C HANGES IN STI SERVICES FOLLOWING INTERVENTION

CHANGES IN STI SERVICES FOLLOWING A COMMUNITY

BASED STI-INTERVENTION PROJECT IN CAMBODIA

Nigel O’ Farrell1 , Phal Sano 2 , Seng Sopheap 2 , Lon Say Heng 2 , Ly Penh Sun 2 ,
Mean Chhi Vun 2 , Knut Fylkesnes 3 and Peter Godwin 2

1
Pasteur Suite, Ealing Hospital, London, and London School of Hygiene and Tropical
Medicine; 2 National Center for HIV/ AIDS, Dermatology and STDs (NCHADS), Ministry of Health,
Phnom Penh, Cambodia; 3 Center for International Health, University of Bergen, Norway

Abstract. The aim of this study was to assess changes in sexually transmitted infections (STI)
related care following a STI project with a particular focus on registered brothel-based (direct)
female sex workers (DFSWs) in four border provinces of Cambodia. A survey of health care
facilities providing STI care was undertaken and the results compared with a baseline survey
done two years previously. The main components of the project were: renovation of the STI
clinics, STI training, formation of mobile teams, provision of STI drugs, and the introduction of
basic laboratory tests at STI clinics. Interviews were held with health care providers and STI
patients and a manual check was made of the STI register and special forms for DFSWs.
Clinical management of STI cases was assessed for DFSWs, women with vaginal discharge
and men with urethral discharge. Advice given to clients about condom use, partner notifica-
tion and STI education was assessed and availability of STI drugs was reviewed. STI clinic
attendance by DFSWs each month increased from 72% (296/ 412) to 93% (459/ 496). The
proportion of DFSWs diagnosed with presumed STIs decreased from 86.5% (256/296) to 25.5%
(117/ 459) and cervicitis from 32.8% (135/ 412) to 12.6% (58/ 459). The percentage of men
attending STI clinics decreased from 26.9% (251/ 933) to 9.4% (102/ 1,080). The proportion of
presumed STI cases/ all cases attending health centers decreased from 7.0% (934/ 13,177) to
4.3% (739/ 17,224). The introduction of laboratory tests coincided with a marked reduction in
DFSWs diagnosed with cervicitis. Further validation studies are required to determine whether
this reduction was accompanied by a real decrease in gonorrhea and chlamydia.

INTRODUCTION b ased ) FSW (IFSW) t est ed H IV p o sit ive

(NCHADS, 2003a). High STI rates were also
The extent of the spread of human im- id ent ified init ially in a 1 9 9 6 survey w it h
m uno d efic ienc y virus (HIV) in Cam b o d ia prevalences in DFSW of gonorrhea of 35% and
makes it one of the worst affected countries chlamydia in 22% although these had reduced
in Asia (UNAIDS/WHO, 2006). Particularly high to 14% and 12%, respectively by 2001 (Ryan
HIV rates are found in female sex workers et al, 1998; NCHADS, 2001). However, a re-
(FSW). In 2003, 21% of direct (brothel-based) cent survey in 2005 showed similar levels to
FSW (DFSW) and 12% of indirect (non-brothel- 2001 with prevalences of gonorrhea of 13%
and chlamydia of 14% (NCHADS et al, 2005).
Correspondence: Dr Nigel O’Farrell, Pasteur Suite,
Ealing Hospital, Uxbridge Road, London UB1 3HW, These high STI levels have prevailed despite
UK. a well established 100% condom use cam-
Tel: 44 208 967 5746/ 5760; Fax: 44 208 967 5677 paign targeting DFSWs that started in the
E-mail: nigel.o’ farrell@eht.nhs.uk, nigel.ofarrell@ late 1990’s (WHO, 2003). An essential com-
lshtm.ac.uk ponent of this program has been high quality

