GRADE Notes 4: how to use GRADE when there is "no" evidence? A case study of using unpublished registry data

Area of Science:

• Medical Research Methodology
• Evidence-Based Medicine
• Data Science

Background:

• Trustworthy clinical guidelines depend on systematic reviews of published evidence.
• The GRADE Working Group offers guidance for developing recommendations when direct evidence is absent.
• Generating primary data from registries offers an alternative when published literature is insufficient.

Purpose of the Study:

• To present a case example of using registries for primary data generation.
• To address guideline development questions lacking published evidence.
• To explore an alternative to traditional systematic reviews.

Main Methods:

• Utilizing clinical and statistical expertise to leverage available registries for primary data generation.
• Focusing on guideline questions with no direct published literature.
• Applying registry data to answer clinically significant questions.

Main Results:

• Out of 54 guideline questions for pediatric peritoneal dialysis infections, 25 lacked evidence.
• Unpublished registry data provided primary information for 12 of these 25 questions.
• Registry data supplemented existing evidence for an additional nine questions.

Conclusions:

• Generating primary evidence from unpublished registry data is valuable when direct evidence is scarce.
• This approach is particularly beneficial for rare diseases or vulnerable populations.
• Transparency in data reporting and analysis plans is essential when using unpublished data.