Implementing GRADE

On September 15, 2009, Ethan Balk made this presentation at the 2009 Annual Conference. Select to access the PowerPoint® presentation (274 KB) (Plugin Software Help).

Slide 1

Implementing GRADE

Experience of the Tufts Center for Kidney Disease Guideline Development and Implementation
Tufts Medical Center, Boston MA

Guideline methodology since 2000 (18 GL)
Implementing GRADE since 2007 (6 GL)

Ethan Balk, MD MPH
Associate Director

Slide 2

Tufts Center for Kidney Disease
Guideline Development and Implementation

• Evidence Review Team for KDOQI & KDIGO
  • Kidney guidelines
• Full-time guideline & systematic review methodologists
  • ~6 FTE
    • MDs, "guideline fellows", research associates (MPH-level)
• Work Group of ~15-20 international domain experts
• 18 month process; 3 meetings

Slide 3

Advantages compared to previous systems

• Formal approach, standardized steps & tables
• Gets Work Group to consider may aspects of evidence as they pertain to guidelines
  • Overall evidence quality, consistency, applicability
  • Rank outcomes (clinically important outcomes)
  • Balance benefits and harms
• Simpler descriptions of recommendation levels (1, 2) and evidence quality (A, B, C, D)
  • Less arbitrary, more consistent connection between level of recommendations and quality of evidence
  • Eliminates reinventing the wheel

Slide 4

Disadvantages compared to previous

• Increased organizational workload
  • ? # summary tables for specific outcomes
  • Evidence Profile creation resource intensive
  • Training Work Group in GRADE methods, Evidence Profiles complicated
• Increases # of arbitrary formal decisions to maintain consistency
  • Study quality: (A/B/C)? Methodological quality (0, -1, -2)
  • • 2A + 4B + 1C = -1?
    • 2B + 2C = -1? -2?
  • Translating consistency? (0, -1, -2)
  • Translating directness? (0, -1, -2)

Slide 5

Responses to use of GRADE

• Work Group likes because:
  • It sets a standard structure and seems less arbitrary
  • Simplifies decision-making about strength and content of recommendation
  • Focuses on clinically important outcomes
• But, large # of steps =? resources, time
  • Difficult for domain experts
  • Some frustration about highly detailed work of filling every cell of Evidence Profile
  • Added work load may diminish participation