Introduction
The reporting of infected persons by name to public health departments (name-based reporting or case surveillance) is standard practice for the surveillance of many infectious diseases that pose a public health threat, including syphilis, gonorrhea, and tuberculosis [1]. AIDS has been a reportable condition in the United States since it was recognized in 1981. A national AIDS surveillance system has been considered a key element in efforts to monitor the spread of HIV infection in the United States, and has been used to allocate federal funds for HIV treatment and prevention to regions most affected by the epidemic [2].
Although AIDS is a reportable condition, making HIV infection a reportable condition has become a central controversy in HIV prevention and surveillance efforts in the United States [3-5]. There have been concerns, particularly among men who have sex with men (MSM), that reporting persons with HIV by name might compromise privacy and that the resulting registry could be used to discriminate against persons with HIV. Concern about HIV reporting has been greater than that about AIDS reporting, in part because persons reported with AIDS historically had a short life expectancy. In contrast, persons with HIV may now live for decades and face a much longer period during which confidentiality is a concern. Critics have charged that the fear of having one's name reported will strongly deter at-risk persons from being tested, resulting in the delayed diagnosis of HIV infection, increased HIV transmission, and lost opportunities for counselling about HIV risk reduction [6].
In the United States, the Centers for Disease Control and Prevention (CDC) has recommended HIV case surveillance to improve the monitoring of the epidemic, the targeting of prevention efforts, and the allocation of public funding for treatment and social services [2]. Other advocates of name-based HIV case reporting have argued that it can also strengthen partner notification and the linking of infected persons to health and social services. Recent advances in HIV treatments have diminished the capacity of AIDS case reporting alone to monitor the epidemic, because the rates of progression to AIDS have been lowered dramatically by current treatments [7,8]. Treatment advances have thus strengthened arguments for focusing disease surveillance efforts on HIV diagnosis [9]. As of January 2000, 33 out of 50 states in the United States have name-based reporting of HIV cases in adults. In western Europe, the use of new HIV treatments has also altered the role of AIDS surveillance in monitoring the HIV epidemic. As of 1999, some form of HIV reporting was taking place in 15 out of 18 western European countries [10].
Surveillance systems require follow-up methods to eliminate duplicate reports, complete missing information, and investigate unusual HIV variants or modes of transmission. The CDC recommends name-based HIV surveillance as the most efficient and effective method for accomplishing this follow-up in the United States. States that conduct name-based HIV surveillance do not forward the names of HIV-infected persons to CDC. Most countries in western Europe with HIV reporting systems do not use name-based reporting to eliminate duplicates, but use other personal identifiers such as a combination of birthdate and initials [10].
Despite important concerns about the possible deterrent effects of HIV reporting policies on the willingness to be tested, little is known about the knowledge of HIV reporting policies or their influence on decisions about testing. To address these issues, we conducted the HIV Testing Survey (HITS). We selected populations with significant risk of HIV exposure, for whom HIV testing would be strongly recommended. Our objectives were to assess the knowledge of state HIV reporting policies and the importance of concerns about name-based reporting as a reason for having delayed HIV testing or not having been tested.
Methods
Study design and study population
HITS was an anonymous cross-sectional survey conducted in the United States from December 1995 to November 1996. Four states included in this analysis had name-based HIV reporting, which had been in effect since 1991 or earlier (Arizona, Colorado, Missouri, and North Carolina). Name-based reporting is carried out only for confidential HIV tests, in which the test is recorded in the patient's medical record. All four states that used name-based HIV reporting also provided one or more anonymous HIV testing sites, from which name-based reporting of positive results is not possible. The results from one additional state that participated in the survey were excluded for three reasons: (i) it differed from other states because it lacked an anonymous test option and there were not other similar states with which to group it; (ii) only MSM and heterosexuals could be recruited because of a low prevalence of injection drug use; (iii) there was evidence that some interviews performed in the state were not valid. Of the four states without name-based HIV reporting, two began reporting by unique identifier (UI) in 1994 (Maryland and Texas); the remaining two used neither form of reporting (Oregon and New Mexico). UI reporting is an alternative to name-based reporting that aims to provide some of the surveillance benefits of HIV reporting by name, such as the elimination of duplicate reports, while reducing privacy concerns by avoiding use of the person's name. UI reporting systems use information such as birth date and part of the Social Security number to create a unique code that is reported instead of a name [11-13].
