Open AccessDesigning a Strategy Trial for the Management of Gout: The Use of a Modified Delphi Panel
Author(s) -
Solomon Daniel H.,
Weissman Joel S.,
Choi Hyon,
Atlas Steven J.,
Berardinelli Cesar,
Dedier Julien,
Fischer Michael A.,
Fitzgerald John,
Hinteregger Erica,
Johnsen Brianne,
Marini Diana D.,
McLean Robert,
Murray Fred,
Neogi Tuhina,
Oertel Lynn B.,
Pillinger Michael H.,
Riggs Kevin R.,
Saag Ken,
Suh Dong,
Watkins James,
Barry Michael J.
Publication year - 2021
Publication title -
acr open rheumatology
Language(s) - English
Resource type - Journals
ISSN - 2578-5745
DOI - 10.1002/acr2.11243
Subject(s) - voting , delphi method , delphi , randomized controlled trial , primary care , medicine , clinical trial , family medicine , medical education , psychology , computer science , artificial intelligence , political science , politics , law , operating system
Objective Disagreement exists between rheumatology and primary care societies regarding gout management. This paper describes a formal process for gathering input from stakeholders in the planning of a trial to compare gout management strategies. Methods We recruited patients, nurses, physician assistants, primary care clinicians, and rheumatologists to participate in a modified Delphi panel (mDP) to provide input on design of a trial focused on optimal management for primary care patients with gout. The 16 panelists received a plain‐language briefing document that discussed the rationale for the trial, key clinical issues in gout, and aspects of trial design. The panelists also received information and considerations on nine voting questions (VQs), judged to be the key design questions. Cognitive interviews with panelists ensured that the VQs were understood by the range of panelists involved in the mDP. Panelists were asked to score all VQs from 1 (definitely no) to 9 (definitely yes). Two voting rounds were conducted—round 1 by email and round 2 by video conference. Results The VQs were modified through the cognitive interviews. The round 1 voting resulted in consensus on eight items, with consensus defined as median voting score in the same tercile (1‐3, 4‐6 or 7‐9). Re‐voting at the meeting (round 2) reached consensus on the remaining item. Conclusion An mDP with various stakeholders facilitated consensus on the design of a trial of different management strategies for chronic gout. This method may be useful for designing trials of clinical questions with substantial disagreement across stakeholders.