CARBALLO-DIÉGUEZ, ALEX PhD*; REMIEN, ROBERT H. PhD*; BENSON, DEBORAH A.†; DOLEZAL, CURTIS PhD*; DECENA, CARLOS U. MA*; BLANK, SUSAN MD†
PARTNER NOTIFICATION (PN) is the long-standing public health practice of notifying the sexual partners (“partners”) of individuals diagnosed with a sexually communicable disease (“index cases”) of possible exposure to a pathogen and the need for medical evaluation. Three forms of PN are generally used: provider referral, patient referral, and contract referral. In provider referral, the health provider treating the index case elicits information on sexual partners and notifies those partners of possible exposure. In patient referral, the index case is encouraged to notify partners on his or her own. Contract referral combines elements of both provider and patient referral: the index case is encouraged to conduct the notification within a certain period, but if the notification does not take place, partners are notified by the health-care provider. 1 The Centers for Disease Control and Prevention has issued PN guidelines under the rubric HIV Partner Counseling and Referral Services. 2
PN has been used with varying degrees of success for sexually transmitted diseases (STDs) such as syphilis, gonorrhea, and chlamydia, illnesses for which there are known treatments and cures. In the case of HIV, the absence of a cure and the lifetime duration of the infectious period make the use of PN more controversial. Some sectors, especially in the public health establishment, support PN; others oppose it and equate PN to “bedroom police.”3–5
Macke and Maher 6 conducted a systematic review of the literature on the effectiveness of PN strategies for syphilis, gonorrhea, chlamydia, and HIV in the United States. They found both good evidence that PN detects previously undiagnosed infections and fair evidence that more partners are notified and medically evaluated through provider referral than through patient referral. Nevertheless, the authors point out that it is not known whether PN is acceptable to partners or which referral strategy (provider, patient, or contract referral) is most acceptable. These issues require exploration.
In 1998, New York State legislators passed an amendment to the public health law (chapter 163) mandating that both sexual partners and injection-drug-using (IDU) partners of individuals newly diagnosed with HIV be notified of their risk of HIV exposure. In the ensuing hearings of public testimony before the New York State AIDS Advisory Council, community groups expressed reservations about the law, fearing that individuals, particularly people of color, would shy away from testing for fear of loss of confidentiality. In focus group interviews conducted in 1999 with minority community–based-organization service providers in the five boroughs of New York City (NYC), 7 concerns about the possible negative effects of the partner-notification law were also strongly voiced. Therefore, it became evident that there was a need to explore public perceptions of PN.
This study was undertaken before the PN law was implemented in NYC on June 1, 2000. The aims of the study were to assess the willingness of individuals seeking HIV testing to (1) give contact information about sex and needle-sharing partners to the HIV counselors in the hypothetical case that the HIV test were positive and (2) contact sex partners on their own to suggest HIV testing. Furthermore, the study sought to assess and compare the respondents’ emotional reactions to being referred to the clinics by Department of Health (DOH) employees or by index cases.
Investigators from the HIV Center for Clinical and Behavioral Studies and members of the STD Control Program of the NYC DOH jointly developed a one-page survey covering the main topics of interest for this study. It explored how many different partners of each gender the respondent had had in the prior 2 months, with how many of them the respondent had had vaginal or anal sex, and with how many of them sex was not protected by condoms. Subsequently, the survey explored how many of those unprotected sex partners the respondent could contact if he or she wished to, and, in the event of the respondent's HIV-positivity, how many of these “traceable” partners the respondent would be willing to contact on his or her own to suggest HIV testing, as well as for how many traceable partners the respondent would be willing to give contact information to a counselor. Questions on sexual behavior were limited to the 2 months before the interview to avoid recall problems observed with longer periods. 8
Concerning substance use, we asked, “During the past 2 months, on how many occasions did you use street drugs (drugs not prescribed by a physician)?” Those whose answer was not zero were asked with how many different people they shared needles or works to inject drugs and with how many different people they shared needles or works that were not clean. The subsequent questions paralleled those of the sexual behavior section, asking how many of such partners the individual would be willing to contact on his or her own to suggest HIV testing and for how many traceable partners the respondent would be willing to give contact information to a counselor.
An equivalent Spanish version of the survey was produced with use of the method of translation and back-translation 9,10 and was printed on the other side of the English questionnaire. The survey instrument and study design were approved by the institutional review boards of the participating institutions.
