Finding Empirical Evidence: D Search Strategy Tips
Download as PPT, PDF1 like331 views
This document outlines a search strategy to investigate the risk factors, impacts, causes and context of unhealthy eating in Australia. It provides keywords and concepts related to nutrition, obesity, and epidemiology. Boolean logic operators and search techniques like truncation and phrase searching are described to construct an effective search string combining these concepts and filtering results to focus on the issue in an Australian context.
Finding Empirical Evidence: D Search Strategy Tips
- 1. Search Strategy Risk Factor: impact, causes, context in Australia Identifying key concepts, terms: Unhealthy Eating OR Impact OR Australia OR “poor diet” Nutrition Obesity causes consequences Reasons Epidemiology Aetiology Western Australia WA • Use AND to combine concepts and identify results that include both
- 2. Boolean Logic AND retrieves all the articles with both ‘Diabetes’ and ‘Indigenous’ OR retrieves all the articles containing either ‘Aboriginal’ or ‘Indigenous’ or both NOT retrieves all the articles containing ‘Indigenous’ but not those about ‘Canada’
- 3. Search Tips Phrases and Truncation: smok* = smoke, smoker, smoking Austral* = Australia, Australian, Australians, Australasian 4. Use Quotation Marks for Phrases “cigarette smoking”
- 4. Search Strategy Risk Factor: impact, causes, context in Australia Adding Boolean logic, truncation and phrases: “Unhealthy Eating” OR Impact OR Australia OR “poor diet” Nutrition Obesity Obes*e/ity causes consequences Reason*s epidemiology Etiology Western Australia WA Austral* ia, ian, s • Use quotations “…….” to search for a phrase • Use asterix * to find all term derivatives • Use OR to search for/identify all similar concepts • Use AND to combine concepts and identify results that include both AND AND
- 5. Search Strategy Constructing a search string (“unhealthy eating” OR “poor diet” OR obes*) AND (“austral*) AND (impact OR cause* OR epidemiology OR etiology OR aetiology) In most databases enclosing terms in brackets will help order how concepts will be searched.
- 6. Study Type Search Tips Randomised Control Trial Questions: Patient/population/problem = who or what is of interest Intervention = what intervention is being investigated Comparison group = what alternatives can be compared to the intervention group Outcome of interest = what should the intervention achieve, what improvements? Useful References: CEBM Study Design and PHCRIS Formulating a Question http://guides.library.uwa.edu.au/systematic_reviews
- 7. Search Strategy Population/ Problem Intervention Comparison Outcome Body mass index In Children 0-4 Physical Activity in Nurseries No exercise Reduction in BMI Weight Exercise Not in nurseries Physical activity to prevent obesity in young children, cluster randomised controlled trial. DOI
- 8. Search Strategy Your turn – activity 1b •See if you can build on your search strategy concepts. • Think about how you might join your concepts with AND, OR operators to refine your search to build a search string. •Are there any particular interventions or comparators you want to focus your search to at this stage. Use worksheet provided.
- 9. Contact the University Library http://library.uwa.edu.au 9346 7570 – Medical and Dental Library hmslibrarians-lib@uwa.edu.au http://facebook.com/UWALibrary @UWALibrary
- 10. Contact the University Library http://library.uwa.edu.au 9346 7570 – Medical and Dental Library hmslibrarians-lib@uwa.edu.au http://facebook.com/UWALibrary @UWALibrary
Editor's Notes
- #2: SO how is your search strategy going, have by you been able to identify other similar terms that will help you identify all records about your topic. Coming up with other keywords/terms used to describe your topic and thinking about how you combine them. Other terms you could add: Nutrition Obesity For impact: Do you know of any medical terms that could help: Epidemiology – study of the distribution and determinants of health related states or events Aetiology Use OR to search for/identify bring all similar concepts together. Use AND to combine your core concepts together and make sure you only identify results that relate to your topic focus. Other terms you could add: Nutrition Obesity For impact: Do you know of any medical terms that could help: Epidemilogy – study of the distribution and determinants of health related states or events Aetiology It is worth putting some thought into what your concepts are and how you might
- #3: Recap on Boolean: Here is a Venn diagram to clarify how the use of booleans, like AND, OR and NOT work in helping you to refine your search strategy. Note: the NOT excludes items containing both ‘indigenous’ and Canada – potentially relevant results? Use NOT with caution, generally for refining a search only or for multiple meanings of the same word, eg China (country) vs china (porcelain).
- #4: Otherways you can refine your search include the use of the truncation symbol (usually an asterisk) to search for all derivatives of a word. Using quotation marks to search for a phrase. Two of the most common, but a variety of other search limiters you can use and found in many academic databases.
- #5: Here is an expanded example of boolean and search limiters might be used to better refine and control your search strategy.
- #6: Your next step would be to think about how you might express this search in a search engine like Google or in a sophisticated database like Medline. This is often expressed as a search string.. Generally at an earlier scoping phase of research you will use broad terms, however as you refine your research focus you would add terms and look for ways you can use the AND boolean in particular to increase specificity. Academic database often have many built in ways you can limit and filter search results. Highly constructed search strings won’t work in every situation, in particular won’t work so well when you are searching the grey literature – non published works ususally found on the web in government or specialist organisation websites. Not necessarily use all of the terms developed, you are starting a process of seeing which work best in the different resources we will search.
- #7: As you develop your RCT focus for your assessment, you will want to further refine your question and the concepts that relate to it to fit into the PICO framework, clearly identifying the population or problem that is a of interest to you, the specific type of intervention you want to evaluate and analyse and the comparison group. This level of refinement to your search strategy is also of relevance when you are searching for and wanting to identify if there have been systematic reviews carried out in the RCT research focus you will be proposing. At the moment our sample search is focussed on a more qualitative search of Population or problem and outcome.
- #8: Here is an expanded example of boolean and search limiters might be used to better refine and control a PICO based search strategy. As you search the literature you may find more related terms that will help you to refine your search so you hone in on specific studies of interest to your intervention.