Department of Anesthesiology, University of Rochester School of Medicine and Dentistry, Rochester, NY, USA Department of Anesthesiology and Pain Medicine, University of Washington, Seattle, WA, USA Queen's University, Kingston, ON, Canada Analgesic Solutions, Natick, MA, USA Tufts University, Boston, MA, USA United States Food and Drug Administration, Bethesda, MA, USA Departments of Biostatistics and Computational Biology and Neurology, University of Rochester School of Medicine and Dentistry, Rochester, NY, USA California Pacific Medical Center Research Institute, San Francisco, CA, USA CHDI, New York, NY, USA Departments of Anesthesiology and Neurology and Center for Human Experimental Therapeutics, University of Rochester School of Medicine and Dentistry, Rochester, NY, USA.
Pain. 2012 Dec;153(12):2415-2421. doi: 10.1016/j.pain.2012.08.009. Epub 2012 Sep 15.
Recommendations for harms (ie, adverse events) reporting in randomized clinical trial publications were presented in a 2004 extension of the Consolidated Standards of Reporting Trials (CONSORT) statement. Our objectives were to assess harms reporting in 3 major pain journals (European Journal of Pain, Journal of Pain, and PAIN®) to determine whether harms reporting improved following publication of the 2004 CONSORT recommendations, and to examine study factors associated with adequacy of harms reporting. A total of 101 randomized, double-blind, noninvasive pharmacologic trials were identified in the 2000-2003 (epoch 1) and 2008-2011 (epoch 2) issues of these journals. Out of 10 reporting recommendations, the mean number fulfilled was 6.08 (SD2.65). Although more harms recommendations were fulfilled in epoch 2 (m(2)=6.49, SD2.66) than in epoch 1 (m(1)=5.39, SD2.52; P=0.04), only the recommendation to report harms per arm was satisfied by >90% of trials in epoch 2, whereas <60% reported withdrawals due to harms. Several trial characteristics (study design, participant type, pain type, frequency of treatment administration, treatment administration method, sponsor, and number of randomized participants) were significantly associated with harms reporting. However, when trial characteristics and epoch were entered into a multiple regression analysis, only trials studying pain patients, those using oral treatments, and industry-sponsored trials were associated with better harms reporting. Despite some improvement in harms reporting, greater improvement is needed to provide informative, consistent reporting of adverse events and safety in analgesic clinical trials.
2004 年 CONSORT 声明的扩展版提出了关于随机临床试验出版物中危害（即不良事件）报告的建议。我们的目的是评估三大疼痛期刊（《欧洲疼痛杂志》、《疼痛杂志》和《疼痛》）中的危害报告情况，以确定在 2004 年 CONSORT 建议发布后，危害报告是否有所改进，并检查与危害报告充分性相关的研究因素。在这些期刊的 2000-2003 年（时期 1）和 2008-2011 年（时期 2）的问题中，共确定了 101 项随机、双盲、非侵入性药物治疗试验。在 10 项报告建议中，平均满足的数量为 6.08（SD2.65）。尽管在时期 2 （m(2)=6.49，SD2.66）满足的危害建议数多于时期 1（m(1)=5.39，SD2.52；P=0.04），但只有报告每个手臂的危害的建议在时期 2 中被>90%的试验满足，而由于危害而退出的试验报告比例<60%。一些试验特征（研究设计、参与者类型、疼痛类型、治疗给药频率、治疗给药方法、赞助商和随机参与者数量）与危害报告显著相关。然而，当将试验特征和时期纳入多元回归分析时，只有研究疼痛患者的试验、使用口服治疗的试验和由行业赞助的试验与更好的危害报告相关。尽管危害报告有所改进，但仍需要进一步改进，以提供信息丰富、一致的镇痛临床试验中不良事件和安全性报告。
