Secondary Logo

Journal Logo

Supplement Article

Understanding the Intention to Undergo Regular HIV Testing Among Female Sex Workers in Benin

A Key Issue for Entry Into HIV Care

Batona, Georges MSc*,†; Gagnon, Marie-Pierre PhD*,†; Simonyan, David A. MSc*; Guedou, Fernand A. MD, PhD*,‡; Alary, Michel MD, PhD*,§,‖

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: March 1, 2015 - Volume 68 - Issue - p S206-S212
doi: 10.1097/QAI.0000000000000452

Abstract

INTRODUCTION

The HIV epidemic in West Africa is mainly concentrated among high-risk groups, especially female sex workers (FSWs) and their sexual partners. Formal commercial sex triggers a higher incidence of HIV acquisition and onward transmission than would be suggested by existing approaches to assessing HIV epidemics.1 In sub-Saharan Africa, FSWs are disproportionally affected by HIV. Women already account for the majority (60%) of Africans living with HIV, and FSWs are, on average, 14 times more likely to be infected with HIV than other women.2

Although, in Benin as a whole, HIV prevalence is moderate (1.2%), it is highly concentrated among FSWs (20.9%) and their sexual partners (3.9%).3,4 This context justifies preventive interventions supported by the National AIDS Control Program (NACP), which specifically target FSWs. FSWs are offered dedicated health services, including HIV and sexually transmitted infections (STI) services, referred to as “service adapté” (SA). SAs are specific units recognized by the NACP and dedicated to FSWs, offering primary health care, free STI care, HIV voluntary counseling and testing, and referral to HIV care centers. One large SA in Cotonou is also an HIV care centre. HIV voluntary counseling and testing (VCT) is highly promoted in this population as it constitutes an entry point for prevention, treatment, support, and HIV care. The health benefits associated with VCT are supported by studies demonstrating its central role in providing better HIV care and potentially reducing risk behaviors.5,6 Because of their continuous exposure to the risk of HIV infection related to sex work (sex with multiple partners, condom breaking, or nonuse of condoms), FSWs are particularly exhorted to undergo periodic (every 3 months) HIV testing to know their serological status.7

Strengthening the promotion of HIV testing among FSWs becomes increasingly important with the promising development of innovative prevention approaches. Early HIV diagnosis and ARV treatment can reduce the risk of HIV transmission by about 96% among discordant couples.8 Individual clinical benefits and potentially enormous public health benefits in slowing the spread of the infection will be possible only if the most affected groups, who play a pivotal role in transmitting HIV, make greater use of testing services and become aware of their serological status early.

Despite the fact that a high proportion of FSWs in Benin (77.6%) reported having been HIV tested,3 this behavior is still not at all routine. Several factors may influence the intention to adopt this behavior on a regular basis. Knowledge, individual beliefs, personal experiences, and the quality of interaction with health care providers in SAs, and social and work environment, can determine FSWs' motivation to be HIV tested regularly at an SA. In addition, program evaluations in Benin show that utilization rates for these SAs remain low compared with the estimated number of FSWs.9 Activities promoting HIV testing are sometimes inadequate at the local level, and existing practices fall short of recommended best practices.

No study has been conducted in Africa on determinants of FSWs' behavioral intention to be HIV tested on a regular basis. The intervention approach, oriented toward individuals and aiming at a planned change in health behavior, should be based on previous understanding of factors explaining the phenomenon of interest in a defined population and a specific context. This understanding is necessary to choose the most appropriate methods and adapt the content of messages to the target population's characteristics and needs.10

To better understand the factors affecting FSWs' intention to be HIV tested at an SA in the next 3 months, we conducted a cross-sectional study based on an extension of Ajzen's Theory of Planned Behavior.11 The study was conducted in 6 cities in Benin (Cotonou, Abomey-Calavi, Abomey-Bohicon, Parakou, Porto-Novo, and Savalou) where a high number of FSWs were benefiting from preventive programs under the guidance of NACP.