The NYC DOH STD Control Program operates 14 sites throughout the five city boroughs, where anonymous and confidential HIV testing is offered on a walk-in basis; 10 of these sites also offer STD diagnostic and treatment services. All sites participated in the study. A per-clinic target number of interviews was established, based on the prior year's HIV-testing rates per location. The investigators met with each clinic's HIV-testing counselors, enlisted their collaboration, and instructed them on a script for requesting voluntary participation in the survey from all individuals presenting for HIV testing. All clients who underwent pre-HIV-test counseling were invited to respond to an anonymous self-administered questionnaire while they were waiting to have their blood drawn. Participants then deposited the survey in a locked box and were given a $3 Metrocard (good for two subway rides) as compensation.
The survey was conducted between January 20 and May 31, 2000. Recruitment start and finish dates varied per clinic. As soon as partial summary results per clinic became available, the investigators met with clinic counselors to discuss the data and their interpretation.
The NYC DOH routinely elicits and records demographic information from people requesting HIV tests at STD clinics. A summary of demographic data corresponding to all STD clients seen for HIV testing during the survey period was produced by the NYC DOH for comparison with the survey results.
There were 1372 survey participants, representing 11% of the total number of clients seeking confidential and anonymous HIV testing at all 14 NYC DOH STD clinics during the survey period. Survey respondents’ average age was 30 years (range, 14–74 years; 5% of the sample was younger than 18 years), while the average age of the clinics’ clients was 31 years (range, 13–83 years; t = 2.73;P = 0.006). Table 1 presents the demographic characteristics of survey respondents in comparison with those of the total HIV testing client population. There were statistically significant differences in gender, ethnicity, and health insurance coverage. Compared with the total clinic HIV-testing client population, survey respondents were less likely to be men and African American and more likely to have health coverage. More survey respondents reported having Medicaid, a form of public health insurance, than the clinic population.
One third of survey respondents reported that they had gone to the clinic the day of the survey at someone's request: sex partner, 59%; DOH employee, 8%; and other (from most to least frequent: friends, family members, doctors, and personnel at residential facilities), 33%. We cannot assume, however, that all sex partners referring individuals to the clinic were HIV-infected; for instance, they might have had another STD or they may have requested HIV testing as a precondition of nonuse of condoms. Also, not all referrals by DOH employees are strictly due to PN; for example, counselors may refer individuals diagnosed with other communicable diseases for HIV testing.
Most people referred to the STD clinic reported feeling comfortable (44%) or neutral (33%) about the request; only 24% said they felt uncomfortable. However, there were differences according to the source of referral. Of those referred by a sex partner, the largest proportion (105; 42%) felt comfortable; 81 (33%) felt neutral; and 61 (25%) felt uncomfortable. Of those referred by the DOH, the largest proportion (16; 46%) felt uncomfortable; 10 (27%) felt neutral; and 9 (26%) comfortable (chi-square P = 0.03). Given a choice, 31% of respondents would have chosen to be told to go to the clinic by the person who may have exposed them to HIV; 21% would have chosen their physician; 10% would have chosen a DOH employee; and 38% would have chosen others (friends were ranked first, followed by family members).
The majority of respondents (71%) had previously been tested for HIV (mean time lapse from prior testing was 1 year 11 months; median, 1 year 1 month; range, 2 weeks to 14 years 9 months); 29% had never been tested before. Of those previously tested, 8% reported having tested HIV-positive. Clinic personnel reported that, among other reasons, HIV-positive individuals sometimes seek retesting because they want confirmation of results, they need a documented result to apply for public benefits, they hope that undergoing antiretroviral therapy may render the HIV test results negative, or they have a new partner and they use the retest as an opportunity to disclose their HIV status (claiming they did not know about the infection before).
Those respondents who had not tested HIV-positive before were asked whether, before being advised to attend the clinic and since their last HIV test (if applicable), they thought they might have been exposed to HIV; 72% responded no and 28% yes.
An open-ended question inquired why respondents had not been tested until the day of the survey. The most frequent response was “fear.” Many respondents stated that they underwent the HIV test at regular intervals—ranging from 3 months to 1 year—and were just waiting for the “due” time. Others said they did not feel at risk, they did not have time, or that a potentially risky incident had taken place only recently.
