An important component of HIV control is early diagnosis through frequent HIV testing.1–3 Early diagnosis and treatment has individual health benefits and a preventive effect.4,5 Although much of the public health response to HIV has focused on increasing the proportion of the population that is tested for HIV,6–8 the importance of a high frequency of testing has only been recognized more recently.9,10 Few interventions to increase frequency of testing have been subjected to randomized controlled trial (RCT), and all have used rapid HIV tests. Two RCTs offered oral rapid tests to men who have sex with men (MSM) to use at home and demonstrated a significant increase in testing frequency over time, whereas another offering finger-stick rapid tests in a clinic did not.11–13 These studies provided oral tests for free; however in practice, they must be purchased. Furthermore, oral tests have a longer seroconversion window-period than antigen–antibody laboratory tests and are more likely to miss seroconverting individuals with high HIV RNA loads.14,15
Australian guidelines recommend that annual testing for HIV and sexually transmitted infections (STI) for MSM increasing to 4 per year for men with 10 or more partners in 6 months.16 However, only about half of Australian MSM have HIV tests at least annually.17,18 Models from the United Kingdom suggest that a 3-fold increase in testing would be needed to diagnose more than 90% of HIV infections within 1 year.19 Barriers to HIV testing have been identified, and many result from the requirement for a clinical consultation, including leaving work, travelling, waiting, and sometimes, paying to see a clinician who orders the test and sometimes repeating the process for the result.20,21 Health services can change the way they provide testing to increase convenience and reduce costs, and some major sexual health centers have introduced express clinics, but these may not be conveniently located for all.22,23 By contrast, pathology collection centers, where doctors send patients for blood tests, are found in most suburbs and towns and would give MSM easy access to sensitive antigen–antibody laboratory HIV tests. These centers typically have some extended hours such as Saturday mornings or before 9 am, enabling workers to attend. Medical care (including antiretroviral therapy) is subsidized for Australian citizens, and laboratory HIV tests are usually free. We hypothesized that by removing the requirement for patients to see a clinician to order an HIV test, the increased convenience of access to pathology collection centers might increase the rate of testing.
Aiming to reduce practical barriers to testing faced by MSM, we conducted an RCT in which we gave MSM pathology forms allowing free HIV and syphilis tests at a pathology collection network, without requiring a consultation before every test, and we measured the effect of this on the frequency of testing.
Participants were randomized 1:1, without blinding, either to a standard of care control group who were required to have a consultation with testing or to an intervention group who were allowed HIV and syphilis serology at multiple locations in Melbourne, without the need for a consultation, over a 1-year period. Trial registration: Australian New Zealand Clinical Trials Registry ID ACTRN12614000760673.
Inclusion criteria were men aged 18 years or older, reporting sex with another man within the previous year, having a negative HIV test at recruitment or within the previous 6 weeks, and intending to remain in the state of Victoria for at least 6 months.
Study Setting, Consent, and Recruitment
Recruitment occurred at the Melbourne Sexual Health Centre (MSHC), the only public sexual health centre in Melbourne, Australia, a city of 4.7 million. During the 2015-2016 financial year, the MSHC tested 5587 MSM, performing 9668 tests of which 52 (0.54%) were positive. Clinicians offered study participation to eligible patients, referring them to a research nurse for verbal and written information about the study. Written consent was obtained from those opting to participate. The study ran from July 2014 to April 2015, and the last participant completed follow-up in April 2016.
Men in the intervention group were issued with 6 pathology request forms for HIV and syphilis testing (free of charge) valid for 12 months at any of 300 pathology collection centers operated by Melbourne Pathology. They were also allowed unlimited HIV and syphilis testing without a consultation at the enrolment site, MSHC. This did not include testing for chlamydia or gonococcal infection, which required consultation. Participants in the intervention group could request a consultation if they wished. Staff meetings and alerts in patient electronic records ensured that participants in the intervention group were offered serology without consultations, when they attended the clinic. Access to the intervention was limited to 12 months to ensure that no participant was tested for longer than this without pretest counseling or STI testing. HIV tests at Melbourne Pathology and at the MSHC were fourth-generation antigen–antibody assays, confirmed by Western Blot.
