Old World, New World

Part of the remit for a journal is to look backwards and forwards simultaneously: are we learning lessons from past, prepared future? In this month's issue JCE publish articles that address both these tasks. Of course COVID-19 pandemic present hovers over all our thoughts, also represented here, but need continue develop science, prepare an unpredictable future ever-present. The imperative pre-register clinical trials may seem like ‘old news’. After all, it now 15 years since International Committee Medical Journal Editors stated should not be considered publication unless they included on registry. This development was closely followed by Food Drug Administration Amendment Act 2007 (FDAAA), which first introduced prospect financial penalties non-compliance. aim eradicate or at least make visible non-publication - bias. Yet, here in 2021 publishing paper such as Lindsley et al. demonstrating sample 2000 100 systematic reviews, even taking only those studies published last 10 years, proportion registered less than 40% (261 out 682) [[1]Lindsley K. Fusco N. Teeuw H. Mooij E. Scholten R. Hooft L. Poor compliance trial registration among reviews: cohort study.J Clin Epidemiol. 2020; 132: 79-87Abstract Full Text PDF PubMed Scopus (1) Google Scholar]. problem hidden data does not, course, end there. It nearly 17 An-Wen Chen illustrious group co-authors their landmark highlighting selective outcome reporting reports peer reviewed journals [[2]Chan A.W. Hróbjartsson A. Haahr M.T. Gøtzsche P.C. Altman D.G. Empirical evidence outcomes randomized trials: comparison protocols articles.JAMA. 2004; 291: 2457-2465Crossref (1294) China developing impressive reputation volume quality its scientific research, yet research integrity problems persist there, everywhere else. Jia colleagues United States results important study 300 RCTs mainland China, up August 2019 [[3]Jia Y. Huang D. Wen J. Liang Qureshi Wang al.Association between switching primary reported findings drug China.J 10-17Abstract familiar: whether were advance during study, ‘negative’ more likely switched ‘positive’. Both biases described above characteristically lead inappropriately optimistic perceptions benefit harm, two further describe additional challenges deliver same consequence. first, Urlings performed citation network analyses six biomedical questions. For four probability being cited associated with positive outcomes, perceived authority authors impact factor [[4]Urlings M.J.E. Duyx B. Swaen G.M.H. Bouter L.M. Zeegers M.P. Citation bias other determinants research: networks.J 71-78Abstract (2) second article, Yu demonstrated via randomised distorting effects reading ‘narrative’ review, compared leading higher predictions absolute risk reduction, evaluations proposed under hypothetical ethical review [[5]Yu M. Montroy Fergusson Lalu M.M. Kimmelman Systematic narrative experts different cancer predictions: trial.J 116-124Abstract These show there no room complacency search realism health care. Changes expectations past 3 decades have had effect still work do. We owe debt researchers who shine light into darkness. As indicated above, can forwards, includes number reflect emerging themes epidemiology. Unsurprisingly, has led explosion interest disease modeling prognostic indicators, readers brief invaluable immersion either fields include excellent guides. part series, Mac provide ‘comprehensive guide’ infectious decision-analytic models, early phase local collaboration decision makers [[6]Mac, S, Mishra Ximenes R, Barrett K, Khan YA, Naimark DM, Modeling coronavirus pandemic: comprehensive guide models. J Epidemiol, doi:10.1016/j.jclinepi.2020.12.002.Google name implies, Key Concepts series aims introduce new topical methodological developments. Van Smeden introductory prediction covering diagnosis prognosis. Whilst obvious relevance current introduces growing importance across care due accelerating precision medicine [[7]Van al.Clinical models: versus prognosis..J 2021; 142-145Abstract Another area use artificial intelligence applications within epidemiology synthesis. article van Dijk how text mining electric records shown accurate efficient way identify eligible patients collect cardiovascular manual screening [[8]van W.B. Fiolet A.T.L. Schuit Sammani Groenhof T.K.J. der Graaf al.Text-mining electronic healthcare used tool collection multicenter validation 97-105Abstract (3) Finally, principally about people, so good see submissions recruitment retention participants. issue, Brehaut patient focused theory-guided approach range barriers drivers participation existing tools [[9]Brehaut J.C. Carroll Presseau Richards D.P. Gordon Bénard al.A patient-focused, survey design identified participation.J 106-115Abstract help trialists optimise based behavioral theory, thereby reduce threats internal external validity caused sub-optimal loss follow poor retention.