Vol 39 No. 5 September 2008 867

 SOUTHEAST ASIAN J TROP M ED PUBLIC H EALTH

STI services for DFSW with a focus on high STI management, in particular, the availability
risk and vulnerable groups (Saphonn et al, of STI drugs and condom supply through re-
2004; Godwin et al, 2006). view of clinic supply logs.
The STI-related activities described in this Data collection
report were initiated as part of a larger com- Data were collected in November 2004
munity project to limit HIV in Cambodia, Lao at the same 6 STI clinics and the same 25
PDR and Vietnam (www.jfpr-hiv.org). In Cam- health centers with an integrated STI service
bodia the project was implemented in 4 bor- as collected in 2002. Data were also collected
der provinces with the aim of reducing STIs in at in 7 clinics and 6 health centers without an
FSWs and improving STI services in the gen- integrated STI service (Sano et al, 2004). The
eral population. An initial project baseline sur- following methods of data collection were
vey was done in 2002 (Sano et al, 2004). The used: 1) documentary method by checking the
aim of this study was to compare the baseline STI register, the standard medical history form
results of the survey in 2002 to evaluate STI in STI clinics that were completed whenever
case management with this study in 2004, to DFSWs registered under the 100% condom
both assess changes brought about by the use program attend, and the consultation
project and identify further measures to im- record book in the health center that records
prove STI control. total attendances; 2) semi-structured inter-
views with healthcare managers and provid-
MATERIALS AND METHODS ers; 3) observation of the performance of the
health center with an individual patient. Ques-
The main objective of the study was to tions were asked about the various aspects
detect differences in the quality of STI services of STI-related knowledge and exposure to a
following the intervention through assessment condom demonstration at a STI clinic. Data
of the following: were collected in a similar manner to that in
2002 and included 1) a national interview team
1) Assess the strengths and weaknesses
of 6 interviewers from the National Center for
in the availability of STI and reproductive tract
AIDS, Dermatology and STDs (NCHADS); 4
infection management and identify the propor-
from the STI unit (2 groups in alternation) and
tion of registered DFSWs attending as a part
2 from the technical bureau; 2) two provincial
of the 100% condom use campaign among
coordinators per province for the field survey.
sex establishments in 4 border provinces:
Battambang, Svay Rieng, Prey Veng and Koh Interviews were conducted with the fol-
Kong. lowing 1) STI service supporters: four provin-
cial AIDS office (PAO) managers, 11 opera-
2) Assess the clinical management of
tional district (OD) staff responsible for drugs,
vaginal discharge and cervicitis in women and
2) STI care providers: six STI clinic managers,
urethral discharge in men through review of
11 reproductive health (RH) staff working in
clinic records and adherence to National STI
the gynecology ward and 32 health center
guidelines.
(HC) staff, 3) 44 STI clients selected by con-
3) Assess the proportion of health facility venience sampling during the assessment visit
attendees reporting with STI complaints that for total of 108 interviewees. Some of the staff
receive advice about condom use and part- may have been included in the first survey but
ner notification as assessed during an exit this was very unlikely for the STI clients. In-
survey. formed verbal consent was obtained from all
4) Assess the support components for subjects. Six questionnaires were developed

868 Vol 39 No. 5 September 2008

 C HANGES IN STI SERVICES FOLLOWING INTERVENTION

for the following groups: PAO, STI clinic man- NCHADS.