Results from Melbourne Pathology tests were available by telephone from the MSHC, and negative results from the MSHC were communicated by text message. Individuals in both groups with positive serology were contacted by clinic staff, booked into the MSHC HIV clinic, and removed from the study. Participants in both groups received HIV pretest counseling and were permitted to have HIV and syphilis testing by consultation at any other health service.
Both groups received text messages at months 2, 4, 8, and 10 recommending frequent testing (month 6 was reserved for an e-mail asking about testing). The message also reminded the intervention group that they could have free tests at Melbourne Pathology collection centers and MSHC and linked to a Web page. This provided a telephone number to replace lost pathology request slips and a list of the locations of collection centers.
The primary outcome was the incidence rate of HIV testing over 12 months, excluding tests at the recruitment visit and excluding tests to confirm a positive result.
HIV and syphilis testing at Melbourne Pathology and the MSHC was ascertained from electronic records. Participants were e-mailed at months 6 and 12, and those who did not reply were sent text messages, asking if they had been tested for HIV elsewhere and if so when and where. Researchers contacted external clinics or laboratories to confirm those reported tests and the date of testing. A baseline questionnaire and an online questionnaire at 12 months included questions about acceptability of HIV testing, perceived barriers to testing, sexual behavior, and quality of life (using the SF6D assessment).24 Quality of life was assessed to determine whether this was affected by the intervention. The rate at which participants had their first HIV test (excluding subsequent tests) and the rate of HIV testing in the first 6 months were included as secondary outcomes because 1 trial showed an initial increase among MSM offered rapid testing that was not sustained over 18 months.11 The rate of clinical consultations was assessed in case this varied as a result of HIV tests being performed at external pathology collection centers rather than the clinic. Other secondary outcomes were the rate of syphilis testing and the acceptability of testing. We were unable to obtain Medicare data to monitor the rate of chlamydia and gonorrhoea testing in general practice, as originally planned, because the application process would have significantly delayed the trial.
MSM in a recent study at the MSHC had approximately 1.6 tests per year (SD 1.4).11 A sample of 192 participants in each study group provided 80% power to detect an increase in testing from 1.6 to 2.0 per year (25% increase), with a type 1 error of 5%. We added 10% to replace HIV-positive diagnoses and losses to follow-up, arriving at a recruitment target of 422.
Randomization and Masking
Participants were randomized with a computer-generated random sequence in blocks of 10 (5 intervention, 5 control) by a researcher not associated with the study. A research assistant also not associated with the trial put study allocations into 422 numbered, opaque, sealed envelopes, which were sequentially opened for each recruit. Research personnel enrolling patients (S.W. and R.W.) were unaware of the allocation until the envelope was opened in front of the patient. The study design required participants and clinic staff to be aware of the allocation.
We used intention-to-treat analysis to calculate the incidence rate of HIV testing in the intervention and control study arms, after excluding men with positive HIV serology at recruitment. Men who withdrew from the study, died, or were lost to follow-up were censored at their last contact with the study team. Men who acquired incident HIV infection during follow-up were censored at the time of their first reactive test. Kaplan–Meier methods were used to calculate time under observation and rates of testing. We estimated the incidence rate ratio (IRR) between the 2 arms of the study using Poisson regression with robust SEs to account for repeated measures from the same individual. To determine whether the intervention had any effect in men who identified difficulties with testing in the baseline questionnaire, we performed post hoc analyses that were restricted to those who reported difficulties, such as getting an appointment, cost, or waiting times. Logistic regression was performed to assess the impact of sexual behavior reported in the final questionnaire on the rate of testing (quartiles of the number of condomless anal sex partners during the study). Analyses were performed with Stata version 13 (StataCorp, College Station, TX). This study was approved by the Alfred Hospital Ethics committee.
