Using an HIV transmission model in 30 states and Washington DC, we simulated ending Ryan White services from Parts C and D and the Ending the HIV Epidemic and Minority AIDS initiatives in February 2026. We projected 23 883 additional infections by 2030 (95% credible interval 2812-53 813) - a 17.6% (2-39.5%) excess. A 'conservative' estimate of suppression losses projected 8.1% (4.4-12.3%) more infections. Ryan White services are critical to preventing US HIV transmission.

BACKGROUND:People with HIV disproportionately experience homelessness and other forms of housing instability. Understanding how housing status influences HIV clinical outcomes remains limited, especially among individuals engaged in care. METHODS:We evaluated associations between housing status and HIV care markers using clinical data and patient-reported measures from the Centers for AIDS Research Network of Integrated Clinical Systems (CNICS) cohort of adults receiving HIV care across the US, between 2019-2025. We used relative risk regression and linear regression to estimate associations between self-perceived housing instability and outcomes including HIV viral suppression (<200 copies/mL), self-reported antiretroviral therapy (ART) use and adherence (visual analog scale), and CD4 cell count, adjusted for demographic characteristics, year, and site. RESULTS:Among 6,873 individuals in clinical care, nearly 10% reported some form of housing instability at their most recent visit: 4.6% "Unstable," 3.3% "Homeless," and 1.7% "Don't know." Compared to stably housed individuals, the prevalence of viral suppression was lower among those experiencing homelessness (PR=0.82, 95% CI: 0.77-0.89) and unstable housing (PR=0.94, 95% CI: 0.90-0.98), as was the prevalence of ART use, mean ART adherence, and mean CD4 count. Results were similar when stratified by substance use and depression and when compared to a community-based cohort of people with HIV. CONCLUSION:Housing instability is associated with a lower prevalence of HIV viral suppression even among individuals engaged in clinical care. These results highlight lack of access to stable housing as a structural barrier to successful management of HIV and to ending the HIV epidemic in the US.

Objectives To understand the trade-offs between different statistical modeling approaches, using real world data with small sub-populations, with rare exposures and increasingly rare outcomes. In particular, to compare adjusted regression, inverse probability weighting, and matching. Methods Data for these analyses came from the RADAR (N=1,134) and combined CNICS/JHHCC (N=14,434) cohorts. We estimated prevalence ratios (PRs) for self-reported use of specific substances comparing subpopulations (SP), SP-1 vs SP-3 and SP-2 vs SP-3, where SP-1 was the largest proportion (92% in RADAR, 81% in CNICS/JHHCC), SP2 was moderate proportion (18% in CNICS/JHHCC) and SP-3 was the smallest proportion (8% in RADAR, 1% in CNICS/JHHCC) of the population. We calculated PRs using 1) unadjusted relative risk regression (RR); and adjusted estimates controlling for age, race/ethnicity, study site, and year of interview using: 2) standard adjustment in RR; 3) stabilized inverse probability of treatment weighting (IPTW); and 4) matching with up to 3 matches from SP-1 or SP-2 per SP-3 participant. Results For most substances, all methods yielded consistent estimates. There were large weights in some of the IPTW analyses and in three cases these resulted in substantially divergent estimates. For the comparison between SP-1 and SP-3, the estimate for smoking was 1.3-fold greater in the matched analysis (PR=1.33, 95% CI: 1.02-1.75) than in IPTW (PR=1.03, 95% CI: 0.78-1.37 ATE and PR=1.05, 95% CI: 0.88-1.26 ATT). Even more extreme divergence in estimates was observed for differences between SP-2 and SP-3 with respect to methamphetamine/amphetamine, (IPTW-ATE: PR=1.51, 95% CI: 0.79-2.89; IPTW-ATT: PR=2.36, 95% CI: 1.36-4.12); vs Matching: PR=2.71, 95%CI: 1.47-4.99) and cocaine (IPTW-ATE: PR=1.15, 95% CI: 0.55-2.38; IPTW-ATT: PR=1.81, 95% CI: 1.03-3.18); vs Matching: PR=1.38, 95%CI: 0.76-2.53). Conclusion The combination of a rare exposure and a rare outcome can produce challenges for commonly used confounding adjustment strategies, and it is often best to compare different modeling approaches to gain greater insight.

