vol 8 February 2023
Review
needed. This study aimed to systematically review the evidence of interventions used to improve health and the wider determinants of health for all sex worker populations living in high-income countries.
Methods We have adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.31 Our review protocol was registered with PROSPERO in November, 2019 (CRD42019158674). Our team included authors with lived experience, and authors who had worked with and continue to work with sex workers, to ensure the Review’s relevance and contextual insight in interpretation of the data.
Search strategy and selection criteria We conducted a systematic literature search in six data- bases (MEDLINE, Embase, PsycINFO, CINAHL, the Cochrane Library, and Web of Science). We used a combination of subject headings and keyword searching related to sex work and health interventions (appendix pp 2–3). Grey literature was also searched using EthOS, OpenGrey, and Social Care Online, the Global Network of Sex Work Projects, the Sex Work Research Hub, and by contacting academic experts and people with lived experience of sex working. Further studies were iden- tified through searching reference lists and citations of included studies. Studies were restricted to those published in English between Jan 1, 2005 and Dec 16, 2021.
Eligibility criteria Eligibility was defined using population, intervention, control, and outcomes criteria. The included population were current sex workers, which we defined as people who had exchanged sex for money, drugs, or other goods within the past 12 months. Trafficking and indirect sex work (in which there is no physical contact of any kind with the client) were not included. We included studies with sex workers aged 18 years and older in high-income countries, as defined by The World Bank.32 Any intervention with data specifically for sex workers was included. Studies with populations that did not entirely consist of sex workers, and for which—following contact with the authors—disaggregated sex-worker-specific data were not available, were excluded. If the majority of a study population was older than 18 years, and the data was specific to sex workers, the study was still included even if disaggregated data was not available following author contact. Any intervention that studied outcomes related to health or the wider determinants of health (eg, housing and welfare support) was included. Studies of sex work laws were excluded as these were investigated in a systematic review in 2018.22 Control groups were not specified a priori.
The review included all quantitative study designs to summarise study effectiveness: randomised con- trolled trials, quasi-experimental studies (ie, uncontrolled or controlled before-and-after studies), observational studies (ie, cohort, case-control, time series, and cross- sectional), and mixed-methods studies with a quantitative component.
Data extraction and quality assessment Titles and abstracts were single-screened for inclusion by one of two reviewers (MB or BM). Remaining articles were double screened at full-text review by two independent reviewers (LCP and BM). Discrepancies were resolved through discussion.
Data extraction was done by one of three reviewers (LJ, LW, or HB) with accuracy checked by a second reviewer (LJ, KW, LW, or HB). Discrepancies were resolved through discussion or decided by a third reviewer (LCP) when they could not be resolved. A spreadsheet was used to extract a standard set of data on study and population characteristics, design, intervention, control, outcome, and results.
Rigour was assessed using the Effective Public Health Practice Project’s Quality Assessment Tool for Quantitative Studies,33 chosen due to its comprehensive assessment of both observational and experimental studies, and showed reliability and validity.34 Criteria assessed include selection bias, study design, confounders, blinding, data collection methods, withdrawals, intervention integrity, and analyses.
Data synthesis Due to heterogeneity in method, interventions, and outcomes, we used descriptive vote counting35 alongside
See Online for appendix
Figure 1: Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram
18 611 records identified through database searching
8984 records after duplicates removed
9750 records screened
9550 records excluded on basis of title or abstract
200 full-text articles assessed for eligibility
20 studies included in narrative synthesis
123 additional records identified through other sources
180 full-text articles excluded 66 no intervention described 31 reviews without primary data 21 outcomes not described quantitatively 16 population not primarily adult sex workers 28 low-income or middle-income country population 13 no published full text available
5 study focused on criminalisation
www.thelancet.com/public-health Vol 8 February 2023 e143
Review Co
un tr
y St
ud ie
s ( n)
Se x
w or
ke r
po pu
la ti
on In
te rv
en ti
on se
tt in
g or
c on
te xt
In te
rv en
ti on
Co m
pa ris
on o
r co
nt ro
l O
ut co
m es
Q ua
lit y
as se
ss m
en t*
Ra nd
om is
ed co
nt ro
lle d
tr ia
ls †
W on
g et
a l
(2 01
9) 37
H on
g Ko
ng 12
7 Fe
m al
e se
x w or
ke rs
ag
ed 1
8 ye
ar s a
nd
ol de
r w ho
h ad
w
or ke
d in
H on
g Ko
ng in
th e
pa st
6
m on
th s a
nd w
ho
sp ok
e C an
to ne
se o
r Pu
to ng
hu a
Se rv
ice ce
nt re
Re sil
ie nc
e pr
og ra
m m
e co
ns ist
in g
of
six 1
-h se
ss io
ns d
es ig
ne d
to
im pr
ov e
se lf-
es te
em , s
el f-
effi ca
cy ,
an d
co pi
ng sk
ill s;
it us
ed
ps yc
ho ed
uc at
io n,
co gn
iti ve
- be
ha vi
ou ra
l s tr
at eg
ie s,
an d
so cia
l le
ar ni
ng p
rin cip
le s
St an
da rd
se rv
ice s p
ro vi
de d
by
N GO
s t ha
t a re
sp ec
ia lis
ed in
w
or ki
ng w
ith fe
m al
e se
x w
or ke
rs , w
hi ch
in clu
de d
ou tr
ea ch
v isi
ts , H
IV a
nd S
TI
te st
in g,
a nd
so cia
l a ct
iv iti
es
Po st
-in te
rv en
tio n
an d
at fo
llo w
-u p
af te
r 3
m on
th s,
th er
e w er
e sig
ni fic
an t
im pr
ov em
en ts
in re
sil ie
nc e,
se lf-
es te
em , a
nd
ps yc
ho lo
gi ca
l d ist
re ss
; t he
re w
as n
o sig
ni fic
an t d
iff er
en ce
in se
lf- effi
ca cy
o r s
tr es
s, or
S TI
te st
in g
be tw
ee n
gr ou
ps ; d
es pi
te n
o sig
ni fic
an t c
ha ng
e in
co nd
om u
se in
la st
se
xu al
tr an
sa ct
io n,
th er
e w as
a si
gn ifi
ca nt
in
cr ea
se in
re gu
la rit
y o f c
on do
m u
se in
th e
pa st
w ee
k in
th e
in te
rv en
tio n
gr ou
p
M od
er at
e
Su rra
tt a
nd
In cia
rd i
(2 01
0) 38
US A
80 6
St re
et -b
as ed
fe m
al e
se x w
or ke
rs w
ho u
se
dr ug
s ( he
ro in
, co
ca in
e, o
r b ot
h)
w ho
w er
e ag
ed
18 –5
0 ye
ar s
Un sp
ec ifi
ed Se
x w or
ke r-f
oc us
ed in
te rv
en tio
n— tw
o 60
-m in
se ss
io ns
d el
iv er
ed b
y pe
er s;
de ve
lo pe
d in
co lla
bo ra
tio n
w ith
se x w
or ke
rs a
nd fo
cu se
d on
ho
w to
re du
ce th
e H
IV ri
sk s t
ha t s
ex
w or
ke rs
fa ce
; a lso
in co
rp or
at ed
vo
lu nt
ar y t
es tin
g fo
r H IV
, he
pa tit
is B,
a nd
h ep
at iti
s C
N ID
A st
an da
rd in
te rv
en tio
n— tw
o 60
-m in
se ss
io ns
d el
iv er
ed
by p
ee rs
; p ro
vi de
d ge
ne ra
l ed
uc at
io n
on H
IV a
nd S
TI ri
sk
re du
ct io
n an
d vo
lu nt
ar y t
es t f
or
H IV
, h ep
at iti
s B , a