Clinical staff referred 443 men to the study from July 24, 2014 to April 22, 2015, and 422 were eligible and consented (Fig. 1). Eight were not eligible because they did not plan to remain in Victoria, and 13 declined to participate. Three men had positive HIV tests on recruitment and were excluded from the analysis. Fifteen men were censored during the observation period because they were lost to follow-up or withdrew. One man died, and 1 man from the intervention arm was diagnosed with incident HIV during follow-up. He commenced antiretroviral therapy within a month. Study follow-up concluded on April 22, 2016. Of the 419 randomized HIV-negative men, the analysis included 211 from the intervention arm and 208 from the control arm, and 200/211 (94·8%) in the intervention arm and 202/208 (97.1%) in the control arm were followed for the full 12 months.
Men in the 2 study arms reported similar numbers of HIV tests and sexual partners in the previous year and had similar education and employment histories (Table 1).
HIV Testing Rates
Study participants had a total of 885 HIV tests. The intervention arm had 453 tests during 205.6 person-years, an incidence rate of 2.20 [95% confidence interval (CI): 2.01 to 2.41] tests per year. The control arm had 432 tests during 204.0 person years or a test incidence rate of 2.12 (95% CI: 1.92 to 2.33) tests per year. The IRR was 1.04 (intervention:control) (95% CI: 0.89 to 1.22; P = 0.63; Table 2). The mean intervals between tests were 166.02 days (95% CI: 151.56 to 181.72) and 172.29 days (95% CI: 156.76 to 190.23) in the intervention and control arms, respectively. There was no difference in the incidence rate of testing when the analysis was restricted to the first HIV test [IRR, 1.05; (95% CI: 0.85 to 1.30; P = 0·65; Figure 2). Proportions of participants who had any HIV tests during the study were 168/211 (79.6%) in the intervention arm and 166/208 (79.8%) in the control arm (P = 0.96). In the first 6 months of the study, the intervention arm had 218 tests during 104.3 person-years or a rate of 2.09 (95% CI: 1.82 to 2.39) tests per year. The control arm had 193 tests in 102 ·7 person-years or a rate of 1.88 (95% CI: 1.62 to 2.16) tests per year. The IRR (1.12; 95% CI: 0.92 to 1.36) was not significantly different between arms (P = 0.25).
Of 211 men in the intervention arm, 49 (23·2%; 95% CI: 18.0 to 29.5) had 101 HIV tests without consultations at pathology collection centers and 22 HIV tests without consultation at the MSHC. Men in the intervention arm had 60 fewer tests at the MSHC and 20 fewer tests requested in general practice than men in the control arm (Table 3). Because all tests in general practice required a consultation, men in the control arm had 432 tests from 432 HIV test–related consultations, but men in the intervention arm had 453 tests from 330 HIV test–related consultations. Men in the intervention arm had 330 HIV test-related consultations in 205·6 person years or a rate of 1.61 (95% CI: 1.44 to 1.79) consultations per year, and men in the control arm had 432 HIV test–related consultations in 204·0 person years or a rate of 2.12 (95% CI: 1.92 to 2.33) consultations per year. The IRR was 0.76 (95% CI: 0.66 to 0.88; P = 0.0001).
The men in the intervention arm had equal numbers of syphilis and HIV tests at pathology centers because the pathology forms requested both tests. At the MSHC, men in the intervention arm had 284 syphilis tests (12 more than HIV tests) and men in the control arm had 348 syphilis tests (16 more than HIV tests). All syphilis tests performed without an HIV test were to further investigate positive syphilis serology. Participants were not asked whether they had syphilis tests elsewhere. To remove any effect from tests repeated to confirm negative results, we excluded tests less than 4 weeks apart, and there was no difference in the IRR for HIV testing (1.04; 95% CI: 0.89 to 1.22).
The final questionnaire asked participants about their experience of HIV and syphilis testing, and responses were obtained from 260 (64.7%) of the 402 who completed 12 months of follow-up: 126/200 (63%) from the intervention arm and 134/202 (66.3%) from the control arm. There was no significant difference between study arms in proportions agreeing or disagreeing with the statements: “It was easy to get a test if I wanted to,” “I should have had more tests over the past year,” nor in questions about travel, cost, or waiting for results (see Table A, Supplemental Digital Content, https://links.lww.com/QAI/B144). However, fewer men in the intervention arm [57/125 (45.6%)] than in the control arm [83/132 (62.9%)] agreed with the statement “Waiting in clinics for a test takes too long” (P = 0.008).