BACKGROUND:Pain is common in persons with HIV, who often report mental health symptoms and substance use, all of which may affect viral suppression. OBJECTIVE:To investigate correlates of self-reported pain and associations with viral suppression among people with HIV. METHODS:We analysed data from 3422 self-interviews contributed by 1626 unique patients in continuity care in the Johns Hopkins HIV Clinical Cohort between October 2013 and December 2018. Self-reported pain was measured using the EuroQol-5D-3L ("Have you had any pain or discomfort today?"). The correlations included PTSD, anxiety, depressive symptoms, and substance use. Viral suppression (HIV RNA ≤ 200 copies/mL) is routinely collected for clinical care and was abstracted between 9 months prior to 1 week after the self-interview. RESULTS:Pain was more prevalent in patients who were older, female, white, and who reported mental health symptoms or who had a recent or past prescription for an opioid. The prevalence ratio between pain and viral non-suppression was consistent with no association and also a weak positive association (1.07, 95% CI: 0.95, 1.21). CONCLUSIONS:Tailored interventions for mental health may improve pain and HIV outcomes.

OBJECTIVE:Evaluate the impact of initiating integrase strand transfer inhibitor (INSTI) - vs. nonnucleoside reverse transcriptase inhibitor (NNRTI) - or protease inhibitor-based regimens on weight and glycemia among people with HIV (PWH) and type 2 diabetes. DESIGN:Cohort study. METHODS:From multisite HIV cohorts in the United States and Canada (2007-2022), we identified PWH with diabetes who newly initiated INSTI-based, NNRTI-based, or protease inhibitor-based therapy. We used inverse probability of treatment-weighted (IPTW) generalized linear models to compare changes in weight and hemoglobin A1c (HbA1c) at approximately 12 months after antiretroviral therapy (ART) initiation. We assessed time to at least 5% weight gain and to glucose-lowering therapy augmentation or HbA1c increase at least 0.5 percentage points with IPTW Cox regression. RESULTS:Among 1279 PWH with diabetes, 548 initiated an INSTI-based regimen, 511 an NNRTI-based regimen, and 220 a protease inhibitor-based regimen. Compared with NNRTI users, INSTI users had greater adjusted mean weight gain [2.1 kg; 95% confidence interval (CI), 1.1-3.1], while the difference in HbA1c change was modest (0.2%; 95% CI, 0.0-0.5); both changes were similar between INSTI and protease inhibitor users. INSTI users had higher risk of at least 5% weight gain [adjusted hazard ratio (aHR), 1.35; 95% CI, 1.15-1.59] and glucose-lowering therapy augmentation or HbA1c increase at least 0.5% (aHR, 1.19; 95% CI, 1.02-1.41) compared with NNRTI users although similar risk vs. protease inhibitor users. CONCLUSION:Among PWH with diabetes, initiating INSTIs conferred modestly greater weight gain and worse glycemic outcomes vs. NNRTIs, but outcomes comparable to PIs, which is recognized for adverse metabolic effects. These findings inform the metabolic implications of INSTIs and support clinical monitoring after ART initiation.

Identifying people with prior HIV care experience in clinical cohorts requires historical HIV data. We examined the availability of historical antiretroviral therapy (ART) prescriptions, historical CD4 counts, historical HIV viral loads, suppressed viral load at enrollment, and historical AIDS defining conditions in United States-based clinical cohorts in the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD). At enrollment, we classified participants as new to care, transfer (without a gap in care), re-engaging in care (transfer after a gap in care), or previously in care with unknown transfer status. From 2000 to 2022, transfers increased from 30% to 68% while new to care declined from 31% to 21%. Historical ART prescriptions and suppressed viral load at enrollment became increasingly common data types for identifying prior HIV care, specifically after 2012. Among participants not new to care with ≥1 known historical HIV date, the median time between earliest known HIV date and enrollment increased from 1.9 years in 2000 to 5.6 years in 2022. While some data types may serve as practical proxies, we recommend that HIV cohorts systematically collect and assess all 5 data types to improve identification of participants new to care and adopt our algorithm specific to their context.

