Candy Bridget, Vickerstaff Victoria, Jones Louise, King Michael
Marie Curie Palliative Care Research Department, Division of Psychiatry, University College London, 6th Floor, Maple House, 149 Tottenham Court Road, London, W1T 7NF, UK.
Division of Psychiatry, University College London, 6th Floor, Maple House, 149 Tottenham Court Road, London, W1T 7NF, UK.
Trials. 2018 Feb 22;19(1):110. doi: 10.1186/s13063-018-2503-0.
Inadequate description of non-pharmacological complex interventions in trial publications means that they cannot be replicated or assessed for generalisability. There are published guidelines on how to describe an intervention, such as those from the CONSORT Group. However, there have been few evaluations of whether intervention reporting is improving.
We aimed to assess whether descriptions of multicomponent, non-pharmacological interventions evaluated in randomised trials are improving. To do so, we chose trials of educational and psychotherapeutic interventions to promote adherence to therapy, and compared those published between 2002 and 2007 (Time-1) with those between 2010 and 2015 (Time-2). These time periods were chosen to concord with the publication in 2008 of the CONSORT extension statement of reporting guidelines for non-pharmacological treatment which included items on intervention description. We assessed 19 items, based on the CONSORT Statement and the more recent Template for Intervention Description and Replication Checklist (TIDieR). Two reviewers independently extracted data. We created a quality score of the eight items we considered key information for replication and assessment of generalisability (setting, provider, recipient, comparator, intervention intensity, how it was conducted, existence of a manual or protocol, and detail of whether there was an assessment of fidelity). Score per item was '1' if reported adequately and '0' if not.
Of the eligible trials, 42 were published in Time-1 and 134 published in Time-2. The trials included were published in 112 peer-reviewed journals, 52 of these journals currently require authors to follow the CONSORT Statements, while only one recommended adherence to the TIDieR. Most items of CONSORT and TIDieR were reported by more than half of the trials at both time points. Few trials reported fidelity. A large proportion of the trials did not report the existence of a manual or protocol, or what the comparator group received. We found no statistically significant improvement in the eight-item quality score (Time-1: mean 5.71 (standard deviation (SD) 1.09), Time-2: 5.87 (SD 1.28), p = 0.49).
We found no overall evidence that reporting the specifics of multicomponent, non-pharmacological interventions is improving. Details to replicate interventions remain lacking, impairing best implementation or meaningful further research. Editorial endorsement of reporting checklists needs to be more extensive.
试验出版物中对非药物综合干预措施的描述不充分，这意味着它们无法被复制或评估其普遍性。关于如何描述干预措施已有一些已发表的指南，例如CONSORT小组的指南。然而，对于干预措施报告是否有所改进的评估却很少。
我们旨在评估随机试验中评估的多成分非药物干预措施的描述是否有所改进。为此，我们选择了教育和心理治疗干预措施的试验，以促进对治疗的依从性，并将2002年至2007年期间发表的试验（时间1）与2010年至2015年期间发表的试验（时间2）进行比较。选择这些时间段是为了与2008年发表的关于非药物治疗报告指南的CONSORT扩展声明相一致，该声明包括干预措施描述的项目。我们根据CONSORT声明和更新的干预措施描述与复制清单模板（TIDieR）评估了19个项目。两名评审员独立提取数据。我们为八个项目创建了一个质量得分，我们认为这些项目是复制和评估普遍性的关键信息（设置、提供者、接受者、对照、干预强度、实施方式、是否存在手册或方案，以及是否有保真度评估的详细信息）。如果报告充分，每个项目得分为“1”，否则为“0”。
在符合条件的试验中，42项在时间1发表，134项在时间2发表。纳入的试验发表在112种同行评审期刊上，其中52种期刊目前要求作者遵循CONSORT声明，而只有一种建议遵循TIDieR。CONSORT和TIDieR的大多数项目在两个时间点都有超过一半的试验报告。很少有试验报告保真度。很大一部分试验没有报告手册或方案的存在，也没有报告对照组接受了什么。我们发现八项质量得分没有统计学上的显著提高（时间1：平均5.71（标准差（SD）1.09），时间2：5.87（SD 1.28），p = 0.49）。
我们没有发现总体证据表明多成分非药物干预措施的具体报告有所改进。仍然缺乏复制干预措施的详细信息，这损害了最佳实施或有意义的进一步研究。报告清单的编辑认可需要更广泛。