We recruited three populations at risk of recent exposure to HIV: MSM, heterosexuals who suspected they had a sexually transmitted disease (STD), and injection drug users (IDU). We aimed to recruit approximately 100 persons from each of these populations in each state (300 persons per state), using consistent recruitment methods in each state. We enrolled slightly more than 100 in many states because some eligibility criteria (such as recent HIV exposure risks) were difficult to establish before the interview. In most states, payments of US$10-20 were used as an incentive to participate. Participants had to be at least 18 years of age, had to reside in the state in which the interview was conducted, and had to provide informed consent.
MSM were recruited at gay bars. To be eligible, MSM had to have had sex with another man in the past 12 months. Recruitment was conducted in four or more bars in seven states; in the remaining state, recruitment was conducted in only one bar because of logistical problems in recruiting from additional sites. To diversify the sample, efforts were made to recruit from at least one bar primarily serving African Americans or Latinos, one bar in a non-urban area (population under 50 000), and one bar serving primarily young (< 25 years old) gay men. Efforts were also made to recruit on different days and at different hours in each bar. Clients were sampled by counting off every third to fifth person (depending on how many persons were present) in the bar, or entering the bar and inviting them to participate in the interview. Most recruitment was done by members of community-based organizations who were familiar with the setting and trained to work in the study; some health department personnel who were familiar with the setting also recruited participants.
Heterosexuals were recruited from STD clinics. These participants had to have been sexually active with a person of the opposite sex in the past 12 months, to consider themselves heterosexual, and be attending the clinic because they suspected they had a STD. To eliminate persons at the clinic for follow-up of an infection and to reduce the influence of having recently received counselling about HIV testing, potential participants were excluded if they had been to the STD clinic in the past 90 days. Recruitment and interviewing were done before clients received HIV counselling from a clinician. To recruit similar numbers of men and women, we stratified recruitment by sex.
IDU were recruited through street outreach. Participants had to have injected drugs in the past 12 months. To avoid biased self-report, this criterion was not stated to potential participants. Participants who reported during the interview that they had not injected drugs in the past 12 months received payment, but were excluded from analysis. To reach a range of participants, staff in most states selected three to five locations for recruitment: street corners and other sites where drugs are bought and sold, parks and other locations in which drugs are injected, shooting galleries, and needle exchange sites. Recruitment was performed by members of community-based organizations or health department personnel who were familiar with the recruitment setting.
Measures and statistical analysis
Participants underwent an anonymous, structured interview conducted by trained study staff in a private space. The interview obtained information about the participant's demographic background, HIV risk behaviors, and HIV testing history. As would be expected, there was some overlap of HIV risk categories. Of the MSM recruited at gay bars, 9% had also injected drugs in the past 12 months, as had 8% of the persons recruited at STD clinics. Of the persons recruited from sites associated with injection drug use, 4% were men who had been sexually active with men in the past 12 months. Because combining risk categories such as MSM and IDU did not appreciably alter analyses, we categorized participants' risk of HIV exposure according to the type of site at which they were recruited.
Participants who had been tested for HIV were asked whether they had delayed HIV testing for any of a list of reasons. If they had delayed testing they were then asked to choose the main reason for having delayed testing. Untested participants were given the same list of reasons, but the reasons were framed as reasons for not being tested. One item was designed to measure concern about reporting: 'because you were worried your name would be reported to the government if you tested positive'. This item was based on data from focus groups, which were used to determine how best to ask about concern about name-based HIV reporting. To be as sensitive as possible in detecting concerns about reporting as a reason for delaying or avoiding testing, we assumed that participants could be influenced by the reporting policy of their state, even if they were not certain of the policy.
We asked participants whether reporting by name or by UI code was carried out in their state (see Cited Here... 2 for the questions). We categorized respondents as correctly identifying the reporting policy in their state if they answered 'yes' to the question describing their state's reporting policy, and also answered 'no' or 'don't know' to questions describing reporting policies that were not used in their state. Participants who had already tested HIV positive were excluded from the analysis of knowledge of testing policies because they may have learned more about their state's reporting policies as a result of testing positive, and because they would be unlikely to test in the future.
