AIDS Care, 2015 Vol. 27, No. 5, 581–586, http://dx.doi.org/10.1080/09540121.2014.987104
HIV testing in the critical care setting: views of patients, family members and health providers from urban South Africa C. Martina, M. Masotea*, A. Hatchera,b, V. Blacka, W.D.F. Ventera and F. Scorgiea,c a
Wits Reproductive Health and HIV Institute (WRHI), Faculty of Health Sciences, University of Witwatersrand, Johannesburg, South Africa; bDivision of HIV/AIDS, University of California San Francisco, San Francisco, CA, USA; cCentre for Health Policy, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa (Received 24 April 2014; accepted 10 November 2014) As antiretroviral treatment has led to decreased morbidity, HIV testing policy has increasingly shifted towards routine, provider-initiated approaches. Yet, few studies have examined the acceptability of provider-initiated HIV testing in the intensive, or critical care setting, where knowledge of HIV status is likely to impact on clinical management but explicit consent for testing is difficult to obtain. We conducted qualitative research in an urban hospital and clinic in Johannesburg. In-depth interviews were conducted among HIV testing clients (n = 20), recently discharged critical care patients (n = 13) and family members of critical care patients (n = 14). One focus group discussion was held with health care providers (n = 10). HIV testing in critical care was viewed as acceptable but challenging to implement. An overarching theme of ambivalence emerged from patients and families, who saw HIV testing as a pre-requisite to appropriate clinical care, but were concerned about the quality of its delivery. While providers were aware of the current “no testing without consent” policy, they expressed frustration in cases when testing was in the patient’s best interest but consent could not be obtained. Furthermore, providers found it stressful to weigh up patient confidentiality against medical necessity when assessing patients’ “best interests”. Without specific guidelines, they often developed pragmatic, ad hoc ways to resolve dilemmas around testing in critical care. Our findings suggest that HIV testing guidelines specific to the critical care setting may help providers do their jobs more ethically and transparently. Provider-initiated approaches are likely to be acceptable to patients and may improve clinical outcomes, but training and support in policy implementation and ethical decision-making are essential. Keywords: HIV testing; provider-initiated testing and counselling; critical care; informed consent; policy guidelines
Introduction Antiretroviral therapy (ART) has substantially improved the prognosis and longevity of HIV-infected patients (Braitstein et al., 2006; Ogoina et al., 2012). Improved access to ART in sub-Saharan Africa has steadily shifted the HIV testing debate from one of “exceptionalism” towards the benefits of routine testing in clinical settings (Baggaley et al., 2012). Since 2007, WHO has recommended “opt-out”, provider-initiated HIV testing and counselling (PITC) in health care settings, alongside scale-up of client-initiated HIV testing and counselling (HCT; WHO & UNAIDS, 2007). The PITC approach was adopted by the South African Government in 2008, to support access to care and treatment to the approximately 5.7 million South Africans then HIV-positive (UNAIDS, 2008). Currently, PITC is recommended in several clinical settings, primarily in antenatal and post-natal clinics, TB facilities and sexually transmitted infection clinics (Department of Health, 2010). Globally, routine HIV testing has not yet been meaningfully integrated into critical care settings. In the USA, up to 40% of patients with HIV admitted to critical care are *Corresponding author. Email: [email protected] © 2014 Taylor & Francis
unaware of their HIV status (Huang, Quartin, Jones, & Havlir, 2006). Patients with an AIDS-defining illness and the poorest prognosis could receive the greatest benefit from accelerated access to ART (Casalino et al., 2004) – but this would necessitate HIV testing. Routine HIV testing in critical or intensive care units (ICUs) would enable timely linkages to HIV care and treatment and generate gains in clinical management and qualityadjusted life years (Barton & Lawson, 2012). Obtaining informed consent from critically ill patients is complex, however, as patients are likely to be unconscious, cognitively impaired or simply too ill to make an active decision about testing. Clinical necessity may therefore require some forfeiting of patient confidentiality and autonomy. This complexity is only partially addressed by the latest South African HCT guidelines (Department of Health, 2010). For unconscious patients, the guidelines follow the WHO recommendation that consent be sought from a proxy; in this case, “the patient’s next-of kin, guardian or other caregiver”, and that “in the absence of such a person, health care providers should act according to the best interests of the patient concerned” (WHO &