ager, OD chief, RH staff, HC staff responsible Statistical analysis
for STIs and STI clients. No interviewee re-
Microsoft Excel was used for data entry
fused study participation.
and comparison between proportions was
What was implemented? done using the chi-square test. A p-value
The main components of the project were ≤0.05 was considered statistically significant.
as follows: renovation and re-equipping of STI
clinics, STI training and re-training of health RESULTS
care workers from both STI clinics and other
STI clinics
health centers, provision of STI drugs, re-
agents and other consumables, formation of The overall numbers of patients increased
mobile STI teams, advocacy meetings for lo- by 15.8% from 933 to 1,080 (Table 1). Women
cal authorities, staff salary supplements and attending the program increased from 38%
supervision. In addition, FSW-specific com- (352) to 42% (454), p= 0.049 whilst the num-
ponents comprised local promotion cam- ber of men decreased from 26.9% (251) to
paigns and events, information, education and 9.4% (102), p< 0.001. The number of FSW
communication materials and promotion of the attending overall increased from 35.0% (330)
100% condom use program. to 49.0% (529), p < 0.001 and registered
DFSW increased from 71.8% (296) to 93%
Laboratory services were introduced at the
(459), p<0.001.
STI clinics. This involved the use of wet mount
microscopy for the detection of Trichomonas Presumed STIs in attendees, that is cases
vaginalis, clue cells and yeast, a Gram’s stain diagnosed syndromically or through a risk-
for vaginal discharge specimens to identify can- based algorithm, decreased significantly in
dida infections and clue cells, a methylene blue FSW from 86.5% (296) to 25.5% (117), p <
stain of cervical secretions for the detection of 0.001. This reflects a significant reduction in
white cells ≥10 per high power field and a rapid those diagnosed with cervicitis as the propor-
plasma reagin test for syphilis. Previously the tion of cervicitis cases in those diagnosed with
criteria for diagnosing cervicitis (covering both a STI was similar in both surveys, whilst the
gonorrhea and chlamydia infections) was com- percentage of cervicitis cases decreased from
prised one or more of the following: yellow dis- 32.8% (135/412) in 2002 to 12.6% (58/459)
charge from the cervix, cervical erosion or in 2004.
bleeding, yellow secretions on cervical swab The quality of care decreased slightly
or deep pain on bimanual examination. The between 2002 and 2004 but this was not sta-
detection of white cells ≥10 per high power field tistically significant (Table 1). In 2002 all ure-
from a cervical smear was introduced as an thritis cases were diagnosed and treated cor-
additional single diagnostic indicator for the rectly whilst in 2004 93% (80/86) were diag-
second survey. No swab tests were done for nosed c orrec tly and 95.3% (82/ 86) w ere
men. In health centers without STI care, no STI treated correctly. The proportion of cases di-
tests were carried out but women with vaginal agnosed correctly per protocol as cervicitis
discharge were treated for Candida and bac- decreased from 99.3% (134/135) to 93.1%
terial vaginosis only. (54/58) though all were treated correctly.
The study protocol and questionnaires Other health centers
were reviewed and approved by the Ministry In Health Centers with STI care (Table 2),
of Health of Cambodia and implemented by all presumed STI cases decreased from 7.0%

Vol 39 No. 5 September 2008 869

870

Table 1
Female sex worker attendances, STI diagnoses, and treatment in 6 provincial STI clinics in 2002 and 2004.

STI attendances, diagnoses and treatment Year p-value

2002 2004

Average monthly attendances in last 3 months

SOUTHEAST ASIAN J TROP M ED PUBLIC H EALTH

Total attendances (men and women) 933 1,080
Females in general population 352 (38%) 454 (42%) 0.049
Men 251 (26.9%) 102 (9.4%) <0.001
Sex workers 330 (35%) 529 (49%) <0.001
All new STI cases/Total attendances last month 345/933 (37%) 455/1,080 (42.1%) 0.018
Urethral discharge/ All STI cases in last month 81/345 (23.5%) 86/455 (18.9%) 0.115
Urethral discharge diagnosed correctly 81/81 (100%) 80/86 (93.0%) 0.825
Urethral discharge treated correctly 81/81 (100%) 82/86 (95.3%) 0.912
Registered female sex workers 412 496
Cases in registered female sex workers in last month
Attended for check-up 296/412 (71.8%) 459/496 (93%) <0.001
STI cases in FSWs /FSWs visiting for check-up 256/296 (86.5%) 117/459 (25.5%) <0.001
STI cases in FSWs/STI cases in FSWs visiting for check-up 93/256 (36.3%) 48/117 (41%) 0.385
Vol 39 No. 5 September 2008

