OBJECTIVES

: The volume of systematic reviews is exponentially growing, and umbrella reviews are increasingly being used to synthesize such evidence and simplify interpretation. However, methods are not fully established, and challenges exist. Using the highly published topic of proton-pump inhibitors (PPIs) and adverse outcomes, our study objective was to conduct an umbrella review, as well as identify methodological challenges and targeted solutions.

METHODS

: The review was conducted in accordance with PRISMA, and recent umbrella review guidance. Notably, methodology was adapted to suit the evidence-base and optimize quality. Key methodological challenges involved: 1) Distilling large amounts of similar evidence; 2) Restricting the eligible target population; 3) Optimally presenting results given variety of effect measures reported and volume of outcomes assessed; and 4) Handling interpretation of observational data.

RESULTS

: Across 42 systematic reviews, the most comprehensive meta-analyses were selected for each outcome based on publication date and sample size, to avoid double-counting and simplify interpretation (with discussion of discrepancies for non-selected data). Studies focusing on specific or high-risk patient groups were excluded to increase generalizability. To ensure data were broadly comparable and optimize results interpretation, only meta-analyses were included. Effect measure type, size and significance of reported associations, combined with sample size, were graphically presented in condensed formats. Given the predominance of observational evidence, we optimized quality assessment by not only conducting AMSTAR 2 on the meta-analyses, but also compiling existing Newcastle-Ottawa Scale data for primary studies. By applying these methodologies, our review identified results for a larger number of adverse outcomes, including significant associations for fracture, kidney disease, pneumonia, enteric infection, gastric cancer, and gastrointestinal events, compared with previous literature.

CONCLUSIONS

: This case-study demonstrates that umbrella reviews can fill existing gaps in the clinical literature, and that review methods can be tailored to increase rigour and generate more reliable high-level information.