Recent years have seen the advent of increasingly effective treatments for HIV through combination antiretroviral therapy (ART).¹ The evidence suggests that early treatment, accompanied by high adherence, improves long-term health² and may lower health care costs over the patient’s lifetime.^3,4 Clinical guidelines have been revised to recommend that therapy be started earlier in the course of disease to take advantage of the health benefits that earlier initiation confers.⁵ Recent studies have indicated that universal testing and appropriately prescribed ART for HIV-infected persons would not only benefit the individual but would substantially decrease HIV transmissions by decreasing HIV viral load in the community.^6,7 A recent clinical trial, HPTN 052, showed that early initiation of ART significantly reduces HIV transmission to a partner.⁸ For the full benefits of these “test and treat” strategies to be realized, HIV-infected persons must enter care and take ART as prescribed before they experience symptoms.⁹ However, information on acceptance of ART in developed countries is limited, particularly among HIV-infected persons who feel well and are not already in care.

Although many researchers and clinicians recognize the advantages of timely entry to care and acceptance of ART, evidence suggests that some HIV-infected persons may not share these beliefs or may place greater importance on perceived disadvantages. For example, studies have documented reluctance to take ART and fears about medication side effects.^10–17 However, because almost all studies have focused on HIV-infected persons already engaged in medical care, little is known about how medication-related factors influence prospective decisions to enter care.^18,19 An important exception is a recent study by Pollini and colleagues²⁰ who found that, compared to those in care, persons who had never initiated HIV care were more likely to mistrust HIV medications and health care providers.

To explore this issue, we reviewed structured and qualitative interview data from the Never in Care (NIC) Project: HIV-infected adults who were aware of their infection but had never received HIV medical care were asked about their perceptions of HIV medical care and barriers to entering care. Although other barriers to entering care were cited (e.g., financial, not wanting to think about or disclose HIV status, depression), a prominent theme that emerged from qualitative interview data was an equation of HIV care with medications and strong concerns about taking HIV medications, resulting in a reluctance to enter care. To better understand the mechanisms through which these concerns about medications serve as a barrier to care, we describe NIC respondents’ perceptions of the disadvantages and advantages of entering HIV medical care and accepting treatment and explore how beliefs about medications informed respondents’ ideas about care.

NIC was a multisite project conducted in Indiana, New Jersey, New York City, Philadelphia, and Washington State during 2008–2010. NIC methods have been described in detail elsewhere,^18,21 but in brief the mixed-method design consisted of a 30-min structured, face-to-face interview in all five project areas and additional 60-min qualitative interviews in Indiana, Philadelphia, and Washington State. NIC respondents were sampled from the Enhanced HIV/AIDS Reporting System (eHARS), a population-based HIV surveillance system. NIC-eligible persons had received, and acknowledged, a diagnosis of HIV infection at least 90 days before sample selection; were at least 18 years old; spoke English; resided in the project area; and had not received HIV medical care. Failure to receive HIV medical care was initially determined by lack of CD4 T-lymphocyte or HIV viral load laboratory test reports to eHARS and later confirmed with participants during recruitment and interview. Informed consent was obtained from all respondents. NIC design and instruments were approved by the Institutional Review Boards (IRBs) of the Centers for Disease Control and Prevention and participating state or local health departments.

A structured questionnaire was used in all five project areas. This 30-min interview collected information on demographics, barriers to entering HIV care, use of complementary and alternative medicines, and interactions with general medical providers.

In Indiana, Philadelphia, and Washington State, a maximum of 25 respondents per area were recruited for an additional, qualitative interview. Most (79%) interviews were recorded by using a digital audio recording device and transcribed verbatim. For 21% of interviews, the interviewer took detailed handwritten notes because respondents refused to be recorded or the equipment failed. An interviewer observation form was used to document supplementary information (e.g., pertinent body language or facial expressions) that was not captured in the audio recording.

Six study researchers developed a qualitative interview codebook containing structural and thematic codes using a standardized iterative process.²² Differences were resolved through discussion and when needed, modification of coding criteria. Two researchers then independently coded the qualitative interview transcripts by using NVivo version 8 (QSR International, Victoria, Australia), averaging 98% intercoder reliability. We used SAS, version 9.3 (SAS Institute, Inc., Cary, NC), to analyze the structured interview data.

In this sample of HIV-infected persons who had never entered HIV care, reluctance to enter care was influenced by reluctance to take medications (informed by distrust of medical care providers), and alternatives to conventional medical care were preferred, the combined effect of which may have been to deter respondents from entering HIV care. As increasing numbers of clinicians endorse early treatment with ART, this finding cautions us that a return to a “Hit early, hit hard” message may deter a significant proportion of HIV-infected persons from entering HIV care. Almost one third of our structured-interview respondents said that concerns about medications had kept them from accessing care during the past 3 months, and almost half of qualitative-interview respondents expressed concerns about taking medications.

For persons with these concerns, the other benefits of HIV care (e.g., monitoring immune system function through laboratory tests, access to ancillary services) should be emphasized. In addition, messages to these patients should challenge the idea that HIV care entails the passive acceptance of medicines and reframe entering care as a way of taking control and being proactive about one’s health through a partnership with a supportive care team. Most important, this communication should begin at the time a positive test result is delivered. For persons who are unwilling to immediately enter HIV care, this communication should continue in supportive nonmedical settings (e.g., HIV/AIDS service organizations, peer support groups), since it may take some time for newly diagnosed persons to accept their diagnoses.²³ After linkage to and engagement in care has been established, individual decisions about whether to begin ART can then be made with full knowledge of clinical status and accurate information about the risks and benefits of treatment.