nd h
ep at
iti s C
Fo llo
w -u
p at
m on
th s 3
a nd
6 ; b
ot h
in te
rv en
tio ns
sh ow
ed si
gn ifi
ca nt
d ec
re as
es in
nu
m be
r o f d
ay s u
sin g
al co
ho l o
r d ru
gs ,
re du
ce d
oc ca
sio ns
o f s
ex w
or k w
hi le
d ru
nk o
r on
d ru
gs , r
ed uc
ed u
np ro
te ct
ed v
ag in
al a
nd
or al
se x,
a nd
re du
ce d
ph ys
ica l a
nd se
xu al
vi
ct im
isa tio
n; se
x w or
ke r-f
oc us
ed
in te
rv en
tio n
w as
b et
te r a
t r ed
uc in
g se
xu al
ab
us e
an d
un pr
ot ec
te d
or al
se x
at 6
m on
th s;
no d
iff er
en ce
b et
w ee
n in
te rv
en tio
ns in
d ru
g us
e re
du ct
io n,
se xu
al co
nt ac
t r ed
uc tio
n, o
r ph
ys ica
l a bu
se
M od
er at
e
Co ho
rt st
ud ie
s ( on
e gr
ou p,
p re
-in te
rv en
ti on
a nd
p os
t- in
te rv
en ti
on )†
Ci gr
an g
et a
l (2
02 0)
39
US A
91 Fe
m al
e se
x w or
ke rs
ag
ed 1
8 ye
ar s a
nd
ol de
r w ith
a t l
ea st
on
e pr
ev io
us a
rre st
fo
r a d
ru g
off en
ce
an d
w ho
w er
e cu
rre nt
ly in
ja il
M on
tg om
er y C
ou nt
y Ja
il, D
ay to
n, O
H , U
SA Tw
o br
ie f m
ot iv
at io
na l i
nt er
vi ew
s w
ith p
sy ch
ol og
ist s,
w hi
ch h
el pe
d pa
rt ici
pa nt
s i de
nt ify
a nd
p rio
rit ise
th
ei r t
op co
nc er
ns fo
r l ea
vi ng
ja il,
as
w el
l a s d
ev el
op p
la ns
to a
dd re
ss
th es
e co
nc er
ns
N A
N um
be r o
f a rre
st s s
ig ni
fic an
tly re
du ce
d fro
m
a m
ea n
of 3
·6 6
in th
e 12
m on
th s b
ef or
e in
ca rc
er at
io n
to a
m ea
n of
1 ·9
7 in
th e
12 m
on th
s f ol
lo w
in g
in ca
rc er
at io
n (p
<0 ·0
01 );
fin di
ng sa
fe a
nd a
ffo rd
ab le
h ou
sin g
an d
co nt
ro lli
ng u
se o
f d ru
gs w
er e t
he tw
o m
os t
co m
m on
ly ci
te d
co nc
er ns
M od
er at
e
Ra nd
om is
ed co
nt ro
lle d
tr ia
ls †
M ur
na n
et a
l (2
01 8)
40
US A
68 Fe
m al
e se
x w or
ke rs
se
ek in
g su
bs ta
nc e
us e t
re at
m en
t, w
ith
ch ild
re n
ag ed
8–
16 ye
ar s i
n th
ei r
ca re
O ne
in te
rv en
tio n
gr ou
p in
th e
pa rt
ici pa
nt ’s
ho m
e,
on e
in te
rv en
tio n
gr ou
p in
o ffi
ce se
tt in
g
Tw o
in te
rv en
tio n
gr ou
ps d
el iv
er ed
ov
er 6
m on
th s:
12 se
ss io
ns o
f E BF
T in
a h
om e o
r o ffi
ce se
tt in
g; E
BF T
ta rg
et s d
ys fu
nc tio
na l f
am ily
in
te ra
ct io
ns th
at a
re th
ou gh
t t o
co nt
rib ut
e t o
th e d
ev el
op m
en t,
m ai
nt en
an ce
, a nd
re so
lu tio
n of
pr
ob le
m s
W om
en ’s
H ea
lth
Ed uc
at io
n— 12
se ss
io ns
o f
in di
vi du
al p
sy ch
oe du
ca tio
na l
tr ea
tm en
t o ve
r 6 m
on th
s, fo
cu sin
g on
to pi
cs su
ch a
s fe
m al
e an
at om
y, h
um an
se xu
al
be ha
vi ou
r, pr
eg na
nc y
an d
ch ild
bi rt
h, a
nd S
TI s a
nd se
xu al
ris
k be
ha vi
ou r
W om
en w
ho re
ce iv
ed a
n EB
FT in
te rv
en tio
n re
po rt
ed a
g re
at er
re du
ct io
n in
d ru
g us
e an
d de
pr es
siv e
sy m
pt om
s, an
d gr
ea te
r im
pr ov
em en
t i n
m ot
he r–
ch ild
in te
ra ct
io ns
co
m pa
re d
w ith
th e
co nt
ro l g
ro up
W ea
k
Su rra
tt e
t a l
(2 01
4) 41
US A
56 2
Af ric
an A
m er
ica n
st re
et -b
as ed
fe m
al e
se x w
or ke
rs w
ho u
se
dr ug
s ( co
ca in
e,
cr ac
k co
ca in
e,
he ro
in , o
r a ll t
hr ee
) ag
ed 1
8– 50
ye ar
s
An o
ffi ce
n ea
r t w
o of
th
e m
aj or
se x w
or k
ar ea
s i n
M ia
m i,
FL ,
US A
Fi ve
se ss
io ns
o f a
st re
ng th
s- ba
se d,
pr
of es
sio na
l–p ee
r i nt
er ve
nt io
n de
liv er
ed o
ve r 8
w ee
ks ,
pr of
es sio
na ls
w er
e ca
se m
an ag
er s;
pe er
s w er
e re
co ve
rin g
ad di
ct s o
r fo
rm er
se x w
or ke
rs ; p
ee rs
a lso
re
m ai
ne d
in co
nt ac
t w ith
th e
pa rt
ici pa
nt s t
hr ou
gh ou
t t he
6-
m on
th st
ud y,
p ro
vi di
ng o
ng oi
ng
su pp
or t f
or se
rv ice
li nk
ag e
Fi ve
se ss
io ns
o f a
st re
ng th
s- ba
se d,
p ro
fe ss
io na
l-o nl
y in
te rv
en tio
n de
liv er
ed o
ve r
8 w
ee ks
b y
a ca
se m
an ag
er
Th e
in clu
sio n
of a
p ee
r f ac
ili ta
to r p
ro vi
de d
no
ad di
tio na
l b en
efi t i
n re
du cin
g H
IV ri
sk
be ha
vi ou
rs o
r i nc
re as
in g
he al
th -s
er vi
ce u
se ;
ho w
ev er
, t he
re w
er e
sig ni
fic an
t c ha
ng es
in
bo th
in te
rv en
tio n
an d
co nt
ro l g
ro up
s i n
so m
e H
IV ri
sk b
eh av
io ur
s ( cr
ac k u
se , a
nd n
um be
r o f
se xu
al p
ar tn
er s)
a nd
in cr
ea se
d us
e o f r
eg ul
ar
m ed
ica l c
ar e
an d
ad di
ct io
n se
lf- he
lp g
ro up
s; th
os e w
ho w
er e
H IV
p os
iti ve
a t b
as el
in e
ha d
a gr
ea te
r r ed
uc tio
n in
H IV
ri sk
b eh
av io
ur (e
g,
cr ac
k co
ca in
e u se
, n um
be r o
f s ex
ua l p
ar tn
er s,
an d
fre qu
en cy
o f u
np ro
te ct
ed v
ag in
al se
x) in
bo
th g
ro up
s
W ea
k
(T ab
le 1
co nt
in ue
s o n
ne xt
p ag
e)
e144 www.thelancet.com/public-health Vol 8 February 2023
Review
Co un
tr y
St ud
ie s (
n) Se
x w
or ke
r po
pu la
ti on
In te
rv en
ti on
se tt
in g
or c
on te
xt In
te rv
en ti
on Co
m pa
ris on
o r
co nt
ro l
O ut
co m
es Q
ua lit
y as
se ss
m en
t*
(C on
tin ue
d fro
m p
re vi
ou s p
ag e)
Co ho
rt a
na ly
ti ca
l s tu
di es
(t w
o gr
ou ps
, p re
-in te
rv en
ti on
a nd
p os
t- in
te rv
en ti
on )†
Ki m
e t a
l (2
01 5)
26
Ca na
da 54
5 (d
isa gg
re ga
te d
sa m
pl e
siz e
af te
r e xc
lu sio
n of
p eo
pl e
ag ed
14
–1 8
ye ar
s)
St re
et -b
as ed
fe m
al e
se x w
or ke
rs a
ge d
≥1 4
ye ar
s ( in
clu di
ng
tr an
sg en
de r
w om
en )
W IS
H d
ro p-
in ce
nt re
W om
en -o
nl y,
se x-
w or
k- sp
ec ifi
c dr
op -in
se rv
ice fo
r s tr
ee t-
ba se
d se
x w
or ke
rs , w
hi ch
p ro
vi de
d fo
od a
nd
dr in
k, sh
ow er
s, clo
th in
g, re
po rt
s o f
re ce
nt cl
ie nt
v io
le nc
e (n
am ed
b ad
da
te sh
ee ts
), H
IV p
re ve
nt io
n re
so ur
ce s (
eg , c
on do
m s a
nd
sy rin
ge s)
, a nd
re fe
rra ls
to so
cia l a
nd
he al
th se
rv ice
s; al
so , p
ee r e
du ca
tio n
an d
su pp
or t p
ro gr
am m
es ,
an d
ou tr
ea ch
n ur
sin g
ca re
so
m et
im es
o ffe
re d
Fe m
al e
se x w
or ke
rs w
ho d
id n
ot
us e t
he W
IS H
d ro
p- in
se rv
ice Se
rv ice
u se
w as
a ss
oc ia
te d
w ith
o ld
er a
ge ,
In di
ge no
us P
eo pl
es a
nc es
tr y,
in je
ct in
g dr
ug
us e,
e xc
ha ng
e o f s
ex fo
r d ru
gs , a
nd a
cc es
sin g
se xu
al a
nd re
pr od
uc tiv
e he
al th
se rv
ice s;
60 %
v isi
te d
W IS
H d
ur in
g th
e 3
ye ar
s o f f
ol lo
w -
up ; t
he se
rv ice
s m os
t f re
qu en
tly a
cc es
se d
w er
e fo
od , m
ak e-
up , c
lo th
in g,
a nd
p rim
ar y
nu rs
e ca
re
W ea
k
De er
in g
et a
l (2
01 1)
42
Ca na
da 24
2 Ci
sg en
de r a
nd
tr an
sg en
de r w
om en
st
re et
-b as
ed se
x w
or ke
rs ag
ed
≥1 4
ye ar
s w ho
sm
ok ed
(n ot
in
clu di
ng m
ar iju
an a)
or
in je
ct ed
ill ici
t dr
ug s i
n th
e l as
t m
on th
; a ut
ho rs
st
at ed
o nl
y a sm
al l
nu m
be r o
f pa
rt ici
pa nt
s l ik
el y t
o be
ag ed
< 18
ye ar
s
Pe er
-le d
m ob
ile
ou tr
ea ch
p ro
gr am
m e
(t he
M AP
, o r t
he M
AP
va n)
A ni
gh tly
o ut
re ac
h se
rv ice
st aff
ed
by a
d riv
er , a
su pp
or t w
or ke
r, an
d a
pe er
-s up
po rt
w or
ke r;
it pr
ov id
es a
sa
fe sp
ac e
fo r w
om en
to re
st , e
at ,
an d
dr in
k, a
nd o
ut re
ac h
st aff
di
st rib
ut e
re po
rt s o
f r ec
en t c
lie nt
vi
ol en
ce (n
am ed
b ad
d at
e re
po rt
s) ;
H IV
p re
ve nt
io n
re so
ur ce
s di
st rib
ut ed
(e g,
co nd
om s,
lu br
ica nt
, an
d sy
rin ge
s) a
nd re
fe rra
ls m
ad e t
o he
al th
a nd
so cia
l s up
po rt
, a nd
d ru
g tr
ea tm
en t s
er vi
ce s
W om
en w
ith in
th e
co ho
rt w
ho
di d
no t a
cc es
