This seventh article in a seven part series presents the Core GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach for moving from evidence to recommendations or policy decisions. Core GRADE users make strong recommendations for an intervention versus a comparator when the desirable consequences clearly outweigh the undesirable consequences, and a conditional (weak) recommendation when the balance is less clear. Primary considerations in deciding on recommendations considering an individual patient perspective include balance of benefits, harms, and burdens; the certainty of evidence; and values and preferences. Secondary considerations, most important from a population perspective, include costs, feasibility, acceptability, and equity. Moving from evidence to recommendations begins with considering evidence regarding patients’ values and preferences and choosing the smallest difference in each outcome that patients perceive as important (the minimal important difference). Core GRADE users construct statements that make clear the values and preferences underlying their recommendations. In general, Core GRADE users make strong recommendations only when certainty of evidence is high or moderate. When evidence certainty is low, recommendations will be conditional under all but special circumstances ### Summary points

中文翻译：

---

核心 7 级：从证据转向建议和决策的原则

这是由七部分组成的系列中的第七篇文章，介绍了从证据转向建议或政策决策的核心 GRADE（建议评估、开发和评估分级）方法。当理想的后果明显超过不良后果时，核心 GRADE 使用者强烈推荐干预与对照，当平衡不太明确时，提出有条件的（弱）推荐。在考虑个体患者的观点时决定推荐的主要考虑因素包括：利、弊和负担的平衡;证据质量;以及价值观和偏好。从人口的角度来看，次要考虑因素是最重要的，包括成本、可行性、可接受性和公平性。从证据到建议，首先要考虑有关患者价值观和偏好的证据，并选择患者认为重要的每个结果的最小差异（最小重要差异）。核心 GRADE 用户构建的语句清楚地说明了其建议背后的价值和偏好。一般来说，核心 GRADE 用户只有在证据质量为高或中等时才会提出强有力的建议。当证据质量低时，除特殊情况外，建议在所有情况下都是有条件的 ### 总结要点