Transition to endemicity: Understanding COVID-19.

The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and its associated disease, coronavirus disease 2019 (COVID-19), has caused a devastating pandemic worldwide. Here, we explain basic concepts underlying the transition from an epidemic to an endemic state, where a pathogen is stably maintained in a population. We discuss how the number of infections and the severity of disease change in the transition from the epidemic to the endemic phase and consider the implications of this transition in the context of COVID-19.

Cryptic transmission of SARS-CoV-2 and the first COVID-19 wave.

Considerable uncertainty surrounds the timeline of introductions and onsets of local transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) globally1-7. Although a limited number of SARS-CoV-2 introductions were reported in January and February 2020 (refs.8,9), the narrowness of the initial testing criteria, combined with a slow growth in testing capacity and porous travel screening10, left many countries vulnerable to unmitigated, cryptic transmission. Here we use a global metapopulation epidemic model to provide a mechanistic understanding of the early dispersal of infections and the temporal windows of the introduction of SARS-CoV-2 and onset of local transmission in Europe and the USA. We find that community transmission of SARS-CoV-2 was likely to have been present in several areas of Europe and the USA by January 2020, and estimate that by early March, only 1 to 4 in 100 SARS-CoV-2 infections were detected by surveillance systems. The modelling results highlight international travel as the key driver of the introduction of SARS-CoV-2, with possible introductions and transmission events as early as December 2019 to January 2020. We find a heterogeneous geographic distribution of cumulative infection attack rates by 4 July 2020, ranging from 0.78% to 15.2% across US states and 0.19% to 13.2% in European countries. Our approach complements phylogenetic analyses and other surveillance approaches and provides insights that can be used to design innovative, model-driven surveillance systems that guide enhanced testing and response strategies.

Multimodeling approach to evaluating the efficacy of layering pharmaceutical and nonpharmaceutical interventions for influenza pandemics.

When an influenza pandemic emerges, temporary school closures and antiviral treatment may slow virus spread, reduce the overall disease burden, and provide time for vaccine development, distribution, and administration while keeping a larger portion of the general population infection free. The impact of such measures will depend on the transmissibility and severity of the virus and the timing and extent of their implementation. To provide robust assessments of layered pandemic intervention strategies, the Centers for Disease Control and Prevention (CDC) funded a network of academic groups to build a framework for the development and comparison of multiple pandemic influenza models. Research teams from Columbia University, Imperial College London/Princeton University, Northeastern University, the University of Texas at Austin/Yale University, and the University of Virginia independently modeled three prescribed sets of pandemic influenza scenarios developed collaboratively by the CDC and network members. Results provided by the groups were aggregated into a mean-based ensemble. The ensemble and most component models agreed on the ranking of the most and least effective intervention strategies by impact but not on the magnitude of those impacts. In the scenarios evaluated, vaccination alone, due to the time needed for development, approval, and deployment, would not be expected to substantially reduce the numbers of illnesses, hospitalizations, and deaths that would occur. Only strategies that included early implementation of school closure were found to substantially mitigate early spread and allow time for vaccines to be developed and administered, especially under a highly transmissible pandemic scenario.

Quantifying the importance and location of SARS-CoV-2 transmission events in large metropolitan areas.

Detailed characterization of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) transmission across different settings can help design less disruptive interventions. We used real-time, privacy-enhanced mobility data in the New York City, NY and Seattle, WA metropolitan areas to build a detailed agent-based model of SARS-CoV-2 infection to estimate the where, when, and magnitude of transmission events during the pandemic's first wave. We estimate that only 18% of individuals produce most infections (80%), with about 10% of events that can be considered superspreading events (SSEs). Although mass gatherings present an important risk for SSEs, we estimate that the bulk of transmission occurred in smaller events in settings like workplaces, grocery stores, or food venues. The places most important for transmission change during the pandemic and are different across cities, signaling the large underlying behavioral component underneath them. Our modeling complements case studies and epidemiological data and indicates that real-time tracking of transmission events could help evaluate and define targeted mitigation policies.

