Journal of the International AIDS Society

Introduction: Adolescent girls and young women (AGYW) in southern Africa are disproportionately affected by HIV. Despite increasing availability of HIV pre-exposure prophylaxis (PrEP), uptake and sustained use remain low. Existing service delivery models may not adequately meet the needs of AGYW, particularly in remote settings. We conducted a discrete choice experiment (DCE) to assess preferences for PrEP service delivery among AGYW living in Lesotho, a country with one of the highest HIV incidence rates globally. Methods: The DCE was conducted among AGYW (16-24 years) in two districts in Lesotho. Participants completed a series of binary choice tasks comparing hypothetical PrEP service delivery scenarios defined by six attributes: service location, provider type, provider characteristics, provider confidentiality, PrEP product type, and the combination of additional prevention services offered. Preferences were analysed using mixed logit and latent class models. Results: A total of 537 AGYW (median age 19 years, IQR 17-22) were included. Provider confidentiality was the strongest driver of choice, with non-confidential providers significantly less preferred ({beta} = -0.58; 95% CI -0.69 to -0.46). Compared with nurses, services delivered by ComBaCaL CHWs were preferred (0.17; 0.01 to 0.33), while those provided by doctors were less preferred (-0.15; -0.30 to 0.00). Younger female providers were preferred over older female providers (0.20; 0.04 to 0.36). Compared with the daily oral pill, both the 2-monthly injectable (-0.24; -0.39 to -0.08) and the vaginal ring (-1.02; -1.20 to -0.82) were less preferred. Differences in preferences were observed across age groups and districts. Latent class analysis identified two preference profiles, indicating variation in preferences for delivery and product characteristics. Conclusions: Preferences for PrEP delivery among AGYW in Lesotho were strongly influenced by provider confidentiality. Among some AGYW, there was openness to decentralised delivery, particularly through CHWs and community-based models, which may reduce access barriers in remote settings. Product preferences were varied, and not all options were acceptable. Differences by age group and district indicate that no single delivery model will meet all needs. Building on the current standard of care, offering acceptable options in accessible and confidential ways may support PrEP uptake.

Introduction Viral load monitoring is central to assessing antiretroviral therapy (ART) effectiveness, yet timely access remains challenging in resource-constrained settings. Point-of-care (POC) triage tests may improve ART monitoring efficiency by identifying clients requiring confirmatory viral load testing while reducing unnecessary testing among those likely to be virally suppressed. We explored perceptions of integrating a POC triage test that measures interferon-gamma-induced protein 10 (IP-10) - a chemokine strongly correlated with HIV viral load - into routine ART monitoring among people living with HIV (PLHIV) on ART, healthcare providers, and HIV programme stakeholders. Methods This qualitative study was nested within a clinical evaluation of the IP-10 POC triage test in two primary healthcare facilities in Maputo Province, Mozambique (2023-2024). We conducted three rounds of interviews with PLHIV on ART who underwent IP-10 testing, and one-off interviews with healthcare providers and HIV programme stakeholders across different health system levels. PLHIV were purposively sampled to capture diverse IP-10 and viral load outcomes. Interviews explored experiences of ART monitoring, perceptions of the IP-10 POC test, and implementation considerations. Data were analysed thematically using an inductive-deductive approach. Results Routine viral load monitoring was widely valued and understood as essential for treatment adherence and effectiveness, but participants described barriers including laboratory delays, access challenges, and health system constraints. The IP-10 POC triage test was broadly acceptable; same-day results were perceived to reduce anxiety, support adherence, and enable timely clinical decision-making. Providers and stakeholders emphasised its potential to improve monitoring efficiency by prioritising clients who require confirmatory viral load testing and adherence support. Concerns were raised regarding test accuracy and the need to maintain confirmatory viral load testing, underscoring the importance of clear communication and client education. Successful implementation would require training, workflow integration, and quality assurance. Conclusions An IP-10 POC triage test could strengthen ART monitoring by enabling same-day identification of clients requiring confirmatory viral load testing and targeted adherence support. By reducing unnecessary viral load testing for virally suppressed clients, it may contribute to more efficient monitoring and support differentiated care approaches. Careful integration into existing ART monitoring algorithms will be critical to maximise impact.