METHODS

Theoretical Framework

This study is based on an extension of the Theory of Planned Behavior (TPB), which is particularly useful in predicting health-related behaviors.12,13 The TPB was introduced to consider behaviors that are not entirely under the individual's volitional control, that is, when there are obstacles to the adoption of the behavior. Many behaviors, such as those related to HIV prevention and access to care, require resources, skills, and opportunities over which individuals lack complete control. Ajzen11 suggested that behavior is under the combined influence of 2 factors: intention (INT) and perceived behavioral control (PBC). According to the TPB, INT is the main direct determinant of the behavior. It represents the expression of an individual's motivation to adopt a given behavior or not. Three factors define INT as follows: attitude (ATT), subjective norms (SN), and PBC.11 ATT is the result of the subjective analysis of the advantages and disadvantages of adopting a given behavior and is determined indirectly by behavioral beliefs (BB).11 SN reflects an individual's perception of the expectations of all significant persons in their environment regarding the adoption of a given behavior. This variable may also be defined indirectly by normative beliefs (NB) that refer to the perception of the degree of approval by persons or groups important to the individual in question.11 We measured NB, instead of directly measuring SN, as the 2 were previously judged equivalent by Gagné and Godin.14 The third factor is PBC, which represents the individual's perceived capacity to adopt the behavior. PBC is also determined by control beliefs (CB) that represent the extent to which particular factors can hamper or facilitate the behavior.

According to the TPB, variables related to the sociodemographic context of individuals do not directly affect their intention but rather have an impact through beliefs and other determinants of the theory.11

The TPB has already been successfully used to understand HIV testing among other population groups.15–18 Moreover, the TPB is open to the integration of constructs from other theories, insofar as they can significantly increase the explained variance in intention.19 In addition, various authors suggest that a combination of theoretical models could prove to be effective in predicting and understanding health-related behaviors.20,21

Different theoretical variables may influence HIV-testing behavior; thus we added 4 variables from other theories to our theoretical framework: descriptive norms (DN), habits (HAB), moral norms (MN), and perceived risk (PR). DN refers to the perceived prevalence of behavior in the target population.22 HAB and MN are 2 theoretical constructs proposed by Triandis.23 Both constructs were found to have residual effects on intention and behavior in studies of HIV prevention behaviors.19,24 HAB refers to having adopted a behavior with some frequency in the past. MN is the personal assessment of the appropriateness of behavior. PR is one of the main constructs from the health belief model (HBM),25 showing a link to health-related behaviors, especially in the prevention of HIV.26 PR is the self-reported measurement of PR, constituting the level of risk that an individual assigns to a behaviour.27Figure 1 presents the integrated theoretical framework used in this study.

F1-18
FIGURE 1:
Theoretical framework: extended version of TPB. Constructs inside the circle were added to TPB (adapted from Ajzen11). Adaptations are themselves works protected by copyright. So in order to publish this adaptation, authorization must be obtained both from the owner of the copyright in the original work and from the owner of copyright in the translation or adaptation.

Population Studied and Sample

The population under examination consists of FSWs from diverse sex work sites of Beninese cities with a high density of FSWs. We used cluster sampling, with the probability of a given sex work site being selected based on its size, as determined by the most recent sex work mapping carried out in Benin by NACP. This mapping was based on an exhaustive nationwide census of FSWs. Details on the mapping process may be found in Supplemental Methods, Supplemental Digital Content, https://links.lww.com/QAI/A598.

Data Collection

Data 2collection was carried out in November 2013, through face-to-face interviews by investigators experienced in conducting behavioral studies in the sex work environment. The questionnaire was administered to all eligible FSWs present on selected sites when visited, after they provided signed informed consent. To ensure anonymity, a coding number was allocated to each questionnaire. The study was approved by the ethics committee of the Centre Hospitalier Universitaire de Québec (Quebec City, Canada) and by the National Health Research Ethics Committee in Benin.

