코크란 문헌고찰 작성 입문 Introduction to writing a Cochrane review

학습 목적 Course learning objectives 코크란 체계적 문헌고찰의 준비에 필요한 전반적 방법을 이해함 코크란 체계적 문헌고찰을 위한 연구계획서 초안을 작성할 수 있음 연구 수행 시 추가적 도움을 받는 방법을 알게 됨 These are the overall objectives for the two-day introductory course for review authors.

개요 Outline 코크란 연합 (The Cochrane Collaboration) 소개 코크란 문헌고찰의 목적 및 구조 문헌고찰 작성의 논리 In this session, we’ll begin by given you an introduction to The Cochrane Collaboration as an organisation – how we work, and what our aims are. Then we’ll look at the aims of Cochrane systematic reviews, and an overview of how we go about writing them. Finally, we’ll look at some of the practical aspects – the logistics of writing a review. There’s more information about The Cochrane Collaboration in Chapter 1 of the Handbook. See Chapter 1 of the Handbook

The Cochrane Handbook for Systematic Reviews of Interventions 문헌고찰의 전체 과정에 필수적 지침서 아래 방법으로 참조 가능 online www.cochrane.org/training/cochrane-handbook 추가 및 교정된 내용 포함됨 RevMan의 Help 메뉴 단행본 구매 (Wiley Blackwell) Throughout this workshop, we will be referring to the Cochrane Handbook of Systematic Reviews. This Handbook is an essential guide for review authors, outlining in detail all the methods we will be discussing during the workshop, as well as more advanced topics, and I strongly recommend that everyone makes use of it. Whenever you have a question about your review, try checking the Handbook first. Throughout the workshop, we’ll give references to the appropriate chapters of the Handbook where you can look for further information. The Handbook is available free of charge, either via the internet or from the Help menu of the RevMan software that we use to write Cochrane reviews. You can access the RevMan version without an internet connection. If you choose, it is possible to purchase a hard copy of the Handbook for around US$50 or €50 from either Wiley Blackwell online, amazon.com or your local bookseller. Note: Participants often comment that it is difficult to print from these electronic versions. PDF files of each chapter are available in the Handbook Editorial Advisory Committee files folder in Archie. As most authors are not familiar with Archie at this stage, trainers may wish to demonstrate this later in the workshop (e.g. in a break or during the RevMan session), or recommend that authors contact their CRG for advice if they wish to access the pdf files later. Look out for pointers to relevant chapters

Archie Cochrane “주 전공 및 세부 전공 분야별로 연관된 모든 무작위 대조 연구에 대한 비판적 요약을 체계적으로 수행하지 못하고, 그 내용을 주기적으로 갱신하지 않았던 것은 진실로 우리 전문가들이 중대하게 비판 받아야 할 점이다.” Cochrane 1979 This is Archie Cochrane, an epidemiologist from the UK. In 1979, Archie suggested that it would be a great benefit to the medical profession if all the results of all the randomised trials were collected together and summarised, so they could be better used to inform decision-making. This was such a good idea, that when the Cochrane Collaboration was formed in 1993, it was named in his honour. “It is surely a great criticism of our profession that we have not organised a critical summary, by specialty or subspecialty, adapted periodically, of all relevant randomised controlled trials.” Cochrane 1979 Photograph: Cardiff University Library, Cochrane Archive, University Hospital Llandough

지향 및 목적 Mission and aims 국제적 비영리 단체로서, 보건의료 중재법의 효과에 대한 체계적 문헌고찰 결과를 항상 사용할 수 있도록 준비, 유지하고 그 활용을 장려함으로써, 관련 정보를 충분히 파악하고 고지받은 상태에서 보건의료 의사결정을 내리도록 돕는 것을 지향함 아래 목표를 지향함 협력적이고 능률적임 비뚤림이 없고, 신뢰할 수 있으며, 최신 결과를 반영함 시의 적절하며, 활용 가능함 The Cochrane Collaboration is an international, non-profit organisation. Our aim is to help people make well-informed decisions about health. We do that by writing and disseminating systematic reviews, so that all the relevant evidence is available when people come to make decisions, whether it’s their own health or their family’s, whether they are health professionals providing care, or policy makers making decisions about health systems or community-level programs. Within the Cochrane Collaboration we try to work according to some important principles. We try to be collaborative, encouraging participation and building on the enthusiasm of individuals. We try to be efficient – many of our contributors are volunteers, and we want to be sure that we are making the most of everyone’s contribution. We try to provide information that is unbiased, reliable and up to date. We also want our reviews to be relevant internationally and across different contexts of health care, and we want our information to be accessible, which is why, for example, we make The Cochrane Library available free of charge in all low-income countries.

