Searching for Systematic Reviews CEBM 5-Day Workshop on Teaching Evidence-Based Practice 10th September 2007 Gemma Healy, Information Scientist, email@example.com PART 1: INTRODUCTION 1.1 Welcome! Bienvenidos! Khush amaadiid! Bienvenue! Willkommen! Foon ying! The aim of this session is to help you locate systematic reviews whether you have 2 minutes to spare, or a couple of hours at your disposal! In the course of the session we shall look at how to prioritise where to search for systematic reviews and let you get hands-on experience searching the best of these sources. We shall formulate clinical questions, devise search terms and learn how to select systematic reviews from the mixed bag of references some searches can throw up. You will get the opportunity to talk about your experiences of searching and what problems you encountered trying to find systematic reviews. Please feel free to share too, any juicy tips you have discovered! 1.2 Our starting point Let’s start by reminding ourselves what systematic reviews are: "Systematic reviews are concise summaries of the best available evidence that address sharply defined clinical questions" (Mulrow, 1987). Systematic reviews use explicit and reproducible methods to find, select, appraise, and synthesise evidence Systematic reviews can be summaries of different types of evidence, e.g. randomised controlled trials, observational studies, diagnostic studies or even other systematic reviews Systematic reviews may contain meta-analyses but remember not all meta-analyses may have conducted a systematic review!! The difference between Reviews and Systematic Reviews Systematic Reviews Reviews “freestyle” idiosyncratic narrative robust explicit reproducible Figure 1: 1.3 Who uses systematic reviews and for what purpose(s)? Well performed systematic reviews provide the most reliable and useful answers to healthcare questions. Systematic reviews have different users and purposes e.g.: Health professionals who need a quick answer to a clinical question Health professionals who are undertaking a research project as part of their CPD Guideline developers who need to identify the complete evidence base on which to draw up recommendations Trialists who need to provide justification for conducting research 1.4 Your Experiences Now let’s focus on you: Have you searched for systematic reviews/provided training in searching for systematic reviews in the past? Why were you searching for systematic reviews? Did you encounter any problems? Did you discover any useful tips? 1.5 Dependents when searching for systematic reviews Accessiblity of sources – computer access is needed and not all resources are free Familiarity with sources – we are creatures of habit and we may not like to try new places! Search skills – we all have different levels of skill/experience e.g. using Boolean operators (i.e. AND OR NOT), formulating questions Familiarity with how systematic reviews are reported – how can we tell if a reference is a systematic review if we only have a title and abstract? TIME and PURPOSE – where and how we search for systematic reviews is greatly influenced by how much time we have and the reason(s) for locating them. PART 2: LET’S GET SEARCHING 2.1 Let’s choose a question The quickest way to get you searching for systematic reviews is for us to choose a question and to search together. In real life, you will have different amounts of time available for you to do this, so we shall see where is best to search if you have 2 minutes, 20 minutes or 2 hours at your disposal. Example question: What is the evidence for the use of acupuncture in adults for treating insomnia? Patient/Population – Adults with insomnia Intervention - Acupuncture Comparison – None Outcome – A good night’s sleep! a 2 minute search - where do we look? Scenario: A patient attends your clinic for an appointment. They have brought with them a printout from the Internet that claims that acupuncture will cure their insomnia. They ask you, their doctor, whether they should try it out. Source bmjuppdates database http://bmjupdates.com/ index.asp How easy is it to access it? Free registration How easy is it to find systematic reviews on it? Very easy. It is possible to specify the type of research you are looking for. Can I trust what I find on it? Yes. Only systematic reviews that meet certain quality criteria are included. Instructions: 1. Go to http://bmjupdates.com/index.asp, click on “Register”, fill out your details 2. Click on “Search”, 3. Using the quick search box, type in insomnia and acupuncture 4. 2 systematic reviews are displayed, click on “View it” to see one of the full references (and the comments on clinical relevance and newsworthiness). 5. Click on “View the Abstract” to read the systematic review’s conclusion via PubMed. N.B. It is possible to do a detailed search but remember to select discipline and category along with entering search terms otherwise you will retrieve no search results a 20 minute search – where do we look? Scenario: You are a nurse undertaking a literature review of complementary therapies for insomnia. Part of the review is to assess whether acupuncture is an effective treatment for insomnia. Source TRIPdatabase http://www.tripdatabase. com/index.html How easy is it to access it? Free access worldwide How easy is it to find systematic reviews on it? Very easy. It is possible to specify the type of research you are looking for. Can I trust what I find on it? Yes. Only systematic reviews from sources that have been critically appraised are included Not exactly. Not every reference found using Clinical Queries will be a systematic review, and it is still necessary to appraise the systematic reviews that you find. PubMed (Clinical Queries) http://www.ncbi.nlm.nih. gov/sites/entrez Free access worldwide Quite easy. Using the Clinical Queries search, you can easily isolate references that are likely to be systematic reviews. Instructions for TRIP: 1. Go to http://www.tripdatabase.com/index.html, and click on the Advanced Search link. 2. Type in insomnia, select a title/text search, click on Go. There are 1119 results. 