Objectives. To examine the associations of routinely collected measures of socioeconomic disadvantage with HIV care continuum outcomes in 2024. Methods. In a cohort of people with HIV in Baltimore, Maryland, we compared restricted mean time spent in 4 states of the longitudinal care continuum (viral suppression, nonsuppression, having a 12-month gap in viral load monitoring, and death) by income, housing status, incarceration, insurance, and neighborhood deprivation. Results. All measures of socioeconomic disadvantage were associated with fewer days spent alive and more days with a nonsuppressed viral load. Individual income and neighborhood deprivation were independently associated with survival and viral suppression: among participants in low-deprivation neighborhoods, those with lower income spent 7 more days with a nonsuppressed viral load (95% confidence interval [CI] = 0, 16) and 2 fewer days alive (95% CI = 0, 4); among participants with higher income, those living in higher-deprivation neighborhoods spent 14 more days with a nonsuppressed viral load (95% CI = 4, 24) and 2 fewer days alive (95% CI = 0, 5). Conclusions. The routine collection of data on socioeconomic disadvantage can help assess risk of poor clinical outcomes among persons with HIV. (Am J Public Health. Published online ahead of print August 13, 2026:e1-e10. https://doi.org/10.2105/AJPH.2026.308484).

Although data may capture continuous event times or event times with high resolution (e.g., day), some statistical analyses require the discretization of time into intervals and assigning each event (i.e., outcome or loss to follow-up [LTFU]) to the start or end of an interval. First, using a simulated example, we showed that outcomes should be assigned to the end of the interval. Next, we considered four approaches for assigning LTFU events in a simulated example and in 20 real datasets. Comparing the resulting cumulative risk curves with the curve using continuous time, one approach always had the least error: assigning LTFU to the start or end of the interval, depending on which was closest to the continuous event time. This approach was superior to always censoring at the beginning or end of the interval.

BACKGROUND:Integrase strand transfer inhibitor (INSTI) initiation has been associated with diabetes in antiretroviral therapy (ART)-naive people with HIV. We aimed to examine the effect of switching to INSTIs on incident diabetes in ART-experienced people with HIV. METHODS:In this target trial emulation, we retrospectively used individual-level data from 27 longitudinal cohorts of people with HIV in the USA and Canada. We included participants aged at least 18 years without diabetes who had used non-nucleoside reverse transcriptase inhibitors (NNRTIs) or protease inhibitors for at least 180 days (in 2016-22) but had never used an INSTI. We used data from any clinical encounters in which participants continued an NNRTI or protease inhibitor versus switched to an INSTI, with a follow-up period of up to 5 years. The effect of switching to INSTIs on incident diabetes was estimated with weighted Cox regression with robust variance. We further assessed whether the effect varied by time since the switch and was explained by weight gain in the first year. FINDINGS:13 071 participants were followed up from 2702 encounters in which they switched to an INSTI from an NNRTI, 54 766 encounters in which they continued an NNRTI, 1714 encounters in which they switched to an INSTI from a protease inhibitor, and 26 599 encounters in which they continued a protease inhibitor. Switching from protease inhibitors to INSTIs conferred an adjusted hazard ratio (HR) of 1·38 (95% CI 1·06-1·80) for incident diabetes, whereas switching from NNRTIs to INSTIs conferred an adjusted HR of 1·10 (0·87-1·39). The diabetes risk was higher during the first 2 years after switching from protease inhibitors to INSTIs (HR 1·67, 95% CI 1·21-2·30), but not thereafter (1·08, 0·75-1·57; pinteraction=0·064). The effect of switching from protease inhibitors to INSTIs on diabetes did not appear to be explained by weight gain. In the sensitivity analysis in which weight gain did not exceed 5% in the first year after, the switch from NNRTIs to INSTIs had a HR of 1·03 (95% CI 0·79-1·36) and the switch from protease inhibitors to INSTIs had a HR of 1·37 (1·02-1·84). INTERPRETATION:The increased diabetes risk after switching from protease inhibitors to INSTIs highlights a metabolic implication of regimen change and could warrant close monitoring early after switch, regardless of weight gain. FUNDING:US National Institutes of Health.

