The problem with searching for is that researchers, when writing thei abstracts and titles, don't indicate study types when they aren't doing a clinical trial. They now identify clinical trials fairly well and don't do too badly with systematic reviews (although I do find non-standard names for systematic reviews)..
But, epi studies such as case-controls, cohorts and especially cross-sectionals are poorly described in abstracts. Same with observational studies. So, it wouldn't be possible to develop a very effective filter for observational studies. I've tried for other study types:
https://sph.uth.tmc.edu/current-students/library/search-filters/
My dream is that all study authors follow the reporting guidelines that are appropriate for their study type. There's a wonderful site with guidelines compiled: http://www.equator-network.org. If authors followed their respective guidelines for reporting (such as MOOSE, PRQISMA, etc.), then it would be so much easier to create filters.
Helena M. VonVille, MLS, MPH
Library Director
University of Texas School of Public Health
Houston, TX
As for your search, I would be inclined to search by data sets of interest. Or maybe terms such as secondary analysis. There is a MeSHA term for data sets-- . It's not perfect but it does pick up some of thos large data sets. Also: Insurance claim review
That's another Mesh term. If you were going to search PubMed:
databases, Factual[mesh] OR Insurance claim review[mesh] OR medicare[mesh] OR "Medical Expenditure Panel Survey"[tiab] OR ((Medicare[tiab] OR Medicaid[tiab]) AND claims[tiab])) or hcup[tiab] OR "h-cup"[tiab] OR meps[tiab] OR administrative data[tiab] OR insurance claims[tiab] OR (Secondary analysis[tiab] AND dataset*[tiab])
This wil only work in PubMed. It's not perfect-- I'd need to play with it more. But it retrieves over 99k hits. Combine with other broad concepts to narrow. If you have other data sets you're interested in, add those to the search following the strategy I employed.
Helena
[log in to unmask]<mailto:[log in to unmask]>
713-500-9131
713-500-9125 (fax)
________________________________________
From: Evidence based health (EBH) [[log in to unmask]] On Behalf Of Arjun Mehta [[log in to unmask]]
Sent: Friday, January 13, 2012 11:33 PM
To: [log in to unmask]
Subject: Search Strategy for Database studies-Observation studies
Hello All,
I am a graduate student studying health outcomes research. The discussion of search strategy for systematic review gave very good understanding of use of different databases.
I am working on different databases like Medical Expediture Panel Survey, Medicare, Medicaid and claims database. I wonder whether any search filters are developed for searching the studies for observation studies. The search strategy can identified studies observation or comparative effectiveness research performed using any of the datasets only and would be able to remove the primary data collections studies.
I would really appreciate the help and support of all group members.
Thank you very much.
Sincerely,
AM
|