Data Interpretation in Analgesic Clinical Trials With Statistically Nonsignificant Primary Analyses: An ACTTION Systematic Review

JOURNAL OF PAIN

卷:16

Review

Gewandter, Jennifer S.¹  McKeown, Andrew¹  McDermott, Michael P.^2,3  Dworkin, Jordan D.⁵  Smith, Shannon M.¹  Gross, Robert A.^3,4  Hunsinger, Matthew⁶  Lin, Allison H.⁷  Rappaport, Bob A.⁷  Rice, Andrew S. C.⁸  Rowbotham, Michael C.⁹  Williams, Mark R.¹  Turk, Dennis C.¹⁰  Dworkin, Robert H.¹ 

[1] Univ Rochester, Sch Med & Dent, Dept Anesthesiol, Rochester, NY 14642 USA

[2] Univ Rochester, Sch Med & Dent, Dept Biostat & Computat Biol, Rochester, NY 14642 USA

[3] Univ Rochester, Sch Med & Dent, Dept Neurol, Rochester, NY 14642 USA

[4] Univ Rochester, Sch Med & Dent, Dept Pharmacol & Physiol, Rochester, NY 14642 USA

[5] Haverford Coll, Haverford, PA 19041 USA

[6] Univ Pacific, Sch Profess Psychol, Hillsboro, OR USA

[7] United States Food & Drug Adm, Silver Spring, MD USA

[8] Univ London Imperial Coll Sci Technol & Med, Dept Surg & Canc, London, England

[9] Calif Pacific Med Ctr, San Francisco, CA USA

[10] Univ Washington, Dept Anesthesiol & Pain Med, Seattle, WA 98195 USA

关键词: Spin;    misrepresentation;    randomized clinical trials;    ACTTION;    systematic review;   

DOI  :  10.1016/j.jpain.2014.10.003

来源: Elsevier

【 摘 要 】

Peer-reviewed publications of randomized clinical trials (RCTs) are the primary means of disseminating research findings. Spin in RCT publications is misrepresentation of statistically nonsignificant research findings to suggest treatment benefit. Spin can influence the way readers interpret clinical trials and use the information to make decisions about treatments and medical policies. The objective of this study was to determine the frequency with which 4 types of spin were used in publications of analgesic RCTs with nonsignificant primary analyses in 6 major pain journals. In the 76 articles included in our sample, 28% of the abstracts and 29% of the main texts emphasized secondary analyses with P values <.05; 22% of abstracts and 29% of texts emphasized treatment benefit based on nonsignificant primary results; 14% of abstracts and 18% of texts emphasized within-group improvements over time, rather than primary between-group comparisons; and 13% of abstracts and 10% of texts interpreted a nonsignificant difference between groups in a superiority study as comparable effectiveness. When considering the article conclusion sections, 21% did not mention the nonsignificant primary result, 22% were presented with no uncertainty or qualification, 30% did not acknowledge that future research was required, and 8% recommended the intervention for clinical use. Perspective: This article identifies relatively frequent spin in analgesic RCTs. These findings highlight a need for authors, reviewers, and editors to be more cognizant of how analgesic RCT results are presented and attempt to minimize spin in future clinical trial publications. (C) 2015 by the American Pain Society. Published by Elsevier Inc. All rights reserved

【 授权许可】

Free