Questionnaire Development

We developed the questionnaire in accordance with the methodology recommended by Ajzen,28 applying the approach described by Gagné and Godin.11,29 The process is described in detail in paragraph 2 of the Supplemental Methods (see Supplemental Digital Content,https://links.lww.com/QAI/A598). A preliminary version of the questionnaire was developed and tested for internal consistency (in terms of Cronbach's alpha or Spearman's correlation if fewer than 3 items) and temporal stability. Cronbach's alpha coefficient was 0.84, 0.77, 0.65, and 0.76 for INT, CB, NB, and BB, and Spearman's correlation was 0.80, 0.66, 0.60, and 0.64 for PBC, MN, DN, and ATT, respectively. The temporal stability analysis scores obtained were very satisfactory (0.53–0.85), and the constructs were quite internally consistent. For most items, a Likert-type scale with 5 levels of response (strongly disagree [1] to strongly agree [5]), illustrated by an image increasing in size was used. This final questionnaire is included in the Supplemental Methods (see Supplemental Digital Content,https://links.lww.com/QAI/A598).

Statistical Analysis

We first performed descriptive analyses (frequencies, mean values, and SDs) for all variables, using SAS 9.3 (SAS Institute Inc., Cary, NC). Next, we applied structural equation modeling using SPSS 18 Amos (Amos Development Corporation, Crawfordville, FL) to identify theoretical variables that significantly predict intention and determine respective weight. We checked the final model fit criteria using the overall model fit χ2 test (chi-square divided by its degrees of freedom [CMIN/DF]), the comparative fit index, and the Tucker–Lewis coefficient. Finally, we applied multiple group analysis to test the modifying effects of the following external variables on the final model: age (≤25 vs. >25 years); nationality (Beninese vs. foreigner); municipality (Cotonou vs. outside Cotonou); education level (educated vs. unschooled); religion (Christian vs. other religion); duration in sex work (less than 1 vs. 1 year or more); involvement as peer educator (involved as peer educator vs. not involved as peer educator); HIV testing history (HIV tested already vs. not HIV tested); and parental status (biological mother of at least 1 child vs. no biological children).

RESULTS

Sociodemographic and Behavioral Characteristics

A total of 455 FSWs were invited to participate in the study, of whom 450 agreed for an acceptance rate of 98.9%. Among these, 3 FSWs were excluded from further analyses because their positive HIV status was already known.

Study participants' sociodemographic and behavioral characteristics are provided in Tables 1 and 2, respectively. The average age was 30.1 years (SD = 8.4 years). FSWs were mostly from Benin, followed by Nigeria. Almost half (49%) of FSWs were divorced, and 72.7% were biological mothers.

T1-18
TABLE 1:
Sociodemographic Characteristics and Description of Psychosocial Variables Related to the Intention of Being HIV Tested in the Next 3 Months Among FSWs in Benin (n = 447)
T2-18
TABLE 2:
Behavioral Characteristics of FSWs in Benin Related to HIV Testing History, Perceived Risk of Being Infected by HIV, and the Intention of Being HIV Tested (n = 447)

A total of 87% of FSWs had already been HIV tested during their lifetime, and 65.3% had had at least 1 HIV test during the past year. Among those who had already been tested, 40% reported that they were last tested in the previous 3 months and 21% between the last 3 and the last 6 months. A strong intention to undergo HIV testing in the next 3 months, that is, a response of 4 or higher on the 5-point Likert-scale, was expressed by 69.4% of FSWs.

Determinants of Intention

As a first step, we tested the predictive ability of direct variables from the TPB (ATT, SN, and PBC). The TPB variables predicted 40% of the variance in intention, with PBC (β = 0.48, P < 0.001) and ATT (β = 0.20, P < 0.001) as the main predictors.

In a second step, we integrated the TPB variables and external variables (DN, HAB, PR, and MN) in a single model. The extended TPB model explained over half of the variance in intention (R2 = 55%). In this final model, the significant determinants of intention included PBC (β = 0.26, P < 0.001); DN (β = 0.24, P < 0.001); CB (β = 0.22, P < 0.001); HAB (β = 0.20, P < 0.001); ATT (β = 0.12, P = 0.01); PR (β = 0.07, P = 0.03); and NB (β = −0.07, P = 0.03) (Table 3). MN did not contribute to the variance of intention and was removed from the final model. Figure 2 shows the final path analysis model. The fit criteria of this final model were all satisfactory: overall model fit χ2 test was nonsignificant (CMIN = 16.225, DF = 9, CMIN/DF = 1.803, P = 0.062); comparative fit index = 0.995; Tucker-Lewis Index = 0.996; root mean square error approximation = 0.042.

