Integrating HIV clinical records with population-based surveillance data allows the study of health care seeking behaviours, access to care, and predictors of patient outcomes. We implemented a graph-based record linkage algorithm to deduplicate and link HIV clinical and population-based surveillance records in an HIV-endemic setting in rural South Africa. We linked four data sources to create the Africa Health Research Institute (AHRI) Unified Data Platform: AHRI’s Health and Demographic Surveillance System (HDSS), AHRI Clinic and Hospital Information System (AHRILink), National Health Laboratory Service (NHLS), and Three Integrated Electronic Registers (TIER.Net) HIV care and treatment records. HDSS data were collected between January 1, 2000, and July 31, 2024, through repeated household surveys of over 140,000 individuals. Clinical and laboratory data were obtained for one hospital and 17 clinics in Hlabisa, KwaZulu-Natal, covering the HDSS surveillance area. We implemented a probabilistic record linkage algorithm trained and validated on a subset of records with national identity numbers. We assessed linkage accuracy, computed descriptive statistics for the linked database, and estimated the HIV care cascade for this population. A total of 986,832 records were successfully linked across the four databases, achieving a sensitivity of 92.7

OBJECTIVES:Network meta-analysis (NMA) with individual participant data can estimate how treatment effects change with patient characteristics. Yet cost-effectiveness analyses typically use population-average effects. We introduce a framework that incorporates NMA-derived heterogeneous treatment effects into cost-effectiveness analysis using a risk-modeling approach. METHODS:We first derived a baseline risk score for each patient using a prognostic model. This risk score was then used as an effect modifier in a network meta-regression to estimate risk-specific treatment effects. These effects were incorporated into the cost-effectiveness model to estimate the incremental cost-effectiveness ratios and net monetary benefits as functions of the baseline risk score. We demonstrated the approach using data from observational and randomized studies in relapsing-remitting multiple sclerosis, comparing dimethyl fumarate, glatiramer acetate, and placebo. RESULTS:Risk-dependent treatment effects from the prediction-NMR framework led to substantial variation in cost-effectiveness across the baseline risk distribution. When these treatment effects were incorporated into the cost-effectiveness model, incremental cost-effectiveness ratios increased steadily across baseline risk quintiles, from 48 811 Swiss Francs [CHF] (US $62 034) /QALY in the lowest-risk group to 212 870 CHF (US $270 536)/QALY in the highest. Dimethyl fumarate has a higher net monetary benefit up to a baseline risk of 55%, after which glatiramer acetate becomes the preferred option. CONCLUSIONS:Our findings show that integrating baseline risk modeling with NMA and cost-effectiveness analysis provides more informative decision-making than relying on average effects. Treatment value can vary substantially across the risk spectrum, indicating that optimal therapy selection is strongly dependent on individual patient risk.

BACKGROUND:Women with human immunodeficiency virus-1 (HIV) (WWH) have a higher cervical cancer risk than women without HIV. The timing and extent to which HIV viremia and immunodeficiency contribute to cervical carcinogenesis remain incompletely understood. METHODS:We conducted a cohort study using medical claims data from a South African HIV program (2011-2022) and calculated incidence rates of cervical precancer and cancer. Cox proportional hazards models assessed associations with CD4 cell count and HIV ribonucleic acid (RNA). We tested summary and point-in-time CD4 and HIV RNA measures with 6-36 months lag periods. Models were adjusted for age, calendar year, and antiretroviral therapy (ART) initiation; fully adjusted models included both CD4 count and HIV RNA. RESULTS:Over 66 000 WWH contributed more than 300 000 person-years; 1202 WWH developed moderate dysplasia (incidence rate: 394/100 000 person-years), 1237 severe dysplasia (405/100 000 person-years), 211 carcinoma in situ (66/100 000 person-years), and 257 cervical cancer (70/100 000 person-years). Lower CD4 counts were strongly associated with higher rates of cervical dysplasia, carcinoma in situ, and cancer, independent of HIV RNA. Lowest CD4 count over 30 months was the most informative measure for cervical dysplasia while CD4 count lagged by 24 months was most informative for cervical cancer. Higher HIV RNA was associated with increased risk of precancer and cancer in models unadjusted for CD4 count, with associations attenuated after adjustment. CONCLUSIONS:Maintaining high CD4 counts and achieving viral suppression through early ART initiation, along with using CD4 cell count for risk stratification in cervical screening, may help improve cervical cancer prevention among WWH in South Africa.