Responses from 36 of the 49 men (73.5%) who used the pathology forms were compared with 85 responses from the 162 men in the intervention arm (52.5%) who did not use the forms. Nonusers were more likely to have lost or forgotten about the pathology forms [42/85 (49.4%)] compared with 6/36 (16.7%) users of the intervention (P = 0.001). Nonusers were also more likely to prefer to see a clinician for swabs and urine tests (52/85; 61.2%) than users (9/36; 25.0%; P = 0.001) (see Table B, Supplemental Digital Content, https://links.lww.com/QAI/B144). Users reported that the intervention was convenient (30/34; 88.2%), more expedient than attending a clinic (29/34; 85.3%), and easy to get to and cheaper than a general practitioner (24/34; 70.6%). Proportions were substantially higher among users than nonusers (P < 0.001 for all). More than 75% of users were not concerned that the pathology centers might not be gay friendly (see Table B, Supplemental Digital Content, https://links.lww.com/QAI/B144). The questionnaire asked men their opinions on having tests without consultations at the MSHC: 68/122 in the intervention arm (55.7%) agreed that they had forgotten about this, and 64/122 (52.5%) preferred to have swabs and urine tests.
There was no difference between study arms in the HIV test incidence when the analysis was restricted to men reporting difficulties with testing in the baseline questionnaire, those who said they would test more frequently if they could do so without waiting to see a clinician, or those who said they would test more frequently at pathology centers near work or home (see Table C, Supplemental Digital Content, https://links.lww.com/QAI/B144).
Questionnaire respondents reported a median of 6 condomless anal sex partners during the study (interquartile range, 2–13). The rate of testing in both study arms increased by 25% for each increase in quartile of partner number (IRR, 1.25; 95% CI: 1.16 to 1.36; P < 0.001).
Quality of Life
Mean utility scores at baseline were 0·734 (95% CI: 0.711 to 0.757) in the intervention arm and 0·737 (95% CI: 0.717 to 0.757) in the control arm. Of the 402 men followed to 12 months, 228 (56.7%) completed the SF6D questionnaire, and utility scores decreased by a mean of 0·01 (95% CI −0.032 to 0.012) in the intervention arm and −0.002 (95% CI −0.023 to 0.019) in the control arm, and the difference between arms was not significant (P = 0.61).
In this RCT of HIV and syphilis testing without the need for practitioner consultations, men in the intervention arm had tests at the same rate as controls who were required to attend a consultation for every test. The testing rate also did not increase in subgroups identifying problems with cost or access to testing, and even the men who predicted that they would test more frequently if these barriers were removed did not do so. Furthermore, there was no discernible adverse effect on individuals with no change in quality of life scores. However, a secondary outcome was a 24% reduction in clinical consultations by men in the intervention arm without any change in testing frequency. This suggests that costs associated with testing can be reduced and that there is no overuse of testing when access is not regulated by clinicians. This study found that making clinic-based HIV testing more available and convenient for clinic attendees will not in itself be sufficient to achieve increases in testing frequency, but it may help manage demand and reduce costs.
The main strength of this study is that it measures the effect of the intervention on the rate of HIV testing over time rather than simply measuring uptake of testing. However, it has limitations. First, our results are from MSM who test relatively frequently and attended a sexual health service, and so they may not apply to those who test less frequently or who have difficulty getting to clinical services. Second, the response rate to the 12-month questionnaire was 65% or less for some questions, and nonresponders may have answered differently. This did not affect measurement of the rate of testing. Third, we did not obtain STI testing data from outside clinics, preventing measurement of any reduction in bacterial STI screening associated with the intervention.
