UBC Faculty Research and Publications

A dose-response relationship between exposure to a large-scale HIV preventive intervention and consistent… Deering, Kathleen N; Boily, Marie-Claude; Lowndes, Catherine M; Shoveller, Jean; Tyndall, Mark W; Vickerman, Peter; Bradley, Jan; Gurav, Kaveri; Pickles, Michael; Moses, Stephen; Ramesh, Banadakoppa M; Washington, Reynold; Rajaram, S; Alary, Michel Dec 29, 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12889_2011_Article_3671.pdf [ 625.29kB ]
Metadata
JSON: 52383-1.0223644.json
JSON-LD: 52383-1.0223644-ld.json
RDF/XML (Pretty): 52383-1.0223644-rdf.xml
RDF/JSON: 52383-1.0223644-rdf.json
Turtle: 52383-1.0223644-turtle.txt
N-Triples: 52383-1.0223644-rdf-ntriples.txt
Original Record: 52383-1.0223644-source.json
Full Text
52383-1.0223644-fulltext.txt
Citation
52383-1.0223644.ris

Full Text

REVIEW Open AccessA dose-response relationship between exposureto a large-scale HIV preventive intervention andconsistent condom use with different sexualpartners of female sex workers in southern IndiaKathleen N Deering1*, Marie-Claude Boily2†, Catherine M Lowndes3†, Jean Shoveller1†, Mark W Tyndall4†,Peter Vickerman5†, Jan Bradley6†, Kaveri Gurav7†, Michael Pickles2†, Stephen Moses8,9†, Banadakoppa M Ramesh7,9†,Reynold Washington7†, S Rajaram7†, Michel Alary6,10,11†AbstractBackground: The Avahan Initiative, a large-scale HIV preventive intervention targeted to high-risk populationsincluding female sex workers (FSWs), was initiated in 2003 in six high-prevalence states in India, includingKarnataka. This study assessed if intervention exposure was associated with condom use with FSWs’ sexual partners,including a dose-response relationship.Methods: Data were from a cross-sectional study (2006-07) of 775 FSWs in three districts in Karnataka. Surveymethods accounted for the complex cluster sampling design. Bivariate and multivariable logistic regression wasused to separately model the relationships between each of five intervention exposure variables and five outcomesfor consistent condom use (CCU= always versus frequently/sometimes/never) with different sex partners, includingwith: all clients; occasional clients; most recent repeat client; most recent non-paying partner; and the husband orcohabiting partner. Linear tests for trends were conducted for three continuous intervention exposure variables.Results: FSWs reported highest CCU with all clients (81.7%); CCU was lowest with FSWs’ husband or cohabitingpartner (9.6%). In multivariable analysis, the odds of CCU with all clients and with occasional clients were 6.3-fold[95% confidence intervals, CIs: 2.8-14.5] and 2.3-fold [95% CIs: 1.4-4.1] higher among FSWs contacted byintervention staff and 4.9-fold [95% CIs: 2.6-9.3] and 2.3-fold [95% CIs: 1.3-4.1] higher among those who everobserved a condom demonstration by staff, respectively, compared to those who had not. A significant dose-response relationship existed between each of these CCU outcomes and increased duration since first contactedby staff (P=0.001; P=0.006) and numbers of condom demonstrations witnessed (P=0.004; P=0.026); a dose-responserelationship was also observed between condom use with all clients and number of times contacted by staff(P=0.047). Intervention exposure was not associated with higher odds of CCU with the most recent repeat client,most recent non-paying partner or with the husband or cohabiting partner.Conclusion: Study findings suggest that exposure to a large-scale HIV intervention for FSWs was associated withincreased CCU with commercial clients. Moreover, there were dose-response relationships between CCU withclients and increased duration since first contacted by staff, times contacted by staff and number of condomdemonstrations. Additional program effort is required to increase condom use with non-commercial partners.* Correspondence: kdeering@cfenet.ubc.ca† Contributed equally1Division of AIDS, Department of Medicine, Faculty of Medicine, University ofBritish Columbia, Vancouver, CanadaFull list of author information is available at the end of the articleDeering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8© 2011 Deering et al; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.BackgroundSex work-related harms are linked inextricably to thesocial, economic, policy, and physical environments ofsex workers. Individual behaviour (high- or low-risk)both shapes and is shaped by individual and environ-mental factors [1,2]. There has thus been increasingrecognition of the importance of using structural andcommunity-level strategies that modify sex work envir-onments to reduce risk and promote health among sexworkers and their clients, and in particular, improvecondom use with sex partners [3-6]. Notably, inresponse to high rates of HIV and sexually transmittedinfections (STIs) among female sex workers (FSWs) inthe early 1990s, several countries in east Asia instituteda 100% condom use campaign intended to increasesocial acceptance of condoms, influence men to agree touse condoms and empower FSWs to demand condomuse with clients, as well as increase access to STI testingand treatment. This programme is thought to have con-tributed to dramatic declines in STIs and HIV in Thai-land and Cambodia, as well as influence similarcampaigns across Asia [7-9]. The Sonagachi Project inKolkata, India, implemented a community empower-ment model for FSWs that framed health risks to sexworkers as occupational hazards, focusing on addressingcommunity- and individual-level factors influencing riskfor HIV. Subsequently, large increases in condom usehave been observed and HIV prevalence remains low inFSWs associated with the Sonagachi Project [10,11].Another large-scale intervention designed to reduceHIV infection rates among groups with high HIV risk(FSWs; men-who-have-sex-with-men; injection drugusers) and groups that bridge high- and low-risk groups(clients of FSWs) is Avahan, the India AIDS Initiative[12,13], which began in 2003 in the six states with thehighest HIV prevalence in India. Using communityinvolvement and mobilization strategies, combined withcondom promotion and increased STI clinical servicesamong these populations, the ongoing Avahan AIDSInitiative addresses proximal and distal determinants ofrisk. The Avahan AIDS Initiative aims to increase con-dom use among groups at high risk for HIV by modify-ing their environments to enable individuals to usecondoms [14]. For FSWs, this is achieved through acombination of approaches. Avahan includes peer-ledoutreach to increase awareness of condoms and abilityto negotiate condom use with clients [15] and efforts toincrease the availability of and access to condoms andSTI testing and treatment centres [12]. The programalso includes actions to improve community mobiliza-tion and involvement. FSWs have played importantroles in mapping local hotspots, informing outreachplans, developing peer networks in communities andparticipating in training and implementation of Avahansurveys [16,17]. The program has also supported thedevelopment and operation of safer sex work spaces,including sex work drop-in centres and collectives,where women can rest safely, take classes (e.g., literacytraining) and interact with staff and other FSWs [18-20].Legal empowerment training has also been offered to25,000 FSWs across Karnataka state, to improve legalliteracy and inform FSWs about their legal rights [19].The evaluation of this large-scale intervention remainschallenging, as is the case for many similar evaluationefforts where conventional methods (e.g., randomizedcontrol