Between empathy and anger: healthcare workers' perspectives on patient disengagement from antiretroviral treatment in Khayelitsha, South Africa -a qualitative study

Global HIV mortality has declined substantially with ART scale-up, yet disengagement from care remains a barrier to achieving optimal outcomes. South Africa, with the world’s largest ART programme, still experiences challenges in retention and adherence that compromise viral suppression. Disengagement contributes to transmission risk, resistance, advanced disease, and health system burden. Prior research identifies multiple determinants of disengagement (stigma, mental health, non-disclosure, access barriers), and indicates that anticipated punitive treatment by HCWs can deter re-engagement. However, most evidence reflects patient perspectives; HCWs’ views are underexplored. This study aimed to understand HCWs’ perspectives on disengagement from ART in Khayelitsha to inform interventions that support re-engagement and person-centred care.

Prior studies document substantial disengagement from ART in South Africa, with notable proportions interrupting care within two years. Disengagement is associated with virologic failure, drug resistance, advanced HIV disease, and increased hospitalisation. Patients often re-enter care when severely ill, reinforcing cyclical engagement patterns. Reported barriers include stigma, mental health disorders (notably depression), non-disclosure to family/partners, structural constraints, and negative clinic experiences. Anticipated or experienced punitive attitudes from HCWs are a major deterrent to re-engagement in sub-Saharan Africa. While HCWs are key gatekeepers shaping the care experience, literature focusing specifically on HCWs’ perspectives is limited, creating a gap in designing effective, empathetic, and person-centred service responses.

Design: Descriptive qualitative study using in-depth, semi-structured interviews. Setting: Ubuntu Clinic, Khayelitsha (peri-urban, low-resourced area in Western Cape, South Africa) with >8000 PLHIV on ART. The study informed Médecins Sans Frontières’ ‘Welcome Service’ intervention to support return-to-care patients and address HCW attitudes. Participants: Purposive sample of 30 HCWs (clinicians: doctors, nurses; support/non-clinical staff: counsellors, social workers, data clerks, security guards, occupational therapists). Eligibility: >3 months working at Ubuntu, permanently based, age >18, provided informed consent. Data collection: Semi-structured interviews (45–60 minutes), primarily in English; some in isiXhosa with translation during transcription. Interviews explored experiences with patients interrupting and re-engaging with ART, perceived reasons for disengagement, and HCWs’ reactions. Interviews were conducted privately, audio-recorded, and continued to saturation. Data analysis: Inductive thematic analysis following Braun and Clarke. Transcripts coded in NVivo 12 Pro. Initial codes and themes developed by the lead analyst were discussed and refined with senior researchers; a separate coding framework was compared to ensure rigour. Ethics: Approved by the University of Cape Town HREC and MSF ERB; written informed consent obtained; participants anonymised.

Sample characteristics: 30 HCWs; 70% female; age range 25–65 years (most 25–34); 67% support staff and 33% clinicians; 60% employed by Department of Health and 40% by NGOs. Time in HIV field: 27% <1 year; 20% >16 years (reported as >10 years: 37%); Time at Ubuntu: 33% <1 year; 20% >16 years. Emergent themes: 1) Disengagement is a public health problem: HCWs perceived disengagement as increasingly common and burdensome, leading to advanced illness, ART resistance, complex treatment needs, and hospitalisations that strain the health system. 2) Empathy and understanding: HCWs recognised complex, often justified drivers of disengagement. - Mental health challenges: Depression, psychosis, and poor insight/judgment were seen as major contributors; addressing mental health improved adherence for some. - Non-disclosure: Non-disclosure to family/partners impeded daily pill-taking and clinic attendance; HCWs acknowledged disclosure dilemmas and potential stigma but emphasised supportive disclosure to trusted individuals. - Concern for patients: Many HCWs expressed worry and sympathy, especially for young or severely ill patients, and a desire to understand reasons to tailor support. 3) Anger and frustration: HCWs reported frustration with patients who interrupt ART, citing increased workload (more frequent visits, group counselling, intensified monitoring) and a perception that some patients lack responsibility for their health. - Responsibility narratives: Many stated adherence is ultimately the patient’s responsibility; HCWs can educate and support but cannot assume patients’ life-management tasks. - Unjustifiable reasons: Some reasons (travel without planning, pill fatigue with low pill burden) were considered excuses; HCWs taking chronic meds themselves tended to show less tolerance, comparing their own adherence to patients’. Overall: HCWs hold contradictory feelings—empathy when disengagement seems beyond patients’ control, and anger when perceived as avoidable—shaping their interactions and potential gatekeeping roles.

The study illuminates HCWs’ dual reactions—empathy and anger—toward patients who disengage, clarifying how perceptions of responsibility and system burden inform attitudes. Empathy arises when disengagement relates to factors perceived beyond patient control (e.g., mental illness, disclosure challenges), aligning with evidence linking depression and non-disclosure to non-adherence. Frustration is driven by heavy workloads, complex re-engagement processes, and beliefs that patients should prioritise health. These attitudes can impact patient experiences and re-engagement, potentially perpetuating disengagement cycles. Findings support person-centred, holistic approaches acknowledging competing life priorities, augmented mental health and psychosocial support (including task-shifted counselling), and HCW-focused interventions that address stress, biases, and cognitive appraisals (e.g., CBT/ABC frameworks). Supporting HCWs with stress management, debriefing, and supervision may reduce burnout and negative interactions, improving retention and outcomes.

HCW–patient relationships around ART engagement are complex. HCWs view disengagement as a major public health concern and express both empathy and anger toward patients who interrupt treatment. To improve long-term retention, interventions should be person-centred and address mental health, disclosure support, and social contexts. Equally, programmes must support HCWs—through psychosocial support, skills-building, and mentorship—to reduce punitive attitudes, manage workload-related stress, and foster empathetic, constructive interactions that facilitate re-engagement and adherence.

- Single-site study in one primary care clinic limits generalisability to other settings. - Did not directly observe how HCWs’ feelings manifest during patient encounters; the impact on patient behaviour requires further study. - Heterogeneity among HCW roles suggests role-specific dynamics that were not fully disentangled; future research should examine differences across cadres. - Self-reported perspectives may be influenced by social desirability or recall biases.