Background: Retention to care and viral load suppression are essential components for effective HIV management, particularly among adolescents and young adults aged 15-24 years, who remain vulnerable to treatment challenges. This study aimed to assess factors associated with poor retention in care and viral load suppression among young people receiving antiretroviral therapy (ART) at Mpilo Centre of Excellence (MCoE) in Bulawayo, Zimbabwe, with the objective to guide youth-friendly interventions and improve health outcomes. Methods: A mixed methods cross-sectional study was conducted involving 110 HIV-positive youths aged 15-24 years on ART, recruited through systematic sampling and surveyed between November and December 2024. Data was collected using structured questionnaires, focus group discussions, in-depth interviews, and key informant interviews. Quantitative data were analyzed using descriptive statistics and logistic regression models to identify factors linked to viral load suppression, while qualitative data underwent thematic analysis. Results: Viral load suppression was achieved by 68.19% of participants, who met the viral suppression criterion of <50 copies/ml. Analysis identified several significant predictors via multivariable logistic regression. Younger adolescents (15-19 years) had lower odds of achieving suppression compared to older youths (20-24 years) (Adjusted Odds Ratio [AOR]: 0.81; 95% Confidence Interval [CI]: 0.67-0.97; p=0.041), while female participants demonstrated higher suppression rates than males (AOR: 0.43; 95% CI: 0.21-0.96; p=0.032). Absence of adherence challenges to ART emerged as a strong predictor of suppression (AOR: 0.12; 95% CI: 0.03-0.72; p=0.018), and perceived lack of clinical staff support was associated with a threefold higher risk of unsuppressed viral load (AOR: 3.01; 95% CI: 1.34-7.69; p=0.046). Lower treatment self-efficacy negatively impacted suppression odds (AOR: 2.65; 95% CI: 1.11-7.83; p=0.046), and lack of friend support for clinic visits reduced the likelihood of suppression (AOR: 0.31; 95% CI: 0.09-0.89; p=0.001). Qualitative findings confirmed that persistent barriers--including stigma, limited family support, economic hardship, school and work commitments--compromised both retention and adherence among adolescents and young adults. Conclusion: Younger age, male sex, ART adherence challenges, lack of clinical staff support, and lower treatment self-efficacy were significantly associated with poor viral suppression among 15-24-year-olds at Mpilo Centre of Excellence. These findings underscore the need for tailored adolescent- and youth-friendly services, enhanced adherence support, and improved treatment literacy to strengthen retention in care and viral suppression. Addressing these factors is critical for advancing progress towards UNAIDS 95-95-95 targets and reducing HIV transmission among Zimbabwean youth.

IntroductionUnderstanding provider preferences for the design of HIV treatment packages could enhance the implementation of programs to support the adoption of long-acting injectable antiretroviral therapy (LAI ART) by people living with HIV who are interested in initiating this treatment modality. MethodsWe recruited providers from New York City (NYC), Rockland, Putman, and Westchester County Ryan White Part A Medical Case Management (MCM) programs to complete a discrete choice experiment (DCE) containing twelve tasks with two alternatives and an opt-out option, with additional survey questions about implementation readiness and choice motivations. The alternatives included four attributes--Type of ART Medication (monthly or bimonthly LAI ART), Service Location and Mode, Support for Clients, and Rewards for Clients-- with 2-4 levels each. We ran latent class multinomial logit analyses (LCA) with 1-5 classes to estimate preferences and explore hypothesis-free preference heterogeneity. We estimated attribute influence using relative importances and preferences using zero-centered part-worth utilities for each level. ResultsOne hundred seventy-seven providers completed the survey (July 2022-January 2023). About half (52%) were 40-59 years old, 72% identified as women, and the plurality (41%) identified as Latino/a. We chose the two-group LCA solution. Bimonthly LAI ART was preferred over monthly LAI ART overall and in both groups. Group 1 (n=45) preferred more traditional adherence supports (e.g., injections at the clinic by appointment, injection appointment reminders) whereas Group 2 (n=132) preferred more client-centered supports (e.g., injections at home by appointment, free transportation to injection appointments if at a clinic). Both groups preferred higher monetary value gift cards for clients for every on-time injection. The top-ranking motivations indicated that participants prioritized patient convenience over job satisfaction and administrative or financial feasibility for the agency. The scores for all implementation measures indicate readiness to implement LAI ART in both groups. ConclusionsOur implementation science-focused study suggests that providers of MCM services in NYC and surrounding counties are motivated to offer services to support clients access and adherence to LAI ART. More work is needed to understand how programs have, in fact, integrated supports for LAI ART into their services.

BackgroundSustained retention in care supports continuous access to antiretroviral therapy, routine clinical monitoring, and long-term viral suppression. ObjectiveTo compare the effectiveness of interventions for improving retention in care among people living with HIV (PLHIV). DesignSystematic review and network meta-analysis Data sourcesPubMed, Embase, CINAHL, PsycINFO, Web of Science, and the Cochrane Library from 1995 to December 2024. Eligibility criteriaRandomised controlled trials (RCTs) evaluating interventions to improve retention in care, viral load suppression, or quality of life (QoL) among PLHIV, compared with standard of care (SoC) or other interventions. Data extraction and synthesisPairs of reviewers independently screened studies, extracted data, and assessed risk of bias using ROBUST-RCT. We conducted a fixed-effect frequentist network meta-analysis and rated interventions categories relative to SoC based on effect estimates effects and the certainty of evidence.. Dichotomous outcomes were summarized as odds ratios (ORs) with 95% confidence intervals (CIs), and continuous outcomes as mean differences (MDs) with 95% CI. ResultsEighty-four trials enrolling 107 137 PLHIV evaluated 13 intervention categories. For retention in care, five interventions supported by moderate or high certainty evidence proved superior to SoC: multi-month dispensing (OR 2.02, 95% CI 1.32 to 3.09), task shifting (OR 1.94, 95% CI 1.42 to 2.66), differentiated service delivery (OR 1.47, 95% CI 1.22 to 1.76), behavioural counselling (OR 1.36, 95% CI 1.21 to 1.54), and supportive interventions (OR 1.31, 95% CI 1.11 to 1.55). For viral load suppression, two interventions supported by moderate or high certainty evidence proved superior to SoC: task shifting (OR 2.07, 95% CI 1.25 to 3.43) and behavioural counselling (OR 1.34, 95% CI 1.11 to 1.67). Across outcomes, no intervention demonstrated convincing superiority over other active interventions. ConclusionsAmong 13 intervention categories, only a subset provided moderate or high-certainty evidence of superiority to the standard of care, and no superiority to other interventions. Persistent evidence gaps for key populations, diverse settings, and long-term outcomes support the need for context-sensitive and patient-centred interventions. RegistrationPROSPERO CRD42024589177 Strengths and limitations of this study[tpltrtarr] This systematic review followed Cochrane methods and was reported in accordance with PRISMA-NMA guidelines. [tpltrtarr]The network meta-analysis integrated direct and indirect evidence to compare multiple intervention categories within a single framework. [tpltrtarr]Risk of bias and certainty of evidence were assessed using ROBUST-RCT and the GRADE approach for network meta-analysis, respectively. [tpltrtarr]Some networks were sparse, and limited representation of key populations and long-term follow-up constrained the strength and generalisability of inferences.

