Skip to main content

Languages

Poor interpretation of quality assessment results in diagnostic accuracy reviews

Date and Location

Session: 

P2.036

Date

Saturday 21 September 2013 - 10:30 - 12:00

Location

Presenting author and contact person

Presenting author

Eleanor Ochodo

Contact person

Eleanor Ochodo
Abstract text
Background: Interpreting and presenting results in systematic reviews without taking into account the outcome of quality assessment has been shown to be common in systematic reviews of interventions, but may also play a role in reviews of test accuracy studies. Drawing conclusions or making recommendations without considering the risk of bias and limited applicability of included studies may lead to unwarranted optimism about the value of the corresponding test. Objectives: To identify and compare approaches used to make conclusions out of the results of quality assessment in diagnostic accuracy reviews, and to provide guidance on recommended methods. Methods: We searched MEDLINE and EMBASE for test accuracy reviews published between May and September 2012. We examined the abstracts and main texts of these reviews to check whether and how the results of quality assessment were taken into account when drawing conclusions. Data was extracted by one author; a sample was checked by a second author. Results: Our search identified 53 eligible reviews. Of these, 49 (92%) had formally assessed the methodological quality of included studies; Twenty-two (45%) distinguished high quality from low quality studies using summary scores (n=13), summary graphs (n=5) and other methods. Overall, only 5 articles (9%) incorporated quality assessment results in the conclusions of the abstracts; in the main texts, only 14 (26%) reviews considered results of quality assessment in the recommendations or conclusions. Examples of approaches used to mention and interpret quality assessment in the sample of reviews can be found in the table. Conclusions: Almost all recent reviews of test accuracy studies evaluate the quality of eligible studies, but only 3 out of 10 reviews consider quality assessment when drawing conclusions or making recommendations. We recommend the results of quality assessment be always factored into the conclusions of reviews, to limit misleading presentation of the performance of the diagnostic tests under review.
Attachments