We used Fisher's exact test or chi square with continuity correction to test for associations between categorical variables. Multiple logistic regression analyses were performed using Statistical Analysis Software (SAS Institute, Inc., Cary, NC, USA). The study was reviewed by institutional review boards at the University of California, San Francisco, the University of California, Berkeley, the Centers for Disease Control and Prevention, and the health departments of the states that required such review.
Results
From December 1995 to November 1996, we interviewed 2599 individuals. We excluded 195 of these interviews. Data about the recruitment site were missing for two persons and 193 did not meet site-specific eligibility criteria. In the latter category, we excluded 34 IDU because they had not injected drugs in the past 12 months, 78 men recruited at gay bars because they had not had male-male sex in the past 12 months, and 81 persons recruited at STD clinics because they did not consider themselves heterosexual or had not been sexually active with a person of the opposite sex in the past 12 months. The characteristics of the remaining 2404 participants are shown in Table 1.
Knowledge of HIV reporting policies
When asked whether their state used name-based HIV reporting, more than half the participants responded that they did not know (Fig. 1). In states with name-based reporting, more MSM (42%) knew that their state used name-based reporting than did heterosexuals (22%) or IDU (27%) (P < 0.001, comparing MSM with each other group). The same proportion of respondents in states with UI reporting as respondents in states with name-based reporting thought their state used name-based reporting (30%). These responses suggested that many participants may have confused UI reporting with name-based reporting; this mistake was more common among heterosexuals (37%, P < 0.001) and IDU (30%, P = 0.1) than among MSM (23%). In states with neither named or UI reporting, 16% of MSM, 16% of heterosexuals, and 24% of IDU thought their state used name-based reporting.
Overall, 15% of participants could correctly identify the HIV reporting policy in their state (Table 2). The state's reporting policy was correctly identified by 24% of MSM, 11% of heterosexuals, and 12% of IDU (P < 0.001, comparing MSM with each other group). Using a logistic regression model that adjusted for other participant characteristics, heterosexuals were less likely than MSM to know their state's reporting policy [odds ratio (OR) 0.3; 95% confidence interval (CI) 0.2-0.5]; IDU were also less likely than MSM to know their state's reporting policy (OR 0.4; 95% CI 0.3-0.6). Other factors in the model that were associated with knowing the HIV reporting policy in one's state were living in a state with name-based HIV reporting, being a woman, and being African American (Table 2).
Effect of concern about reporting on delaying or avoiding testing
We examined reasons participants had delayed HIV testing and reasons participants had avoided testing. Of 17 reasons participants could choose, worry about having one's name reported to the government if you tested HIV positive was the eighth most commonly selected reason for delaying testing, and the ninth most common reason for avoiding testing. The two most common reasons for avoiding testing were 'you didn't want to think about being HIV positive (48%)', and 'you were afraid of finding out you were HIV positive (48%)'.
When we compared participants who had never been HIV tested according to whether their state did or did not use name-based reporting, there were no statistically significant differences in the number who chose worry about reporting as a reason for avoiding testing (Table 3). To characterize more fully the participants who chose worry about name-based reporting as a reason for avoiding testing, we compared untested participants who chose this as a reason (n = 76, 16%) and untested participants who did not choose this as a reason (n = 405, 84%). Race/ethnicity, education, and income were not significantly associated with concern about reporting as a reason for avoiding testing. Of the 76 participants who chose concern about reporting as a reason for avoiding testing, 51 (67%) were unsure whether their state used name-based reporting, 18 (24%) thought their state used name-based reporting, and seven (9%) thought their state did not use name-based reporting.
Among participants who had been tested, 26% of IDU who lived in states with name-based reporting compared with 13% of those in states without name-based reporting chose worry about reporting as a reason for having delayed testing (P < 0.001;Table 4). Significantly more IDU in states with name-based reporting (6%) than IDU in states without name-based reporting (2%) chose this as the main reason they had delayed testing. Although 26% of MSM in states with name-based reporting compared with 19% of MSM in states without name-based reporting chose this as a reason for having delayed testing, the difference was not statistically significant. There was also no difference between states with and without name-based reporting in the proportion of MSM who selected worry about reporting as main reason for delaying testing. We found no increase among heterosexuals in the frequency of choosing concern about reporting as reason for delaying HIV testing in states with name-based reporting.