trials of communities) are unethical and/orimpractical to implement [13,21]. A multi-pronged eva-luation framework is necessary to gain an overall under-standing of an intervention’s impact [21]. This includesan examination of programmatic (e.g., numbers of peereducators, clinics or services to meet the population’sneeds) and health indicators (e.g., increases in condomuse, decreases in HIV or STI incidence). The consis-tency of study results from a combination of studydesigns, including transmission dynamics modelling (e.g., testing hypotheses while taking into account uncer-tainty in parameter estimates), cost-effectiveness analy-sis, surveillance and epidemiological approaches, cantogether provide a stronger understanding of the effec-tiveness of the intervention [22].As part of this comprehensive evaluation framework,the objective of the current analysis was to determine ifthe Avahan AIDS Initiative had an impact on condomuse amongst FSWs in urban areas of three districts inKarnataka State, India. HIV prevalence among FSWs was12.7% in Bangalore district, 15.7% in Bellary and 33.9% inBelgaum in the mid-2000s [23]. Specifically, we assessedwhether five variables measuring intervention exposurewere associated with consistent condom use (CCU) (i.e.100%) among FSWs with: (1) all clients on the mostrecent day worked; (2) their current occasional clients (i.e., clients who FSWs are not familiar with and who visitFSWs once); (3) their most recent repeat client (i.e., regu-lar clients who FSWs are familiar with and who visitFSWs more than once); (4) their most recent non-payingpartner and (5) their husband or cohabiting partner.MethodsStudy design and samplingDuring 2006-07, in-depth face-to-face interviews (Spe-cial Behavioural Surveys, SBS) were conducted with 775FSWs in three districts in Karnataka state, located insouthern India. A probability sampling method wasemployed, using time-location cluster sampling withnormalized weights calculated to account for the com-plex sampling design. Sampling methods were similar tothose reported by Ramesh et al [23] for other studiescarried out among FSWs in Karnataka state.Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 2 of 14Survey organization and methodsThe surveys were implemented by the CHARME-Indiaproject in collaboration with the Institute of PopulationHealth and Clinical Research (IPHCR), St John’s MedicalCollege, and the Karnataka Health Promotion Trust(KHPT), Bangalore, India, the Centre hospitalier affiliéuniversitaire de Québec (CHA), Québec, Canada, andthe University of Manitoba, Winnipeg, Canada. The sur-veys were administered face-to-face by trained inter-viewers in the local language (Kannada) and wereconducted anonymously, with no names or personalidentifiers recorded. Ethics approval was attained fromthe CHA and the University of Manitoba as well as St.John’s Medical College.OutcomesThe first outcome, CCU with all clients (includingboth occasional and repeat) during all instances of sex-ual intercourse in the most recent day worked wasderived by dividing the reported number of instancesof sexual intercourse in which condoms were used bythe reported total number of instances of sexual inter-course in the most recent day worked. This was usedto create a dichotomized variable of 100% versus<100% of instances of sexual intercourse in which con-doms were used. The remaining four outcomesdescribed CCU with FSWs’ different sexual partners,including: commercial sex clients (their current occa-sional clients; their most recent repeat client); andnon-commercial partners (their most recent non-pay-ing partner who was neither a husband nor the maincohabiting partner; and their husband or main cohabit-ing partner (if they had one)). These outcomes werederived from survey items about general condom fre-quency with each type of partner (e.g. “How often doyou use condoms with <this partner>?”). Condom usewas considered to be CCU with their partners, if theyanswered ‘always’, as opposed to ‘often’, ‘sometimes’ or‘never’.Explanatory factorsWe examined five variables measuring exposure to theintervention, including: if FSWs had been contacted byintervention staff; if FSWs had been given condoms byintervention staff; the duration of time since contactedby intervention staff (years), which was specific to eachdistrict and limited to the total number of years womencould have been exposed to the intervention (the year/month of the start of the intervention subtracted fromthe year/month of the survey – 1.5-2.5 years); the num-ber of times in the past month FSWs had been con-tacted by intervention staff; and the number of condomdemonstrations by intervention staff that FSWs hadseen in the past month.For each model, we adjusted for social and environ-mental factors that may influence condom use. Socialfactors included age, marital status (married versusunmarried, including those of the Devadasi tradition, aform of temple-based sex work whereby women arededicated through marriage to gods or goddesses)[24-26], age at first sex, age at first sex work and dura-tion of sex work. Environmental factors included districtof residence, education (literacy) and having sex work assole income (no other paid work versus any, includingnon-agricultural labour, petty business, maid servant,agricultural labour, handicrafts and other). It alsoincluded FSWs’ working environment, which was repre-sented by the type of solicitation (independent orthrough a middleman/pimp) and the place of solicitationof clients of FSWs, which was grouped into three cate-gories: home-based (home; rented room), brothel-based(lodge; dabha [road-side lodge-type establishment];brothel); and public-places-based (vehicle; bar/nightclub; public places, such as bus stops, train stations andthe street).Statistical analysisStatistical analysis was conducted using survey methodsin SAS Version 9.1 [27], taking into account the sam-pling clusters and weights. FSWs sampled from thesame clusters are assumed to be more similar to eachother than they are to FSWs from different clusters; sur-vey methods account for this by estimating the overallvariance from the variation among the clusters [28].Descriptive statistics were calculated for sample charac-teristics. The prevalence of outcomes was calculated foreach variable describing exposure to the intervention.Bivariate and multivariable logistic regression was usedto model the relationship between the condom use out-comes and the five variables describing exposure to theintervention. Five separate logistic models were createdfor each of the five dichotomous outcomes, for a totalof 25 separate models. Inclusion into multivariable mod-els was based on significance at the P<0.10-level fromWald chi-squared tests in bivariate regression analyses,or if they were perceived to be important confounders apriori (district, typology of sex work). Each single inter-vention variable was forced into the five different multi-variable models to examine the independent relationshipbetween intervention exposure and CCU. Two interven-tion exposure variables were dichotomous (ever beencontacted by intervention staff, ever seen a condomdemonstration by intervention staff), while three wereoriginally continuous (duration since first contacted byintervention staff, number of times contacted by inter-vention staff, number of condom demonstrations givenby staff). The continuous variables were categorizedprior to analysis. To examine a dose-responseDeering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 3 of 14relationship, a linear