582
C. Martin et al.
UNAIDS, 2013). But the South African guidelines ultimately sidestep consent challenges specific to the ICU setting; critical care patients are not listed among its “special populations” nor are considerations for deciding on the “best interests” of the patient provided. Such oversight is problematic in a country where HIV remains endemic and stigmatised (Merten et al., 2010). HIV prevalence in 2012 for adults (15–49 years) was estimated at 18.8%, with up to 62.2% reportedly unaware of their status (Shisana et al., 2014). While consent for HIV testing has been explored in local antenatal settings (Groves, Maman, Msomi, Makhanya, & Moodley, 2010) and psychiatric hospitals (Joska, Kaliski, & Benatar, 2008), to our knowledge, this has not been assessed in South African critical care settings. Given complex ethical and clinical dilemmas in this setting, the views of patients and providers are important to understand. We conducted exploratory qualitative research on the perceptions of HIV testing among patients recently discharged from critical care, their family members and health providers, and among self-referred clients at an HCT clinic. We examined opinions on the acceptability of testing patients in critical care for HIV without consent from either the patient or their proxy, particularly in situations where knowledge of HIV status likely impacts on clinical management. Methods Study design Participants were recruited from: a district hospital with an eight-bed ICU and a primary health care (PHC) clinic in Johannesburg, South Africa. Participants (18 years and older) were recruited using convenience sampling and categorised as: (1) 20 adults presenting for voluntary HCT at the PHC clinic; (2) 13 patients recently discharged from ICU; (3) 14 family members of patients recently discharged from ICU; and (4) 10 health providers working with critically ill patients in the hospital. Data were collected from individual in-depth interviews (IDIs) with the HCT clients, former ICU patients and family members (n = 47), and from one focus group discussion (FGD) held with health providers (one doctor, six professional nurses, two staff nurses and one enrolled auxiliary nurse). Data collection We developed, pilot-tested and revised an interview guide containing descriptions of hypothetical scenarios relating to unconsented HIV testing1 in critical care settings, followed by semi-structured questions to explore participants’ responses to these scenarios. Participants were interviewed in their language of choice (English, isiZulu or seSotho), while the FGD was held in English. They provided written, informed consent
for participation, with separate consent provided for audio recordings. The University of the Witwatersrand Human Research Ethics Committee approved the study (HREC M120289). Data management and analysis Audio files were transcribed in their original language and then translated into English where necessary. Transcripts were coded in QSR Nvivo 9, using a thematic approach to data analysis (Braun & Clarke, 2006). An initial coding framework was developed based on the research questions, data collection themes and current literature. Transcripts were then coded into “broad codes”. A preliminary research report was created by assembling excerpts related to each code, analysed inductively for “second-order” constructs (Miles & Huberman, 1994) and discussed among the research team to optimise validity (Miles & Huberman, 1984). Results The legitimacy of unconsented HIV testing Overall, participants expressed a wide range of views on HIV testing without consent in critical care. Some rejected it in this setting altogether. One family member felt that the patient would be justified in taking legal action, but only if the result was positive: If I am negative it’s my advantage and the doctor’s luck. He’ll survive. But if I am positive it’s another thing. I’ll be angry. Take it to the management, call the media and sue him, because I wouldn’t have given him authority. [Family member (FM); 27 years, man]
Similarly, some saw unconsented testing in this setting as a human rights violation. Picturing herself in this situation, a family member reflected: I would feel that my right as a human being has been violated, but I think over time I will understand why it was done. It wasn’t in a way an injustice to me; it was more to help me. (FM; 32 years, woman)
As this quote suggests, for some the rejection of unconsented HIV testing was tempered by recognition that while ethically problematic, it was often important for optimised treatment: They should know what they are treating me for. Let’s say I am admitted with stab wounds, and they treat me for that and my wounds take time to heal, I should be relaxed at the time and they should […] advise me to take a test. Then we would know your status and give you correct treatment. (ICU patient; 28 years, man)
Others felt it important to wait until the patient regained consciousness and could provide informed consent. Where
AIDS Care unconscious patients simply had to be tested (to facilitate treatment in ICU), one believed that test results should not be disclosed to them once conscious: “If he [health provider] tests you and you are positive, he must just keep quiet. Yes it will be for him to be able to do his job, [but he must] keep the results” (FM; 27 years, man). By contrast, other HCT clients and former ICU patients expressed trust in doctors’ decisions, depicting them as “knowledgeable” and only inclined to test for HIV where indicated. They further recognised that waiting for patients to regain consciousness before testing and administering appropriate treatment could be fatal: Critical care is life and death, it’s emergency unit, I wouldn’t mind. Honestly, I wouldn’t mind because being at critical care, its intensive care unit, it’s something else. It’s different from normal wards. (HCT client; 26 years, woman)
Health providers concurred that knowledge of their patients’ HIV status in this setting was vital for clinical management. One recalled an instance where a favourable outcome resulted from the testing (without consent) of a critically ill patient: The patient didn’t know, the husband didn’t know, nobody knew about the patient’s HIV status. So, when the results came back the patient was positive and the CD4 count was less than 200, and the patient was commenced on ARVs immediately. In this case for me it benefited the patient more than it benefited us. The patient’s condition changed from critically ill. The patient was awake and stronger due to the treatment. (Female professional nurse)
Providers also feared that a clinical mistake resulting from ignorance of the patient’s HIV status could lead to poor clinical outcomes and potential legal action. Consequently, they were caught between possibly “losing” patients and being sued by them: I mean if the patient is HIV negative you would not think of PCP [Pneumocystis Pneumonia], [but] the earlier you treat PCP, the better the outcome. So if you don’t test the patient, you stand a chance of losing that patient. However, if you start the patient on Bactrim and they don’t have PCP and they don’t have HIV they have the right to come back and sue you as a doctor. (Female doctor)
At the other end of the spectrum, we encountered the view that knowledge of HIV status in ICUs was so important, it should be routine, with or without consent. This conviction was expressed by almost all the family members interviewed. Some even advocated mandatory testing beyond critical care settings. Provider confusion and ambivalence Many health providers expressed confusion and ambivalence about the legal and moral acceptability of testing
583
patients in their care for HIV without consent. Some challenged the assumption that testing unconscious patients without their consent would benefit them, since many newly diagnosed patients experienced emotional distress upon learning of their HIV status. In contrast, some providers described an atmosphere of HIV exceptionalism and “over-protection”, which made the decision to test for HIV inherently fraught. Given the larger emotional and medical issues triggered by a new HIV diagnosis, they expressed difficulty determining what was in the “best interest” of patients – as recommended by current policy. Added to this difficulty was the need for a calculation around scarce resources. Several providers believed that only patients with real potential to recover should be admitted to ICU, an assessment that was influenced by knowledge of the patient’s HIV status: In this hospital the bed occupancy is 570, so out of this 570 we (critical care) are only having 8 beds. So we need to make sure that we use these beds very wisely. [Doctors] need to make sure that they take a patient that is going to be able to benefit from the system. (Female professional nurse)
To address these challenges, providers developed informal, pragmatic strategies to facilitate clinical management on a case-by-case basis, without violating ethical principles. One such strategy used the “stability” of the patient as a yardstick for determining if HIV testing should occur: “if the patient is stable, there is no indication … it [testing] can wait” (female professional nurse). Another approach was to “test whenever necessary, and tell the patient afterwards” – in other words, seeking informed consent after the fact, and even re-testing once the patient had regained consciousness. A female staff nurse explained: “We know that we are not allowed to test them. So you have to develop the strategy, once the patient gets better you ask for permission to test the patient”. Interestingly, patients themselves revealed knowledge of this pretence: Whether we want it or not, doctors do test. Once your sample is gone they do all the tests, even HIV … when you agree you want to be tested for HIV, that’s when they come and pretend now they are taking your blood for HIV test. Your test for HIV is already done. [Re-testing] it’s just for you to know the news. (FM; 31 years, man)
Many providers described existing HCT guidelines as unclear and ambiguous, thereby failing to guide decisionmaking. Specifically, they believed it was time for policy on HIV testing in critical care to permit unconsented testing, particularly when an HIV diagnosis would impact on clinical management. Health providers’ ambivalence over HIV testing in critical care also carried an implicit critique of health
