Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

A practical approach to predict expansion of evidence networks

a case study in treatment-naive advanced melanoma

Halfpenny, Nicholas J.A.a; Scott, David A.b; Thompson, Juliette C.a; Gurung, Binua; Quigley, Joan M.a,c

doi: 10.1097/CMR.0000000000000513
ORIGINAL ARTICLES: Basic science
Buy
SDC

Network meta-analysis (NMA) is a statistical method used to produce comparable estimates of efficacy across a range of treatments that may not be compared directly within any single trial. NMA feasibility is determined by the comparability of the data and presence of a connected network. In rapidly evolving treatment landscapes, evidence networks can change substantially in a short period of time. We investigate methods to determine the optimum time to conduct or update a NMA based on anticipated available evidence. We report the results of a systematic review conducted in treatment-naive advanced melanoma and compare networks of evidence available at retrospective, current, and prospective time points. For included publications, we compared the primary completion date of trials from clinical trials registries (CTRs) with the date of their first available publication to provide an estimate of publication lag. Using CTRs we were able to produce anticipated networks for future time points based on projected study completion dates and average publication lags which illustrated expansion and strengthening of the initial network. We found that over a snapshot of periods between 2015 and 2018, evidence networks in melanoma changed substantively, adding new comparators and increasing network connectedness. Searching CTRs for ongoing trials demonstrates it is possible to anticipate future networks at a certain time point. Armed with this information, sensible decisions can be made over when best to conduct or update a NMA. Incorporating new and upcoming interventions in a NMA enables presentation of a complete, up-to-date and evolving picture of the evidence.

aICON Health Economics, Abingdon, Oxon

bDiabetes Research Centre, University of Leicester, Leicester General Hospital, Leicester, UK

cHealth Research Board, Grattan House, Dublin, Republic of Ireland

Correspondence to Nicholas J.A. Halfpenny, MSci, ICON Health Economics, 100 Park Drive, Milton Park, Abingdon, Oxon, OX14 4RY, UK Tel: +44 186 532 4930; e-mail: nicholas.halfpenny@iconplc.com

Received May 27, 2018

Accepted August 27, 2018

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.