Introduction

Emergency Departments (ED) are a potential venue to expand access to HIV counselling and testing (HCT) services to otherwise missed populations [1–3]. In the United States, ED-based HCT has had a significant impact on improving the rates of HIV diagnosis and subsequent linkage to care [4–6]. Unfortunately, the expansion of this strategy to low- and middle- income countries (LMIC) has been limited [7]. Implementing ED-based HCT will help improve access to HIV treatment and care for populations who do not regularly interact with the healthcare system through common and more established channels, such as the private health sector [2, 8]. South Africa has the highest prevalence of HIV in the world [9, 10], with much of its HCT resources predominantly directed to primary health care centers and antenatal clinics, neglecting other high-volume venues, such as EDs [2, 11]. Wherein, ED-based HCT is not routinely implemented due to lack of standardized training, competing clinical priorities, and in-sufficient resource allocation [3, 12–14]. However, a significant proportion of the population in South Africa relies on EDs as the sole point of contact with the healthcare system [15], as such ED-based HCT in South Africa has the potential to extend necessary HIV services to populations not captured through other testing venues. Furthermore, recent studies demonstrate a high prevalence of HIV in the ED (28%), of which almost a third of patients were unaware of their status [16].

Successful implementation of ED-based HIV testing interventions, rest on understanding the feasibility and organizational readiness for adoption [17]. Understanding patient and provider attitudes to ED-based testing can inform organizational readiness for change and, thus inform the successful implementation of HCT services [18, 19]. While there is anecdotal understanding of local attitudes towards ED-based testing programs in LMICs, there is no standardized, validated assessment tool currently available. Attitudes, defined as a complex mental state involving beliefs, feelings, values, and dispositions to act in certain ways [20], are essential to determine in order to suitably evaluate acceptability, appropriateness, and feasibility of a complex intervention. Several survey tools already exist to capture patient and provider attitudes to HIV testing programs, however, none of them have been validated for use outside of the US. In this study we propose to use existing survey instruments within our context, to develop a short-validated survey instrument assessing patient and provider attitudes towards ED-based HCT across contexts.

Methods

Results

A total of 104 patient and 132 provider surveys were completed (Table 1). Among patient responses, a majority were female (n = 54, 51.9%), between the ages of 30 and 39 years (n = 36, 34.6%), of African race (n = 86, 80.8%), had completed some or all of high school (n = 79, 75.9%), were unemployed (n = 65, 62.5%), and had previously been tested for HIV (n = 67, 64.4%). Among provider responses a majority were female (n = 93, 69.9%), between the ages of 20 and 29 years (n = 49, 36.8%), of African race (n = 107, 80.4%), and had been practicing medicine for 0 to 4 years (n = 48, 36.1%). Further, provider responses indicate 45 (33.8%) were physicians, 52 (39.1%) were registered nurses, 19 (14.3%) were nursing assistants/practitioners, and 16 (12.8%) were case managers.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of respondents by site.

https://doi.org/10.1371/journal.pone.0252372.t001

Patient attitudes to testing

Overall, 92.3% (n = 96) patients supported ED-based HCT. A majority of patients (80.8%, n = 84) agreed that if offered, they would assent to testing, and 72.1% (n = 75) patients reported wanting to learn more about HIV. However, 71.2% (n = 74) patients also stated that if ED-based testing were implemented, they would prefer an HIV counsellor to carry out testing and disclose results, instead of a physician or nurse. Patients’ knowledge on HIV infection, transmission and management was generally good, wherein the median score from the nine questions on HIV knowledge was 8 [CI: 7–9].

Factor analysis and scores.

Survey validation using exploratory factor analysis resulted in a 17-question instrument (from the original 36 survey questions) with seven factors, accounting for 74.6% of the total variance of the sample. Questions with a factor loading less than 0.4 and those cross-loading on more than one factor were omitted. Factors with only one question loading on to it were also omitted.