584
C. Martin et al.
system inefficiencies. Some pointed out that it was not the role of critical care staff to provide HIV education and testing. Were the PHC system working well, they argued, patients admitted to tertiary level hospitals would already be aware of their HIV status. The role of counselling All participants emphasised the importance of patient support upon regaining consciousness and acknowledged the complexity of counselling at this point. As in other settings, counselling was seen as essential for helping patients cope with the implications of a new diagnosis: Counselling is important; people can kill themselves after the results. To know that they are HIV positive when they are not strong in mind is a problem. You must show them that HIV is just like any kind of sickness and they must take it easy. (HCT client; 37 years, man)
Counselling was also portrayed as necessary for educating patients about available treatment and care options. HCT clients felt that in the absence of counselling, they would not want to be tested at all. Others from this group downplayed the importance of counselling, however, because they doubted the quality of HIV counselling in public facilities. And yet others eschewed counselling altogether because they felt their knowledge of HIV was adequate, saying they “hear about this thing every day” (FM; 20 years, woman). Among health providers, the challenge was integrating counselling into an already full schedule of duties. They expressed frustration around the time counselling takes: I’ll tell you one thing about the public sector, right, we don’t have the time to sit and listen to them grieving. You tell them we did a test and tell them what HIV is and in 5 minutes you have to be done because I have sixty other patients to see. (Female doctor)
Even more challenging were situations where patients relied on health providers to disclose their status to family members on their behalf. While this assistance helped to facilitate ongoing care within the system, providers emphasised the additional emotional involvement and empathy it demanded of them. Finally, while discussing HIV test results with patients was clearly difficult, one provider described it as ultimately rewarding and suggested that it brought a sense of catharsis following discussion of a “hidden issue”.
without consent in this setting. Some 15 years after, Bhagwanjee, Muckart, Jeena, and Moodley (1997) identified the acceptance of a positive result as more problematic for patients than the absence of informed consent in itself; our study similarly found that those who favoured legal action against health facilities for unconsented testing only considered this necessary when HIV results were positive. Health providers in our study emphasised that accurate HIV diagnosis influences treatment decisions and supported provider-initiated approaches to testing in critical care. These views are consistent with global shifts in perceptions of HIV from a terminal illness to a manageable, chronic condition and follow international trends towards making HIV testing routine in the standard of care (Baggaley et al., 2012; Drayton, Keane, & Prentice, 2010). However, it is precisely because patients in critical care often lack the capacity to make decisions about their care that PITC per se is perhaps not a “viable model” for this setting (Dodd & Pryce, 2012). When a patient is unconscious, after all, the provider cannot initiate a conversation about the importance of HIV testing nor can the patient “opt out” if they wish – as the PITC model requires. Awareness of this dilemma possibly underlies the ambivalence and frustration expressed by health providers in our study, who are left to determine what is in the “best interests of the patient”. This includes deciding whether and when to test; how to manage proxy consent; and when (and if) to disclose test results to the patient or indeed to family members. Current HCT guidelines provide little guidance on these matters and are silent in particular on the disclosure of results to patients upon regaining consciousness. Health providers called for a new model of testing appropriate to critical care; their recommendation that unconsented testing be permitted in this context has also been raised elsewhere (Halpern et al., 2007). An important finding is the use of informal – but ethically questionable – strategies by health providers to manage consent challenges and assist clinical decisionmaking in ICUs. While this illustrates pragmatic agency in compensating for a lack of clear policy guidance, it also draws attention to the importance of ongoing training and managerial support in the implementation of PITC (Evans & Ndirangu, 2009), along with staff training in ethics. Transparent guidelines are urgently needed to set out viable procedures for HIV counselling and testing in critical care and to provide guidance on proxy consent and disclosure of results.
Discussion Most patients and family members saw HIV testing as an important pre-requisite for appropriate clinical care in ICUs, but were concerned about the ethics of testing
Study limitations These findings do not aim to be representative, but they do raise some key practicable insights for exploration in
AIDS Care future research. Given the study’s formative nature, sample sizes were small and convenience-based, requiring confirmation in a larger qualitative study. Social desirability bias may have influenced participants’ responses, especially if they viewed the researcher as belonging to the medical establishment.
Conclusion This formative research occurs alongside a growing trend towards routine PITC in clinical settings (WHO & UNAIDS, 2007). Many settings lag behind in establishing the policy guidance required to support PITC in practice, however, and our study illustrates implications for health providers, patients and family members of this lacuna. As elsewhere (Barton & Lawson, 2012; Dodd & Pryce, 2012), we believe that more decisive policy guidance, tailored to the conditions in critical care settings, would improve patient care for the critically ill in South Africa. Acknowledgements We thank the Gauteng Department of Health and authorities of participating health facilities, staff and participants. Thanks also to Mothepane Phatsoane and Petronella Goliath-Soyizwapi for assistance with conducting the FGD and Nokuphumelela Zwane for assistance with Zulu translations. Natasha Davies made useful contributions to the design and analysis, as did Jonathan Stadler and the Social Science Forum of Wits RHI. Andrew Black provided contact references for the Department of Health.