test for trends across categories foreach of the three continuous intervention exposure vari-ables and each CCU outcome was conducted. The med-ian of each category was taken, and the exposurevariable was treated as a continuous variable. Oddsratios (ORs) and adjusted odds ratios (AORs) and their95% confidence interval [95% CIs] were reported forlogistic regression and P-values were reported for thetests for trends. All P-values reported are two-sided.ResultsSample characteristicsTable 1 presents characteristics of the overall sample ofFSWs in three districts in Karnataka state. The samplesizes for Belgaum, Bellary and Bangalore districts were208, 198 and 369 (N=775) respectively, and the medianage of FSWs across the three districts was 30 years(interquartile range [IQR]=25-35; mean=30.3 years). Ofthe total sample, the majority of women, 348 (55.6%),primarily solicited clients in public places, while 245(26.2%) solicited clients from their homes and 182(18.2%) women solicited clients in brothels. Overall, 371(52.5%) women in the sample were divorced, separatedor widowed, 229 (26.0%) were currently married, 119(15.5%) were Devadasi and 56 (5.9%) were other womenwho were never married (Table 1).Relationship between the intervention and condom useCommercial sex clientsThe sample of 775 FSWs all reported having occasionalclients. Of these women, 433 had repeat clients. Overall,585 (81.7%) of FSWs reported CCU with all clients inthe most recent day worked, 530 (69.5%) womenreported CCU with current occasional clients and 269(57.5%) women reported CCU with their most recentrepeat client. CCU with all clients was higher amongFSWs who had ever been contacted by intervention staffcompared to those who had not (84.6% versus 65.6%),as was CCU with occasional clients (71.9% versus53.5%) (Figure 1a). The same patterns were observed forFSWs who had ever been given condoms by interven-tion staff compared to those who had not (CCU with allclients: 86.6% versus 65.8%; CCU with occasional clients:73.9% versus 53.6%) (Figure 1b). CCU was approxi-mately the same with the most recent repeat client forwomen who had ever been contacted by interventionstaff compared to those who had not (57.7% versus56.2%) and for those who had ever seen a condomdemonstration (57.9% versus 58.0%) compared to thosewho had not.CCU with all clients in the most recent day worked,with occasional clients and with the most recent repeatclient, increased overall as the duration of time sincefirst contact by intervention staff increased, but onlysteadily increased with increased duration for CCU withoccasional clients (Figure 1c). CCU with all clientsincreased steadily as the number of times contacted bystaff in the past month increased (Figure 1d). CCU washighest with occasional clients and with the most recentrepeat client among women who had been contacted <5times (relative to women who had never been contactedor who had been contacted 5+ times). Finally, and con-sistent with the previous outcome, CCU with all clientsin the last day worked, with occasional clients and themost recent repeat client increased with the number ofcondom demonstrations observed in the last month, butlevelled off and decreased (substantially for the two lat-ter outcomes) at two condom demonstrations in the lastmonth (Figure 1e).In bivariate analysis, all five intervention variableswere significantly associated (on a P<0.10 significancelevel) with CCU with all clients in the most recent dayworked and CCU with occasional clients (Table 1).Other explanatory variables that were significant on aP<0.10-level, or included in multivariable models apriori, are also listed in Table 1. In multivariable analysis(Table 2), after adjusting for social and environmentalfactors, the odds of CCU with all clients in the mostrecent day worked and CCU with occasional clientswere 6.3-fold [95% CIs 2.8-14.5] and 2.3-fold [95% CIs:1.3-4.1] higher among FSWs who had ever been con-tacted by intervention staff and 4.9-fold [95% CIs 2.6-9.3] and 2.3-fold [95% CIs: 1.3-4.1] higher among FSWswho had ever been given condoms by intervention staff,compared to those who never had been. None of theintervention exposure variables were significantly asso-ciated with CCU with the most recent repeat client inbivariate or multivariable analysis (Tables 1 and 2).In bivariate analysis testing for trends, CCU with allclients in the most recent day worked was significantlyassociated with an increased duration since first con-tacted by intervention staff, the number of times con-tacted by intervention staff, and the number of condomdemonstrations seen by staff in the last month. CCUwith occasional clients was significantly associated withan increased duration since first being contacted byintervention staff, and the number of condom demon-strations seen by staff in the last month. In multivariableanalysis, significant tests for trends indicated a dose-response relationship between CCU with all clients inthe most recent day worked and CCU with occasionalclients, and increased duration since first being con-tacted by staff (P=0.001, P=0.006 respectively). For bothof these outcomes, significant tests for trends also indi-cated a dose-response relationship between CCU withall clients in the most recent day worked and CCU withoccasional clients and increased numbers of condomdemonstrations witnessed (P=0.004 and P=0.026Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 4 of 14Table 1 Sample characteristics and bivariate associations (unadjusted odds ratios [OR]) and 95% confidence intervals(CIs): sample characteristics and bivariate associations between social, environmental and intervention exposurefactors and consistent condom use with commercial sex clients1,2,3Proportion (n) or mean/median (interquartilerange=IQR)N=775OUTCOMESCondoms used ineach occasion ofsexual intercourse withall clients in the mostrecent day workedConsistent condomuse with occasionalclientsConsistent condomuse with most recentrepeat clientOR [95% CIs] P OR [95% CIs] P OR [95% CIs] PSOCIALAge (years) 30.3/30 (25-35)1.00 [0.97-1.03] 0.935 0.97 [0.93-1.00] 0.081 0.97 [0.93-1.02] 0.188Marital statusDevadasiNever marriedDivorced/Separated/WidowedCurrently married15.5 (119)5.9 (56)52.5 (371)26.0 (229)1.73 [0.73-4.08]0.65 [0.26-1.63]0.93 [0.49-1.76]1.0 (ref)0.268 2.27 [1.18-4.34]1.02 [0.47-2.22]0.94 [0.54-1.64]1.0 (ref)0.031 1.30 [0.64-2.66]1.26 [0.50-3.18]0.77 [0.28-2.09]1.0 (ref)0.729ReligionHindu (versus other - Islam/Christian/Jain) 89.1 (682) 1.01 [0.51-1.98] 0.981 1.27 [0.69-2.33] 0.445 0.53 [0.20-1.41] 0.206Age at first sex (years) 15.5/15 (14-17)1.12 [1.01-1.24] 0.036 1.07 [0.94-1.22] 0.295 1.06 [0.98-1.16] 0.137Age at first sex work (years) 23.8/23 (18-29)0.99 [0.96-1.03] 0.734 0.96 [0.93-0.99] 0.006 0.97 [0.92-1.03] 0.269Duration of sex work (years) 6.5/5 (2-10)1.01 [0.96-1.05] 0.744 1.00 [0.95-1.06] 0.892 0.99 [0.95-1.04] 0.786ENVIRONMENTALDistrictBelgaumBellaryBangalore26.8 (208)25.6 (198)47.6 (369)0.54 [0.30-1.00]0.82 [0.41-1.63]1.0 (ref)0.124 1.34 [0.81-2.23]1.58 [0.91-2.73]1.0 (ref)0.237 0.89 [0.32-2.43]2.50 [0.89-7.04]1.0 (ref)0.005Literate (versus cannot read/write) 27.2 (227) 1.47 [0.84-2.58] 0.177 1.67 [1.05-2.64] 0.029 2.27 [1.02-5.05] 0.044Sex work sole income (versus has other paidwork)35.0 (301) 0.73 [0.45-1.17] 0.186 0.86 [0.57-1.30] 0.468 0.75 [0.32-1.78] 0.519Independent solicitation (versus solicitationby a manager)77.4 (555) 0.84 [0.47-1.48] 0.543 0.89 [0.54-1.47] 0.640 1.38 [0.52-3.65] 0.523TypologyBrothelPublic placesHome18.2 (182)55.6 (348)26.2 (245)0.55 [0.28-1.06]0.93 [0.52-1.65]1.0 (ref)0.161 1.09 [0.58-2.07]0.68 [0.41-1.11]1.0 (ref)0.142 0.88 [0.42-1.83]0.66 [0.25-1.71]1.0 (ref)0.689INTERVENTION EXPOSUREEver contacted by intervention staff (versusnot ever contacted)85.5 (632) 2.88 [1.56-5.32] <0.001 2.23 [1.31-3.82] 0.003 1.06 [0.42-2.68] 0.901Had a condom demonstration byintervention staff (versus never had acondom demonstration)82.0 (591) 3.37 [1.93-5.88] <0.001 2.45 [1.37-4.39] 0.003 1.00 [0.40-2.48] 0.992Duration since first contacted by interventionstaffHas not been contactedLess than one year (greater than zero)One yearTwo to three yearsTest for trends15.4 (143)36.1 (240)28.0 (198)20.6 (154)1.0 (ref)3.38 [1.65-6.92]2.08 [1.04-4.18]3.47 [1.64-7.33]0.0040.0581.0 (ref)1.69 [0.94-3.04]2.37 [1.23-4.55]2.85 [1.45-5.59]0.0120.0041.0 (ref)0.65 [0.19-2.29]1.41 [0.61-3.28]1.51 [0.65-3.50]0.4640.165Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 5 of 14respectively). Finally, a dose-response relationship wasalso observed between CCU with all clients and numberof times contacted by staff (P=0.047).Non-commercial partnersOf the total sample, 226 FSWs reported having a non-paying sexual partner in the last year (who was neitherthe husband nor main cohabiting partner) and 354 hada husband or cohabiting partner. Overall, 68 (31.1%)and 40 (9.6%) reported CCU with their most recentnon-paying partner and their husband or cohabitingpartner respectively (Table 1). In contrast to CCU withall clients and with occasional clients, CCU with theirmost recent non-paying partner was higher amongFSWs who had never been contacted by interventionstaff compared to those who had been contacted (35.8%versus 31.4%); the same was true for CCU with theirhusband or cohabiting partner (15.5% versus 8.8%).CCU with their most recent non-paying partner washigher for those who had seen a condom demonstrationcompared with those who had not (34.0% versus 26.6%),while CCU with their husband or cohabiting partnerwas higher for those who had never seen a condomdemonstration compared to those who had (12.9% ver-sus 9.3%) (Figures 2a-2b). Figure 2c demonstrates howCCU with both types of partners decreased and thenincreased as the duration of time since first contactedby intervention staff increased. CCU with both theirmost recent non-paying partner and CCU with theirhusband or cohabiting partner decreased as the numberof times contacted by staff in the past month increased(Figure 2d). CCU with the most recent non-paying part-ner initially increased as the number of condom demon-strations witnessed increased, and then dropped byalmost half for women who had seen three or moredemonstrations in the past month. CCU with the hus-band or cohabiting partner increase slightly for onecompared to zero demonstrations, then decreased stea-dily with increasing number of condom demonstrations(Figure 2e).In bivariate logistic regression, only the variable num-ber of times contacted by intervention staff was signifi-cantly associated with CCU with FSWs’ most recentnon-paying partner (on a P<0.10 significance level). Inmultivariable analysis, after adjusting for social andenvironmental factors, none of the intervention expo-sure variables were significantly associated with CCUwith their non-paying partner or husband or cohabitingpartner (Table 3). In bivariate analysis testing for trends,CCU with FSWs’ most recent non-paying partner wasinversely associated with the number of times contactedTable 1 Sample characteristics and bivariate associations (unadjusted odds ratios [OR]) and 95% confidence intervals(CIs): sample characteristics and bivariate associations between social, environmental and intervention exposure fac-tors and consistent condom use with commercial sex clients1,2,3 (Continued)Number of times contacted by interventionstaffZeroFive or fewer (greater than zero)Greater than fiveTest for trends15.1 (146)63.2 (486)21.7 (137)1.0 (ref)2.55 [1.36-4.78]3.38 [1.32-8.66]0.0060.0751.0 (ref)2.54 [1.47-4.42]1.80 [0.62-5.27]0.0030.8211.0 (ref)1.20 [0.53-2.75]0.85 [0.23-3.15]0.5070.603Number of condom demos seen past monthby staffZeroOneTwoThree or greaterTest for trends18.0 (160)23.0 (183)22.2 (180)36.8 (228)1.0 (ref)1.99 [1.10-3.61]4.72 [2.28-9.77]4.48 [1.96-10.28]<0.0010.0011.0 (ref)1.77 [0.90-3.49]4.51 [2.37-8.60]2.23 [1.11-4.49]<0.0010.0991.0 (ref)1.09 [0.49-2.43]2.08 [0.94-4.61]0.73 [0.21-2.46]0.1300.499Condoms used in all occasions of sexualintercourse with clients in the most recentday worked81.7 (585)Consistent condom use with all occasionalclients69.5 (530)Consistent condom use with most recentrepeat client57.5 (269)Consistent condom use with most recentnon-paying partner31.1 (68)Consistent condom use with husband/cohabiting partner9.6 (40)1Not all of the outcomes have the same denominator, as the sample was subset to women with different types of partners for each outcome; condoms used inall occasions of sexual intercourse with clients in the most recent day worked has a smaller denominator than consistent condom use with all occasional clientsbecause of missing data in the former outcome.2Consistent condom use is defined as reporting always (100%) using condoms.3The total N for each factor may not add up to 775 due to missing values.Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 6 of 14Figure 1 Relationship between indicators of intervention exposure and consistent condom use (CCU). These include CCU with all commercialclients of female sex workers (FSWs) in the most recent day worked, CCU with occasional clients and CCU with the most recent repeat client,based on the results of special behavioural surveys in Karnataka state: (a) CCU vs. ever been contacted by intervention staff; (b) CCU vs. everseen a condom demonstration by intervention staff; (c) CCU vs. time since first contacted by programme staff; (d) CCU vs. number of timescontacted by staff in the past month; and (e) CCU vs. number of condom demonstrations by staff observed by FSWs in the past month.Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 7 of 14by intervention staff and CCU with their husband orcohabiting partner was inversely associated with thenumber of condom demonstrations by intervention staff.In multivariable analysis testing for trends, CCU withFSWs’ husband or cohabiting partner remained inverselysignificantly associated with the number of condomdemonstrations by intervention staff (P=0.05).DiscussionThe results from our analysis suggest that exposure to alarge-scale HIV prevention initiative in Karnataka, India,was associated with higher reported consistent condomuse (CCU) among women engaged in sex work withtheir commercial sex clients. After adjusting for socialand environmental factors, a strong independent asso-ciation was observed between CCU with all clients inthe most recent day worked and CCU with occasionalclients, and five measures of intervention exposure.Moreover, a significant dose-response relationship wasobserved between these two outcomes and increasedduration since first contacted by intervention staff, aswell as number of condom demonstrations seen by staffin the last month. There was also a significant dose-response relationship observed between CCU with allclients and the number of times contacted by staff inthe past month. In multivariable analysis, interventionexposure was not significantly associated with increasedCCU with FSWs’ most recent repeat commercial client,their most recent non-paying partner or their husbandor cohabiting partner.The association between increased intervention expo-sure and increased CCU with all clients