Linear mixed effect model estimates of fixed effects (ADAS-cog, MMSE, ADL, NPI).

* indicates the main effects after dropping the insignificant group x time interaction.
Time × group interaction effect was observed in the scores of MMSE (F107.776 = 3.251, p = 0.025). The scores in both groups increased in week 12 and 24, by 1.72±2.59 in YHD group and 1.27±2.23 in DH group, but there is no significant difference between groups during the 24-week treatment. Compared with week-24, the scores of ADAS-cog in the follow up in week 48 showed a decrease of 0.40±1.63 in YHD group, but significantly less than the decrease of 1.38±2.68 in DH group (p<0.05).
Linear mixed effect model estimates of fixed effects (ADAS-cog, MMSE, ADL, NPI).

* 表示剔除无统计学意义的组×时间交互效应后的主效应。在简易精神状态检查表（Mini-Mental State Examination, MMSE）得分中观察到组×时间交互效应（F(107.776)=3.251，p=0.025）。两组评分在第12周及第24周时均有所升高：YHD组升高值为1.72±2.59，DH组为1.27±2.23；但在整个24周治疗周期内，两组间评分无显著差异。与第24周相比，第48周随访时的阿尔茨海默病评估量表-认知子量表（Alzheimer's Disease Assessment Scale-Cognitive Subscale, ADAS-cog）得分中，YHD组下降0.40±1.63，但其下降幅度显著低于DH组的1.38±2.68（p<0.05）。固定效应的线性混合效应模型估计结果，涵盖指标包括ADAS-cog、MMSE、日常生活能力量表（Activities of Daily Living, ADL）及神经精神问卷（Neuropsychiatric Inventory, NPI）。