Sex workers clients in Italy: results of a phone survey on HIV risk behaviour and perception



Ilaria MulieriI; Flavia SantiI; Anna ColucciI; Emanuele Fanales-BelasioII; Pietro GalloII; Anna Maria LuziI

IDipartimento di Malattie Infettive, Parassitarie ed Immunomediate, Istituto Superiore di Sanità, Rome, Italy
IICentro Nazionale AIDS, Istituto Superiore di Sanità, Rome, Italy





INTRODUCTION: Sex workers (SW) clients represent a bridge population for HIV transmission from high risk to low risk general population.
MATERIALS AND METHODS: A cross-sectional anonymous telephone survey was carried out at the AIDS and Sexually Transmitted Infections Helpline of National Institute of Health in Italy. The questionnaire was proposed on a voluntary basis to a sample of 119 subjects from helpline users.
RESULTS: The 119 participants were all males, aged between 19 and 59 years and mostly accessed female prostitutes. Vaginal intercourses with SW were more frequently reported, followed by passive oral, active oral sex and active anal intercourses. Cumulatively, 86.6% and 84.6% of vaginal and anal intercourses were respectively reported as regularly protected by condom.
DISCUSSION: The telephone interview allowed an eased access, a high response rate and a standardised evaluation of questions
It is necessary a constant monitoring of the characteristics, behaviour, risk perception and testing of SW clients in Italian and other populations.

Key words: HIV; clients of sex workers; commercial sex; survey HIV; risk perception




Since the early years of epidemic, HIV infection has progressively expanded towards population groups, as heterosexual men and women, who do not perceive themselves as "at risk" and do not consider the possibility of HIV testing [1].

This scenario is particularly alarming since, within the last years, the proportion of people who discovered to have HIV infection in an advanced stage has gradually increased up to more than half of all diagnosed AIDS cases. Data of the National Institute of Health in Italy (Istituto Superiore di Sanità, ISS) AIDS Operational Centre (COA) have shown that, among new infections, the distribution of HIV ways of transmission has undergone remarkable changes in the last 28 years: injecting drug users (IDUs) have decreased from 76.2% in 1985 to 5.3% in 2012, while heterosexuals and men who have sex with men (MSM) increased from 1.7% to 42.7% and from 6.3% to 37.9%, respectively [2].

The persons who were unaware of their HIV-positive status have prevalently got infected through sexual intercourses. Since sexual behaviour is a key factor in the transmission of HIV and sexually transmitted diseases (STDs), sex workers (SW) including prostitutes and transsexuals are often referred as high-risk group. It has been documented that SW have a higher prevalence of sexually transmitted diseases and have a key role in the incidence of HIV and STDs in the general population, constituting a cause of concern for both public health and economic resources exploited for specific preventive interventions [3].

Thus, SW are a group of population highly vulnerable to both HIV infection and STDs and, through clients, can transmit infection to general population. Indeed, a comprehensive study conducted in China shows that SW clients represent a real bridge-population for the transmission of HIV from a high risk group to low risk general population [4]. SW clients have therefore become a key target for prevention programs, including the promotion of HIV testing [4-6].

HIV testing of general population is however, a hard-to-get goal due to the unawareness of the infection risk. The latest data from the COA showed that among persons who underwent HIV testing in 2012, 22.4% did it due to the presence of HIV-related symptoms, 16.5% following potential risky behaviours, 15.0% within investigations for other pathologies or surgery-associated procedures, 4.7% within routine check-up, 2.9% following the diagnosis of a STD, 2.9% within gynaecological check-up during pregnancy/childbirth/abortion, 2.7% following unprotected sex with a HIV-positive partner, 2.3% within routine checks in therapeutic communities or in care services for injection drug users (IDU), 2.1% within blood donation screening procedures, and 2.7% for other reasons. Cumulatively, it was computed that only a small percentage (16.5%) of tested people had a clear perception of HIV infection risk.

