| 证据分级方法： |
1.GRADE evidence quality rating:
high-very confident that the observations are close to the true values, and further research is unlikely to change the credibility of the observations.
Medium-medium confidence in the observed value: the true value may be close to the observed value, but there is still the possibility that there is a difference between the two, and further research may change the credibility of the observation and may change the results of the observation.
Low-the degree of confidence in the observation is limited: the real value may be very different from the observation, and further research is very likely to change the credibility of the observation, and is likely to change the results of the observation.
Very low-there is little confidence in the observed value: the real value is likely to be different from the observed value, the true value may be very different from the observed value, and the results of the observed value are very uncertain.
2.Recommendation:
through preliminary research, this guide is screened according to the actual clinical situation according to evidence retrieval, core group and expert discussion to determine the final clinical problems. The corresponding recommendations are formed based on evidence, combination cost, patient preferences and values, and all recommendations are graded according to evidence quality and recommendation intensity. Delphi method is used for two rounds of expert consensus and revised according to expert suggestions. The recommendation is finally formed after many steps, such as pre-clinical investigation, expert discussion, evidence retrieval and screening, expert consensus and so on. |