Finding influential subjects in a network using a causal framework.

Researchers across a wide array of disciplines are interested in finding the most influential subjects in a network. In a network setting, intervention effects and health outcomes can spill over from one node to another through network ties, and influential subjects are expected to have a greater impact than others. For this reason, network research in public health has attempted to maximize health and behavioral changes by intervening on a subset of influential subjects. Although influence is often defined only implicitly in most of the literature, the operative notion of influence is inherently causal in many cases: influential subjects are those we should intervene on to achieve the greatest overall effect across the entire network. In this work, we define a causal notion of influence using potential outcomes. We review existing influence measures, such as node centrality, that largely rely on the particular features of the network structure and/or on certain diffusion models that predict the pattern of information or diseases spreads through network ties. We provide simulation studies to demonstrate when popular centrality measures can agree with our causal measure of influence. As an illustrative example, we apply several popular centrality measures to the HIV risk network in the Transmission Reduction Intervention Project and demonstrate the assumptions under which each centrality can represent the causal influence of each participant in the study.

Evaluating the Mediating Role of Recall of Intervention Knowledge in the Relationship Between a Peer-Driven Intervention and HIV Risk Behaviors Among People Who Inject Drugs.

Peer-driven interventions can be effective in reducing HIV injection risk behaviors among people who inject drugs (PWID). We employed a causal mediation framework to examine the mediating role of recall of intervention knowledge in the relationship between a peer-driven intervention and subsequent self-reported HIV injection-related risk behavior among PWID in the HIV Prevention Trials Network (HPTN) 037 study. For each intervention network, the index participant received training at baseline to become a peer educator, while non-index participants and all participants in the control networks received only HIV testing and counseling; recall of intervention knowledge was measured at the 6-month visit for each participant, and each participant was followed to ascertain HIV injection-related risk behaviors at the 12-month visit. We used inverse probability weighting to fit marginal structural models to estimate the total effect (TE) and controlled direct effect (CDE) of the intervention on the outcome. The proportion eliminated (PE) by intervening to remove mediation by the recall of intervention knowledge was computed. There were 385 participants (47% in intervention networks) included in the analysis. The TE and CDE risk ratios for the intervention were 0.47 [95% confidence interval (CI): 0.28, 0.78] and 0.73 (95% CI: 0.26, 2.06) and the PE was 49%. Compared to participants in the control networks, the peer-driven intervention reduced the risk of HIV injection-related risk behavior by 53%. The mediating role of recall of intervention knowledge accounted for less than 50% of the total effect of the intervention, suggesting that other potential causal pathways between the intervention and the outcome, such as motivation and skill, self-efficacy, social norms and behavior modeling, should be considered in future studies.

Rapid review and meta-analysis of serial intervals for SARS-CoV-2 Delta and Omicron variants.

