Abstract
Background:
In the United States, men who have sex with men (MSM) are known to disproportionately have HIV. The authors sought to describe the acceptability of providing at-home dried blood spot specimen collection kits for HIV testing among MSM.
Methods:
Between August 2010 and December 2010, the authors recruited Internet-using, HIV-negative or -unknown MSM to participate in a 12-month study of behavioral risks. Eligible participants were mailed an at-home HIV test.
Results:
Of the 896 men who were sent a test kit, 735 (82%) returned the kit. Returning a test kit was significantly associated with race (P = .002), highest level of education (P = .012), and annual income (P = .026). The adjusted odds of black, non-Hispanic men returning a test kit were about half of the odds of white, non-Hispanic men returning a test kit (adjusted odds ratios: 0.49; 95% confidence intervals: 0.31-0.78).
Conclusions:
Men who have sex with men are willing to provide biological specimens as part of an Internet-based HIV prevention study.
Introduction
The HIV epidemic continues to disproportionally affect men who have sex with men (MSM). The HIV incidence rates among MSM are increasing in North America and several other regions of the world. 1,2 Although MSM represent approximately 2.9% of the US population, they accounted for 63% of all new HIV infections and 82% of infections among newly infected men in 2010. 3,4 In 2010, black men had the highest number of estimated HIV infections, followed by white men and Hispanic men. 3 Additionally, there is evidence that HIV incidence is increasing among young MSM. Even though there was not a significant change in HIV incidence among MSM from 2006 to 2009, there was a 48% increase in new diagnoses among young, black/African American MSM during that time period. 5 Among all young Americans aged 13 to 29 years, only MSM experienced significant increases in incidence from 2006 to 2009. 5 Since 2010, the proportion of new HIV infections from male-to-male sexual contact continued to grow and was 70% in 2014. 3
Regular testing remains a central focus of managing the HIV epidemic among MSM. 6,7 Increasing awareness of HIV infection helps initiate newly diagnosed men on treatment and is the entry point into the HIV care continuum. 8 Research has shown that starting antiretroviral therapy early is associated with better clinical outcomes 9 and a reduction in HIV reservoir. 10 Beyond the individual health benefits of beginning treatment early, regular testing and knowledge of status are essential for the implementation of new interventions that aim to reduce HIV transmission. High-profile biomedical interventions, such as treatment as prevention 11 and preexposure prophylaxis, 12,13 can only be promoted and implemented with knowledge of HIV status. It is estimated that 13% of individuals with HIV are unaware of their status, 14 and this lack of awareness may be much higher among MSM, especially young MSM and MSM of color. 7
The US Centers for Disease Control and Prevention (CDC) recommends sexually active MSM test at least annually; some men with higher risks may benefit from testing as often as every 3 to 6 months. 15 However, data from the National HIV Behavioral Surveillance System and other online behavior surveys indicate that a large number of MSM do not adhere to that recommendation. 7,16,17 Although there is a variety of explanations for why MSM are not meeting the HIV testing recommendation, there has been some indication that men would be willing to use at-home HIV test kits to overcome some of the barriers that prevent the access of traditional services. 18 –20
The first at-home test kit licensed in the United States required an individual to use a dried blood spot collection kit, return it to a specimen collection site, and receive results by phone and/or online. This method protects the individual’s privacy and potential feelings of stigma, but the results are not available instantly. In 2012, after enrollment for this study was completed, the Food and Drug Administration (FDA) approved the OraQuick in-home HIV test (OraSure Technologies, Inc.), which allowed individuals to swab their gums and receive their test results in 20 minutes. 21
We sought to describe acceptability and feasibility of providing at-home dried blood spot specimen collection kits for HIV testing. The objective of our analysis was to describe factors associated with returning a home-based specimen collection kit for HIV testing among MSM participating in an online behavioral risk study. Developing a better understanding of factors associated with MSM returning at-home HIV test kits may help improve methods for utilizing at-home specimen collection procedures. At-home specimen collection can be used as biological end points in online prevention studies and potentially help understand how home testing products could be used to promote more frequent HIV testing among MSM.
Methods
Recruitment
We analyzed baseline data from MSM who participated in the “Checking In” study, a prospective study of Internet research methods. 22 From August 2010 to December 2010, men were recruited online through selective placement of banner advertisements shown on social networking sites such as MySpace, Facebook, Adam4Adam, and Black Gay Chat. On MySpace, exposures targeted self-identified gay and bisexual males. On Facebook, advertisements targeted men who indicated they were interested in men on their profile page. No specific targeting was done on Adam4Adam and Black Gay Chat.