likely reflectshigher condom use with occasional clients, which con-stitute the majority of commercial clients in Karnataka.On a micro-level, condom use with occasional clientslikely improved due to regular contact between FSWsand peer outreach workers (i.e., members of local sexworker communities), who were responsible forTable 2 Multivariable associations1 (adjusted odds ratios [AOR]) and 95% confidence intervals (95% CIs): multivariableassociations between social, environmental and intervention exposure factors and consistent condom use withcommercial sex clients. Five models (MODEL1-MODEL5) were constructed for each of the five explanatory variables forintervention exposure and each outcome, for 15 models total.MODEL1 OUTCOME: Consistent condom2 use within different sexualpartnershipsCondoms used in eachoccasion of sexualintercourse with clients inthe most recent dayworkedConsistent condomuse with occasionalclientsConsistent condomuse with most recentrepeat clientINTERVENTION EXPOSURE AOR [95% CIs] P AOR [95% CIs] P AOR [95% CIs] P1 Ever contacted by intervention staff (versus not evercontacted)6.32 [2.76-14.47] <0.001 2.30 [1.30-4.08] 0.006 1.07 [0.34-3.33] 0.9932 Had a condom demonstration by intervention staff (versusnever had a condom demonstration)4.89 [2.57-9.30] <0.001 2.30 [1.30-4.07] 0.009 0.88 [0.29-2.66] 0.8123 Duration since first contacted by intervention staffHas not been contactedLess than one year (greater than zero)One yearTwo-three yearsTest for trends1.0 (ref)5.80 [2.54-13.29]5.38 [2.06-14.06]11.76 [3.80 -36.33]1.0 (ref)1.77 [0.98-3.20]2.72 [1.33-5.56]3.71 [1.55-8.87]1.0 (ref)0.65 [0.19-2.19]1.53 [0.51-4.65]2.55 [0.79-8.23]4 Number of times contacted by intervention staffZeroFive or fewer (greater than zero)Greater than fiveTest for trends1.0 (ref)5.10 [2.28 -11.43]7.66 [2.27-25.87]1.0 (ref)2.65 [1.42-4.96]1.99 [0.78-5.08]1.0 (ref)1.12 [0.36-3.51]0.87 [0.21-3.60]5 Number of condom demos by staff seen past monthZeroOneTwoThree or greaterTest for trends1.0 (ref)3.41 [1.69-6.86]7.76 [3.08-19.54]4.88 [2.17-10.97]1.0 (ref)1.48 [0.72-3.04]4.59 [2.26-9.29]2.13 [1.14-3.97]1.0 (ref)1.08 [0.42-2.76]1.97 [0.80-4.87]0.70 [0.21-2.35]0.8750.1430.561/1 Models were all adjusted for variables that were included a priori and variables that were significantly associated with each outcome on the p<0.10-level inbivariate analysis. For all three outcomes of condom use, a priori variables included typology of sex work (place of solicitation) and district; for condoms used ineach occasion of sexual intercourse with clients in the most recent day worked, models were also adjusted by marital status, age and age at first sex; for condomuse with occasional clients, models were also adjusted by age, marital status, literacy, age at first sex work; for condom use with most recent repeat client,models were also adjusted by education.2 Consistent condom use is defined as reporting always (100%) using condoms./ Test for trend not significant in bivariate analysis, and was not tested in multivariable models.Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 8 of 14Figure 2 Relationship between indicators of intervention exposure and consistent condom use (CCU). These included CCU with the most recentnon-paying partner of female sex workers (FSWs) (who was neither the husband nor the main cohabiting partner) and FSWs’ husband orcohabiting partner, based on the results of special behavioural surveys in Karnataka state: (a) CCU vs. ever been contacted by intervention staff;(b) CCU vs. ever seen a condom demonstration by intervention staff; (c) CCU vs. time since first contacted by programme staff; (d) CCU vs.number of times contacted by staff in the past month; (e) CCU vs. number of condom demonstrations by staff observed by FSWs in the pastmonth.Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 9 of 14providing condoms to FSWs, giving demonstrations ofcorrect condom use and facilitating conversations aboutrisk and vulnerability [15]. Of note, our exposure vari-ables measuring contacts by peers were not independentof our variables measuring condom demonstrations bypeers. Although intervention exposure variables couldnot directly capture the influence of community involve-ment or mobilization strategies, peers also encouragedmembership in community groups and were proponentsof community mobilization, which is intended to facili-tate condom negotiation by FSWs and use with clientsthrough both individual-level and collective empower-ment and agency [14,18]. Interestingly, CCU with clientsin this analysis was highest for FSWs who had seen twocondom demonstrations by staff in the previous monthand lower for FSWs who had seen three or more. CCUwith occasional clients and repeat clients was also higheramong FSWs who had been contacted <5 times com-pared to those who had been contacted 5+ times. Theseresults could suggest that there may be a point whereincreased contact by staff or education about correctcondom use by intervention staff will not improve con-dom use [29]. Resources may be better directed to otherfeatures of the intervention if additional increases incondom use are to be observed. CCU was found tosteadily increase with increased duration since first con-tacted by the intervention. This effect may not havelevelled off (as with the previous two intervention expo-sure variables) over time due to the limited amount oftime since the intervention began in some districts (var-ied from 1.5-2.5 years). Condom use may also havenaturally increased over time in southern India(reflected in the duration since first contacted by staff)albeit likely at slower rates than in if the interventionwas not present. Condom use may not have reached100% in all commercial sex acts due to the timing ofsurvey data collection (e.g., condom use may stillincrease with increased exposure to the intervention).There may also be groups of highly vulnerable FSWswho may be unable to negotiate condom use with allclients who refuse to wear condoms, even if exposed tothe intervention. Condom use with commercial clientswas relatively high for those FSWs who reported thatthey were not exposed to the intervention. This may bedue to the presence of other HIV prevention pro-grammes in place prior to Avahan. SBS surveys werealso implemented 7-19 months after Avahan was intro-duced in different districts, and an independent analysisretrospectively assessing condom use confirmed thatcondom use increased notably after Avahan wasTable 3 Multivariable associations1 (adjusted odds ratios [AOR]) and 95% confidence intervals (95% CIs): multivariateassociations between social, environmental and intervention exposure factors and condom use with non-commercialpartners. Five models (MODEL1-MODEL5) were constructed for each of the five explanatory variables for interventionexposure and each of the two outcomes, for 10 models total.Consistent condomuse2 with most recentnon-paying partnerConsistent condomuse2 with husband orcohabiting partnerAOR [95% CIs] P AOR [95% CIs] P1 Ever contacted by intervention staff (versus not ever contacted) 1.40 [0.47-4.18] 0.542 0.35 [0.11-1.16] 0.0852 Had a condom demonstration by intervention staff (versus never had a condom demonstration) 1.72 [0.60-4.91] 0.311 0.50 [0.17-1.43] 0.1943 Duration since first contacted by intervention staffHas not been contactedLess than one year (greater than zero)One yearTwo-three yearsTest for trends1.0 (ref)1.05 [0.34-3.29]1.52 [0.43-5.40]2.74 [0.64-11.74]1.0 (ref)0.39 [0.11-1.44]0.48 [0.13-1.78]0.20 [0.03-1.43]4 Number of times contacted by intervention staffZeroFive or fewer (greater than zero)Greater than fiveTest for trends1.0 (ref)1.66 [0.56-4.93]0.43 [0.09-2.09]1.0 (ref)0.42 [0.13-1.36]0.43 [0.11-1.66]5 Number of condom demos by staff seen past monthZeroOneTwoThree or greaterTest for trends1.0 (ref)3.43 [0.95-12.48]3.39 [0.89-12.97]1.02 [0.28-3.67]1.0 (ref)0.86 [0.23-3.15]0.64 [0.18-2.27]0.33 [0.10-1.00]1 Models were all adjusted for variables that were included a priori and variables that were significantly associated with each outcome on the p<0.10-level inbivariate analysis. For all three outcomes, a priori variables included district and typology of sex work (place of solicitation); for condom use with the husband orcohabiting partners, models were also adjusted by age at first sex work.2 Consistent condom use is defined as reporting always (100%) using condoms./ Test for trend not significant in bivariate analysis, and was not tested in multivariable models.Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 10 of 14introduced [30]. Condom use may also be high due tothe indirect impacts of Avahan (e.g., through increasedpeer awareness of condoms or increased condom avail-ability [31]). Improved condom availability was also akey feature of the intervention [31]. This facilitated con-dom use simply by increasing access, but also likely byincreasing social acceptance of condoms throughincreased visibility and presence. Other interventionsincorporating these program elements have shown suc-cess in improving condom use among FSWs [10,11,32].Results from this study are supported by other studiesshowing similar results. These include observational stu-dies suggesting that condom use as reported by clients[33] and FSWs [30,34] increased after the introductionof the intervention, as well as studies suggesting thatcondom availability to FSWs increased substantiallysince the intervention began [31]. Increases in condomuse among high-risk groups could have important impli-cations for HIV and STI prevalence in Karnataka. Senti-nel surveillance and observational studies have founddecreasing trends in terms of HIV and STIs amongFSWs in Karnataka state since the intervention wasintroduced [34,35]. Moreover, results are also consistentwith mathematical modelling indicating that the increasein condom use after initiation of the intervention wasconsistent with decreasing HIV epidemiological trendsover multiple rounds of survey data collection in Karna-taka state [36,37]. There is evidence to suggest thatincreased condom levels can be sustained over time inthis population [34]. Nevertheless, continued monitoringof condom use levels and assessments of the impact ofobserved increases in condom use on reducing HIV andSTIs is important for a long-term and comprehensiveunderstanding of the impact of the intervention.While a higher probability of CCU with all clients in themost recent day worked and occasional clients wasobserved for FSWs with increased exposure to the inter-vention, the same patterns were not observed amongFSWs for CCU with their most recent repeat client, non-paying partner and their husband or cohabiting partner.Moreover, CCU with FSWs’ husband or cohabiting part-ner decreased significantly with increasing numbers ofcondom demonstrations seen by intervention staff in theprevious month, when testing for trends. It is not clearwhy this was observed, but of note, CCU with FSWs’ hus-band or cohabiting partner was very low and the absoluteproportions did not vary substantially according to thenumber of condom demonstrations seen (10.1% to 13.8%).The reasons for low condom use within non-commercialand regular commercial sexual partnerships of FSWs arecomplex. These may include power disparities that favourthe male partner [4,38,39], including an economic depen-dence on longer-term male partners [40,41]. The use ofcondoms may not be acceptable in non-commercialrelationships, if there is greater longevity, trust and inti-macy within the partnership; the use of condoms may alsobe perceived as a symbol of infidelity and foster mistrust[42]. Women may agree to not use condoms with repeatclients in exchange for these partners providing a stableform of longer-term income or because they feel they canassess if their partner is not infected with HIV after theyhave seen him several times. Further research is requiredto better understand condom use with non-commercialand regular commercial partners of FSWs; in particular,understanding gender-based interpersonal factors thatinfluence condom use and preferences by both partners,as well as environmental factors (e.g. favourable societalviews of condoms) that could be incorporated into inter-ventions to increase condom use. The female condomcould play an increasing role in HIV prevention withinthese partnerships. In addition, recent promising findingsof the effectiveness of microbicides indicate that microbi-cides could play an important role in HIV prevention asan alternative to condoms for women whose partner doesnot use condoms [43].There are several limitations to this study. This study isbased on self-reported data from cross-sectional surveyscollected in three districts and is not experimental indesign. This study was based on data collected only fromFSWs and should be considered alongside studies thatshow consistent results in other populations (e.g., cli-ents), using other data sources to assess exposure to theintervention, and in similar settings. The condom useoutcomes used in this study did not specify a timeframe.However, the questions were intended to capture recent(e.g., condom use with the last 10 clients) or currentaverage behaviour. We relied on self-reported answers toquestions that may be perceived as sensitive (e.g. consis-tency of condom use), and the questions are thereforesusceptible to social desirability bias [44]. This may haveoverestimated the relationship between interventionexposure and CCU, and it is possible that women withincreased condom use are more likely or able to beaccessed by intervention-related services and programsrather than the other way around. However, we believethat this is unlikely, since the total sample size was rela-tively large, particularly for a marginalized and hiddenpopulation, and the cluster sampling design was used tomake the sample as representative as possible, with com-plex survey methods accounting for the sampling design.Additionally, our results suggest not only a relationshipbetween exposure to the intervention and increased con-dom use, but a dose-response relationship betweenincreased exposure to the intervention and increasedcondom use with all clients and occasional clients.The impact of increases in condom use among FSWsand their clients on the HIV epidemic in southern Indiashould continue to be assessed. Large-scale HIVDeering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 11 of 14prevention programs targeted at groups with high HIVrisk could in theory also have an indirect impact onHIV prevalence within general populations [45]; how-ever, results from mathematical modelling suggest thatthe current observed decrease in antenatal clinic HIVprevalence in India was likely not caused by the FSW-targeted intervention, although it is likely too early toassess the impact of the intervention on bridge groupsor groups with low risk for HIV [46]. Since many clientsof FSWs have other longer-term and/or non-payingpartners such as wives, as well as other FSW partners,and since condom use is low in these partnerships, cli-ents provide an important transmission “bridge”between FSWs