Search filters include both terms to retrieve quality results for inquiry types, as well as search lines with study types to help sort results for critical appraisal. For some clinical questions it may be appropriate to combine portions of more than one filter.
** Indicates See Search Notes
1. Sensitivity and Specificity[mh] OR Diagnostic Errors[mh] OR Likelihood Functions[mh] OR Reproducibility of Results[mh] OR Area Under Curve[mh] OR Probability[mh] OR accurac*[tw] OR false positive*[tw] OR false negative*[tw] OR false rate*[tw] OR likelihood[tw] OR probabilit*[tw] OR receiver operat*[tw] OR roc[tw] OR sensitiv*[tw] OR specificity[tw] OR observer variation*[tw] OR predictive value*[tw] OR differential diagnosis[mh] **
2. etiolog*[ti] OR aetiolog*[ti] OR Incidence[mh] OR Prevalence[mh] OR incidence[tw] OR prevalence[tw] OR MeSH term /etiology OR etiology[sh] **
3. Cross-Sectional Studies[mh] OR Epidemiologic Studies[mh] OR Case-Control Studies[mh] OR Retrospective Studies[mh] OR Cohort Studies[mh] OR Longitudinal Studies[mh] OR Follow-up Studies[mh] OR Prospective Studies[mh] OR Comparative Study[pt]
4. Meta Analysis[pt] OR meta analy*[tw] OR metaanaly*[tw] OR systematic review*[tw] OR "published studies"[tw] OR medline[tw] OR embase[tw] OR data synthesis[tw] OR data extraction[tw] OR Cochrane Database Syst Rev[ta] **
If retrieval is low, consider including systematic review filter from PubMed Clinical Queries: systematic[sb]
If appropriate, consider including the MeSH term: meta-analysis as topic[mh]
5. Guideline[pt] OR Practice Guideline[pt] OR guideline*[tw] OR Guidelines[mh] OR Consensus Development Conference, NIH[pt] OR Consensus Development Conference[pt] OR consensus[tw] OR Consensus[mh] OR recommend*[ti]
If appropriate, consider including the MeSH term: guidelines as topic[mh]
** We suggest using MeSH terms and textwords in combination with this filter or MeSH terms/diagnosis if retrieval is very large. In most cases, "diagnostic use" as a subheading won't be needed but should be considered for diagnostic testing.
** Depending on the nature of the question consider terms such as screening[tw] OR mass screening[mh] , as well as searching the U.S. Preventive Services Task Force Recommendations: http://www.ahrq.gov/clinic/prevnew.htm
** Statement 1:Probability[mh] includes:
Bayes Theorem
Likelihood Functions
Markov Chains
Odds Ratio
Proportional Hazards Model
Risk
-Logistic Models
-Risk Assessment
-Risk Adjustment
-Risk Factors
Uncertainty
** Statement 2: The goal of statement 2 is to elicit common and serious etiologies which may impact upon diagnostic approach or "work-up" of a condition or disease. If retrieval seems too small, consider also searching in the abstracts for these textwords: etiolog*[tw] OR aetiolog*[tw].
** Statement 4: Systematic review terms applied to the diagnostic search strategy will identify studies that have used a systematic approach to searching, assessing and summarizing the results.
** The following MeSH terms and textwords can sometimes prove useful in focusing searches that deal with diagnostic questions involving some sort of decision-making process.
Algorithms[mh]
Decision Support Techniques[mh]
Decision making[mh]
Judgment[mh]
prediction rule*[tw]
decision*[tw] OR decide*[tw]
judg*[tw]
If appropriate, consider incorporating the concept of outcome*[tw] into your strategy in order to elicit the outcomes of making a particular diagnosis.
When working on a diagnosis question, consider discussing with your author whether or not "screening" concepts should be included in the search.
Questions such as, "What are the causes of..." implies "How do I approach the diagnosis of..." or "What is the best diagnostic approach to..."
In addition to the terms indicated in statement 2, and depending upon the nature of the question itself, or the condition/disease, you may want to consider incorporating some of the following MeSH terms related to etiology and epidemiology:
Age Distribution
Age Factors
Disease Susceptibility
-Genetic Predisposition to Disease
Family Characteristics
Health Status
-Geriatric Assessment
Sex Distribution
Sex Factors
The following outline provides the author with a general approach to developing an answer to an FPIN diagnostic question:
Evidence Based Answer
Evidence Summary
-Prevalence and incidence
-Common Etiologies
-Serious Etiologies
-Complete list of Etiologies
-Impact
Recommendations from Others
References
Clinical Commentary
PubMed search filters were developed by FPIN Librarians at the University of Washington. For more information about the FPIN search filters, please contact the FPIN Librarian Coordinator .