Article Text
Abstract
Objective To describe a pilot programme that consisted of identifying, contacting and providing HIV testing to members of social and sexual networks of HIV-infected youth patients as a part of routine clinical care at an adolescent HIV clinic in Baltimore, Maryland, USA.
Methods Forty-nine sexually active HIV-infected adolescent patients were interviewed about their social and sexual contacts at a routine HIV clinic visit. A trained community health worker located these referred social and sexual contacts, and encouraged them to make an appointment for HIV counselling, testing and referral (CTR) services.
Results During a period of 18 months, 26 index youths provided locating information on 53 first-generation contacts and these 53 contacts provided information on 16 second-generation contacts. A total of 32 contacts received counselling services and 25 were tested for HIV infection, yielding three new HIV-positive individuals.
Conclusion As a part of standard care for regular visits of HIV-infected youth patients, interviewing about their social and sexual contacts could be a viable strategy in identifying high-risk youths for HIV infection and subsequent CTR services.
- Adolescent
- HIV testing
- intervention studies
- referral
- STD clinic
Statistics from Altmetric.com
It is estimated that half of all new HIV infections in the USA occur among individuals under 25 years of age,1 2 and a quarter of Americans currently living with HIV do not know their infection status.3 The most recent Centers for Disease Control and Prevention (CDC) guidelines as well as the Society for Adolescent Medicine recommend that all sexually active adolescents should be routinely screened for HIV infection;4 5 however, adolescents at the highest risk of HIV are often the most difficult to reach for screening.6 7 This suggests the need for more proactive strategies to identify and screen high-risk adolescents for early HIV detection, both for treatment and prevention of further disease transmission.
Network theory argues that people tend to socialise and have sex within a community of those who share a similar demographic and behavioural profile.8–10 Furthermore, the social and sexual networks that include HIV-positive index individuals are more likely to include additional network members with HIV than do networks that do not include HIV-positive indexes.11 12 Therefore, the challenge is to organise strategies that target networks with identified HIV-infected index patients so that there are opportunities to reach out to other network members to provide HIV counselling, testing and referral (CTR) services. In line with this theory, one means of increasing the identification of HIV-positive adolescents is to offer CTR services to the social and sexual contacts of HIV-infected index adolescents and young adults.
Respondent-driven sampling has been experimentally used to sample at-risk populations for HIV/AIDS in the USA and internationally.13 14 Other network sampling methods that have also been used to recruit hard-to-reach populations, such as men who have sex with men (MSM), include standard partner counselling and referral services, and time space sampling (eg, alternative venue-based sampling). These different strategies and sampling methods have been effectively used to recruit drug users and at-risk MSM.15 16 However, there are no published reports detailing the use of social and sexual network methods to recruit and provide HIV testing for at-risk adolescents as part of routine clinical services.
This paper describes the results of a pilot programme implemented to recruit social and sexual contacts at one adolescent HIV clinic in Baltimore, Maryland, USA. The pilot programme used social network interviews (SNI) facilitated by a community health worker (CHW) to identify adolescents and young adults at risk of HIV infection, in order to offer CTR services and possible engagement in care. The objectives of this paper were to describe the pilot programme, including SNI conducted with members of networks to which HIV-infected adolescents in care belong; and to discuss whether SNI effectively reach high-risk youths by comparing risk behaviours of HIV-infected adolescents in care with those of their contacts.
Methods
Setting and pilot programme
The Harriet Lane Intensive Primary Care Clinic at the Johns Hopkins Hospital (Adolescent HIV Clinic) has offered care to children and adolescents for HIV infection for over 20 years. In addition to medical care, the Adolescent HIV Clinic offers social and mental health services to help HIV-infected patients manage the variety of challenges they may face and improve their chances of medication adherence. The Adolescent HIV Clinic provides services to both behaviourally as well as perinatally acquired HIV infections in youths from birth to 25 years of age. The majority of the youths in the Adolescent HIV Clinic with perinatally acquired HIV infection are currently adolescents and are exhibiting an equal amount of high-risk behaviours as their peers with behaviourally acquired HIV infection. The Adolescent HIV Clinic serves HIV-infected patients until they are 25 years old, at which time they are transferred to adult services.
Both HIV-infected youths at the Adolescent HIV Clinic and high-risk youth patients who visit other clinics in the Johns Hopkins Hospital receive annual or bi-annual risk reduction programmes. An HIV risk reduction programme was developed following the HIV Health Education and Risk Reduction Guidelines issued by the CDC17 as a way to educate and allow for candid discussion with HIV-positive youths. The risk reduction programme was offered at the Adolescent HIV Clinic, and it provided skill building concerning the prevention of HIV transmission using multiple formats, including literature, discussion and role playing. Partner notification is offered through the risk reduction programme and information, such as names and contacts of sex partners and at-risk social friends, who may require outreach efforts were collected.
All HIV-infected youth patients and at-risk contacts who chose to visit the Adolescent HIV Clinic for HIV testing received the risk reduction programme as it is a part of the standard of care for adolescent HIV infection. All HIV-infected youth patients also received the programme more frequently if they presented with a sexually transmitted infection. Often, youths would name different sex partners each time they were interviewed.