Disclosure statement No potential conflict of interest was reported by the authors.
Funding This project was supported by USAID PEPFAR [grant number AID-674-A-12-00021] – USAID HSS Programme.
Note 1. In this article, “unconsented HIV testing” and “HIV testing without consent” refer to situations where consent is obtained neither from the patient nor their proxy, such as next-of-kin, guardian or other caregiver (as advocated by current HCT guidelines).
References Baggaley, R., Hensen, B., Ajose, O., Grabbe, K. L., Wong, V. J., Schilsky, A., … Hargreaves, J. (2012). From caution to urgency: The evolution of HIV testing and counselling in Africa. Bulletin of the World Health Organization, 90, 652–658B. doi:10.2471/BLT.11.100818 Barton, S., & Lawson, A. (2012). Universal testing for HIV in intensive care units should be the norm: A call to action. Journal of the Intensive Care Society, 13(2), 98–99.
585
Bhagwanjee, S., Muckart, D. J. J., Jeena, P. M., & Moodley, P. (1997). Does HIV status influence the outcome of patients admitted to a surgical intensive care unit? A prospective double blind study. BMJ, 314, 1077–1081. doi:10.1136/ bmj.314.7087.1077a Braitstein, P., Brinkhof, M. W., Dabis, F., Schechter, M., Boulle, A., Miotti, P., … Egger, M. (2006). Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: Comparison between low-income and highincome countries. Lancet, 367, 817–824. doi:10.1016/ S0140-6736(06)68337-2 Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. doi:10.1191/1478088706qp063oa Casalino, E., Wolff, M., Ravaud, P., Choquet, C., Bruneel, F., & Regnier, B. (2004). Impact of HAART advent on admission patterns and survival in HIV-infected patients admitted to an intensive care unit. AIDS, 18, 1429–1433. doi:10.1097/01.aids.0000131301.55204.a7 Department of Health. (2010). National HIV counselling and testing policy guidelines. Pretoria: National Department of Health. Dodd, M., & Pryce, A. (2012). A national survey of HIV testing in intensive care: Moving forwards. Journal of the Intensive Care Society, 13(2), 136–139. Drayton, R., Keane, F., & Prentice, E. (2010). Patients’ attitudes towards increasing the offer of HIV testing in primary and secondary care. International Journal of STD and AIDS, 21, 563–566. doi:10.1258/ijsa.2010.010085 Evans, C., & Ndirangu, E. (2009). The nursing implications of routine provider-initiated HIV testing and counselling in subSaharan Africa: A critical review of new policy guidance from WHO/UNAIDS. International Journal of Nursing Studies, 46, 723–731. doi:10.1016/j.ijnurstu.2008.11.003 Groves, A. K., Maman, S., Msomi, S., Makhanya, N., & Moodley, D. (2010). The complexity of consent: Women’s experiences testing for HIV at an antenatal clinic in Durban, South Africa. AIDS Care, 22, 538–544. doi:10.1080/09540120903311508 Halpern, S. D., Metkus, T. S., Fuchs, B. D., Ward, N. S., Siegel, M. D., Luce, J. M., & Curtis, J. R. (2007). Nonconsented human immunodeficiency virus testing among critically ill patients: Intensivists’ practices and the influence of state laws. Archives of Internal Medicine, 167, 2323–2328. doi:10.1001/archinte.167.21.2323 Huang, L., Quartin, A., Jones, D., & Havlir, D. V. (2006). Intensive care of patients with HIV infection. New England Journal of Medicine, 355, 173–181. doi:10.1056/ NEJMra050836 Joska, J. A., Kaliski, S. Z., & Benatar, S. R. (2008). Patients with severe mental illness: A new approach to testing for HIV. South African Medical Journal, 98, 213–217. Merten, S., Kenter, E., McKenzie, O., Musheke, M., Ntalasha, H., & Martin-Hilber, A. (2010). Patient-reported barriers and drivers of adherence to antiretrovirals in sub-Saharan Africa: A meta-ethnography. Tropical Medicine and International Health, 15(Suppl. 1), 16–33. doi:10.1111/j.13653156.2010.02510.x Miles, M. B., & Huberman, A. M. (1984). Drawing valid meaning from qualitative data: Toward a shared craft. Educational
586
C. Martin et al.
Researcher, 13(5), 20–30. doi:10.3102/0013189X01300 5020 Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage. Ogoina, D., Obiako, R. O., Muktar, H. M., Adeiza, M., Babadoko, A., Hassan, A., … Tabi-Ajayi, E. (2012). Morbidity and mortality patterns of hospitalised adult HIV/AIDS patients in the era of highly active antiretroviral therapy: A 4-year retrospective review from Zaria, Northern Nigeria. AIDS Research and Treatment, 2012, 940580. doi:10.1155/2012/940580
Shisana, O., Rehle, T., Simbayi, L. C., Zuma, K., Jooste, S., Zungu, N., … Onoya, D. (2014). South African national HIV prevalence, incidence and behaviour survey, 2012. Cape Town: Human Sciences Research Council. UNAIDS. (2008). Report on the global HIV/AIDS epidemic 2008. Geneva: Author. WHO & UNAIDS. (2007). Guidance on provider-initiated HIV testing and counseling in health facilities. Geneva: World Health Organization. WHO & UNAIDS. (2013). Guiding principles on ethical issues in HIV surveillance. Geneva: World Health Organisation.