s t he
M AP
v an
du
rin g
th e
18 -m
on th
fo llo
w -u
p
O ve
r t he
1 8-
m on
th st
ud y
pe rio
d, 4
2% o
f re
po rt
s f ro
m se
x w or
ke rs
st at
ed u
se o
f t he
va
n; th
os e u
sin g
th e
va n
w er
e m
or e
lik el
y t o
be w
or ki
ng w
ith te
n or
m or
e cli
en ts
p er
w ee
k,
w or
ki ng
in is
ol at
ed p
ub lic
sp ac
es , a
nd u
sin g
th e W
IS H
d ro
p- in
ce nt
re se
rv ice
(l in
ke d
to th
e va
n) ; t
ho se
w ho
u se
d th
e pr
og ra
m m
e w er
e m
or e
lik el
y t o
ha ve
u se
d in
pa tie
nt a
dd ict
io n
tr ea
tm en
t s er
vi ce
s i n
th e
pa st
6 m
on th
s; th
er e w
as n
o sig
ni fic
an t r
el at
io ns
hi p
be tw
ee n
us e o
f t he
v an
a nd
a cc
es sin
g ou
tp at
ie nt
d ru
g tr
ea tm
en t;
yo ut
h ag
ed ≤
24 ye
ar s w
er e
sig ni
fic an
tly le
ss li
ke ly
to a
cc es
s t he
v an
W ea
k
Bu rn
et te
et
a l (
20 09
)14
US A
53 3
W om
en e
xc ha
ng in
g se
x fo
r m on
ey o
r dr
ug s i
n th
e pa
st
12 m
on th
s
O ut
pa tie
nt o
r re
sid en
tia l t
re at
m en
t pr
og ra
m m
e
Va rio
us d
ru g
tr ea
tm en
t pr
og ra
m m
es a
cr os
s 7 1
fa cil
iti es
in
th e U
SA (m
et ha
do ne
a nd
n on
- m
et ha
do ne
; r es
id en
tia l a
nd
ou tp
at ie
nt );
an cil
la ry
se rv
ice s a
lso
off er
ed —
m ed
ica l s
er vi
ce s,
m en
ta l
he al
th se
rv ice
s, an
d ps
yc ho
so cia
l se
rv ice
s
W om
en n
ot re
po rt
in g
ex ch
an gi
ng se
x fo
r m on
ey o
r dr
ug s a
t b as
el in
e
12 m
on th
s a ft
er d
isc ha
rg e
fro m
tr ea
tm en
t, se
x w or
ke rs
h ad
a lo
w er
li ke
lih oo
d of
ab
st in
en ce
fr om
d ru
gs a
nd a
lco ho
l t ha
n no
n- se
x w or
ke rs
u sin
g th
e se
rv ice
; o f a
ll w om
en in
th
e pr
og ra
m m
e, th
er e w
as a
re du
ct io
n of
30
% in
th os
e re
po rt
in g
se x w
or k,
a nd
a
re du
ct io
n of
se x w
or k t
ra ns
ac tio
ns in
th os
e w
ho co
nt in
ue d
se x w
or k;
w om
en w
ith a
lo
ng er
d ur
at io
n of
tr ea
tm en
t a nd
th os
e w ho
re
ce iv
ed m
or e
m en
ta l h
ea lth
a nd
ps
yc ho
lo gi
ca l s
er vi
ce s w
er e
m or
e lik
el y t
o ha
ve st
op pe
d se
x w or
k; ce
ss at
io n
of se
x w or
k w
as p
re di
ct iv
e o f l
es s f
re qu
en t d
ru g
an d
al co
ho l u
se a
nd fe
w er
m en
ta l h
ea lth
sy
m pt
om s a
t f ol
lo w
-u p
W ea
k
(T ab
le 1
co nt
in ue
s o n
ne xt
p ag
e)
www.thelancet.com/public-health Vol 8 February 2023 e145
Review Co
un tr
y St
ud ie
s ( n)
Se x
w or
ke r
po pu
la ti
on In
te rv
en ti
on se
tt in
g or
c on
te xt
In te
rv en
ti on
Co m
pa ris
on o
r co
nt ro
l O
ut co
m es
Q ua
lit y
as se
ss m
en t*
(C on
tin ue
d fro
m p
re vi
ou s p
ag e)
Co ho
rt st
ud ie
s ( on
e gr
ou p,
p re
-in te
rv en
ti on
a nd
p os
t- in
te rv
en ti
on )†
Pa rk
e t a
l (2
02 0)
43
US A
10 3
Fe m
al e
se x w
or ke
rs
ag ed
≥ 18
ye ar
s w ho
w
er e u
sin g
op io
id s
ob ta
in ed
il le
ga lly
(ie
, h er
oi n,
fe nt
an yl
, pr
es cr
ip tio
n op
io id
pi
lls p
ur ch
as ed
o n
th e
st re
et , o
r a ll
th re
e)
Va rio
us se
tt in
gs (e
g,
m ob
ile v
an , s
tu dy
offi
ce , f
as t f
oo d
re st
au ra
nt s,
an d
ho m
e vi
sit s)
Tr ai
ni ng
a nd
p ro
vi sio
n of
fi ve
fe
nt an
yl te
st st
rip s t
o te
st d
ru g
sa m
pl es
fo r f
en ta
ny l a
nd re
la te
d an
al og
ue s t
ha t h
av e
in cr
ea se
d ris
k of
o ve
rd os
e an
d de
at h;
a lso
pr
ov id
ed b
rie f o
ve rd
os e
ris k
as se
ss m
en t,
ta ilo
re d
ha rm
re
du ct
io n
ad vi
ce , a
nd n
al ox
on e
N A
1 m
on th
la te
r, th
er e w
er e
sig ni
fic an
t re
du ct
io ns
in il
lic it
op io
id u
se , i
nj ec
tio n
dr ug
us
e, b
en zo
di az
ep in
e u se
, a nd
so lit
ar y d
ru g
us e;
8 4%
o f w
om en
u se
d th
e st
rip s;
of th
e 48
w ho
h ad
a p
os iti
ve te
st re
su lt,
6 3%
st ill
us
ed th
ei r d
ru g
as o
rig in
al ly
in te
nd ed
; o th
er s
us ed
a ra
ng e o
f h ar
m re
du ct
io n
be ha
vi ou
rs
af te
r p os
iti ve
re su
lts
W ea
k
De ck
er e
t a l
(2 01
7) 44
US A
60 St
re et
-b as
ed a
nd
ve nu
e- ba
se d
fe m
al e
se x w
or ke
rs
M ob
ile v
an th
at w
en t
to b
ot h
a st
re et
lo
ca tio
n an
d a
ve nu
e kn
ow n
to h
av e
a lo
t o f
se x t
ra de
a ct
iv ity
IN SP
IR E—
a sin
gl e
br ie
f s em
i- st
ru ct
ur ed
d isc
us sio
n w
ith a
n ou
tr ea
ch w
or ke
r a bo
ut im
pr ov
in g
sa fe
ty a
nd re
du cin
g H
IV ri
sk , a
s w el
l as
h ow
to a
cc es
s v io
le nc
e su
pp or
t se
rv ice
s; su
pp le
m en
te d
w ith
w al
le t-
siz ed
ca rd
su m
m ar
isi ng
in
fo rm
at io
n; d
ev el
op ed
in
co lla
bo ra
tio n
w ith
fe m
al e
se x
w or
ke rs
N A
12 w
ee ks
a ft
er co
m pl
et in
g th
e in
te rv
en tio
n,
pa rt
ici pa
nt s u
se d
m or
e sa
fe ty
b eh
av io
ur s,
m ad
e m
or e u
se o
f s ex
ua l v
io le
nc e
an d
tr affi
ck in
g su
pp or
t p ro
gr am
m es
, a nd
w er
e m
or e
aw ar
e o f h
ow to
re po
rt v
io le
nc e t
o th
e po
lic e t
ha n
th ey
w er
e be
fo re
; t he
y w er
e m
or e
lik el
y t o
en ga
ge in
co nd
om n
eg ot
ia tio
n an
d le
ss li
ke ly
to h
av e
se x w
ith cl
ie nt
s w hi
le u
sin g
al co
ho l o
r d ru
gs th
an b
ef or
e; th
er e w
as n
o ch
an ge
in P
TS D
or d
ep re
ss iv
e sy
m pt
om s
W ea
k
Li tc
hfi el
d et
a l (
20 10
)45
UK 34
Ad ul
t f em
al e
se x
w or
ke rs
u sin
g he
ro in
Ge ne
ra l p
ra ct
io ne
r-l ed
pr
im ar
y ca
re d
ru g
tr ea
tm en
t s er
vi ce
th at
in
co rp
or at
ed a
sp ec
ifi c
cli ni
c t ar
ge te
d at
fe
m al
e se
x w or
ke rs
Ta rg
et ed
d ru
g tr
ea tm
en t
pr og
ra m
m e—
pr es
cr ib
ed o
pi oi
d su
bs tit
ut io
n (t
yp ica
lly m
et ha
do ne
); se
xu al
h ea
lth in
te rv
en tio
ns a
nd
ad vi
ce , a
nd k
ey -w
or ki
ng a
nd
ps yc
ho so
cia l s
up po
rt w
er e
al so
av
ai la
bl e
N A
Af te
r 1 ye
ar , o
nl y
33 %
o f p
ar tic
ip an
ts w
er e
st ill
se
x w or
ke rs
, q ua
lit y o
f l ife
h ad
im pr
ov ed
, a nd
he
ro in
u se
h ad
re du
ce d
(in u
rin e
sa m
pl es
, 87
% p
os iti
ve fo
r h er
oi n
at b
as el
in e,
7 2%
a t
1 y ea
r)
W ea
k
W ar
d an
d Ro
e- Se
po w
itz
(2 00
9) 9
US A
29 o
ve ra
ll: 1
1 in
re
sid en
tia l
pr og
ra m
m e a
nd
18 in
p ris
on
(in ca
rc er
at ed
fo
r n on
-s ex
w
or ki
ng cr
im es
)
Fe m
al e
se x w
or ke
rs Tw
o se
tt in
gs —
a co
m -
m un
ity -b
as ed
re
sid en
tia l p
ro gr
am m
e an
d a
m od
er at
e- se
cu rit
y pr
iso n
Th e
Es ub
a pr
og ra
m m
e— a
ps yc
ho -
ed uc
at io
na l t
he ra
py g
ro up
de
sig ne
d to
in cr
ea se
a w
ar en
es s o
f ab
us e
an d
vi ol
en ce
w hi
le te
ac hi
ng
an ge
r m an
ag em
en t a
nd
co m
m un
ica tio
n sk
ill s,
an d
de ve
lo pi
ng so
cia l s
up po
rt ; o
ne
se ss
io n
pe r w
ee k
fo r 1
2 w
ee ks
, e ac
h la
st in
g 2
h; se
ss io
ns fa
cil ita
te d
by a
do
ct or
al st
ud en
t a nd
cl in
ica l s
oc ia
l w
or ke
r
N A
Re du
ct io
n in
tr au
m a
sc or
es in
b ot
h gr
ou ps
, w
ith th
e pr
iso n
gr ou
p ha
vi ng
g re
at er
ch an
ge
(b ut
th ey
a lso
h ad
h ig
he r b
as el
in e t
ra um
a sc
or es
th an
th e
re sid
en tia
l p ro
gr am
m e)
W ea
k
Bo w
se r e
t a l
(2 00
8) 46
US A
18 9
Fe m
al e
se x w
or ke
rs
w ho
u se
d ru
gs Pr
og ra
m m
e ho
us e
Da y
pr og
ra m
m e
la st
in g
12 m
on th
s, pr
ov id
in g
m ea
ls, H
IV ri
sk -r
ed uc
tio n
ed uc
at io
n, a
nd o
ne -o
n- on
e ps
yc ho
lo gi
ca l c
ou ns
el lin
g
N A
By th
e en
d of
th e
in te
rv en
tio n,
th er
e w as
a
sig ni
fic an
t r ed
uc tio
n in
p ol
yd ru
g us
e w ith
al
co ho
l; al
so si
gn ifi
ca nt
im pr
ov em
en t i
n ho
us in
g se
cu rit
y an
d a
sig ni
