A proposed methodology for uncertainty extraction and verification in priority setting partnerships with the James Lind Alliance: an example from the Common Conditions Affecting the Hand and Wrist Priority Setting Partnership

Background: To report our recommended methodology for extracting and then confirming research uncertainties - areas where research has failed to answer a research question - derived from previously published literature during a broad scope Priority Setting Partnership (PSP) with the James Lind Alliance (JLA).

Methods: This process was completed in the UK as part of the PSP for "Common Conditions Affecting the Hand and Wrist", comprising of health professionals, patients and carers and reports the data (uncertainty) extraction phase of this. The PSP followed the robust methodology dictated by the JLA and sought to identify knowledge gaps, termed "uncertainties" by the JLA. Published Cochrane Systematic Reviews, Guidelines and Protocols, NICE (National Institute for Health and Care Excellence) Guidelines, and SIGN (Scottish Intercollegiate Guidelines Network) Guidelines were screened for documented "uncertainties". A robust method of screening, internally verifying and then checking uncertainties was adopted. This included independent screening and data extraction by multiple researchers and use of a PRISMA flowchart, alongside steering group consensus processes. Selection of research uncertainties was guided by the scope of the Common Conditions Affecting the Hand and Wrist PSP which focused on "common" hand conditions routinely treated by hand specialists, including hand surgeons and hand therapists limited to identifying questions concerning the results of intervention, and not the basic science or epidemiology behind disease.

Results: Of the 2358 records identified (after removal of duplicates) which entered the screening process, 186 records were presented to the PSP steering group for eligibility assessment; 79 were deemed within scope and included for the purpose of research uncertainty extraction (45 full Cochrane Reviews, 18 Cochrane Review protocols, 16 Guidelines). These yielded 89 research uncertainties, which were compared to the stakeholder survey, and added to the longlist where necessary; before derived uncertainties were checked against non-Cochrane published systematic reviews.

Conclusions: In carrying out this work, beyond reporting on output of the Common Conditions Affecting the Hand and Wrist PSP, we detail the methodology and processes we hope can inform and facilitate the work of future PSPs and other evidence reviews, especially those with a broader scope beyond a single disease or condition.