Analysis of opinions from studies citing use of Core Outcome Measures in Effectiveness Trials outcomes taxonomy supports it use in health research: descriptive study

Objectives

To support the use of the Core Outcome Measures in Effectiveness Trials taxonomy for health outcomes, its suitability and applicability must be assessed more broadly beyond that of its early piloting phases. Demonstration of its suitability in practice would provide further support for its use in aiding the development of core outcome sets (COS), systematic reviews, and searching in online resources, thereby aiding knowledge dissemination.

Study Design and Setting

A citation analysis identified published studies where the taxonomy had been applied. Analysis of these publications aimed to understand the type of publication, clinical area, reason for taxonomy use or adaptation, and any comments made by researchers who had applied the taxonomy.

Results

Of 315 papers identified, 200 were sampled and 193 publications relating to 184 projects were analyzed. Nearly one-third (58, 30%) of publications were related to the development of COS and half (98, 51%) were related to the development of reviews (systematic, scoping, and literature). In two-thirds (123, 67%) of the projects the taxonomy was applied for the classification of health outcomes and the vast majority (117, 95%) of these did so without making any changes.

Conclusion

This research confirms the taxonomy is sufficiently comprehensive and granular for the classification of all patient outcomes in health research. Its application can highlight a lack of attention being paid toward outcomes most important to patients. We encourage the adoption of this classification system to facilitate evidence searching.

Plain Language Summary

Patient health outcomes measure things that happen to patients relating to their health. These outcomes include clinical measures (such as blood pressure), life impact measures (such as effects on physical functioning), use of resources (such as number of hospital appointments), survival (such as how long someone survives after surgery), and harms (such as adverse events following treatment). In 2018, we developed a classification system (called a taxonomy) to help researchers organize the types of outcomes that they are collecting (perhaps when carrying out a clinical trial) or reporting (eg, when combining results from many studies in a systematic review). This taxonomy was designed to help researchers to more clearly present their results, as well as to allow them to search for outcomes online in a more organized manner. We used outcomes taken from many trials and reviews when developing the taxonomy, to make sure that all types of outcomes were covered, but we wanted to make sure that researchers found the taxonomy helpful in practice. This study has been carried out to understand the opinions of researchers who mentioned the taxonomy in their publications, to find out whether they thought any types of outcomes were missing. The study found that researchers found the taxonomy useful, either to classify their research outcomes or because they found our definitions of the different types of outcomes useful. Sometimes researchers wanted to be able to classify “who” reported the outcomes (such as patients) or wanted more specific definitions of the different types of outcomes included. We would encourage them to create subcategories within our taxonomy categories to cover these outcome features. A few researchers felt that they needed to add some extra outcome types, but this was because they did not realize that these outcomes would fit within one of the existing taxonomy groups. In summary, this study found that the taxonomy is useful to researchers to classify “what” patient health outcomes are recorded in their studies, and researchers are very welcome to contact us if they have any questions about how to use it or how to classify particular outcomes.