and the general population [45,47]. Sincecondom use tends to be relatively low among FSWs’repeat clients and non-paying partnerships, these part-ners can also provide a transmission bridge to FSWs[48].If the intervention’s influence on condom use variesby type of commercial sex client (e.g. occasional com-pared to repeat clients) and the patterns of sexual struc-ture vary geographically (e.g., districts such as Bangalorehave higher fractions of occasional clients and lowernumbers of repeat clients per month), we would expectto observe different intervention effects across the threedistricts in Karnataka. CCU with non-paying partnerswas also much lower in Bangalore (12.8%) compared tothe other two districts (45.7% and 41.6% respectively),indicating that the importance of these partnerships inthis district may be more pronounced, and should beconsidered in this district more than others when plan-ning interventions. Exploring the relative role of differ-ent patterns of FSW-client sexual structure andvariation in the numbers of different types of partnerson HIV transmission in Karnataka, India, using simula-tion studies, would be useful to further improve theimpact of the intervention [3,24,49-51].ConclusionsIn summary, study findings suggest that the exposure toa large-scale HIV preventive intervention among FSWswas associated with increased condom use with occa-sional clients, with a dose-response relationship, butthat it did not seem to influence condom use withrepeat clients, non-paying partners and with the hus-band or cohabiting partner. Future research should bedirected toward understanding why condom useremains relatively low with non-commercial partnersand new strategies should be investigated and developedspecifically to increase condom use by these partners.AcknowledgementsKND is supported by a Postdoctoral Fellowship (Bisby Award) from theCanadian Institutes of Health Research. We would like to thank AdammaAghaizu for early contributions to the study. Support for this study wasprovided by the Bill and Melinda Gates Foundation. The views expressedherein are those of the authors and do not necessarily reflect the officialpolicy or position of the Bill and Melinda Gates Foundation.This article has been published as part of BMC Public Health Volume 11Supplement 6, 2011: Learning from large scale prevention efforts – findingsfrom Avahan. The full contents of the supplement are available online atURL.Author details1Division of AIDS, Department of Medicine, Faculty of Medicine, University ofBritish Columbia, Vancouver, Canada. 2Department of Infectious DiseasesEpidemiology, Imperial College, London, UK. 3HIV and STI Department,Health Protection Services – Colindale, Health Protection Agency, London,UK. 4Faculty of Medicine, University of Ottawa, Ottawa, Canada. 5LondonSchool of Hygiene and Tropical Medicine, London, UK. 6URESP, Centre derecherche FRSQ du CHA universitaire de Québec, Québec, Canada.7Karnataka Health Promotion Trust, Bangalore, India. 8Department of MedicalMicrobiology, University of Manitoba, Winnipeg, Canada. 9Department ofCommunity Health Sciences, University of Manitoba, Winnipeg, Canada.10Département de médecine sociale et préventive, Université Laval, Québec,Canada. 11Institut national de santé publique du Québec, Québec, Canada.Authors’ contributionsKND contributed to the conceptual design of the study, conducted thestudy and the analysis and drafted the manuscript; MCB, CML, PV, MP andMA participated in the conceptual design of the study and coordination,made substantial contributions in the interpretation of data; JS, MWT, SMand JB made substantial contributions in the interpretation of data andcritically revised the manuscript for important intellectual content; BMR,KG, SR, RW made substantial contributions to the acquisition andmanagement of the data. All authors read and approved the finalmanuscript.Competing interestsThe authors declare that they have no competing interests.Published: 29 December 2011References1. Rhodes T: The ‘risk environment’: a framework for understanding andreducing drug-related harm. Int J Drug Policy 2002, 13:85-94[http://www.sciencedirect.com/science/article/pii/S0955395902000075].2. Rhodes T, Singer M, Bourgois P, Friedman SR, Strathdee SA: The social andstructural production of HIV risk among injecting drug users. Soc Sci Med2005, 61:1026-1044.3. Buzdugan R, Halli SS, Cowan FM: The female sex work typology in Indiain the context of HIV/AIDS. Trop Med Int Health 2009, 14:673-687.4. Shannon K, Kerr T, Allinott S, Chettiar J, Shoveller J, Tyndall MW: Social andstructural violence and power relations in mitigating HIV risk of drug-using women in survival sex work. Soc Sci Med 2008, 66:911-921.5. Harcourt C, Donovan B: The many faces of sex work. Sex Transm Infect2005, 81:201-206.6. Shannon K, Rusch M, Shoveller J, Alexson D, Gibson K, Tyndall MW, MakaProject Partnership: Mapping violence and policing as an environmental-structural barrier to health service and syringe availability amongsubstance-using women in street-level sex work. Int J Drug Policy 2008,19:140-147.7. Jana S, Rojanapithayakorn W, Steen R: Harm reduction for sex workers.The Lancet 2006, 367:814.8. Hanenberg R, Rojanapithayakorn W, Kunasol P, Sokal D: Impact ofThailand’s HIV-control programme as indicated by the decline ofsexually transmitted diseases. Lancet 1994, 344:243-245.9. Rojanapithayakorn W: The 100% condom use programme in Asia. ReprodHealth Matters 2006, 14:41-52.10. Jana S, Basu I, Rotheram-Borus MJ, Newman PA: The Sonagachi Project: asustainable community intervention. AIDS Educ Prev 2004, 16:405-414.11. Cohen J: HIV/AIDS in India: Sonagachi sex workers stymie HIV. Science2004, 304:506.12. Avahan, the India AIDS Initiative - the Business of HIV prevention atScale. New Delhi, India: The Bill & Melinda Gates Foundation; 2008.Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 12 of 1413. Piot P: Setting new standards for targeted HIV prevention: the Avahaninitiative in India. Sex Transm Infect 2010, 86(Suppl 1):i1-i2.14. Sarkar S: Community engagement in HIV prevention in Asia: going from‘for the community’ to ‘by the community’–must we wait for moreevidence? Sex Transm Infect 2010, 86(Suppl 1):i2-i3.15. Managing HIV Prevention from the Ground Up: Avahan’s Experience inPeer Led Outreach at Scale in India.. New Delhi, India: The Bill & MelindaGates Foundation; 2008.16. Blanchard JF, Bhattacharjee P, Kumaran S, Ramesh BM, Kumar NS,Washington RG, Moses S: Concepts and strategies for scaling up focusedprevention for sex workers in India. Sex Transm Infect 2008, 84(Suppl 2):ii19-ii23.17. Steen R, Mogasale V, Wi T, Singh AK, Das A, Daly C, George B, Neilsen G,Loo V, Dallabetta G: Pursuing scale and quality in STI interventions withsex workers: initial results from Avahan India AIDS Initiative. Sex TransmInfect 2006, 82:381-385.18. Halli SS, Ramesh BM, O’Neil J, Moses S, Blanchard JF: The role of collectivesin STI and HIV/AIDS prevention among female sex workers in Karnataka,India. AIDS Care 2006, 18:739-749.19. Beattie TS, Bhattacharjee P, Ramesh BM, Gurnani V, Anthony J, Isac S,Mohan HL, Ramakrishnan A, Wheeler T, Bradley J, et al: Violence againstfemale sex workers in Karnataka state, south India: impact on health,and reductions in violence following an intervention program. BMCPublic Health 2010, 10:476.20. Reza-Paul S, Beattie T, Syed HU, Venukumar KT, Venugopal MS, Fathima MP,Raghavendra H, Akram P, Manjula R, Lakshmi M, et al: Declines in riskbehaviour and sexually transmitted infection prevalence following acommunity-led