IntroductionSustaining long-term viral suppression among people living with HIV (PLHIV) remains a major public health challenge in sub-Saharan Africa, despite widespread access to antiretroviral therapy (ART). Evidence on time to first-line ART failure and its predictors at a national scale remains limited, particularly for dolutegravir (DTG)-based regimens. We aimed to estimate the time to first-line ART failure and identify associated predictors among PLHIV in Tanzania using national programmatic data. MethodsWe conducted a retrospective cohort study using routinely collected data from the National Care and Treatment Clinic database (CTC-2) in Tanzania. The analysis included PLHIV aged [&ge;]11 years who initiated first-line ART between January 2017 and December 2021 and had at least six months of follow-up. Time to first-line ART failure was defined as the duration from ART initiation to the first documented virological failure (viral load [&ge;]1,000 copies/mL). Kaplan-Meier methods were used to estimate failure-free survival, and Cox proportional hazards models were used to identify predictors of failure. Non-proportional hazards for DTG-based regimens were addressed using an extended Cox model with a time-varying coefficient. ResultsThe final analytic cohort comprised 36,764 individuals and 789 first-line treatment failure events. Median follow-up time varied across regimen groups. Failure-free survival differed significantly by regimen anchor (log-rank p<0.001). In multivariable Cox models, age and gender were significantly associated with treatment failure. DTG-based regimens demonstrated a time-varying effect: compared with non-DTG regimens, DTG was associated with a substantially lower hazard of failure early after initiation, with the protective effect attenuating over time. Estimated hazard ratios for DTG versus non-DTG regimens were 0.37, 0.67, and 1.22 at 6, 12, and 24 months of follow-up, respectively. ConclusionsIn this large national cohort, the risk of first-line ART failure varied by regimen and patient characteristics. DTG-based regimens showed strong early protection against failure, but this effect diminished over time, highlighting the importance of continued virological monitoring after ART initiation. Time-to-event analyses using routine programmatic data provide important evidence for optimizing ART delivery and informing HIV programme decisions in Tanzania and similar settings.

Introduction: Differentiated service delivery (DSD) models aim to reduce time healthcare providers spend with DSD clients, increasing time available for non-DSD clients. We measured nurses' time allocation and explored their experiences with DSD models in South Africa. Methods: We conducted time and motion observations and surveyed nurses at 24 public primary healthcare facilities across two SENTINEL study rounds (09/2022-07/2023 and 11/2023-07/2024). We report median time nurses spent by activity, model of care, and interaction type. Log binomial regression investigated factors associated with high direct nurse-client interaction (above median minutes) and extended work-days ([&ge;]9 hours), and estimated adjusted risk ratios (aRR). Survey questions were related to client care, additional time availability, and policy changes post DSD implementation, with key themes presented alongside illustrative quotes. Results: 176 nurses (88% female, median age 44) were observed for 344 working days; of these, 60 (34%) participated in the provider survey. Nurses spent a median of 293 minutes (53% of their work-day) on direct nurse-client interaction, 89 minutes (22%) on client-support or facility-related tasks, and the remainder on other activities including personal breaks. Time spent per client was similar across conventional care clients (11 [IQR: 8-15] minutes) but ranged between 9 (7-13) to 11 (8-15) minutes for DSD clients; number of direct nurse-client interactions did not differ meaningfully. Nurses at facilities with 2,000-3,999 total remaining on ART (TROA) (aRR 1.56, 95% CI: 1.02-2.37) and in urban areas (aRR 1.43, [1.08-1.89]) had more direct nurse-client interactions than those at facilities with <1,999 TROA and in rural areas, respectively. Nurses at facilities with 4,000+ TROA (aRR 2.22, [1.36-3.63]) and those observed in SENTINEL 3.0 (aRR 1.53, [1.13-2.07]) were more likely to work standard or longer workdays than those at lower TROA facilities (<1,999), those in SENTINEL 2.0 and urban areas. Nurses reported DSD models improved client care (90%), freed up time (60%), and changed clinic procedures and policies (60%). Conclusions: While DSD models did not significantly reduce direct nurse-client interaction time, nurses reported improved client care and gained additional time. DSD impact may vary by facility context. As DSD implementation expands, effective time reallocation may enhance facility performance and provider productivity.