fic an
t r ed
uc tio
n in
th
e nu
m be
r o f n
ig ht
s p er
m on
th sp
en t i
n ja
il;
no si
gn ifi
ca nt
ch an
ge in
e m
pl oy
m en
t
W ea
k
Sh er
m an
et
a l (
20 06
)47
US A
50 Fe
m al
e se
x w or
ke rs
ag
ed 1
8– 45
ye ar
s w
ho h
ad u
se d
he ro
in o
r c oc
ai ne
a t
le as
t o nc
e w ee
kl y
in
th e
pa st
m on
th
O ffi
ce in
th e t
ar ge
t ne
ig hb
ou rh
oo d
(in
Ba lti
m or
e, M
D, U
SA ;
ho w
ev er
, e xa
ct
ne ig
hb ou
rh oo
d un
sp ec
ifi ed
)
Si x
2- h
se ss
io ns
th at
ta ug
ht H
IV ri
sk
re du
ct io
n an
d th
e m
ak in
g,
m ar
ke tin
g, a
nd se
lli ng
o f j
ew el
le ry
; th
er e w
er e o
pp or
tu ni
tie s t
o se
ll je
w el
le ry
th at
w as
m ad
e as
w el
l; se
ss io
ns to
ok p
la ce
tw ice
p er
w ee
k ov
er a
3 -w
ee k
pe rio
d
N A
3 m
on th
s a ft
er co
m pl
et in
g th
e in
te rv
en tio
n,
th er
e w er
e sig
ni fic
an t r
ed uc
tio ns
in th
e nu
m be
r o f t
ra ns
ac tio
na l s
ex p
ar tn
er s,
al l s
ex
pa rt
ne rs
, a nd
d ai
ly in
je ct
io n
an d
no n-
in je
ct io
n dr
ug u
se ; w
om en
w ho
e ar
ne d
m or
e m
on ey
th ro
ug h
je w
el le
ry sa
le s h
ad a
sig
ni fic
an tly
re du
ce d
nu m
be r o
f s ex
tr ad
e pa
rt ne
rs a
t f ol
lo w
-u p
W ea
k
(T ab
le 1
co nt
in ue
s o n
ne xt
p ag
e)
e146 www.thelancet.com/public-health Vol 8 February 2023
Review
Co un
tr y
St ud
ie s (
n) Se
x w
or ke
r po
pu la
ti on
In te
rv en
ti on
se tt
in g
or c
on te
xt In
te rv
en ti
on Co
m pa
ris on
o r
co nt
ro l
O ut
co m
es Q
ua lit
y as
se ss
m en
t*
(C on
tin ue
d fro
m p
re vi
ou s p
ag e)
Cr os
s- se
ct io
na l s
tu di
es †
St ew
ar t e
t a l
(2 02
0) 48
US A
50 Fi
rs t 5
0 w
om en
at
te nd
in g
th e
cli ni
c; 31
(6 2%
) o f
50 re
po rt
ed
tr an
sa ct
io na
l s ex
fo r
fo od
, s he
lte r,
or d
ru gs
; h ow
ev er
, au
th or
s w er
e co
nt ac
te d
an d
st ro
ng ly
su sp
ec t
th at
a ll
cli en
ts
en ga
ge d
in
tr an
sa ct
io na
l s ex
O ut
re ac
h cli
ni c a
t a
dr op
-in ce
nt re
fo r
pe op
le e
xp er
ie nc
in g
ho m
el es
sn es
s
3- h
w ee
kl y
cli ni
c i nv
ol vi
ng a
n in
fe ct
io us
d ise
as e
ph ys
ici an
, a
nu rs
e, a
nd a
m ed
ica l s
oc ia
l w
or ke
r; th
e fu
ll- tim
e m
ed ica
l s oc
ia l
w or
ke r w
as p
re se
nt o
n sit
e d ur
in g
ea ch
cl in
ic se
ss io
n an
d on
n on
-c lin
ic da
ys fo
r c ar
e co
or di
na tio
n; cl
in ic
pr ov
id es
p rim
ar y
m ed
ica l c
ar e
an d
ha rm
re du
ct io
n in
te rv
en tio
ns fo
r dr
ug tr
ea tm
en t,
in clu
di ng
bu
pr en
or ph
in e–
na lo
xo ne
, f am
ily
pl an
ni ng
, a nd
tr ea
tm en
t o f S
TI ,
Pr EP
, a nd
H IV
ca re
N A
Pr im
ar y
re as
on s f
or se
ek in
g ca
re w
er e
sk in
an
d so
ft ti
ss ue
in fe
ct io
n, S
TI a
nd H
IV
sc re
en in
g, a
nd u
rin ar
y t ra
ct in
fe ct
io n;
fo
ur (1
0% ) o
f 3 9
te st
ed w
om en
h ad
un
pl an
ne d
pr eg
na nc
ie s;
fo ur
(1 0%
) o f
42 w
er e
po sit
iv e
fo r H
IV , o
f w hi
ch tw
o w
er e
ne w
d ia
gn os
es ; o
pi oi
d de
te ct
ed in
u rin
e o f
31 w
om en
— ni
ne in
iti at
ed b
up re
no rp
hi ne
– na
lo xo
ne a
nd th
re e
al re
ad y
co nn
ec te
d to
a
tr ea
tm en
t p ro
gr am
m e;
1 1
(4 8%
) o f 2
3 te
st ed
po
sit iv
e fo
r T ric
ho m
on as
va gi
na lis
, fiv
e (1
8% ) o
f 2 8
fo r g
on or
rh oe
a, fi
ve (1
8% ) o
f 28
fo r c
hl am
yd ia
, n on
e o f 1
3 fo
r s yp
hi lis
, an
d 15
(3 9%
) o f 3
8 fo
r h ep
at iti
s C ; H
IV P
rE P
pr es
cr ib
ed to
1 7
w om
en
W ea
k
Ba ar
s e t a
l (2
00 9)
49
N et
he rla
nd s
25 9
Fe m
al e
se x w
or ke
rs
in b
ro th
el s,
clu bs
, er
ot ic
m as
sa ge
sa
lo ns
, e ro
tic b
ar s,
w in
do w
se x w
or k,
an
d se
x w or
k zo
ne s
O ut
re ac
h an
d cli
ni c
Fr ee
, t ar
ge te
d na
tio na
l h ep
at iti
s B
va cc
in at
io n
pr og
ra m
m e
fo r s
ex
w or
ke rs
(i nc
lu de
d m
en a
nd
w om
en —
bu t t
he st
ud y o
nl y
lo ok
ed
at w
om en
a nd
o th
er g
ro up
s a t h
ig h
ris k)
; a lso
sc re
en ed
fo r h
ep at
iti s B
a t
tim e o
f v ac
cin at
io n;
co m
m un
ity
he al
th st
aff p
er io
di ca
lly v
isi te
d va
rio us
se x w
or ke
r l oc
at io
ns (e
g,
st re
et s a
nd b
ro th
el s)
a nd
o ffe
re d
va cc
in es
th er
e
N A
20 5
(7 9%
) p ar
tic ip
an ts
w er
e aw
ar e t
ha t t
he y
co ul
d ob
ta in
fr ee
h ep
at iti
s B v
ac cin
at io
n;
va cc
in at
io n
up ta
ke —
of a
t l ea
st o
ne d
os e—
w as
63
% (8
2% o
f w ho
re ce
iv ed
it th
ro ug
h th
is pr
og ra
m m
e) ; t
ho se
w ho
re ce
iv ed
th e
va cc
in at
io n
ha d
w or
ke d
in a
n ar
ea fo
r l on
ge r
an d
w er
e m
or e
lik el
y t o
w or
k be
hi nd
w in
do w
s th
an th
os e w
ho d
id n
ot re
ce iv
e t he
va
cc in
at io
n; th
e m
os t i
m po
rt an
t r ea
so n
fo r
no n-
pa rt
ici pa
tio n
w as
la ck
o f t
im e;
o f t
ho se
w
ho st
ar te
d th
e va
cc in
e co
ur se
, 74
(7 9%
) o f 9
4 pa
rt ici
pa nt
s r ec
ei ve
d th
re e o
r m
or e
va cc
in at
io ns
, 1 5
(1 6%
) r ec
ei ve
d tw
o va
cc in
at io
ns , a
nd fi
ve (5
% ) r
ec ei
ve d
on e
va cc
in at
io n;
re as
on s f
or n
ot fi
ni sh
in g
w er
e va
ca tio
n, ch
an ge
d w
or k
lo ca
tio n,
fo rg
et tin
g,
la zin
es s,
an d
di d
no t k
no w
; 7 5%
re ce
iv ed
th ei
r fir
st v
ac cin
at io
n at
a n
ou tr
ea ch
lo ca
tio n
W ea
k
Ja ns
se n
et a
l (2
00 9)
50
Ca na
da 10
0 St
re et
-b as
ed fe
m al
e se
x w or
ke rs
a ge
d ≥1
6 ye
ar s;
au th
or s
st at
ed th
at o
nl y
a sm
al l n
um be
r o f
pa rt
ici pa
nt s a
ge d
<1 8
ye ar
s w er
e in
clu de
d in
th is
st ud
y;
di sa
gg re
ga te
d da
ta
co ul
d no
t b e
ob ta
in ed
Pe er
-le d
m ob
ile
ou tr
ea ch
p ro
gr am
m e
(t he
M AP
, o r t
he M
AP
va n)
An a
ll- w
ee k
ni gh
tly o
ut re
ac h
se rv
ice st
aff ed
b y
a d riv
er , a
su pp
or t
w or
ke r,
an d
a pe
er -s
up po
rt w
or ke
r; pr
ov id
es a
sa fe
sp ac
e fo
r w om
en to
re
st , e
at , a
nd d
rin k;
o ut
re ac
h st
aff
di st
rib ut
e re
po rt
s o f r
ec en
t c lie
nt
vi ol
en ce
(n am
ed b
ad d
at e
re po
rt s)
; H
IV p
re ve
nt io
n re
so ur
ce s
di st
rib ut
ed (e
g, co
nd om
s, lu
br ica
nt ,
an d
sy rin
ge s)
a nd
re fe
rra ls
m ad
e t o
he al
th a
nd so
cia l s
up po
rt , a
nd d
ru g
tr ea
tm en
t s er
vi ce
s.
St re
et se
x w or
ke rs
n ot
a cc
es sin
g M
AP 94
% o
f M AP
u se
rs sa
id th
e va
n m
ad e t
he m
fe
el sa
fe r—
16 %
re ca
lle d
a t im
e it
pr ev
en te
d ph
ys ica
l a ss
au lt,
1 0%
re ca
lle d
a t im
e it
pr ev
en te
d se
xu al
a ss
au lt
W ea
k
(T ab
le 1
co nt
in ue
s o n
ne xt
p ag
e)
www.thelancet.com/public-health Vol 8 February 2023 e147
Review
a narrative synthesis36 to summarise findings, following guidance from the Cochrane Handbook for Systematic Reviews of Interventions.35
For the narrative synthesis, categories of intervention were developed based on the included papers, and interventions with multiple components were allocated to as many categories as relevant. Intervention com- ponents identified were education and empower ment, drug treatment, sexual and reproductive health care, other health care (eg, vaccination, screening, and primary care), and welfare. We summarised the papers in each intervention category according to four main areas: the nature of the interventions, outcomes reported, what was effective, and what was ineffective. We report outcomes as described in the studies but recognise that outcomes relating to cessation or reduction of sex working might not be wanted or important for many sex workers.
