Daniel T. Citron

Background: Antiretroviral therapy (ART) has greatly reduced HIV-related mortality and improved life expectancy in sub-Saharan Africa since the early 2000s. However, HIV-related mortality remains high. To guide intervention priorities, we used seven established HIV simulation models to identify where in the HIV care cascade deaths occur in Zimbabwe, Malawi, and South Africa. This multi-model approach provided a more comprehensive and robust assessment than could be achieved using any single modelling approach. Methods and findings: To assess alignment, each model generated key HIV metrics from 2000 to 2045, including total population, HIV prevalence, annual new infections, and proportions of people with HIV (PWH) who were diagnosed, on ART, and virally suppressed. HIV-related deaths were estimated annually and categorised by whether they occurred in individuals who were (1) undiagnosed, (2) diagnosed but had not yet started ART, (3) on ART, or (4) had interrupted ART. The models showed strong consistency with population estimates and HIV prevalence across the different settings. There was, however, variation in the absolute number of HIV-related deaths between models; in Zimbabwe in 2025, this ranged from 3,666 to 17,475; in Malawi from 4,580 to 17,835; and in South Africa from 39,470 to 114,968. In Zimbabwe, all models consistently indicated that PWH receiving ART made up the largest proportion of HIV-related deaths, though this proportion differed across models, ranging from 35% to 64%. This was followed by HIV-related deaths amongst those who had interrupted ART, ranging from 17% to 29%. Similar trends were seen in Malawi. In South Africa, a high proportion of HIV-related deaths occurred amongst people who had interrupted ART (23% to 74% in 2025) as well as those who were currently receiving ART (23% to 56%). Models are reliant on empirical data and limited data availability—in this instance, lack of national death registries—is a key constraint on modelling studies. Conclusion: Absolute numbers of HIV-related deaths vary substantially between models, highlighting wider issues around ascertainment of death in the region. However, our results do consistently show that most deaths are occurring amongst PWH on ART in Zimbabwe and Malawi, and amongst those who have interrupted treatment in South Africa, suggesting interventions around adherence counselling and retention in care need to be prioritised.

Background: In eastern and southern Africa, HIV and depression lead to substantial morbidity and mortality. Each condition affects the other: people living with HIV (PLHIV) have higher depression rates, while depression increases HIV acquisition and hinders treatment. The combined impact of HIV-depression interactions on HIV and mental health remains poorly understood. Methods: We adapted the EMOD-HIV simulation model to incorporate depression incidence, recovery and relapse. We modelled a bidirectional scenario between HIV and depression: HIV increased depression incidence; depression increased HIV incidence and reduced HIV care outcomes. In a counterfactual scenario, all HIV-depression interactions were removed. We estimated how interactions impacted depression episodes, HIV acquisitions, HIV-related deaths and HIV treatment benefits in a high-prevalence region of Western Kenya, 1985–2035. Results: Without interactions, the model projected 1.23 million (95% CI 1.21M to 1.24M) HIV acquisitions, 658 000 (95% CI 650 000 to 666 000) HIV deaths and 12.88 million (95% CI 12.87M to 12.89M) depression episodes. HIV-depression interactions increased depression episodes by 9.76% (95% CI 9.67% to 9.87%), HIV acquisitions by 12.3% (95% CI 12.0% to 12.6%) and HIV deaths by 12.5% (95% CI 12.2% to 12.9%). These interactions also diminished HIV treatment benefits, with 7.72% (95% CI 7.39% to 8.07%) fewer HIV acquisitions and 3.88% (95% CI 3.71% to 4.05%) fewer deaths averted per person-year on treatment. In secondary analyses, increased depression incidence among PLHIV had the greatest effect on HIV outcomes, contributing 7.26% of acquisitions and 7.55% of deaths. Conclusion: HIV-depression interactions have worsened both epidemics. Depression’s effects on HIV incidence have been especially deleterious, while its effect on HIV care engagement is increasingly important.

