A new version of RevMan has been released and includes some major changes to functionality for Cochrane DTA reviews. The changes in DTA reviews are “structural changes” which means that once a DTA review has been imported into the new RevMan 5.2 it will no longer be compatible with RevMan 5. This will have implications for review authors and CRGs who are currently preparing new DTA reviews, or looking to update DTA reviews. As RevMan 5.2 is a mandatory update, all CRGs and review authors should, by now, have made this change. We have produced a document outlining what changes have been made (see "Software Development") so that you can envisage how it will affect your review. The most major change is to the methodological quality assessment but there are also changes to subheadings of the background section of DTA protocols and reviews concerned with the mappingof the clinical pathway.
Methodological Quality Assessment
RevMan 5 used the original QUADAS framework for quality assessment, which is explained in detail in Chapter 9 of the Cochrane DTA Handbook (available at dta.cochrane.org). Many of you took part in a survey reporting your experiences and problems in using this tool. Based on these views, and experiences of others outside the Collaboration, an updated version of QUADAS, known as QUADAS-2, has been developed. This new version aims to resolve issues with particular items, and provides a more useful summary of study quality which separately assesses “risk of bias” and “applicability”. It is thus quite different from the original QUADAS. QUADAS-2 has recently been published in Annals of Internal Medicine (http://www.ncbi.nlm.nih.gov/pubmed/22007046).
QUADAS-2 has a different structure to the original QUADAS, and it will not be possible to fully automate a change between these tools. Although much of the detailed information for a QUADAS-2 assessment will have been collected in an original QUADAS assessment, updating from one assessment process to the other will be a manual process that review authors need to undertake, although we doubt that you will need to re-read the original papers. Thus there are advantages to changing to QUADAS-2 if you are in the early stages of your review.
We are in the process of updating the DTA Handbook to support review authors and hope this will be available in March 2012. Until this is available they should consult the published QUADAS-2 paper together with the data structure document to ensure that correct data will be collected. The QUADAS-2 background document is available on the QUADAS-2 website, http://www.quadas.org.
Overview of changes to subheadings:
Authors should note that a DTA protocol opened in 5.2 will automatically be populated with the revised headings (as outlined below) but for full reviews heading changes are only activated when a user chooses to convert to QUADAS-2.
An additional level 2 mandatory subheading has been introduced into the background sections of DTA protocols and reviews - ‘Clinical Pathway’. The clinical pathway subheading is supported by 3 optional level 3 subheadings (prior tests, role of index test(s) and alternative test(s)) that are designed to prompt authors to consider the important elements of their overview of the clinical pathway. In the background and methods sections of DTA protocols and reviews, the term ‘comparator tests’ has been removed and review authors are requested to consider all tests whose accuracy is being evaluated in the DTA review under the ‘index test(s)’ level 2 mandatory subheadings.
Background(level 1 mandatory heading)
Target condition being diagnosed (level 2 optional subheading)
The target condition is the condition of interest that the index test(s) are attempting to detect. This may be a disease or disease stage.
Index test(s)(level 2 optional subheading)
Any test whose accuracy is being verified using a reference standard test as part of the DTA review. Index tests may or may not include standard diagnostic practice. Where a review includes more than one index test a comparative evaluation of test accuracy may be included as an objective.
Clinical pathway (level 2 mandatory subheading)
Outline the existing clinical pathway of patients. Detail how patients might present, the point in the existing pathway that participants would be considered for testing with the index test or tests and the role of the index test(s).Authors may choose to use the level 3 optional subheadings to help with their description of the clinical pathway.
Prior test(s)(level 3 optional subheading)
What tests might participants in the review receive prior to receiving the index test(s).This should include clinical history and examination if appropriate. Alternatively the healthcare setting (community, primary or secondary care) in which participants will present may be used as an indication of the type and number of tests they might have received prior to receiving the index test.
Role of index test(s) (level 3 optional subheading)
The role of index test(s) may be that they are to be used in addition to existing tests (add on), to replace existing tests (replacement) or be used to decide which patients should receive further testing (triage).
Alternative test(s) (level 3 optional subheading)
A description of the possible diagnostic tests and strategies that could be used in clinical practice, that aren’t evaluated in this review. This will help place the review in context of all available diagnostic options.
Rationale (level 2 optional subheading)
The background helps set the rationale for the review, and should justify why the questions being asked are important. It might also mention why this review was undertaken and how it might relate to a wider review of a general problem.
RevMan 5.2 is now available.
1) Reviews published: DTA Reviews published before the release of RevMan 5.2 will continue to appear in the Cochrane Library using the RevMan 5 presentation format until they are updated. Review authors with published reviews need not work to convert from QUADAS to QUADAS-2 until they are updated. However, once a CRG has switched to RevMan 5.2, any edit to a review (even only correcting a typo) will mean that the next published version will require completion of QUADAS-2 for every study.
2) Reviews under development: Reviews close to completion are in the most difficult situation. With immediate effect all protocols and reviews must be converted to RevMan 5.2. DTA review authors who are close to completion should discuss processes with their CRG editorial team.
3) Review authors who have not yet started data extraction or who are only a small way through their review should switch to undertaking data extraction and quality assessment using QUADAS-2. It is possible to convert from RevMan 5 review to RevMan 5.2. Quality assessments made using the QUADAS-2 framework should not be entered into RevMan 5.
4) All new protocols should state that quality assessment will be done using QUADAS-2. Protocols currently in the editorial process will be asked to change.