Of the 669 female participants, 452 (71%) reported having sex with men, 60 (9%) with women, and 35 (5%) with both within the prior 2 months. (Unfortunately, the question on vaginal sex [With how many of these female partners did you have vaginal or anal sex?] appears to have lacked specificity for women who have sex with women. It could have been understood as meaning oral–genital contact, fingering, use of dildos, etc. Given this lack of clarity, we have decided to exclude data on sexual risk behavior among women from the present report.) Of the 700 male participants, 458 (67%) reported having sex with women, 106 (15%) with men, and 17 (2%) with both. Approximately one third of the respondents reported having no heterosexual sexual activity during the prior 2 months, and more women (23%) than men (18%) reported neither male nor female partners (chi square P = 0.019).
Table 2 shows that 90% or more of the heterosexually active women and men had engaged in penetration (vaginal or anal), versus three fourths of MSM who had had anal penetration. More than 80% of heterosexually active respondents had had unprotected sex, compared with less than 60% of MSM. Overall, women and men who had had unprotected genital sex overwhelmingly reported they were able to contact their partners if they so wanted, although MSM were less likely to be able to do so. Respondents expressed willingness both to provide PN information on at least one partner and to notify at least one partner on their own in the event of HIV-positivity. MSM were as willing as heterosexually active respondents to contact partners on their own. However, heterosexual men (excluding men with bisexual behavior) where significantly more likely (P ≤ 0.05) than MSM to give information to the DOH. While there was a trend toward a difference between heterosexual women and MSM, that difference did not reach statistical significance (P = 0.08). No interethnic differences were observed in terms of willingness to contact partners or give information to the DOH.
All sexually active respondents were asked how many different male and female sexual partners they had had during the prior 2 months. Table 3 shows that more women than men had only one sex partner.
Table 4 shows that, excluding those partners with whom respondents did not have unprotected vaginal or anal sex and those they could not contact even if they wanted, the “notification burden,” i.e., the number of identified partners to be notified by either the DOH or the respondent, ranges from 0.95 male per female index case to 1.46 males per male index case. This figure corresponds to sexual partners during the prior 2 months and could be larger if longer periods were considered.
Of the 1234 individuals who responded to the questions on substance use, 1117 (90.5%) said they had not used street drugs in the prior 2 months. Only 117 (8%) acknowledged any use. Although the NYC DOH does not routinely collect information on substance use among individuals seeking HIV tests in a comparable format, clinic counselors estimate that their proportion is higher than 8%. In the opinion of clinic counselors, some respondents may have not considered marijuana, for example, a street drug since it is often not purchased in the street (dealers, who can be beeped, deliver marijuana to residences); also, it may be considered “herbal” and natural. It is also possible that, despite the anonymity of the survey, respondents did not want to acknowledge illegal activity. Within these constraints, the data showed that 85% of substance users would be willing to give contact information to the DOH on those partners with whom they had had unprotected sex and they could contact; drug-using women were less likely than non-drug-using women to provide information on male partners with whom they had unprotected sex (68% versus 91%;t test P = 0.028). There were no differences between drug-using and non-drug-using men. In terms of willingness to notify partners on their own, 97% of substance users reportedly would do it, and there were no significant differences between men and women. Of the 117 substance users, only seven reported sharing drug paraphernalia.
This study shows that immediately before the implementation of the Partner Notification Law in NYC, more than 90% of heterosexually active NYC DOH STD clients and more than 80% of MSM in this population could contact most of those sex partners with whom they had had unprotected vaginal or anal sex. Furthermore, almost all individuals who could contact sexual partners expressed willingness to notify them of their potential exposure to HIV in the event of HIV-positivity, and about 90% of heterosexually active individuals and 80% of MSM expressed willingness to provide contact information to the DOH for PN purposes.
These outcomes are in sharp contrast with community advocates’ views that forecasted strong resistance to PN among at-risk individuals. What are the reasons for this discrepancy? We suggest some possible interpretations.