Background: Differentiated service delivery (DSD) is increasingly available for HIV treatment. DSD has been shown to improve treatment retention, but DSD modalities incur higher costs than the clinic-based standard-of-care (SoC). We conducted a cost-effectiveness (CE) analysis to assess what DSD modalities, in what settings, would constitute an efficient use of limited HIV program resources. Methods: We adapted two validated mathematical models (EMOD-HIV and HIV Synthesis) to project HIV trends (incidence, prevalence, mortality), disability-adjusted life years (DALYs), and costs (2021 USD) arising from DSD versus SoC over 2022–2062 in four settings: South Africa, Malawi, Zambia, and a collective representation of African low- and middle-income countries (LMICs). We compared three DSD modalities: healthcare worker-managed community adherence groups (CAG), client-managed urban adherence group (UAG), and home ART delivery (HomeART). We calculated incremental cost-effectiveness ratios (ICERs) of DSD versus SoC from the health system perspective using country-specific CE thresholds, and performed one-way sensitivity analyses for key assumptions. Results: Community adherence groups (ICER: $274–$604/DALY averted) and UAG (ICER: $590–$720/DALY averted) were cost-effective for all country/model settings. HomeART was dominated by UAG in all settings. In nearly all settings, CE estimates of CAG were robust to uncertainty in DSD effectiveness (except Zambia), DSD costs, CE threshold (except South Africa), HIV-associated disability weights, and discount rates. Cost-effectiveness of UAG was highly sensitive to uncertainty in DSD effectiveness in all settings. Conclusions: Community adherence groups and UAG can provide cost-effective alternatives to the clinic-based SoC in multiple African settings.

Background: Of countries with high HIV prevalence, Zambia had the largest proportion of funding to its HIV program from the United States President's Emergency Plan for AIDS Relief (PEPFAR)—84% at the start of 2025. Abrupt withdrawal of bilateral aid in January 2025 disrupted HIV services. This study aimed to estimate the health and epidemiological consequences of the disruptions, and to what extent impacts could be mitigated by restoring services. Methods: We leveraged a previously developed HIV agent-based network transmission model, Epidemiological MODeling software for HIV, calibrated to Zambian HIV data at the provincial level. Health authorities leading the Zambian HIV program identified data and assumptions regarding impacts of aid disruptions by province and associated uncertainty ranges. We simulated disruptions lasting 3 months, 1 year, 4 years, or unabated, versus a counterfactual of no disruptions, over 2025–2060. Outcomes included additional HIV infections, deaths, and prevalence. Results: Unabated disruptions added 3.3 million HIV acquisitions (8.8x more than no disruption) and 1.6 million HIV deaths (5.3x), with the largest number among women (1.5 million acquisitions, 790 933 deaths) and the largest proportional increase among children (21.6x acquisitions, 20.8x deaths). Restoring services within 3 months would limit additional acquisitions to 54 863 (+13.1%) and additional deaths to 32 550 (+8.7%). HIV prevalence would increase by 4.5x if disruptions were unabated through 2060, but would not change (0.0x) if services were restored within 3 months. Conclusions: Rapid restoration of HIV services disrupted by the 2025 bilateral aid withdrawal could save >1.5 million lives and prevent epidemic resurgence in Zambia.

In the first year of the COVID-19 pandemic, digital contact tracing smartphone applications were developed for the purpose of reducing transmission by encouraging individuals to voluntarily quarantine after exposure. Privacy protocols used in many of these apps prevented longitudinal linkage of users’ contact network data, making it impossible to observe whether users reduced social contact following digital exposure notification. We used anonymous data from one particular app deployment, which used alternative privacy measures and retained longitudinal interaction data, to evaluate changes in the users’ time in proximity to other users after digital exposure notification. We analyzed a dataset of the anonymous app users in a US university community January − March 2021, before COVID-19 vaccines were widely available. Users provided consent for their anonymous data to be used for research. We conducted pre-post analysis of contact time with other users before and after using the app to self-report being a positive COVID-19 case, and before and after exposure notification by the app. We further conducted a retrospective case-control study by matching users who did and who did not receive exposure notification based on having similar contact time with other users prior to notification. In pre-post analysis, we found that users who reported being a positive case to the app reduced their total contact time with others by 29% (p = 0.002) in the 7-day period immediately after time of the report vs. the 7-day period before the time of the report. However, we observed no statistically significant change in contact time before vs. after exposure notification. We also observed no statistically significant difference in contact time compared to matched controls who received no notification. Our findings suggest that digital contact tracing app users affiliated with this US university did not reduce contact with others after receiving exposure notification. Our study provides one of the first rigorous evaluations of the effectiveness of digital exposure notifications during a time when communities were reliant on non-pharmaceutical interventions to reduce disease spread.