To quantitatively analyse results, we used vote counting, which can be used when outcomes are measured hetero- geneously between studies.35 Vote counting compares the number of studies in which a particular outcome improved with the number of studies in which that outcome did not improve, based only on the direction of effect and therefore with no measure of the magnitude of effect. All studies that measured outcomes before and after an intervention were included. For randomised controlled trials, both the intervention and control groups were included separately if enough information was available. We did this as most controls were well designed interventions that contributed important results to the Review. Intervention categories mirrored the narrative synthesis; the exception to this was that multicomponent interventions were categorised separately both to prevent double counting and because their effectiveness relies on the entirety of the intervention. Outcomes were grouped together into categories. Only outcome categories measured in two or more different interventions were included. If multiple outcomes were reported within one category for a particular intervention, only the primary outcome was used. If no primary outcome was identified and the results were not all in a single direction, the intervention was labelled as having mixed results for that outcome. No intervention had an outcome (or group of outcomes) that deteriorated within an outcome category. We display these data within a harvest plot, which provides a visual summary of the vote counting.35 Additionally, we produced a standard binary metric (benefit or mixed results), which we used to calculate a proportion, 95% CI (binomial exact calculation), and p value (binomial probability test) to show the evidence for each intervention category’s effectiveness across all outcome measures.
Results Overview 18 611 studies were identified through database searching and 123 through additional methods. After de-duplication
Co un
tr y
St ud
ie s (
n) Se
x w
or ke
r po
pu la
ti on
In te
rv en
ti on
se tt
in g
or c
on te
xt In
te rv
en ti
on Co
m pa
ris on
o r
co nt
ro l
O ut
co m
es Q
ua lit
y as
se ss
m en
t*
(C on
tin ue
d fro
m p
re vi
ou s p
ag e)
St ur
ro ck
et
a l (
20 07
)51
Au st
ra lia
71 M
al e
an d
fe m
al e
se x
w or
ke rs
in b
ro th
el s
O ut
re ac
h cli
ni cs
co
nd uc
te d
at b
ro th
el s
Tw o
nu rs
es a
nd o
ne se
xu al
h ea
lth
ed uc
at or
co nd
uc te
d w
ee kl
y cli
ni cs
fo
r 4 w
ee ks
, t hr
ee o
r f ou
r t im
es p
er
ye ar
N A
O ne
(2 %
) o f 6
3 ha
d ch
la m
yd ia
, n on
e o f
51 h
ad sy
ph ili
s, fo
ur (3
3% ) o
f 1 2
ha d
he pa
tit is
A, 3
4 (5
5% ) o
f 6 2
ha d
he pa
tit is
B,
th re
e (9
% ) o
f 3 4
ha d
he pa
tit is
C; o
f t ho
se w
ith
he pa
tit is
C, tw
o w
er e
ne w
ca se
s— bo
th
re po
rt ed
in je
ct io
n dr
ug u
se a
nd w
er e
re fe
rre d
to g
en er
al p
ra ct
iti on
er s;
42 (6
2% ) o
f 6 8
re tu
rn ed
fo r t
he ir
re su
lts ; s
ev en
(1 0%
) o f
68 re
ce iv
ed tr
ea tm
en t;
tw o
se x w
or ke
rs
re ce
iv ed
p os
t- co
ita l c
on tr
ac ep
tio n;
68
(9 6%
) o f 7
1 w
er e
fe m
al e
se x w
or ke
rs
W ea
k
W on
g (2
00 7)
52
H on
g Ko
ng 24
5 St
re et
-b as
ed fe
m al
e se
x w or
ke rs
Cl in
ic— ou
tr ea
ch
w or
ke rs
in vi
te d
fe m
al e
se x w
or ke
rs
Tw ice
p er
m on
th cl
in ic
pr ov
id in
g ce
rv ica
l s m
ea r t
es tin
g; fo
llo w
-u p
ex pl
an at
io n
of re
su lts
a cc
om pa
ni ed
by
a le
tt er
fo r a
g yn
ae co
lo gi
st
N A
23 6
ce rv
ica l s
m ea
rs d
on e,
2 9
(1 2%
) s m
ea rs
w
er e
at yp
ica l;
ni ne
(3 1%
) o f t
he 2
9 w
er e
re fe
rre d
fo r f
ur th
er m
an ag
em en
t, an
d 13
(4 5%
) w er
e u nc
on ta
ct ab
le
W ea
k
Lo m
ax e
t a l
(2 00
6) 53
UK 24
St re
et -b
as ed
se x
w or
ke rs
Ge ni
to ur
in ar
y m
ed ici
ne cl
in ic
Af te
r a lo
ca l s
yp hi
lis o
ut br
ea k,
se x
w or
ke rs
w er
e in
vi te
d to
a cl
in ic
fo r
ST I a
nd H
IV sc
re en
in g;
tr
an sp
or ta
tio n
w as
p ro
vi de
d to
a nd
fro
m th
e cli
ni c
N A
24 w
om en
se en
; 1 5
ca se
s o f s
yp hi
lis w
er e
di ag
no se
d in
1 4
w om
en (o
ne w
as a
ca se
o f
re in
fe ct
io n)
; o nl
y fo
ur w
er e
gi ve
n in
tr am
us cu
la r b
en za
th in
e be
nz yl
pe ni
cil lin
(t
re at
m en
t o f c
ho ice
), th
e re
st d
ec lin
ed a
nd
w er
e gi
ve n
su bo
pt im
al o
ra l t
re at
m en
t; th
er e
w as
o ne
n ew
H IV
d ia
gn os
is an
d se
ve ra
l o th
er
ST Is
al so
d ia
gn os
ed
W ea
k
EB FT
=e co
lo gi
ca lly
b as
ed fa
m ily
th er
ap y.
M AP
=M ob
ile A
cc es
s P ro
je ct
. N A=
N ot
a pp
lic ab
le . N
GO =n
on -g
ov er
nm en
ta l o
rg an
isa tio
n. N
ID A=
N at
io na
l I ns
tit ut
e o n
Dr ug
A bu
se . P
rE P=
pr e-
ex po
su re
p ro
ph yl
ax is.
P TS
D= po
st -t
ra um
at ic
st re
ss d
iso rd
er .
ST I=
se xu
al ly
tr an
sm itt
ed in
fe ct
io n.
W IS
H =W
om en
’s In
fo rm
at io
n Sa
fe H
av en
. * Su
bc at
eg or
ise d
by st
ud y d
es ig
n hi
er ar
ch y. †S
tu dy
ca te
go ris
at io
ns u
se d
ar e d
er iv
ed fr
om th
e Eff
ec tiv
e Pu
bl ic
H ea
lth P
ra ct
ice P
ro je
ct ’s
Q ua
lit y A
ss es
sm en
t T oo
l f or
Q
ua nt
ita tiv
e St
ud ie
s.33
Ta bl
e 1 : S
um m
ar y
an d
ch ar
ac te
ris ti
cs o
f i nc
lu de
d st
ud ie
s
e148 www.thelancet.com/public-health Vol 8 February 2023
Review
and initial screening, 200 were reviewed in full. 20 studies were included in the final review (figure 1). Summary characteristics and categorisations of included studies are presented in table 1 and table 2. The appendix (p 4) shows a map detailing the number of studies included in each intervention category by country.
Most studies were from North America. 11 (55%) focused on street-based sex workers, and nearly all exclusively studied female sex workers. Eight (40%) of the inter ventions were multicomponent. 18 (90%) of the
inter ventions (90%) took place in a context where sex work was fully or partly criminalised at the time of study. Interventions were primarily based in static locations, although seven (35%) studies included outreach compo nents. The most common outcomes measured related to drug use and drug harm reduction, sexual risk behaviours, and mental health and wellbeing. No harms associated with the interventions were reported. All studies presented limitations in sampling strategy. Most used convenience or snowball sampling. A few used repeated time-space sampling of mapped sex worker districts to improve systematicity.26,38,41,42,49 Four (20%) studies were randomised controlled trials, but all had limitations including non-systematic recruit- ment strategies,37,38,40,41 an absence of information on the randomisation process,38,40 and no data on loss to follow-up.40 Only one was reported using Consolidated Standards of Reporting Trials guidelines.37
The harvest plot (figure 2) summarises evidence for effectiveness within each intervention category across
Studies (n=20)
Country
USA9,14,38–41,43,44,46–48 11 (55%)
Canada26,42,50 3 (15%)
UK45,53 2 (10%)
Hong Kong37,52 2 (10%)
Australia51 1 (5%)
Netherlands49 1 (5%)
Sex work legal context in intervention setting at time of study*
Full criminalisation of sex working9,14,38–41,43,44,46–48 11 (55%)
Partial criminalisation of sex working26,37,42,45,50,52,53 7 (35%)
Criminalisation of the purchase of sex 0
Regulation of sex working49,51 2 (10%)
Full decriminalisation 0
Sex worker’s sex, gender, or both†
Female9,14,26,37–53 20 (100%)
Male51 1 (5%)
Transgender women26,42 2 (10%)
Location of sex work
Street based26,38,41–43,45–47,50,52,53 11 (55%)
Brothel or indoor based51 1 (5%)
Street based and indoor based44,49 2 (10%)
Unclear where sex work takes place9,14,37,39,40,48 6 (30%)
Study design‡
Randomised controlled trial37,38,40,41 4 (20%)
Cohort analytic (two groups, pre-intervention and post-intervention)14,26,42
3 (15%)
Cohort (one group, pre-intervention and post- intervention)9,39,43–47
7 (35%)
Cross-sectional study48–53 6 (30%)
Quality rating
Strong 0
Moderate37–39 3 (15%)
Weak9,14,26,40–53 17 (85%)
Intervention setting
Outreach locations40,42–44,49–51 7§ (35%)
Static site9,14,26,37–41,45–48,52,53 14§ (70%)
Single component interventions
Overall9,14,37–41,43,44,49,51,53 12 (60%)
Education and empowerment37–41,44 6
Drug treatment14,43 2
Sexual and reproductive health care51,53 2
Other health care9,49 2
(Table 2 continues in next column)
Studies (n=20)
(Continued from previous column)
Multicomponent interventions†
Overall26,42,45–48,50,52 8 (40%)
Education and empowerment26,42,45–48,50,52 7
Drug treatment45,48 2
Sexual and reproductive health care26,42,48,50,52 5
Other health care26,48,52 3
Welfare26,42,46,47,50 5
Peer involvement
Developed with peer workers44 1 (5%)
Developed and delivered with peer workers26,38,41,42,50 5 (25%)
No peer involvement reported9,14,37,39,40,43,45–49,51–53 14 (70%)
Outcomes measured†
Drug use and drug harm reduction38,41,43,45–47,50 7 (35%)
Sexual risk behaviours14,37,38,41,44,45,47 7 (35%)
Sex worker safety38,44,50 3 (15%)
Mental health and wellbeing9,37,40,44,45 5 (25%)
Criminal activity39,46 2 (10%)
Outcomes related to wider determinants46 1 (5%)
Awareness of health-care and support services44 1 (5%)
Use of other health-care and support services37,41,43 3 (15%)
Sexually transmitted infection treatment48,51,53 3 (15%)
Other health-care outcomes26,38,41,42,48–52 6 (30%)
Data are presented as n or n (%). *Legal context categorisations from Platt and colleagues (2018).22 Full criminalisation prohibits all aspects of sex work and selling and buying sex; partial criminalisation criminalises only some aspects; in criminalisation of purchase of sex models, the sale of sex is legal but clients are criminalised; and regulatory models allow the sale of sex in some settings or conditions. Full decriminalisation removes all criminality of sex work while still prohibiting violence and coercion of sex workers. †Can be in more than one category. ‡Study categorisations used are derived from the Effective Public Health Practice Project’s Quality Assessment Tool for Quantitative Studies.33 §One study evaluated both outreach and static interventions.