Cervicitis cases in FSWs/ STI cases in FSWs visiting for check-up 135/256 (52.7%) 58/117 (49.6%) 0.571
Cervicitis cases in FSWs in first visit / STI cases in FSWs visiting for check-up 61/256 (23.8%) 27/117 (23.1%) 0.874
Cervicitis cases in FSWs diagnosed correctly 134/135 (99.3%) 54/58 (93.1%) 0.822
Cervicitis cases in FSWs treated correctly 134/134 (100%) 54/54 (100%) 1.0
Vol 39 No. 5 September 2008

Table 2
STI attendances at health centers with and without STI care in 2002 and 2004 in the last month.
STI issue Health centers with STI service Health centers without STI service

2002 (n=25) 2004 (n=26) p-value 2002 (n=7) 2004 (n=6) p-value

STI cases/Total clients for all conditions 934/13,177 (7.0%) 739/17,224 (4.3%) <0.001 286/3,253 (8.9%) 229/3,170 (7.2%) 0.021

C HANGES
New STI cases/Total new clients for all conditions 763/10,293 (7.4%) 575/12,427 (4.6%) <0.001 205/2,896 (7.1%) 140/2,686 (5.2%) 0.004
Follow-up STI cases/Total follow-up clients that 171/2,884 (5.6%) 164/4,167 (3.9%) <0.001 81/357 (22.7%) 89/484 (18.4%) 0.125
visited HCs for all conditions

IN
STI SERVICES FOLLOWING INTERVENTION
Female STI cases/Total STI cases that visited HC 701/934 (75.1%) 624/739 (84.4%) <0.001 234/286 (81.8%) 227/229 (99.1%) <0.001
New female STI cases/Total new female STI cases 552/701 (78.7%) 482/575 (83.8%) 0.021 155/234 (66.2%) 138/140 (98.6%) <0.001
that visited HC
New vaginal discharge cases/Total new female 469/552 (85%) 435/482 (90.2%) 0.011 142/155 (91.6%) 126/138 (91.3%) 0.925
STI cases
New cases of cervicitis/Total new cases of vaginal 305/469 (65%) 271/435 (62.2%) 0.393 74/142 (52.1%) 59/126 (46.7%) 0.388
discharge
New urethral discharge cases/Total new STI cases 137/763 (18%) 74/575 (12.9%) 0.012 39/205 (19%) 2/140 (1.4%) <0.001
that visited HC
New cases of cervicitis in vaginal discharge 144/305 (47.2%) 174/271 (64.2%) <0.001
diagnosed correctly/Total new cases of cervicitis
New cases of cervicitis in vaginal discharge treated 127/144 (88.2%) 146/174 (83.9%) 0.275
correctly/Total new cases of cervicitis diagnosed
correctly
Urethral discharge cases diagnosed correctly 82/137 (59.9%) 72/74 (97.3%) 0.030
Urethral discharge cases treated correctly 54/82 (65.8%) 67/72 (93.1%) 0.182
871
 SOUTHEAST ASIAN J TROP M ED PUBLIC H EALTH