Data on both HIV infection prevalence and incidence may be underestimated due to people inadequate perception of risk, availability to testing and access to diagnostic services. There is in fact clear evidence that HIV testing in European countries is still strongly associated with a history of STDs in people who had a poor self-perception of risk [6].

For this reason, specialists within STDs strongly encourage the broadening of HIV testing to people who have risk due to unprotected sexual behaviours [7].

In some studies, based on prevailing theories on health behaviours such as the Health Action Process Approach (HAPA) and the Health Belief Model, the perception of HIV infection risk has been measured by the assessment of self-perceived likelihood of contracting the infection [8-12]. This experimental approach can be effectively applied in order to understand the behaviour related to HIV or other sexually transmitted diseases and to support both specifically targeted interventions.

The perception of HIV risk is one of the most studied research topics [13-22]. HIV risk perception should - at least theoretically - induce individuals to take precautionary measures and/or to refrain from risky behaviours. Therefore, attempts to increase both the individual awareness and risk perception are common approaches to interventions for HIV prevention. In particular, the relationship between the perception of risk and risk behaviour has not been fully investigated, probably due to conceptual problems. It would be therefore necessary to pay special attention to clients of SW which are neither a homogeneous and well-outlined population nor easily identifiable and accessible within specific preventive interventions [23].



Study sample

An anonymous cross-sectional telephone survey has been carried out for 3 months (December 5th, 2012 - March 5th, 2013 in the usual weekly schedule Monday to Friday from 13:00 to 18:00) within the AIDS and Sexually Transmitted Infections Helpline (Telefono Verde AIDS/IST, TVA-IST) of National Institute of Health in Italy (Istituto Superiore di Sanità, ISS), co-financed by the Italian Ministry of Health and managed by the Psycho-Socio-Behavioural-Research-Communication and Education Operative Unit (UO RCF) of Department of Infectious, Parasitic and Immune-Mediated Diseases of ISS. Population included into the study consisted in persons, aged between 19 and 59 years, reporting to have been clients of prostitutes and/or transsexuals. A total number of 155 persons responding to inclusion criteria and selected through a non-probability sampling type, were asked to enter the study and 119 persons, who gave their consent, were specifically recruited.

Interview and questionnaire

Phone counselling intervention was performed as for all calls at TVA-IST Helpline, according to a standardized procedure. At the completion of the phone counselling intervention, an anonymous questionnaire was proposed, on voluntary basis, to persons/users upon adequate information on both the characteristics and purpose of the study. The questionnaire included 13 questions related to: a) socio-demographic information (province of origin, gender, age, marital status, education level, employment); b) sexual behaviour in the last six months (type of intercourses, degree of protection); c) degree of perception of risk of HIV or other STDs; d) HIV testing (reasons to have or not HIV testing; type and availability of diagnostic services structures); e) history of STDs.

Collection and analysis of data

Data from the interview and questionnaires were adequately stored within a specifically designed data-entry software ensuring the protection of sensitive data. The demographic characteristics, referred "potentially at risk" behaviours, sexual risk perception and HIV testing of the participants were analyzed by descriptive statistics. Continuous numerical variables were displayed as median, upper quartile and lower, while categorical variables were displayed as observed frequencies and proportions. The Chi-squared test was used to assess statistical significance in relation between condom use and educational level or presence of a stable relationship. It is assumed that statistical significance occurs where p-value is lower than 0.05.

All analyses were performed using STATA Software (version 13.0).



Characteristics of study sample

Thirty-six out of 155 subjects selected during the session period did not agree to answer to the questionnaire due to reported lack of time, lack of will or anxiety condition.

The socio-demographic characteristics of enrolled 119 persons are shown in Table 1.

The survey participants are all males aged between 19 and 59 years old with a median age of 34 years (range 28-40).

All of them declared to have Italian nationality and a significant proportion (82.4%) reported a high level of education (> 8 years).

The 59.7% of persons did not report to be in a stable relationship, while the remaining 40.3% of study sample reported to have a wife or stable partner.

