aDepartment of Primary Care and Population Sciences and Royal Free Centre for HIV Medicine, Royal Free and University College Medical School, University College London, London, UK; and bHIV Unit, Royal Free Hampstead NHS Trust Hospital, London UK.
Received: 7 September 2000;
revised: 21 December 2000; accepted: 10 January 2001.
Sponsorship: Camden and Islington Health Authority provided financial support.
An increasing number of gay men who seek a voluntary HIV test, including those who test positive, have had at least one previous test with a negative result (known as repeat testers) [1–3]. Clinics that record the date and result of previous HIV tests provide the opportunity for measuring HIV incidence, because this can only be estimated among individuals who have already tested negative [4–11]. In this London testing clinic, between 1997 and 1998, nearly three-quarters of gay men were repeat testers . Gay men with a history of three or more previous negative HIV tests were significantly more likely to report high-risk sexual behaviour than those who had had one, two or no previous tests. High-risk sexual behaviour was defined as unprotected anal intercourse with a partner of unknown HIV status or who was HIV positive . Information on sexual risk behaviour was, however, self-reported and was restricted to the 3 months before the most recent test. Does self-reported sexual behaviour in the previous 3 months fully reflect repeat testers’ exposure to risk since their last negative HIV test? To answer this question we examined whether the incidence of HIV infection among repeat testing gay men was also elevated among those with three or more previous negative HIV tests. In addition, time trends in HIV incidence among repeat testers were investigated.
The methods have already been described  as has the HIV testing algorithm used in this clinic . In brief, between September 1997 and July 1998, all individuals (n = 2100) attending the same-day HIV testing clinic at the Royal Free Hampstead NHS Trust Hospital, London, were invited to complete an anonymous self-administered questionnaire concerning sexual risk behaviour in the previous 3 months (including type, number and HIV status of partner(s) for unprotected penetrative sex), the number of previous HIV tests, the date of the last test and the result. Their HIV test result was linked to the questionnaire by a numeric code that preserved confidentiality and anonymity. Of the 1580 individuals (75% response) who completed the questionnaire, 470 were gay men, of whom 337 (72%) were repeat testers with a previous negative test result.
Exposure to risk for repeat testers was defined as the interval between the last negative test according to the self-reported date and the current test. Only men whose previous test was at least 3 months before their current test were eligible for the analysis. Excluding individuals who may have been in the window period at the time of their last test ensured that all new diagnoses of HIV were a result of transmission that had occurred since the previous negative test. We also excluded from the analysis repeat testers who did not provide information on the number of previous tests, the date of their last test or their age (total exclusions n = 62). Incidence rates were calculated by dividing the number of newly diagnosed cases of HIV among repeat testers by the person-years of exposure since the last test. For those diagnosed HIV positive, the precise date of infection was unknown. However, among those who had tested negative 3–12 months previously, infection must have been recently acquired. Ninety-five per cent confidence intervals (CI) for rates were calculated using the Poisson distribution; rates were compared between subgroups using Poisson regression.
Among the 275 gay male repeat testers included in the analysis (median age 31 years, median time since last HIV test 24 months), 151 (55%) had had one to two previous tests, whereas 124 (45%) reported three or more previous tests. Twelve men tested HIV positive. The overall HIV incidence was 1.8 per 100 person-years (95% CI 0.9, 3.2, Table 1). HIV incidence was higher for men reporting three or more previous HIV tests (3.6%) than for those reporting one or two previous tests (1.1%) (incidence ratio 3.3, 95% CI 1.1–10.5, P = 0.04, Table 1). HIV incidence was also elevated among men whose previous test was within the past 12 months (4.7%) compared with those who had tested more than 12 months before (1.4%) (incidence ratio 3.4, 95% CI 1.0–11.2, P = 0.05, Table 1). Among the 108 men who had tested negative 3–12 months previously, all new diagnoses of HIV (n = 4) were among men (n = 67) with a history of three or more previous HIV tests (HIV incidence 8.0%; four cases/49.9 person-years at risk). There was no significant difference in HIV incidence between younger (below the median age) and older men (overall 2.1 versus 1.7%, respectively, P = 0.9; tested in the previous 12 months, 6.3 versus 2.7%, P = 0.5). Nonetheless, the highest HIV incidence was observed among younger (i.e. below the median age) men with a history of three or more previous HIV tests who had last tested negative in the previous 12 months (10.7%, three cases/28.1 person-years).
In this London clinic, HIV incidence was higher for gay men reporting three or more previous negative HIV tests than for those with one or two previous tests. This has not been reported before and provides supporting evidence for our earlier finding of elevated levels of high-risk sexual behaviour among gay male multiple repeat testers. Forty-two per cent of men with a history of three or more previous negative HIV tests reported high-risk unprotected anal intercourse compared with 25% of men with one, two or no previous tests (P = 0.002) . For some gay men, repeatedly receiving a negative HIV test result may have had a disinhibiting effect, providing reassurance and reinforcing patterns of high-risk sexual behaviour. A similar phenomenon was observed in a longitudinal study in the USA among men who tested negative for HIV. During follow-up, the incidence of sexually transmitted diseases increased after receiving a negative HIV test result compared with the incidence before the test . The number of previous tests was not reported. Although post-test counselling provides an opportunity to discuss risk reduction with repeat testing gay men, only limited time is available for this. Health promotion initiatives outside the clinic should be developed, targeting those men for whom a repeat negative test result may have a disinhibiting effect.