First, the sample may lack representativeness. Compared with the population seen for HIV testing at the 14 STD clinics during the survey period, survey respondents differed at a statistically significant level in age, gender, race/ethnicity, and health coverage. However, the large size of the sample contributes to this result, despite the apparent similarity of both groups. Furthermore, concerning health coverage, for example, individuals may have given less honest responses in the face-to-face interviews with the counselors than in the anonymous survey: 71% of clinic clients seeking HIV testing told the counselors they had no health insurance (compared with 50% of survey respondents), and only 2% of the former acknowledged being covered under Medicaid (compared with 19% of the latter). Perhaps concerns about confidentiality led individuals to deny they had public health coverage.
Second, the self-selection of respondents may affect the results, especially since our data did not allow us to discriminate between respondents choosing anonymous over confidential testing (both available at all STD clinics). It could be argued that if our survey respondents were mainly choosing confidential testing, they might have higher trust in the system than anonymous clients. However, in studying a cross-sectional probability sample of HIV-infected people in five states with name-based HIV-surveillance, Osmond et al 11 found no differences between the number of partners notified by those testing anonymously versus those testing confidentially.
Third, the survey was administered immediately after completion of a pre–HIV test counseling session. The issues raised during the session and the recall of past sexual risk behavior, both in the session and while filling out the survey, may have heightened a sense of duty to warn partners about possible HIV exposure. Furthermore, if the interaction with the counselor was satisfactory and clients felt treated with empathy, respect, and confidentiality, they may have become more trustful of the system and willing to provide partner contact information. Respondents may have also given socially desirable responses.
Fourth, because HIV status was not known, in most cases, at the time of the survey, the questions about willingness to notify a partner were hypothetical. Intentions may not translate into actual behavior. In the words of a counselor, “It's like skydiving: you may want to do it, but you really don't know what it's like until you are faced with it.” Also, 98% of individuals seeking HIV testing at NYC DOH STD clinics are uninfected and therefore may never be confronted with the dilemma of having to notify a partner. Nevertheless, intentions may constitute an essential building block of an effective PN program.
Within these constraints, the results of this study provide substantial evidence to be taken into consideration for policy development. It is important to note that respondents favor patient referral over provider notification. Given a choice, three to one preferred to be notified by the person who may have exposed them to HIV rather than by a DOH employee; furthermore, three fourths of respondents who had actually been referred by a partner felt either neutral or comfortable about the referral. Among those referred by DOH employees, twice as many respondents felt uncomfortable about the referral, versus those referred by sex partners. These results probably reflect the preference of respondents for a sexual partner who takes responsibility for informing them about the risk of infection rather than delegating the obligation to a stranger. Given the literature reports 12–14 that show that not every infected individual notifies current sexual partners of his/her infection and even fewer notify past sexual partners, our findings suggests that the preferred system is that of contract notification in which the first step in the process is the voluntary notification of a partner by the index case, followed by provider notification if the initial step does not take place.
About two thirds of respondents had tested HIV-negative before and were being retested. Although the range of time elapsed since the prior test varied considerably, there were indications that many people get tested periodically (every 3, 6, or 12 months) as a health check-up. This affords both positive and negative interpretations. On the positive side, routine test-seekers may be vigilant about their health, especially if we consider that three fourths of them thought they had not been exposed to HIV; also, if they were to seroconvert, these individuals would find out about it early and could seek prompt treatment. On the negative side, individuals who repeatedly test negative may get a false reassurance from the result and subsequently engage in further risk behavior.
The finding that fear of a positive diagnosis is the most frequent deterrent to HIV testing presents a challenge for HIV prevention strategists. Stressing the availability of relatively effective HIV therapies to encourage testing may create the public perception that HIV is no longer a serious threat. On the other hand, if HIV prevention campaigns underscore that no one has been cured from AIDS yet and that therapies are very burdensome, people may continue to avoid HIV testing out of fear of finding out the results.
Compared with heterosexually active individuals, fewer MSM report engaging in penetration and fewer MSM are able to contact their partners; this may indicate that a higher number of MSM engage in oral sex or mutual masturbation with anonymous partners. Also, fewer MSM than heterosexually active respondents engage in unprotected penetration. The fact that fewer MSM than heterosexuals would give contact information to the DOH may result from the effect of political activism in the gay community against name disclosure or general mistrust of authorities.
In terms of number of partners and consistent with cultural norms, women tended to have fewer sexual partners than men. MSM tended to have more partners than heterosexual men and women. The notification burden ranges approximately from 1 to 1.5 partners per respondent.