A segment of the routine clinical care for HIV-infected youth patients at the Adolescent HIV Clinic was distinguished as our pilot programme consisting of three major services: offering the risk reduction programme; conducting a paper-based SNI and offering an HIV test. The pilot programme was implemented at the Adolescent HIV Clinic between January 2006 and June 2007. The goals of the pilot programme were to identify, contact and provide HIV testing to members of social and sexual networks of HIV-infected youth patients. There was no informed consent nor compensation provided for the completion of the SNI as it was not implemented as a research instrument, but as a part of standard care at the Adolescent HIV Clinic. The analysis of these data was conducted with de-identified data and was declared exempt from human subjects research by our institution's review board.
Social network interview
When the providers or social workers identified patients under the age of 25 years as newly diagnosed with HIV, or currently HIV-infected youths visited the Adolescent HIV Clinic for their routine care, these patients participated in the pilot programme. The patients thus served as index youths in the pilot programme. The index youths were referred to the CHW who provided the risk reduction programme and an SNI. The SNI was administered once by the CHW, who received a 4-day training session including how to protect the privacy of the HIV-infected patients. The CHW at the Adolescent HIV Clinic collaborated with the Baltimore City Health Department and their CHW.
At the beginning of SNI, the CHW explained to the index youths the importance of informing all sex partners of their possible exposure to HIV and assessed the index youths' plan to end any potential routes of HIV transmission. The CHW then obtained as much information about the sex partners and at-risk social friends as possible from the index youth patient. The index youths were assured that their privacies were protected, and that their sexual and at-risk social contacts would not be provided the information about the source of the referrals. Following the SNI, the CHW provided their business cards to the index youths and encouraged them to inform their sex partners and at-risk social friends to visit the Adolescent HIV Clinic for confidential testing.
Locating contacts
The CHW attempted to locate and meet with all referred contacts. If located, the CHW attempted to meet the contact face to face but sometimes utilised telephone contact. When the CHW met with the referred contacts, she introduced herself generically as a public health worker and explained that the contacts' names were referred to her as someone who may have been exposed to or should be tested for HIV. The CHW encouraged the referred contacts to make appointments for HIV screening the same day at either the Adolescent HIV Clinic or Baltimore City Health Department Clinic. When the referred contacts visited and registered at one of two clinics, they were offered CTR and the risk reduction programme. If the referred contacts tested positive for HIV, they also completed the SNI. The de-duplication process was to ensure a particular referred contact was included in the index's sexual/social network only once. This study was not designed to de-duplicate reported indexes if they were referred back by counselled contacts.
Statistical analysis
Summary statistics as percentages and means were calculated to describe those index youth patients who referred their friends and sex partners and those who were referred. Due to the clustering nature of referred contacts and index youths in the same sexual/social network, generalised estimating equations were used to assess the association of the probability of engaging in risk behaviours and being a counselled contact. Risk behaviours were measured as binary outcomes, and the explanatory variable was a binary variable that indicated 1 as being a counselled contact and 0 as being an index youth. Continuous outcome variables included the number of sexually transmitted infections in the past year and the number of sex partners in the past 3 months. A concurrent partnership was identified when a respondent reported multiple partners in an overlapping period. We created a dichotomised variable (1: ≥two partners, 0: <two partners) indicating a presence of concurrent partners from the self-reported number of sex partners in the past 3 months. While considering these outcomes as count data, rate ratios (RR) with 95% CI were reported for Poisson regression models. Due to the extremely high reported numbers, the number of lifetime partners was dichotomised by median. OR and corresponding 95% CI were reported for binary outcomes. The pilot programme reported here was not a research study; thus sample size calculation was not relevant. All analyses were performed using SAS (version 9.1).
Results
In the period from January 2006 to June 2007, a total of 301 youth patients (49 HIV-infected and 252 high-risk youths) were recruited for the risk reduction programme at the Adolescent HIV Clinic. The 49 HIV-infected youths served as index youths for the pilot programme, and they were interviewed about sexual and social contacts. Approximately half of the index youths (26/49) provided locatable contacts' names on a total of 53 first-generation contacts (mean=2.0 contacts per index). The remaining 23 index youths provided no pertinent information concerning their sexual or social contacts. The CHW located 24 contacts of the referred 53 first-generation contacts. These 24 contacts referred 16 locatable social and sexual contacts (second-generation contacts) during their SNI; eight were located. A total of 32 first and second generation contacts received the risk reduction programme. Twenty-seven per cent of those 32 contacts were originated from social networks and 73% were originated from sexual networks.
The mean and range of days from the time of the index patients' risk assessment interview to the time of contacts being located and interviewed was 41 days and 0–293 days, respectively. Almost half of the contacts (47%) were located and interviewed on the same day as the index or the first-generation contact. The 32 contacts who received the risk reduction programme included two with a previous HIV diagnosis, five who refused HIV testing and 25 who agreed to be tested. As a result of the SNI followed by the HIV test, three individuals were newly diagnosed with HIV (see figure 1).
