Changes in At-Risk Behavior for HIV Infection among HIV-Positive Persons in Italy

Abstract

Many HIV-positive persons reportedly continue to engage in at-risk behavior. We compared the sexual and drug-using practices of HIV-positive persons before and after the diagnosis of HIV infection to determine whether their behavior had changed. To this end, in 2006, we conducted a cross-sectional study involving clinical centers in five Italian cities. Each center was asked to enroll 100 persons aged 18 years or older who had a diagnosis of HIV infection that dated back at least 2 years. Data were collected with a specifically designed questionnaire, administered during a structured interview. The McNemar χ² test was used to compare the data before and after the diagnosis. A total of 497 persons participated (65.5% males; median age of 40 years; age range, 34–45 years). The most common exposure categories were: heterosexual contact (43.4%), homosexual contact (27.2%), and injecting drug use (20.6%). Although the percentage of drug users significantly decreased after diagnosis, 32.4% of injectors continued to use drugs, and approximately half of them exchanged syringes. Regarding sexual behavior, after diagnosis there was a significant decrease in the number of sexual partners and in stable relationships and an increase in condom use, both for persons with stable partners and those with occasional partners, although the percentage varied according to the specific sexual practice. These results indicate that though at-risk behavior seems to decrease after the diagnosis of HIV infection, seropositive persons continue to engage in at-risk practices, indicating the need for interventions specifically geared toward HIV-positive persons.

Introduction

I n Western countries, the success of highly active antiretroviral therapy (HAART) has led to a steady increase in the number of persons living with HIV infection, who may constitute a source of infection for others. Several studies have indicated that many HIV seropositive persons continue to engage in at-risk behavior and that awareness of being seropositive does not always translate into behavioral changes, for instance, safer sexual practices.^1

–6 Moreover, in recent years, there has been an increase in the incidence of other sexually transmitted infections (STI) among persons who have been HIV-positive for a number of years, suggesting that a proportion of HIV-infected persons nonetheless engage in unprotected sex.^7
–9 It has also been reported that persons undergoing HAART (or those with an undetectable viral load) more frequently have unprotected sex and occasional partners than do persons not undergoing HAART, increasing the probability of transmitting strains that are more aggressive and resistant to drugs.^1,10

–13 The objective of the present study was to determine whether the diagnosis of HIV infection leads to changes in sexual and drug-using behavior.

Methods

In 2006, we conducted a cross-sectional multicentre study involving five major public clinical centers (i.e., hospital infectious-disease clinics and STI clinics) in five large cities in Italy: Turin and Milan, in northern Italy; Rome in central Italy; and Foggia-Bari and Palermo in southern Italy. Each of the centers was asked to include 100 consecutive HIV patients aged 18 years or older who had been diagnosed at least 2 years prior to the study.

A specifically designed questionnaire was used to collect the following information: sociodemographic, personal, and clinical data; information on drug-using behavior; information on sexual behavior; reasons for engaging in unprotected sex; and risk perception. All of the questions referred to the periods before and after the diagnosis of HIV infection. The questionnaire, which was anonymous, was administered during a structured interview performed in 2005 by trained medical personnel. Written informed consent was obtained from all persons before completing the questionnaire.

The McNemar χ² test was used to compare the data before and after the diagnosis of HIV infection, to identify any changes. The test for paired data compares the changes in behavior of the same individuals before and after the HIV diagnosis and was only applied to persons who provided information for both periods. We analyzed the results using version 15 of SPSS software (Statistical Package Social Science; SPSS Inc., Chicago, IL).

Results

Of the 497 persons asked to participate in the study; 487 agreed. Of the 10 (1.97%) who refused, 7 stated that they thought it was useless to respond to the questions and 3 felt that the questionnaire violated their privacy. Males represented 65.5% of the participants, and the median age was 40 years (range, 34–45 years). Most participants (85.2%) were Italian; of the 72 non-Italians, 38.8% had been living in Italy for more than 10 years. Regarding marital status, 50.7% were single, 32.7% were married, 11.2% were separated or divorced, and 5.3% were widowed. Primary school had been completed by 40.8% of the participants; 35.2% had attended high school, yet only 12.5% graduated. One third of the participants reported that they earned a monthly salary of 1,000–2,000 Euro.