Although NIC respondents had not entered HIV care, many perceived care as having several disadvantages and few advantages. Respondents voiced concerns about prescription practices in terms of receiving appropriate treatment and preserving future treatment options, mistrust of medications and medical providers, a sense that medications do more harm than good, and an ambivalence toward the life-preserving properties of medications in light of an assumed negative impact on quality of life. The respondents who had medication-related barriers to entering care were more likely to mention dislike and mistrust of health care workers, and they were more likely to use complementary and alternative medications and self-care to treat their HIV infection.

Among patients who have entered HIV care, fear of side effects, mistrust of medications and health care workers, concerns about the effects of medications on quality of life, and a preference for alternative medicine and self-care have been associated with lower acceptance of ART,^{11,13–17,24–26} although at least one study found no association between conspiracy beliefs and access to care.²⁷ However, research on the role of mistrust of medications and health care workers as barriers to entering care has been limited. Negative feelings about medications, a preference for self-care over medications, and distrust of health care providers have been found to be barriers to care among those who have dropped out of care.¹⁰ In studying a sample of HIV-infected women who had entered care, Raveis and colleagues²³ found that negative attitudes toward health care providers were associated with delayed entry to HIV care. A recent quantitative study by Pollini and colleagues²⁰ found that negative ideas about medications and health care providers are common among HIV-infected persons who have not entered care, although these factors were not associated with being in care once other factors were considered. In a longitudinal study of a small number of gay men who had not entered HIV care, researchers found that reluctance to receive ART, concerns about side effects, and mistrust of medical care providers affected engagement in care.¹² These factors, along with a preference for traditional healers rather than medical staff, were associated with suboptimal health care utilization among HIV-infected Zambians.²⁶ Our analysis supports the limited work in this area in that it also finds that concerns about medications and health care providers are common, and extends it by articulating the pathways through which fears about medications are related to decisions to not enter care.

Our results suggest that to improve linkage to care among persons who have concerns about HIV medications, time should be set aside during the delivery of the positive test result to describe the care process and explain that care also includes prevention measures and other services to ensure the well-being of patients. Staff trained in conducting psychosocial evaluations should ask questions about perceptions of HIV medications, use of complementary and alternative therapies, and trust in the health care system and health care workers. Information on current ART regimen options, the risks and benefits of ART, and ART side effects should be provided, but patient concerns should be taken seriously and discussed in a manner that encourages reflection.¹⁴ These discussions should include a full assessment of the patient’s values and sociocultural context as they relate to HIV care.¹⁷

Although information about care and medications may be most accurately communicated by a medical provider, persons with negative perceptions of HIV medications may never enter a care relationship and therefore never have the opportunity to receive this information. In addition, as the experiences of NIC respondents indicate, they may not trust medical providers and thus prefer alternative therapies to information from the medical establishment. Distrust of health care institutions and systems has been found to affect the utilization of health care^28–30 and may resonate more among people of color,^31–34 who accounted for most of the new diagnoses of HIV infection in the United States in 2009.³⁵ Respondents who did not describe themselves as white were somewhat more likely to have medication-related barriers to entering HIV care. Because NIC respondents’ ideas about HIV care were derived from their experiences in the community, it may be more effective to confront their negative perceptions through community-based interventions. Programs should consider peer-based and patient navigation strategies that use active referrals, which have proved effective in improving linkage to care and care utilization.^36–39 Ongoing support and outreach in nonmedical settings may also facilitate linkage to, and engagement in, care for persons with concerns about medications.

This study has certain limitations. Our sample was drawn from five health department jurisdictions, but we were not able to contact a large proportion of potentially eligible persons. In addition, half of all interviews (50% of structured and 52% of qualitative) were conducted in one jurisdiction, Indiana. Compared to structured participants, a higher percentage of qualitative interview participants reported they had not received HIV medical care during the past 3 months because of medication concerns (38% versus 27%). Therefore, our findings may not represent all HIV-infected persons in the United States who have not received HIV care. However, the findings illustrate the beliefs and perceptions of some persons who have received a diagnosis of HIV infection but who have never been in care—an understudied population. Another potential limitation is that because of logistical constraints, all qualitative interviews were conducted with English-speaking respondents. In addition, most respondents were male; however, the proportion of female NIC respondents (23%) is similar to the proportion of women in the United States who received a diagnosis of HIV infection in 2009 (24%).³⁵

Our results indicate that medication concerns may be barriers to HIV care for a substantial proportion of persons who do not enter care after receiving a diagnosis of HIV infection. We suggest that such concerns be elicited and that psychosocial barriers to entering care be assessed at the time of delivery of a positive test result and throughout the linkage-to-care process, followed by support and outreach in nonmedical settings to accommodate persons who are unwilling to enter care immediately. Building relationships of support and trust with nonmedical personnel may be the first step toward empowering HIV-infected persons to establish control of their medical care and make informed decisions about treatment.