Skip to main content

Table 3 Summary of main features in GRADE, NICE DAP and the NHMRC systems for moving from evidence to making recommendations

From: Guidelines for guideline developers: a systematic review of grading systems for medical tests

 

GRADE

NHMRC system

NICE DAP system

Evidence on accuracy

Using a considered judgment process, derive presumed PIO for the four accuracy groups: TP, TN. FP, FN

Evidence base aspects considered:

Diagnostic accuracy aspects considered:

 

- Validity

- Number of studies

- Inclusiveness of underlying data

- Level of evidence

- Meta analysis techniques

- Risk of bias

- Cut off points

 

- Uncertainty of data

Factors considered when moving to recommendations

Then consider:

 

-

- Quality of evidence gathered

Clinical impact: not explicitly explained

Clinical effectiveness:

- Patient’s values and preferences

Generalisibility: how well does the body of evidence match the the body of evidence match thepopulation and clinical setting being targeted by the guideline

- Nature and quality of evidence derived from expert,

- Costs

Applicability: is the evidence body relevant to the Australian healthcare context and culture

- Lay members and stakeholder judgments

- Benefits vs harms

Evidence base:

- Uncertainty of evidence and differences in evidence gathered under research conditions vs in actual

 

- Number of studies

- Clinical practice

 

- Level of evidence

- Greater benefits or harms in subgroups

 

- Risk of bias

- Risks and /or benefits of technology from patients perspective

   

- Position of technology in overall care pathway and available alternative treatments

  

Consistency: not explicitly explained

Cost Effectiveness:

   

- Impact on patient outcomes

   

- Robustness and appropriateness of model

   

- Plausibility of inputs and assumptions made in economic model

   

- Evaluation of the modeling approach and related evidence effectiveness ratios (ICERs) generated by the models

   

- Range and plausibility of the incremental cost-

   

- Likelihood of decision error and consequences

   

- Degree of clinical need of patients under consideration

   

- Potential for long term benefits of innovation