The most common exposure categories for HIV were: heterosexual contact (43.4%), homosexual contact (27.2%), and injecting drug use (20.6%). Of the participants, 58.6% had been diagnosed with HIV infection before 1996, and 52.5% were in clinical stage A (Centers for Disease Control and Prevention [CDC] classification) upon enrolment in the study.

Of the participants, 52.0% (n = 253) reported having used drugs (injecting or noninjecting). The drug-using behavior before and after the diagnosis of HIV infection is reported in Table 1. After diagnosis, there was a statistically significant decrease (p < 0.01) in the use of all drugs. Nonetheless, of the 138 persons who reported injecting drug use, 60% continued to use injecting drugs after diagnosis, and about half of these individuals engaged in syringe exchange. The proportions of individuals represented by the different exposure categories did not significantly vary when comparing the period before diagnosis to the period after diagnosis.

Table 1.

Drug Use Before and After the Diagnosis of HIV Infection (253 Persons; 52.0% of the Total Study Participants)

	Before diagnosis		After diagnosis		McNemar test for paired data
Type of drug	n	%	n	%	χ²	p value
Heroin					51.8	0.000
Yes	139	54.9	80	31.6
No	114	45.1	173	68.4
Total	253		253
Cocaine					28.4	0.000
Yes	125	49.4	76	30.0
No	128	50.6	177	70.0
Total	253		253
Cannabis					52.4	0.000
Yes	176	69.6	118	46.6
No	77	30.4	135	53.4
Total	253		253
Alcohol					10.3	0.001
Yes	104	41.1	83	32.8
No	149	58.9	170	67.2
Total	253		253
Injecting use					42.9	0.000
Yes	138	54.5	82	32.4
No	114	45.1	164	64.8
No answer	1	0.4	7	2.8
Total	253		253
Syringe exchange					53.7	0.000
Yes	113	44.7	40	15.8
No	104	41.1	160	63.3
No answer	36	14.2	53	20.9
Total	253		253

The data regarding sexual behavior before and after the diagnosis of HIV infection are shown in Table 2. The median age of first sexual contact was 17 years (interquartile range, 15–18). There was a significant decrease (p < 0.000) in the number of sexual partners after diagnosis: the proportion of those who reported having had more than 2 partners decreased from 83.2% to 54.2%, in particular: the percentage decreased from 24.6% to 19.5% for those with 3–5 partners, from 23.4% to 15.6% for those with 6–20 partners, and from 35.1% to 19.1% for those with more than 20 partners. However, the probability of having more partners could be higher before diagnosis because the median time of exposure to sexual contact was 12 years before diagnosis (i.e., the time elapsed between the first sexual contact and the diagnosis of HIV infection) and 9 years after diagnosis (the time elapsed between the interview and the diagnosis).

Table 2.

Sexual Behaviour Before and After the Diagnosis of HIV Infection (487 Persons)

	Before diagnosis		After diagnosis		McNemar test for paired data
	n	%	n	%	χ²	p value
Number of sexual partners					113.47	0.000
≤2	81	16.6	219	45.0
>2	405	83.2	264	54.2
No answer	1	0.2	4	0.8
Total	487	100.0	487	100.0
Sexual orientation					0.48	0.48
Homosexual/bisexual	180	37.0	164	33.7
Heterosexual	302	62.0	284	58.3
No answer	5	1.0	39	8.0
Total	487	100.0	487	100.0
Sex for money or drugs					16.68	0.000
Yes	64	13.1	33	6.8
No	413	84.8	433	88.9
No answer	10	2.1	21	4.3
Total	487	100.0	487	100.0
Sex with sex workers					22.37	0.000
Yes	78	16.0	35	7.2
No	381	78.2	416	85.4
No answer	28	5.8	36	7.4
Total	487	100.0	487	100.0
Sex when infected (or partner infected) with an STI					1.90	0.168
Yes	122	25.1	106	21.7
No	337	69.2	352	72.3
No answer	28	5.7	29	6.0
Total	487	100.0	487	100.0
Condom use when infected (or partner infected) with an STI					1.10	0.118
Yes	65	53.3	77	72.6
No	54	44.3	25	23.6
No answer	3	2.5	4	3.8
Total	122	100.0	106	100.0

STI, sexually transmitted infection.

