"As a patient I do not belong to the clinic, I belong to the community." Co-developing a multi-level, person-centred tuberculosis stigma intervention in Cape Town, South Africa.

Background: Anticipated, internal, and enacted stigma are major barriers to TB care engagement, and directly impact patient well-being. Unfortunately, targeted stigma interventions are lacking. We aimed to co-develop a person-centred stigma intervention with TB-affected community members and health workers in South Africa. Methods: Using a community-based participatory research approach, we conducted ten group discussions with people diagnosed with TB (past or present), caregivers, and health workers (total n=87) in Khayelitsha, Cape Town. Group discussions were facilitated by TB survivors. Discussion guides explored experiences and drivers of stigma and used human-centred design principles to co-develop solutions. Recordings were transcribed, coded, thematically analysed and then further interpreted using the socio-ecological model. Results: Intervention components across socio-ecological levels shared common behaviour change strategies, namely education, empowerment, engagement, and innovation. At the individual level, participants recommended counselling to improve TB knowledge and provide ongoing support. TB survivors can guide messaging to nurture stigma resilience by highlighting that TB can affect anyone and is curable, and provide lived experiences of TB to decrease internal stigma. At the interpersonal level, support clubs and family-centred counselling were suggested to dispel TB-related myths and foster support. At the institutional level, health worker stigma reduction training informed by TB survivor perspectives was recommended. Consideration of how integration of TB/HIV care services may exacerbate TB/HIV intersectional stigma and ideas for restructured service delivery models were suggested to decrease anticipated and enacted stigma. At the community level, participants recommended awareness-raising events led by TB survivors, including TB information in school curricula. At the policy level, solutions focused on reducing the visibility generated by a TB diagnosis and resultant stigma in health facilities and shifting tasks to community health workers. Conclusions: Decreasing TB stigma requires a multi-level approach. Co-developing a person-centred intervention with affected communities is feasible and generates stigma intervention components that are directed and implementable. Such community-informed intervention components should be prioritised by TB programs, including integrated TB/HIV care services.

Accuracy of digital chest x-ray analysis with artificial intelligence software as a triage and screening tool in hospitalized patients being evaluated for tuberculosis in Lima, Peru.

Tuberculosis (TB) transmission in healthcare facilities is common in high-incidence countries. Yet, the optimal approach for identifying inpatients who may have TB is unclear. We evaluated the diagnostic accuracy of qXR (Qure.ai, India) computer-aided detection (CAD) software versions 3.0 and 4.0 (v3 and v4) as a triage and screening tool within the FAST (Find cases Actively, Separate safely, and Treat effectively) transmission control strategy. We prospectively enrolled two cohorts of patients admitted to a tertiary hospital in Lima, Peru: one group had cough or TB risk factors (triage) and the other did not report cough or TB risk factors (screening). We evaluated the sensitivity and specificity of qXR for the diagnosis of pulmonary TB using culture and Xpert as primary and secondary reference standards, including stratified analyses based on risk factors. In the triage cohort (n = 387), qXR v4 sensitivity was 0.91 (59/65, 95% CI 0.81-0.97) and specificity was 0.32 (103/322, 95% CI 0.27-0.37) using culture as reference standard. There was no difference in the area under the receiver-operating-characteristic curve (AUC) between qXR v3 and qXR v4 with either a culture or Xpert reference standard. In the screening cohort (n = 191), only one patient had a positive Xpert result, but specificity in this cohort was high (>90%). A high prevalence of radiographic lung abnormalities, most notably opacities (81%), consolidation (62%), or nodules (58%), was detected by qXR on digital CXR images from the triage cohort. qXR had high sensitivity but low specificity as a triage in hospitalized patients with cough or TB risk factors. Screening patients without cough or risk factors in this setting had a low diagnostic yield. These findings further support the need for population and setting-specific thresholds for CAD programs.

Accuracy of digital chest x-ray analysis with artificial intelligence software as a triage and screening tool in hospitalized patients being evaluated for tuberculosis in Lima, Peru.

