Discussion
Participants described a number of factors that influenced their decision to accept testing, including curiosity, reassurance of negative status, convenience, and opportunity. Bringing the test to patients in the ED removed logistical and psychological barriers that are known to prevent people from seeking out testing in traditional venues. In addition, treating HIV like any other health problem helped patients feel comfortable about the HIV testing process. With regard to refusal of HIV testing in the ED, we found that reasons were having been tested recently and wanting to focus on the medical issue that brought the patient to the ED, consistent with other studies. Other reasons were "not wanting to know" and fear of confidentiality violations. The role of patients' perception of HIV risk in testing decisions was more complex. The perception of being at risk for HIV infection was certainly a motivation to accept testing, as at voluntary and counseling testing sites, however, we found that the perception of being at low risk for HIV infection was a reason for both refusal and acceptance, as it allowed individuals to feel comfortable accepting a test that they may not have sought elsewhere.
We discovered that many decliners provided logical reasons for refusing the test. Even so, decliners viewed HIV testing in general as important and interpreted the offer to test as an expression of concern on the part of the medical establishment. However, our data suggest that even patients who support HIV testing and are aware of its benefits may choose not to test because they prefer to live in uncertainty rather than face psycho-social consequences such as partner discord or discrimination based on HIV status. Indeed, the decliners in our study were more likely to describe instances of HIV stigma, even though there was no conscious acknowledgment of HIV stigma in the decision-making process.
There are several limitations to this study. This qualitative data is hypothesis-generating rather than definitive, and it may not be generalizable to other ED HIV testing programs. In addition, the interviews were done in busy EDs with patients who had pressing medical issues, thus participants may not have been as reflective as they would have been in other settings. Since the goal of this investigation was to look across rather than within programs, we did not assess how operational aspects of the three different models of ED HIV testing may have affected acceptance or refusal of testing. We were only able to recruit one decliner from the site that used clinician-initiated testing, as referrals of decliners at that site had to come directly from ED clinicians who had multiple competing priorities and may have been too busy to refer patients to the study. Finally, we did not systematically ascertain when patients last tested, since at the time of this study, these programs did not have policies on repeat testing. To our knowledge, there are no published guidelines on repeat HIV testing in the ED. In general, the 2006 CDC guidelines suggest at least annual testing of high-risk individuals with repeat testing of other individuals based on clinical judgment. Thus, it is important to acknowledge that repeat testing may not have been necessary for some of the individuals who cited recent testing as a reason for refusal. Indeed, the optimal interval for repeating an HIV test in the ED is an important area of future research.