Prediction models for mild cognitive impairment in older adults: a systematic review and meta-analysis

To systematically review published studies on risk prediction models for mild cognitive impairment (MCI) in older adults.

Systematic review and meta-analysis of observational studies.

A systematic search was conducted across nine Chinese and English databases, ranging from their foundation to March 20, 2025, using a combination of subject terms and free words. The Prediction Model Risk of Bias Assessment Tool (PROBAST) checklist was used to assess the risk of bias and applicability. Discrimination measures for models with the area under the curve (AUC) values from the validated models were pooled by meta-analysis using Stata software (version 17.0), with heterogeneity assessed through the I² index and Cochrane Q test.

We included 24 studies with 72 prediction models. The sample sizes ranged from 119 to 6434 participants across the studies. The most frequently used predictors were education level and age. The reported AUC valves ranged from 0.35 to 0.97. The results of the PROBAST evaluation showed that all studies were at high risk of bias. The pooled AUC value of the seven validated models was 0.83 (95 % confidence interval: 0.76, 0.91), indicating a good level of discrimination.

The development of risk prediction models for MCI in older adults is still in its infancy. Future studies should focus on developing new models with larger samples, rigorous study designs, and multicenter external validation.