BACKGROUND: The serial interval is the period of time between symptom onset in the primary case and symptom onset in the secondary case. Understanding the serial interval is important for determining transmission dynamics of infectious diseases like COVID-19, including the reproduction number and secondary attack rates, which could influence control measures. Early meta-analyses of COVID-19 reported serial intervals of 5.2 days (95% CI: 4.9-5.5) for the original wild-type variant and 5.2 days (95% CI: 4.87-5.47) for Alpha variant. The serial interval has been shown to decrease over the course of an epidemic for other respiratory diseases, which may be due to accumulating viral mutations and implementation of more effective nonpharmaceutical interventions. We therefore aggregated the literature to estimate serial intervals for Delta and Omicron variants. METHODS: This study followed Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. A systematic literature search was conducted of PubMed, Scopus, Cochrane Library, ScienceDirect, and preprint server medRxiv for articles published from April 4, 2021, through May 23, 2023. Search terms were: ("serial interval" or "generation time"), ("Omicron" or "Delta"), and ("SARS-CoV-2" or "COVID-19"). Meta-analyses were done for Delta and Omicron variants using a restricted maximum-likelihood estimator model with a random effect for each study. Pooled average estimates and 95% confidence intervals (95% CI) are reported. RESULTS: There were 46,648 primary/secondary case pairs included for the meta-analysis of Delta and 18,324 for Omicron. Mean serial interval for included studies ranged from 2.3-5.8 days for Delta and 2.1-4.8 days for Omicron. The pooled mean serial interval for Delta was 3.9 days (95% CI: 3.4-4.3) (20 studies) and Omicron was 3.2 days (95% CI: 2.9-3.5) (20 studies). Mean estimated serial interval for BA.1 was 3.3 days (95% CI: 2.8-3.7) (11 studies), BA.2 was 2.9 days (95% CI: 2.7-3.1) (six studies), and BA.5 was 2.3 days (95% CI: 1.6-3.1) (three studies). CONCLUSIONS: Serial interval estimates for Delta and Omicron were shorter than ancestral SARS-CoV-2 variants. More recent Omicron subvariants had even shorter serial intervals suggesting serial intervals may be shortening over time. This suggests more rapid transmission from one generation of cases to the next, consistent with the observed faster growth dynamic of these variants compared to their ancestors. Additional changes to the serial interval may occur as SARS-CoV-2 continues to circulate and evolve. Changes to population immunity (due to infection and/or vaccination) may further modify it.

Incidence of SARS-CoV-2 infection and associated risk factors among staff and residents at homeless shelters in King County, Washington: an active surveillance study.

Homeless shelter residents and staff may be at higher risk of SARS-CoV-2 infection. However, SARS-CoV-2 infection estimates in this population have been reliant on cross-sectional or outbreak investigation data. We conducted routine surveillance and outbreak testing in 23 homeless shelters in King County, Washington, to estimate the occurrence of laboratory-confirmed SARS-CoV-2 infection and risk factors during 1 January 2020-31 May 2021. Symptom surveys and nasal swabs were collected for SARS-CoV-2 testing by RT-PCR for residents aged ≥3 months and staff. We collected 12,915 specimens from 2,930 unique participants. We identified 4.74 (95% CI 4.00-5.58) SARS-CoV-2 infections per 100 individuals (residents: 4.96, 95% CI 4.12-5.91; staff: 3.86, 95% CI 2.43-5.79). Most infections were asymptomatic at the time of detection (74%) and detected during routine surveillance (73%). Outbreak testing yielded higher test positivity than routine surveillance (2.7% versus 0.9%). Among those infected, residents were less likely to report symptoms than staff. Participants who were vaccinated against seasonal influenza and were current smokers had lower odds of having an infection detected. Active surveillance that includes SARS-CoV-2 testing of all persons is essential in ascertaining the true burden of SARS-CoV-2 infections among residents and staff of congregate settings.

Zika virus evolution and spread in the Americas.

Although the recent Zika virus (ZIKV) epidemic in the Americas and its link to birth defects have attracted a great deal of attention, much remains unknown about ZIKV disease epidemiology and ZIKV evolution, in part owing to a lack of genomic data. Here we address this gap in knowledge by using multiple sequencing approaches to generate 110 ZIKV genomes from clinical and mosquito samples from 10 countries and territories, greatly expanding the observed viral genetic diversity from this outbreak. We analysed the timing and patterns of introductions into distinct geographic regions; our phylogenetic evidence suggests rapid expansion of the outbreak in Brazil and multiple introductions of outbreak strains into Puerto Rico, Honduras, Colombia, other Caribbean islands, and the continental United States. We find that ZIKV circulated undetected in multiple regions for many months before the first locally transmitted cases were confirmed, highlighting the importance of surveillance of viral infections. We identify mutations with possible functional implications for ZIKV biology and pathogenesis, as well as those that might be relevant to the effectiveness of diagnostic tests.

Covariate-constrained randomization with cluster selection and substitution.