Table 2 shows the mean scores for the 17 survey questions used to evaluate patient attitudes. A mean score of 3 represents an overall neutral attitude, while a mean score more or less than 3 represents an overall positive or negative attitude, respectively. Patients demonstrated an overall mean score of 3.27, indicating a neutral attitude towards ED-based HCT. For example, question 5 ‘I want to learn more about HIV’ has a mean score of 3.48, demonstrating patient willingness to engage and learn more. Question 10 ‘The ED should offer HIV testing’ has a mean score of 4.43, showing a strongly positive attitude towards having ED-based HCT. On the other hand, question 7 ‘The ER and hospital can test you for HIV without asking for your consent’ has a mean score of 2.77, indicating the presence of concern or mistrust of clinical facilities in this population.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Mean patient and provider scores by question.

https://doi.org/10.1371/journal.pone.0252372.t002

The seven patient factors were identified as representing, ‘confidentiality’, ‘counseling & consent’, ‘openness to HIV knowledge’, ‘stigma around HIV testing’, ‘ED-based testing’, ‘social support’, and ‘stigma around HIV infection’ (attitudes questions included in each factor can be viewed in S3 Table). Mean scores for each factor in Table 3 signify the baseline attitudes of patients towards ED-based HCT. Overall, patients showed a positive attitude for each factor. Patient factor scores show that ‘confidentiality’ and ‘stigma around HIV testing’ have the least positive influence on patients’ attitudes towards ED-based HCT.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Mean patient and provider scores by factor (re-scaled to mean of 3.27).

https://doi.org/10.1371/journal.pone.0252372.t003

Discussion

To conquer the HIV epidemic in South Africa, innovative testing venues are actively being sought. The ED is one such venue. While ED-based HCT has been extensively evaluated in North America and Europe, it remains relatively novel in this region [2, 8, 19, 32]. Although research supports that expanding HCT services to EDs has the potential to close critical coverage gaps, successful implementation of ED-based HCT programs depends significantly on the willingness and commitment of ED patients and providers to adopt the intervention. In this study we develop and present a refined survey instrument to evaluate attitudes towards ED-based HCT, which can allow for more precise and targeted interventions, as well as a more efficient and standardized assessment of attitudes towards HCT across contexts. Overall, results suggest that ED patients and providers have a neutral attitude to ED-based HCT. Factor analysis demonstrated patients reporting favorably on ‘counseling and consent’ and ‘ED-based testing’, but raised notable concerns regarding ‘confidentiality’ ‘stigma around HIV testing’, and ‘stigma around HIV infection’. Similarly, providers reported favorably on the ‘benefits of HIV testing’, but raised concerns around ‘comfort with providing HIV testing’.

Patient acceptance of ED-based HCT, in South Africa and elsewhere, is encouragingly high [1, 4–6, 24, 33, 34]. Our study revealed that 92% of patients expressed support for ED-based HCT, and 80% were willing to undergo testing, if available–findings which are consistent with other contexts [3, 5, 24, 33, 35]. Despite this reassuringly high acceptance combined with innovative measures undertaken by the government to reach all vulnerable, high-risk populations with HCT, an estimated 10% of HIV positive South Africans remain unaware of their status and 15% of all new infections globally continue to occur here [10]. In our study, patients of African race displayed significantly more negative attitude towards ‘counselling and testing’. In this setting, refusal of HCT services by this demographic may be explained by structural violence exercised over decades, encompassing institutional racism, environments with elevated burden of disease, and stigmatizing social norms preventing underserved populations from getting adequate healthcare [36, 37]. Wherein, individuals are asked to govern their own access to health care rather than relying on the state to provide necessary services, likely discouraging them from wanting to engage with the healthcare system. This further emphasizes the need to provide quality services for counselling and obtaining informed consent with transparency, especially as 80.8% of patients presenting to the ED were African. We also found, patients who were unemployed displayed a significantly more negative attitude towards ‘ED-based testing’, potentially stemming from a lack of awareness, misconceptions around HIV infection and transmission, and a lack of trust in the government [38]. Considering 62.5% of patients presenting to the ED reported being unemployed, they are another important demographic to capture.