T3-18
TABLE 3:
Parameters for Variables Predicting Intention to be HIV Tested in the Next 3 Months Among FSWs in Benin (Standardized Regression Weights)
F2-18
FIGURE 2:
Final theoretical framework including significant variables predicting intention to be HIV tested in the next 3 months among FSWs in Benin.

Our model also shows that the indirect variables (BB and CB) determine their corresponding direct variables (ATT and PBC), which in turn determine INT. Furthermore, CB has a significant direct and positive effect on intention, whereas NB (used to measure SN) is significantly and negatively associated with intention (β = −0.07, P = 0.03).

The multigroup analysis of the influence of external variables (demographic and behavioral) on the final predictive model shows that 2 variables are significant modifying factors of the associations between the theoretical variables of this model and intention. These are lifetime history of HIV testing (χ2 = 106.405; DF = 10, P < 0.001) and having children (χ2 = 18.582; DF = 10, P < 0.045).

On the one hand, we observe that DN, NB, and PR are strong negative predictors of intention among FSWs who have never been HIV tested. However, PBC, CB, DN, and HAB (in descending order of importance) are positive predictors of intention among those who have already been HIV tested. On the other hand, the importance of these predictors changes for a biological mother: DN, PBC, CB, HAB, and ATT (in descending order of importance). ATT no longer appears to be a significant predictor of intention among childless FSWs. Furthermore, the weight of the other determinants of intention differs for women without children and mothers (more details of the results are provided in Tables S1 and S2; see Supplemental Digital Content,https://links.lww.com/QAI/A598).

DISCUSSION

In this study, we find that 87.1% of FSWs have been HIV tested in their lifetime. This proportion is similar to the percentage reported in a 2012 national biological and behavioral survey (77.6%).3 The percentage observed in Benin is much higher than reported elsewhere, such as in Guinea26 and China,30 where only 26.6% and 48% of FSWs, respectively, reported earlier HIV testing.

Our study revealed a higher proportion of FSWs in Benin who had had an HIV test during the past year (65.3%) than that reported by UNAIDS and WHO in 2009.31 In the latter report, in pooled data from 45 countries, the median proportion of FSWs who knew their HIV status through a recent test was estimated at 38%. However, this proportion was slightly higher in sub-Saharan Africa (39%).31

These encouraging results in Benin may be, at least partially, related to successive interventions targeting FSWs, particularly actively supported by the Canadian project initiated in 1992 and lasting until 2006. Later, the project was continued and expanded by the NACP. In addition, pregnant women, whether FSWs or not, are offered HIV testing by the mother-to-child transmission prevention program. Because 72.7% of our study participants were biological mothers, many of them were probably offered the HIV test during their prenatal visit.

Despite the seemingly high proportion observed in Benin, it is important to mention that prescheduled periodic HIV testing would be more beneficial for FSWs than one-time testing, owing to their continuous exposure to the infection. This would be a key step toward the early detection of the disease and, therefore, toward early care and ultimately a breaking of the transmission cycle.

Our results indicate that 55% of the variance of intention can be explained by this model, and that the most important determinants in descending order of effect are PBC, DN, CB, HAB, ATT, PR, and NB.

To our knowledge, this is the first study conducted among FSWs using the TPB to explore determinants of intention to undergo regular HIV testing. Previously published studies that applied TPB were conducted among groups other than FSWs.15–18 They showed its effectiveness in predicting the intention to be HIV tested.

Several previous studies targeting FSWs were not based on psychosocial theories.32–35 These studies focused mainly on identifying either facilitating factors or barriers to adoption of VCT.