In the 84.9% of study sample a job was reported, while the 5.9% resulted as unemployed and the 9.2% as a student.

Within study sample 112 persons (94.1%) reported having had sex with prostitutes, and only 7 (5.9%) with transsexuals.

Sexual behaviour

As shown in Table 2, the mostly reported sexual intercourse in the last six months is the vaginal intercourse (68.9%), followed by passive oral sex (36.1%), active oral sex (28.6%), anal sex (21.8%), and massage/masturbation (21.0%). The data also show that among the 119 examined subjects, 35 (29.4%) had more than one type of sexual intercourses.

Respondents had also been asked how frequently they used condom (always, sometimes, never) in vaginal, anal, and active oral intercourses which are at potential risk for HIV transmission.

Among reported vaginal intercourses, condom use was not consistent (never or sometimes) in the 13.4% of cases due to usage difficulty, choice of "coitus interruptus", unexpected penetration, carelessness.

Active oral sex was unprotected in the 55.9% of cases reported, mainly due to either the consideration that it is not a risky intercourse or lack of information on female condoms.

Among those reporting anal intercourses, consistent condom use was declared in the 84.6% of cases.

The number of persons who reported at least one HIV risky sexual intercourse, including unprotected or partly protected vaginal, anal, and active oral intercourses, resulted to be 27 (22.7%).

Of note, the number of unprotected intercourses resulted to inversely correlate, although without statistical significance, with the level of education and was reported in a higher extent, although not statistically significantly, by persons without a stable partnership than those with a stable partnership (26.8% vs 16.7% respectively).

Risk perception

Among persons reporting vaginal, anal or active sexual intercourses, the perception of HIV risk (as high or moderate) was 77.7%. Of those, the risk perception was slightly higher for persons reporting consistent condom protection than those reporting no or occasional condom protection (79.1% vs 74.1%, respectively) (see Table 3).

The perception of HIV infection risk changed according to the type of sexual intercourse.

Specifically, among persons reporting vaginal, anal and oral intercourses the risk was perceived (as high or moderate) in the 78.0%, 84.6% and 73.5% of cases, respectively. Of note, risk perception resulted different between persons reporting regular or non-regular/non-use of condom (81.8% vs 77.6%, 75.0% vs 86.4%, 79.0% vs 66.7% for vaginal, anal and active oral intercourses, respectively).

HIV testing

Within study sample 34.0% of persons said that they had never undergone HIV testing. Among these, 83.3% of cases had the awareness they have not risked HIV infection.

Among the 66.0% of persons who reported HIV testing, in 36.0% of cases it was performed once.

No particular preference between public or private health structures was reported for HIV testing and in about 50.0% of the sample public health structures were indicated.



HIV infection spreading in general population occurs prevalently through sexual activity. Although infection prevalence in general population is quite low, SW bears a higher infection risk and can spread HIV to low-risk general population.

The true extent of SW potential and occurred HIV transmission to their clients and general population has to be however fully characterised since SW clients have not the full perception of infection risk and do not easily undergo HIV testing. To this aim, specific interventions targeted to increase in SW clients the awareness of HIV infection risk require to adequately define their profile and characteristics, including sexual behaviour and risk perception.

This study was designed as a continuation of a previous investigation carried out by the TVA-IST Helpline on SW clients, which has allowed characterising the main features of this target population [24]. To this aim, a specific survey was designed and performed by researchers of TVA-IST Helpline to gain deeper and detailed information on SW client's sexual behaviour and HIV risk perception.

Looking through the gathered data, the survey shows that, among TVA-IST Helpline users, those who report sex with SW and have been recruited in the study are mainly Italian young adults in an age range of 19-48 years (median 34 years). These data are similar to those found in previously published studies, in which the age of targeted subjects ranges between 25 and 39 years [25-27].

Among Helpline users selected for the survey, 23.0% refused to participate to the study; this amount, already observed in other telephone surveys carried out by TVA-IST Helpline is generally associated with the fact that the phone tool - although meeting the needs of the persons/users - may not fit conditions as the available time, the emotional state or even the personal context, thus hampering the full and adequate participation at the survey [24].