In this clinic, between 1997 and 1998, the incidence of HIV among gay men who had tested negative within the previous 12 months (i.e. for whom exposure to risk was recent) was 4.7 per 100 person-years. This is broadly consistent with HIV incidence estimates reported between 1988 and 1996 among gay men seeking voluntary repeat HIV tests in other London testing clinics (5.0–6.5 per 100 person-years) [9–11]. In both our analysis and the earlier studies, HIV incidence rates were higher among those reporting a shorter interval between consecutive tests. Taken together, these data suggest that annual HIV incidence rates among gay men seeking a repeat HIV test in London remained stable between 1988 and 1998 with no evidence of a decline. This finding is consistent with behavioural surveillance data, which showed that levels of high-risk sexual behaviour among London gay men also remained stable during most of that period [14,15]. A recent report of increasing high-risk sexual behaviour among London gay men since late 1997 raises the possibility of HIV incidence now increasing in this population .
Our analysis relied on the self-reported date of the last test, which may be subject to recall error; the number of seroconversions was small; and the incidence of HIV among repeat testers may differ from that among first-time testers or those who did not come back for a repeat test. Nonetheless, our findings are broadly consistent with those found in other studies [9–11].
To conclude, in this London clinic, the incidence of HIV was elevated among gay men reporting three or more previous negative HIV tests. Furthermore, there was no evidence of a decline in HIV incidence among gay male repeat testers in London between 1988 and 1998.
The authors would like to thank Amanda Jones, same-day HIV testing clinic staff, HIV counsellors and all those who completed a questionnaire for their support.
1. Phillips KA, Paul J, Kegeles S, Stall R, Hoff C, Coates TJ. Predictors of repeat HIV testing among gay and bisexual men.
AIDS 1995, 9: 769 –775.
2. Norton J, Elford J, Sherr L, Miller R, Johnson M. Repeat HIV testing a London same-day testing clinic.
AIDS 1997, 11: 773 –781.
3. Leaity S, Sherr L, Wells H, Evans A, Miller R, Johnson M, Elford J. Repeat HIV testing: high-risk behaviour or risk reduction strategy?
AIDS 2000, 14: 547 –552.
4. McFarland W, Kellogg TA, Dilley Katz MH. Estimation of human immunodeficiency virus (HIV) seroincidence among repeat anonymous testers in San Francisco.
Am J Epidemiol 1997, 146: 662 –664.
5. Kellogg T, McFarland W, Katz M. Recent increases in HIV seroconversion among repeat anonymous testers in San Francisco.
AIDS 1999, 13: 2303 –2304.
6. Suligoi B, Giuliani M, Galai N, Balducci M, and the STD Surveillance Working Group. HIV incidence among repeat HIV testers with sexually transmitted diseases in Italy.
AIDS 1999, 13: 845 –850.
7. Weinstock H, Sweeney S, Satten GA, Gwinn M, for the STD clinic HIV Seroincidence Study Group. HIV seroincidence and risk factors among patients repeatedly tested for HIV attending sexually transmitted disease clinics in the United States 1991 to 1996.
J Acquir Immune Defic Syndr 1998, 19: 506 –512.
8. Meyer L, Couturier E, Brossard Y. et al
. Trends in HIV infection among sexually transmitted disease patients in Paris.
AIDS 1996, 10: 401 –405.
9. Waight PA, Miller E. Incidence of HIV infection among homosexual men.
BMJ 1991, 303: 311. 311.
10. Miller E, Waight PA, Tedder RS, Sutherland S, Mortimer PP, Shafi MS. Incidence of HIV infection in homosexual men in London, 1988–94.
BMJ 1995, 311: 545. 545.
11. Miller E, Waight P, Mercey D, Parry J, Newham J, Nicoll A. Incidence of HIV infection in homosexual men attending a sexually transmitted disease clinic in London measured by a novel unlinked anonymous method.
AIDS 1999, 13: 143 –144.
12. Squire SB, Elford J, Bor R. et al
. Open access clinic providing HIV-1 antibody test results on day of testing: the first twelve months.
BMJ 1991, 302: 1383 –1386
13. Otten MW, Zaidi AA, Wroten JE, Witte JJ, Peterman TA. Changes in sexually transmitted disease rates after HIV testing and posttest counselling, Miami, 1988 to 1989.
Am J Pub Health 1993, 83: 529 –533.
14. Hickson FCI, Reid DS, Davies PM, Weatherburn P, Beardsell S, Keogh PG. No aggregate change in homosexual HIV risk behaviour among gay men attending the Gay Pride Festivals, United Kingdom, 1993–1995.
AIDS 1996, 10: 771 –774.
15. Nardone A, Dodds JP, Mercey D, Johnson AM. Active surveillance of sexual behaviour among homosexual men in London.
Commun Dis Public Health 1998, 1: 197 –201.
16. Dodds JP, Nardone A, Mercey DE, Johnson AM. Increase in high risk sexual behaviour among homosexual men, London 1996–98: cross-sectional, questionnaire study.
BMJ 2000, 320: 1510 –1511.