Explore the vast range of titles available.

For researchers using observational data, a useful way to answer a causal question is to design the target trial that would answer it and then emulate its protocol. The example of the HIV-treatment-as-prevention strategy illustrates the benefits of this approach.

Observational databases are often used to study causal questions. Before being granted access to data or funding, researchers may need to prove that “the statistical power of their analysis will be high.” Analyses expected to have low power, and hence result in imprecise estimates, will not be approved. This restrictive attitude towards observational analyses is misguided. A key misunderstanding is the belief that the goal of a causal analysis is to “detect” an effect. Causal effects are not binary signals that are either detected or undetected; causal effects are numerical quantities that need to be estimated. Because the goal is to quantify the effect as unbiasedly and precisely as possible, the solution to observational analyses with imprecise effect estimates is not avoiding observational analyses with imprecise estimates, but rather encouraging the conduct of many observational analyses. It is preferable to have multiple studies with imprecise estimates than having no study at all. After several studies become available, we will meta-analyze them and provide a more precise pooled effect estimate. Therefore, the justification to withhold an observational analysis of preexisting data cannot be that our estimates will be imprecise. Ethical arguments for power calculations before conducting a randomized trial which place individuals at risk are not transferable to observational analyses of existing databases. If a causal question is important, analyze your data, publish your estimates, encourage others to do the same, and then meta-analyze. The alternative is an unanswered question.

When using observational data, quantifying the effect of treatment duration on survival outcomes is not straightforward because only people who live for a long time can receive treatment for a long time. This problem doesn’t apply to randomised trials because people are classified based on the treatment duration they are assigned, rather than the treatment duration that they achieve. This approach accepts that dead people do not deviate from their assigned treatment strategy. By transferring this insight to the analysis of observational data, we can follow three steps to estimate the effect of treatment duration from observational data without the bias of naive comparisons between long term and short term users. The first step is cloning people to assign them to multiple treatment strategies. The second step is censoring clones when they deviate from their assigned treatment strategy. The third step is performing inverse probability weighting to adjust for the potential selection bias introduced by censoring. The procedure can be used to compare any treatment strategies that are sustained over time. Cloning, censoring, and weighting eliminates immortal time bias in the estimates of absolute and relative risk, which helps researchers focus their attention on other biases that may be present in observational analyses and are not so easily eliminated.

Pragmatic trials are designed to address real-world questions about care options. This article addresses issues that may arise from per-protocol and intention-to-treat analyses of such trials, outlines alternative analytic approaches, and provides guidance on how to choose among them.

Many countries are experiencing a resurgence of COVID-19, driven predominantly by the delta (B.1.617.2) variant of SARS-CoV-2. In response, these countries are considering the administration of a third dose of mRNA COVID-19 vaccine as a booster dose to address potential waning immunity over time and reduced effectiveness against the delta variant. We aimed to use the data repositories of Israel's largest health-care organisation to evaluate the effectiveness of a third dose of the BNT162b2 mRNA vaccine for preventing severe COVID-19 outcomes. Using data from Clalit Health Services, which provides mandatory health-care coverage for over half of the Israeli population, individuals receiving a third vaccine dose between July 30, 2020, and Sept 23, 2021, were matched (1:1) to demographically and clinically similar controls who did not receive a third dose. Eligible participants had received the second vaccine dose at least 5 months before the recruitment date, had no previous documented SARS-CoV-2 infection, and had no contact with the health-care system in the 3 days before recruitment. Individuals who are health-care workers, live in long-term care facilities, or are medically confined to their homes were excluded. Primary outcomes were COVID-19-related admission to hospital, severe disease, and COVID-19-related death. The third dose effectiveness for each outcome was estimated as 1 – risk ratio using the Kaplan-Meier estimator. 1 158 269 individuals were eligible to be included in the third dose group. Following matching, the third dose and control groups each included 728 321 individuals. Participants had a median age of 52 years (IQR 37–68) and 51% were female. The median follow-up time was 13 days (IQR 6–21) in both groups. Vaccine effectiveness evaluated at least 7 days after receipt of the third dose, compared with receiving only two doses at least 5 months ago, was estimated to be 93% (231 events for two doses vs 29 events for three doses; 95% CI 88–97) for admission to hospital, 92% (157 vs 17 events; 82–97) for severe disease, and 81% (44 vs seven events; 59–97) for COVID-19-related death. Our findings suggest that a third dose of the BNT162b2 mRNA vaccine is effective in protecting individuals against severe COVID-19-related outcomes, compared with receiving only two doses at least 5 months ago. The Ivan and Francesca Berkowitz Family Living Laboratory Collaboration at Harvard Medical School and Clalit Research Institute.

Many analyses of observational data are attempts to emulate a target trial. The emulation of the target trial may fail when researchers deviate from simple principles that guide the design and analysis of randomized experiments. We review a framework to describe and prevent biases, including immortal time bias, that result from a failure to align start of follow-up, specification of eligibility, and treatment assignment. We review some analytic approaches to avoid these problems in comparative effectiveness or safety research.

Effect estimates may be biased when the study design or the data analysis is conditional on a collider—a variable that is caused by two other variables. Causal directed acyclic graphs are a helpful tool to identify colliders that may introduce selection bias in observational research.

In this issue, Weiss discusses “generalizing” inferences from randomized trials to other populations [1]. However, he does not explicitly define what “generalizing” means, assumes that “generalizing” the results of a randomized trial has a single goal, and reduces generalizability to a binary subjective judgment—findings are either generalizable or not generalizable. A growing literature (e.g., [1–13]) precisely defines the several meanings and goals of extending inferences from randomized trials to another population, and describes analyses whose findings go beyond simple binary judgements. Here, we provide a non-technical overview of this literature. First, we briefly review the main concepts, then we outline the available study designs and statistical approaches.

Observational analyses for causal inference often rely on real world data collected for purposes other than research. A frequent goal of these observational analyses is to use the data to emulate a hypothetical randomized experiment, i.e., the target trial, that mimics the design features of a true experiment, including a clear definition of time zero with synchronization of treatment assignment and determination of eligibility. We review a recent observational analysis that explicitly emulated a target trial of screening colonoscopy using insurance claims from U.S. Medicare. We then compare this explicit emulation with alternative, simpler observational analyses that do not synchronize treatment assignment and eligibility determination at time zero and/or do not allow for repeated eligibility. This empirical comparison suggests that lack of an explicit emulation of the target trial leads to biased estimates, and shows that allowing for repeated eligibility increases the statistical efficiency of the estimates.

Recent observational studies have found a higher risk of reinfection with the omicron variant of SARS-CoV-2 in people who received a third covid-19 booster dose. This finding has been interpreted as evidence of immune imprinting of covid-19 vaccines. This article proposes an alternative explanation: that the increased risk of reinfection in individuals vaccinated with a booster compared with no booster is the result of selection bias and is expected to arise even in the absence of immune imprinting. To clarify this alternative explanation, this article describes how previous observational analyses were an attempt to estimate the direct effect of vaccine boosters on SARS-CoV-2 reinfections—an effect that cannot be correctly estimated with observational data. Causal diagrams (directed acyclic graphs), data simulations, and analysis of real world data are used to illustrate the mechanism and magnitude of this bias, which is the result of conditioning on a collider.