코크란 연합의 구조 Our structure 100 개 이상의 국가에서 28,000 명 이상의 사람들이 참여하고 있음 대부분 자발적 참여자임 탈중앙화된 구조 52 개의 코크란 리뷰 그룹 Cochrane Review Groups (CRGs) 특정 보건의료 분야 별로 분류 코크란 문헌고찰 저자들의 최우선 연락처 국제적이며 다학제적임 사무국 및 코크란 편집국 (영국) 방법론 연구그룹 (Methods Groups), Fields, Centres, 소비자 네트워크 (Consumer Network) 등이 지원함 대체로 정부 연구비로 지원됨. Worldwide, the Cochrane Collaboration has over 27,000 people in 100 countries contributing in some way, whether as authors, editors, peer referees, etc. Almost all these people are volunteers. We operate a very decentralised structure. We have a small Secretariat in Oxford and an Editor in Chief’s office in London, but the majority of the work of the Collaboration is done by Cochrane Review Groups (CRGs). There are currently 52 CRGs, each responsible for supporting and publishing reviews in a different area of health (e.g. Heart Group, Breast Cancer Group, Infectious Diseases Group, etc.). The CRGs work internationally and across disciplines. If you live in Argentina and your review is on diabetes, you will work with the Metabolic and Endocrine Disorders Group in Germany. If your review is on arthritis, you will work with the Musculoskeletal Group, whether you are a GP, rheumatologist, podiatrist or occupational therapist. As authors, your CRG is always your first point of contact, and they will support you throughout the review process. The Collaboration is supported by a range of other groups. Methods Groups bring together leading researchers in statistics, bias, literature searching, etc., to develop the methods we use to write our reviews, and they’re always working to see how our methods can be improved. Fields promote the development and use of systematic reviews in areas that cut across health topics (e.g. primary care, child health), but aren’t involved in the editorial role that CRGs provide. Centres provide support across a geographical area, providing training and support to authors, and promoting systematic reviews and supporting the other Cochrane entities located in their region. We are also supported by the Cochrane Consumer Network, an international network of consumer volunteers who act as peer referees and authors, and help to communicate and disseminate the results of reviews. Cochrane activities are generally funded by government grants in the countries where they each group is located. We have a strict policy against commercial funding of our activities. We do generate some funds from subscriptions to The Cochrane Library, and these fund the Secretariat and Editorial Unit, as well as providing a small amount of funding for special projects, such as software and methods development.

국제적 분포도 An international organisation To give you an idea of our international reach, this map shows only the Centres and their Branches, not the Review Groups. You can see that we are working on every continent, and that helps us keep our perspective on what are the important questions to answer, where the evidence might come from, and how the evidence might be applied in different contexts. It’s important for you, as authors, to always keep that international perspective in mind – your review may be read and used by people anywhere in the world, in contexts very different to your own. www.cochrane.org/contact/centres_map

All our reviews are published online in The Cochrane Library. ASK: Has anyone accessed Cochrane reviews in the Library before? The part of the library containing Cochrane reviews is called the Cochrane Database of Systematic Reviews. As of August 2010 it contained 4331 reviews and 1993 protocols for reviews in progress. The Library is published monthly, and currently has an impact factor of 5.65. On the Library homepage, you’ll also find editorials, special collections of reviews (e.g. around H1N1 flu, disaster relief following major natural disasters, etc.), podcasts and resources for your local journal club. To find out more about what’s in the Library, explore the Learn menu. www.thecochranelibrary.com

One thing you might like to learn more about are the other databases included in the Library – there’s more to the Library than Cochrane reviews. It contains several other databases, including the Cochrane Central Register of Controlled Trials, containing over 625,000 records of randomised and quasi-randomised trials, as well as other databases of non-Cochrane reviews, economic evidence, research methodology studies, and more. Any time you search the Library, you can also find results from these other databases, just by clicking the links along the top of your search results.