Abstract Female sex workers (FSW) in sub-Saharan Africa experience disproportionately high risks of HIV infection. Mathematical models are widely used to assess the contribution of sex workers and other key populations to HIV transmission dynamics and to inform targeted programmes. However, many rely on simplifying assumptions, such as stable sex worker characteristics and constant HIV transmission risk over time. These assumptions may be unrealistic and could bias modelled estimates. We used the South African Thembisa model to assess how alternative assumptions about FSW age, duration of sex work, and client-to-FSW transmission risk affect modelled HIV outcomes. We compared six scenarios that combined constant and increasing FSW age and sex work duration with constant and early-epidemic declining (exponentially or exposure-dependent) transmission risk. Each scenario was calibrated to HIV prevalence data from population-based and sex worker-specific surveys. Scenarios that allowed both FSW characteristics and transmission risk to vary over time showed the best agreement with external data, most closely reproducing HIV incidence, prevalence, and viral suppression estimates from a 2019 national sex worker survey (incidence ∼5 per 100 person-years, prevalence 61-62%, viral suppression ∼60%), and producing incidence rate ratios more consistent with estimates from the broader eastern and southern Africa region. By contrast, the scenario assuming constant FSW characteristics and transmission risk overestimated HIV incidence and underestimated prevalence and viral suppression. At the same time, this time-invariant specification attributed a much larger share of new HIV infections to sex work, with commercial sex work accounting for more than 20% of new infections in 2025, compared with 9-13% under time-varying assumptions. Overall, our findings show that HIV model estimates for sex workers are highly sensitive to modelling assumptions. Incorporating time-varying FSW parameters yields estimates that are more consistent with empirical data and support more reliable programme planning and evaluation. Author Summary Female sex workers in sub-Saharan Africa face much higher risks of HIV infection than other women. Mathematical models are often used to understand why and to guide prevention programmes. Yet many of these models make simple assumptions about sex workers - for example, that their average age stays the same over time, that they spend a fixed number of years in sex work, or that the chance of HIV passing from a client to a sex worker never changes. In reality, these factors changed over time. In this study, we used South Africa’s national HIV model to test how changing these assumptions affects the results. We compared different versions of the model and checked which ones best matched national sex worker survey data. We found that the model worked better when we allowed sex workers to become older over time, to spend longer in sex work, and the risk of passing on HIV to decline. Our findings show that mathematical models can give very different answers depending on how they represent the lives and experiences of sex workers. More realistic assumptions lead to more accurate estimates and can help ensure that programmes focus support where it is most needed.

Cervical cancer is the leading cause of cancer death among women in Zambia, where widespread HIV infection elevates cervical cancer risk. However, HIV’s contribution to Zambia’s cervical cancer burden and the extent to which antiretroviral therapy (ART) has mitigated this burden, remain poorly quantified. We adapted HPVsim, an existing open-source mathematical model simulating human papillomavirus (HPV) and cervical cancer natural history, to capture the dual burden of cervical cancer and HIV in Zambian women. We extended HPVsim by incorporating country-specific HIV incidence data and modeled how HIV alters cervical disease progression through CD4 cell count-dependent modification of HPV susceptibility and disease severity. We calibrated the model to GLOBOCAN cervical cancer estimates and cervical cancer incidence rate ratios by HIV status estimated from the Zambian Cancer Registry. We estimated that by 2025, 49.9% of all cervical cancer cases in Zambia could be attributed to HIV co-infection. Although ART prevented at least an estimated 5.4% of cervical cancer cases nationally and 8.7% of cases among women with HIV, HIV remains a major driver of the cervical cancer burden. HIV-stratified prevention strategies will be essential for Zambia to achieve the World Health Organization’s cervical cancer elimination targets.