HIV preventive intervention among female sex workersin Mysore, India. AIDS 2008, 22(Suppl 5):S91-S100.21. Boily MC, Lowndes CM, Vickerman P, Kumaranayake L, Blanchard J, Moses S,Ramesh BM, Pickles M, Watts C, Washington R, et al: Evaluating large-scaleHIV prevention interventions: study design for an integratedmathematical modelling approach. Sex Transm Infect 2007, 83:582-589.22. Szklo M, Nieto FJ: Epidemiology: Beyond the Basics. New York: AspenPublishers; 1999.23. Ramesh BM, Moses S, Washington R, Isac S, Mohapatra B, Mahagaonkar SB,Adhikary R, Brahmam GNV, Paranjape RS, Subramanian T, Blanchard JF, IBBAStudy Team: Determinants of HIV prevalence among female sex workersin four south Indian states: analysis of cross-sectional surveys in twenty-three districts. AIDS 2008, 22(Suppl 5):S35-S44.24. Blanchard JF, O’Neil J, Ramesh BM, Bhattacharjee P, Orchard T, Moses S:Understanding the social and cultural contexts of female sex workers inKarnataka, India: implications for prevention of HIV infection. J Infect Dis2005, 191(Suppl 1):S139-S146.25. O’Neil J, Orchard T, Swarankar RC, Blanchard JF, Gurav K, Moses S:Dhandha, dharma and disease: traditional sex work and HIV/AIDS inrural India. Soc Sci Med 2004, 59:851-860.26. Orchard T: In this life: the impact of gender and tradition on sexualityand relationships for Devadasi sex workers in rural India. Sexuality andCulture 2007, 11:3-27[http://www.springerlink.com/content/u38121502n51gt8q/].27. SAS. Cary, NC, USA: SAS Institute Inc; 2002-03, 9.1, pp 9.1.28. Cole SR: Analysis of complex survey data using SAS. Comput MethodsPrograms Biomed 2001, 64:65-69.29. Aral SO, Blanchard J, Lipshutz J: STD/HIV prevention intervention: efficacy,effectiveness and population impact. Sex Transm Infect 2008, 84(Suppl 2):ii1-ii3.30. Lowndes CM, Alary M, Verma S, Demers E, Bradley J, Jayachandran AA,Ramesh BM, Moses S, Adhikary R, Mainkar MK: Assessment of interventionoutcome in the absence of baseline data: ‘reconstruction’ of condomuse time trends using retrospective analysis of survey data. Sex TransmInfect 2010, 86(Suppl 1):i49-i55.31. Bradley J, Moses S, Blanchard JF, Rajaram S, Ramesh BM, Verma S, Alary M:Assessing reported condom use among female sex workers in southernIndia through examination of condom availability. Sex Transm Infect 2010,86(Suppl 1):i44-i48.32. Kerrigan D, Telles P, Torres H, Overs C, Castle C: Community developmentand HIV/STI-related vulnerability among female sex workers in Rio deJaneiro, Brazil. Health Educ Res 2007, 23:137-145.33. Lipovsek V, Mukherjee A, Navin D, Marjara P, Sharma A, Roy KP: Increasesin self-reported consistent condom use among male clients of femalesex workers following exposure to an integrated behaviour changeprogramme in four states in southern India. Sex Transm Infect 2010,86(Suppl 1):i25-i32.34. Ramesh BM, Beattie TS, Shajy I, Washington R, Jagannathan L, Reza-Paul S,Blanchard JF, Moses S: Changes in risk behaviours and prevalence ofsexually transmitted infections following HIV preventive interventionsamong female sex workers in five districts in Karnataka state, southIndia. Sex Transm Infect 2010, 86(Suppl 1):i17-i24.35. HIV Fact Sheets: Based on HIV Sentinel Surveillance Data in India, 2003-2006.. New Delhi, India: National AIDS Control Organization (NACO); 2007.36. Pickles M, Foss AM, Vickerman P, Deering K, Verma S, Demers E,Washington R, Ramesh B, Moses S, Blanchard J, et al: Interim modellinganalysis to validate reported increases in condom use and assess HIVinfections averted among female sex workers and clients in southernIndia following a targeted HIV prevention programme. Sex Transm Infect2010, 86(Suppl 1):i33-i43.37. Dureau J, Boily M-C, Vickerman P, Pickles M, S I, Ramesh B, Verma S,Moses S, Alary M, Kalogeropoulos K: What has been the impact of theAvahan HIV intervention on condom use among female sex workers?Insights from a fully Bayesian inference methodology. InternationalSociety for Sexually Transmitted Diseases Research; Quebec City, Canada 2011.38. Wang Y, Li B, Song DM, Ding GY, Cathy E: Power relation and condomuse in commercial sex behaviours. Biomed Environ Sci 2007, 20:302-306.39. Wojcicki JM, Malala J: Condom use, power and HIV/AIDS risk: sex-workersbargain for survival in Hillbrow/Joubert Park/Berea, Johannesburg. SocSci Med 2001, 53:99-121.40. Luke N: Exchange and condom use in informal sexual relationships inurban Kenya. Economic Development and Cultural Change 2006, 54:319-348[http://www.jstor.org/pss/10.1086/497011].41. Luke N: Confronting the ‘sugar daddy’ stereotype: age and economicasymmetries and risky sexual behavior in urban Kenya. Int Fam PlanPerspect 2005, 31:6-14.42. Varga CA: The condom conundrum: barriers to condom use amongcommercial sex workers in Durban, South Africa. Afr J Reprod Health 1997,1:74-88.43. Abdool Karim Q, Abdool Karim SS, Frohlich JA, Grobler AC, Baxter C,Mansoor LE, Kharsany ABM, Sibeko S, Mlisana KP, Omar Z, et al:Effectiveness and safety of tenofovir gel, an antiretroviral microbicide,for the prevention of HIV infection in women. Science 2010,329:1168-1174.44. Hanck SE, Blankenship KM, Irwin KS, West BS, Kershaw T: Assessment ofself-reported sexual behavior and condom use among female sexworkers in India using a polling box approach: a preliminary report. SexTransm Dis 2008, 35:489-494.45. Vickerman P, Foss AM, Pickles M, Deering K, Verma S, Eric Demers,Lowndes CM, Moses S, Alary M, Boily MC: To what extent is the HIVepidemic driven by commercial sex? A modelling analysis. AIDS 2010,24:2563-2572.46. Boily MC, Pickles M, Vickerman P, Buzdugan R, Isac S, Deering KN,Blanchard JF, Moses S, Lowndes CM, Ramesh BM, et al: Usingmathematical modelling to investigate the plausibility of attributingobserved antenatal clinic declines to a female sex worker interventionin Karnataka state, India. AIDS 2008, 22(Suppl 5):S149-S164.47. Lowndes CM, Alary M, Meda H, Gnintoungbé CA, Mukenge-Tshibaka L,Adjovi C, Buvé A, Morison L, Laourou M, Kanhonou L, Anagonou S: Role ofcore and bridging groups in the transmission dynamics of HIV and STIsin Cotonou, Benin, West Africa. Sex Transm Infect 2002, 78(Suppl 1):i69-i77.48. Lowndes CM, Alary M, Gnintoungbé CAB, Bédard E, Mukenge L, Geraldo N,Jossou P, Lafia E, Bernier F, Baganizi E, et al: Management of sexuallytransmitted diseases and HIV prevention in men at high risk: targetingclients and non-paying sexual partners of female sex workers in Benin.AIDS 2000, 14:2523-2534.49. Coffee M, Lurie MN, Garnett GP: Modelling the impact of migration onthe HIV epidemic in South Africa. AIDS 2007, 21:343-350.50. Deering KN, Vickerman P, Moses S, Ramesh BM, Blanchard JF, Boily MC: Theimpact of out-migrants and out-migration on the HIV/AIDS epidemic: acase study from south-west India. AIDS 2008, 22(Suppl 5):S165-S181.Deering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 13 of 1451. Female Sex Workers and Their Clients in the Epidemiology and Controlof Sexually Transmitted Diseases.. New York, USA: McGraw Hill;BlanchardJF, Moses S , 4 2007:.doi:10.1186/1471-2458-11-S6-S8Cite this article as: Deering et al.: A dose-response relationship betweenexposure to a large-scale HIV preventive intervention and consistentcondom use with different sexual partners of female sex workers insouthern India. BMC Public Health 2011 11(Suppl 6):S8.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitDeering et al. BMC Public Health 2011, 11(Suppl 6):S8http://www.biomedcentral.com/1471-2458/11/S6/S8Page 14 of 14

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0223644/manifest

Comment

Related Items