社论：在《高级护理杂志》上发表文献综述的未来方向

分享智慧

共同成长

Full text

Doing a literature review has become an increasingly important type of research in recent years as the need to provide a synthesis of all the research relating to a particular topic is ever more crucial. The reasons for this are well described; nurses and healthcare professionals rely on research to provide evidence-based care. In many areas, though not all, we are fortunate enough to have a large body of evidence which can inform practice. It has long been accepted that it is not practical for busy practitioners to read every academic paper relating to their practice; indeed some have tried to calculate how long this would take and have demonstrated that such a task is incompatible with everyday duties (Fraser & Dunstan, 2010). Furthermore, any one academic paper provides just one piece of a jigsaw, it is important to see the whole picture of research rather than rely on just one piece. This is why well conducted literature reviews are so important in developing an evidence-based approach to care.

There are any different methods for doing a literature review. One of the most established methods is the original method for doing a systematic review and meta-analysis developed by the Cochrane Collaboration (Higgins et al., 2022). Their method for doing a systematic review and meta-analysis is widely accepted as the gold standard for doing a review of research that explores ‘what works?’, that is to identify if intervention or drug A is better than intervention or drug B. Given the focus of these review questions, these Cochrane Collaboration reviews contain mainly randomized controlled trials. Cochrane Collaboration review are often heralded as one of the greatest inventions of recent times as they summarize the evidence about the effectiveness of a care treatment or intervention and enable us to provide optimal care based on summaries of empirical evidence rather than relying on experience or tradition. In addition to intervention reviews, other types of reviews including diagnostic test accuracy and prognosis reviews are now published.

When the Cochrane Collaboration was established in 1993, methods for doing other forms of literature reviews, for example of qualitative studies, were also under development but were in their infancy. For example, Noblit and Hare published their text for the synthesis of qualitative studies in 1988. This early method for synthesizing qualitative methods was followed by other approaches, for example Walsh and Downe in 2005 and Thomas and Harden in 2008. At around this time, Whittemore and Knalf published their well-known approach to the synthesis of different types of research and non-research papers, using an integrative review (Whittemore & Knafl, 2005). These published approaches to doing a literature review reflect just a few methods that are available for those doing a literature review, which has a broader scope that identifying effective treatments or interventions.

The benefit of developing methods for doing a literature review for questions that fall outside the remit of the traditional systematic review is widely acknowledged as this brings rigour to a method that was previously undefined and often categorized, somewhat vaguely as a ‘narrative review’ (Greenhalgh, 2014). However it is also well documented that methods for doing a review have proliferated in recent times (Aveyard & Bradbury-Jones, 2019; Booth et al., 2012). There are many different names of reviews—Aveyard and Bradbury-Jones identified in excess of 35 different names used in published papers to describe a review—though it is not clear how different these methods really are in practice. Furthermore, stated methods for doing a review are not always adhered to in published literature reviews (Aveyard & Bradbury-Jones, 2019), and reviews are sometimes given alternative names that do not seem to correspond to an existing method (Aveyard & Bradbury-Jones, 2019). In addition, the term ‘systematic’ is often applied to a review name, presumably to indicate rigour, which can lead to confusion as it is not yet clear whether the term ‘systematic’ is one that is (or should be) synonymous with a literature review with meta-analysis or whether it should be used as an umbrella term to describe any review which has been undertaken in a systematic way. There is currently no clear consensus on this—there is also no right or wrong—but the inconsistent use of the term in different contexts is potentially confusing for readers who might wonder how a ‘systematic integrative review’ differs from an ‘integrative review.’

Aveyard and Bradbury-Jones (2019) called for clarity and consolidation in the methods used by researchers when doing a literature review. It was felt that the many different names and approaches to doing a review is potentially confusing for both readers and authors of reviews. This is especially so when a review was given a new name which did not seem to correspond to an existing approach and where the methods used were not clearly described. Adherence to a published method for doing a literature review was advocated to ensure that authors build on and consolidate existing methods rather than develop further approaches which might further complicate the field. Detailed adherence was also recommended to ensure that those doing a review attend to each aspect of the review and provide detail to demonstrate rigour. Yet in making such recommendations, we also need to be mindful of the need to be responsive to new methods for doing a literature review, where these are justified and which should not be precluded by rigid adherence to pre-specified methods.