We also compared the sexual and drug-using behavior of persons undergoing HAART and those not undergoing HAART. There were no significant differences between the two groups, probably because 83% of our study sample was undergoing therapy and the remaining 17% was nonetheless regularly followed by the physicians at the clinical centers and received information and psychological support.

After diagnosis, the percentage of persons who engaged in sex in exchange for money or drugs significantly decreased (from 13.1% to 6.8%). When considering males and females separately, this percentage decreased from 9.7% to 4.5% among males and from 19.8% to 11.0% among females. Moreover, the percentage of males who had sex with sex workers significantly decreased (from 24.2% to 11.0%).

Of the participants, 89.1% (n = 434) reported that they had had a stable relationship before the diagnosis of HIV infection; the percentage decreased to 77.4% (n = 377) after diagnosis (p < 0.000). (Table 3). This decrease remained significant when conducting the analysis separately by gender: in particular, 85.8% of males had a stable relationship before the HIV diagnosis and 74.5% after diagnosis; 95.9% of females had a stable relationship before diagnosis and 82.6% after diagnosis.

Table 3.

Sexual Behaviour with Stable Partner (434 Persons) and Occasional Partners (400 Persons) Before and After the Diagnosis of HIV Infection

	Before diagnosis		After diagnosis		McNemar test for paired data
	n	%	n	%	χ²	p value
Stable partner					27.75	0.000
Yes	434	89.1	377	77.4
No	53	10.9	110	22.6
Total	487	100.0	487	100.0
Condom use: vaginal sex					118.07	0.000
Always	24	5.5	150	39.8
Not always/never	323	74.5	122	32.4
No answer	87	20.0	105	27.8
Total	434	100.0	377	100.0
Condom use: anal sex					86.49	0.000
Always	18	4.1	120	31.8
Not always/never	292	67.3	126	33.5
No answer	124	28.6	131	34.7
Total	434	100.0	377	100.0
Condom use: oral-genital sex					26.03	0.000
Always	4	0.9	34	9.0
Not always/never	372	85.7	273	72.4
No answer	58	13.4	70	18.6
Total	434	100.0	377	100.0
Occasional partners					89.11	0.000
Yes	400	82.1	283	58.1
No	87	17.9	204	41.9
Total	487	100.0	487	100.0
Condom use: vaginal sex					65.33	0.000
Always	41	10.3	107	37.8
Not always/never	254	63.5	65	23.0
No answer	105	26.2	111	39.2
Total	400	100.0	283	100.0
Condom use: anal sex					68.36	0.000
Always	42	10.5	115	40.6
Not always/never	267	66.8	91	32.2
No answer	91	22.7	77	27.2
Total	400	100.0	283	100.0
Condom use: oral-genital sex					31.24	0.000
Always	11	2.7	49	17.3
Not always/never	329	82.3	188	66.4
No answer	60	15.0	46	16.3
Total	400	100.0	283	100.0

After diagnosis, there was a significant decrease (p < 0.000) in the percentage of persons who never (or not always) used a condom with their stable partner for vaginal sex (from 74.5% to 32.4%), anal sex (from 67.3% to 33.5%), and oral–genital sex (from 85.7% to 72.4%). The percentage of males who never (or not always) used a condom during vaginal sex with their stable partner decreased significantly, from 54.8% before diagnosis to 16.8% after diagnosis; significant decreases were also observed for anal sex (from 66.5% to 25.8%) and oral-genital sex (from 76.8% to 54.2%). The percentage of females who never (or not always) used a condom with their stable partner also decreased significantly for vaginal sex (from 87.2% to 40.7%), anal sex (from 49.4% to 26.7%), and oral–genital sex (from 77.3% to 60.5%).

Of the participants, 82.1% (n = 400) declared that they had had casual sexual contact before diagnosis, whereas this percentage decreased to 58.1% (n = 283) after diagnosis (p < 0.000; Table 3). This decrease remained significant when conducting the analysis separately by gender: in particular, 94.2% of males had had an occasional partner before diagnosis and 68.7% after diagnosis; 61.0% of females had had a causal partner before diagnosis and 39.5% after diagnosis.