3. Type in acupuncture, and repeat. There are 634 results. 4. Type in #1 AND #2. There are 48 records, 11 of which are systematic reviews. 5. Click on the systematic reviews link – our search was a little sensitive so we have to scan the results for subject relevance. Click on the “I” to view conclusions only. TOP TRIP TIP: Click on the “See Synonyms Used” to get a quick list of synonyms to use for more in depth searches in other databases. Instructions for PubMed Clinical Queries: 1. Go to http://www.ncbi.nlm.nih.gov/sites/entrez 2. Click on the Clinical Queries link on the left-hand side menu, scroll down to “Find Systematic Reviews”. 3. Type in ((insomnia OR insomnia OR sleep disorder OR agrypnia OR sleeplessness OR hyposomnia OR wakefulness OR insomniac) AND (acupuncture or needle therapy)). These are synonyms you will have found through TRIP. 4. Your search will have looked for occurrences where these terms are mentioned in the title or abstract and PubMed will have automatically searched on relevant subject terms too. 5. There are 6 references for you to scan through. Some references you recognise from your previous searches, but how do you know if the others are systematic reviews? E.g. Chen et al 2007 cf Kalavapalli et al 2007 cf Li et al 2005. SUGGESTIONS FOR “SIFTING” SYSTEMATIC REVIEWS: 1. If the title or abstract states explicitly that it is a systematic review – then assume that it is such, until a full appraisal can be completed. 2. If there is no explicit statement then it is reasonable to expect that the abstract of a systematic review will include where it has searched. If it mentions the search of more than one source - assume that it is a systematic review, if it mentions a search of one source only – then it is possibly a systematic review, if it does not mention any searching – then this is probably a “freestyle” review or not a review at all!! a 2 hour search – where do we look? Scenario: You are part of a guideline group developing an evidence-based guideline on the assessment, diagnosis and treatment of chronic pain and insomnia. The guideline will include recommendations about the use of acupuncture for insomnia. Source EMBASE CINAHL Other specialty databases e.g. AMED, BNI How easy is it to access it? Subscription required. Check with your library provider for access How easy is it to find systematic reviews on it? Not so easy. These databases contain a real mixed bag of literature. Methodological search filters are required to isolate systematic reviews. Can I trust what I find on it? Not exactly. Not every reference found using a methodological search filter will be a systematic review, and it is still necessary to appraise the LILACs http://bases.bireme. br/cgibin/wxislind. exe/iah/online/ ?IsisScript=iah/ iah.xis&base= LILACS&lang=i IndMed http://indmed. nic.in/. Conference abstracts Free access worldwide Not so easy. Search terms relating to systematic reviews need to be used. Variable – depending on the conference organisers and the general rights to access their websites’ contents. Variable – some will have evidence categories, others will require search terms to isolate systematic reviews Google www.google.com Free access worldwide Not so easy. Search terms relating to systematic reviews need to be used. systematic reviews that you find. Not exactly. Not every reference found using terms related to systematic reviews will be a systematic review, and it is still necessary to appraise the systematic reviews that you find. It will still be necessary to appraise the systematic reviews that you find, and this might be difficult if the authors have not published full papers yet. Danger Danger Danger!!! No controls of any kind are in place, but if you have a very careful eye you might find something that all the other sources failed to throw up. PART 3: Glossary AMED AMED is a unique bibliographic database produced by the Health Care Information Service of the British Library. It covers a selection of journals in three separate subject areas: several professions allied to medicine, complementary medicine, palliative care. More info @ http://www.bl.uk/collections/health/amed.html. bmjupdates A new advanced information service to support clinical practitioners to keep up to date and to find current best evidence for practice, provided by the BMJ Publishing Group and the Health Information Research Unit at McMaster University. Quality filters are used to select articles from over 110 premier clinical journals. Cochrane reviews are also included. More info @ http://ebn.bmj.com/ cgi/content/full/8/2/39. BNI BNI is a nursing and midwifery database, covering over 200 UK journals and other English-language titles. It is produced by a partnership of UK libraries serving nurses, and is the most current nursing database for UK journals. More info @ http://www.gcal.ac.uk/library/resources/guides/bni.htm. Boolean Operators Boolean operators are when “AND” “OR” “NOT” are used in searching