Epidemiology

Test-negative designs (TNDs) can be used to estimate vaccine effectiveness by comparing the relative rates of the target disease and symptomatically similar diseases among vaccinated and unvaccinated populations. However, the diagnostic tests used to identify the target disease typically suffer from imperfect sensitivity and specificity, leading to biased vaccine effectiveness estimates. Here we present a solution to this problem via a Bayesian statistical model which can either incorporate poin...

BackgroundThe relationship between antibody levels (more generally, a scalar measure of immune protection) at the time of exposure to infection (so-called exposure-proximal correlates of protection) and the risk of infection given exposure is of central interest in evaluating the evolution of immune protection conferred by prior infection and/or vaccination. A version of the test-negative study design (TND), adapted from vaccine effectiveness studies, has been used to assess this relationship. H...

Effective antimicrobial stewardship requires unbiased assessment of the benefits and harms of different treatment strategies, considering both immediate patient outcomes and patterns of antimicrobial resistance. In principle, these benefits and harms can be expressed as causal contrasts between treatment strategies and therefore should be ideally suited for study under the potential outcomes framework. However, causal inference in this setting is complicated by interference between individuals (...

The diagnosis of infectious disease syndromes such as fever, diarrhea, and pneumonia is complicated by the potential for shedding or carriage of putatively etiologic pathogens among individuals experiencing symptoms due to other causes. Symptomatic individuals among whom a pathogen is detected, but whose symptoms are caused by other factors, may be misclassified by diagnostic criteria based on pathogen detection. Case-control studies are commonly undertaken to estimate vaccine effectiveness, and...

SO_SCPLOWUMMARYC_SCPLOWTest-negative designs, a form of case-cohort studies, have been commonly used to assess infectious disease interventions. Early examples of the design included the evaluation of seasonal influenza vaccine in the field. Recently, they have also been widely used to evaluate the efficacy of COVID-19 vaccines in preventing symptomatic disease for different variants (Evans and Jewell, 2021). The design hinges on individuals being tested for the disease of interest; upon recruit...

Host adaptive immune responses may protect against infection or disease when a pathogen is repeatedly encountered. The hazard ratio of infection or disease, given previous infection, is typically sought to estimate the strength of protective immunity. However, variation in individual exposure or susceptibility to infection may introduce frailty bias, whereby a tendency for infections to recur among individuals with greater risk confounds the causal association between previous infection and susc...

Vaccine effectiveness (VE) studies are subject to biases due to depletion of at-risk persons or of highly susceptible persons at different rates from different groups (depletion-of-susceptibles bias), a problem that can also lead to biased estimates of waning effectiveness, including spurious inference of waning when none exists. An alternative study design to identify waning is to study only vaccinated persons, and compare for each day the incidence in persons with earlier or later dates of vac...

Vaccines can provide protection against infection or limit disease progression and severity. Vaccine efficacy (VE) is typically evaluated independently for different outcomes, but this can cause biased estimates of VE. We propose a new analytical framework based on a model of disease progression for VE estimation for infections with multiple possible outcomes of infection: Joint analysis of multiple outcomes in vaccine efficacy trials (JAMOVET). JAMOVET is a Bayesian hierarchical regression mode...

The naive approach to estimating the effects of a vaccine on asymptomatic infections, which compares the risk of asymptomatic infection among vaccinated and unvaccinated individuals, can be misleading because it is comprised of two effects: the vaccine preventing asymptomatic infections and the vaccine converting symptomatic to asymptomatic infections. When the latter effect is strong, vaccines can appear harmful with respect to asymptomatic infections. Using a causal principal stratification fr...

Randomized controlled trials (RCTs) have shown high efficacy of multiple vaccines against SARS-CoV-2 disease (COVID-19), and recent studies have shown the vaccines are also effective against infection. Evidence for the effect of each of these vaccines on ability to transmit the virus is also beginning to emerge. We describe an approach to estimate these vaccines effects on viral positivity, a prevalence measure which under the reasonable assumption that vaccinated individuals who become infected...

Methods of causal inference are used to estimate treatment effectiveness for non-randomized study designs. The propensity score (i.e., the probability that a subject receives the study treatment conditioned on a set of variables related to treatment and/or outcome) is often used with matching or sample weighting techniques to, ideally, eliminate bias in the estimates of treatment effect due to treatment decisions. If multiple treatments are available, the propensity score is a function of the ad...

BackgroundDuring epidemics with emerging infections, diagnostic tests directly inform model-based decision-making and thereby shape infection control strategies. However, diagnostic accuracy studies (DTA) assessing the validity of these tests must be conducted under severe time and data constraints. We investigated whether the integration of adaptive designs and of epidemic spread modelling for prevalence prediction can accelerate DTA studies during epidemics with emerging infections without com...

Establishing dose-response relationships from observational data is challenging due to confounding and sample selection bias. Standard causal methods adjust for confounding but typically require knowledge of covariate distributions in the target population--often via a well-defined probability sampling scheme. We propose the Covariate Adjusted Logit Model (CALM), which generalizes log-linear structural mean models for binary exposures to continuous exposures by modeling a relative dose-response ...

The test-negative design (TND) has become a widely used observational study design for evaluating vaccine effectiveness, especially during the COVID-19 pandemic. Traditionally, TND has been viewed as a variant of the case-control study and largely limited to use with logistic regression models. In this paper, we first establish that TND can be framed as a special case of a cohort study, thereby opening the door to a wider range of analytical approaches. We then introduce the Prentice, Williams, ...

BackgroundVaccines can prevent severe disease by preventing infection or by reducing progression among those who become infected. Vaccine effectiveness against progression given infection is often used to quantify this second mechanism, but it conditions on infection, which is itself affected by vaccination. As a result, this estimand lacks a clear causal interpretation and may behave non-intuitively over time. MethodsWe introduce a conceptual framework that models protection against infection ...

The extent and duration of immunity following SARS-CoV-2 infection are critical outstanding questions about the epidemiology of this novel virus, and studies are needed to evaluate the effects of serostatus on reinfection. Understanding the potential sources of bias and methods to alleviate biases in these studies is important for informing their design and analysis. Confounding by individual-level risk factors in observational studies like these is relatively well appreciated. Here, we show how...

Sample pooling of CoViD-19 PCR tests has been recently proposed as a low cost alternative to individual tests. We show that sample pooling is efficient as long as the fraction of the population infected is relatively small. Fisher information theory suggests a rule of thumb that for low infection rates p, pooling 2/p samples is close to optimal. We present a simple strategy for survey design when not even a ballpark estimate of the infection rate is available.

In real world data (RWD) studies, observed datasets are often subject to left truncation, which can bias estimates of survival parameters. Standard methods can only suitably account for left truncation when survival and entry time are independent. Therefore, in the dependent left truncation setting, it is important to quantify the magnitude and direction of estimator bias to determine whether an analysis provides valid results. We conduct simulation studies of common RWD analytic settings in ord...

Epidemiologic data often violate common modeling assumptions of independence between subjects due to study design. Statistical separation is also common, particularly in the study of rare binary outcomes. Statistical separation for binary outcomes occurs when regions of the covariate space have no variation in the outcome, and separation can negatively impact the validity of logistic regression model parameters. When data are correlated, we generally use multi-level modeling for parameter estima...

BackgroundTwo required inputs to mathematical models of sexually transmitted infections are the average duration in epidemiological risk states (e.g., selling sex) and the average rates of sexual partnership change. These variables are often only available as aggregate estimates from published cross-sectional studies, and may be subject to distributional, sampling, censoring, and measurement biases. MethodsWe explore adjustments for these biases using aggregate estimates of duration in sex work...