GRADEmetoden

GRADEmetoden is a systematic approach used to evaluate the quality of evidence and the strength of clinical or policy recommendations. It originated in the early 2000s as part of the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) working group, which was initially based at the Oxford Centre for Evidence-Based Medicine. The method is now widely adopted by international guideline development organisations, including the World Health Organization, the Cochrane Collaboration and the National Institute for Health and Care Excellence.

The core of GRADEmetoden involves judging evidence across several domains: risk of bias, consistency of results, directness of evidence, precision of estimates, and potential publication bias. Each domain can lead to upgrading or downgrading the confidence placed in the evidence. After assessing the study-level evidence, guideline panels evaluate the balance between desirable and undesirable consequences, patient values and preferences, resource implications, equity, acceptability, and feasibility. This synthesis determines whether recommendations are strong or weak (conditional), and whether the underlying evidence is high, moderate, low, or very low quality.

GRADEmetoden has several practical advantages. It provides a transparent, structured framework that facilitates comparison across studies and transparency in decision making. It encourages explicit discussion of how health outcomes and trade‑offs are weighed. However, critics point out that the method can be resource‑intensive, requires specialized training, and may still omit nuanced contextual factors in some settings.

Despite these limitations, GRADEmetoden remains a cornerstone of evidence‑based health policy, formulating consensus recommendations that aim to improve clinical practice and public health interventions worldwide.