Discussion
Despite the controversy over name-based HIV reporting in policy debates, we are not aware of other studies that have assessed the knowledge of HIV reporting policies in the United States. The level of knowledge of reporting policies was low among the at-risk populations we studied in eight states. Although more MSM compared with the other two groups we studied knew the reporting policy in their state, more than three-quarters did not. This low level of knowledge suggests that reporting policies were not a substantial factor in HIV testing decisions for most individuals in our study. Other researchers have not found a decrease in HIV testing rates after the implementation of a name-based HIV reporting policy [14].
We found that in most of the groups we studied, the proportion of participants who stated that they had delayed or avoided testing was similar in states with and without name-based HIV reporting. The one exception was that more IDU in states with than in states without name-based reporting described delaying testing because of worry about reporting; we did not find that more IDU in states with than in states without name-based reporting had avoided testing because of this concern. Concern about named reporting was infrequently the main reason for delaying or avoiding testing in any group, with the highest proportion being 6% among IDU in states with name-based reporting. Overall, these results suggest that name-based reporting policies were not an important deterrent to testing in our study population.
We assumed that name-based reporting policies might influence testing decisions even if participants were uncertain whether there was reporting. When we restricted our analysis to the smaller number of persons who knew the reporting policy in their state, there were no differences in the frequency of concern about reporting as a reason for delaying or avoiding HIV testing related to the state policy.
In considering the possible deterrent effects of name-based HIV reporting, it is important to recognize that the states in our study that used name-based reporting also offered anonymous testing at specific sites. Anonymous testing may have served as a safety valve through which the persons most concerned about name-based reporting could be tested without being reported. Current guidelines from the US CDC recommend providing anonymous HIV testing sites to encourage HIV testing [15]. In addition, home HIV test collection kits now provide a way of getting anonymous HIV testing throughout the United States.
Limitations and summary
Although we strove for a diverse population at risk of HIV exposure, we used a convenience sample rather than a population-based sample. The limitation of conducting population-based studies by using random sampling is that large numbers are needed to include modest numbers of persons at high risk of HIV exposure. Although we interviewed large numbers of persons at high risk of HIV exposure, our results may not be generalizable to all high-risk populations. For example, MSM who do not go to gay bars might have responded differently; similar issues apply to the other populations we recruited. We also did not include states, such as New York and California, where the prevalence of HIV infection is highest. It is possible that at-risk populations in the states in our study do not adequately represent those in other states. Most participants had been HIV tested. The smaller numbers of untested participants limited our statistical power to detect modest increases in concern about reporting as a reason for not testing in states with name-based reporting.
Despite these limitations, our results provide clear evidence that knowledge of HIV reporting policies was low among at-risk populations in the states we studied. This suggests that despite the controversy about reporting policies, they are not a central factor in HIV testing decisions among most at-risk persons. MSM were more likely to know the actual reporting policy in their state. However, three-quarters of MSM still did not know the actual policy. IDU were more likely to report delaying testing because of concerns about reporting if they lived in states with name-based reporting. Otherwise, we did not find evidence that participants perceived concern about reporting as a reason for delaying or avoiding HIV testing more frequently if they were from states with name-based reporting than if they were not. These findings also suggest that name-based HIV reporting policies were not a major deterrent to HIV testing among the populations we studied. An anonymous HIV testing option and strong confidentiality protections are factors that may have contributed to the lack of apparent deterrent effects of name-based HIV reporting that we found.
Acknowledgements
The authors wish to acknowledge helpful reviews of the paper, which were provided by Drs Bernard Lo and Ted Hecht.
References
1. Centers for Disease Control and Prevention. Case definition for infectious conditions under public health surveillance. MMWR 1997, 46 (no. RR10): 1 -42.
2. Centers for Disease Control and Prevention. Guidelines for national human immunodeficiency virus case surveillance, including monitoring for human immunodeficiency virus infection and acquired immunodeficiency syndrome. MMWR 1999, 48 (no. RR-13): 1 -27.