The use of publishing guidelines raises important questions. Currently on the EQUATOR -network website (https://www.equator-network.org/reporting-guidelines/prisma/), there is reference to publishing guidelines which authors are generally required to adhere to prior to publication. These include the updated PRISMA guidelines (Page et al., 2021) for systematic reviews with meta-analysis and mixed methods review and the ENTREQ guidelines (Tong et al., 2012) for qualitative reviews. However due to the many different types of literature reviews, there is not an associated publishing guideline for every type of literature review. Therefore, it is not currently possible to require that all those doing a literature review adhere to an existing publication guideline as this risks the shoehorning some methods inappropriately into guidelines that may not be applicable. Both authors and editors need to be mindful of this. Therefore, the universal recourse to a publishing guideline is not necessarily a practical option. Whether existing guidelines, which are specific to certain review types, could be broadened to encompass a wider range of review types is a topic currently under discussion (Aveyard et al., 2021).

Despite the proliferation of names for doing a literature review, there are many common features of a review. All reviews have a focused research question with clear inclusion criteria and advocate a planned search strategy to demonstrate that the most relevant papers are included. Many reviews recommend a process of data extraction and critical appraisal to ensure the relevant aspects and relative merits of each paper are considered. All types of review require an appropriate level of synthesis so that new findings come from the review; often these findings should reflect a new interpretation of the papers which would not be evident from the reading each paper individually and should avoid simple lists of findings (Thorn, 2017). It is at the point of synthesis that a clear divergence between different types of reviews can often be found. Those doing a systematic review with meta-analysis will undertake a numerical re-analysis of the statistics in the included papers and where this is possible, the results will be presented as a meta-analysis though it should be noted that this is not always possible. Those doing an analysis involving the interpretation of the findings of different papers will present their results narratively.

At the Journal of Advanced Nursing, we are committed to the publication of well-conducted literature reviews. We welcome reviews that adhere to an established published method which is clearly referenced rather than review types where the method is unclear. Where there is the need for a new type of review, this should be clearly indicated in the paper and an explicit description of the method should be given. All reviewers should provide a clear and transparent account of the method used to do their review and any deviation from the published method should be explained. This should include a search strategy following clearly stated inclusion criteria, methods for data extraction and critical appraisal of the included papers. The approach to data analysis should be fully explained and should be in accordance with the method used to guide the review. We expect adherence to publication guidelines where this is appropriate and will happily discuss with authors where this is not the case. We also welcome contributions that develop the discussion surrounding doing a literature review and the complexities therein. We recognize that this is a complex area and it is one that with our authors, editors and readers we look forward to engaging with in further debate so that we move the science of doing a literature review forward.

全文翻译（仅供参考）

近年来，文献综述已成为一种越来越重要的研究类型，因为提供与特定主题相关的所有研究的综合需求变得越来越重要。其原因已得到很好的描述；护士和医疗保健专业人员依靠研究来提供循证护理。在许多领域，虽然不是全部，但我们很幸运拥有大量可以为实践提供信息的证据。长期以来，人们普遍认为忙碌的从业者阅读与其实践相关的每篇学术论文是不切实际的；事实上，有些人试图计算这需要多长时间，并证明这样的任务与日常工作不相容（Fraser & Dunstan,  2010）。此外，任何一篇学术论文都只提供了一个拼图，重要的是要看到研究的全貌，而不是仅仅依靠一张。这就是为什么进行良好的文献综述对于开发基于证据的护理方法如此重要的原因。

进行文献综述有任何不同的方法。最成熟的方法之一是由 Cochrane Collaboration 开发的系统评价和荟萃分析的原始方法（Higgins et al.,  2022）。他们进行系统回顾和荟萃分析的方法被广泛接受为对探索“什么有效？”的研究进行回顾的黄金标准，即确定干预或药物 A 是否优于干预或药物 B。作为这些综述问题的重点，这些 Cochrane Collaboration 综述主要包含随机对照试验。Cochrane Collaboration 综述通常被誉为近期最伟大的发明之一，因为它们总结了有关护理治疗或干预有效性的证据，使我们能够根据经验证据的总结提供最佳护理，而不是依赖经验或传统。除了干预审查之外，现在还发布了其他类型的审查，包括诊断测试准确性和预后审查。

当 Cochrane Collaboration 于 1993 年成立时，进行其他形式的文献综述的方法，例如定性研究，也正在开发中，但还处于起步阶段。例如，Noblit 和 Hare 在1988 年发表了他们的定性研究综合文本。这种用于综合定性方法的早期方法被其他方法采用，例如 2005 年的 Walsh 和 Downe 以及2008 年的 Thomas 和 Harden  。大约在这个时候，Whittemore 和 Knaf 发表了他们著名的方法来综合不同类型的研究和非研究论文，使用综合评论 (Whittemore & Knafl,  2005）。这些已发表的文献综述方法仅反映了可供文献综述者使用的几种方法，其范围更广，可以确定有效的治疗或干预措施。