Copyright of AIDS Care is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
HIV testing in the critical care setting: views of patients, family members and health providers from urban South Africa C. Martina, M. Masotea*, A. Hatchera,b, V. Blacka, W.D.F. Ventera and F. Scorgiea,c a
Wits Reproductive Health and HIV Institute (WRHI), Faculty of Health Sciences, University of Witwatersrand, Johannesburg, South Africa; bDivision of HIV/AIDS, University of California San Francisco, San Francisco, CA, USA; cCentre for Health Policy, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa (Received 24 April 2014; accepted 10 November 2014) As antiretroviral treatment has led to decreased morbidity, HIV testing policy has increasingly shifted towards routine, provider-initiated approaches. Yet, few studies have examined the acceptability of provider-initiated HIV testing in the intensive, or critical care setting, where knowledge of HIV status is likely to impact on clinical management but explicit consent for testing is difficult to obtain. We conducted qualitative research in an urban hospital and clinic in Johannesburg. In-depth interviews were conducted among HIV testing clients (n = 20), recently discharged critical care patients (n = 13) and family members of critical care patients (n = 14). One focus group discussion was held with health care providers (n = 10). HIV testing in critical care was viewed as acceptable but challenging to implement. An overarching theme of ambivalence emerged from patients and families, who saw HIV testing as a pre-requisite to appropriate clinical care, but were concerned about the quality of its delivery. While providers were aware of the current “no testing without consent” policy, they expressed frustration in cases when testing was in the patient’s best interest but consent could not be obtained. Furthermore, providers found it stressful to weigh up patient confidentiality against medical necessity when assessing patients’ “best interests”. Without specific guidelines, they often developed pragmatic, ad hoc ways to resolve dilemmas around testing in critical care. Our findings suggest that HIV testing guidelines specific to the critical care setting may help providers do their jobs more ethically and transparently. Provider-initiated approaches are likely to be acceptable to patients and may improve clinical outcomes, but training and support in policy implementation and ethical decision-making are essential. Keywords: HIV testing; provider-initiated testing and counselling; critical care; informed consent; policy guidelines
Introduction Antiretroviral therapy (ART) has substantially improved the prognosis and longevity of HIV-infected patients (Braitstein et al., 2006; Ogoina et al., 2012). Improved access to ART in sub-Saharan Africa has steadily shifted the HIV testing debate from one of “exceptionalism” towards the benefits of routine testing in clinical settings (Baggaley et al., 2012). Since 2007, WHO has recommended “opt-out”, provider-initiated HIV testing and counselling (PITC) in health care settings, alongside scale-up of client-initiated HIV testing and counselling (HCT; WHO & UNAIDS, 2007). The PITC approach was adopted by the South African Government in 2008, to support access to care and treatment to the approximately 5.7 million South Africans then HIV-positive (UNAIDS, 2008). Currently, PITC is recommended in several clinical settings, primarily in antenatal and post-natal clinics, TB facilities and sexually transmitted infection clinics (Department of Health, 2010). Globally, routine HIV testing has not yet been meaningfully integrated into critical care settings. In the USA, up to 40% of patients with HIV admitted to critical care are *Corresponding author. Email: [email protected] © 2014 Taylor & Francis
unaware of their HIV status (Huang, Quartin, Jones, & Havlir, 2006). Patients with an AIDS-defining illness and the poorest prognosis could receive the greatest benefit from accelerated access to ART (Casalino et al., 2004) – but this would necessitate HIV testing. Routine HIV testing in critical or intensive care units (ICUs) would enable timely linkages to HIV care and treatment and generate gains in clinical management and qualityadjusted life years (Barton & Lawson, 2012). Obtaining informed consent from critically ill patients is complex, however, as patients are likely to be unconscious, cognitively impaired or simply too ill to make an active decision about testing. Clinical necessity may therefore require some forfeiting of patient confidentiality and autonomy. This complexity is only partially addressed by the latest South African HCT guidelines (Department of Health, 2010). For unconscious patients, the guidelines follow the WHO recommendation that consent be sought from a proxy; in this case, “the patient’s next-of kin, guardian or other caregiver”, and that “in the absence of such a person, health care providers should act according to the best interests of the patient concerned” (WHO &