개요 Overview 코크란 연합 (The Cochrane Collaboration) 소개 코크란 체계적 문헌고찰의 목적 및 구조 문헌고찰 작성의 논리 So that’s the Cochrane Collaboration. But what about our reviews? How are Cochrane reviews put together, and what do we aim to achieve by systematically reviewing the evidence?

왜 체계적 문헌고찰이 필요한가? Why systematic reviews? 전반적 연구 현황을 알 수 있는 효과적인 방법 검색에 필요한 시간 절약 비판적 평가 결과 해석 각 연구 간 차이점 탐색 의사결정 시 신뢰할 만한 근거 제공 관련 정보 선택 시 비뚤림 배제 보건의료, 정책, 향후 연구에 유용함 There are three main reasons why systematic reviews are useful: first, they are efficient, bringing together the whole body of research answering a specific question, and saving readers the time and expertise required to locate, appraise and interpret the results. secondly, unlike a single study, a systematic review can help us explore the differences between studies, identifying those factors that influence the effect of an intervention. thirdly, they aim to be reliable, presenting an unbiased, comprehensive picture of the available evidence, that can then be used to support decisions in health care, in policy, or to identify areas where future research may be needed. Traditional narrative reviews have not had this advantage – it’s often not clear how the evidence was selected for inclusion in reviews – perhaps only studies conducted in the last five years, only studies published in English, only studies conducted in large US teaching hospitals, or only studies that agree with the author’s opinion.

체계적 문헌고찰의 핵심 요소 Key features of a systematic review 연구목적의 명확한 서술 선정 기준의 사전 정의 명백하며 재현 가능한 연구방법론 체계적 문헌 검색 선정된 연구의 타당도 평가 분석결과의 체계적 합성 및 결과 제시 There are key features that should go into any systematic review, whether Cochrane or not: first, a review should have clearly stated objectives, specifying the question to be answered, and determining what evidence will be sought. the objectives should determine clear eligibility criteria, so that all the relevant evidence is identified, and clear, objective decisions can be made about which studies are included in or excluded from the review. throughout the review, all the methods used should be explicit and transparent, so that in theory, someone could follow along, repeat what you’ve done, and come to the same conclusions, or at least see clearly where they might have made a different decision. once the eligibility and methods are planned, there is a systematic search of the literature, to find all the relevant evidence. the included studies are then carefully assessed, identifying those that might be at risk of bias or inaccurate results. finally, the results of all those included studies are systematically reported, and a synthesis of their findings is presented.

코크란 체계적 문헌고찰의 과정 Steps of a Cochrane systematic review 연구 질문 설정 선정 기준 계획 연구 방법 계획 연구자료 검색 선정 기준 적용 자료 수집 분석 대상 연구들의 비뚤림 위험 평가 결과 분석 및 제시 분석결과의 해석 및 결론 도출 문헌고찰 개선 및 주기적 갱신 연구주제 등록 연구계획서 출판 For a Cochrane review, we translate those principles into these steps. We start by formulating the question, and by registering the proposed review title with the relevant Cochrane Review Group, so we can keep track of which reviews are in progress. Then, we plan our eligibility criteria and methods in advance, and these are published as a protocol in The Cochrane Library. Only then do we proceed with the review: running a systematic search to identify all the included studies, collecting information from each study, assessing each study for bias, analysing and interpreting their results. The review is then published. Cochrane reviews add another step to this process: every two years, we aim to update our reviews, to keep up with newly published studies. Some reviews are updated more frequently, if the literature is fast-moving and many new studies are published. Some are updated less frequently, if there are few new studies and the findings of the review are clear. We’ll be working through each step in this process during this workshop, and looking at it in more detail. 문헌고찰 출판 문헌고찰의 갱신 출판