开发方法对传统系统评价范围之外的问题进行文献综述的好处已得到广泛认可，因为这为以前未定义且经常分类的方法带来了严格性，有点模糊地称为“叙述性综述”（Greenhalgh ,  2014 年）。然而，也有据可查的是，最近进行审查的方法激增（Aveyard & Bradbury-Jones,  2019 ; Booth et al.,  2012）。评论有许多不同的名称——Aveyard 和 Bradbury-Jones 在已发表的论文中确定了超过 35 个不同的名称来描述评论——尽管目前尚不清楚这些方法在实践中的实际差异有多大。此外，已发表的文献综述（Aveyard & Bradbury-Jones,  2019）并不总是遵循所述的综述方法，并且有时会给综述赋予似乎与现有方法不对应的替代名称（Aveyard & Bradbury-Jones ,  2019）。此外，“系统性”一词通常用于评论名称，大概是为了表示严谨，这可能会导致混淆，因为尚不清楚“系统性”一词是否是（或应该）与文献综述和荟萃分析，或者它是否应该作为一个概括性术语来描述任何以系统方式进行的综述。目前对此没有明确的共识——也没有对错之分——但该术语在不同上下文中的不一致使用可能会让读者感到困惑，他们可能想知道“系统综合评价”与“综合评价”有何不同。

Aveyard 和 Bradbury-Jones（2019 年）) 要求研究人员在进行文献综述时使用的方法要清晰和统一。人们认为，进行评论的许多不同名称和方法可能会使评论的读者和作者感到困惑。当评论被赋予一个似乎与现有方法不对应的新名称并且所使用的方法没有明确描述时，尤其如此。提倡坚持已发表的文献综述方法，以确保作者建立和巩固现有方法，而不是开发可能使该领域进一步复杂化的进一步方法。还建议详细遵守，以确保进行审核的人员参加审核的各个方面并提供详细信息以证明其严谨性。然而在提出这样的建议时，

出版指南的使用提出了重要的问题。目前在 EQUATOR -network 网站 ( https://www.equator-network.org/reporting-guidelines/prisma/ ) 上，有对出版指南的参考，作者通常在出版前必须遵守这些指南。其中包括更新的 PRISMA 指南 (Page et al.,  2021 )，用于通过荟萃分析和混合方法进行系统评价，以及 ENTREQ 指南 (Tong et al., 2012 )) 进行定性审查。然而，由于文献综述的类型很多，因此并没有针对每种文献综述的相关出版指南。因此，目前不可能要求所有进行文献综述的人都遵守现有的出版指南，因为这有可能将某些方法不适当地硬塞进可能不适用的指南中。作者和编辑都需要注意这一点。因此，普遍求助于出版指南不一定是一种实用的选择。是否可以扩大特定于某些审查类型的现有指南以涵盖更广泛的审查类型是当前正在讨论的主题（Aveyard 等人，  2021 年）。

尽管进行文献综述的名称激增，但综述有许多共同特征。所有评论都有一个具有明确纳入标准的重点研究问题，并提倡有计划的搜索策略以证明包含最相关的论文。许多评论推荐了数据提取和批判性评估的过程，以确保考虑每篇论文的相关方面和相对优点。所有类型的审查都需要适当的综合水平，以便从审查中产生新的发现；通常，这些发现应该反映对论文的新解释，这在单独阅读每篇论文时并不明显，并且应该避免简单的发现列表（Thorn，  2017）。正是在综合点上，通常可以发现不同类型的评论之间存在明显的差异。那些通过荟萃分析进行系统审查的人将对所包含论文中的统计数据进行数值重新分析，并且在可能的情况下，结果将作为荟萃分析呈现，尽管应该注意这并不总是可能的. 那些进行涉及解释不同论文结果的分析的人将叙述他们的结果。

在高级护理杂志，我们致力于出版进行良好的文献综述。我们欢迎遵循明确引用的既定已发布方法的评论，而不是方法不明确的评论类型。如果需要新的审查类型，应在论文中明确指出，并明确说明方法。所有审稿人都应清楚、透明地说明审稿方法，并应解释与已发布方法的任何偏差。这应包括遵循明确规定的纳入标准、数据提取方法和对纳入论文的批判性评估的搜索策略。应充分解释数据分析的方法，并应与用于指导审查的方法一致。我们希望在适当的情况下遵守出版指南，并在不适合的情况下很乐意与作者讨论。我们也欢迎围绕文献综述及其复杂性展开讨论的贡献。我们认识到这是一个复杂的领域，我们期待与我们的作者、编辑和读者一起参与进一步的辩论，以便我们推动文献综述科学向前发展。

原文链接：

https://onlinelibrary.wiley.com/doi/10.1111/jan.15206

THE

END