Table 2: Characteristics of included studies by category
www.thelancet.com/public-health Vol 8 February 2023 e149
Review
different outcomes. A total of 15 interventions from 12 studies could be included within the harvest plot. Nine (60%) of 15 included interventions were focused on education and empowerment and many showed improvements in one or more outcome. Multi- component interventions showed potential benefit, although only three (15%) interventions were included and all were of low quality. Only a small amount of evidence could be included for drug treatment and other health-care interventions; however, drug treatment was a central component of one of the multicomponent interventions (Litchfield and colleagues [2010]).45 No studies based on sexual and reproductive health care could be included. With the exception of Decker and colleagues (2017),44 which only included a peer-design element, all other interventions involving peers included both a design and delivery element and showed potential benefit. Three outreach interventions were included, of which one showed potential benefits across outcomes,40 and two showed mixed results.43,44
We analysed the number of positive outcomes (potential benefit) to the number of total outcomes reported per intervention category using the binomial exact calculation and binomial probability test (table 3). Education and empowerment and multicomponent interventions showed a greater proportion of positive outcomes than would have been expected by chance, suggesting their potential effectiveness, whereas the little evidence for drug treatment and other health-care interventions precludes clear insight.
Education and empowerment Seven interventions (six single component,37–41,44 and one multi component47) focused on education and empower- ment, and four multicomponent interventions had a small educational component, but did not detail what was provided.45,46,51,52 Of the seven, three focused on street-based sex workers,38,41,47 one on street-based and indoor-based sex workers,44 and in the other three the authors did not state the sex worker population that the intervention was targeting.37,39,40 Three were of moderate quality,37–39 and four were of weak quality.40,41,44,47
A few studies used health behaviour models that recognise structural and environmental vulnerabilities contributing to HIV and sexual health risk.37,47 Structural determinants were addressed through enhancing sex worker self-efficacy and condom negotiation skills,37,47 as well as teaching strategies to minimise risk of violence.38,44 Several used psychological therapies— five were individually administered,37–39,41,44 and one used family therapy between mothers who were sex workers and their children.40 Two were developed and delivered in collaboration with peer sex workers.38,41 Key study outcomes for these interventions related to sexual risk behaviours,37,38,41,47 drug use and drug harm reduction,38,41,47 mental health and wellbeing,37,40,44 use of other health-care and support services,37,41 and criminal activity.39
All interventions showed a level of effectiveness, but most only measured outcomes at 3 months after intervention.37,44,47 A brief intervention that provided
Figure 2: Harvest plot of evidence for interventions to improve health and wider determinants in sex workers by intervention category The harvest plot is a supermatrix showing the direction of effect for outcome categories across different categories of intervention. Each bar represents an intervention and is labelled by a footnote, which can be identified below. Taller bars represent interventions from studies with a moderate-quality assessment; shorter bars are interventions from studies with a low-quality assessment. Darker purple bars are static interventions. Lighter purple bars are outreach. Striped bars are interventions that involved peer design, delivery, or both. Solid bars had no peer involvement. *Murnan et al (2018)40—home intervention. †Murnan et al (2018)40—office intervention. ‡Surratt and Inciardi (2010)38—sex worker-focused intervention. §Surratt and Inciardi (2010)38— National Institute on Drug Abuse intervention. ¶Surratt et al (2014)41— professional–peer intervention. ||Decker et al (2017).44 **Wong et al (2019)37—resilience programme. ††Burnette et al (2009).14 ‡‡Cigrang et al (2020).39 §§Park et al (2020).43 ¶¶Ward and Roe-Sepowitz (2009)9—prison group intervention. ||||Ward and Roe-Sepowitz (2009)9—community group intervention. ***Litchfield et al (2010).45 †††Sherman et al (2006).47 ‡‡‡Bowser et al (2008).46
Mixed results
Potential benefit
Substance misuse
Sexual risk behaviours
Sex worker safety
Mental health and wellbeing
Crime
Support service use
Mixed results
Potential benefit
* † ‡ § ¶
‡
‡
§ ¶
§
¶
||
||
|| **
**
** ††
‡‡
§§
§§
¶¶ ||||
***
***
***
†††
†††
‡‡‡
‡‡‡
A Education and empowerment B Drug treatment
C Other health care D Multicomponent interventions
* †
Substance misuse
Sexual risk behaviours
Sex worker safety
Mental health and wellbeing
Crime
Support service use
e150 www.thelancet.com/public-health Vol 8 February 2023
Review
information on strategies to improve sex worker safety and reduce the risk of violence affected safety behaviours and use of relevant support programmes 12 weeks later.44 A six-session resilience-promoting programme showed improvements in resilience, self-esteem, and condom use 3 months later.37 A 12-session family therapy programme showed greater reductions in drug use and depressive symptoms than a psychoeducational pro- gramme with only sex workers.40 One programme helped sex workers to develop negotiation skills with different sexual partner types alongside teaching jewellery-making skills.47 3 months post-intervention, there were reductions in transactional and total sex partners, as well as injection and non-injection drug use. In another intervention, female sex workers in prison were provided two brief motivational interviews to help identify and problem solve their greatest concerns for post-release.39 This intervention led to a reduction in the number of arrests in the 12-month period after release.
Two studies including peer sex workers in the development and delivery of an intervention showed mixed results.38,41 Both were randomised controlled trials with interventions showing similarly positive outcomes to control groups, which were high quality. One found that a strengths-based programme did not show additional effectiveness when incorporating a peer facilitator over a case manager alone.38 The other study showed that a sex worker-focused HIV risk education programme, developed and delivered in collaboration with sex workers, led to a significantly greater reduction in unprotected oral sex and episodes of sexual violence than the US National Institute on Drug Abuse standard intervention at 6-month follow-up. However, other HIV risk outcomes were similar to the standard intervention.41
Drug treatment Drug treatment was provided by four interventions (two single component14,43 and two multicomponent45,48). Three specifically targeted sex worker populations—a drug treatment clinic for street-based female sex workers,45
a one-off harm reduction intervention for street-based female sex workers,43 and a one day per week clinic offering an array of primary care and harm reduction services to an unspecified sex worker population.48 The fourth study compared outcomes between sex workers (no subpopulation identified) and non-sex workers using US-Government-funded drug treatment programmes across 71 facilities.14 Three interventions were at static locations,14,45,48 and all studies were low quality.
The primary care clinic studied by Stewart and colleagues48 found that 31 (62%) of 50 women seen had opioids in their urine. Of these, nine (29%) of 31 started opioid substitution therapy (OST) and three (10%) were already in OST programmes. In the harm reduction inter vention, women were given naloxone, harm reduction advice, and self-administered tests for detecting the presence of fentanyl in drugs.43 Fentanyl has a higher risk of overdose and death compared with heroin. 1 month after intervention, opioid and injection drug use, as well as solitary drug use, had reduced. However, fentanyl detection in drugs did not lead to changes in harm reduction behaviours for most people. The other two studies on drug treatment found a significant decrease in drug use at the end of drug programmes (one focused on heroin,45 the other included various drugs14), and a reduction in the number of women still engaging in sex work.14,45 Burnette and colleagues14 found that those still involved in sex work were doing significantly less sex work than they had before. Both interventions provided physical and mental health services alongside OST, which led to improvements in mental health and wellbeing. Burnette and colleagues14 found higher use of mental health services was associated with increased probability of cessation of sex work at follow-up, which in turn was associated with lower drug use, higher abstinence rates, and fewer mental health symptoms.14
Sexual and reproductive health care Two interventions provided STI screening, STI treat- ment, and HIV pre-exposure prophylaxis through sexual health outreach clinics in brothels,51 and a 1 day per week, multicomponent primary care intervention for an unspecified sex worker population.48 Two multi- component welfare services for street-based sex workers provided free condoms and lubricants.26,42,50 One study described the management of a syphilis outbreak in street-based sex workers in east London.53 Through partnership with a charity providing outreach to sex workers, women with suspected syphilis were invited to the charity’s drop-in centre, from where they were driven to a nearby genitourinary medicine clinic. Reproductive health-care service components included pregnancy testing,48 contraceptive prescriptions and advice,45,52 and post-coital contraception.51 No details of service provision, uptake, or acceptability were provided and no study focused on reproductive health care.
Number of potentially beneficial outcomes of total outcomes
95% CI of the proportion of beneficial outcomes
p value
Education and empowerment
17 (80%) of 21 58·1–94·6 0·007
Drug treatment 0 (0%) of 2 0–84·2 0·50
Other health care 0 (0%) of 2 0–84·2 0·50
Multicomponent interventions
7 (100%) of 7 59·0–100 0·016
For each intervention category, a p value was calculated using the binomial probability test to determine the chance that the true proportion of potentially beneficial outcomes of total outcomes was 0·50. Accompanying exact 95% binomial CIs are also displayed.
Table 3: Effectiveness of intervention categories across different outcome measures
www.thelancet.com/public-health Vol 8 February 2023 e151
Review
Studies were observational and low quality. Two pro- vided information on STI treatment.51,53 Sturrock and colleagues51 invited sex workers with positive results back for treatment. 42 (62%) of 68 participants returned for their results and seven (17%) of 42 returning sex workers received treatment. In the syphilis outbreak, epi demiological treatment (ie, treatment based on probable exposure) was provided to all sex workers.53 Most individuals declined intramuscular penicillin— the best available treatment—and many instead chose oral antibiotics, which are a suboptimal alternative. 13 (93%) of 14 sex workers were followed up.