After diagnosis, there was a significant decrease (p < 0.000) in the percentage of persons who never (or not always) used a condom with occasional partners for vaginal sex (from 63.5% to 23.0%), anal sex (from 66.8% to 32.2%), and oral-genital sex (from 82.3% to 66.4%). The percentage of males who never (or not always) used a condom during vaginal sex with an occasional partner decreased significantly, from 55.2% before diagnosis to 14.2% after diagnosis; significant decreases were also observed for anal sex (from 68.4% to 21.6%) and oral–genital sex (from 79.0% to 46.8%). The percentage of females who never (or not always) used a condom with their occasional partner also decreased significantly for vaginal sex (from 47.7% to 12.2%), anal sex (from 31.4% to 14.0%), and oral–genital sex (from 47.7% to 24.4%). Decreases in the percentage of persons who never (or not always) used a condom were also observed among heterosexuals and homosexuals/bisexuals.

Regarding the reasons for not using a condom, 35.5% of the participants reported that they wanted to “share everything” with their partner; 33.9% reported difficulties in using a condom (e.g., it was uncomfortable or it ruined “the mood”); 28.7% thought that the specific sexual behavior constituted a low risk (specifically: 18.5% “because anal or oral contact is not a risk”; 5.5% “because my viral load is low”; and 4.7% “because I am taking medication”); and 7.0% because they had feelings of distrust or disinterest toward themselves and others.

Regarding the percentage of persons who engaged in sexual activity in the presence of STIs other than HIV (whether the individual him/herself was infected or their partner), there were no significant changes after diagnosis. In particular, one fourth of the persons who reported having sex while infected with an STI did not use a condom.

Discussion

The use of HAART has led to an increase in the number of persons living with HIV infection, yet many of them continue to engage in at-risk behavior because the mere awareness of being seropositive does not translate into safer sexual practices.¹⁴ Although counseling could play a role in modifying sexual behavior, it requires trained personnel and time. In Italy, there are approximately 150,000 HIV-positive persons.¹⁵ In the past, injecting drug users accounted for the majority of HIV/AIDS cases, yet in the past decade, transmission through heterosexual contact has become of paramount importance also in Italy.

To the best of our knowledge, this is the first study in Italy to explore behavioral changes before and after the diagnosis of HIV infection among HIV-positive persons. With regard to drug use, although there was a significant decrease in the use of all drugs following the diagnosis of HIV infection, nearly one third of the persons reported that they continued to use drugs intravenously and half of them continued to share syringes. This finding is consistent with the results of international studies which have reported the continued presence of at-risk behavior among drug users (i.e., the exchange of syringes and drug-using equipment), although drug-reduction campaigns (e.g., distribution of syringes free of charge, use of methadone) have produced a reduction in these practices.^16,17 In Italy, most injecting drug users attend public drug-dependency centers and receive substitutive therapy, psychological assistance, and accurate information on at-risk behavior. Thus these persons are constantly followed, as shown by the data from the National AIDS Registry, which show that there is a low proportion of drug users with a delay in diagnosis of HIV and that access to HAART is timely.¹⁸

The analysis of sexual behavior also revealed a significant decrease in the number of partners after diagnosis. However, nearly one fifth of the participants reported that they had had more than 20 sexual partners after having been diagnosed. Nonetheless, it should be stressed that the number of partners may have been lower after HIV diagnosis because the number of years that had elapsed between diagnosis and the interview was lower than the number of years between the first sexual contact and the diagnosis of HIV infection. After the diagnosis of HIV infection, the percentage of persons who engaged in sex with sex workers or who performed sex for payment significantly decreased, despite reports that behavior related to prostitution or sex for payment is the most difficult to change, even when the person is aware of being HIV positive.^1,19,20 With regard to the percentage of persons who engaged in sexual activity in the presence of STIs other than HIV (whether the individual him/herself was infected or their partner), there were no significant changes after diagnosis, which could indicate a low perception of risk regarding STIs. In particular, one fourth of the persons who reported having sex while infected with an STI did not use a condom. The increase in the incidence of diagnoses of STIs among both HIV-positive homosexuals and heterosexuals since 2000 suggests that there has been an increase in at-risk sexual behavior, as reported by other studies.^21
–23

An increase in condom use after HIV diagnosis was observed among both males and females and with both stable and occasional partners, although the percentage varied with type of sexual practice. However, more than one third of the persons with a stable partner reported that they did not use a condom when engaging in vaginal or anal sex, and two thirds did not do so when engaging in oral–genital sex; a similar trend was found with respect to occasional partners.

