Classifying the levels of recommendations in guidelines why GRADE? Magorzata Baa, MD, PhD Chair of Epidemiology and Preventive Medicine Department of Hygiene and Dietetics Guidelines what are they Guidelines what are they Guidelines what are they IOM, 2011 Clinical Practice Guidelines are statements that include recommendations intended to optimize patient care that are informed by a systematic review of evidence and an
assessment of the benefits and harms of alternative care options. Decision making and EBM Clinical scenario Clinical expertise Values and preferences evidence Standards 1. Transparency: details on guideline development and funding are explicit and publicly accessible 2. Management of conflicts of interest 3. Multidisciplinary guideline panel composition 4. Use of systematic reviews for guideline questions
5. Rating strength of recommendations 6. Clear articulation of recommendations 7. External review by a full spectrum of stakeholders 8. Updating Guideline quality 42 reviews of 626 guidelines 1980-2007 AGREE Instrument mean score 1988-2007 % in domains 70 60 50 40 30 20 10 0 S
pe co d an p p ur e os S r de
l ho ke a t m ve l vo in t en ur go i R
of m op l ve de t en it ar Cl Alonso-Coello P. Qual Saf Health Care 2010;19:e58 y
d an en es r p n tio a t p Ap y lit
bi a lic it Ed n li ir a o n pe e d nc de
e m co Re m en d de in ss la l
a er ov Guideline quality Domain scores over time 80 70 60 50 40 30 20
10 0 1988-1992 (9)SS 1993-1997 (102)SS 1998-2002 (291)SS Alonso-Coello P. Qual Saf Health Care 2010;19:e58 2003-2007 (206)NS Process Team COI Rules of work
Problem/topic evidence PICO (T) Groups of patients Management options Important outcomes Methods Evidence quality Benefits vs harms
Benefits vs costs Recommendation Process Problem/topic Systematic review Existing or new one Assessment of its validity evidence Evidence quality
Benefits vs harms Benefits vs costs Recommendation Cochrane review Antibiotics in acute otitis media Venekamp et al., Cochrane Database of Systematic Review Antibiotics in acute otitis media - Venekamp et al., Cochrane Database of Systematic Review GRADE
Systematic review Quality of evidence (levels of evidence) Problem/topic Evidence Evidence quality Benefits vs harms Benefits vs costs
Recommendation Hierarchy of Evidence Systematic review of RCTs Individual RCT Observational studies Basic research Clinical experience ADA grading for Standards of Medical Care in Diabetes Level Evidence
A Well-conducted large RCTs or properly conducted metaanalyses of RCTs B Well conducted cohort studies, registry, case-control study or properly conducted metaanalyses of cohort studies C RCTs with methodological flaws, biased observational studies, case series E Consensus of experts no studies
Diabetes Care 2018;41(Suppl. 1):S1S2 NCCN categories of evidence Categor Evidence (quality, quantity, consistency) and consensus y 1 High level evidence (i.e., high-powered RCTs or metaanalyses); uniform consensus (>=85%) 2A Lower level evidence; uniform consensus (>=85%) 2B Lower level evidence; consensus (>=50% but <85%) 3
Any level of evidence; major disagreement (>=25%) https://www.nccn.org/professionals/physician_gls/categories_of_consensus.aspx European Society of Cardiology Level What it means A Multiple RCTs/ metaanalyses B Single RCT/ large nonRCT C
Consensus/small studies, retrospective, registries Class What it means I Evidence/general agreement intervention beneficial, useful, effective II Conflicting evidence/ divergent opinions IIa Evidence/opinion in favour
IIb Evidence/opinion less well established III Evidence/general agreement intervention not useful/effective, may be harmful ESC. Recommendations for Guideline Production. www.escardio.org GRADE Systematic review Summary of findings table (SoF) Quality of evidence Summary of
results Problem/topic Evidence Evidence quality Benefits vs harms All important outcomes All risk groups Benefits vs costs
Recommendation GRADE quality of evidence US GRADE network GRADE Systematic review Summary of findings table (SoF) Quality of evidence Summary of results Quality of body of evidence for each important outcome
Each important outcome assigned relative weight Decision which outcome critical Problem/topic Evidence Evidence quality Benefits vs harms Benefits vs costs
Recommendation Quality of evidence Cochrane review Antibiotics in acute otitis media Venekamp et al., Cochrane Database of Systematic Review Process Problem/topic Evidence Lowest quality of evidence for any critical outcome
Evidence quality Benefits vs harms Benefits vs costs Recommendation GRADE Balance of benefits and harms Benefit certain Benefit uncertain Problem/topic
Evidence Evidence quality Benefits vs harms Benefits vs costs Recommendation Balance of desirable and undesirable effects For
strongconditional Against conditionalstrong GRADE Strength of recommendation strong (recommended) weak (suggested) Problem/topic Evidence Evidence quality
Benefits vs harms Benefits vs costs Recommendation problem recommendation Standards 1. Transparency: details on guideline development and funding are explicit and publicly accessible 2. Management of conflicts of interest 3. Multidisciplinary guideline panel composition
4. Use of systematic reviews for guideline questions 5. Rating strength of recommendations 6. Clear articulation of recommendations 7. External review by a full spectrum of stakeholders 8. Updating Thank you