BackgroundRoutine viral load monitoring is central to assessing treatment effectiveness in HIV care, and dolutegravir (DTG)-based regimens are now preferred in many treatment programmes. However, national routine data analyses comparing 48-week viral load suppression across antiretroviral therapy initiation regimens in Tanzania remain limited. MethodsWe conducted a retrospective cohort analysis using routinely collected HIV programme data from Tanzanias National AIDS, STIs and Hepatitis Control Programme database. After de-duplication and data processing, the working analysis warehouse contained 49,547 patients and 1,008,137 visits. The primary analysis included 6,991 patients with a valid viral load measured 48 weeks after initiation of antiretroviral therapy. Viral suppression was defined as a viral load <1,000 copies/mL. We compared suppression between DTG-based and non-DTG-based initiation groups and across individual initiation regimens. Treatment change episodes and early DTG switching patterns were summarized as secondary analyses. ResultsOf the 6,991 included patients, 6,113 (87.4%) achieved viral load suppression at 48 weeks. Suppression was higher among DTG initiators than non-DTG initiators (917/1,000, 91.7% vs. 5,196/5,991, 86.7%). TDF+3TC+EFV was the most common non-DTG initiation regimen, whereas TDF+3TC+DTG was the most common regimen among DTG initiators. ConclusionsViral suppression at 48 weeks was high overall but was higher among patients initiated on DTG-based regimens than among those initiated on non-DTG regimens. By anchoring outcomes to a fixed post-initiation time point, this study complements existing Tanzanian evidence on viral load testing uptake and geographic variation. It provides regimen-specific insights into the effectiveness of early treatment under routine programme conditions.

Background Interruptions in HIV care pose a major challenge to achieving HIV control goals in many countries, with 30% of clients who initiate antiretroviral therapy (ART) in South Africa experiencing an interruption of >28 days during their first six months on treatment. South Africa introduced revised guidelines in 2023 to improve outcomes during this early treatment period, but guideline compliance remains incomplete and gaps in the support provided to both clients and providers to optimize service delivery and health outcomes. Protocol BRIDGE (Behavioral Risk Identification and Decision Guidance for Engagement) is a mixed-methods evaluation of a package of light-touch, low-cost interventions aimed at improving the experiences of both clients and providers of care, increasing compliance with the 2023 guidelines, supporting clients to remain in care, and ultimately reducing the incidence of missed visits during the early treatment period. Components of the BRIDGE Retention Toolkit include an intervention navigator to help clients self-assess areas of vulnerability for disengagement from care and identify appropriate interventions; client roadmap to explain the treatment journey for the early treatment period; WhatsApp-based counseling tool for clients; guideline reference for providers; and tracing job aids. The tookit will be piloted at 6-8 public sector primary health facilities for a one-month period. The primary outcome will be the probability of returning less than 28 days late for the next scheduled clinic visit, assessed using electronic medical record data for the pilot and comparison sites. Pilot outcomes will be compared to both their own probabilities prior to the pilot and to probabilities from comparable non-pilot facilities. Implementation outcomes to be assessed using qualitative interview data from both clients and providers will include reach, implementation fidelity, adoption (uptake), costs, feasibility, appropriateness, and acceptability. Discussion The evaluation will assess the implementation and preliminary effectiveness of a set of interventions designed to improve client outcomes during the early HIV treatment period. If some or all of the BRIDGE tools are found to be helpful and/or are associated with a reduction in missed clinic visits, they will comprise a readily scalable and affordable intervention to help address a major barrier in large-scale HIV treatment programs.

Introduction: Viral load (VL) monitoring is the gold standard for antiretroviral therapy (ART) monitoring. Still, due to limited funds and infrastructure, many people living with HIV (PLHIV) in low- and middle-income countries do not receive timely VL testing. We evaluated the clinical performance and end-user acceptability of a prototype interferon gamma-induced protein 10 (IP-10) point-of-care (POC) test as a rule-out triage tool to identify individuals unlikely to have unsuppressed VL in PLHIV in Mozambique. Methods: A mixed-methods study was conducted between November 2023 and November 2024 at two primary healthcare facilities in Maputo Province. We enrolled 1,057 PLHIV on ART from stable and specialized risk clinics. Clinical performance of the IP-10 POC test (index test) was compared against plasma HIV VL (reference test; unsuppressed defined as >1000 copies/mL). Socio-demographic and clinical predictors of false-positive results were identified using multivariable logistic regression. Immediate acceptability was assessed through exit interviews on a subset of 43 PLHIV. Results: Among participants (71.7% female; median age 41.4 years), 12.0% had unsuppressed VL. The IP-10 POC test demonstrated high sensitivity (90.6%) and moderate specificity (35.6%). Specificity was higher in clinics treating stable patients (44.5% 95%CI: 39.7-49.3) compared to specialized risk clinics (26.5% 95%CI: 21.1-28.9). The proportion of false-positive results was also higher in patients attending specialized risk clinics. Independent predictors of false positivity included enrolment in a one-stop TB/HIV clinic (aOR=2.99 95%CI: 1.09-8.15), cotrimoxazole use (aOR=2.16, 95% CI: 1.13-4.13), and obesity (aOR=3.47 95%CI: 1.74-6.93). Acceptability was high: 70% of participants appreciated the test simplicity and rapid results, and 95.3% expressed interest in future testing. Most patients preferred finger-prick collection over venous draws. Conclusions: The IP-10 POC test is a highly sensitive triage tool, demonstrating superior performance among stable PLHIV enrolled in differentiated service delivery models like six-month multi-month dispensing. While factors associated with co-infections can reduce specificity, the test's high acceptability and potential to reduce confirmatory VL test demand suggests it could serve as a viable triage strategy for optimizing resources particularly in stable care pathways with a lower prevalence of inflammatory comorbidities. This could enable health systems to reallocate intensive monitoring toward higher-risk populations.