Users who clicked on a banner advertisement were directed to an online eligibility screening questionnaire. In order to be eligible, men had to be at least 18 years of age; have had at least 1 male sex partner in the past 12 months; be non-Hispanic white, black, non-Hispanic, or Hispanic ethnicity; be willing to participate in an online research study for 12 months of follow-up; be willing to be randomized to Internet-based or short messaging system (SMS)-based follow-up surveys as part of the prospective study; and be willing to take and return an HIV test through specimen self-collection. The research study was reviewed and approved by Emory University Institutional Review Board (IRB00031326).
Eligible men were presented with an online consent module, and those who provided consent were asked to provide their e-mail address and a mobile phone number. To validate participants’ e-mail address, a link to the baseline survey was e-mailed to the provided e-mail address. Participants were required to click on the link within the e-mail to initiate the baseline survey. To validate their mobile phone number, participants received an SMS with a 3-digit code, which was required in order to proceed through the baseline survey.
Participants who validated their phone number and e-mail address were asked to complete a baseline survey that included questions on demographic characteristics, the use of the Internet to meet sex partners, recent sexual risk behaviors, and HIV/STD testing history. The survey also included questions about participants’ care-seeking behaviors and detailed information on recent sex partners (such as HIV status and partner concurrency). At the conclusion of the 30-minute baseline survey, men who did not report being HIV positive were asked to provide a valid mailing address for the at-home HIV test kit.
To maintain confidentiality, participants were given the opportunity to provide a work address or the address of a friend or family member. Kits were sent in plain packaging to protect participant privacy. Based on the previously identified factors shown to be associated with willingness to return a test kit, a picture of the kit was included in the registration process so that men were assured the shipping boxes were discreet and kits unidentifiable. 18 The at-home specimen collection for HIV testing used a commercial, dried blood spot collection system manufactured by Home Access Health Corporation (Hoffman Estates, Illinois). The kit required participants to perform a single finger prick and provide enough blood to fill a dime-sized circle on the specimen card. Participants then mailed the dried blood spot card back to the laboratory in a prepaid mailer. Study staff followed up with participants who had not returned their test kits after 14 days. The cards were tested using standard serological methods. Participants received a US$15 incentive for completing their baseline survey and US$20 once Home Access received their test kit. Participants were able to choose between 2 methods of payment: PayPal or an electronic gift card to Amazon.com.
Per Home Access Health protocol, participants were asked to call the FDA-approved medical call center to receive their HIV test results within 7 days of the ship date. Most participants with a negative HIV result were routed to an automated message system and participants with a positive or indeterminate HIV test result were routed to a trained HIV counselor. The counselor had the ability to provide referrals to over 19 000 organizations around the country that could provide resources to HIV-infected individuals. Participants with an insufficient or expired sample were able to request a new test kit. All HIV test results were transmitted securely from Home Access Health to Emory research staff. If a participant had not received their result within 14 days, research staff called and encouraged them to call Home Access for their result.
Participants who tested positive on their baseline HIV test were given the option to continue with follow-up surveys about their care-seeking behavior and continue to receive incentives. Participants who tested positive also spoke to Home Access counselors about linkage to HIV care. These records were kept by Home Access and shared in aggregate form with Emory University study staff.
Analysis
Data were analyzed using SAS software package (version 9.4; Cary, North Carolina). The primary analytic outcome was returning a test kit to the Home Access laboratory. We described demographic and behavioral characteristics of men who provided a mailing address and were sent a test kit. We included the following variables that have previously been shown to be associated with HIV testing: age, 23 –25 race/ethnicity, 25 –27 education, 26,27 and condomless anal sex. 27 Bivariate analyses were conducted to report crude associations of these characteristics with the outcome of interest. Statistical significance of crude associations was assessed using Wald χ2 tests with an α = 0.05.
Characteristics that were statistically significant were entered into a multivariate logistic regression model with an outcome of kit return. Any variables that had been previously shown to be associated with HIV testing were also entered into the model, even if they were not statistically significant in our data. All possible two-way interaction terms were assessed using a likelihood ratio test. 28 Collinearity was assessed by identifying condition indices higher than 30. 28 Adjusted odds ratios (aORs) and 95% confidence intervals were reported for all variables in the final model.