582
C. Martin et al.
UNAIDS, 2013). But the South African guidelines ultimately sidestep consent challenges specific to the ICU setting; critical care patients are not listed among its “special populations” nor are considerations for deciding on the “best interests” of the patient provided. Such oversight is problematic in a country where HIV remains endemic and stigmatised (Merten et al., 2010). HIV prevalence in 2012 for adults (15–49 years) was estimated at 18.8%, with up to 62.2% reportedly unaware of their status (Shisana et al., 2014). While consent for HIV testing has been explored in local antenatal settings (Groves, Maman, Msomi, Makhanya, & Moodley, 2010) and psychiatric hospitals (Joska, Kaliski, & Benatar, 2008), to our knowledge, this has not been assessed in South African critical care settings. Given complex ethical and clinical dilemmas in this setting, the views of patients and providers are important to understand. We conducted exploratory qualitative research on the perceptions of HIV testing among patients recently discharged from critical care, their family members and health providers, and among self-referred clients at an HCT clinic. We examined opinions on the acceptability of testing patients in critical care for HIV without consent from either the patient or their proxy, particularly in situations where knowledge of HIV status likely impacts on clinical management. Methods Study design Participants were recruited from: a district hospital with an eight-bed ICU and a primary health care (PHC) clinic in Johannesburg, South Africa. Participants (18 years and older) were recruited using convenience sampling and categorised as: (1) 20 adults presenting for voluntary HCT at the PHC clinic; (2) 13 patients recently discharged from ICU; (3) 14 family members of patients recently discharged from ICU; and (4) 10 health providers working with critically ill patients in the hospital. Data were collected from individual in-depth interviews (IDIs) with the HCT clients, former ICU patients and family members (n = 47), and from one focus group discussion (FGD) held with health providers (one doctor, six professional nurses, two staff nurses and one enrolled auxiliary nurse). Data collection We developed, pilot-tested and revised an interview guide containing descriptions of hypothetical scenarios relating to unconsented HIV testing1 in critical care settings, followed by semi-structured questions to explore participants’ responses to these scenarios. Participants were interviewed in their language of choice (English, isiZulu or seSotho), while the FGD was held in English. They provided written, informed consent
for participation, with separate consent provided for audio recordings. The University of the Witwatersrand Human Research Ethics Committee approved the study (HREC M120289). Data management and analysis Audio files were transcribed in their original language and then translated into English where necessary. Transcripts were coded in QSR Nvivo 9, using a thematic approach to data analysis (Braun & Clarke, 2006). An initial coding framework was developed based on the research questions, data collection themes and current literature. Transcripts were then coded into “broad codes”. A preliminary research report was created by assembling excerpts related to each code, analysed inductively for “second-order” constructs (Miles & Huberman, 1994) and discussed among the research team to optimise validity (Miles & Huberman, 1984). Results The legitimacy of unconsented HIV testing Overall, participants expressed a wide range of views on HIV testing without consent in critical care. Some rejected it in this setting altogether. One family member felt that the patient would be justified in taking legal action, but only if the result was positive: If I am negative it’s my advantage and the doctor’s luck. He’ll survive. But if I am positive it’s another thing. I’ll be angry. Take it to the management, call the media and sue him, because I wouldn’t have given him authority. [Family member (FM); 27 years, man]
Similarly, some saw unconsented testing in this setting as a human rights violation. Picturing herself in this situation, a family member reflected: I would feel that my right as a human being has been violated, but I think over time I will understand why it was done. It wasn’t in a way an injustice to me; it was more to help me. (FM; 32 years, woman)
As this quote suggests, for some the rejection of unconsented HIV testing was tempered by recognition that while ethically problematic, it was often important for optimised treatment: They should know what they are treating me for. Let’s say I am admitted with stab wounds, and they treat me for that and my wounds take time to heal, I should be relaxed at the time and they should […] advise me to take a test. Then we would know your status and give you correct treatment. (ICU patient; 28 years, man)
Others felt it important to wait until the patient regained consciousness and could provide informed consent. Where
AIDS Care unconscious patients simply had to be tested (to facilitate treatment in ICU), one believed that test results should not be disclosed to them once conscious: “If he [health provider] tests you and you are positive, he must just keep quiet. Yes it will be for him to be able to do his job, [but he must] keep the results” (FM; 27 years, man). By contrast, other HCT clients and former ICU patients expressed trust in doctors’ decisions, depicting them as “knowledgeable” and only inclined to test for HIV where indicated. They further recognised that waiting for patients to regain consciousness before testing and administering appropriate treatment could be fatal: Critical care is life and death, it’s emergency unit, I wouldn’t mind. Honestly, I wouldn’t mind because being at critical care, its intensive care unit, it’s something else. It’s different from normal wards. (HCT client; 26 years, woman)
Health providers concurred that knowledge of their patients’ HIV status in this setting was vital for clinical management. One recalled an instance where a favourable outcome resulted from the testing (without consent) of a critically ill patient: The patient didn’t know, the husband didn’t know, nobody knew about the patient’s HIV status. So, when the results came back the patient was positive and the CD4 count was less than 200, and the patient was commenced on ARVs immediately. In this case for me it benefited the patient more than it benefited us. The patient’s condition changed from critically ill. The patient was awake and stronger due to the treatment. (Female professional nurse)
Providers also feared that a clinical mistake resulting from ignorance of the patient’s HIV status could lead to poor clinical outcomes and potential legal action. Consequently, they were caught between possibly “losing” patients and being sued by them: I mean if the patient is HIV negative you would not think of PCP [Pneumocystis Pneumonia], [but] the earlier you treat PCP, the better the outcome. So if you don’t test the patient, you stand a chance of losing that patient. However, if you start the patient on Bactrim and they don’t have PCP and they don’t have HIV they have the right to come back and sue you as a doctor. (Female doctor)
At the other end of the spectrum, we encountered the view that knowledge of HIV status in ICUs was so important, it should be routine, with or without consent. This conviction was expressed by almost all the family members interviewed. Some even advocated mandatory testing beyond critical care settings. Provider confusion and ambivalence Many health providers expressed confusion and ambivalence about the legal and moral acceptability of testing
583
patients in their care for HIV without consent. Some challenged the assumption that testing unconscious patients without their consent would benefit them, since many newly diagnosed patients experienced emotional distress upon learning of their HIV status. In contrast, some providers described an atmosphere of HIV exceptionalism and “over-protection”, which made the decision to test for HIV inherently fraught. Given the larger emotional and medical issues triggered by a new HIV diagnosis, they expressed difficulty determining what was in the “best interest” of patients – as recommended by current policy. Added to this difficulty was the need for a calculation around scarce resources. Several providers believed that only patients with real potential to recover should be admitted to ICU, an assessment that was influenced by knowledge of the patient’s HIV status: In this hospital the bed occupancy is 570, so out of this 570 we (critical care) are only having 8 beds. So we need to make sure that we use these beds very wisely. [Doctors] need to make sure that they take a patient that is going to be able to benefit from the system. (Female professional nurse)
To address these challenges, providers developed informal, pragmatic strategies to facilitate clinical management on a case-by-case basis, without violating ethical principles. One such strategy used the “stability” of the patient as a yardstick for determining if HIV testing should occur: “if the patient is stable, there is no indication … it [testing] can wait” (female professional nurse). Another approach was to “test whenever necessary, and tell the patient afterwards” – in other words, seeking informed consent after the fact, and even re-testing once the patient had regained consciousness. A female staff nurse explained: “We know that we are not allowed to test them. So you have to develop the strategy, once the patient gets better you ask for permission to test the patient”. Interestingly, patients themselves revealed knowledge of this pretence: Whether we want it or not, doctors do test. Once your sample is gone they do all the tests, even HIV … when you agree you want to be tested for HIV, that’s when they come and pretend now they are taking your blood for HIV test. Your test for HIV is already done. [Re-testing] it’s just for you to know the news. (FM; 31 years, man)
Many providers described existing HCT guidelines as unclear and ambiguous, thereby failing to guide decisionmaking. Specifically, they believed it was time for policy on HIV testing in critical care to permit unconsented testing, particularly when an HIV diagnosis would impact on clinical management. Health providers’ ambivalence over HIV testing in critical care also carried an implicit critique of health