BACKGROUND: An ongoing cluster-randomized trial for the prevention of arboviral diseases utilizes covariate-constrained randomization to balance two treatment arms across four specified covariates and geographic sector. Each cluster is within a census tract of the city of Mérida, Mexico, and there were 133 eligible tracts from which to select 50. As some selected clusters may have been subsequently found unsuitable in the field, we desired a strategy to substitute new clusters while maintaining covariate balance. METHODS: We developed an algorithm that successfully identified a subset of clusters that maximized the average minimum pairwise distance between clusters in order to reduce contamination and balanced the specified covariates both before and after substitutions were made. SIMULATIONS: Simulations were performed to explore some limitations of this algorithm. The number of selected clusters and eligible clusters were varied along with the method of selecting the final allocation pattern. CONCLUSION: The algorithm is presented here as a series of optional steps that can be added to the standard covariate-constrained randomization process in order to achieve spatial dispersion, cluster subsampling, and cluster substitution. Simulation results indicate that these extensions can be used without loss of statistical validity, given a sufficient number of clusters included in the trial.

Designing effective control of dengue with combined interventions.

Viruses transmitted by Aedes mosquitoes, such as dengue, Zika, and chikungunya, have expanding ranges and seem unabated by current vector control programs. Effective control of these pathogens likely requires integrated approaches. We evaluated dengue management options in an endemic setting that combine novel vector control and vaccination using an agent-based model for Yucatán, Mexico, fit to 37 y of data. Our intervention models are informed by targeted indoor residual spraying (TIRS) experiments; trial outcomes and World Health Organization (WHO) testing guidance for the only licensed dengue vaccine, CYD-TDV; and preliminary results for in-development vaccines. We evaluated several implementation options, including varying coverage levels; staggered introductions; and a one-time, large-scale vaccination campaign. We found that CYD-TDV and TIRS interfere: while the combination outperforms either alone, performance is lower than estimated from their separate benefits. The conventional model hypothesized for in-development vaccines, however, performs synergistically with TIRS, amplifying effectiveness well beyond their independent impacts. If the preliminary performance by either of the in-development vaccines is upheld, a one-time, large-scale campaign followed by routine vaccination alongside aggressive new vector control could enable short-term elimination, with nearly all cases avoided for a decade despite continuous dengue reintroductions. If elimination is impracticable due to resource limitations, less ambitious implementations of this combination still produce amplified, longer-lasting effectiveness over single-approach interventions.

Inverse probability weighted estimators of vaccine effects accommodating partial interference and censoring.

Estimating population-level effects of a vaccine is challenging because there may be interference, that is, the outcome of one individual may depend on the vaccination status of another individual. Partial interference occurs when individuals can be partitioned into groups such that interference occurs only within groups. In the absence of interference, inverse probability weighted (IPW) estimators are commonly used to draw inference about causal effects of an exposure or treatment. Tchetgen Tchetgen and VanderWeele proposed a modified IPW estimator for causal effects in the presence of partial interference. Motivated by a cholera vaccine study in Bangladesh, this paper considers an extension of the Tchetgen Tchetgen and VanderWeele IPW estimator to the setting where the outcome is subject to right censoring using inverse probability of censoring weights (IPCW). Censoring weights are estimated using proportional hazards frailty models. The large sample properties of the IPCW estimators are derived, and simulation studies are presented demonstrating the estimators' performance in finite samples. The methods are then used to analyze data from the cholera vaccine study.

Spillover benefit of pre-exposure prophylaxis for HIV prevention: evaluating the importance of effect modification using an agent-based model.

We developed an agent-based model using a trial emulation approach to quantify effect measure modification of spillover effects of pre-exposure prophylaxis (PrEP) for HIV among men who have sex with men (MSM) in the Atlanta-Sandy Springs-Roswell metropolitan area, Georgia. PrEP may impact not only the individual prescribed, but also their partners and beyond, known as spillover. We simulated a two-stage randomised trial with eligible components (≥3 agents with ≥1 HIV+ agent) first randomised to intervention or control (no PrEP). Within intervention components, agents were randomised to PrEP with coverage of 70%, providing insight into a high PrEP coverage strategy. We evaluated effect modification by component-level characteristics and estimated spillover effects on HIV incidence using an extension of randomisation-based estimators. We observed an attenuation of the spillover effect when agents were in components with a higher prevalence of either drug use or bridging potential (if an agent acts as a mediator between ≥2 connected groups of agents). The estimated spillover effects were larger in magnitude among components with either higher HIV prevalence or greater density (number of existing partnerships compared to all possible partnerships). Consideration of effect modification is important when evaluating the spillover of PrEP among MSM.

