New pyramid proposal for the levels of scientific evidence according to SIGN

High-quality scientific evidence requires approaches to evaluate the research methodologies. Thus, degrees of recommendation for medical procedures or health interventions have been established.1,2 Internal validity assessment (for methodology and bias), external validity assessment (for reproducibility in the target population), and analysis of impact ought to be performed. Clinical practice guidelines are documents intended to provide recommendations for making decisions about medical procedures.3 Nevertheless, some clinical practice guidelines are not properly structured and lack updated scientific evidence. Therefore, clinical practice guideline quality control ensures that potential biases in their elaboration have been marked and recommendations have been internally and externally validated. The Appraisal for Guidelines, Research and Evaluation in Europe evaluates the methodological rigor (a key factor for clinical practice guideline quality) and the transparency with which the clinical practice guidelines are elaborated. The two most used measurement scales, the Grading of Recommendations, Assessment, Development, and Evaluation scale (GRADE) and the Scottish Intercollegiate Guidelines Network (SIGN) scale, were designed to assess and categorize systematic reviews and clinical practice guidelines regarding therapeutic procedures. The SIGN scale also focuses on the reduction of systematic errors or bias risk.4 The pyramid for levels of scientific evidence classifies research according to its relevance, but not all the evidence placed at the same level have the same quality. Murad et al.5 proposed an alternative depiction of this pyramid with wavy lines dividing the quality levels (following the GRADE approach) (Fig. 1, above). This study aimed to present a new didactic proposal of the level of scientific evidence pyramid according to the SIGN scale, by depicting each level of scientific evidence that leads to recommendation degree (Fig. 1, below). This study was approved by the Institutional Committee of Ethics in Research, IMT “DAC” National University of San Marcos (CIEI-2020-003).Fig. 1.: New pyramid proposal according to SIGN. (Above) Pyramid proposed according to GRADE. (Used with permission from [email protected].) (Below) New proposal for a pyramid of evidence levels according to SIGN.The SIGN approach is different from others that only evaluate the methodological design. This causes overestimation in the categorization of the level of scientific evidence and recommendation degrees. Another limitation is the need to complement the bias risk assessment with additional instruments. For treatment studies, the SIGN proposal categorizes the level of scientific evidence with numbers (from 1 to 4) to generate the recommendation degree, identified with letters (from A to D) (Fig. 1, below). The levels of scientific evidence categorized as “1++” and “1+” generate a “type A” recommendation degree. The levels of scientific evidence categorized as “1−” and “2−” are considered “not recommended” due to high bias risk. The level of scientific evidence categorized as “2++” generates a “type B” recommendation degree. The level of scientific evidence categorized as “2+” generates a “type C” recommendation degree. The levels of scientific evidence categorized as “3” and “4” generate a “type D” recommendation degree. This new pyramid proposal responds better to the levels of scientific evidence for clinical application. DISCLOSURE The authors have no financial interest to declare in relation to the content of this article.