Mann 2012

Appraisal of: Mann R, Gilbody S M. Should methodological filters for diagnostic test accuracy studies be used in systematic reviews of psychometric instruments? A case study involving screening for postnatal depression. Syst Rev 2012: 1 (1):9. DOI: 10.1186/2046-4053-1-9.

Reviewer(s):

Julie Glanville

Caroline Higgins

Full Reference:

Mann R, Gilbody S M. Should methodological filters for diagnostic test accuracy studies be used in systematic reviews of psychometric instruments? A case study involving screening for postnatal depression. Syst Rev 2012: 1 (1):9. DOI: 10.1186/2046-4053-1-9.

Short description:

This is a small exploratory case study. It compared two MEDLINE search filters for psychometric diagnostic test accuracy studies in postnatal depression with a search using the recommended Cochrane approach of searching the target condition and index test without a search filter. Cochrane recommendations were based on assessment of biochemical laboratory tests rather than psychometric tests. A reference set of 6 records was identified using forward citation searching from a single key seed paper. The Cochrane recommended search and using the CRD filter were both 100% sensitive but low precision (5.2%), whereas the Vincent filter had higher precision but only 66.6% sensitivity. Since the Cochrane search was less complex than using a filter, the authors recommend the Cochrane approach of searching for the target condition and index test without a search filter, to identify DTA studies of psychometric tests. The authors note that filters may be suboptimal because psychometric tests are often associated with terms related to reliability and validity rather than diagnostic accuracy terms.

Limitations stated by the author(s):

The reference set may not be a gold standard because it was arrived at using a citation search, but the authors note that the topic has a small literature so the probability of missing studies may be small.

Limitations stated by the reviewer(s):

This is a small reference set and as the authors acknowledge was not identified using standard methods such as handsearch or relative recall. However, the authors' conclusions are conservative as a result.

Study Type:

Primary research: Comparison of filter use versus non-filter use

Related Chapters:

Diagnostic accuracy

Tags:

• diagnostic test accuracy studies

• search filters