The mean (SD) age of these 26 HIV-infected index youths and 32 counselled contacts was 19.5 years (2.5) and 20.9 years (4.8), respectively. The majority of index youths and contacts were African American (80% and 97%, respectively). Approximately 60% of index youths and counselled contacts had concurrent sex partners. Whereas 47% of index youths used a condom at last vaginal sex, only 10% of counselled contacts reported condom use at last vaginal sex. The number of lifetime partners reported by index youths ranged from one to 300, with a median of seven, and counselled contacts reported a range of four to 200, with a median of 10. Additional risk characteristics and behaviours are presented in table 1. Generalised estimating equation results showed that counselled contacts were less likely (OR 0.15; 95% CI 0.05 to 0.53) to use a condom at last vaginal sex than index youths, and the odds of using alcohol in the past 3 months and having more than eight lifetime sex partners was 3.6 (OR 3.60; 95% CI 1.11 to 11.7) and 3.9 times greater (OR 3.90; 95% CI 1.17 to 13.1), respectively, for counselled contacts than index youths. The results from the Poisson model showed that, on average, the number of sex partners over the period of 3 months was 55% more for counselled contacts than index youths (RR 1.55; 95% CI 1.14 to 2.14).
Not surprisingly, some HIV-infected index youths were hesitant to disclose the names and contact information of their sex partners. The CHW found that a common reason given for refusing to provide information about sex partners was that these sex partners were ‘not significant enough’ for index youths to care if those individuals would get tested and treated for HIV, and index youths did not think there was any chance in the future to have sex with those partners again. In these cases, the CHW used various strategies to explain how an untreated infection can travel back to the index youth or their social friends indirectly through intermediary members in the social and sexual network. The reactions of counselled contacts varied: some felt angry when they were reached by the CHW that they had been named; others were interested in getting tested.
Discussion
Our pilot programme, a segment of youths' routine HIV clinic visit care including a SNI, was carried out by a female CHW and successfully identified three new HIV-positive individuals. The CHW implemented a SNI to index HIV-positive youth patients and subsequently to contacts referred by those index youths. She reached approximately half of referred contacts on the same day when these referrals were made. We expected two groups of individuals (ie, index and referred contacts) to present similar demographic and behavioural characteristics; however, there were substantial differences. Referred contacts were less likely to have engaged in protected sex, more likely to have used alcohol, and reported a greater number of lifetime sex partners than index youths. Referred contacts were encouraged to make an appointment for confidential testing and they visited clinics voluntarily. Therefore, we believe that the difference between index youths and referred contacts in the results could be explained by the selection of each group voluntarily to receive an HIV test, which may reflect individuals' perceived risk of HIV.
The study reported here has a number of strengths. It highlighted a novel concept of a more proactive outreach approach embedded in existing care services. This paradigm shift has led to the discovery of three new HIV-positive cases, who may not otherwise have been identified if same routine procedures were provided without our new concept of the outreach programme.
SNI, also known as social networks testing, is defined in the current CDC guidelines for a community-based strategy as a way of identifying persons with an undiagnosed HIV infection. Our study using the pilot programme, which was conducted in the period from January 2006 to June 2007, was a timely effort, as the 2006 guidelines recommended incorporating new approaches to STD eradication, including patient-delivered partner therapy.
The utility of SNI or social networks testing could be evaluated by variables including, but not limited to, the number of new test positives identified (n=3), the percentage of tests provided among test-offered individuals (85%), the percentage of people tested who have engaged in high-risk behaviours (100%) and the percentage of those receiving testing who would not otherwise have done so (100%). As a result, this study supports that our pilot programme is result-oriented and also a cost-effective model, as there was virtually no additional cost or infrastructures that had to be implemented, particularly for the pilot programme.
Outreach programmes using SNI could be improved in order to reach a larger population. SNI could potentially be more efficient in reaching a larger proportion of referred contacts in a short period of time by using multiple CHW. Our pilot programme utilised one female CHW who interviewed both sex partners and social friends, including both female and male interviewees. Future outreach programmes could possibly improve the response rate by using same-gender interviewers once the gender of social and sexual contacts was identified.
Conclusion
With only one CHW charged with interviewing and locating contacts, three individuals were newly diagnosed with HIV infection. These three incidence cases were unaware of their HIV infection status before our pilot programme, and this suggests that current strategies for identifying HIV positives are insufficient.
A number of studies have called for innovative approaches to HIV prevention, screening and treatment, particularly among young minority populations.17–19 Studies of marginalised populations such as young African American MSM indicate that social networks may function to increase HIV transmission risk independent of individuals' behavioural risk factors.20 Attempting to contact social and sex partners of HIV-infected youth patients could be a valuable approach to identify high-risk individuals to offer CTR who would not otherwise seek it.
Key messages
Current strategies for identifying undiagnosed HIV-infected youth are insufficient and need more proactive approaches to provide HIV CTR services.
SNI provided by a CHW led to the discovery of three new HIV-positive cases, who may not otherwise have been identified.
The utilisation of SNI embedded in routine HIV clinic care could be a viable approach to identify high-risk individuals in networks of HIV-infected youths to offer CTR.
References
Footnotes
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.