Despite differences in methodological and theoretical approaches, there are some convergences between our results and previous studies in Africa26,36 and elsewhere32,33,37,38 regarding beliefs associated with PBC among FSWs. The importance of beliefs perceived by FSWs as barriers or facilitators of intention or behavior has already been underlined in previous studies.26,33,36,38 Some FSWs who had never visited the SA were reluctant to undergo VCT, citing fear of breaches of confidentiality because of the stigma associated with HIV and AIDS.26

Lack of confidentiality may undermine VCT. This is particularly crucial for vulnerable populations, such as FSWs, which would otherwise bear the double burden of social exclusion and stigma.36 This factor was considered a salient belief that underlies PBC and CB and has been reported several times in the literature26,32,33,37,38 as an obstacle to the use of VCT services. However, the perception of support from peers, managers of prostitution sites, and close family members is a factor facilitating the acceptance of VCT.32,33,37,38 Wang et al33 reported a positive peer influence that could promote utilization of VCT clinics. This supports the influence of normative factors (subjective or DN) in using VCT services.

In our study, PR explained a small proportion of the variance in intention. These results corroborate those of Aho et al26 who report the high acceptability of VCT associated with a self-perceived high risk of HIV among FSWs. Wang et al33 also emphasized that PR was a motivating factor regarding undergoing VCT. However, another study39 suggested that people who perceive themselves at risk are less motivated to be tested although they are more likely to be infected.

ATT is a significant determinant of intention in our study. These results are in line with Beattie et al37 and with the findings of Wang et al32 and Wang et al,33 which reveal that knowledge about HIV and VCT and the benefits of HIV testing (eg, perceived benefits of help in case of the positive test) is a factor favoring its use, whereas a low level of knowledge, along with fear, anxiety, and perceived negative consequences (rejection, stigma, and discrimination) are the main barriers. As in our study, having already undergone VCT is a factor associated with FSWs' motivation to go for VCT.33

The observed modifying effect of lifetime history testing in the overall prediction model indicates that future HIV testing promoting interventions should target and adapt message content according to the personal history of HIV testing. Indeed, the effect of CB and DN on INT to be tested during the next 3 months changes significantly and positively among FSWs with any history of HIV testing. This suggests that previous testing could have a positive influence on CB of FSWs, by increasing their perception of being able to undergo HIV testing, as well as their beliefs about the behavior of their peers (DN).

The main strength of our study is the use of an extended model of the TPB that appeared relevant in explaining FSWs' intention to undergo HIV testing. The addition of other variables enhanced the prediction of intention, after considering the relative importance of direct variables in the original theoretical framework. In addition, the large sample size reinforced our results' validity and statistical power. Furthermore, the data collection procedures (direct personal interviews) and the study instrument used (a validated questionnaire with a visual scale) helped to limit missing data and to improve the quality of the data, despite the difficult prostitution environment. The final model fitted our data well, allowing us to obtain reasonable estimates of the model and achieve good internal validity.

Despite our efforts to limit recall bias and social desirability when collecting data, they cannot be totally ruled out. Indeed, during the face-to-face interviews with FSWs, who are often stigmatized and socially marginalized, the tendency to acquiescence, impatience to finish with the interview so as not to miss customers seeking their services by telephone, and/or social desirability may have affected participants' responses. We had no knowledge of a differential effect of this type of bias on the structural modeling analyses. However, such a bias may have led to an overestimation or underestimation of the scores of the variables.

This study is the first of its kind in Benin, and its results may be useful in developing the content of interventions to maintain and increase regular HIV testing among FSWs. Our findings suggest that a theory-based program of educational intervention should focus on reducing practical barriers and overcoming perceived obstacles to HIV testing. In addition, programs should strongly consider the descriptive norm as an important determinant of FSW intention to be HIV tested, using peer educators' influence and experience to convince them to undergo regular HIV testing.

ACKNOWLEDGMENTS

The authors thank the participants for the time they devoted to the study and all stakeholders in the field who were involved in this study.