(934/13,177) in 2002 to 4.3% (739/17,224) increased significantly from 35.1% and 46%,
p< 0.001) in 2004. Similar significant de- respectively, to 100% in both, as did those
creases were also seen in both new (first visit) witnessing a condom demonstration.
and follow-up cases. Overall there were sig- In health centers with integrated services,
nificant increases in all female and new female all the parameters of STI-related knowledge
presumed STI cases and the proportion of increased, except for partner notification which
vaginal d isc harge c ases/ all STI c ases in remained about the same. Knowledge of HIV
women. New cases of cervicitis in those with transmission modes and treatment compli-
vaginal discharge diagnosed correctly/ total ance increased significantly.
new cervicitis cases, increased significantly.
Urethral discharge cases decreased signifi- DISCUSSION
cantly from 18.0% (137/763) to 12.9% (74/
575), (p= 0.012). There was a reduction in Coverage of registered DFSW increased
the number of health centers working in both from 72% (296/412) to 93% (459/496). This
the morning and afternoon (66% to 33%). high coverage was facilitated by the efficient
In health centers without STI care, there implementation of the 100% condom use pro-
were significant decreases in both all and gram that targets registered FSWs. However,
new presumed STI cases (Table 2) whilst all this improvement was also associated with a
female and new female STI cases increased significant decrease in the number of cervicitis
significantly. Urethral discharge cases de- cases diagnosed from 32.8% (135/ 412) to
creased significantly from 19% (39/ 205) to 12.6% (58/459) following the introduction of
1.4% (2/ 140) (p<0.001). basic STI laboratory tests. Whilst our survey
did not undertake molecular STI testing it is in-
Drug stock outs, defined as out of stock
teresting to note that in the National STI survey
for longer than 1 week in the last 6 months,
a few months later, the respective prevalences
were reported in 12/25 of Health Centers with
of gonorrhea and/or chlamydia in DFSWs in two
an integrated STI service in 2004 compared
of the four provinces covered was 10% in Prey
with 18/25 in 2002.
Veng and 25% in Battambang (Guy Morineau,
Exit survey personal communication) whilst the number of
In STI clinics (Table 3), the proportion of cervicitis cases in FSW in these provinces were
subjects w ith a correct know ledge of HIV 19.3% and 3.8%, respectively (data not shown).
transmission modes and prevention methods This finding may be relevant in explaining the

Table 3
STI education prevention and care in STI clients assessed at exit interview in 2002 and 2004.

STI Clinics Health centers with integrated STI service

STI-related knowledge,
and exposure to condom 2002 2004 p-value 2002 2004 p-value
demonstration in STI clients N = 37 N = 13 N=7 N = 25

STI transmission mode 13 (35.1%) 13 (100%) <0.001 0 (0%) 14 (56%) 0.010

Prevention methods 17 (46%) 13 (100%) 0.001 1 (16.7%) 12 (48%) 0.195
Compliance 35 (94.6%) 13 (100%) 1.000 5 (71.4%) 25 (100%) 0.042
Partner notification 24 (64.9%) 11 (85%) 0.294 4 (57.1%) 14 (56%) 1.000
Condom demonstration 16 (43.2%) 11 (85%) 0.012 1 (14.4%) 15 (60%) 0.083