개요 Overview 코크란 연합 (The Cochrane Collaboration) 소개 코크란 체계적 문헌고찰의 목적 및 구조 문헌고찰 작성의 논리 Before we move on to the detail of these methods, let’s look first at some of the practical steps – the logistics of getting started with a Cochrane review. You’ll find more information about this in Chapter 2 of the Handbook. See Chapter 2 of the Handbook

동기 Motivation 서로 상충되는 근거의 해결책 임상진료 지원 향후 연구의 필요성 파악 의사결정을 돕기 위한 근거 요약 근거가 불확실한 분야를 파악함 변이(variations) 탐색 현행 임상진료의 적절/부적절성 판단 향후 연구의 필요성 파악 Each author will have different motivations for wanting to address a particular question. The overarching aim is to summarise the evidence, to help people understand and support decision-making. This overarching goal should help us select questions that need answering, and plan our reviews to make sure the right information is gathered and presented to answer those questions. Sometimes systematic reviews can resolve conflicting evidence around particular interventions, helping us reach a clear conclusion about their effectiveness. In practice, a systematic review can help us reach a decision about interventions we’re uncertain about, or about the pros and cons of the different options or variations in current practice. They can help us confirm that current practice is best practice, or they can help us identify problems. Importantly, they can identify gaps, where we don’t have enough evidence to answer our questions, and where future research is needed. 의사결정을 돕기 위한 근거 요약

문헌고찰 연구진 The review team 문헌고찰은 2명 이상의 연구진이 수행해야 함. 이중 검정이 필요함 문헌고찰에 포함된 연구자료의 선정기준 충족 여부 자료 추출 및 입력 비뚤림 위험 평가 상이한 분야의 전문가 임상 분야 (다양한 전공) 체계적 문헌고찰 방법론 전문가 (통계학자 포함) 사용자 관점 고려 (소비자, 임상전문가, 보건의료 환경 등) 자문 그룹의 구성을 고려할 것 All Cochrane reviews must be written by more than one person. This helps to share the workload, but it has other important advantages. There are some points in the review process where work should be double-checked, such as the final decisions about including studies in the review, data collection and entry, and decisions about the risk of bias of your included studies. Also, multiple authors can provide different perspectives and different areas of expertise, which can improve the quality of your review. You may have authors from different disciplines, or different geographical settings. First time review authors are encouraged to work with others who are experienced, and having someone on the team with methods expertise is always an advantage. Your CRG may have minimum criteria for the skills and members of your review team. It’s important to consider the users of your review from the beginning of this process – this can include different professional perspectives, but the consumer perspective is particularly important. Consumers may have vital insight into the scope of your review, the important outcomes to be measured, accessibility of the intervention, and variability in the circumstances and perspectives of affected individuals. Consumers are usually included in the peer review stage of protocols and reviews, but can also be involved directly, throughout the process. The perspectives of people from different settings in which the health issue or intervention occurs should also be considered, for example reviews of malaria should be informed by the perspective of people from affected low- and middle-income countries. Interventions that are not specific to resource-poor settings should also consider the different access, delivery and outcome challenges in those areas. Your CRG may be able to recommend consumers and colleagues active in the field who may be able to advise on your review. There are options available as to how to engage these perspectives in the review. You may wish to involve people as authors, or consult them for advice which is acknowledged appropriately in the published review. You may wish to set up a formal Advisory Group for your review, including a range of stakeholders to be regularly consulted for discussion and advice during the review process, including scoping the review, identifying important background materials, interpreting the findings, and disseminating the findings to relevant users.

