The editors of the US family medicine and primary care journals (i.e. Journal of Family Practice, Journal of the American Board of Family Practice, American Family Physician, Family Medicine, and the British Medical Journal-USA) have developed a grading scale for research papers to assess the strength of using their results for making recommendations to patients*. It is called “SORT”, which stands for “Strength of Recommendation Taxonomy.” This scale was developed for uniformity of research assessment, since over 100 different types of scoring methods are used by various journals, making it difficult to compare the relative value of papers. SORT uses the following system:
Letters rank the value of making a health recommendation based on a GROUP of research papers addressing the same issue. “A” is the highest, and “C” is the lowest score.
A: Health recommendation based on consistent and good-quality evidence of changes in endpoints that matter to patients (e.g. longevity, symptom improvement, quality of life)
B: Health recommendation based on inconsistent or limited-quality evidence of changes in endpoints that matter to patients
C: Health recommendation based on consensus, usual practice, opinion, case studies, or evidence of changes in physiological parameters that may or may not result in changes of endpoints that matter to patients (e.g. blood pressure, blood chemistry, physiological function, or pathological findings)
Numbers rank the value of INDIVIDUAL research papers, including reviews and meta-analysis.
1: Quality prospective cohort studies, meta-analysis, and systematic reviews
2. Case-control studies and meta-analysis or systematic reviews based on inconsistent findings
3. Same as C (above)
*Reference: “Strength of Recommendation Taxonomy (SORT): A patient-centered approach to grading evidence in the medical literature.” Ebell MH et al., J Am Board Fam Pract 17 2004 59