Frequently Asked Questions about the CDC Tier-Classified Guidelines Database
Posted on by
Over the last several years, OPHG has hosted a “Tier Table” database of genomic applications (i.e., clinical scenarios involving genomic testing) sorted into one of three tiers using a method described in Clinical Pharmacology and Therapeutics, 2014. The Tier Table is being replaced with our new Tier-Classified Guidelines Database, which we hope will promote more efficient and rapid classification. The key procedural difference in populating the new database is that, as opposed to presenting tier level codes for genetic testing clinical scenarios (as in the Tier Table), it is the guideline documents, and related evidence sources, that are assigned a tier level (representing the highest tier level that could be assigned to a contained genetic testing clinical scenario). For more information, including important caveats, please read our Frequently Asked Questions below, and see our blog: Introducing the CDC Tier-Classified Guidelines Database.
What is, and is not, in the Tier-Classified Guidelines Database?
Guidelines identified in our weekly horizon scanning are included. While the Genomics and Precision Health Database of horizon scanning results classifies a guideline ‘product type’ to include: clinical, laboratory, and public health practice guidelines, summaries of guidelines and articles about guidelines, policy statements, etc., only clinical and public health practice guidelines are included in the Tier-Classified Guidelines Database. Although clinical scenarios themselves are not assigned a Tier level, at least one clinical scenario involving genetic testing must be the subject of a recommendation (either for or against use, or neither for nor against use with insufficient evidence findings) in order for a guideline to be assigned a Tier level and included in the database. In the future, it is expected that additional types of guidance, such as links to FDA drug labels and CMS coverage decisions, will be included in the database.
How are Tier Levels Assigned?
Sorting is done using criteria (shown above) after reviewing titles and abstracts, and consulting the full text of articles when needed. Reviews are not rigorous, being done relatively quickly, and necessarily involve some degree of subjectivity. While some errors are likely in the process, at least two reviewers must agree on each Tier level classification. Reviewer disagreements are resolved through discussion, and additional reviewers may be called upon as needed for majority voting resolution.
What is the Practical Significance of a Tier Designation?
The population of Tier 1 guidelines should be relatively enriched in genomic applications whose implementation has some support from evidence-informed processes (for example, practice guidelines based on systematic reviews, CMS coverage determinations, etc.). Tier 2 guidelines should likewise be enriched with genomic applications with there is support for implementation, but with more uncertainty, and Tier 3 with genomic applications where there are recommendations against implementation. It is important to remember that the highest Tier level is assigned to each guideline, so a Tier 1 guideline could include separate recommendations that meet criteria for Tiers 1, 2, and 3. Likewise, Tier 2 guidelines could include recommendations meeting criteria for Tiers 2 and 3.
What do the search results mean?
Guidelines in the database that contain the search terms will be displayed in the search results. It is important to remember that the tier level is assigned to each guideline, NOT to the specific search terms. For example, a Tier 1 guideline has at least one recommendation that meets criteria for Tier 1, but also could include tier 2 or 3 recommendations. DO read the guidelines to find out about specific search terms.
DON’T rely on the overall tier level classification of the guideline for the specified search terms.
How might the Tier-Classified Guidelines Database be Useful?
Given that Tier 1 guidelines should be enriched with evidence-informed recommendations supporting implementation of genomic applications, these documents might be worth reading in more depth to identify the clinical scenarios that are (and are not) recommended for implementation. Similar uses might be practicable for Tier 2 and Tier 3 guidelines, for those interested in identifying genomic applications where continuing evidence development has (and has not) been recommended.
What are the Main Caveats in Using the Tier-Classified Guidelines Database?
As mentioned above, the review process is not rigorous, and some level of subjectivity is involved, so some level of error must be inherent to the Tier assignment process. Since the highest Tier level is assigned to each guideline, and many guidelines contain multiple recommendations, Tier 1 and Tier 2 guidelines may contain recommendations that meet criteria for more than one Tier level. While documents are coded by gene, disease, drug, etc., they are only intended to assist in searching the corpus of documents, and neither these features, nor specific tests or clinical scenarios are assigned a Tier level – it is only the guideline document that is assigned a Tier in this process. Findings presented here may be useful as a starting point for people interested in identifying relevant evidence sources, however, the corresponding full evidence sources must be consulted in order to accurately and effectively interpret and employ their findings and recommendations. The process was not intended to capture all of the information necessary to inform clinical or public health practice or policymaking, such as coverage decisions. Tier level classifications should not be construed as an endorsement or official position of CDC.