As already confirmed by literature studies [28], a high level of education has been observed among the users with only a few cases reporting a low level of education (< 8 years).

The data also show that in 40.0% of the survey sample a wife or stable partner was reported, indicating that the demand of sex workers is not necessarily associated to inadequate affection or sexual fulfillment. The presence of stable partnership could therefore suggest that the persons targeted in the study may represent a bridge between high-risk groups (sex workers) and low-risk groups (companions, wives, girlfriends, recurring female partners), thus playing an important role in the transmission of HIV [29].

As foreseen by a previous study and literature [28-30], the survey found out that the most requested type of intercourse was the vaginal - 69.0% of cases - followed by active or passive oral sex (64.6%), and by anal intercourse (22.0%).

As expected, a high percentage of SW clients (over 80.0%) reported the consistent use of condom protection in both vaginal and anal intercourses, while less than half of them regularly protected active oral sex, mainly due to the underestimation of the infection risk and/or missed use of female condoms due to inadequate knowledge or availability.

The perception of HIV infection risk was not necessarily correlated to the usage of condom protection since it resulted to be high in both persons reporting protected or unprotected intercourses, also if slightly higher values were observed within vaginal and active oral sex. The risk perception is probably also due to stigma and irrational fear related to HIV infection, particularly within SW, or to the feeling of guilt which generally follows the access to sex workers. Nonetheless, condom use persists as inadequate in a relevant component of study sample, maybe due to carelessness, desire of condom-free sex or high condom costs. As regards risk perception with oral sex, a lack of knowledge was generally found within study participants, with an evident trend to underestimate risk in the active oral sex. The inadequate knowledge or use of female condom probably contributes to the biased perception of HIV risk related to active oral sex.

HIV testing was performed by at least 66.0% of the study sample, also if not necessarily following the access to SW. Of note, it was performed once in the 36.0% of cases, evidencing that the high-risk perception does not lead to adequate testing, may due to not easily accessible diagnostic services in the territory or irrational fear or shame associated to HIV testing.

It was however not possible to find an evident association between the perception of risk related to SW access and the implementation of HIV testing because of the short time period: the report of sexual behaviour refers to the six months preceding the survey while the questions on HIV testing refer to a wider and earlier time period (Have you ever made an HIV antibody detection test?)

We underscore that the study presents the typical limitations of any telephone survey that use self-reports by the respondent/user and it was necessary to adopt a convenience - non probabilistic - sampling due to the TVA-IST Helpline organization and mission. However, the exploit of structured telephone interview allowed an eased access, a high response rate and a standardised evaluation of questions related to HIV associated behavioural risks within a non-easily accessible population group.



Despite these potential experimental limitations and although the investigation sample can not be considered fully representative of the whole Italian population of SW customers, study data allow to define the characteristics, sexual behaviour and HIV risk perception of the targeted group. In particular, the evidence of inadequate risk perception, condom use and HIV testing in a relevant part of SW clients was found, suggesting the need of implementing the interventions aimed to increase the general awareness of sexually associated infection risk.

As already underlined, SW customers are a relevant target bridge from high-risk groups to low-risk groups and it is highly important to specifically reach this population group in order to encourage the adoption of safe behaviours and HIV testing.

It is also necessary, through further specifically targeted studies and surveys, a constant monitoring of the characteristics, behaviour, risk perception and testing of SW clients in Italian and other populations.


We wish to thank all the personnel of the TVA-IST Helpline who have been working in the HIV/AIDS/STD's phone counselling activity for twenty-seven years, answering daily to questions from users, together with Pierpaolo Ricci for the careful revision of the manuscript. In particular we wish to thank Eleonora Lichtner, Filippo Maria Taglieri and Rudi Valli for the administration of the survey questionnaire.

Conflicts of interest statement

None declared.



