Methodological flaws in“diagnostic accuracy of blood biomarkers for Alzheimer's disease and amnestic mild cognitive impairment: A meta-analysis”

We read with interest the review by Chen et al. They intended to examine the diagnostic accuracy of blood-based biomarkers for detecting Alzheimer's disease and amnestic mild cognitive impairment. We believe that there were substantial methodological flaws in their meta-analysis. These methodological flaws included no comprehensive literature search details, neglect of the negative result research, no prespecified cut-off values, erroneous data input in their meta-analysis, and the issue of prevalence determined by the included studies. These factors potentially contributed to overestimation of the discriminative accuracy of blood-based biomarkers. Subsequently, the conclusion that blood-based biomarkers are effective tools for detecting Alzheimer's disease is debatable without correction of these methodological flaws and providing robust and trustworthy estimates.