A platform trial design for preventive vaccines against Marburg virus and other emerging infectious disease threats.

BACKGROUND: The threat of a possible Marburg virus disease outbreak in Central and Western Africa is growing. While no Marburg virus vaccines are currently available for use, several candidates are in the pipeline. Building on knowledge and experiences in the designs of vaccine efficacy trials against other pathogens, including SARS-CoV-2, we develop designs of randomized Phase 3 vaccine efficacy trials for Marburg virus vaccines. METHODS: A core protocol approach will be used, allowing multiple vaccine candidates to be tested against controls. The primary objective of the trial will be to evaluate the effect of each vaccine on the rate of virologically confirmed Marburg virus disease, although Marburg infection assessed via seroconversion could be the primary objective in some cases. The overall trial design will be a mixture of individually and cluster-randomized designs, with individual randomization done whenever possible. Clusters will consist of either contacts and contacts of contacts of index cases, that is, ring vaccination, or other transmission units. RESULTS: The primary efficacy endpoint will be analysed as a time-to-event outcome. A vaccine will be considered successful if its estimated efficacy is greater than 50% and has sufficient precision to rule out that true efficacy is less than 30%. This will require approximately 150 total endpoints, that is, cases of confirmed Marburg virus disease, per vaccine/comparator combination. Interim analyses will be conducted after 50 and after 100 events. Statistical analysis of the trial will be blended across the different types of designs. Under the assumption of a 6-month attack rate of 1% of the participants in the placebo arm for both the individually and cluster-randomized populations, the most likely sample size is about 20,000 participants per arm. CONCLUSION: This event-driven design takes into the account the potentially sporadic spread of Marburg virus. The proposed trial design may be applicable for other pathogens against which effective vaccines are not yet available.

Modeling the Impacts of Clinical Influenza Testing on Influenza Vaccine Effectiveness Estimates.

BACKGROUND: Test-negative design studies for evaluating influenza vaccine effectiveness (VE) enroll patients with acute respiratory infection. Enrollment typically occurs before influenza status is determined, resulting in over-enrollment of influenza-negative patients. With availability of rapid and accurate molecular clinical testing, influenza status could be ascertained before enrollment, thus improving study efficiency. We estimate potential biases in VE when using clinical testing. METHODS: We simulate data assuming 60% vaccinated, 25% of those vaccinated are influenza positive, and VE of 50%. We show the effect on VE in 5 scenarios. RESULTS: Vaccine effectiveness is affected only when clinical testing preferentially targets patients based on both vaccination and influenza status. Vaccine effectiveness is overestimated by 10% if nontesting occurs in 39% of vaccinated influenza-positive patients and 24% of others. VE is also overestimated by 10% if nontesting occurs in 8% of unvaccinated influenza-positive patients and 27% of others. Vaccine effectiveness is underestimated by 10% if nontesting occurs in 32% of unvaccinated influenza-negative patients and 18% of others. CONCLUSIONS: Although differential clinical testing by vaccine receipt and influenza positivity may produce errors in estimated VE, bias in testing would have to be substantial and overall proportion of patients tested would have to be small to result in a meaningful difference in VE.

Estimating Population-Level Effects of the Acellular Pertussis Vaccine Using Routinely Collected Immunization Data.