IntroductionHIV testing for children of women living with HIV (WLHIV) is an efficient method of diagnosing HIV in children. We analyzed pooled data from 13 Population-based HIV Impact Assessments (PHIA) conducted from 2015 through 2019 to determine the gap in diagnosing HIV in children of WLHIV. MethodsIn each PHIA, children younger than 15 years in a subset of households were sampled for HIV testing. Mother-reported responses on childs status were linked to maternal interviews and biomarker data. Analysis was restricted to children whose mothers were alive, older than 15 years and aware of their HIV-positive status prior to the survey. We calculated weighted proportions of children who were never previously tested and proportion of children living with HIV (CLHIV) with no evidence of antiretroviral treatment (ART) use (categorized as newly diagnosed). Survey weights were pooled across all PHIAs to account for survey design and nonresponse. ResultsOf 4,234 WLHIV, 3,436 were aware of their HIV status and had at least one child (n=6,173) for whom responses were obtained. Of the 6,173 children, 43.5% (n=2,371) were reported as never been tested. Overall, 5,500 children provided blood for HIV testing during the survey. Newly diagnosed test positivity was 1.7% (90/5,191); 2.9% (61/2,120) among those with reported unknown HIV status and 0.9% (29/3,071) among those with reported HIV negative status. Among children with reported HIV positive status, 94.5% were confirmed by survey testing and of these, 91% had antiretrovirals (ARVs) detected. ConclusionsOver 40% of children of WLHIV who were aware of their HIV positive status had never been tested for HIV. HIV positivity ranged between 0.9% to 2.9% while 9.0% of children known to be HIV positive were not on ART. The study calls for renewed efforts to enhance testing of children and treatment linkage for those diagnosed with HIV.

BackgroundAdolescents and young adults with perinatally acquired HIV (APHIV) face complex psychosocial and structural challenges that may undermine resilience, a modifiable psychosocial determinant of treatment engagement, and health outcomes. Evidence on peer-led interventions targeting resilience among APHIV in South Asia remains limited. We evaluated resilience and its correlates among participants in the ImPossible Fellowship, a peer-led mentorship intervention in India. MethodsWe conducted a cross-sectional evaluation of 216 APHIV following completion of the 24-month ImPossible Fellowship in southern India in 2024. Surveys administered by trained youth investigators assessed sociodemographic, educational, and clinical characteristics. Resilience was measured using the Child and Youth Resilience Measure-Revised (CYRM-R), a validated multidimensional tool capturing personal and relational resilience dimensions. Low resilience was defined as CYRM-R threshold score [&le;]33rd percentile. Multivariate logistic regression identified independent correlates of low resilience, and sensitivity analyses explored alternative CYRM-R thresholds. ResultsParticipants had a mean age of 18.7 years (range 9-24); 50% had no surviving parents, and 43% lived in child care institutions. Median resilience scores were high (74, Interquartile range [IQR] 69-78), and 91% achieved viral suppression. In multivariate analyses, three factors were independently associated with low resilience: loss of both parents (adjusted odds ratio [aOR] 4.35, 95% CI 2.09-9.06), school discontinuation (aOR 2.43, 95% CI 1.10-5.34), and self-reported communication barriers at HIV clinics (aOR 5.83, 95% CI 2.69-12.64). These associations were consistent across sensitivity analyses at alternative resilience thresholds. At the most stringent threshold of low resilience (CYRM-R score [&le;]15th percentile), unsuppressed viral load also emerged as a significant correlate, suggesting that treatment failure may be concentrated among those with the most severely compromised resilience. ConclusionsAPHIV participating in a peer-led mentorship program demonstrated high overall resilience and viral suppression, but also revealed addressable vulnerabilities mapping to specific programmatic priorities. Peer-led models offer a promising foundational platform; however, complementary structural and psychosocial enhancements targeting these modifiable determinants are essential to optimize outcomes for those facing the greatest cumulative adversity.

Background Studies show 53 to 74% of women living with HIV experience postpartum ART adherence challenges. Viral load testing is a delayed indicator, highlighting the need for culturally appropriate screening tools to identify at-risk women early. This study examined the association between non-viral suppression and constructs within the AIDS Clinical Trials Group (ACTG) adherence questionnaire among women in Uganda to inform timely, targeted interventions to improve adherence. Methods The ACTG was adapted, and postpartum participants completed ACASI or Provider-Assisted Interviews (PAIs). Self-efficacy, social support, anxiety, depression, viral load, and clinical factors were analysed using mixed-effects logistic models over a 1-year period. Results Of 166 women, 21 completed ACASI and 145 PAIs. 4.2% (7/166) were not virally suppressed at baseline, and their non-suppression status was consistent throughout one year of follow-up. High self-efficacy scores were associated with 27% lower odds of viral non-suppression (Odds Ratio [OR], 0.73; 95% CI, 0.54, 0.98). High depression scores were associated with 22% higher odds of non-suppression (OR 1.22;95% (1.01,1.49). Other variables, including age, Body Mass Index, duration on ART, marital status, employment, education level, tap water, and travel time from home to clinic, were not associated with viral suppression in the covariate-adjusted analyses. Median self-efficacy and depression scores were 8 (IQR 1,9) and 1.2 (IQR 0,16), respectively. Focused group discussion data showed high acceptability and feasibility of using the ACTG adherence questionnaire in Uganda. Conclusion Lower self-efficacy and higher depression scores on the ACTG adherence questionnaire can help identify Ugandan women at risk of viral non-suppression in HIV programs. Keywords WLHIV, Antiretroviral Therapy, Adherence, Audio Computer Assisted Self Interview, Viral load

