Have you ever heard that song by U2 and thought how easily that could apply to searching for evidence? Perhaps not, but it certainly reflects how many health professionals feel when faced with the task of searching for research evidence to support their practice.
Since the mid-1980s and particularly since the rise of the Internet and the World Wide Web in the 1990s, there has been an 'information revolution' in how knowledge is created, stored, catalogued, published and indexed. This information revolution has had such an impact on how we do business, live our lives or even simply process data that we are now experiencing what commentators call 'information overload'. An interesting comment was made in the late 1990s that a particular event with momentous implications went unnoticed: it was the day (perhaps in the late 1970s or early 1980s), when information was transformed from a 'scarce commodity' that had to be found, processed, marketed and sold, into something that was suddenly in excess supply.
Such is the rate of data being made available every day that the information revolution makes it virtually impossible for people to read everything on any particular subject. And in this scenario people who work in the health sciences are no exception. In the health sciences, Internet search engines and oft-used databases such as MEDLINE, Cochrane Library, EMBASE, PsycINFO and CINAHL cannot hope to catalogue or index everything. So, if you are new to research and you are looking for evidence about nursing practice, how would you go about it? There are essentially three key things you need to keep in mind when searching for evidence: how to frame the terms or key words you use; which databases and search engines to use; and how to assess the quality of the information you find.
According to one study , nurses felt more confident asking colleagues, peers and searching the Internet to find information to inform their practice than using databases such as MEDLINE or CINAHL. The sorts of issues identified in this study are commonplace and understandable. Health professionals are busy providing care and so time and access to information can prove difficult. While many would consider their search skills to be somewhat successful, many are still not confident when searching the Internet or specific databases.
The importance of basing practice on research evidence has never been higher, but health professionals across the board are still struggling with how to identify relevant information and appraise its quality in order to make use of it in their clinical decision-making.
The ability to search bibliographic databases effectively is an essential skill for anyone undertaking research in health. However, before you proceed 'helter skelter' to a computer terminal, click on a database and type in what you think are the key words or phrases on a topic, you first have to consider the exact information you are looking for. This may well mean rephrasing your topic as a question and reframing it into its major concepts, for example: firstly, the patient population or problem being addressed; secondly, the intervention under consideration; thirdly, comparing the intervention, where relevant; and fourthly, clinical outcomes of interest.
For example, you may want to investigate 'peanut allergies in children' and if you set this query out as follows:
Patient >Children
Exposure >Peanuts
Outcome >Allergy
You then start to think about alternative words or terms, synonyms or spelling variations for each concept, for example:
Children >Child, pediatric,paediatric, infant, toddler
Peanuts >Nuts
Allergy > Aversion, sensitivity, hypersensitivity, reaction
Having established a comprehensive terminology, nurses and other health care workers should then explore the essential 'core' of databases that provide access to citations for most original studies and reviews published in the healthcare literature.
Each database has its own style, parameters and entry mechanism for searching. The best method of finding articles is to use the index terms or subject headings that the indexers have assigned to each article as it is prepared for entry into the database, for example Medical Subject Headings or MeSH terms in MEDLINE. MeSH is a thesaurus of controlled terms to index articles; each article will have a number of terms assigned to it in order to represent the key concepts that it covers. Searching in this way will retrieve only material that is relevant to that concept.
Alternatively, searching using terms from the title and abstract (called textword or free text searching) is effective for retrieving information on new ideas. For example, textword searching would probably be the best approach when looking for information on the effects of household smoking-a relatively new topic of study.
However, the free text approach needs to be treated with caution because it is not as effective when looking for information on a topic that has many different names or terms. For example, decubitus ulcers are also called bedsore, bedsores, pressure ulcer, pressure ulcers, pressure sore, pressure sores, and can be referred to by terms such as 'tissue viability'. In this instance, searching using the database index terms (decubitus ulcer for MEDLINE and pressure ulcer in CINAHL) would be more accurate.
The 'advanced search' option can also be useful because this allows you to refine your search. Again, before starting your search think about what search terms to use. Break up your topic or question into individual concepts, for instance a search for citations on the effect of postnatal depression on the family has two concepts - postnatal depression and family. The secret is to search for each term separately and then combine them, based on using subject headings, index or thesaurus terms allocated to each article to describe its subject content. These will assist you to find articles about your subject, but which use different terminology from your free text search phrase. For example, a free text or 'Google-style' search for postnatal depression will omit many articles on that topic because they will have been given the descriptor postpartum depression.
It is critical in any field of research, especially if you are new to the profession, to understand the (unwritten and obviously biased towards librarians) 11th commandment: Thou shalt know thy search engine!
The Internet is awash with search engines like Google, GoogleScholar, Altavista, Excite, MSN and Yahoo. These are very popular because they have a browsing capacity, quickly retrieve materials and are easy to use. However, the problem facing health care researchers when using these types of search engines is to whittle down the results to a realistic, workable size.
The 'Advanced Search' feature in Internet search engines such as Google.com or GoogleScholar.com are often ignored, but are well worth looking at. They can really help you to develop very focused searches on any evidence-based health topic. This does not necessarily mean that you find the exact citations you need, but it can be a useful addition to your search strategy and can help refine the information you retrieve.
The issue of quality is important in evidence-based health care. Database developers such as MEDLINE, Cochrane Library and CINAHL have quality control mechanisms, but this is not generally the case with the Internet. There are, however, certain criteria that researchers can use when evaluating the quality of information on the Internet:
Health professionals need information to inform their practice, but it is not always easy to find. While the focus of this article has been on databases and the Internet it does not mean that 'hand searching' journals for information should be discounted. In fact it is a great addition to electronic searching. Researchers' different needs are best met by using specific sources or combinations of sources. Nonetheless, as academic libraries scale down their hard copy collections due to space and budgetary considerations, researchers in the health sciences must take a proactive and thoughtful approach to how they search for information.