OBJECTIVES:To evaluate the diagnostic performance, particularly specificity, of the MucorGenius® PCR assay for detecting pulmonary mucormycosis in bronchoalveolar lavage fluid (BALF) samples from at-risk patients. METHODS:This is a retrospective diagnostic accuracy study using prospectively collected BALF samples. All consecutive BALF samples obtained from patients who were considered to be at risk for invasive mould infections (IMIs) were prospectively collected. Patients were retrospectively classified according to European Organization for Research and Treatment of Cancer and Mycoses Study Group Education and Research Consortium and adapted Invasive Fungal Diseases in Adult Patients in Intensive Care Unit definitions. All samples were retrospectively tested with the MucorGenius® assay. RESULTS:A total of 1407 BALF samples obtained from 1330 patients had been included and tested for Mucorales DNA. A total of 256 patients (19.6%) fulfilled European Organization for Research and Treatment of Cancer and Mycoses Study Group Education and Research Consortium host factors and 664 (49.9%) Invasive Fungal Diseases in Adult Patients in Intensive Care Unit host factors. Proven or probable pulmonary mucormycosis was routinely diagnosed (without Mucorales PCR) in four patients (0.3%), 26 patients had proven or probable invasive pulmonary aspergillosis (IPA) (2%), and 25 (1.9%) had possible IMI. Overall, 32 positive MucorGenius® results had been observed. Per patient the MucorGenius® assay gave a specificity of 98.6% (95% CI: 97.8-99.2; n = 1251/1269) and a sensitivity of 100% (95% CI: 39.8-100; n = 4/4). Two cases with IPA were routinely diagnosed with mixed Mucorales infection and three additional IPA cases had a positive MucorGenius® PCR. In total, 5 of 26 IPA cases were therefore diagnosed with a mixed mould infection. Six of the 25 possible IMI cases (24%) also turned out positive on MucorGenius® PCR. Nineteen Mucorales PCR-positive samples had been obtained from patients without any routinely diagnosed IMI. CONCLUSIONS:We observed a near-to-perfect specificity of the MucorGenius® assay in BALF, making diagnosis of pulmonary mucormycosis very likely in patients with a positive PCR. In addition, the test was able to identify mucormycosis in a relevant proportion of cases with possible IMI and probable IPA, potentially indicating otherwise missed mixed-mould infections.

Background:Major depressive disorder is one of the most common, burdensome, and costly psychiatric disorders worldwide in adults. Pharmacological and non-pharmacological treatments are available; however, because of inadequate resources, antidepressants are used more frequently than psychological interventions. Prescription of these agents should be informed by the best available evidence. Therefore, we aimed to update and expand our previous work to compare and rank antidepressants for the acute treatment of adults with unipolar major depressive disorder. Methods:We did a systematic review and network meta-analysis. We searched Cochrane Central Register of Controlled Trials, CINAHL, Embase, LILACS database, MEDLINE, MEDLINE In-Process, PsycINFO, the websites of regulatory agencies, and international registers for published and unpublished, double-blind, randomised controlled trials from their inception to Jan 8, 2016. We included placebo-controlled and head-to-head trials of 21 antidepressants used for the acute treatment of adults (≥18 years old and of both sexes) with major depressive disorder diagnosed according to standard operationalised criteria. We excluded quasi-randomised trials and trials that were incomplete or included 20% or more of participants with bipolar disorder, psychotic depression, or treatment-resistant depression; or patients with a serious concomitant medical illness. We extracted data following a predefined hierarchy. In network meta-analysis, we used group-level data. We assessed the studies' risk of bias in accordance to the Cochrane Handbook for Systematic Reviews of Interventions, and certainty of evidence using the Grading of Recommendations Assessment, Development and Evaluation framework. Primary outcomes were efficacy (response rate) and acceptability (treatment discontinuations due to any cause). We estimated summary odds ratios (ORs) using pairwise and network meta-analysis with random effects. This study is registered with PROSPERO, number CRD42012002291. Findings:We identified 28 552 citations and of these included 522 trials comprising 116 477 participants. In terms of efficacy, all antidepressants were more effective than placebo, with ORs ranging between 2·13 (95% credible interval [CrI] 1·89-2·41) for amitriptyline and 1·37 (1·16-1·63) for reboxetine. For acceptability, only agomelatine (OR 0·84, 95% CrI 0·72-0·97) and fluoxetine (0·88, 0·80-0·96) were associated with fewer dropouts than placebo, whereas clomipramine was worse than placebo (1·30, 1·01-1·68). When all trials were considered, differences in ORs between antidepressants ranged from 1·15 to 1·55 for efficacy and from 0·64 to 0·83 for acceptability, with wide CrIs on most of the comparative analyses. In head-to-head studies, agomelatine, amitriptyline, escitalopram, mirtazapine, paroxetine, venlafaxine, and vortioxetine were more effective than other antidepressants (range of ORs 1·19-1·96), whereas fluoxetine, fluvoxamine, reboxetine, and trazodone were the least efficacious drugs (0·51-0·84). For acceptability, agomelatine, citalopram, escitalopram, fluoxetine, sertraline, and vortioxetine were more tolerable than other antidepressants (range of ORs 0·43-0·77), whereas amitriptyline, clomipramine, duloxetine, fluvoxamine, reboxetine, trazodone, and venlafaxine had the highest dropout rates (1·30-2·32). 46 (9%) of 522 trials were rated as high risk of bias, 380 (73%) trials as moderate, and 96 (18%) as low; and the certainty of evidence was moderate to very low. Interpretation:All antidepressants were more efficacious than placebo in adults with major depressive disorder. Smaller differences between active drugs were found when placebo-controlled trials were included in the analysis, whereas there was more variability in efficacy and acceptability in head-to-head trials. These results should serve evidence-based practice and inform patients, physicians, guideline developers, and policy makers on the relative merits of the different antidepressants. Funding:National Institute for Health Research Oxford Health Biomedical Research Centre and the Japan Society for the Promotion of Science.Appeared originally in Lancet 2018; 391:1357-1366.