Background: Malaria elimination is the goal for Bioko Island, Equatorial Guinea. Intensive interventions implemented since 2004 have reduced prevalence, but progress has stalled in recent years. A challenge for elimination has been malaria infections in residents acquired during travel to mainland Equatorial Guinea. The present article quantifies how off-island contributes to remaining malaria prevalence on Bioko Island, and investigates the potential role of a pre-erythrocytic vaccine in making further progress towards elimination. Methods: Malaria transmission on Bioko Island was simulated using a model calibrated based on data from the Malaria Indicator Surveys (MIS) from 2015 to 2018, including detailed travel histories and malaria positivity by rapid-diagnostic tests (RDTs), as well as geospatial estimates of malaria prevalence. Mosquito population density was adjusted to fit local transmission, conditional on importation rates under current levels of control and within-island mobility. The simulations were then used to evaluate the impact of two pre-erythrocytic vaccine distribution strategies: mass treat and vaccinate, and prophylactic vaccination for off-island travellers. Lastly, a sensitivity analysis was performed through an ensemble of simulations fit to the Bayesian joint posterior probability distribution of the geospatial prevalence estimates. Results: The simulations suggest that in Malabo, an urban city containing 80% of the population, there are some pockets of residual transmission, but a large proportion of infections are acquired off-island by travellers to the mainland. Outside of Malabo, prevalence was mainly attributable to local transmission. The uncertainty in the local transmission vs. importation is lowest within Malabo and highest outside. Using a pre-erythrocytic vaccine to protect travellers would have larger benefits than using the vaccine to protect residents of Bioko Island from local transmission. In simulations, mass treatment and vaccination had short-lived benefits, as malaria prevalence returned to current levels as the vaccine’s efficacy waned. Prophylactic vaccination of travellers resulted in longer-lasting reductions in prevalence. These projections were robust to underlying uncertainty in prevalence estimates. Conclusions: The modelled outcomes suggest that the volume of malaria cases imported from the mainland is a partial driver of continued endemic malaria on Bioko Island, and that continued elimination efforts on must account for human travel activity.

Activity Space Maps are a novel global-scale movement and mobility data set which describes how people distribute their time through geographic space. The maps are intended for use by researchers for the purposes of epidemiological modeling. Activity Space Maps are designed to complement existing digitally-collected mobility data sets by quantifying the amount of time that people spend in different locations. This information is important for estimating the duration of contact with the environment and the potential risk of exposure to disease. More concretely, the type of information contained in Activity Space Maps will make it easier to model the spatial transmission patterns of vector-borne diseases like malaria and Dengue fever. We will discuss the motivation for designing Activity Space Maps, how the maps are generated from mobile phone user app location history data, and discuss an example use case demonstrating how such data may be used together with spatial epidemiological data to advance our understanding of spatial disease patterns and the relationship between travel behaviors and infection risk.

Background Malaria elimination is the goal for Bioko Island, Equatorial Guinea. Intensive interventions implemented since 2004 have reduced prevalence, but progress has stalled in recent years. A challenge for elimination has been malaria infections in residents acquired during travel to mainland Equatorial Guinea. We quantify how off-island contributes to remaining malaria prevalence on Bioko Island, and investigate the potential role of a pre-erythrocytic vaccine in making further progress towards elimination. Methods We simulated malaria transmission on Bioko Island using a model calibrated based on data from the Malaria Indicator Surveys (MIS) from 2015-2018, including detailed travel histories and malaria positivity by rapid-diagnostic tests (RDTs), as well as geospatial estimates of malaria prevalence. Mosquito population density was adjusted to fit local transmission, conditional on importation rates under current levels of control and within-island mobility. We evaluated the impact of two pre-erythrocytic vaccine distribution strategies: mass treat and vaccinate, and prophylactic vaccination for off-island travelers. We propagated uncertainty through the model through an ensemble of simulations fit to the Bayesian joint posterior probability distribution of the geospatial prevalence estimates. Results The simulations suggest that in Malabo, an urban city containing 80\% of the population, there are some pockets of residual transmission, but a large proportion of prevalence is attributable to malaria importation by travelers. Outside of Malabo, prevalence was mainly attributable to local transmission. We assess the uncertainty in the local transmission vs. importation to be lowest within Malabo and highest outside. Using a pre-erythrocytic vaccine to protect travelers would have larger benefits than using the vaccine to protect residents of Bioko Island from local transmission. In simulations, mass treatment and vaccination had short-lived benefits, as malaria prevalence returned to current levels as the vaccine's efficacy waned. Prophylactic vaccination of travelers resulted in longer-lasting reductions in prevalence. These projections were robust to underlying uncertainty in prevalence estimates. Conclusions The modeled outcomes suggest that the volume of malaria cases imported from the mainland is a partial driver of continued endemic malaria on Bioko Island, and that continued elimination efforts on must account for human travel activity.