Patients were concerned about confidentiality related to HCT in the ED. In a busy clinical environment with high patient volumes, where beds are separated only by curtains, ensuring confidentiality when testing and disclosing results remains a challenge. Moreover, misconceptions around HIV testing, such as the assumption that those agreeing to test are likely HIV positive, augment resistance to ED-based HCT. Challenges to maintaining confidentiality is inherent in an ED-based program design, where conducting HCT bedside is as much a problem as stepping away to a dedicated space [23, 26, 28, 39]. It is then imperative for facilities to continually sensitize their staff so as not to add to or reinforce any stigma constraining people’s testing decisions. Embedding HCT within routine clinical pathways can ensure the service is normalized, reducing resistance [40, 41].

In comparison to other studies evaluating challenges to implementing ED-based HCT in low-resource and high-prevalence regions, patients in our study strongly supported the provision for opt-in consent and counselling prior to testing. This is in contrast to the growing body of literature proposing an opt-out strategy eliminating the need for obtaining separate consent, as well as a flexible counselling program tailored to those who choose it [35, 42]. Although CDC guidelines state that risk reduction and prevention counselling should not be required as part of an HIV screening program, this recommendation must be viewed in context. In well-functioning networks, it may be assumed once a patient is diagnosed for HIV, they will be linked to care, and will receive appropriate counselling along with follow up care. However, in a setting such as an ED in a resource-poor context, ensuring linkage to care is still a challenge. Additionally, the historically unequal distribution of power and resources between racial groups in South Africa, giving rise to the idea of healthcare services being a commodity–available only to those who can afford it–has resulted in a lack of trust in the government [43]. This is exemplified in our data as patients agree that ‘the hospital already tests every patient for HIV without telling them about it’ and ‘if I have been to the hospital, and no one told me I had AIDS or HIV, then I am negative’. Therefore, while streamlining HCT into ED processes is beneficial, it should be approached with caution to avoid patients believing they could be tested without their knowledge.

Prior to this study there were no routine HCT efforts in these EDs or in the hospitals at large, despite South Africa’s nationally recommended provider-initiated HCT strategy in all health care settings as a standard component of medical care [44]. While 96% of providers believed all ED patients should know their HIV status, 52% believed HCT takes up too much of their time and interferes with their primary role, and 76% reported a shortage of staff and other resources, that would make implementing such a program a significant challenge. Frequently reported barriers to HCT, as well as other preventive programs in the ED, including domestic violence and alcohol screening in other settings, parallel the concerns noted by providers in our study [12, 27]. Several studies have explored the feasibility of a parallel model, wherein dedicated staff are hired to conduct HCT, as a routine part of clinical care in the ED [45]. A parallel model can alleviate the burden of providers taking on an additional role allowing them to focus on patients’ clinical complaints, also addressing concerns around shortage of staff. In line with this approach, 71% of patients and 89% of providers reported preferring an HIV counsellor for HCT, over a doctor or nurse, likely increasing acceptability. Requiring sustainable funding to hire and support dedicated staff is a drawback to consider, however, if suitably integrated into routine ED operations, ED-based HCT has been found to be persuasively cost-effective [46].

Our study also revealed a surprising amount of provider stigma towards people living with HIV, with almost 70% believing that HIV positive patients indulge in risky activities, and 20% believing that HIV positive patients present a threat to their health and thus have the right to refuse care. In the 1980s there was worldwide debate as to the ethical and moral obligations of healthcare workers to care for HIV positive patients [47]. Science organizations such as the World Medical Association, and professional bodies within South Africa and internationally have clearly worded statements that healthcare workers may not ethically refuse to treat a patient whose condition is within the physician’s current realm of competence, solely because the patient is seropositive [48]. The challenge however is breaking the barrier of stigma. To combat stigma in health facilities, interventions must focus on the individual, environmental, and policy levels, this means increasing awareness of what stigma is among healthcare workers, providing supplies and equipment necessary to practice universal precautions and prevent occupational transmission of HIV, and developing clear institutional guidance to stop discriminatory behavior [49].