CRG 편집 과정 CRG editorial processes 연구주제 등록 중복 방지 연구주제가 적절하고 명확한지 확인 지속적인 지원 궁금점에 대한 우선 문의처 표준화된 연구방법 및 리뷰 양식 템플릿 (예; 자료 추출 양식) 문헌 검색 통계 전문가 출판 동료 평가 제도 출판 여부 결정 Once you’ve put together your team, your first step is to register your title with the relevant CRG. This prevents duplication, so that no-one else is working on the same topic, and also ensures that your topic is appropriate and well-defined. CRGs may choose not to register your proposed topic. For example, some groups prioritise topics to be registered based on global health priorities and burden of disease. Alternatively, your topic may overlap with an existing review, or it may wish to consider a question that is not aligned with the scope of the Cochrane Collaboration, for example questions of aetiology or prevalence. The CRG may wish to negotiate the topic with you, for example they may feel the scope is too broad to be manageable, or alternatively they may wish you to expand the scope of your review to cover one or more related interventions for which a review is needed. Once the review is registered, the CRG will provide ongoing support and assistance. As authors, they are your first point of contact for help with your review. Importantly, your CRG has experts in literature searching and statistical methods who can help with your review. The CRG may provide templates and tools for your review. Finally, the CRG is responsible for editing and publishing your review, including managing the peer review process. Like any journal, your CRG reserves the right not to publish a protocol or review if it doesn’t meet the Cochrane Collaboration’s editorial standards.

연구주제 등록 양식 Title registration form 연구진 관련 정보 성명, 전문 분야, 이해 상충 역할 및 책임 분담 연구 질문 제기 대상 집단 및 질환, 평가하고자 하는 치료기술/중재방법, 평가도구 및 문헌고찰에 포함시킬 원 자료의 연구 디자인 활용가능한 자원 파악 연구계획서/문헌고찰 작성 예상 소요 시간 추정 The Cochrane Library 에 연구결과 출판 동의 문헌고찰의 주기적 갱신 서약 At the title registration stage, you will be asked to complete a Title Registration Form to outline your plans. ASK: has anyone already submitted one of these forms? What did it include? CRGs need as much detail as possible to help clarify the scope of your proposed review, and decide whether it should be registered. As well as scoping out your topic in detail, you will be asked to provide other details, such as the members of your review team, including areas of expertise and systematic review experience. The timeframes for completion are important. The CRG will monitor your progress, and will reserve the right to assign the topic to a different team of authors if the review is not progressing. The timeline for a review is usually up to two years, depending on the time and resources available and the complexity of the topic. All Cochrane authors agree to publish their review first in The Cochrane Library. You may wish to co-publish your review in other peer reviewed journals or books – and many journals are enthusiastic about co-publishing Cochrane reviews, often in a shorter form. The Collaboration is happy for this to happen with one essential condition: Cochrane reviews must remain free for dissemination in any and all media, without restriction from any of them. Cochrane authors grant the Collaboration a worldwide licence to disseminate Cochrane reviews, and cannot sign exclusive copyright over to any other journal or publisher. See Chapter 2 of the Handbook and contact your CRG for more information if you would like to make arrangements for co-publication. Reviews previously published in other journals can be converted and published as Cochrane reviews. This often requires a lot of work and modification, such that the review can essentially be treated as a different publication. If the Cochrane version is essentially the same as a previously published version, permission should be sought from the previous publishers.

이해관계에 대한 고지 Declaration of interest 영리단체에 의한 연구지원은 불허함 기타 지원단체들 역시, 연구결과의 출판을 지연, 방해하거나 연구자의 독립성을 훼손할 수 없음 잠재적 이해관계는 모두 공개되어야 함 재정적 이해 (예: 모든 종류의 연구지원금 및 현물지원) 개인적 이해 (예: 문헌고찰 연구진 중, 자신의 연구가 해당 문헌고찰에 잠정적으로 포함될 [가능성이 있는]경우) “...the only person who does not have some sort of vested interest in a subject is somebody who knows nothing about it at all.” (Smith R. Conflict of interest and the BMJ. BMJ 1994; 308: 4-5). The Title Registration Form will also ask you and your co-authors to complete a declaration of interest, and you will be asked to do so again before the publication of your protocol and the completed review. We take this very seriously, and Cochrane authors are asked to be as transparent as possible about any interested, commercial or in kind, that may be perceived as a conflict, whether or not you think the conflict is important. Importantly, direct commercial sponsorship of reviews is prohibited. See Code of Conduct, Box 2.6.a in the Handbook