1. Suligoi B, Rezza G. Infezione da HIV: stime di incidenza e sistemi di sorveglianza. Epid Prev 2003;27:67.         

2. Camoni L, Boros S, Regine V, Ferri M, Santaquilani M, Pugliese L, Suligoi B. Aggiornamento delle nuove diagnosi di infezione da HIV e dei casi di AIDS in italia al 31 dicembre 2012. Not Ist Super Sanità 20013;26(9) Suppl. 1.         

3. Gaspari V, D'Antuono A, Bellavista S, Trimarco R, Patrizi A. Prostitution, sexual behaviour and STDs. G Italian Dermatol Venerol 2012;147(4) 349-55.         

4. Jin X, Smith K, Ray Y, Chen R, Ding G, Yao Y, Wang H, Quian H-Z, Chan D, Wang G, Wang N. HIV prevalence and risk behaviours among male clients of female sex workers in Yunnan, China. J Acquir Imune Defic Syndr 2010;53(1):131-5. DOI: 10.1097/ QAI.0b013e3181b3552a.         

5. Shaw S, Deering KN, Sushena RP, Isac S, Ramesh BM, Washington R, Moses S, Blanchard JF. Prevalence of HIV and sexually transmitted infections among clients of female sex workers in Karnataka, India: a cross sectional study. BMC Public Health 2011;11(Suppl. 6):S4. DOI: 10.1186/1471-2458-11-S6-S4.         

6. Fenton KA. Screening men for Chlamydia trachomatis infection: have we fully explored the possibilities? Common Dis Public Health 2000;3(2):86-9.         

7. Goldberg DJ, Emslie JA, Smyth W, Reid D. A system for surveillance of voluntary HIV testing: results of the first 2 years, 1989-1990. AIDS 1992;6(12):1555.         

8. Weinstein ND. Testing 4 competing theories of health-protective behaviour. Health Psychol 1993;12:324-33.         

9. Schwarzer R. Self-efficacy in the adoption and maintenance of health behaviours. Theoretica approaches and new model. In: Schwarzer R (Ed.). Self-efficacy. Though control of action. Washington, DC: Hemisphere; 1992. p. 217-243.         

10. Schwarzer R. Self-regulatory processes in the adoption and maintenance of health behaviour. J Health Psychol 1999;4:115-27. DOI: 10.1177/135910539900400208.         

11. Janz NK, Becker MH. The Health Belief Model - a decade later. Health Educ 1984;Q11:1-47. DOI: 10.1177/109019818401100101        

12. Rosenstock IM. Historical origins of Health Belief Model. Health Educ Monogr 19742:328-35.         

13. Crisp BR, Barber JG, Ross MW, Wodak A, Gold J, et al. Injecting drug-users and HIV/AIDS - risk behaviours and risk perception. Drug Alcohol Depend 1993;33:73-80. DOI: 10.1016/0376-8716(93)90035-O        

14. Prohaska TR, Albrecht G, Levy JA, Sugrue N, Kim JH. Determinants of self-perceived risk for AIDS. J Health Soc Behav 1990;31:384-94. DOI: 10.2307/2136821        

15. Bailey SL, Ouellet LJ, Mackesy-Amiti ME, Golub ET, Hagan H, et al. Perceived risk, peer influences, and injection partner type predict receptive syringe sharing among young adult injection drug users in five US cities. Drug Alcohol Depend 2007;91:S18-S29. DOI: 10.1016/j.drugalcdep.2007.02.014        

16. Do M, Meekers D. Multiple sex partners and perceived risk of HIV infection in Zambia: attitudinal determinants and gender differences. AIDS Care 2009;21:1211-21. DOI: 10.1080/09540120902730047        

17. Essien EJ, Ogungbade GO, Ward D, Fernandez-Esquer ME, Smith CR, et al. Injecting drug use is associated with HIV risk perception among Mexican Americans in the Rio Grande Valley of South Texas, USA. Public Health 2008;122:397-403. DOI: 10.1016/j.puhe.2007.07.023        

