Sullivan, Patrick S. DVM, PhD; Lansky, Amy PhD; Drake, Amy MPH; for the HITS-2000 Investigators
About 75% of HIV-infected people in the United States know their HIV serostatus. 1 One of the prevention priorities for the Centers for Disease Control and Prevention (CDC) is to increase the percentage of HIV-infected people who are aware of their infection to 95% by the year 2005. 2 The CDC’s Serostatus Approach to Fighting the Epidemic (SAFE) strategy 3 and the CDC’s Advancing HIV Prevention initiative 4 emphasize the need to promote testing and knowledge of serostatus. Understanding barriers to HIV testing and knowledge of HIV serostatus is an important part of evaluating progress toward these strategic goals and is an approach to understanding how current HIV counseling and testing efforts can be improved to promote knowledge of HIV status.
The impact of HIV counseling and testing may be greatly reduced if persons tested for HIV do not return to receive posttest counseling and HIV test results. Failure to return for HIV test results (FTR) has been previously evaluated, but most previous reports used data from a single testing venue or the CDC Counseling and Testing System and usually represented clients testing at a publicly funded HIV counseling and testing venue. 5–12 We analyzed data from a multistate interview project of persons at risk for HIV infection recruited in various venues to evaluate self-reported FTR and reasons for FTR.
The HIV Testing Survey (HITS) is an anonymous cross-sectional survey that has been conducted in the United States several times since 1996. 13 Our data come from the survey conducted in 2000; during this project year, data were collected in 8 project areas in 7 US states/cities: Florida, Illinois, Kansas, Nevada, New York City, New York State, Texas, and Washington. Participants from 3 groups of people at risk for HIV infection were recruited from 3 corresponding venues: men who have sex with men (MSM) from gay bars, injection drug users (IDUs) through street outreach, and high risk heterosexuals (HRHs) from sexually transmitted disease (STD) clinics. For each venue, specific sites were identified through formative research, which included review of written reports (eg, local HIV/AIDS surveillance reports), key informant interviews, and observations at some of the potential interview sites. Sites were selected by project staff based on the feasibility of conducting interviews there, and criteria were determined locally to obtain a diverse sample of each risk group.
Recruitment and face-to-face interviewing were performed by members of local community-based organizations or health department personnel familiar with the populations of interest. All interviewers were trained at a national interviewer training program, which included sessions on general interviewing techniques, role playing, and specific training on reducing bias in recruitment and reducing social desirability (obsequiousness) bias. In all venues, study staff used systematic sampling to select potential participants. The HITS was reviewed for human subject research protections at the CDC and in participating project areas.
The target sample size was at least 100 HIV-negative or untested persons from each of the 3 high-risk populations in each project area. Participants had to be at least 18 years of age, to have resided for at least 6 months in the state in which the interview was conducted, and to provide informed consent. There were additional behavioral criteria for inclusion in analysis: MSM were included in the analysis if they reported having sex with another man in the past 12 months, IDUs were included if they reported injecting drugs in the past 12 months, and HRHs were included if they reported being sexually active with a person of the opposite sex (but not a person of the same sex) in the past 12 months and were attending the clinic because they suspected they had an STD. In addition to these criteria, any participant who reported being HIV-positive was excluded from analysis.
The interview obtained information about respondents’ demographic background, self-reported HIV testing history, sexual behaviors, drug use, and access to prevention programs. The focus of this article is on the questions pertaining to HIV testing and obtaining test results.
FTR was based on respondents’ answers to the question, “Have you ever been tested for HIV but did not get your results?” Those who answered “yes” to this question were categorized as having failed to return for test results. The question measuring perceived risk was “How would you describe your chances of having HIV right now?” This was measured on a scale of 1 (lowest chance) to 10 (highest chance).
To assess reasons for failing to return for test results, we asked respondents who had failed to return for test results at least once to answer “yes” or “no” to 3 reasons that were read to them (Table 1) and then asked if there were any other reasons why they failed to return. Responses to these questions are not mutually exclusive; “other” reasons were grouped by frequency.
We wanted to assess the characteristics of tests for which participants failed to return. The questionnaire, however, did not ask participants about the specific test(s) for which they failed to return for results. Instead, we examined characteristics of participants’ most recent HIV test: whether it was confidential or anonymous and where it took place.