Background: Pre-exposure prophylaxis (PrEP) has demonstrated a significant reduction in HIV infections among men who have sex with men (MSM), however, low medication adherence hinders its preventative effectiveness. Traditional approaches, such as health education and face-to-face inquiry (HEF), have demonstrated certain efficacy in improving PrEP adherence. However, these methods are resource-intensive and often plagued by delays, rendering timely and precise interventions challenging. This randomized controlled trial aims to assess the effectiveness of an intervention comprising AI-chatbot for PrEP (PrEP-bot) and Smart pillbox (SPB) (PrEP-bot-SPB) strategy to improve PrEP adherence among MSM compared to HEF.Methods and analysis: A three-arm, multicenter, open-lable RCT will be conducted with Chinese MSM [&ge;]18 years. A total of 300 participants will be recruited through three sources, including hospitals, community-based organizations (CBOs) and peer referral in five cities: Shenzhen, Beijing, Qingdao, Hangzhou and Zhengzhou. After completing baseline survey, participants will be randomized evenly into interventions or control groups: the PrEP-bot group, the PrEP-bot-SPB group, and the HEF control group. Participants in the PrEP-bot group will be granted access to an AI-chatbot agent through WeChat. This agent will: 1) generate personalized PrEP medication plans; 2) provide medication reminders and PrEP-related health check-ups notifications; 3) inquire about missed doses to deliver tailored interventions; 4) answer participant questions about PrEP using guideline-based knowledge. Participants in the PrEP-bot-SPB group will receive both the SPB and the PrEP-bot interventions. SPB could delivers medication reminders. Participants in HEF group will receive a health education pamphlet introducing PrEP and knowledge related to PrEP medication adherence at baseline and face-to-face inquiry every three months. Outcomes will be assessed for both short-term and medium-to-long-term effects. The primary objective is the effectiveness in improving PrEP adherence measured by self-report, Eight-Item Morisky medication adherence scale (MMAS-8) and concentration of Tenofovir in dried blood spots (DBS) (PrEP adherence [&ge;]90%) at 3 months follow-up. Secondary outcomes include: 1) effectiveness in preventing HIV infection measured by HIV-self test (HIVST); 2) effectiveness of PrEP-related health check-ups; 3) the effectiveness, feasibility, acceptability, and user satisfaction with the PrEP-bot; 4) effectiveness in improving PrEP adherence at 6-month, 9-month and 12-month follow-up periods. All participants will receive quarterly follow-up visits during the 12-month study period. Intention-to-treat analysis and per protocol set (PPS) analysis will be used.Results: Recruitment and enrollment of participants began in January 2026 and is currently ongoing.Discussion: This study is expected to establish a novel AI-based intervention model for PrEP, providing innovative strategies for HIV control among MSM populations. If the PrEP-bot is proven non-inferior to HEF, it could offer users real-time, precise, and personalized interventions while simultaneously addressing PrEP-related inquiries and health check-ups reminders. Importantly, this approach would achieve significant reductions in resource requirements for implementation and maintenance and be more cost-effective. With the ongoing advancement of AI technologies, PrEP-bot holds substantial promise for widespread implementation in PrEP adherence, potentially revolutionizing HIV prevention for MSM in China through this innovative intervention modality.Trial registration: ChiCTR2500111280 (Chinese Clinical Trial Registry). Date of registration: 29 October 2025.

Background: Advanced HIV disease (AHD) remains a major contributor to HIV-related morbidity and mortality despite widespread antiretroviral therapy (ART) access in sub-Saharan Africa. Although treatment-related adverse drug reactions (ADRs) may compromise treatment outcomes, evidence on the relationship between AHD and ADR occurrence remains limited. This study aimed to determine the prevalence and identify factors associated with AHD, characterize treatment-related ADR and assess the association between AHD and ADR occurrence among people living with HIV receiving ART in Dar es Salaam, Tanzania. Methods: We conducted a multicenter cross-sectional study among 1,513 people living with HIV receiving ART at selected HIV care and treatment clinics in Dar es Salaam, TanzaniaFor this adolescent/adult cohort, AHD was operationally defined as WHO clinical stage III/IV disease and/or baseline CD4 count <200 cells/mm3. Treatment-related ADRs were defined as participant-reported and/or clinically documented ART-related adverse events identified during routine HIV care, including both current and retrospectively reported events. Modified Poisson regression with robust standard errors was used to estimate crude and adjusted risk ratios (RRs) with 95% confidence intervals (CIs). Results: Among 1,508 participants with sufficient information for classification, 961 (63.7%) had AHD. Factors independently associated with AHD included age [&ge;]50 years (aRR 1.10, 95% CI 1.01-1.20), underweight nutritional status (aRR 1.17, 95% CI 1.00-1.35), and concomitant medication use (aRR 1.19, 95% CI 1.03-1.37), while DTG-based ART was associated with lower AHD prevalence (aRR 0.78, 95% CI 0.68-0.90). Overall, 569 participants (38.0%) reported at least one ADR. Composite AHD was not independently associated with ADR occurrence (aRR 0.95, 95% CI 0.82-1.11), but baseline CD4 <200 cells/mm3 was associated with increased ADR risk (aRR 1.20, 95% CI 1.02-1.41). Comorbidity (aRR 1.66, 95% CI 1.42-1.93) was the strongest correlate of ADR occurrence. Conclusion: AHD remains highly prevalent among people living with HIV receiving ART in Tanzania. While composite AHD was not independently associated with ADR occurrence, severe immunosuppression, comorbidity burden, and concomitant medication exposure were associated with increased ADR risk. These findings suggest that immunologic severity and broader clinical complexity may be more informative predictors of ART-related toxicity than composite syndromic AHD classification alone. Strengthened early diagnosis, differentiated advanced HIV care, integrated pharmacovigilance strategies, and routine medication risk assessment are needed.