584
C. Martin et al.
system inefficiencies. Some pointed out that it was not the role of critical care staff to provide HIV education and testing. Were the PHC system working well, they argued, patients admitted to tertiary level hospitals would already be aware of their HIV status. The role of counselling All participants emphasised the importance of patient support upon regaining consciousness and acknowledged the complexity of counselling at this point. As in other settings, counselling was seen as essential for helping patients cope with the implications of a new diagnosis: Counselling is important; people can kill themselves after the results. To know that they are HIV positive when they are not strong in mind is a problem. You must show them that HIV is just like any kind of sickness and they must take it easy. (HCT client; 37 years, man)
Counselling was also portrayed as necessary for educating patients about available treatment and care options. HCT clients felt that in the absence of counselling, they would not want to be tested at all. Others from this group downplayed the importance of counselling, however, because they doubted the quality of HIV counselling in public facilities. And yet others eschewed counselling altogether because they felt their knowledge of HIV was adequate, saying they “hear about this thing every day” (FM; 20 years, woman). Among health providers, the challenge was integrating counselling into an already full schedule of duties. They expressed frustration around the time counselling takes: I’ll tell you one thing about the public sector, right, we don’t have the time to sit and listen to them grieving. You tell them we did a test and tell them what HIV is and in 5 minutes you have to be done because I have sixty other patients to see. (Female doctor)
Even more challenging were situations where patients relied on health providers to disclose their status to family members on their behalf. While this assistance helped to facilitate ongoing care within the system, providers emphasised the additional emotional involvement and empathy it demanded of them. Finally, while discussing HIV test results with patients was clearly difficult, one provider described it as ultimately rewarding and suggested that it brought a sense of catharsis following discussion of a “hidden issue”.
without consent in this setting. Some 15 years after, Bhagwanjee, Muckart, Jeena, and Moodley (1997) identified the acceptance of a positive result as more problematic for patients than the absence of informed consent in itself; our study similarly found that those who favoured legal action against health facilities for unconsented testing only considered this necessary when HIV results were positive. Health providers in our study emphasised that accurate HIV diagnosis influences treatment decisions and supported provider-initiated approaches to testing in critical care. These views are consistent with global shifts in perceptions of HIV from a terminal illness to a manageable, chronic condition and follow international trends towards making HIV testing routine in the standard of care (Baggaley et al., 2012; Drayton, Keane, & Prentice, 2010). However, it is precisely because patients in critical care often lack the capacity to make decisions about their care that PITC per se is perhaps not a “viable model” for this setting (Dodd & Pryce, 2012). When a patient is unconscious, after all, the provider cannot initiate a conversation about the importance of HIV testing nor can the patient “opt out” if they wish – as the PITC model requires. Awareness of this dilemma possibly underlies the ambivalence and frustration expressed by health providers in our study, who are left to determine what is in the “best interests of the patient”. This includes deciding whether and when to test; how to manage proxy consent; and when (and if) to disclose test results to the patient or indeed to family members. Current HCT guidelines provide little guidance on these matters and are silent in particular on the disclosure of results to patients upon regaining consciousness. Health providers called for a new model of testing appropriate to critical care; their recommendation that unconsented testing be permitted in this context has also been raised elsewhere (Halpern et al., 2007). An important finding is the use of informal – but ethically questionable – strategies by health providers to manage consent challenges and assist clinical decisionmaking in ICUs. While this illustrates pragmatic agency in compensating for a lack of clear policy guidance, it also draws attention to the importance of ongoing training and managerial support in the implementation of PITC (Evans & Ndirangu, 2009), along with staff training in ethics. Transparent guidelines are urgently needed to set out viable procedures for HIV counselling and testing in critical care and to provide guidance on proxy consent and disclosure of results.
Discussion Most patients and family members saw HIV testing as an important pre-requisite for appropriate clinical care in ICUs, but were concerned about the ethics of testing
Study limitations These findings do not aim to be representative, but they do raise some key practicable insights for exploration in
AIDS Care future research. Given the study’s formative nature, sample sizes were small and convenience-based, requiring confirmation in a larger qualitative study. Social desirability bias may have influenced participants’ responses, especially if they viewed the researcher as belonging to the medical establishment.
Conclusion This formative research occurs alongside a growing trend towards routine PITC in clinical settings (WHO & UNAIDS, 2007). Many settings lag behind in establishing the policy guidance required to support PITC in practice, however, and our study illustrates implications for health providers, patients and family members of this lacuna. As elsewhere (Barton & Lawson, 2012; Dodd & Pryce, 2012), we believe that more decisive policy guidance, tailored to the conditions in critical care settings, would improve patient care for the critically ill in South Africa. Acknowledgements We thank the Gauteng Department of Health and authorities of participating health facilities, staff and participants. Thanks also to Mothepane Phatsoane and Petronella Goliath-Soyizwapi for assistance with conducting the FGD and Nokuphumelela Zwane for assistance with Zulu translations. Natasha Davies made useful contributions to the design and analysis, as did Jonathan Stadler and the Social Science Forum of Wits RHI. Andrew Black provided contact references for the Department of Health.