Metabolomic epidemiology has expanded rapidly, but publications often lack sufficient detail for readers to assess study design, analytical methods, sources of bias, and the robustness and reproducibility of findings. Existing reporting recommendations in epidemiology and metabolomics do not fully address the specific challenges that arise when these fields are combined. To improve the completeness and transparency of reporting, we developed the Strengthening the Reporting of Metabolomic Epidemiology (STROBE-MetEpi) statement, an extension of the original STROBE guidance for observational research. The STROBE-MetEpi checklist includes 31 items and subitems covering the Title, Abstract, Introduction, Methods, Results, Discussion, and Other Information sections of metabolomic epidemiology studies. This explanation and elaboration document is intended to complement the STROBE-MetEpi statement by explaining the rationale for each checklist item and providing published examples of transparent reporting. It applies to studies using metabolomic profiling to explore human health, but not to randomized trials, methodological studies, reviews, or multi-omics studies. As with previous STROBE explanation and elaboration documents, its purpose is to improve how studies are reported, not to prescribe how they should be conducted. The STROBE-MetEpi statement and this accompanying document should support authors, reviewers, editors, and readers in improving the reporting, appraisal, interpretation, and reproducibility of metabolomic epidemiology research.

BACKGROUND:Long-acting injectable cabotegravir and rilpivirine (LAI cabotegravir and rilpivirine) is recommended as maintenance therapy for people living with HIV who achieved viral suppression on oral antiretroviral therapy (ART). However, its effect on drug resistance evolution in resource-limited settings remains uncertain. We aimed to assess this effect under different roll-out strategies and explore key factors for resistance. METHODS:We extended the Modelling Antiretroviral Drug Resistance In South Africa (MARISA) model to assess the effect of introducing LAI cabotegravir and rilpivirine in South Africa from 2025 to 2045. The HIV transmission rate was recalibrated using incidence estimates from the Thembisa model. We considered three strategies: dolutegravir-based ART as per current guidelines; LAI cabotegravir and rilpivirine as a switching option for those with viral suppression; and LAI cabotegravir and rilpivirine for both switching and ART initiation. We assumed faster development of integrase strand transfer inhibitor (INSTI) resistance in individuals with viraemia (≥1000 copies/mL) on LAI cabotegravir and rilpivirine than those with viraemia on dolutegravir-based ART. We evaluated their effect on pre-treatment drug resistance (PDR) and transmitted drug resistance (TDR) for INSTI and rilpivirine, explored resistance mitigation strategies, and identified key uncertainties through one-at-a-time local and global sensitivity analyses. FINDINGS:By 2045, using LAI cabotegravir and rilpivirine for both switching and ART initiation was estimated to lead to higher INSTI resistance levels, with 30·6% INSTI PDR and 11·3% INSTI TDR at 30% LAI cabotegravir and rilpivirine coverage, compared with 14·0% INSTI PDR and 6·9% INSTI TDR with continued oral dolutegravir-based ART. Rilpivirine resistance under the LAI cabotegravir and rilpivirine initiation and switch strategy was estimated to increase to 41·0% PDR and 30·0% TDR, compared with 6·7% and 4·5% with oral dolutegravir-based ART. Resistance levels were similar for LAI cabotegravir and rilpivirine as a switching option for those with viral suppression. A course of oral ART after LAI cabotegravir and rilpivirine interruption (oral bridging) decreased the risk of drug resistance due to the