The data provided by substance users show that they were also very likely to be able to contact sex partners and to a large degree be willing to alert them of possible HIV exposure. This gives further support to the findings of Levy and Fox 15 and contradicts stereotypes about substance users not having concern about their peers. There was more reluctance to give contact information to the DOH, particularly among women. This finding coincides with the observations reported by Rogers et al 16 of their qualitative study of HIV-infected substance users and drug counselors in NYC.
Although intention to notify is an important component of the process of PN, there are many other important steps: ability of professionals to elicit partner information from infected persons, success in reaching partners unaware of possible exposure, willingness of partners to be tested, and treatment follow-up for those infected.
Finally, there is no doubt that individuals who are unaware of their HIV infection may benefit by being notified of their exposure to HIV and being offered the possibility of state-of-the-art treatment. Yet, from an epidemiologic perspective, as Cowan et al 1 state it, “partner notification for viral STDs including HIV is only worthwhile if people who become aware of their infectivity modify their behavior to reduce their risk of further disease transmission.”1
There are few data available on this issue. 17,18 Further research is needed to address this very important concern.
1. Cowan FM, French R, Johnson AM. The role and effectiveness of partner notification in STD control: a review. Genitour Med 1996; 72: 247–252.
2. Centers for Disease Control and Prevention, US Department of Health and Human Services. HIV partner counseling and referral services: guidance. Atlanta: Centers for Disease Control and Prevention, 1998.
3. Bayer R, Toomey K. HIV prevention and the two faces of partner notification. Am J Public Health 1992; 82: 1158–1164.
4. Potterat JJ. Contact tracing's price is not its value [editorial]. Sex Transm Dis 1997; 24: 519–521.
5. Potterat JJ, Muth SQ, Muth JB. Partner notification early in the AIDS era: misconstruing contact tracers as bedroom police. Res Soc Policy 1998; 6: 1–15.
6. Macke BA, Maher JE. Partner notification in the United States: an evidence-based review. Am J Prev Med 1999; 17: 230–242.
7. New York City Prevention Planning Group. HIV Prevention Plan 2001. 4 (Report available from the New York City Department of Health, Prevention Planning Unit, 225 Broadway, New York, NY).
8. Catania JA, Gibson DR, Chitwood DD, Coates TJ. Methodological problems in AIDS behavioral research: influences on measurement error and participation bias in studies of sexual behavior. Psychol Bull 1990; 108: 339–362.
9. Marin G, Marin BV. Research with Hispanic Populations. Applied Social Research Methods Series. Vol. 23. Riverside, California: Sage Publications, 1991.
10. Zea MC, Reisen CA, Poppen PJ. Psychological well-being among Latino lesbians and gay men. Cult Divers Ethnic Minor Psychol 1999; 5: 371–379.
11. Osmond DH, Bindman AB, Vranizan K, et al, for the Multistate Evaluation of Surveillance for HIV Study Group. Name-based surveillance and public health interventions for persons with HIV infection. Ann Intern Med 1999; 131: 775–779.
12. Perry S, Ryan J, Fogel K, Fishman B, Jacobsberg L. Voluntarily informing other of positive HIV test results: patterns of notification by infected gay men. Hosp Commun Psychiatry 1990; 41: 549–551.
13. Perry SW, Card C, Moffatt M, Ashman T, Fishman B, Jacobsberg L. Self-disclosure of HIV infection to sexual partners after repeated counseling. AIDS Educ Prev 1994; 6: 403–411.
14. Wolitski RJ, Rietmeijer CAM, Goldbaum GM, Wilson RM. HIV serostatus disclosure among gay and bisexual men in four American cities: general patterns and relation to sexual practices. AIDS Care 1998; 10: 599–610.
15. Levy JA, Fox SE. The outreach-assisted model of partner notification with IDU's. Public Health Rep 1998; 113: 160–169.
16. Rogers SJ, Tross S, Doino-Ingersol J, Weisfuse I. Partner notification with HIV-infected drug users: results of formative research. AIDS Care 1998; 10: 415–429.
17. Fenton KA, Peterman TA. HIV partner notification: taking a new look. AIDS 1997; 11: 1535–1546.
18. West GR, Stark KA. Partner notification for HIV prevention: a critical reexamination. AIDS Educ Prev 1997; 9 (suppl B): 68–78.