18. Mehrotra P, Noar SM, Zimmerman RS, Palmgreen P. Demographic and personality factors as predictors of HIV/STD partner-specific risk perceptions: Implications for interventions. AIDS Educ Prev 2009;21:39-54. DOI: 10.1521/aeap.2009.21.1.39        

19. Mitchell MM, Latimer WW. Unprotected casual sex and perceived risk of contracting HIV among drug users in Baltimore, Maryland: evaluating the influence of non-injection versus injection drug user status. AIDS Care 2009;21: 221-30. DOI: 10.1080/09540120801982897        

20. Tenkorang EY, Rajulton F, Maticka-Tyndale E. Perceived risks of HIV/AIDS and first sexual intercourse among youth in Cape Town, South Africa. AIDS Behav 2009;13: 234-45. DOI: 10.1007/s10461-008-9470-5        

21. Wood E, Li K, Miller CL, Hogg RS, Montaner JSG, et al. Baseline self-perceived risk of HIV infection independently predicts the rate of HIV seroconversion in a prospective cohort of injection drug users. Int J Epidemiol 2005;34:152-8. DOI: 10.1093/ije/dyh357        

22. Wright PB, McSweeney JC, Frith SE, Stewart KE, Booth BM. Losing all the pieces: a qualitative study of hiv risk perception and risk reduction among rural african american women who use cocaine. J Drug Issues 2009;39:577-605. DOI: 10.1177/002204260903900306        

23. Rezza G. Epidemie. Origini ed evoluzione. Roma: Carocci Editore; 2010.         

24. Mulieri I, Gallo P, Fanales Belasio E, Colucci A, D'Ippoliti A, Luzi AM, Sexual Behaviours of clients of sex workers reported within phone calls at HIV/AIDS/STIs Italian Helpline, Ann Ist Super Sanità 2013;49(2):183-9. DOI: 10.4415/ANN_13_02_10 .         

25. Goldenberg SM, Gallardo Cruz M, Strathdee S, Nguyen L, Semple S, Patterson TL. Correlates of unprotected sex with female sex workers among male clients in Tijuana, Mexico. NIH 2010;37(5):319-24. DOI: 10.1097/ OLQ.0b013e3181c5334f.         

26. Equipe dell'Unità Operativa UO RCF. Rapporto attivita di counselling telefonico, giugno 1987-2011. Available from: www.iss.it/binary/urcf/cont/Rapporto_Attivit_di_counselling_telefonico_1987_2011.pdf.         

27. Luzi A. L'intervento di counselling telefonico nella prevenzione da HIV. Not Ist Super Sanità 2007;20(4):7-10. Available from: www.iss.it/binary/urcf/cont/Notiziario_Ventennale.pdf.         

28. Rana MS, Nepali B, Sathian B, Aryal RP, Thapalia M, Bhatta DR. Socio-demographic characteristics of the clients of female sex workers and their perspectives, behaviours and attitude on HIV and AIDS. A questionnaire based survey from Pokhara, Nepal. J Clin Diagn Res 2013;7(1):112-7. DOI: 10.7860/JCDR/2012/4391.2683        

29. Goldenberg SM, Gallardo Cruz M, Strathdee S, Nguyen L, Semple S, Patterson TL. Correlates of unprotected sex with female sex workers among male clients in Tijuana, Mexico. Sex Transm Dis 2010;37(5):319-24. doi: 10.1097/OLQ.0b013e3181c5334f.         

30. Monto MA. Prostitution and fellatio. J Sex Res 2001; 38(2):140. DOI: 10.1080/00224490109552081.         



Ilaria Mulieri
Dipartimento di Malattie Infettive
Parassitarie ed Immunomediate
Istituto Superiore di Sanità
Viale Regina Elena 299
00161 Rome, Italy
E-mail: ilaria.mulieri@guest.iss.it

Received on 16 September 2014
Accepted on 18 November 2014

Istituto Superiore di Sanità Roma - Rome - Italy
E-mail: annali@iss.it