When interpreting the results of this study, some limitations need to be considered. The sample cannot be considered as representative of the Italian HIV-positive population, given that the centers were not selected randomly and the study population was a convenience sample. Nonetheless, the participating centers are located throughout the entire country, allowing for good geographic coverage, and the study population mainly consisted of Italian males who had presumably acquired the infection through sexual contact, which reflects the characteristics of persons with AIDS and HIV infection observed in Italy in recent years.¹⁸ The fact that a structured interview was used ensured that the answers to the questions were homogeneous and complete, though the persons being interviewed may have provided responses that they felt were acceptable to the interviewer. Furthermore, the results may have been affected by recall bias, in that the questions on at-risk behavior before diagnosis referred to a period that was, on average, 10 years prior to the interview. Finally, the duration of the period before diagnosis was greater than that following diagnosis, not to mention the fact that the duration of these periods varied greatly from person to person.

Another consideration is that the observed changes in behavior could have been a consequence of prevention measures geared toward persons with HIV infection (i.e., specific collective or individual interventions after the diagnosis of HIV infection). In fact, since the beginning of the HIV epidemic, the World Health Organization (WHO) has recommended pretest and posttest counseling to help persons implement strategies for dealing with their seropositivity and to change behaviors considered to constitute a risk for transmitting the virus.²⁴

The results of this study show that even persons who are aware of being infected can be a source of HIV infection and STIs. Despite that fact that the persons interviewed had known for some time that they were seropositive (nearly half had known for at least 10 years), they continued to engage in at-risk behavior, especially at-risk sexual behavior. Anal and oral–genital sex are those practices for which the lowest percentages of persons use a condom, whether with stable partners or occasional partners. Drug users also continued to engage in at-risk behavior after the diagnosis of HIV infection (intravenous use and syringe exchange).

Conclusions

These results constitute important information for developing prevention activities in Italy to safeguard the health of HIV-positive persons and the public overall. In particular, interventions geared toward couples are needed to prevent the transmission of HIV and to encourage communication and negotiation with respect to condom use. In Italy, information/education campaigns and prevention activities have mainly focused on the general population or specific target groups, such as young persons, drug users, and non-nationals, with less focus on HIV-positive persons. Finally, knowledge of the lifestyles of HIV-positive persons can constitute a foundation for developing support interventions to improve adherence to therapy.

Footnotes

Acknowledgments

The authors wish to thank Debora Lepore for secretarial assistance and Mark Kanieff for revising the manuscript and providing useful suggestions. This study was funded by the VI Programma Nazionale di Ricerca sull'AIDS 2005. None of the authors have any conflict of interest.

Author Disclosure Statement

No competing financial interests exist.

References

McGowan

, Shah

, Ganea

et al. Risk Behavior for transmission of human immunodeficiency virus (HIV) among HIV-seropositive individuals in an urban setting. Clin Infect Dis, 2004; 38:122–127.

Browning

, Evans

, Rees

. Continued high-risk sexual behaviour among HIV-positive people in Wales. Int J STD AIDS, 2003; 14:737–739.

Amirkhaninan

, Kelly

, McAuliffe

. Psychological needs, mental health, and HIV transmission risk behaviour among people living with HIV/AIDS in St Petersburg, Russia. AIDS, 2003; 17:2367–2374.

Centres for Disease Control and Prevention (CDC). High-risk sexual behavior by HIV-positive men who have sex with men—16 sites, united states, 2000–2002. MMWR, 2004; 53:891–894.

Weinhardy

, Kelly

, Brondino

et al. HIV transmission risk behaviour among men and women living HIV/AIDS in 4 cities in the Unites States. J Acquir Immune Defic Syndr, 2004; 36:1057–1066.