Our study complements other RCTs examining the effect of new forms of delivering HIV testing on the frequency of testing over time and comparing these to conventional clinic-based laboratory testing. One trial assessed rapid clinic-based HIV blood tests at the MSHC and showed no increase in testing frequency,11 but 2 other trials, providing oral home HIV tests led to a 1.5- to 2-fold increase in the rate of testing in the intervention group.12,13 All 3 studies recruited from similar populations to the present trial so the ability to test at home is the likely factor that increases testing frequency. The findings of our study contrast with those of observational studies suggesting that the removal of barriers to testing will increase test frequency.25–27 Rapid testing at home or at the point of care makes testing convenient and removes what can be an anxiety provoking wait for the result. Dried blood spot testing has been used to supplement HIV rapid tests in Africa and may be used to detect seroconverters who would otherwise be missed by oral tets.28,29
There are a number of possible reasons why our intervention did not increase the rate of HIV and syphilis testing. First, more than 60% of those who did not use the pathology forms said in the final questionnaire that they preferred to see a clinician for swab and urine tests for STI. Tests for chlamydia and gonorrhoea were not requested on the pathology forms, and this could also explain the low uptake of the intervention. Second, visiting a pathology collection centre may seem no more convenient than visiting a clinic, possibly explaining why home-based oral testing increased testing frequency while our intervention did not. Nevertheless, quality of life responses indicated no harm from the intervention, and the 24% reduction in the rate of clinical consultations indicates the potential for considerable cost savings. It should also be noted that 20% of participants had no HIV test during the study period. Because the rate of testing significantly increased by 25% for each quartile increase in the number of condomless anal sex partners, it is likely that level of HIV risk explains why some did not test.
This study demonstrates that access to HIV and syphilis testing without consultations does not further increase the frequency of testing in an MSM population that is already testing relatively frequently. It does however reduce demand on clinical services, and this may be a reason to offer it, given the importance of access to health care in achieving effective STI control. Because this study was conducted, HIV preexposure prophylaxis (PREP) has become available in many countries. This requires regular testing. However, not all those at risk of HIV can or will use PREP, and so measures to increase testing frequency are still needed. Future research could examine this intervention in less-tested populations, perhaps incorporating testing for chlamydia and gonorrhoea.
The authors thank the study participants and the staff at Melbourne Sexual Health Centre and Dr. Lyn Waring and staff at Melbourne Pathology, for their important contributions to this project.
1. Birger RB, Hallett TB, Sinha A, et al. Modeling the impact of interventions along the HIV continuum of care in Newark, New Jersey. Clin Infect Dis. 2014;58:274–284.
2. Charlebois ED, Das M, Porco TC, et al. The effect of expanded antiretroviral treatment strategies on the HIV epidemic among men who have sex with men
in San Francisco. Clin Infect Dis. 2011;52:1046–1049.
3. Wilson DP, Hoare A, Regan DG, et al. Importance of promoting HIV testing for preventing secondary transmissions: modelling the Australian HIV epidemic among men who have sex with men
. Sex Health. 2009;6:19–33.
4. Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505.
5. Group ISS, Lundgren JD, Babiker AG, et al. Initiation of antiretroviral therapy in early asymptomatic HIV infection. N Engl J Med. 2015;373:795–807.
6. Myers JE, Braunstein SL, Shepard CW, et al. Assessing the impact of a community-wide HIV testing scale-up initiative in a major urban epidemic. J Acquir Immune Defic Syndr. 2012;61:23–31.
7. Noble M, Jones AM, Bowles K, et al. HIV testing among internet-using MSM in the United States: systematic review. AIDS Behav. 2016;21:561–575.
8. Sharma M, Ying R, Tarr G, et al. Systematic review and meta-analysis of community and facility-based HIV testing to address linkage to care gaps in sub-Saharan Africa. Nature. 2015;528:S77–S85.
9. Gray RT, Prestage GP, Down I, et al. Increased HIV testing will modestly reduce HIV incidence among gay men in NSW and would be acceptable if HIV testing becomes convenient. PLoS One. 2013;8:e55449.
10. Fairley CK, Law M, Chen MY. Eradicating syphilis, hepatitis C and HIV in MSM through frequent testing strategies. Curr Opin Infect Dis. 2014;27:56–61.