872 Vol 39 No. 5 September 2008

 C HANGES IN STI SERVICES FOLLOWING INTERVENTION

lack of change in national STI prevalence rates Drug stock-outs were fewer in 2004 than
recently, in that the decrease in cases diag- 2002. This improvement was mainly due to a
nosed as cervicitis, according to the algorithm new focus on drug-supply logistics with addi-
using basic laboratory results, resulted in a de- tional qualified personnel which has enabled
crease in sensitivity and a higher proportion of the roll-out of antiretrovirals in the country.
true gonorrhea/ chlamydia cases going un- As assessed by the exit survey, STI edu-
treated. Our survey also showed that the mean cation and knowledge was good although the
age of DFSWs appears to be increasing. In number of STI clients was small. There were
2001 the mean age was 21.6 but by 2005, this some difficulties with contact tracing but these
had increased to 25 (NCHADS, 2001; NCHADS may reflect the low predictive values for gon-
et al, 2006). This increase in age would be ex- orrhea and chlamydia amongst partners with
pected to be associated with a lower rate of vaginal discharge.
cervical STIs, given that the prevalence of cer-
vical ectopy, a complication for gonorrhea and One factor introduced to Cambodia was
chlamydia, tends to be lower in this older age- salary supplements for STI clinic staff and
group. some provincial AIDS office staff to facilitate
improved quality of care. This model has been
Overall there was a significant reduction
expanded and a full performance-based sal-
in the p rop ortion of p resumed STI c ases
ary incentive scheme supported and sustained
amongst all cases attending health centers in
by DFID has been developed with the even-
this c ommunity w hen c omp ared w ith the
tual aim of the government being to take over
baseline survey. The significant reduction in
these payments (Viney, 2007).
the proportion of male attendees at the clin-
ics suggests a real decrease in the commu- A weakness of the project was its limited
nity burden of STIs. The last Cambodian Be- evaluation of IFSW, a large group that may
havioral Surveillance Survey showed a con- have a high incidence of STIs (Kim et al, 2005;
sist enc y in healt h c are seeking b ehavior Vandepitte et al, 2006). The outreach teams
among men with urethral discharge in the mili- did see many IFSW in the field and encour-
tary, police and taxi drivers; 18-21% attended aged them to visit the STI clinics. Overall, the
public clinics (NCHADS, 2003b). During the project saw 13,673 FSW and 20,220 from the
project there were no apparent changes in the general population (NCHADS, 2005).
provision of STI care reported by the PAO This study also showed that whilst moni-
managers involving other providers, such as toring some process indicators may be an ef-
pharmacies, private doctors or traditional heal- fective way of assessing interventions, actual
ers. This is particularly relevant given the im- measurement of STI levels is still required to
portant of sexual bridging of Cambodian men help interpret changes in STI-related atten-
in spread STIs and HIV and the cost effective- dances. Treatment of cervicitis for all first-time
ness of male intervensions (Gorbach et al, attendees is appropriate given the high mo-
2000; Godwin et al, 2005). bility of FSW in Cambodia (Sopheab et al,
It is unclear why there was an increase in 2006) and the high risk for STI associated with
women with STI attending the clinics. It is not young age. One way of tackling these high
uncommon for women in Cambodia to com- rates of STIs may be the introduction of peri-
plain of vaginal discharge even though they may odic presumptive treatment (PPT), at least for
be at low risk for STIs. We believe that more the short term. This method has clear advan-
women probably attended the STI clinics be- tages over using Gram’s stain and basic mi-
cause of the good reputation of the clinics. croscopy (Thuong et al, 2007) which may not