BackgroundTo promote sustainability and strengthen national ownership of Advanced HIV Disease (AHD) services, a transition was implemented across 22 health facilities in Central Malawi. This transition involved shifting responsibility for key AHD program elements, including clinical service delivery, diagnostics, provider mentorship, and reporting systems, from implementing partner-led implementation to full Ministry of Health (MoH) leadership. This evaluation assessed the impact of this transition on diagnostic coverage, TB preventive therapy (TPT) uptake, and 12-month survival outcomes. MethodsA retrospective cohort study was conducted involving all children and adults enrolled in AHD care during the pre-MoH transition (January 2020-December 2021) and post-MoH transition (January 2023-December 2024) periods. Eligibility followed national AHD criteria: CD4 count <200 cells/mm3, WHO stage 3 or 4 illness, or age <5 years. AHD clients data were abstracted from clinical records and linked across routine facility registers to assess diagnostic and treatment indicators. Kaplan-Meier survival curves, Cox proportional hazards, and Fine and Gray competing risk models were used to evaluate 6 and 12-month mortality and retention as primary outcomes. ResultsA total of 1,044 AHD clients were included (553 pre-transition; 491 post-transition) in the evaluation. Median age increased post-transition (35.9 to 38.5 years, p<0.001). CD4 testing declined (80.7% to 46.0%, p<0.001) testing uptake, while WHO staging and TB diagnostic coverage improved. TB diagnoses decreased (44.5% to 31.2%, p=0.002). TPT uptake dropped from 46.4% to 31.6% (p<0.001). Twelve-month mortality significantly declined from 9.4% to 5.5% (adjusted hazard ratio [aHR]=0.59, 95% CI: 0.37-0.94, p=0.026). Retention in care remained stable (HR=0.86, 95% CI: 0.62- 1.20, p=0.384). ConclusionsTransitioning AHD services to MoH leadership sustained key program outcomes and significantly reduced mortality. Continued mentorship and government ownership were key drivers of success. However, declines in CD4 testing and TPT coverage highlight the need for strengthened diagnostics and preventive care integration. These findings support scaling nationally-led AHD models in high-burden HIV settings.

Background African immigrants in the United States bear a disproportionate HIV burden, with incidence approximately sixfold higher than the general population, yet remain largely absent from targeted prevention research. HIV vulnerability among this population is mediated through relationship and family systems that are restructured by migration, reorganizing household composition, gender norms, trust, and communication patterns through which prevention engagement occurs. Despite this, migration has rarely been examined as a force that transforms the relational contexts shaping engagement with HIV testing, HIV self-testing (HIVST), and pre-exposure prophylaxis (PrEP). Methods The MiST-Pathways Study will use a sequential mixed-methods, community-based pilot design among first-generation African immigrant adults (ages 18-50) residing in New York and Massachusetts. The study will proceed in three phases: Aim 1 will use semi-structured interviews (n = 15) and a structured survey (n = 75) to identify relationship typologies and migration-related relational mechanisms influencing HIV prevention engagement; Aim 2 will employ Palava Hut Conversations (PHC) (an African-centered deliberative method) with up to 30 participants to co-develop and prioritize relationship-tailored intervention components; and Aim 3 will conduct a proof-of-concept assessment of the prioritized component using a single-group pre-post design (n = 24), incorporating surveys and cognitive interviews to assess feasibility, acceptability, and preliminary evidence of mechanism activation. All activities will be conducted virtually via Zoom and WhatsApp, with eligibility screening administered through REDCap. The study has been approved by the University at Buffalo Institutional Review Board (STUDY00010347) and registered at ClinicalTrials.gov (NCT07565584). Discussion This protocol outlines the planned evaluation of a sequentially designed, community-engaged pilot study to examine how migration reshapes relational contexts that influence HIV prevention decision-making among African immigrants. Findings will inform the development of culturally grounded, relationship-tailored prevention strategies and the design of a future, larger-scale intervention study.