문헌고찰에 필요한 자원 Resources for your review 시간 각 저자들이 얼마나 많은 시간을 할애할 수 있는가? 누가, 어떤 일을 해야 하는가? 각각의 업무 수행에 얼마나 많은 시간이 걸리는가? 전체적으로 1-2년까지 소요될 수 있음 전문가 도서관 (데이터베이스 접근성, 도서관 상호 대출, 조언) 통계 전문가 사무 환경 및 기술적 지원 컴퓨터, 인터넷, 전화/컨퍼런스, 프린터, 복사기 The most important resource needed for your review is time. Consider mapping out a time chart of tasks, with target dates, and get a clear picture from each of your co-authors about the amount of time they have available to work on the review. A review may take up to 1-2 years to complete, but this will vary, depending on the amount of time available to dedicate to the review. You will need some expert advice, particularly around literature searching and statistical analysis. Your CRG can assist with these, but if you have access to local experts, this can be very valuable. Your local library will also be important in providing access to electronic literature databases for searching, and for obtaining copies of journal articles once your search has been run. If you don’t have local access to a good library, talk to your CRG. You’ll also need some office supplies and technology – computers and internet access for writing and submitting your review, any resources you need to ensure that the members of your team are communicating regularly, such as by teleconference if you’re not all in the same place, and some basics, like printing. Funding may be available to support priority systematic reviews from the various organisations that fund health research, from service providers, policy agencies, guideline developers, etc. Most Cochrane authors, however, provide their time voluntarily.

Review Manager (RevMan) 필수적 소프트웨어 Archie database를 통해 문헌고찰 억세스 가능 연구계획서/리뷰 템플릿 사용 가능 리뷰 내용 작성 통계 분석 편집 심사 및 출판 ID / 비밀번호 필요 소속된 CRG에 문의 www.ims.cochrane.org/revman 에서 다운로드 가능 Review Manager (RevMan) is the Cochrane Collaboration software that you must use to write your review. RevMan is available to download free of charge from our website, and is specifically designed to help you with your review, providing templates to help you write the text, and performing the statistical analysis for your review. RevMan will also help you save your drafts to our central database, called Archie, and to submit your review for editorial approval. The Archie database is then used to publish the online Cochrane Library. We’ll talk about RevMan and Archie in more detail in a separate session, but a user guide and self-paced tutorial are available from the Help menu in RevMan, and more documentation is available on the website. See user guide and tutorial on the RevMan Help menu

자료 출처 Sources of information The Cochrane Library www.thecochranelibrary.com 코크란 연합 홈페이지 Cochrane Collaboration www.cochrane.org 코크란 리뷰 그룹 홈페이지 (CRG websites) www.cochrane.org/contact/review-groups 문의사항은 Reference Centre 에 연락 가능 www.cochrane.org/contact/centres There are lots of resources available if you’d like more information about the Cochrane Collaboration, Cochrane systematic reviews or the practicalities of the review process. Have a look at our reviews in The Cochrane Library, and the information on the Cochrane Collaboration website, including contact information for all the Cochrane Review Groups. If you have questions and you’re not sure who to contact, you’re also welcome to contact your local Cochrane Centre.

핵심 전달 사항 Take home message 코크란 체계적 문헌고찰은 보건의료 의사결정을 지원하기 위한 신뢰성 있는 정보 제공을 목적으로 함 문헌고찰 작성을 위한 최우선 연락처는 해당 연구주제와 관련된 코크란 리뷰 그룹 (CRG) 임 연구진 구성 및 연구수행에 필요한 자원 파악부터 시작할 것

References Acknowledgements Green S, Higgins JPT, Alderson P, Clarke M, Mulrow CD, Oxman AD. Chapter 1: Introduction. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org. Green S, Higgins JPT (editors). Chapter 2: Preparing a Cochrane review. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org. Cochrane AL. 1931-1971: a critical review, with particular reference to the medical profession. In: Medicines for the year 2000. London: Office of Health Economics, 1979, 1-11. Acknowledgements 편집: Miranda Cumpston 호주 코크란 센터 (Australasian Cochrane Centre) 및 캐나다 코크란 센터(Canadian Cochrane Centre) 교육자료를 원자료로 함 본 교육자료는 Cochrane Methods Board 가 승인하였음 Translated by Kun Hyung Kim, Myeong Soo Lee and Byung-Cheul Shin