Introduction

The Umbrella Review has been identified as one of the four “next-generation” types of systematic reviews, along with prospective meta-analysis, individual participant data meta-analysis, and network meta-analysis, which can contribute to the development of new methodologies for the reliable synthesis of research data.

Definition

An umbrella review systematically collects and evaluates information from multiple systematic reviews and meta-analyses across all clinical outcomes. It is usually applied to controversial issues, areas with multiple meta-analyses providing inconclusive evidence, or research fields with substantial bias that have not been systematically assessed. The goal is to evaluate the reliability and validity of epidemiological evidence using a set of predefined criteria and statistical tests.

Purpose

The purpose of an umbrella review is not to replicate search strategies, assess study eligibility, or evaluate the risk of bias of the included systematic reviews and meta-analyses, but rather to provide an overall picture of the findings for specific questions or phenomena. In contrast to a systematic review or a meta-analysis that is typically limited to a single treatment comparison, an umbrella review can offer a broader overview of multiple treatments. This is particularly useful for informing guidelines and clinical practice when all management options need to be considered.

Benefits of the Review

An umbrella review offers a ready-to-use tool for healthcare decision-makers to gain a clear and comprehensive understanding of a wide field of research. It is no surprise that, although relatively recent, such reviews are gaining increasing influence in the biomedical literature. Their number is steadily rising as their content provides an attractive way of summarizing research data. Depending on the research question, umbrella reviews may include systematic reviews and meta-analyses of observational studies, interventional studies, or Mendelian randomization studies, thus covering a wide range of topics including nutrition, obstetrics and gynecology, psychiatry, neurology, and internal medicine.

Evaluation – Criteria

Although no official consensus has yet been established by recognized organizations on how to assess the epidemiological validity of associations derived from umbrella reviews, a specific ranking system based on predefined criteria has been repeatedly applied by experienced research groups. According to this system, associations are classified into five categories. Convincing associations (Category I) meet criteria such as high statistical significance (p-value < 10^-6), large sample size with more than 1000 cases (or over 20,000 participants for continuous outcomes), no substantial heterogeneity (I^2 < 50%), prediction intervals excluding the null, and no evidence of small-study effects or excess significance bias. Highly suggestive associations (Category II) are defined by a p-value < 10^-6 in random-effects models, more than 1000 cases (or over 20,000 for continuous outcomes), and at least one large individual study showing statistically significant results. Suggestive associations (Category III) are characterized by a p-value < 0.001 and more than 1000 cases (or over 20,000 participants for continuous outcomes). Weak associations (Category IV) are those with statistically significant results (p-value < 0.05) in random-effects models, while not significant associations are defined by p-values ≥ 0.05.

Conclusions

In summary, it is important to emphasize that the main goal of an umbrella review is to provide an overarching overview and assessment of the validity of the available evidence on a specific research topic, drawing from all existing systematic reviews and meta-analyses. Its purpose is not to generate new pooled results through meta-analysis or metasynthesis of existing studies but rather to produce a comprehensive and reliable evaluation that can better inform research, guidelines, and clinical practice.