Kalichman

, Ntseane

, Nthomang

et al. Recent multiple sexual partners and HIV transmission risk among people living with HIV/AIDS in Botwana. Sex Transm Infect, 2007; 83:371–375.

Suligoi

, Giuliani

. Epidemiologia delle malattie a trasmissione sessuale. Riv Gin Cons, 2002; 14:107–114.

Batter

. and the European Study Group. Trends in HIV seroprevalence among patients with sexually transmitted diseases in 17 European sentinel networks, 1990–1996. AIDS, 2000; 14:871–800.

Ward

, Mercer

, Wellings

. Who pays for sex? An analysis of the increasing prevalence at female commercial sex contacts among men in Britain. Sex Transm Infect, 2005; 81:467–471.

10.

Stephenson

, Imrie

, Davis

MMD

et al.

Is use of antiretroviral therapy among homosexual men associated with increased risk of transmission of HIV infection?

Sex Transm Infect, 2003; 79:7–10.

11.

Kalichman

, Rompa

, Austin

, Luke

, DiFonzo

. Viral load, perceived infectivity and unprotected intercourse. J Acquire Immune Defic Syndrome, 2001; 28:303–308.

12.

Miller

, Meyer

, Boufassa

et al. Sexual behaviour changes and protease inhibitor therapy. AIDS, 2000; 4:F33–F39.

13.

Remien

, Halkitis

, O'Leary

et al. Risk perception and sexual risk behaviours among HIV-positive men on antiretroviral therapy. AIDS Behav, 2005; 9:167–176.

14.

Centers for Disease Control and Prevention (CDC); Health Resources and Services Administration; National Institutes of Health; HIV Medicine Association of the Infectious Diseases Society of America. Incorporating HIV prevention into the medical care of persons living with HIV. Recommendations of CDC, the Health Resources and Services Administration, the National Institutes of Health, and the HIV Medicine Association of the Infectious Diseases Society of America. 2003; 52:1–24 MMWR Recomm Rep, 2004; 53:744erratum in

15.

WHO. Epidemiological Fact Sheet on HIV and AIDS. Italy, 2008. www.who.int/globalatlas/predefinedReports/EFS2008/full/EFS2008_IT.pdf Last accessed August 28, 2009Update.

16.

Lindenburg

, Krol

, Smit

et al. Decline in HIV incidence and injecting, but not in sexual risk behaviour, seen in drug users in Amsterdam: A 19-year prospective cohort study. AIDS, 2006; 20:1771–1775.

17.

Hope

, Judd

, Hickma

et al. HIV prevalence among injecting drug users in England and Wales 1990 to 2003: Evidence for increased transmission in recent years. AIDS, 2005; 19:1207–1214.

18.

Centro Operativo AIDS, Istituto Superiore di Sanità. Aggiornamento dei casi di AIDS notificati in Italia al 31 Dicembre 2006. Notiziario Istituto Superiore di Sanità, 2007; 20,Suppl 1:1–23.

19.

Kalichman

. Psychological and social correlates of high-risk sexual behaviour among men and women living with HIV/AIDS. AIDS Care, 1999; 11:415–427.

20.

El-Bassel

, Schilling

, Irwin

et al. Sex trading and psychological distress among women recruited from the streets of Harlem. Am J Public Health, 1997; 87:66–70

21.

Hanson

, Posner

, Hassig

et al. Assessment of sexually transmitted diseases as risk factors for HIV seroconversion in a New Orleans sexually transmitted disease clinic, 1990–1998. Ann Epidemiol, 2005; 15:13–20.

22.

Buchacz

, Greenberg

, Onorato

, Janssen

. Syphilis epidemics and human immunodeficiency virus (HIV) incidence among men who have sex with men in the United States: Implication for HIV prevention. Sex Transm Dis, 2005; 32:S73–79.

23.

Wong

, Chaw

, Kent

, Klausner

. Risk factors for early syphilis among gay and bisexual men seen in an STD clinic: San Francisco, 2002–2003. Sex Transm Dis, 2005; 35:458–463.

24.

World Health Organization—Global Programme on AIDS (WHO/GPA). Counselling for HIV/AIDS. As key to caring: Geneva, 1995.