Introduction: Tuberculosis (TB) transmission in healthcare facilities is common in high-incidence countries. Yet, the optimal approach for identifying inpatients who may have TB is unclear. We evaluated the diagnostic accuracy of qXR (Qure.ai, India) computer-aided detection (CAD) software versions 3.0 and 4.0 (v3 and v4) as a triage and screening tool within the FAST (Find cases Actively, Separate safely, and Treat effectively) transmission control strategy. Methods: We prospectively enrolled two cohorts of patients admitted to a tertiary hospital in Lima, Peru: one group had cough or TB risk factors (triage) and the other did not report cough or TB risk factors (screening). We evaluated the sensitivity and specificity of qXR for the diagnosis of pulmonary TB using culture and Xpert as primary and secondary reference standards, including stratified analyses based on risk factors. Results: In the triage cohort (n=387), qXR v4 sensitivity was 0.91 (59/65, 95% CI 0.81-0.97) and specificity was 0.32 (103/322, 95% CI 0.27-0.37) using culture as reference standard. There was no difference in the area under the receiver-operating-characteristic curve (AUC) between qXR v3 and qXR v4 with either a culture or Xpert reference standard. In the screening cohort (n=191), only one patient had a positive Xpert result, but specificity in this cohort was high (>90%). A high prevalence of radiographic lung abnormalities, most notably opacities (81%), consolidation (62%), or nodules (58%), was detected by qXR on digital CXR images from the triage cohort. Conclusions: qXR had high sensitivity but low specificity as a triage in hospitalized patients with cough or TB risk factors. Screening patients without cough or risk factors in this setting had a low diagnostic yield. These findings further support the need for population and setting-specific thresholds for CAD programs.

"This is an illness. No one is supposed to be treated badly": Community-based stigma assessments in South Africa to inform TB stigma intervention design.

Background: Though TB-related stigma is a recognized barrier to care, interventions are lacking and gaps remain in understanding the drivers and experiences of TB-related stigma. We undertook community-based mixed methods stigma assessments to inform stigma intervention design. Methods: We adapted the Stop TB Partnership stigma assessment tool, and trained three peer research associates (PRAs; two TB survivors, one community health worker) to conduct surveys with people with TB (PWTB, n=93) and caregivers of children with TB (n=24) at peri-urban and rural clinic sites in Khayelitsha, Western Cape, and Hammanskraal, Gauteng Province, South Africa. We descriptively analyzed responses for each stigma experience (anticipated, internal, and enacted), calculated stigma scores, and undertook generalized linear regression analysis. We further conducted 25 in-depth interviews with PWTB (n=22) and caregivers TB (n=3). Using inductive thematic analysis, we performed open coding to identify emergent themes, and selective coding to identify relevant quotes. Themes were organised using the CARD (Constraints, Actions, Risks and Desires) framework. Results: Surveys revealed at least one-third of PWTB and one-quarter of caregivers report experiences of anticipated, internal, and/or enacted stigma, which affected engagement throughout the care cascade. Participants in rural locations (compared to peri-urban) reported higher anticipated, internal, and enacted stigma (ß-coefficient 0.72, 0.71, and 0.74). Interview participants described how stigma experiences, including HIV intersectional stigma, act individually and in concert as key constraints to impede care, and underpins failure to disclose a TB diagnosis, isolation, and exclusion. Stigma resilience arose through understanding that TB can affect anyone and should not diminish self-worth. Risks of stigma, driven by fears related to disease severity and infectiousness, led to care disengagement and impaired psychological wellbeing. Participants desired counselling, identifying a specific role for TB survivors as peer counsellors, and community education. Conclusions: Stigma is highly prevalent and negatively impacts TB care and the well-being of PWTB, warranting its assessment as a primary outcome indicator rather than intermediary contributor to poor cascade outcomes. Multicomponent stigma interventions are needed, including counselling for PWTB and education for health workers and communities. Such interventions must incorporate contextual differences based on gender or setting, and use survivor-guided messaging to foster stigma resilience.