long pharmacokinetic tail, resulting in 22·3% INSTI PDR and 26·0% rilpivirine PDR when using LAI cabotegravir and rilpivirine for both switching and ART initiation. Key factors influencing resistance levels were care disengagement rates, LAI cabotegravir and rilpivirine coverage, and INSTI mutation reversion rates. INTERPRETATION:LAI cabotegravir and rilpivirine roll-out in South Africa should be cautious and targeted. Resistance risks may be mitigated through pre-treatment resistance testing, close monitoring of treatment outcomes, and efforts to increase retention in care. FUNDING:US National Institutes of Health National Institute of Allergy and Infectious Diseases and the Swiss National Science Foundation.

BACKGROUND:Most research on human immunodeficiency virus-1 (HIV-1) viremia and cancer risk is from high-income countries. We evaluated the association between HIV-1 viremia and the risk of various cancer types among people with HIV (PWH) in South Africa. METHODS:We analyzed data from the South African HIV Cancer Match study, based on laboratory measurements from the National Health Laboratory Service and cancer records from the National Cancer Registry from 2004 to 2014. Using Cox proportional hazards models, we estimated hazard ratios (HR) for cancer incidence per unit increase in time-updated Log10 HIV-1 RNA viral load copies/mL. We created partially adjusted (sex, age, calendar year) and fully adjusted models (additionally including time-updated CD4 count). RESULTS:We included 2 770 200 PWH with 10 175 incident cancers; most common were cervical cancer (N = 2481), Kaposi sarcoma (N = 1902), breast cancer (N = 1063), and non-Hodgkin lymphoma (N = 863). Hazard ratios for the association of HIV-1 viremia and cancer risk changed after partial and full adjustment and were generally attenuated for infection-related cancers but tended to increase for infection-unrelated cancers. In the fully adjusted model, HIV-1 viremia was associated with an increased risk of Kaposi sarcoma (HR per unit increase in Log10 HIV-1 RNA viral load: 1.38; 95% confidence interval [CI], 1.35-1.42), leukemia (HR: 1.28; 95% CI, 1.13-1.45), non-Hodgkin lymphoma (HR: 1.24; 95% CI, 1.19-1.29), conjunctival cancer (HR: 1.19; 95% CI, 1.11-1.25), and colorectal cancer (HR: 1.11; 95% CI, 1.02-1.21). Associations with other cancer types were weaker or absent. CONCLUSIONS:Our findings underline the importance of sustained viral suppression for cancer prevention among PWH in South Africa.

The risk of Mycobacterium tuberculosis (Mtb) transmission can be high in crowded clinics. We developed a spatiotemporal model of airborne Mtb transmission based on the Wells-Riley equation. We collected environmental, clinical and person-tracking data in a South African clinic during COVID-19, when community or surgical masks were compulsory and ventilation was increased. We matched person movements with clinical records to identify the spatiotemporal location of infectious TB patients. We modeled the concentration of infectious doses (quanta) and estimated the individual risk of infection. Over five days, video sensors tracked 1,438 clinic attendees. CO2 levels were low (median 431 ppm, IQR 406 ppm-458 ppm); the quanta concentration was higher in the morning than in the afternoon, and highest in the waiting room. The estimated risk of infection per clinic attendee was 0.05% (80%-credible interval (CrI) 0.01%-0.06%). It increased with the number of close contacts with infectious patients and the time spent in the clinic, and was 1.3-fold (95%-CrI 1.2-1.4) higher in scenarios without mask use and 2.1-fold (95%-CrI 0.9-5.0) higher with pre-pandemic ventilation rates, emphasizing the importance of ventilation. Spatiotemporal modeling can identify high-risk areas and evaluate the impact of infection control measures in clinics.