Other health care Other health-care interventions included a trauma-based psychoeducational therapy group for street-based sex workers,9 a vaccination programme for multiple sex worker populations,49 a multicomponent clinic focused on women’s health for street-based female sex workers,52 and multicomponent primary care clinics in the proximity of welfare drop-in centres.26,48 The vaccination programme was a nationally run programme in the Netherlands. The programme provided free hepatitis B vaccinations to sex workers through local community health services, working alongside existing sex worker outreach services, and by community health-service staff periodically visiting various sex work locations, including brothels and streets over several years. All other interventions were at static locations, with the therapy group provided at both a community-based residential centre and a moderate-security prison. All studies were low quality. Several interventions offered referral to other health or social services, but no study gave information on the uptake of this offer.26,38,40,42,50,52
The psychotherapy group participants’ trauma scores decreased significantly in six of ten parameters at the end of the 12-week intervention.9 The decrease was more profound in the prison group than in the residential centre, possibly as their baseline trauma scores were higher. Baars and colleagues49 provided evidence of the effectiveness of the Netherlands’ hepatitis B vaccination programme. Through a cross-sectional survey of 259 sex workers working in various settings across three cities, they found that 2 years after programme initiation, 205 (79%) of 259 were aware of the programme and 163 (63%) of 257 had received at least one dose—134 (82%) of 163 through the programme. Of those who started the vaccine programme, 74 (79%) of 94 received all three vaccinations. Those who had been vaccinated were more likely to have worked in an area for longer and 75% reported receiving their first vaccination at an outreach location. Wong evaluated a well-women clinic’s cervical cancer screening intervention for street- based female sex workers in Hong Kong.52 208 (88%) of 236 tested women returned for their smear results, and 13 (45%) of 29 women with atypical smear results were uncontactable. Nine (31%) of the 29 with atypical results
were given referral letters to attend a gynaecologist, but it is not known whether they were seen.
Welfare All interventions that addressed welfare were multi- component and focused on meeting basic needs through providing food and drink, washing facilities, clothing, and a safe space.26,42,46,50 The intervention by Sherman and colleagues47 was the only exception which, alongside teaching better condom negotiation skills, taught female sex workers jewellery-making skills over six sessions. These women then had the opportunity to sell their handmade items at a stand within a hospital. The intervention was designed to address structural deter- minants preventing these women earning a sustainable, alternative income. 3 months after com pletion, there were significant reductions in transactional and total sex partners, as well as injection and non-injection drug use.47 Women who earned more money through market sales had a significantly decreased number of transactional sex partners at follow-up. All welfare interventions were oriented towards street-based sex workers and studies were of low quality.
Three studies focused on two linked interventions in Vancouver, BC, Canada.26,42,50 The inter ventions, both designed for female sex workers, were the Women’s Information Safe Haven (WISH) drop-in centre and a peer-led, van-based outreach programme called the Mobile Access Project (MAP). Those with greater numbers of clients and working in isolated areas were more likely to use the MAP van,42 reflecting the outreach approach used. The studies showed that both services were associated with accessing other health services— inpatient addiction services for the MAP van,42 and sexual and reproductive services for WISH.26 However, the temporality of both relationships is unclear. Both the MAP van and WISH were less likely to be used by younger sex workers compared with older sex workers. Of those who used the MAP van, 94% felt safer when the van was present, 16% recalled a time it had prevented physical assault, and 10% a time it had prevented sexual assault.
Discussion We identified 20 studies, with intervention components divided into education and empowerment, drug treat- ment, sexual and reproductive health care, other health care, and welfare. 12 interventions were single component and eight were multicomponent. Considering the diversity of sex worker populations and their corresponding needs, this was a very small number of studies. There was promising evidence for interventions that focused on education and empowerment and those that were multicomponent. Sherman and colleagues’47 jewellery skills and sexual negotiation strategy workshops were particularly innovative as a multicomponent inter- vention combining empowerment and a focus on the
e152 www.thelancet.com/public-health Vol 8 February 2023
Review
structural determinants of health. Evidence across studies also showed that designing and delivering interventions alongside sex workers was effective. Importantly, only six interventions used co-design or co-delivery. The harvest plot provided unclear results as to the effectiveness of outreach. However, two interventions that involved outreach, but could not be included in the plot because they were cross-sectional studies and did not follow up participants, showed evidence of possible benefit. Both the Netherlands’ hepatitis B vaccination programme49 and the management of a syphilis outbreak in east London53 relied on collaboration with existing outreach services and showed good uptake and retention. Few interventions incorporated reproductive health care,48 and there was no evidence for interventions treating chronic diseases. One intervention provided cervical cancer screening, but many people with atypical results could not be contacted and informed, and it is unclear whether those who were contacted were followed up.52
Previous systematic reviews analysing health inter- ventions for sex workers in high-income countries underscored the need for flexible services which are non- judgemental, built on respect and trust, trauma- informed, and targeted at specific sex worker needs.10,29,30 Our Review also found outreach might be important in ensuring high levels of engagement in some contexts. However, outreach was not always enough to ensure continuity of care. Similar to other systematic reviews,10,29 we found many studies highlighted low levels of follow-up. Three studies were exceptions to this. One involved OST,45 possibly showing the perceived value of this intervention. The other two,49,53 as discussed earlier, collaborated with established outreach services, which might have improved trust and provided a channel by which to follow up individuals.
To the best of our knowledge, this Review is the first compre hensive overview of evidence on sex worker interventions aiming to improve health and wider determinants of health outcomes in high-income countries. Academic databases and grey literature were searched, and both academic experts and people with lived experience of sex work were contacted to ensure we identified all relevant literature. Importantly, we have included authors with lived experience, and authors who have worked with and continue to work with sex workers, from the study’s inception—to develop the search strategy, ensuring relevant grey literature channels were searched and experts in the field contacted, and to ensure findings were relevant, correctly interpreted, and presented with appropriate language and without stigma.
This Review has some limitations. Where stated, the majority of interventions were either primarily or exclusively targeted at street-based sex workers,26,38,41–43,45–47,50,52,53 probably because they are more easily identified by service providers and researchers; are more exposed to structural determinants such as homelessness, poverty, and violence;18,54,55 and typically
have worse health outcomes.56 Therefore, general is- ability of this Review’s findings to other sex worker populations is limited. People engaged in street-based sex work often have a range of different health and social issues, including homelessness,4,18 drug use,10,14 and history of imprison ment,10 emphasising the need for a wider inclusion health approach to service provision and research that addresses multiple, overlapping risk factors and vulnerabilities.28 We reviewed English language studies since 2005 as a pragmatic choice and because an initial scoping search suggested most studies relevant to this Review met these criteria. We did not include qualitative studies that might provide insight into differences in results between studies. Outcomes were highly heterogeneous, often self-reported, and might not be the outcomes that are important for all sex workers. The development of a core outcome set in collaboration with sex workers would help future researchers to ensure that outcomes measured are relevant.57 Methods used by the included studies also represent an important limitation, with only three studies37–39 rated moderate in our quality assessment, and all other studies rated weak (table 2). One common reason for low quality was study design— the most common design was a single group, pre-post cohort study (often referred to as quasi-experimental studies). Additionally, due to the nature of recruiting marginalised populations, all studies presented limi- tations in sampling strategy and most used either convenience or snowball sampling. Finally, the com- plexity and dynamic nature of the legal sex working context in which the interventions took place could not be accounted for in the Review’s findings and is likely to be an explanatory factor for study heterogeneity.
There is scarce investment both in services and research, particularly for sex workers who are not street based. However, a range of interventions are likely to be effective. Services should be developed and delivered in collaboration with sex workers. Interventions that are focused on education and empowerment or those that are multicomponent are likely to be effective, and an outreach or drop-in component could be of benefit in some contexts.58 Future interventions should incorporate components related to chronic diseases given they are an important contributor to sex worker mortality.58 Within the identified studies, almost all interventions were designed exclusively for female sex workers—the only exceptions being two that included transgender women sex workers,26,42 and one that included male sex workers.51 Sex worker services and future research should take a gender-sensitive and inclusive approach. Several studies highlighted that sex workers who were new to working in an area were less likely to access services than those who had been working in an area for longer.26,42,49 Effective information dissemination and outreach could help ensure accessibility. Crucially, repressive policing practices and the criminalisation of
www.thelancet.com/public-health Vol 8 February 2023 e153
Review
sex work have already been shown to adversely affect access to health and social services and sex worker health outcomes.22 Therefore, the effectiveness of any service will always be restricted in settings where sex work is criminalised. Contributors LCP and SAL conceptualised the study. LCP, MB, BM, LJ, LW, KW, and HB were involved in screening and data extraction. All authors were involved in interpretation of the findings. LJ and SAL wrote the first draft. All authors were involved in draft revisions and approving the final draft for submission. All authors had full access to all systematic review data and accept responsibility for the decision to submit for publication.
Declaration of interests SAL and LCP are Pathway Fellows. Pathway is a charity that provides health care to homeless and inclusion health patients, including sex workers. All other authors declare no competing interests.
Acknowledgments We would like to thank Kelsey Johnson for her help with some figures. Although there was no specific funding for this study, LJ, LCP, BS, and SAL were supported by funding from the National Institute for Health Research (NIHR). This study was also supported by the Wellcome Trust through a Wellcome Clinical Research Career Development Fellowship to RWA (206602).
References 1 Hester M. Natasha Mulvihill, Matolcsi A, Sanchez AL, Walker S-J.
The nature and prevalence of prostitution and sex work in England and Wales today. Bristol: University of Bristol, 2019.
2 Havocscope. Prostitution: prices and statistics of the global sex trade. 2015. https://www.havocscope.com/wp-content/uploads/2015/06/ ProstitutionPricesandStatisticsoftheGlobalSexTrade-3.pdf (accessed Sept 9, 2022).
3 Laing DM, Feis-Bryce A, Sanders T, Pitcher J. Written evidence submitted by National Ugly Mugs (CVG0014). July, 2020. https://committees.parliament.uk/writtenevidence/8610/pdf/ (accessed Sept 9, 2022).
4 Inciardi JA, Surratt HL, Kurtz SP. HIV, HBV, and HCV infections among drug-involved, inner-city, street sex workers in Miami, Florida. AIDS Behav 2006; 10: 139–47.
5 Burnette ML, Lucas E, Ilgen M, Frayne SM, Mayo J, Weitlauf JC. Prevalence and health correlates of prostitution among patients entering treatment for substance use disorders. Arch Gen Psychiatry 2008; 65: 337–44.
6 McMahon JM, Tortu S, Pouget ER, Hamid R, Neaigus A. Contextual determinants of condom use among female sex exchangers in East Harlem, NYC: an event analysis. AIDS Behav 2006; 10: 731–41.
7 Jeal N, Salisbury C. A health needs assessment of street-based prostitutes: cross-sectional survey. J Public Health (Oxf) 2004; 26: 147–51.
8 Creighton S, Tariq S, Perry G. Sexually transmitted infections among UK street-based sex workers. Sex Transm Infect 2008; 84: 32–33.