Objectives. To estimate the increase in HIV infections in 11 US states if Ryan White services are interrupted or ended. Methods. We applied a population-level model of HIV transmission to 11 states. We represented the proportion of people with HIV receiving Ryan White AIDS drug assistance, outpatient health services, or support services, and simulated a loss of suppression in each category if services permanently end or return after delays of 1.5 or 3.5 years. Results. Cessation of Ryan White services in 2025 was projected to result in 69 695 additional infections from 2025 to 2030 (95% credible interval [CrI] = 18 943, 123 628), 68% more (95% CrI = 18%, 118%) than if Ryan White were continued. Temporary interruptions of 1.5 and 3.5 years resulted in 26 951 (95% CrI = 7341, 47 534) and 53 594 (95% CrI = 14 645, 94 860) additional infections, respectively. Excess infections varied across states, from a 45% increase in Texas to 126% in Missouri. Conclusions. Projected increases in HIV infections because of disruptions of Ryan White services threaten the progress made in curtailing the US HIV epidemic, illustrating the critical role Ryan White plays in preventing HIV transmission. (Am J Public Health. 2026;116(5):732-735. https://doi.org/10.2105/AJPH.2025.308409).

Plasma HIV-1 RNA viral loads (VLs) are measured via laboratory assays with changing lower limits of quantification over time. We described an approach to produce an analytic-ready dataset of VLs over time and across longitudinal cohorts of adults. A 3-step approach was used: (1) initial data cleaning, (2) data checking with visualization, and (3) final data cleaning. Assumptions, data-driven decisions, and information from cohort-specific data managers produce an analytic-ready dataset of VLs with minimal missing data for date of blood draw, HIV-1 RNA result (copies/mL), below the lower limit of quantification (BLLQ) indicator, and the lower limit of quantification (LLQ). Among 3 663 786 VLs from 186 990 participants (median number of VLs per participant = 12, interquartile range 4-27) measured from 1988 to 2021, 61% of VL records were harmonized via the 3-step approach. The proportion of VLs below the lower limit of quantification increased from 39% to 60% after application of this approach. Changes to LLQ, VL result, and BLLQ indicator variables were made to 45%, 36%, and 22% of VLs, respectively. Stated assumptions, visualized data distributions, and a documented approach to preparing an analytic-ready dataset of pooled individual-level longitudinal data revealed data idiosyncrasies, informed assumptions, and improved the data for research inference.

OBJECTIVE:To synthesize the available evidence and identify gaps in the literature on residential mobility and health among people with HIV. DESIGN:Scoping review. METHODS:We included original research articles, reviews, editorials and commentaries that addressed residential mobility among people with HIV and that were published in English or Spanish between 2010-2025. We extracted key information including the study population, study design, definition of residential mobility, measurement of residential mobility, health outcomes, and main results. RESULTS:We identified 17 original research articles and 7 review articles or commentaries meeting the inclusion criteria. Definitions of residential mobility and measurement approaches differed across the included studies. Only 8 studies measured contextual factors, such as the distance moved or reason for moving, and 5 studies compared residential mobility across subgroups of people with HIV. In 3 tracing studies, residential mobility was the most frequently reported reason for loss to clinic. Associations with other care continuum outcomes, including treatment initiation, adherence, viral suppression, and mortality, were mixed, highlighting the complex and context-dependent relationship between mobility and health among people with HIV. CONCLUSIONS:While residential mobility is recognized as an important driver of care continuum outcomes among people with HIV, further research is needed to elucidate the relevant causal pathways and identify opportunities for intervention.

BACKGROUND:Longitudinal data often include gaps in observation when outcomes (and other variables) are unmeasured due to missed study visits or dropout. We explore the fundamentals of data gaps and use simulation to compare approaches for handling data gaps when estimating outcome incidence. METHODS:We generated a simulation of 1000 individuals across 10 study visits. We used 4 data-generating mechanisms: (1) missingness was independent of the outcome; (2) there was a baseline common cause of missingness and the outcome; (3) there was a time-varying common cause; and (4) the outcome directly affected future missingness. We estimated the risk and rate of the first outcome occurrence (generated as a transient, repeated, and permanent outcome), using crude and adjusted approaches, across 1000 iterations, and compared bias and empirical standard error. RESULTS:Under Scenario 1, in crude analyses, results were unbiased when censoring before a data gap but not when allowing participants to return. Under scenarios 2-4, all crude approaches were biased. Inverse probability of censoring weights and multiple imputation were relatively unbiased across scenarios and outcome types; multiple imputation was more precise. Inverse probability of observation weights was biased when the outcome was permanent and was less precise than either of the other two approaches. CONCLUSION:Crude approaches allowing participants to return following a data gap are not recommended because they can be biased even when missingness and the outcome are independent. Instead, one should either censor or handle the data gap using multiple imputation.

