Background: 

The quality of reporting of randomized controlled trials (RCTs) published in The Journal of Bone & Joint Surgery (JBJS) from 1988 to 2000 was previously analyzed. The purpose of this current study was to analyze the quality of reporting of RCTs published in JBJS from 2001 to 2013 to identify trends over time and potential areas of improvement for future clinical trials.

Methods: 

A manual search of the JBJS database identified RCTs published between January 2001 and December 2013. Quality assessments, using the Detsky quality-of-reporting index (Detsky score), a modified Cochrane risk-of-bias tool, and abstraction of relevant data identifying predictors of quality, were conducted.

Results: 

A total of 5,780 publications were identified in JBJS from 2001 to 2013, with 285 RCTs (4.9%), representing an increase from the prior 13-year period. The overall mean transformed Detsky score (and standard error) increased significantly (p < 0.001) from 68.1% ± 1.67% to 76.24% ± 0.72%. The percentage of multicenter RCTs decreased from 67% to 31%. The percentage of positive trials also decreased from 80% to 50.5%, as did the mean sample size (212 to 166). Regression analysis indicated that trials with an epidemiologist as the first author and nonsurgical trials were significantly associated (p = 0.001) with a higher overall trial quality score. The categories of the lowest mean methodology scores were randomization and concealment, eligibility criteria, and reasons for patient exclusion, as identified with the Detsky score, and patient and assessor blinding, as identified with the risk-of-bias assessment.

Conclusions: 

The quantity and quality of published RCTs in JBJS have increased in the 2001 to 2013 time period compared with the previous time period. Although these improvements are encouraging, trends to smaller, single-center trials were also observed. To efficiently determine the efficacy of orthopaedic treatments and limit bias, high-quality randomized trials of appropriate sample size and rigorous design are needed.

中文翻译：

---

2001年至2013年《骨与关节外科杂志》中随机对照试验质量的评估

背景： 

先前分析了1988年至2000年发表在《骨与关节外科杂志》（JBJS）上的随机对照试验（RCT）的报告质量。本研究的目的是分析2001年至2013年JBJS中发表的RCT的报告质量，以确定随时间变化的趋势以及未来临床试验的潜在改进领域。

方法： 

手动搜索JBJS数据库可确定2001年1月至2013年12月之间发布的RCT。使用Detsky报告质量指数（Detsky评分），改良的Cochrane偏见风险工具以及相关数据的抽象识别进行质量评估进行质量预测。

结果： 

从2001年到2013年，JBJS共发现5780份出版物，有285篇RCT（4.9％），比前13年有所增加。整体平均转换Detsky得分（和标准误）从68.1％±1.67％显着提高（p <0.001）至76.24％±0.72％。多中心RCT的百分比从67％下降到31％。阳性试验的百分比也从80％降至50.5％，平均样本量也从212降至166。回归分析表明，以流行病学家为第一作者的试验和非手术试验与较高的总体试验质量得分显着相关（p = 0.001）。方法学得分最低的类别是随机化和隐蔽性，资格标准以及患者被排除的原因（如Detsky得分所示），

结论： 

与前一个时期相比，JBJS中已发布的RCT的数量和质量在2001年至2013年期间有所增加。尽管这些改进令人鼓舞，但也观察到了较小的单中心试验的趋势。为了有效地确定骨科治疗的功效和限制偏倚，需要具有适当样本量和严格设计的高质量随机试验。