Evidence-informed situation analyses are critical for HIV programming, yet primary data collection is often resource-intensive and time-consuming. Approaches that leverage existing evidence while generating locally grounded insights are needed to inform context-appropriate interventions. We applied the Synthesised Narrative Exploration (SNE) approach in two districts in Inhambane province, southern Mozambique, to inform an HIV project through assessing key dynamics affecting HIV prevention, treatment, and care for women and children. SNE combines structured synthesis of existing country-specific literature with qualitative validation and contextualisation through focus group discussions and semi-structured interviews. Findings from 16 peer-reviewed studies and two national reports were synthesised into short narratives and discussed with 83 community members and key stakeholders. Qualitative data were thematically analysed and validated through community and stakeholder consultations. Study participants confirmed many well-documented barriers across the maternal and infant HIV care cascade and added local insights. Known pre-pregnancy barriers such as limited preconception HIV testing and low ART uptake were confirmed, while strong fertility expectations, low levels of pregnancy planning, gendered decision-making, and fear of partner abandonment following HIV status disclosure were highlighted as factors shaping engagement with HIV services. The participants confirmed barriers during pregnancy reported in earlier studies, including delayed antenatal care initiation and limited male partner involvement, and added the role of household power hierarchies, particularly the influence of mothers-in-law, in shaping HIV testing, disclosure, and ART adherence. Widely documented postpartum disengagement from care was explained by difficulties sustaining non-disclosure and the loss of socially acceptable reasons for continued clinic attendance. The participants proposed locally grounded strategies to address the barriers, including strengthened couple counselling and engagement of influential family members such as mothers-in-law. By integrating existing evidence in a qualitative enquiry, SNE enables efficient generation of contextually rich insights directly relevant to intervention design, offering an approach for strengthening HIV programmes.

IntroductionOn January 20, 2025, the U.S. government froze foreign assistance including for PEPFAR, though a limited waiver for "life-saving" interventions was subsequently granted. PEPFARs 2025 monitoring results, released April 17, 2026, covered only quarter 4 while an earlier inadvertent release included all four quarters. Combining both data sets, we systematically assess facility-level programmatic performance and reporting trends to quantify service disruptions accounting for reporting discrepancies. MethodsWe categorized facilities by reporting continuity across Q1 2024 and Q4 2025 (e.g. continuous, intermittent, dropped, or new) and assessed changes in service delivery by the category of health facility for key HIV treatment, testing, PMTCT, and prevention programming. We additionally analyze changes in employed human resources for health (HRH) reported by PEPFAR. ResultsPEPFAR data included 31,746 facilities and community service sites. 71.3% were classified as continuous reporters, 16.9% intermittent reporters, 2.5% community services, 3.9% dropped in 2025, and 3.1% new in 2025. Total number of people accessing HIV treatment declined modestly by -0.3%, but differed by facility category. Continuous facilities saw a 0.5% increase in people on treatment, while intermittent facilities saw a -1.7% decrease. HIV testing declined -17%. HIV diagnoses declined -13% in continuous facilities, -35% in community services, and -29% in intermittent facilities. PMTCT infant testing and diagnoses declined by -6% and -12% in continuous facilities, respectively, and -60% and -31% in intermittent facilities, respectively. PrEP initiations declined -33%. Total direct service delivery HCWs reduced -62,541 (-24%) ConclusionThese findings reveal substantial disruptions across PEPFAR service areas, with the steepest declines among intermittent and community-based delivery sites, alongside a 24% reduction in direct service delivery healthcare workers. As potentially the final data set PEPFAR will ever release, these findings represent a troubling inflection point. The dismantling of public data systems and accountability structures undermine progress and enable programmatic gaps to develop and go unnoticed that risk allowing HIV resurgence to occur over the coming years.

Delayed diagnosis and poor antiretroviral therapy (ART) adherence remain primary drivers of HIV-related morbidity in low-resource settings, yet real-world AI validation at scale is lacking. We conducted a retrospective validation study using two publicly available, de-identified datasets: a Quality of Care cohort of 27,288 HIV-positive patients on ART across multiple healthcare facilities, and the CEPHIA multi-country assay database comprising 165,444 specimen records from six countries. Four machine learning classifiers were evaluated using 10-fold stratified cross-validation with SMOTE applied strictly to training folds. Explicit data leakage prevention, ablation analysis, calibration assessment, and bootstrap confidence intervals were applied. Economic projections used one-way sensitivity analysis. This study adheres to TRIPOD reporting guidelines. Random Forest achieved AUC-ROC of 0.9753 (95% CI: 0.970-0.975), sensitivity 87.3% (95% CI: 86.4-88.2%), specificity 95.7% (95% CI: 95.2-96.2%), and Brier score 0.079. Ablation testing confirmed robustness (AUC 0.963 without the primary predictor). Temporal validation on held-out future patients yielded AUC 0.772 (95% CI: 0.744-0.802), confirming generalisation across time. Real-world analysis revealed median diagnosis-to-ART delay of 74 days, with 47.3% of patients exceeding 90 days and 36.7% presenting with CD4 below 200 cells per microlitre. Multi-country CEPHIA analysis identified 18.6% HIV recency within the 130-day early-intervention window. Decision curve analysis confirmed net clinical benefit across threshold probabilities 0.03-0.45. Subgroup analysis demonstrated consistent AUC across sex, age, CD4 strata, and WHO staging (max difference 0.051). Economic modelling projected base-case savings of USD 415 per patient (USD 2.07 million per 5,000-patient cohort). These findings provide large-scale empirical evidence that AI-driven informatics can predict ART adherence failure and quantify systemic care gaps, offering a scalable framework for equitable HIV care delivery in resource-limited settings. Prospective external validation is required before clinical deployment.