to either broaden or narrow search results. “OR” is used when you search a set of synonyms e.g. hayfever OR rhinitis. Using “OR” will find you more. “AND” is used when you are searching for more than one concept at a time e.g. cancer AND diet. “AND” will narrow your search. “NOT” is used when you want to limit your search in a particular way e.g. urinary tract infection NOT children, but it must be used with caution as it is easy to loose relevant references. CINAHL The Cumulative Index to Nursing & Allied Health (CINAHL) database provides authoritative coverage of the literature related to nursing and allied health. Virtually all English-language publications are indexed along with the publications of the American Nurses Association and the National League for Nursing. More info @ http://library.wellcome.ac.uk/doc_WT005543.html. The Cochrane Database of Systematic Reviews A database containing all Cochrane Reviews and Cochrane Protocols (Cochrane Reviews in progress) and prepared by members of the Cochrane Collaboration. Cochrane Reviews are considered to use a gold-standard methodology in their preparation. More info @ http://www.update software.com/publications /cochrane/. The Cochrane Library A meta-database of the Cochrane Collaboration available via Wiley Interscience, containing individual databases of good quality primary and secondary research. The Cochrane Database of Systematic Reviews, DARE and the HTA Database are amongst these. More info @ http://www.update software.com/publications/ cochrane/. DARE The Database of Abstracts of Reviews for the Effects of Treatments contains systematic reviews that have been appraised by the Centre for Reviews and Dissemination (CRD), York, UK. More info @ http://www.york.ac.uk/inst/crd/ darefaq.htm EMBASE EMBASE is a biomedical and pharmacological database which gives you access to the most up-to-date information about medical and drug-related subjects. More info @ http://www.elsevier.com/wps/find/bibliographicdatabasedescription.cws_ home/523328/description#description. Google Google is one of the most popular Internet search engines. More info @ http://www.google.co.uk/intl/en/about.html. HTA Database The Health Technology Assessment (HTA) database brings together details of completed and ongoing health technology assessments from around the world. The abstracts in the database are descriptive rather than analytical and do not form critical appraisals of the reports. The database is produced in collaboration with the INAHTA Secretariat, based at SBU, Sweden. More info @ http://www .york.ac.uk/inst/crd/crddatabases.htm#HTA. IndMed IndMed is a bibliographic database of Indian biomedical journals. The database is maintained by the Indian Medlars Centre. More info @ http://indmed.nic.in/. Information Scientist An information Scientist is quite like a librarian except that they have two heads instead of three. Information Scientists specialise in information retrieval; selection, organisation and presentation. In evidence-based medicine they focus on the first two steps of the EBM model i.e. formulating questions and searching and sifting the evidence. LILACS LILACS database - Latin American and Caribbean Health Sciences Literature - is a cooperative database built by the institutions which integrate the Latin American and Caribbean of Health Sciences Information System. The main objectives of this database are the bibliographic control and the dissemination of health scientific-technique literature from Latin American and Caribbean countries, absentee from the international databases. Meta Analysis In statistics, a meta-analysis combines the result of several studies that address a set of related research hypotheses. The first meta-analysis was performed by Karl Pearson in 1904, in an attempt to overcome the problem of reduced statistical power in studies with small sample sizes; analyzing the results from a group of studies can allow more accurate data analysis. Methodological Search Filters Pre-tested search strategies that combine search terms relating to specific study designs so that users of bibliographic databases can quickly find the type of evidence relating to their question. There are methodological search filters available for systematic reviews, RCTs, observational studies, diagnostic studies etc. More info @ http://www.york.ac.uk/inst/crd/intertasc/filters.htm. PubMed PubMed is developed and maintained by the National Center for Biotechnology Information (NCBI) at the U.S. National Library of Medicine (NLM). PubMed lets you search millions of bibliographic citations and abstracts in the fields of medicine, nursing, dentistry, veterinary medicine, the health care system, and preclinical sciences. It provides access to MEDLINE and to articles in selected life sciences journals not included in MEDLINE. More info @ http://www.ncbi.nlm .nih.gov/books/bv.fcgi?rid=helppubmed.chapter.pubmedhelp. Systematic Review A summary of the medical literature that uses explicit methods to perform a thorough literature search and critical appraisal of individual studies and that uses appropriate statistical techniques to combine these valid studies. Systematic reviews are not all equal, and quality issues are important. TRIPdatabase A meta-database containing secondary and primary evidence from over 100 critically assessed evidence resources including The Cochrane Database of Systematic Reviews, DARE, Health Technology Assessment (HTA) Database (from the Cochrane Library). More info @ http://www.tripdatabase.com/aboutus/ index.html.