BACKGROUND:Implementation of long-acting cabotegravir/rilpivirine (LA-CAB/RPV) has been challenging and may limit its uptake. We describe the number of people with HIV (PWH) prescribed LA-CAB/RPV at 9 US academic HIV clinics in the Center for AIDS Research Network of Integrated Clinical Systems and clinic-level variation in LA-CAB/RPV prescriptions. METHODS:Using a sequential explanatory mixed methods design, we analyzed clinical cohort data of PWH prescribed LA-CAB/RPV, then conducted key informant surveys to ascertain each clinic's Exploration-Preparation-Implementation-Sustainment phase in conjunction with implementation determinants mapped to the Consolidated Framework for Implementation Research. We stratified the cohort analysis by phase. RESULTS:Between 21-Jan-2021 and 30-Sep-2024, 1,451 (6%) of 22,379 PWH in care were prescribed LA-CAB/RPV. Among those prescribed, 15% had baseline viral load ≥200 copies/mL (200/1,451). Two sustainment-phase clinics that experienced fewer insurance-related barriers, planned prior to LA-CAB/RPV availability, had a pharmacy team coordinating LA-CAB/RPV services, and customized the electronic health record (EHR) system to track insurance approvals and/or injection appointments prescribed LA-CAB/RPV to 16% (924/5,638) of PWH in care. Four sustainment-phase clinics that had a mix of payor coverage and insurance plans, and adequate staffing prescribed to 3% (213/8,154). Three implementation-phase clinics with difficulty procuring LA-CAB/RPV due to restrictive insurance coverage and/or hospital policy and staffing challenges prescribed to 4% (314/8,587). CONCLUSIONS:Barriers related to insurance coverage and drug procurement impacted the number of PWH prescribed LA-CAB/RPV as did processes to plan for, and adapt to, these barriers. Scale-up efforts should consider team coordination and EHR-facilitated tracking to support the growing number of PWH on LA-CAB/RPV.

Background:Angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) are established antihypertensive treatments that reduce cardiovascular disease (CVD) risk. However, their comparative effectiveness in people with HIV (PWH) is not well examined. This study evaluated the comparative effectiveness of ACEIs and ARBs head-to-head and versus no antihypertensive treatment in preventing primary CVD. Methods:Using a target trial emulation framework and data from the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD), we estimated observational analogs of intention-to-treat (ITT) and per-protocol (PP) effects of antihypertensive treatments in preventing primary CVD (myocardial infarction, non-MI coronary artery disease, stroke, transient ischemic attack, peripheral vascular disease, cardiovascular death) among hypertensive PWH, with subgroup analyses for Black and White PWH. Results:Compared with no antihypertensive treatment, ACEIs and ARBs were both associated with lower CVD risk in PWH, with similar effect sizes in ITT and PP analyses (ACEI ITT adjusted hazard ratio (HR): 0.79, 95% CI [0.70-0.89]; ACEI PP: 0.71 [0.55-0.90]; ARB ITT: 0.87 [0.65-1.16]; ARB PP: 0.37 [0.18-0.76]). Race-stratified ITT and PP analyses suggested somewhat greater risk reductions in White than Black PWH, although differences were not statistically significant. In head-to-head comparisons, ACEIs and ARBs showed comparable effectiveness overall (ITT: 1.14 [0.84-1.55]; PP: 0.54 [0.25-1.18]), and within race strata. Conclusions:Our study found that both ACEIs and ARBs were effective in reducing CVD risk among PWH, with similar effectiveness observed for both medications. The analysis did not reveal statistically significant differences in effectiveness between Black and White PWH for either drug.