BACKGROUND: Measuring and reporting the different population-level effects of the acellular pertussis vaccine on pertussis disease in addition to direct effects can increase the cost-effectiveness of a vaccine. METHODS: We conducted a retrospective cohort study of children born between 1 January 2008 and 31 December 2017, in King County, Washington, who were enrolled in the Washington State Immunization Information System. Diphtheria, tetanus toxoids, and acellular pertussis (DTaP) vaccination data from in the Washington State Immunization Information System was linked with pertussis case data from Public Health Seattle and King County. Census-level vaccination coverage was estimated as proportion of age-appropriately vaccinated children residing in it. Direct vaccine effectiveness was estimated by comparing pertussis risk in fully vaccinated and undervaccinated children. Population-level vaccine effectiveness (VE) measures were estimated by comparing pertussis risk in census tracts in the highest quartile for vaccination coverage with that in the lowest quartile. RESULTS: For direct protection, estimated VE was 76% (95% confidence interval, 63%-84%) in low-vaccination-coverage clusters, and it decreased to 47% (13%-68%) in high-coverage clusters, after adjustment for potential confounders. The estimated indirect VE was 45.0% (95% confidence interval, 1%-70%), the total VE 93.9% (91%-96%), and the overall VE 42.2% (19%-60%). CONCLUSION: Our findings suggest that DTaP vaccination provided direct as well as indirect protection in the highly immunized King County, Washington. Routine DTaP vaccination programs may have the potential to provide not only protection for vaccinated individuals but also for the undervaccinated individuals living in the same area.

Estimates of Inactivated Influenza Vaccine Effectiveness Among Children in Senegal: Results From 2 Consecutive Cluster-Randomized Controlled Trials in 2010 and 2011.

BACKGROUND: We report results of years 2 and 3 of consecutive cluster-randomized controlled trials of trivalent inactivated influenza vaccine (IIV3) in Senegal. METHODS: We cluster-randomized (1:1) 20 villages to annual vaccination with IIV3 or inactivated poliovirus vaccine (IPV) of age-eligible residents (6 months-10 years). The primary outcome was total vaccine effectiveness against laboratory-confirmed influenza illness (LCI) among age-eligible children (modified intention-to-treat population [mITT]). Secondary outcomes were indirect (herd protection) and population (overall community) vaccine effectiveness. RESULTS: We vaccinated 74% of 12 408 age-eligible children in year 2 (June 2010-April 11) and 74% of 11 988 age-eligible children in year 3 (April 2011-December 2011) with study vaccines. Annual cumulative incidence of LCI was 4.7 (year 2) and 4.2 (year 3) per 100 mITT child vaccinees of IPV villages. In year 2, IIV3 matched circulating influenza strains. The total effectiveness was 52.8% (95% confidence interval [CI], 32.3-67.0), and the population effectiveness was 36.0% (95% CI, 10.2-54.4) against LCI caused by any influenza strain. The indirect effectiveness against LCI by A/H3N2 was 56.4% (95% CI, 39.0-68.9). In year 3, 74% of influenza detections were vaccine-mismatched to circulating B/Yamagata and 24% were vaccine-matched to circulating A/H3N2. The year 3 total effectiveness against LCI was -14.5% (95% CI, -81.2-27.6). Vaccine effectiveness varied by type/subtype of influenza in both years. CONCLUSIONS: IIV3 was variably effective against influenza illness in Senegalese children, with total and indirect vaccine effectiveness present during the year when all circulating strains matched the IIV3 formulation. CLINICAL TRIALS REGISTRATION: NCT00893906.

Incorporating Real-time Influenza Detection Into the Test-negative Design for Estimating Influenza Vaccine Effectiveness: The Real-time Test-negative Design (rtTND).

With rapid and accurate molecular influenza testing now widely available in clinical settings, influenza vaccine effectiveness (VE) studies can prospectively select participants for enrollment based on real-time results rather than enrolling all eligible patients regardless of influenza status, as in the traditional test-negative design (TND). Thus, we explore advantages and disadvantages of modifying the TND for estimating VE by using real-time, clinically available viral testing results paired with acute respiratory infection eligibility criteria for identifying influenza cases and test-negative controls prior to enrollment. This modification, which we have called the real-time test-negative design (rtTND), has the potential to improve influenza VE studies by optimizing the case-to-test-negative control ratio, more accurately classifying influenza status, improving study efficiency, reducing study cost, and increasing study power to adequately estimate VE. Important considerations for limiting biases in the rtTND include the need for comprehensive clinical influenza testing at study sites and accurate influenza tests.