Data were analyzed separately for each of the 3 recruitment venues. We calculated odds ratios and 95% confidence intervals using EpiInfo (Centers for Disease Control and Prevention, 2002, available at www.cdc.gov/epiinfo, accessed on November 15, 2003) to determine whether the odds of failing to return varied by race/ethnicity, age, sex, education, or employment within each risk group. We also used Cochran-Mantel-Haenszel statistics to test for a trend in FTR across levels of age and levels of perceived risk for HIV infection for each risk group. For the HRH group, we conducted logistic regression (PROC LOGISTIC; SAS Institute, Cary, NC) to determine whether persons who failed to return for test results and those who did not differed significantly in terms of demographic characteristics (age, race, sex, education, and employment).
From June 2000 to February 2001 (HITS-2000), a total of 5823 people were approached to participate in the HITS. Of these, 1598 declined to participate (27% refusal rate), 874 did not meet residence or age requirements for participation, 40 did not complete the interview, and 91 were excluded because residence or age information was missing on the questionnaire. We excluded 120 participants from gay bars, 109 participants from STD clinics, and 154 participants recruited through street outreach because they did not meet behavioral criteria for inclusion in analysis. We excluded 215 persons who reported being HIV-positive and 15 whose HIV testing status could not be determined accurately from the data. We also excluded 363 who had never been tested for HIV and 3 with missing data on returning for results. After these exclusions, the sample for this analysis comprised 2241 respondents (782 MSM, 697 HRHs, and 762 IDUs).
Failure to Return for Test Results
The proportion of participants categorized as having failed to return for test results varied by risk for HIV infection: 75 (10%) MSM, 136 (20%) HRHs, and 202 (27%) IDUs reported failing to return for test results at least once. Among those who had failed to return for test results at least once, the median number of times they did not return was 1 for MSM (range: 1–9) and HRHs (range: 1–8) and 2 for IDUs (range: 1–28). Tables 2 through 4 present characteristics of those who failed to return for test results and those who did not. The only statistically significant crude odds ratios were found among HRHs (see Table 3): persons who had failed to return for test results were less likely to have completed education beyond high school (vs. high school/General Education Development certificate [GED]) and were less likely to be employed part time (vs. unemployed). The Mantel-Haenszel P values for trend in FTR by age were 0.3, 0.2, and 0.1 for MSM, HRHs, and IDUs, respectively.
The proportion of persons ever failing to return for test results was associated with perceived HIV risk (Table 5). More persons with high perceived risk than those with lower perceived risk failed to return for HIV test results.
For the HRH group, education, employment, age, perceived risk of having HIV, and study site of interview were entered into a logistic regression model. Persons who had failed to return for test results were significantly less likely to have had some college or higher education (vs. high school/GED; adjusted odds ratio = 0.5, range: 0.3–0.9) and were less likely to be working part time (vs. unemployed; adjusted odds ratio = 0.5, range: 0.3–0.9). Because demographic factors were not associated with FTR for MSM and IDUs, logistic regression was not performed for those groups.
Participants who failed to return for test results were asked their reasons for not returning (see Table 1). Among the reasons read to participants from which to choose, the most common reason among MSM for not returning for results was “too busy/forgot”; the most common reason among HRHs and IDUs for not returning for results was “thought testing place would contact me if test was positive.” About one quarter of each group reported not returning for test results because they were afraid to get their results.
Among those who failed to return for test results, the percentage of those who failed to return who were tested anonymously at the last test was 40% for MSM, 15% for HRHs, and 39% for IDUs (Table 6). In terms of where they were tested, the highest percentages of MSM and IDUs (32% and 27%, respectively) were most recently tested through outreach or at a counseling and testing site, and the highest percentage of HRHs with an identified site (19%) were most recently tested at an STD clinic. Few respondents in any group were tested at blood banks.
Our study documented proportions of persons at high risk for HIV infection who failed to return for HIV test results among 3 risk populations. FTR can decrease the prevention value of HIV counseling and testing programs in 2 ways. First, it represents a missed opportunity for posttest counseling. Second, some tests are anonymous; people who are HIV infected but fail to return for their test results cannot be reached and are unlikely to learn of their infection.
This study furthers previously reported public clinic–based studies about FTR by evaluating self-reported FTR among persons recruited at venues other than STD clinics or HIV counseling and testing system (CTS) sites. The proportions of respondents who had failed to return in our analysis (10%–27%) are similar to proportions who failed to return reported in other studies conducted in varying time periods, where proportions ranged from 10% to 71%. 5–12,14–16 Most previous studies, however, were conducted using data from HIV testing at publicly funded clinics only, including STD clinics. 5–12 Our study, on the other hand, recruited participants from venues, including gay bars and street injecting venues, identified through the formative research process. Less than 50% of all respondents reported having their most recent test at a publicly funded clinic.
