Explanation and source of recommendations

Table 1. Criteria for assigning levels of evidence to published studies 

Level 

Criteria  

Studies of diagnosis 

Level 1 

1. Independent interpretation of test results (without knowledge of the result of the diagnostic or gold standard) 

2. Independent interpretation of the diagnostic standard (without knowledge of the test result) 

3. Selection of people suspected (but not known) to have the disorder 

4. Reproducible description of both the test and diagnostic standard 

5. At least 50 patients with and 50 patients without the disorder 

Level 2 

Meets four of the Level 1 criteria  

Level 3 

Meets three of the Level 1 criteria 

Level 4 

Meets one of the Level 1 criteria 

Studies of treatment and prevention 

Level 1A 

Systematic overview or meta-analysis of high-quality RCTs 

1. Comprehensive search for evidence 
2. Authors avoided bias in selecting articles for inclusion 
3. Authors assessed each article for validity 
4. Reports clear conclusions that are supported by the data and appropriate analyses 

OR 

Appropriately designed RCT with adequate power to answer the question posed by the investigators 

1. Patients were randomly allocated to treatment groups 
2. Follow-up at least 80% complete 
3. Patients and investigators were blinded to the treatment* 
4. Patients were analysed in the treatment groups to which they were assigned 
5. The sample size was large enough to detect the outcome of interest 

Level 1B 

Non-randomised clinical trial or cohort study with indisputable results 

Level 2 

RCT or systematic overview that does not meet Level 1 criteria 

Level 3 

Non-randomised clinical trial or cohort study; systematic overview or meta-analysis of Level 3 studies 

Level 4 

Other 

Studies of prognosis 

Level 1 

1. Inception cohort of patients with the condition of interest, but free of the outcome of interest 
2. Reproducible inclusion/exclusion criteria 
3. Follow up of at least 80% of subjects 
4. Statistical adjustment for extraneous prognostic factors (confounders) 
5. Reproducible description of outcome measures 

Level 2 

Meets criterion (a) above, plus three of the other four criteria 

Level 3 

Meets criterion (a) above, plus two of the other criteria 

Level 4 

Meets criterion (a) above, plus one of the other criteria 

*In cases where such blinding was not possible or was impractical (eg intensive versus conventional insulin therapy), the blinding of individuals who assessed and adjudicated study outcomes was felt to be sufficient. 

RCT, randomised controlled trial. 

Source: Adapted from Diabetes Canada Clinical Practice Guidelines Expert Committee. Diabetes Canada 2018 clinical practice guidelines for the prevention and management of diabetes in Canada. Can J Diabetes 2018;42(Suppl 1):S1–S326. 

Table 2. Criteria for assigning grades of recommendations for clinical practice 

Grade 

Criteria 

Grade A 

The best evidence was at Level 1 

Grade B 

The best evidence was at Level 2 

Grade C 

The best evidence was at Level 3 

Grade D 

The best evidence was at Level 4 or consensus 

Source: Adapted from Diabetes Canada Clinical Practice Guidelines Expert Committee. Diabetes Canada 2018 clinical practice guidelines for the prevention and management of diabetes in Canada. Can J Diabetes 2018;42(Suppl 1):S1–S326. 

Table 3. ADA evidence-grading system for Standards of care in diabetes 

Level of evidence 

Description 

A 

Clear evidence from well-conducted, generalisable RCTs that are adequately powered, including: 

• evidence from a well-conducted multicentre trial 
• evidence from a meta-analysis that incorporated quality ratings in the analysis 

Supportive evidence from well-conducted RCTs that are adequately powered, including: 

• evidence from a well-conducted trial at one or more institutions 
• evidence from a meta-analysis that incorporated quality ratings in the analysis 

B 

Supportive evidence from well-conducted cohort studies, including: 

• evidence from a well-conducted prospective cohort study or registry 
• evidence from a well-conducted meta-analysis of cohort studies 

Supportive evidence from a well-conducted case-control study 

C 

Supportive evidence from poorly controlled or uncontrolled studies, including: 

• evidence from randomised clinical trials with one or more major or three or more minor methodological flaws that could invalidate the results 
• evidence from observational studies with high potential for bias (such as case series with comparison with historical controls) 
• evidence from case series or case reports 

E 

Expert consensus or clinical experience 

Source: Adapted from American Diabetes Association Professional Practice Committee. Introduction and methodology: Standards of care in diabetes – 2024. Diabetes Care 2024;47(Suppl_1):S1–S4. 

RCTs, randomised controlled trials.