REFERENCES

1. Alary M, Mishra S, Moses S, et al.. The importance of focused HIV prevention programming for female sex workers and clients in the response to the HIV epidemic in African countries. Sex Transm Dis. 2014;41(suppl l):S14.
2. Baral S, Beyrer C, Muessig K, et al.. Burden of HIV among female sex workers in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Infect Dis. 2012;12:538–549.
3. Direction Nationale de la Protection Sanitaire/Programme National de Lutte contre le SIDA et les IST. Enquête de Surveillance de Deuxième Génération des IST/VIH/SIDA au Bénin (ESDG-2012): travailleuses de Sexe, Routiers et Clients des Travailleuses de Sexe, Rapport Final. Cotonou, Ministère de la Santé du Bénin. 2012.
4. Direction Nationale de la Protection Sanitaire/Programme National de Lutte contre le SIDA et les IST. Enquête de Surveillance de Deuxième Génération des IST/VIH/SIDA au Bénin (ESDG-2008): Travailleuses de Sexe, Routiers et Clients des Travailleuses de Sexe, Rapport Final. Cotonou, Ministère de la Santé du Bénin. 2008.
5. Denison JA, OReilly KR, Schmid GP, et al.. HIV voluntary counseling and testing and behavioral risk reduction in developing countries: a meta-analysis, 1990–2005. AIDS Behav. 2008;12:363–373.
6. Fonner VA, Denison J, Kennedy CE, et al.. Voluntary counseling and testing (VCT) for changing HIV related risk behavior in developing countries. Cochrane Database Syst Rev. 2012;9:CD001224.
7. World Health Organization. Delivering HIV Test Results and Messages for Re-Testing and Counselling in Adults. Geneva, Switzerland: WHO Library Cataloguing-in-publication Data; 2010.
8. Cohen MS, Chen YQ, McCauley M, et al.. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505.
9. Semini I, Batona G, Lafrance C, et al.. Implementing for results: program analysis of the HIV/STI interventions for sex workers in Benin. AIDS Care. 2013;25(suppl 1):S30–S39.
10. Bartholomew LK, Parcel JS, Kok G, Gottlieb NH. Intervention Mapping: Designing Theory- and Evidence-based Health Promotion Programs. 3rd ed. San Francisco, CA: Jossey Bass; 2011.
11. Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Process. 1991;50:179–211.
12. Armitage CJ, Conner M. Efficacy of the theory of planned behavior: a meta-analytic review. Br J Soc Psychol. 2001;40:471–499.
13. Godin G, Kok G. The theory of planned behavior: a review of its applications to health-related behaviors. Am J Health Promot. 1996;11:87–98.
14. Gagné C, Godin G. The theory of planned behavior: some measurement issues concerning belief-based variables. J Appl Social Psychol. 2000;30:2173–2193.
15. Kakoko DC, Astrom AN, Lugoe WL, et al.. Predicting intended use of voluntary HIV counselling and testing services among Tanzanian teachers using the theory of planned behaviour. Soc Sci Med. 2006;63:991–999.
16. Hadera HG, Boer H, Kuiper WAJ. Using the theory of planned behaviour to understand the motivation to learn about HIV/AIDS prevention among adolescents in Tigray, Ethiopia. AIDS Care. 2007;19:895–900.
17. Astrom AN, Nasir EF. Predicting intention to treat HIV-infected patients among Tanzanian and Sudanese medical and dental students using the theory of planned behaviour-a cross sectional study. BMC Health Serv Res. 2009;9:213.
18. Omer S, Haidar J. Applicability of theory of planned behavior in predicting intended use of voluntary HIV counseling and testing services among teachers of Harari region, Ethiopia. Ethiop J Health Dev. 2010;24:96–102.
19. Conner M, Armitage CJ. Extending the theory of planned Behavior: a review and avenues for further research. J Appl Social Psychol. 1998;28:1429–1464.
20. Fishbein M. The role of theory in HIV prevention. AIDS Care. 2000;12:273–278.