Results
From August 2010 to December 2010, we recruited 6104 Internet-using MSM who were eligible to take a baseline assessment. Of these participants, 3524 started the baseline survey, and 1498 completed the baseline survey in its entirety. The analysis data set included 896 men who met eligibility requirements, completed the baseline assessment, and provided a mailing address to which an at-home HIV test kit was sent.
Table 1 describes the demographic and behavioral characteristics of men who were sent test kits. The average age of men who were sent test kits was 30 years (standard deviation [SD] = 10). Among the 896 men sent test kits, 564 (63%) were white non-Hispanic, 157 (17%) were black non-Hispanic, and 175 (20%) were Hispanic. Most participants (80%, n = 718) had some education past high school. Most participants (92%, n = 828) reported only having sex with men in the past 12 months, and 61% (n = 544) reported having unprotected anal intercourse with a man in the past 12 months. Less than half (45%, n = 401) of the participants reported being tested for HIV in the past year. There were 735 men (82%) who returned a test kit to the Home Access laboratory.
Demographic and Behavioral Characteristics of HIV-Negative or -Unknown Men Who Have Sex with Men (MSM), Receiving a Home HIV Test Kit in a National Online HIV Prevention Study, United States, 2010.
Abbreviation: GED, general educational development.
There were 481 (85%) white, non-Hispanic, 115 (73%) black, non-Hispanic, and 139 (79%) Hispanic men who returned kits (Table 2). Returning a test kit was significantly associated with race (P = .002), highest level of education (P = .012), and annual income (P = .026), all of which were included in the final multivariate model. The final model also included 2 variables that were not significant in our data but have been shown to be associated with HIV testing: age 23 –25 and reporting unprotected anal intercourse with a man in the past 12 months. 27 No 2-way interaction term was significant. Controlling for the other variables in the model, the odds of black non-Hispanic men returning a test kit were about half of the odds of white, non-Hispanic men returning a test kit (aOR: 0.49; 95% CI: 0.31-0.78; Table 2). Compared to participants with only a high school or general educational development level education, the odds of returning a test kit are higher among participants with some college education (aOR: 1.67; 95% CI: 1.03-2.69) or a college degree (aOR: 1.76; 95% CI: 1.01-3.06).
Associations between Demographic and Behavioral Factors and Returning a Home HIV Test Kit among HIV-negative or -Unknown Men Who Have Sex with Men (MSM) in an Online HIV Prevention Study, United States, 2010.
Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; cOR, crude odds ratio; GED, general educational development.
Twenty-five men (3.4%) had a positive HIV test result through their at-home test kit (Figure 1). Of these 25 men with HIV-positive results, 11 were white, non-Hispanic (2% seropositivity), 9 were black, non-Hispanic (8% seropositivity), and 5 were Hispanic (4% seropositivity). Seven lived in rural areas (3% seropositivity) and 18 lived in urban areas (3.9% seropositivity). Although men who tested positive for HIV were offered the chance to continue in the 12-month study, only 15 (60%) took at least 1 bimonthly follow-up survey after their HIV diagnosis. Ten men who took at least 1 bimonthly follow-up survey continued through all 12 months of follow-up. Linkage to care information was available on the 15 men who took at least 1 bimonthly follow-up survey. Of the 15 men who took at least 1 bimonthly follow-up survey, 12 men indicated they had seen a physician or medical care provider, and 3 indicated they had not yet been to a physician or medical care provider at the time of their last follow-up survey. Of the remaining 10 men who received a positive HIV test result, 1 man did not get his result and the other 9 did not continue to their first bimonthly follow-up survey. Home Access Health Corporation counselors have limited information on these men who chose not to provide follow-up information as part of the study; however, 2 of these men indicated to Home Access counselors they would follow up with a physician or medical care provider. Home Access counselors suggested consultation with a physician to all participants who received an HIV-positive result. The 3 men who provided follow-up data to us and had not yet accessed care for HIV infection were asked to give reasons for not seeking care. They reported feeling fine, not wanting to tell anyone they have HIV, not having health insurance or enough money to visit a clinic, feeling depressed, not wanting to think about being HIV positive, having other responsibilities such as work, and not knowing where to seek care.
Participants receiving a positive HIV test result on their at-home test kit and care-seeking behavior in an online HIV prevention study, United States, 2010. Test kits were sent to all participants who provided an address and did not self-report as being HIV positive (N = 896).
