Real-world data and evidence in pain research: a qualitative systematic review of methods in current practice

Abstract

The use of routinely collected health data (real-world data, RWD) to generate real-world evidence (RWE) for research purposes is a growing field. Computerized search methods, large electronic databases, and the development of novel statistical methods allow for valid analysis of data outside its primary clinical purpose. Here, we systematically reviewed the methodology used for RWE studies in pain research. We searched 3 databases (PubMed, EMBASE, and Web of Science) for studies using retrospective data sources comparing multiple groups or treatments. The protocol was registered under the DOI:10.17605/OSF.IO/KGVRM. A total of 65 studies were included. Of those, only 4 compared pharmacological interventions, whereas 49 investigated differences in surgical procedures, with the remaining studying alternative or psychological interventions or epidemiological factors. Most 39 studies reported significant results in their primary comparison, and an additional 12 reported comparable effectiveness. Fifty-eight studies used propensity scores to account for group differences, 38 of them using 1:1 case:control matching. Only 17 of 65 studies provided sensitivity analyses to show robustness of their findings, and only 4 studies provided links to publicly accessible protocols. RWE is a relevant construct that can provide evidence complementary to randomized controlled trials (RCTs), especially in scenarios where RCTs are difficult to conduct. The high proportion of studies reporting significant differences between groups or comparable effectiveness could imply a relevant degree of publication bias. RWD provides a potentially important resource to expand high-quality evidence beyond clinical trials, but rigorous quality standards need to be set to maximize the validity of RWE studies.

Pain Research and Treatment: Policy Implications

“…The use of real-world data, or health data that is routinely collected for research purposes, is another novel approach. A systematic review of this methodology demonstrates that, if rigorous quality standards are met, this approach can be valuable in generating high-quality evidence that goes beyond clinical trials (Vollert et al, 2023). Finally, emerging technologies such as machine learning can help construct prognostic profiles (Zmudzki and Smeets, 2023) and therefore aid in treatment personalization and outcome optimization.…”

Introduction to a special issue on big data and pain

“…The availability of publicly accessible databases, self-reported community-based cohorts, and data from wearables have created an increased interest on the methods and qualities of this “real-world evidence” (RWE) data, which has been collected outside the standardised clinical environment. In this systematic review, Vollert et al 15 discuss the methodological approaches required to design and analyse RWE studies in pain research. They address the usefulness, particularly when randomised control trials are very challenging to conduct, potentials, and challenges of using RWE in assessing the effectiveness of pain treatments.…”