Other investigators have reported factors associated with FTR. These factors have included black race 5,7,9,10,12,16 or Hispanic ethnicity, 5,9,12 younger age, 10 recent injection drug use, 10,14 risk for HIV infection other than male-to-male sex, 9,10,12 testing at an STD clinic, 5,9 not coming to the testing site specifically for an HIV test, 6,9 having risk by anonymous sex, 14 and confidential (vs. anonymous) HIV testing. 10 Further, FTR has been reported to be associated with lower levels of social support, 17 knowledge about AIDS or HIV testing, 11,17 and perceived risk for HIV infection. 17 In our analysis, lower educational attainment and unemployment were associated with FTR among HRHs; race/ethnicity and age, however, were not associated with FTR in any of the 3 risk populations we surveyed.
The reasons reported by our participants for FTR should prompt careful evaluation of how pretest counseling for HIV tests is conducted. For example, 21% to 31% of participants who had failed to return for test results reported that they were afraid to get the results. Pretest counseling may provide an opportunity to help clients articulate and manage their fears of learning their HIV infection status. Encouragingly, operational and structural barriers, such as long waiting time to receive results or transportation problems to return for results, were rarely cited by the participants as contributing to FTR.
Respondents frequently reported that they believed the testing site would contact them if test results were positive and that this belief was a reason for not returning for test results. Although the CDC Counseling and Testing guidelines recommend attempting to recontact clients with positive serologic results or high-risk clients with negative results, 18 attempting to contact clients is expensive, may not be successful, and risks compromising their privacy. Further, it is not possible to contact clients who were tested anonymously for HIV. In our survey, 808 respondents (36%) were tested anonymously at their most recent test (we did not collect information about whether the test for which clients failed to return for results was anonymous).
The US Food and Drug Administration (FDA) has recently approved licensure of 2 rapid tests for HIV infection. 19,20 In some settings, using rapid HIV tests may improve the proportion of clients receiving HIV test results. Some participants in our study reported that they failed to return because they were too busy or forgot, moved away from town before the results were available, or were put into jail or released from jail before results were returned. Not surprisingly, using rapid testing leads to high proportions of testers returning for HIV test results 21,22 and may be particularly useful in short-term detention settings.
This study has some limitations. First, our sampling was not population based, so our findings are not representative of those who are in the risk groups targeted in our study or all persons who can be found at the selected venues. This limitation was minimized by conducting extensive formative research for each venue and using systematic sampling within interview sites. Second, we may have recall bias (because we are asking about HIV testing events in the past) or obsequiousness bias (because we are asking about an outcome that may be perceived as socially desirable by the respondents). 23 Third, we did not collect information about the specific test for which results were not obtained but, instead, used the characteristics of the most recent HIV test as a proxy measure. We were thus unable to draw definite conclusions about associations between FTR and place of testing or between FTR and anonymous testing. Fourth, HIV testing conducted as part of screening for blood donation would be expected to increase proportions of respondents who failed to return for HIV test results, because people usually do not return for results from testing performed on donated blood. In our study group, however, the MSM and IDU respondents should be excluded from blood donation based on current FDA donor screening criteria, 24 and our data on the site of last HIV testing suggest that few respondents in any group were recently tested at blood banks. Finally, although the HITS has been conducted in several project years since 1996, we could not evaluate the trend in FTR over time, because each project year had a different set of study sites.
Based on our analysis, we suggest that there are clear roles for rapid testing as part of HIV counseling and testing and that, for some clients, pretest counseling may not have completely realized the opportunity to promote return for HIV test results. 25 Particularly, we recommend that rapid testing with the return of results as quickly as possible be offered and evaluated in short-term correctional settings. Pretest counseling should continue to emphasize the importance of returning to receive HIV test results. Clients who currently do not return because they think they will be contacted if they are positive may be encouraged to return if the benefits of further counseling in addition to return of HIV test results are promoted. Further, for clients who are afraid to learn their HIV test results, pretest counseling may offer opportunities to provide information about availability and benefits of treatments available for persons who test HIV-positive and to discuss strategies to manage fear about the return of results.
Erik Schwab, MA, provided developmental editing and copyediting of the manuscript.
1. Fleming PL, Byers RH, Sweeney PA, et al. HIV prevalence in the United States, 2000. Presented at the 9th Conference on Retroviruses and Opportunistic infections, Seattle, WA. February 2002.
2. Centers for Disease Control and Prevention. HIV Prevention Strategic Plan Through 2005. Atlanta, Centers for Disease Control and Prevention; 2001.
3. Janssen RS, Holtgrave DR, Valdiserri RO, et al. The serostatus approach to fighting the HIV epidemic: prevention strategies for infected individuals. Am J Public Health. 2001;91:1019–1024.
