If the missing data is non-random, that is, associated with other variables, it may lead to some specific types of individuals in the sample. If the missing data is non-random, that is, it is associated with other variables, it may lead to the exclusion of certain types of individuals or situations in the sample from the analysis, thus leading to sample deviation. In addition, if the missing data is related to clinical manifestations or research results, there will be bias. Therefore, data missing is common, which brings great challenges to the analysis and interpretation of clinical research results.