Disseminated Effects in Agent-Based Models: A Potential Outcomes Framework and Application to Inform Preexposure Prophylaxis Coverage Levels for HIV Prevention.

Preexposure prophylaxis (PrEP) for prevention of human immunodeficiency virus (HIV) infection may benefit not only the person who uses it but also their uninfected sexual risk contacts. We developed an agent-based model using a novel trial emulation approach to quantify disseminated effects of PrEP use among men who have sex with men in Atlanta, Georgia, from 2015 to 2017. Model components (subsets of agents connected through partnerships in a sexual network but not sharing partnerships with any other agents) were first randomized to an intervention coverage level or the control group; then, within intervention components, eligible agents were randomized to receive or not receive PrEP. We calculated direct and disseminated (indirect) effects using randomization-based estimators and report corresponding 95% simulation intervals across scenarios ranging from 10% coverage in the intervention components to 90% coverage. A population of 11,245 agents was simulated, with an average of 1,551 components identified. When comparing agents randomized to no PrEP in 70% coverage components with control agents, there was a 15% disseminated risk reduction in HIV incidence (risk ratio = 0.85, 95% simulation interval: 0.65, 1.05). Persons not on PrEP may receive a protective benefit by being in a sexual network with higher PrEP coverage. Agent-based models are useful for evaluating possible direct and disseminated effects of HIV prevention modalities in sexual networks.

Using simulated infectious disease outbreaks to inform site selection and sample size for individually randomized vaccine trials during an ongoing epidemic.

BACKGROUND: Novel strategies are needed to make vaccine efficacy trials more robust given uncertain epidemiology of infectious disease outbreaks, such as arboviruses like Zika. Spatially resolved mathematical and statistical models can help investigators identify sites at highest risk of future transmission and prioritize these for inclusion in trials. Models can also characterize uncertainty in whether transmission will occur at a site, and how nearby or connected sites may have correlated outcomes. A structure is needed for how trials can use models to address key design questions, including how to prioritize sites, the optimal number of sites, and how to allocate participants across sites. METHODS: We illustrate the added value of models using the motivating example of Zika vaccine trial planning during the 2015-2017 Zika epidemic. We used a stochastic, spatially resolved, transmission model (the Global Epidemic and Mobility model) to simulate epidemics and site-level incidence at 100 high-risk sites in the Americas. We considered several strategies for prioritizing sites (average site-level incidence of infection across epidemics, median incidence, probability of exceeding 1% incidence), selecting the number of sites, and allocating sample size across sites (equal enrollment, proportional to average incidence, proportional to rank). To evaluate each design, we stochastically simulated trials in each hypothetical epidemic by drawing observed cases from site-level incidence data. RESULTS: When constraining overall trial size, the optimal number of sites represents a balance between prioritizing highest-risk sites and having enough sites to reduce the chance of observing too few endpoints. The optimal number of sites remained roughly constant regardless of the targeted number of events, although it is necessary to increase the sample size to achieve the desired power. Though different ranking strategies returned different site orders, they performed similarly with respect to trial power. Instead of enrolling participants equally from each site, investigators can allocate participants proportional to projected incidence, though this did not provide an advantage in our example because the top sites had similar risk profiles. Sites from the same geographic region may have similar outcomes, so optimal combinations of sites may be geographically dispersed, even when these are not the highest ranked sites. CONCLUSION: Mathematical and statistical models may assist in designing successful vaccination trials by capturing uncertainty and correlation in future transmission. Although many factors affect site selection, such as logistical feasibility, models can help investigators optimize site selection and the number and size of participating sites. Although our study focused on trial design for an emerging arbovirus, a similar approach can be made for any infectious disease with the appropriate model for the particular disease.