4. Centers for Disease Control and Prevention. Advancing HIV prevention: new strategies for a changing epidemic also emphasize the need to promote testing and knowledge of serostatus. MMWR Morb Mortal Wky Rep. 2003;52:329–332.
5. Centers for Disease Control and Prevention. HIV Counseling and Testing in Publicly Funded Sites Annual Report 1997 and 1998. Atlanta: Centers for Disease Control and Prevention; 2001.
6. Lazebnik R, Hermida T, Szubski R, et al. The proportion and characteristics of adolescents who return for anonymous HIV test results. Sex Transm Dis. 2001;28:401–404.
7. Wiley DJ, Frerichs RR, Ford WL, et al. Failure to learn human immunodeficiency virus test results in Los Angeles public sexually transmitted disease clinics. Sex Transm Dis. 1998;25:342–345.
8. Schluter WW, Judson FN, Baron AE, et al. Usefulness of human immunodeficiency virus post-test counseling by telephone for low-risk clients of an urban sexually transmitted diseases clinic. Sex Transm Dis. 1996;23:190–197.
9. Valdiserri RO, Moore M, Gerber AR, et al. A study of clients returning for counseling after HIV testing: implications for improving rates of return. Public Health Rep. 1993;108:12–18.
10. Slutsker L, Klockner R, Fleming D. Factors associated with failure to return for HIV post-test counseling. AIDS. 1992;6:1226–1227.
11. Catania JA, Kegeles SM, Coates TJ. Psychosocial predictors of people who fail to return for their HIV test results. AIDS. 1990;4:261–262.
12. Molitor F, Bell RA, Truax SR, et al. Predictors of failure to return for HIV test result and counseling by test site type. AIDS Educ Prev. 1999;11:1–13.
13. HIV testing among populations at risk for HIV infection—nine states, November 1995–December 1996. MMWR Morb Mortal Wkly Rep. 1998;47:1086–1091.
14. Solomon L, Moore J, Gleghorn A, et al. HIV testing behaviors in a population of inner-city women at high risk for HIV infection. J Acquir Immune Defic Syndr Hum Retrovirol. 1996;13:267–272.
15. Bindman AB, Osmond D, Hecht FM, et al. Multistate evaluation of anonymous HIV testing and access to medical care. JAMA. 1998;280:1416–1420.
16. Rotheram-Borus MJ, Gillis JR, Reid HM, et al. HIV testing, behaviors, and knowledge among adolescents at high risk. J Adolesc Health. 1997;20:216–225.
17. Stein JA, Nyamathi A. Gender differences in behavioural and psychosocial predictors of HIV testing and return for test results in a high-risk population. AIDS Care. 2000;12:343–356.
18. Centers for Disease Control and Prevention. Revised guidelines for HIV counseling testing and referral. MMWR Recomm Rep. 2001;50(RR19):1–57.
20. US Food and Drug Administration. Letter to OraSure Technologies, November 7, 2002.
21. Kassler WJ, Dillon BA, Haley C, et al. On-site, rapid HIV testing with same-day results and counseling. AIDS. 1997;11:1045–1051.
22. Keenan PA, Keenan JM. Rapid HIV testing in urban outreach: a strategy for improving posttest counseling rates. AIDS Educ Prev. 2001;13:541–550.
23. Sackett DL. Bias in analytic research. J Chronic Dis. 1979;32:51–63.
24. US Food and Drug Administration. Revised Recommendations for the Prevention of Human Immunodeficiency Virus (HIV) Transmission by Blood and Blood Products. Bethesda: Food and Drug Administration; 1992.
25. Centers for Disease Control and Prevention. Technical guidance on HIV counseling. MMWR Recomm Rep. 1993;42(RR02):11–17.
The HITS-2000 Investigators comprise the following individuals: Becky Grigg, Florida Department of Health, Tallahassee, Florida; Fran Eury, Illinois Department of Health, Springfield, Illinois; Farrell Webb, Kansas State University/Kansas Department of Health and Environment, Manhattan, Kansas; Janice Fung, Nevada Department of Human Resources, Las Vegas, Nevada; Lucia Torian, New York City Department of Health, New York, New York; Chris Nemeth, New York State Department of Health, Albany, New York; Marcia Becker, Texas Department of Health, Austin, Texas; Susan Barkan, Public Health–Seattle and King County, Seattle, Washington; and Frederick M. Hecht, University of California at San Francisco, San Francisco, California.
© 2004 Lippincott Williams & Wilkins, Inc.