Background:Tuberculosis (TB) is the leading cause of death among people with HIV and a major global health challenge. Subclinical cardiovascular manifestations of TB are poorly documented in high TB and HIV burden countries. Objectives:The purpose of this study was to quantify the prevalence of cardiovascular involvement in TB patients and investigate changes after completion of anti-TB treatment. Methods:HIV-positive and HIV-negative patients diagnosed with pulmonary TB between October 2022 and November 2023 were enrolled from 2 tertiary care hospitals in Zambia and South Africa. Standardized transthoracic echocardiography (TTE) was conducted at TB diagnosis and after 6 months of anti-TB treatment. Cross-sectional and longitudinal analyses assessed pericardial effusion, thickening, or calcification, with and without signs of pericardial constriction. Results:A total of 286 TB patients (218 [76%] men, 109 [38%] people with HIV, median age 35 years) underwent TTE at TB diagnosis, of whom 105 participants had a second TTE after completion of treatment. At TB diagnosis, 134 (47%) had pericardial effusions, 86 (30%) thickening, 7 (2%) calcifications, 103 (42%) signs of constriction, and 13 (12%) had definite diagnosis of constriction. After TB treatment, pericardial effusions (47% vs 16%, P < 0.001) and pericardial thickenings (30% vs 15%, P = 0.002) became less prevalent. Pericardial calcifications (2% vs 1%, P = 0.4), signs of constrictions (42% vs 38%, P = 0.4), and definite diagnosis of constriction (12% vs 14%, P = 0.8) were similar. Conclusions:Cardiac involvement is frequent in newly diagnosed TB patients. Early pericardial changes may be reversed with anti-TB treatment. Echocardiographic screening facilitates early detection and timely management of cardiovascular involvement in TB patients.

Various statistical and machine learning algorithms can be used to predict treatment effects at the patient level using data from randomized clinical trials (RCTs). Such predictions can facilitate individualized treatment decisions. Recently, a range of methods and metrics were developed for assessing the accuracy of such predictions. Here, we extend these methods, focusing on the case of survival (time-to-event) outcomes. We start by providing alternative definitions of the participant-level treatment benefit; subsequently, we summarize existing and propose new measures for assessing the performance of models estimating participant-level treatment benefits. We explore metrics assessing discrimination and calibration for benefit and decision accuracy. These measures can be used to assess the performance of statistical as well as machine learning models and can be useful during model development (i.e., for model selection or for internal validation) or when testing a model in new settings (i.e., in an external validation). We illustrate methods using simulated data and real data from the OPERAM trial, an RCT in multimorbid older people, which randomized participants to either standard care or a pharmacotherapy optimization intervention. We provide R codes for implementing all models and measures.

Introduction: Little is known about the clinical status of persons with HIV (PWH) who re-engage in care after an interruption. We evaluated the immunological and clinical characteristics of individuals re-engaging in care within the Swiss HIV Cohort Study. Methods: Participants who re-engaged in care after an interruption >14 months with a viral load ≥100 copies/mL were classified as having interrupted ART. We defined late re-engagement as re-engaging with a CD4 cell count of <350 cells/µL or a new CDC stage C disease. Linear and logistic regression models with restricted cubic splines were used to estimate the mean CD4 cell count at re-engagement and the probability of late re-engagement as a function of care interruption duration. Results: Of 14,864 participants with a median follow-up of 10.2 years (IQR 4.7-17.2 years), 2,768 (18.6%) interrupted care, of whom 1,489 (53.8%) re-engaged. Among those re-engaging, 62.3% had interrupted ART. For participants who interrupted ART, the mean CD4 count declined from 374 cells/µL (95% CI 358-391 cells/µL) before the interruption to 250 cells/µL (95% CI 221-281 cells/µL) among those re-engaging after 14 months, and to 185 cells/µL (95% CI 160-212 cells/µL) among those re-engaging after 60 months. The estimated risk of late re-engagement in care was 68.6% (95% CI 62.3-74.4%) for participants who interrupted ART for 14 months and 75.2% (95% CI 68.9-80.6%) for those who interrupted ART for 60 months. Conclusion: Although HIV care interruptions are not very common in Switzerland, the majority of PWH re-engaging after interrupting ART return with late-stage HIV.