Another commonly cited attitudinal barrier rendering providers generally unwilling to offer HCT is discomfort, stemming from the fear of upsetting or offending patients in an already vulnerable setting and the fear of contagion [27, 50]. This is supported by our results, wherein providers expressed discomfort in approaching patients for HCT, in disclosing the results of a positive test, and expressed a right to refuse care to HIV positive patients. Discomfort may be rooted in cultural stigma against individuals positive for HIV, wherein despite gains in knowledge a subset of physicians, across the globe, perceive no ethical difficulty in denying medical care to patients infected with HIV [51]. Fueling this notion, former President of South Africa, Thabo Mbeki and his administration established policy that denied access to treatment for individuals with HIV/AIDS. Despite international protections and a shift in policy in the last few decades, many providers continue to refuse treatment [52]. Interestingly, a majority of providers participating in our survey were nurses or providers who have been in practice for less than 1 to 4 years. Across low-, middle, and high-income settings discriminatory attitudes and unethical behaviours by residents and nursing staff towards HIV positive patients is well documented, underpinned by the perception of patient ignorance towards risk behaviours and a failure to link to care [53, 54]. Similarly, a positive attitude to, and increased comfort with HCT is associated with increased patient acceptance of HCT, as well as an increased sense of familiarity with, and confidence in conducting HCT for providers. Employing a phased approach to implementation, supplemented with regular education and training sessions for providers may address existing misconceptions and knowledge gaps that can lead to changes in attitudes and greater buy-in for the program.

Lastly, while this brief attitudes survey provides some insight to the organizational readiness for an ED-based HIV testing intervention; it only considers attitudes towards HIV–such as stigma, in addition to HIV knowledge, and attitudes towards testing. This may provide some pre-implementation opportunities to optimize patient and provider readiness for ED-based testing programs. Beyond provider concerns that testing would take too much time and the ability to deliver testing in a confidential manner, larger implementation concerns are not addressed. Other structured frameworks that are able to evaluate the broader implementation environment i.e., administrative resources, post-testing care etc. are required. Several such frameworks exist including the Theoretical Domains Framework (TDF), Modes of Diffusion in Service Organization (MDISO) and the Innovation Implementation Framework to name a few [55–57].

There are several limitations of the study to consider. Due to convenience sampling participants assembled for this study may be biased on multiple accounts: (1) those who were willing to complete the survey may be more likely to be supportive of an HCT program than those who declined, (2) patients enrolled in the study may have been easier to approach, spoke the same language as the study staff, had milder injuries, and presented at a time when patient volume was lower, and (3) a majority of providers surveyed were nurses, as physicians were generally fewer in number, but also centrally involved in all cases and hence less available. Thus, the study may under-represent true attitudes held by patients and providers in this context. Our study is unique in that it utilized exploratory factor analysis to better understand key determinants of attitudes towards ED-based HCT. To improve generalizability, it will be useful to see whether data from different samples show similar factor structures.

Conclusion

Our study demonstrated ED patients and providers are generally supportive of ED-based HCT, though despite national policy in place since 2010, scale-up of HCT has remained inadequate. It is imperative patients and providers recognize HIV as an ongoing epidemic; it remains a leading cause of death and illness in South Africa. However, with a neutral attitude, high patient volumes, and disease severity of presenting patients, ED-based HCT is unlikely to result in successful implementation without behavioral interventions. A validated survey instrument can provide a standardized approach to understand and conquer existing barriers to HCT in an ED setting, across contexts.

Supporting information

Acknowledgments

The authors would like to acknowledge the staff at Frere Hospital, Nelson Mandela Academic Hospital, and the Mthatha Regional Hospital Emergency Departments, for making this research possible, the HIV Counseling and Testing team for their dedication and hard work during the study, and Nomazamo Mvandaba for her help in coordinating the study. We would also like to thank Dr. Thomas Quinn and the International HIV STD laboratory at NIH, for their support of the parent WISE study and ongoing collaboration at Walter Sisulu University.