9 Ward A, Roe-Sepowitz D. Assessing the effectiveness of a trauma- oriented approach to treating prostituted women in a prison and a community exiting program. J Aggress Maltreat Trauma 2009; 18: 293–312.
10 Abad N, Baack BN, O’Leary A, Mizuno Y, Herbst JH, Lyles CM. A systematic review of HIV and STI behavior change interventions for female sex workers in the United States. AIDS Behav 2015; 19: 1701–19.
11 Jeal N, Salisbury C, Turner K. The multiplicity and interdependency of factors influencing the health of street-based sex workers: a qualitative study. Sex Transm Infect 2008; 84: 381–85.
12 Footer KHA, Park JN, Allen ST, et al. Police-related correlates of client-perpetrated violence among female sex workers in Baltimore City, Maryland. Am J Public Health 2019; 109: 289–95.
13 Connelly L, Kamerāde D, Sanders T. Violent and nonviolent crimes against sex workers: the influence of the sex market on reporting practices in the United Kingdom. J Interpers Violence 2021; 36: NP3938–63.
14 Burnette ML, Schneider R, Timko C, Ilgen MA. Impact of substance-use disorder treatment on women involved in prostitution: substance use, mental health, and prostitution one year after treatment. J Stud Alcohol Drugs 2009; 70: 32–40.
15 Argento E, Goldenberg S, Shannon K. Preventing sexually transmitted and blood borne infections (STBBIs) among sex workers: a critical review of the evidence on determinants and interventions in high-income countries. BMC Infect Dis 2019; 19: 212.
16 Roxburgh A, Degenhardt L, Copeland J, Larance B. Drug dependence and associated risks among female street-based sex workers in the greater Sydney area, Australia. Subst Use Misuse 2008; 43: 1202–17.
17 Jeal N, Macleod J, Salisbury C, Turner K. Identifying possible reasons why female street sex workers have poor drug treatment outcomes: a qualitative study. BMJ Open 2017; 7: e013018.
18 Surratt HL, O’Grady CL, Kurtz SP, Buttram ME, Levi-Minzi MA. HIV testing and engagement in care among highly vulnerable female sex workers: implications for treatment as prevention models. J Health Care Poor Underserved 2014; 25: 1360–78.
19 Marmot M. Social determinants of health inequalities. Lancet 2005; 365: 1099–104.
20 Rhodes T. Risk environments and drug harms: a social science for harm reduction approach. Int J Drug Policy 2009; 20: 193–201.
21 Shannon K, Goldenberg SM, Deering KN, Strathdee SA. HIV infection among female sex workers in concentrated and high prevalence epidemics: why a structural determinants framework is needed. Curr Opin HIV AIDS 2014; 9: 174–82.
22 Platt L, Grenfell P, Meiksin R, et al. Associations between sex work laws and sex workers’ health: a systematic review and meta-analysis of quantitative and qualitative studies. PLoS Med 2018; 15: e1002680.
23 Potter LC, Horwood J, Feder G. Access to healthcare for street sex workers in the UK: perspectives and best practice guidance from a national cross-sectional survey of frontline workers. BMC Health Serv Res 2022; 22: 178.
24 Mosedale B, Kouimtsidis C, Reynolds M. Sex work, substance misuse and service provision: the experiences of female sex workers in south London. Drugs Educ Prev Policy 2009; 16: 355–63.
25 Socías ME, Shoveller J, Bean C, Nguyen P, Montaner J, Shannon K. Universal coverage without universal access: institutional barriers to health care among women sex workers in Vancouver, Canada. PLoS One 2016; 11: e0155828.
26 Kim SR, Goldenberg SM, Duff P, Nguyen P, Gibson K, Shannon K. Uptake of a women-only, sex-work-specific drop-in center and links with sexual and reproductive health care for sex workers. Int J Gynaecol Obstet 2015; 128: 201–05.
27 WHO. Consolidated guidelines on HIV prevention, diagnosis, treatment and care for key populations. Geneva: World Health Organization, 2016.
28 Luchenski S, Maguire N, Aldridge RW, et al. What works in inclusion health: overview of effective interventions for marginalised and excluded populations. Lancet 2018; 391: 266–80.
29 Turner K, Meyrick J, Miller D, Stopgate L. Which psychosocial interventions improve sex worker well-being? A systematic review of evidence from resource-rich countries. BMJ Sex Reprod Health 2022; 48: e88–100.
30 Jeal N, Macleod J, Turner K, Salisbury C. Systematic review of interventions to reduce illicit drug use in female drug-dependent street sex workers. BMJ Open 2015; 5: e009238.
31 Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement. PLoS Med 2009; 6: e1000097.
32 The World Bank. World Bank Country and Lending Groups. 2021. https://datahelpdesk.worldbank.org/knowledgebase/articles/906519- world-bank-country-and-lending-groups (accessed Dec 3, 2021).
33 Thomas BH, Ciliska D, Dobbins M, Micucci S. A process for systematically reviewing the literature: providing the research evidence for public health nursing interventions. Worldviews Evid Based Nurs 2004; 1: 176–84.
34 Armijo-Olivo S, Stiles CR, Hagen NA, Biondo PD, Cummings GG. Assessment of study quality for systematic reviews: a comparison of the Cochrane Collaboration Risk of Bias Tool and the Effective Public Health Practice Project Quality Assessment Tool: methodological research. J Eval Clin Pract 2012; 18: 12–18.
e154 www.thelancet.com/public-health Vol 8 February 2023
Review
35 McKenzie JE, Brennan SE. Chapter 12. Synthesizing and presenting findings using other methods. In: Higgins J, Thomas J, eds. Cochrane handbook for systematic reviews of interventions. Hoboken, NJ: Wiley-Blackwell, 2021.
36 Popay J, Roberts H, Sowden A, et al. Guidance on the conduct of narrative synthesis in systematic reviews: ESRC methods programme. Lancaster: University of Lancaster, 2006.
37 Wong WC-W, Yuen WW-Y, Tang CS-K, Holroyd E, Fong DY-T. Evaluation and mechanism analysis of HIV prevention programme using resilience framework among female sex workers: a randomised controlled trial. Prev Med Rep 2019; 13: 229–37.
38 Surratt HL, Inciardi JA. An effective HIV risk-reduction protocol for drug-using female sex workers. J Prev Interv Community 2010; 38: 118–31.
39 Cigrang JA, Fedynich AL, Nichting EMS, Frederick SAL, Schumm JA, Auguste CB. Brief motivational interview-based intervention for women in jail with history of drug addiction and sex-trading. Prof Psychol Res Pr 2020; 51: 25–33.
40 Murnan A, Wu Q, Slesnick N. Effects of ecologically-based family therapy with substance-using, prostituting mothers. J Fam Ther 2018; 40: 557–83.
41 Surratt HL, O’Grady C, Kurtz SP, Levi-Minzi MA, Chen M. Outcomes of a behavioral intervention to reduce HIV risk among drug-involved female sex workers. AIDS Behav 2014; 18: 726–39.
42 Deering KN, Kerr T, Tyndall MW, et al. A peer-led mobile outreach program and increased utilization of detoxification and residential drug treatment among female sex workers who use drugs in a Canadian setting. Drug Alcohol Depend 2011; 113: 46–54.
43 Park JN, Tomko C, Silberzahn BE, Haney K, Marshall BDL, Sherman SG. A fentanyl test strip intervention to reduce overdose risk among female sex workers who use drugs in Baltimore: results from a pilot study. Addict Behav 2020; 110: 106529.
44 Decker MR, Tomko C, Wingo E, et al. A brief, trauma-informed intervention increases safety behavior and reduces HIV risk for drug-involved women who trade sex. BMC Public Health 2017; 18: 75.
45 Litchfield J, Maronge A, Rigg T, Rees B, Harshey R, Keen J. Can a targeted GP-led clinic improve outcomes for street sex workers who use heroin? Br J Gen Pract 2010; 60: 514–16.
46 Bowser BP, Ryan L, Smith CD, Lockett G. Outreach-based drug treatment for sex trading women: the Cal-Pep risk-reduction demonstration project. Int J Drug Policy 2008; 19: 492–95.
47 Sherman SG, German D, Cheng Y, Marks M, Bailey-Kloche M. The evaluation of the JEWEL project: an innovative economic enhancement and HIV prevention intervention study targeting drug using women involved in prostitution. AIDS Care 2006; 18: 1–11.
48 Stewart J, Stadeli KM, Green ML, et al. A co-located continuity clinic model to address healthcare needs of women living unhoused with opioid use disorder, who engage in transactional sex in north Seattle. Sex Transm Dis 2020; 47: e5–7.
49 Baars JE, Boon BJ, Garretsen HF, van de Mheen D. Vaccination uptake and awareness of a free hepatitis B vaccination program among female commercial sex workers. Womens Health Issues 2009; 19: 61–69.
50 Janssen PA, Gibson K, Bowen R, Spittal PM, Petersen KL. Peer support using a mobile access van promotes safety and harm reduction strategies among sex trade workers in Vancouver’s Downtown Eastside. J Urban Health 2009; 86: 804–09.
51 Sturrock CJ, Currie MJ, Vally H, et al. Community-based sexual health care works: a review of the ACT outreach program. Sex Health 2007; 4: 201–04.
52 Wong WC-W. Feasibility study of an outreach well-women clinic for female sex workers in Hong Kong. A study of female sex workers in Hong Kong—moving beyond sexual health. Edinburgh: The University of Edinburgh, 2007:148–57.
53 Lomax N, Wheeler H, Anaraki S, Anderson H, Goh B. Management of a syphilis outbreak in street sex workers in east London. Sex Transm Infect 2006; 82: 437–38.
54 Surratt HL, Inciardi JA. HIV risk, seropositivity and predictors of infection among homeless and non-homeless women sex workers in Miami, Florida, USA. AIDS Care 2004; 16: 594–604.
55 Cohan D, Lutnick A, Davidson P, et al. Sex worker health: San Francisco style. Sex Transm Infect 2006; 82: 418–22.
56 Kurtz SP, Surratt HL, Kiley MC, Inciardi JA. Barriers to health and social services for street-based sex workers. J Health Care Poor Underserved 2005; 16: 345–61.
57 Williamson PR, Altman DG, Bagley H, et al. The COMET handbook: version 1.0. Trials 2017; 18 (suppl 3): 280.
58 Potterat JJ, Brewer DD, Muth SQ, et al. Mortality in a long-term open cohort of prostitute women. Am J Epidemiol 2004; 159: 778–85.
Copyright © 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license.
- Interventions to improve health and the determinants of health among sex workers in high-income countries: a systematic review
- Introduction
- Methods
- Search strategy and selection criteria
- Eligibility criteria
- Data extraction and quality assessment
- Data synthesis
- Results
- Overview
- Education and empowerment
- Drug treatment
- Sexual and reproductive health care
- Other health care
- Welfare
- Discussion
- Acknowledgments
- References