BACKGROUND:In response to increasing resistance to non-nucleoside reverse transcriptase inhibitors, millions of people living with HIV have switched to dolutegravir-based antiretroviral therapy, so understanding the possible emergence of dolutegravir resistance is essential. We aimed to predict how dolutegravir resistance in South Africa will change over time. METHODS:For this modelling study, we used the Modelling Antiretroviral Drug Resistance in South Africa (MARISA) model, a deterministic compartmental model calibrated to reproduce the HIV-1 epidemic in South Africa from 2005 to 2035 using data from the International Epidemiology Databases to Evaluate AIDS collaboration and the literature. Key parameters for modelling dolutegravir-resistance evolution were acquisition rates of dolutegravir-resistance mutations, reversion rates of dolutegravir-resistance mutations, the effect of resistance to nucleoside reverse transcriptase inhibitors on dolutegravir-resistance acquisition, the effect of dolutegravir resistance on dolutegravir-treatment efficacy, the probability of transmitting dolutegravir drug-resistance mutations compared with the probability of transmitting wild-type HIV, and the proportion of people with virologic failure on dolutegravir-based antiretroviral therapy with detectable drug levels. Model outcomes were estimated transmitted dolutegravir resistance and estimated acquired dolutegravir resistance. FINDINGS:We estimated a substantial increase in the number of individuals on dolutegravir-based antiretroviral therapy after its introduction in 2020, increasing from 0 to approximately 7 million people (7·08-7·15) living with HIV on dolutegravir in 2035. We estimated the proportion of people living with HIV with viral suppression (ie, viral load <1000 copies per mL) on dolutegravir-based antiretroviral therapy to be 93% (uncertainty range 92·2-94·3) in 2035. We estimated that acquired dolutegravir resistance in people living with HIV on failing dolutegravir-based antiretroviral therapy would increase rapidly, from 18·5% (uncertainty range 12·5-25·4) in 2023 to 41·7% (29·0-54·0) in 2035. For transmitted dolutegravir resistance, we estimated an increase from 0·1% (0·0-0·2) in 2023 to 5·0% (1·9-11·9) in 2035. We estimated that resistance-mitigation strategies involving rapid switching to protease-inhibitor-based antiretroviral therapy could effectively reduce the increase in acquired dolutegravir resistance and slow the increase in transmitted dolutegravir resistance. INTERPRETATION:Although dolutegravir-based antiretroviral therapy maintains high virological suppression, acquired and transmitted dolutegravir resistance are likely to increase. This increase will likely be greater in settings where HIV RNA monitoring, genotypic-resistance testing, and options to switch antiretroviral therapy regimens are scarce. FUNDING:US National Institutes of Health National Institute of Allergy and Infectious Diseases, Swiss National Science Foundation, and University of Zurich Research Priority Program Evolution in Action.

멘델 무작위화(Mendelian randomization, MR) 연구는 조절 가능한 노출(modifiable exposure)이 건강결과에 미치는 인과효과(causal effect)를 더 잘 이해하게 해 주지만, 그 근거는 종종 보고가 불충분함으로 인하여연구 결과의 해석과 적용에 한계가 있을 수 있다. 보고지침은 흔히 무슨 연구를 하고 무엇을 발견했는지 독자가 쉽게 이해하도록 돕는다. STROBE-MR(관찰연구의 멘델 무작위화를 활용한 보고지침)은 MR 연구를 명확하고 투명하게 보고하도록 돕는다. STROBE-MR을 논문 작성에 활용하면 독자, 심사자, 학술지 편집인이 MR 연구의 보고 품질과 완성도를 평가하는 데 도움이 될 것이다. 이 글은 STROBE-MR 체크리스트 20개 항목의 의미와 근거를 설명하고, 각 항목마다 사례를 제시해 독자가 잘 이해할 수 있는 논문 작성법을 설명하려고 하였다.