11. Read TR, Hocking JS, Bradshaw CS, et al. Provision of rapid HIV tests within a health service and frequency of HIV testing among men who have sex with men
: randomised controlled trial. BMJ. 2013;347:f5086.
12. Jamil MS, Prestage G, Fairley CK, et al. Effect of availability of HIV self-testing on HIV testing frequency in gay and bisexual men at high risk of infection (FORTH): a waiting-list randomised controlled trial. Lancet HIV. 2017;4:e240–e250.
13. Katz D, Golden MR, Hughes J, et al. HIV self-testing increases HIV testing frequency among high-risk men who have sex with men
: a randomized controlled trial. 8th IAS Conference on HIV Pathogenesis, Treatment & Prevention, Vancouver, Canada, 19–22 July, 2015.
14. Masciotra S, McDougal JS, Feldman J, et al. Evaluation of an alternative HIV diagnostic algorithm using specimens from seroconversion panels and persons with established HIV infections. J Clin Virol. 2011;52(suppl 1):S17–S22.
15. Tan WS, Chow EP, Fairley CK, et al. Sensitivity of HIV rapid tests compared with fourth-generation enzyme immunoassays or HIV RNA tests. AIDS. 2016;30:1951–1960.
16. STIGMA SiGMAG. Australian STI & HIV testing guidelines 2014 for asymptomatic MSM 2014; Available at: https://stipu.nsw.gov.au/stigma/stihiv-testing-guidelines-for-msm/
. Accessed April 23, 2018.
17. Lin AC, Fairley CK, Dutt K, et al. Testing for HIV among men who have sex with men
needs a paradigm shift in Australia, given the minimal increase between 2003 and 2013 in Melbourne, Australia. Sex Health. 2015;12:373–382.
18. Wilkinson AL, El-Hayek C, Spelman T, et al. “Seek, test, treat” lessons from Australia: a study of HIV testing patterns from a cohort of men who have sex with men
. J Acquir Immune Defic Syndr. 2015;69:460–465.
19. Phillips AN, Cambiano V, Miners A, et al. Potential impact on HIV incidence of higher HIV testing rates and earlier antiretroviral therapy initiation in MSM. AIDS. 2015;29:1855–1862.
20. Prestage G, Brown G, Keen P. Barriers to HIV testing among Australian gay men. Sex Health. 2012;9:453–458.
21. Spielberg F, Branson BM, Goldbaum GM, et al. Overcoming barriers to HIV testing: preferences for new strategies among clients of a needle exchange, a sexually transmitted disease clinic, and sex venues for men who have sex with men
. J Acquir Immune Defic Syndr. 2003;32:318–327.
22. Dean Street Express sexual health service. Available at: http://express.dean.st/
. Accessed March 9, 2017.
23. Sydney Sexual Health Centre Xpress. Available at: https://www.sshc.org.au/OurServices
. Accessed March 9, 2017.
24. Brazier J, Roberts J, Deverill M. The estimation of a preference-based measure of health from the SF-36. J Health Econ. 2002;21:271–292.
25. Chen MY, Estcourt CS. Time to roll out rapid testing for HIV? Yes, but with appropriate safeguards. Sex Health. 2009;6:1–3.
26. Levy ME, Wilton L, Phillips G II, et al. Understanding structural barriers to accessing HIV testing and prevention services among black men who have sex with men
(BMSM) in the United States. AIDS Behav. 2014;18:972–996.
27. Wilkinson AL, Pedrana AE, El-Hayek C, et al. The impact of a social marketing campaign on HIV and sexually transmissible infection testing among men who have sex with men
in Australia. Sex Transm Dis. 2016;43:49–56.
28. Kania D, Bekale AM, Nagot N, et al. Combining rapid diagnostic tests and dried blood spot assays for point-of-care testing of human immunodeficiency virus, hepatitis B and hepatitis C infections in Burkina Faso, West Africa. Clin Microbiol Infect. 2013;19:E533–E541.
29. Luo W, Davis G, Li L, et al. Evaluation of dried blood spot protocols with the bio-Rad GS HIV combo Ag/Ab EIA and Geenius HIV 1/2 supplemental assay. J Clin Virol. 2017;91:84–89.