3. Fordyce E, Sambula S, Stoneburner R. Mandatory reporting of HIV testing would deter blacks and Hispanics from being tested. JAMA 1989, 262: 349. 349.
4. Bayer R. Private acts, social consequences: AIDS and the politics of public health. New York: Free Press; 1989.
5. Gostin LO, Ward JW, Baker AC. National HIV case reporting for the United States. :A defining moment in the history of the epidemic. N Engl J Med 1997, 337: 1162 -1167.
6. Perlman D. States told to track data in HIV cases. San Francisco Chronicle, 10 December 1999. p. 1.
7. Hammer SM, Katzenstein DA, Hughes MD. et al. A trial comparing nucleoside monotherapy with combination therapy in HIV-infected adults with CD4 cell counts from 200 to 500 per cubic millimeter. :AIDS Clinical Trials Group Study 175 Study Team. N Engl J Med 1996, 335: 1081 -1090.
8. Hammer SM, Squires KE, Hughes MD. et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. :AIDS Clinical Trials Group 320 Study Team. N Engl J Med 1997, 337: 725 -733.
9. Steinbrook R. Battling HIV on many fronts. N Engl J Med 1997, 337: 779 -781.
10. Infuso A, Hamers F, Downs A, Alix J. HIV reporting in western Europe: national systems and first European Data. Euro Surveillance 2000, 5: 13 -17.
11. Smith E, Rix BA, Melbye M. Mandatory anonymous HIV surveillance in Denmark: the first results of a new system. Am J Public Health 1994, 84: 1929 -1932.
12. McDonald AM, Gertig DM, Crofts N, Kaldor JM. A national surveillance system for newly acquired HIV infection in Australia. :National HIV Surveillance Committee. Am J Public Health 1994, 84: 1923 -1928.
13. Centers for Disease Control and Prevention. Evaluation of HIV case surveillance of non-named unique identifiers - Maryland and Texas, 1994-1996. MMWR 1998, 46: 1254 -1271.
14. Nakashima AK, Horsley R, Frey RL, Sweeney PA, Weber JT, Fleming PL. Effect of HIV reporting by name on use of HIV testing in publicly funded counseling and testing programs. JAMA 1998, 280: 1421 -1426.
15. Centers for Disease Control and Prevention. HIV partner counseling and referral services. Atlanta, GA, USA: Centers for Disease Control and Prevention; 1998.
Appendix 1
The MESH Study Group includes:
John Ward MD, MPH, and Patricia Fleming PhD (Centers for Disease Control and Prevention, Atlanta, GA, USA); Denise K. Boyd MS, MPH and Vjollca Berisha MD, MPH (Arizona Public Health Department); Cornelis Rietmeijer MD, Kenneth Gershman MD, MPH and Alacey Berumen (Colorado Public Health Department); John Newman and Craig Thompson (Mississippi Public Health Department); Robert Hamm MD, MPH, Kristen Wendt MPH, and Linda Bell (Missouri Public Health Department); Michael Samuel DrPH, Jill Gatwood MS, and Patti Doherty (New Mexico Public Health Department); Delbert E. Williams PhD, Evelyn Foust MPH, and Judy Owen-O'Dowd (North Carolina Public Health Department); Steven Modessitt RN, MPH, Roger Wirt PhD, and David Fleming MD (Oregon Public Health Department); Ann S. Robbins PhD, Sharon A. King MA, and Douglas Hamaker (Texas Public Health Department); Rebecca Cohen DrPH, and Jan Markowitz PhD (Maryland Public Health Department).
Appendix 2
The following questions were used to test knowledge of state reporting policies.
Some places require reporting of positive HIV test results to the health department. There are different ways this is done.
1. In some states your name is reported to the health department if you test HIV positive, unless you got an anonymous test. Is reporting by name done in __________ (say state)?
2. In some states a unique identifier number is reported to the health department if you test HIV positive. The unique identifier number is made using part of your social security number, your birthday, gender, and race. Is unique identifier reporting done in __________ (say state)?
Cited Here...
© 2000 Lippincott Williams & Wilkins, Inc.