Living, Rapid Reviews in a Rapidly Evolving World

Abstract

In their article, Holmer and colleagues reported the final update of a living review that examined the duration of IgG antibody response, the role of previous SARS-CoV-2 infection to prevent reinfection, and the role of antibodies in protection from reinfection. The editorialist discusses the challenge of living reviews as new evidence emerges and the relevant clinical questions evolve.

When the COVID-19 pandemic began, clinicians struggled to make decisions in the absence of any direct evidence. Researchers responded rapidly to fill the knowledge void with many studies of varying quality, using different interventions, cointerventions, comparators, and populations. Simultaneously, the systematic review field mobilized to organize and help collate, manage, and translate the rapid proliferation of research for on-the-ground decision makers (1, 2).

In the face of a voracious need for information but an insufficient, although rapidly developing, evidence base, the living, rapid review was born. Rapid reviews, although poorly defined, emerged out of the acknowledgment that traditional systematic reviews often take more than 12 months to complete, but decisions need to be made faster (3). Previously established methods for updating reviews (4) formed the foundation of the living review. The pandemic brokered this marriage, and since 2020, Annals of Internal Medicine has since published 7 living, rapid reviews related to the COVID-19 pandemic. The publication by Holmer and colleagues (5) marks the completion of 5.

In this second and final update, Holmer and colleagues (5) examine the duration of IgG antibody response, the role of previous SARS-CoV-2 infection to prevent reinfection, and the role of antibodies in protection from reinfection. The authors found low strength of evidence that antibodies are maintained for at least 12 months, although this may be less in immunocompromised groups and with later variants. They found moderate strength of evidence that prior infection with the Delta SARS-CoV-2 variant protected against symptomatic reinfection with the Omicron BA.1 and BA.2 variants (50% to 67%) and was highly protective against severe disease. There was low strength of evidence that prior infection with BA.1 or BA.2 protected against symptomatic reinfection with BA.4 or BA.5 variants for up to 5 months. However there was little evidence on the role of antibody testing to predict future reinfection risk.

Updating a review is often more complicated than rerunning a search, adding more studies to an evidence table, and updating a meta-analysis. In 2016, Garner and colleagues (4) published a consensus and checklist for updating systematic reviews. That guidance suggested that the decision to update a review should start with asking whether the review addresses a current question. If the question is no longer current, the panel recommended that no further update is needed, such as when the intervention is not in use or has been superseded. In 2022, El Mikati and colleagues (6) published a framework for living practice guidelines. Once a guideline is in the living mode, authors note that each update should start with revisiting the relevance of the question and whether any modifications are needed (for example, changing the outcomes of interest) before moving on to updating the systematic evidence review and clinical practice recommendation.

The review by Holmer and colleagues highlights many of the challenges of conducting living, rapid reviews not only when the evidence is rapidly evolving, but the environment and virus itself also continue to evolve. When this review began in 2020, the primary question motivating the review was whether antibodies could be used as a marker of immunity (7). Although the authors could not make this link, in the original review, they did find that most adults with SARS-CoV-2 infection developed antibodies that persisted for at least 3 months. With the advent of vaccination further confusing the measurement of antibodies, in the first update, rather than continuing down the established, but less clinically useful, path on examining evidence for antibodies as a marker of immunity, the authors turned to understanding the role of SARS-CoV-2 infection on reinfection (8). They found that before the Delta and Omicron variants, infection had strong protection against symptomatic reinfections for 7 months compared with unvaccinated, uninfected persons. This second and final update returns to update both questions about the antibody response and protection from reinfection after infection.

Although both Garner and colleagues (4) and El Mikati and colleagues (6) account for the possibility of changing questions in living updates, both suggest or imply that reviewers and guideline developers restrict updates to relatively minor changes in questions. This topic exemplifies the challenges that may arise from broader changes in scope and questions. The authors of this living, rapid review could have considered that changes in the vaccination and variant landscape made the original primary question obsolete and abandoned the review and/or started a new review. Instead, they used a pragmatic approach and evolved the questions to encompass related questions on reinfection risk to remain pertinent to the decisional milieu at the time. They used what they called a SWATH review (9) where updates may focus on different parts of the overarching topic coincident with the emergence of research and interest by decision makers. Challenges in transparently conveying the linkages between the updates and the original review may be offset by increased flexibility in being able to address the question most relevant to decision making.

This living, rapid review demonstrated novel methods to maintain flexibility and relevance in a rapidly changing environment. As interest and experience in living reviews increase, the systematic review field will continue to learn and experiment with different approaches to handle changing questions and methods. The COVID-19 pandemic has spurred interest in living reviews and shown how quickly questions themselves become outdated. However, interest in living reviews continues beyond COVID-19. Research, by nature, means learning and changing context; thus, the challenges around changing questions will certainly not be unique to COVID-19.