Disclosure statement No potential conflict of interest was reported by the authors.
Funding This project was supported by USAID PEPFAR [grant number AID-674-A-12-00021] – USAID HSS Programme.
Note 1. In this article, “unconsented HIV testing” and “HIV testing without consent” refer to situations where consent is obtained neither from the patient nor their proxy, such as next-of-kin, guardian or other caregiver (as advocated by current HCT guidelines).
References Baggaley, R., Hensen, B., Ajose, O., Grabbe, K. L., Wong, V. J., Schilsky, A., … Hargreaves, J. (2012). From caution to urgency: The evolution of HIV testing and counselling in Africa. Bulletin of the World Health Organization, 90, 652–658B. doi:10.2471/BLT.11.100818 Barton, S., & Lawson, A. (2012). Universal testing for HIV in intensive care units should be the norm: A call to action. Journal of the Intensive Care Society, 13(2), 98–99.
585
Bhagwanjee, S., Muckart, D. J. J., Jeena, P. M., & Moodley, P. (1997). Does HIV status influence the outcome of patients admitted to a surgical intensive care unit? A prospective double blind study. BMJ, 314, 1077–1081. doi:10.1136/ bmj.314.7087.1077a Braitstein, P., Brinkhof, M. W., Dabis, F., Schechter, M., Boulle, A., Miotti, P., … Egger, M. (2006). Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: Comparison between low-income and highincome countries. Lancet, 367, 817–824. doi:10.1016/ S0140-6736(06)68337-2 Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. doi:10.1191/1478088706qp063oa Casalino, E., Wolff, M., Ravaud, P., Choquet, C., Bruneel, F., & Regnier, B. (2004). Impact of HAART advent on admission patterns and survival in HIV-infected patients admitted to an intensive care unit. AIDS, 18, 1429–1433. doi:10.1097/01.aids.0000131301.55204.a7 Department of Health. (2010). National HIV counselling and testing policy guidelines. Pretoria: National Department of Health. Dodd, M., & Pryce, A. (2012). A national survey of HIV testing in intensive care: Moving forwards. Journal of the Intensive Care Society, 13(2), 136–139. Drayton, R., Keane, F., & Prentice, E. (2010). Patients’ attitudes towards increasing the offer of HIV testing in primary and secondary care. International Journal of STD and AIDS, 21, 563–566. doi:10.1258/ijsa.2010.010085 Evans, C., & Ndirangu, E. (2009). The nursing implications of routine provider-initiated HIV testing and counselling in subSaharan Africa: A critical review of new policy guidance from WHO/UNAIDS. International Journal of Nursing Studies, 46, 723–731. doi:10.1016/j.ijnurstu.2008.11.003 Groves, A. K., Maman, S., Msomi, S., Makhanya, N., & Moodley, D. (2010). The complexity of consent: Women’s experiences testing for HIV at an antenatal clinic in Durban, South Africa. AIDS Care, 22, 538–544. doi:10.1080/09540120903311508 Halpern, S. D., Metkus, T. S., Fuchs, B. D., Ward, N. S., Siegel, M. D., Luce, J. M., & Curtis, J. R. (2007). Nonconsented human immunodeficiency virus testing among critically ill patients: Intensivists’ practices and the influence of state laws. Archives of Internal Medicine, 167, 2323–2328. doi:10.1001/archinte.167.21.2323 Huang, L., Quartin, A., Jones, D., & Havlir, D. V. (2006). Intensive care of patients with HIV infection. New England Journal of Medicine, 355, 173–181. doi:10.1056/ NEJMra050836 Joska, J. A., Kaliski, S. Z., & Benatar, S. R. (2008). Patients with severe mental illness: A new approach to testing for HIV. South African Medical Journal, 98, 213–217. Merten, S., Kenter, E., McKenzie, O., Musheke, M., Ntalasha, H., & Martin-Hilber, A. (2010). Patient-reported barriers and drivers of adherence to antiretrovirals in sub-Saharan Africa: A meta-ethnography. Tropical Medicine and International Health, 15(Suppl. 1), 16–33. doi:10.1111/j.13653156.2010.02510.x Miles, M. B., & Huberman, A. M. (1984). Drawing valid meaning from qualitative data: Toward a shared craft. Educational
586
C. Martin et al.
Researcher, 13(5), 20–30. doi:10.3102/0013189X01300 5020 Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage. Ogoina, D., Obiako, R. O., Muktar, H. M., Adeiza, M., Babadoko, A., Hassan, A., … Tabi-Ajayi, E. (2012). Morbidity and mortality patterns of hospitalised adult HIV/AIDS patients in the era of highly active antiretroviral therapy: A 4-year retrospective review from Zaria, Northern Nigeria. AIDS Research and Treatment, 2012, 940580. doi:10.1155/2012/940580
Shisana, O., Rehle, T., Simbayi, L. C., Zuma, K., Jooste, S., Zungu, N., … Onoya, D. (2014). South African national HIV prevalence, incidence and behaviour survey, 2012. Cape Town: Human Sciences Research Council. UNAIDS. (2008). Report on the global HIV/AIDS epidemic 2008. Geneva: Author. WHO & UNAIDS. (2007). Guidance on provider-initiated HIV testing and counseling in health facilities. Geneva: World Health Organization. WHO & UNAIDS. (2013). Guiding principles on ethical issues in HIV surveillance. Geneva: World Health Organisation.
Copyright of AIDS Care is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