Vol 39 No. 5 September 2008 873

 SOUTHEAST ASIAN J TROP M ED PUBLIC H EALTH

be available or performed poorly; PPT treat- try of Health, 2001. [Cited 2008 Jan 22]. Avail-
ment in high-risk groups may be a particularly able from: URL: www.nchads.org/docs/publi-
cost effective STI control strategy in Cambo- cations/sss/cambodiastibook.pdf
dia (O’Farrell et al, 2006). NCHADS. HIV sentinel results 2003. Results, trends
and estimates. 2003a. [Cited 2008 Jan 22].
Our study sheds doubt on the effective-
Available from: URL: Nchads.org/ docs/ publi-
ness of basic laboratory tests employed to di-
cation/ dissemination/ HSS2003.ENG.pdf
agnose cervicitis due to gonorrhea and/ or
NCHADS.Cambodia 2003 behavioural surveillance
chlamydia. Further monitoring should focus on
survey. 2003b. [Cited 2008 Jan 22]. Available
the proportion of FSW treated for cervicitis at
from : URL: Nc had s.org/ d oc s/ p ub lic at ion/
each attendance. Supervision should be inten-
BSS/ BSS_2003.pdf
sified if diagnostic criteria change (O’Farrell,
NCHADS. Cambodia. Final report and analysis of
2007) and future surveys should be undertaken
project impact 2001-2004, Community Action
using STI molecular diagnostic techniques. for preventing HIV/ AIDS, 2005.
NCHADS, FHI, ADB, CDC Gap. Cambodia STI sur-
ACKNOWLEDGEMENTS vey 2005. Key risk behaviours and STI preva-
lence. 2006. [Cited 2008 Jan 22]. Available
We wish to thank Linda Morison of the
from: URL: w w w.nchads.org/ docs/ publica-
London School of Hygiene and Tropical Medi- t io ns/ d issem inat io n/ SSS% 2 0 2 0 0 5 % .
cine for her statistical assistance, Dr Francois 20ENG.pdf
Crabbé of the Institute of Tropical Medicine,
O’ Farrell N , C rab b é F, Vic hea O, Sap ho nn V,
Antwerp, and Indu Bhushan and Vincent de Fylkesnes K, Godwin P. Periodic presumptive
Wit of the Asian Development Bank. This treatment for cervicitis in female sex workers
project was financed by the Japan Fund for in Cambodia: a cheap, sustainable HIV pre-
Poverty Reduction-project 9006, administered vent io n st rat eg y. [Ab st rac t M OPE0 5 7 7 ].
by the Asian Development Bank. Toronto: AIDS Conference, August 2006.
O’ Farrell N. Program issues in delivering targeted
REFERENCES STI services through the public sector in the
Greater Mekong Region. Southeast Asian J
Godwin P, O’ Farrell N, Fylkesnes K. JFPR- 9006- Trop Med Public Health 2007; 38: 328-38.
Community Action for preventing HIV/ AIDS,
Ryan CA, Vathiny OV, Gorbach PM, et al. Explosive
Cambodia, Laos, Vietnam. A summary of the
spread of HIV-1 and sexually transmitted dis-
p rojec t imp lementation and results. 2005.
eases in Cambodia. Lancet 1998; 351: 1175.
[Cited 2008 Jan 22]. Availab le from : URL:
Sano P, Sopheap S, Sun LP, Vun MC, Godwin P,
w w w.jfp r- hiv.o rg / finalrep o rt _ADB _J FPR_
O’Farrell N. An evaluation of sexually transmit-
REG_9006.pdf
ted infection case management in heath facili-
Godwin P, O’ Farrell N, Fylkesnes K, Misra S. Five
ties in 4 border provinces of Cambodia. Sex
myths about the HIV epidemic in Asia. Plos
Transm Dis 2004; 31: 713-8.
Med 2006; 3: e426.
Saphonn V, Sopheab H, Sun LP, et al. Current HIV/
Gorbach PM, Sopheab H, Phalla T, et al. Sexual AIDS/ STI epidemic: intervention programs in
bridging by Cambodian men. Sex Transm Dis Cambodia, 1993-2003. AIDS Educ Prev 2004;
2000; 27: 320-6. 16 (suppl A): 64-77.
Kim AA, Sun LP, Chhorvann C, et al. High preva- Sopheab H, Fylkesnes K, Vun MC, O’Farrell N. HIV-
lence of HIV and sexually transmitted infec- related risk behaviours in Cambodia and ef-
tions among indirect sex workers in Cambo- fects of mobility. J AIDS 2006; 41: 81-6.
dia. Sex Transm Dis 2005; 32: 745-51. Thuong NT, Nghia KV, Hau TP, et al. Impact of a
NCHADS. Cambodian STI prevalence survey. Minis- c om m unity STI/ HIV intervention p rojec t in

874 Vol 39 No. 5 September 2008

 C HANGES IN STI SERVICES FOLLOWING INTERVENTION

female sex workers in five border provinces Sex Transm Infect 2006; 82 (suppl 111): 18-25.
of Vietnam. Sex Transm Infect 2007; 83: 376- Viney S. Report on performance based salary in-
82. centives to government staff working on HIV/
UNAIDS/ WHO. AIDS epidemic update: December AIDS programmes in Cambodia. 2007.
2006. [Cited 2008 Jan 22]. Available from: WHO. WHO Guidelines for scaling up the 100%
URL : 2 0 0 6 h t t p : / / d at a. u n aid s . o rg / p u b / c ond om use p rogramme: exp erienc e from
EpiReport/ 2006/ 2006_EpiUpdate_en.pdf Cambodia. 2003. Manila: WHO Regional Of-
Vandepitte J, Lyerla R, Dallabetta G, Crabbé F, Alary fice for the Western Pacific. [Cited 2008 Jan
M, Buve A. Estimates of the number of female 22]. Available from: URL: www.wpro.who.int/
sex workers in different regions of the world. publications/ pub_9290610425.htm

Vol 39 No. 5 September 2008 875