Discussion
These results demonstrate feasibility and acceptability of collecting a biological specimen as part of an Internet-based HIV prevention study. The results indicate a large proportion of men (82%) participating in an online study of behavioral risks are willing to use and return an at-home collection kit for a dried blood spot specimen for HIV testing. There is also indication that at-home testing can effectively identify HIV-infected individuals. The use of at-home test kits in this study identified an HIV positivity of 3.4%, which is comparable to the 3.7% HIV positivity reported by CDC for HIV testing programs among MSM at non–healthcare facilities in 2013. 29 Returning an at-home test kit was not associated with number of male sex partners or having unprotected anal sex, which indicates that men of varying risk profiles are willing to participate in at-home specimen collection. There was also not a significant difference in test kit return between rural or urban areas, suggesting that at-home test kits can be used to reach MSM who live in rural areas.
Previous work suggested offering at-home test kits for free or with an incentive 18,30 may increase the use, and we also believe that much of our retention success is due to compensation and constant contact with participants. Future studies and efforts to implement at-home specimen collection should consider the importance of incentives and validating multiple methods of contact in order to keep participants active.
Individuals of black race, lower education, and lower annual salary were less likely to return a specimen collection home test kit, indicating an added challenge in reaching MSM who are often most impacted by the HIV epidemic: minority men and men with lower annual incomes. Recent research reported that MSM social media users are more likely to request an at-home test kit if they use the Internet to find sexual partners or have not recently been tested for HIV, 31 but neither of these were associated with returning a blood clot specimen in our study. However, we did not collect data on perceived risk, and it is possible that risk perception could influence participation in at-home testing activities.
We collected limited information on linkage to care for participants who received an HIV-positive result. All 25 men who tested positive for HIV were offered the chance to continue in the 12-month study, but only 15 (60%) took at least 1 bimonthly follow-up survey. Of the 15 men who took at least 1 bimonthly follow-up survey, 12 men indicated they had seen a physician or medical care provider and 3 indicated they had not yet been to a physician or medical care provider at the time of their last follow-up survey. Using the same calculation as Seth et al, 29 the minimum linkage to care percentage was 48% (12 of 25), and the maximum was 80% (12 of 15). These numbers are comparable to linkage to care percentages in non–healthcare settings (minimum = 48.9% and maximum = 87.6%) and healthcare settings (minimum = 50.2% and maximum = 85.9%) funded by CDC in 2013. 29 Although our percentages are similar to those seen in clinical settings, concern about linking individuals to care is an understandable barrier to using at-home test kits. Future research should focus on ensuring that all participants who receive a positive result are linked to care. Based on our experience, we recommend that participation in follow-up surveys be expected of those who get a new HIV diagnosis; although they may still choose to discontinue participation, it is critical to get as much follow-up data about linkage as possible.
The results of this study should be interpreted with a few limitations. Although we are confident that all test kits were sent to the requested destination, we have no way of knowing how many kits were actually received by the intended recipient. We categorized participants as living in rural or urban areas depending on the ZIP code of the mailing address that they provided. However, we don’t know if all participants actually lived at that mailing address or in the same ZIP code. When study staff followed up with participants who had not returned their test kit, most men cited reasons related to not receiving the kit or a change of mailing address. We also do not know if any participants provided a blood spot specimen that came from somebody else. Although our sample represents a diverse group of MSM, our Internet-using participants may not be a representative sample of MSM in the United States. Willingness to collect and return blood spot specimens may vary by geographic areas and cultural groups.
Finally, this study was conducted before the approval and release of the OraQuick at-home test kit. Although the OrqQuick test has a slightly lower sensitivity (98.03% versus 99.68%) than blood-based specimens, results are rapidly available to the individual. 32 Although our study had a high proportion of participants who were willing to return a blood specimen, more recent work indicates that MSM would prefer to use a rapid home self-testing kit. 20
Conclusion
These results support the ability to collect biological specimens in online research of MSM. Although there were differences by race, education, and income levels, the majority of men in all groups returned their specimen to be tested. These results are an initial step to future work in developing incentive and communication methods for online HIV prevention research that includes biological specimens. At-home testing can be a useful tool for increasing testing and early detection of HIV infection among rural and minority men in the United States.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: National Center for Minority Health and Health Disparities (1RC1MD004370) and Emory Center for AIDS Research (P30 AI050409).