In this manuscript, we present the results of a series of workshops convened in conjunction with the 2023 Society for Epidemiologic Research annual meeting. The overall objective of the workshops was to develop a set of core competencies for PhD students in epidemiology. The topics presented in the list of competencies are organized using a framework similar to many graduate programs in epidemiology, proceeding from basic to advanced topics. Given the breadth of substantive topics in the fields of epidemiology and public health, this list of competencies focuses on methodologic topics that are relevant to all students, regardless of research interest. The final topic lists were developed based on discussions including a large and diverse group of epidemiologists with different areas of expertise. By creating this resource, we aim to facilitate training of future generations of epidemiologists.

Sequential nested trial (SNT) emulation is a powerful approach for maximizing precision and avoiding time-related biases. However, there exists little discussion about the implied causal estimands in comparison to a real-world single point trial. We used Monte Carlo simulation to compare treatment effect estimates from an SNT emulation that re-indexed patients annually and a SNT emulation with a treatment decision design to the estimates from a single point trial. We generated 5,000 cohorts of 5,000 people with 3 years of follow-up. For the single point trial, patients were randomized to initiate or not initiate treatment at Visit 1. For the SNT emulations, simulated patients could contribute up to two index dates. When disease severity did not modify the treatment effect, both SNT approaches returned treatment effect estimates identical to the single point trial. In the presence of treatment effect modification by disease severity, both SNT approaches returned treatment effect estimates that diverged from the single point trial even after confounding-adjustment. These findings underscore the difficulties of interpreting causal estimands from a SNT emulation: the target population does not correspond to a single time point trial. Such implications are important for communicating study results for evidence-based decision-making.

An impactful epidemiologic question is one that, if answered, could inform meaningful action to reduce the burden of disease in the population it concerns. We propose a set of factors that could be used for discussing, evaluating, and communicating the public health impact of epidemiologic studies. These factors pertain to the burden and distribution of disease, the potential for an intervention to alter the disease burden, and the context in which the study is conducted. The disease burden is characterized by the number of cases, severity or cost of disease, and distribution of disease across the population. The potential for intervention is characterized by the mutability of the exposure itself, the prevalence and distribution of other causes of the disease in the population, the prevalence of the exposure and risk of the outcome under the natural course (before any intervention), and the feasibility of intervening. An epidemiologic question need not be impactful along all these factors to make answering it worthwhile. However, answering epidemiologic questions with more of these factors present will likely have a greater public health impact than answering questions for which these factors are absent. We hope that collecting these factors into a single framework may aid students and senior epidemiologists alike when organizing arguments for the value of their own work or attempting to evaluate the impact of others' work.

Background:Availability of the hepatitis B virus (HBV) vaccine in the United States since 1982 and recommendations for universal/catch-up vaccination of infants and children since the 1990s may be associated with lower HBV prevalence among people with human immunodeficiency virus (HIV; PWH) born after 1980. Methods:Active HBV infection prevalence, defined as the proportion of patients with a positive hepatitis B surface antigen result, was assessed among PWH at entry into a clinical cohort. Patients were categorized into birth-year cohorts of 1940-1959, 1960-1979, or 1980-1999 and then further dichotomized into pre- and post-1980 birth-year cohorts. Log binomial regression was used to assess the association of birth-year cohort with hepatitis B surface antigen positivity, adjusting for race/ethnicity, HIV infection risk factor, baseline HIV viral load and CD4+ cell count, HBV active therapy, and year of/age at cohort entry. Results:Among 5598 PWH, most were male (67%) and Black (77%), with a mean age (SD) of 39.7 (9.6) years at cohort entry. Approximately a third of the participants (30%) identified as men who have sex with men, and 39% reported a history of injection drug use. At cohort entry, the majority had a CD4+ cell count <350/µL and a viral load >1000 copies/mL. The HBV prevalence was 6.7% overall but varied by birth-year cohort: 6.2% for 1940-1959, 7.9% for 1960-1979, and 2.6% for 1980-1999. The risk of HBV infection was lower in the post-1980 than in the pre-1980 cohort (adjusted prevalence ratio, 0.19 [95% confidence interval, .09- .42]). Conclusions:PWH born after 1980 had a lower prevalence of HBV coinfection than those born before 1980, supporting the potential impact of universal childhood HBV vaccination.