21. Godin G, Vézina-Im LA. Les théories de prédiction, dans Godin 2012. Les comportements dans le domaine de la santé. Comprendre pour mieux intervenir. Les presses de l'Université de Montréal 2012:11–40.
22. Sheeran P, Orbell S. Augmenting the theory of planned behavior: Roles for anticipated regret and descriptive norms. J Appl Social Psychol. 1999;29:2107–2142.
23. Triandis HC. Values, attitudes and interpersonal behaviour. In: Page MM, ed. Nebraska Symposium on Motivation, 1979: Beliefs, Attitudes and Values. Lincoln, NE: University of Nebraska Press; 1980.
24. Rivis A, Sheeran P. Descriptive norms as an additional predictor in the theory of planned behaviour: a meta-analysis. Curr Psychol Dev Learn Personal Social. 2003;22:218–233.
25. Rosenstock IM, Strecher VJ, Becker MH. Social learning theory and the health belief model. Health Educ Q. 1988;15:175–183.
26. Aho J, Nguyen VK, Diakité SL, et al.. High acceptability of HIV voluntary counselling and testing among female sex workers: impact of individual and social factors. HIV Med. 2012;13:156–165.
27. Fichtner RR, Wolitski RJ, Johnson WD, et al.. Influence of perceived and assessed risk on STD clinic clients' acceptance of HIV testing, return for test results, and HIV serostatus. Psychol Health Med. 1996;1:83–98.
28. Ajzen I. Constructing a TPB questionnaire: conceptual and methodological considerations. 2006. Available at: http://www-unix.oit.umass.edu/∼aizen/pdf/tpb.measurement.pdf. Accessed September 15, 2008.
29. Gagné C, Godin G. La mesure des variables théoriques et des comportements, dans Godin 2012. Les comportments dans le domaine de la santé. Comprendre pour mieux intervenir. Les presses de l'Université de Montréal. 2012:231–292.
30. Hong Y, Zhang C, Li X, et al.. HIV testing behaviors among female sex workers in Southwest China. AIDS Behav. 2012;16:44–52.
31. OMS, ONUSIDA, UNICEF. Vers un accès universel: étendre les interventions prioritaires liées au VIH/SIDA dans le secteur de la santé : rapport de situation. 2009.
32. Wang Y, Li B, Zheng J, et al.. Factors related to female sex workers' willingness to utilize VCT service: a qualitative study in Jinan city, northern China. AIDS Behav. 2009;13:866–872.
33. Wang Y, Bing L, Jingbin P, et al.. Factors associated with utilization of a free HIV VCT clinic by female sex workers in Jinan city, Northern China. AIDS Behav. 2010;15:702–710.
34. Ngo AD, Ratliff EA, Mccurdy SA, et al.. Health-seeking behaviour for sexually transmitted infections and HIV testing among female sex workers in Vietnam. AIDS Care. 2007;19:878–887.
35. Hong Y, Zhang C, Li X, et al.. HIV testing behaviors among female sex workers in Southwest China. AIDS Behav. 2012;16:44–52.
36. Ogbogu IO, Ogbogu OI, Odutolu O, eds. Trust and Confidentiality in Use of VCT Service by Female Sex Workers. XVth International Conference on AIDS. Bangkok, Thailand. 2004. Abstract C12092.
37. Beattie TS, Bhattacharjee P, Suresh M, et al.. Personal, interpersonal and structural challenges to accessing HIV testing, treatment and care services among female sex workers, men who have sex with men and transgenders in Karnataka state, South India. J Epidemiol Community Health. 2012;66(suppl 2):ii42–ii48.
38. Chakrapani V, Shanmugam M, Michael S, et al.. Testing Barriers and Facilitators Among Population At-Risk in Chennai, India. Chennai, India: India Network for People Living with HIV (INP+); 2008.
39. Johnston L, O'Bra H, Chopra M, et al.. The associations of voluntary counseling and testing acceptance and the perceived likelihood of being HIV-infected among men with multiple sex partners in a South African township. AIDS Behav. 2010;14:922–931.
Keywords:

HIV testing; intention; Theory of Planned Behavior; sex workers; Benin

Supplemental Digital Content

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.