Predicting Chronic Liver Disease in Chinese Adults
Author Information
Author(s): Cong Xue, Song Shuyao, Li Yingtao, Song Kaiyang, MacLeod Cameron, Cheng Yujie, Lv Jun, Yu Canqing, Sun Dianjianyi, Pei Pei, Yang Ling, Chen Yiping, Millwood Iona, Wu Shukuan, Yang Xiaoming, Stevens Rebecca, Chen Junshi, Chen Zhengming, Li Liming, Kartsonaki Christiana
Primary Institution: Peking University
Hypothesis
The study aims to systematically review and validate chronic liver disease prediction models in a diverse Chinese population.
Conclusion
Several models showed good discrimination and calibration, indicating their potential for identifying high-risk populations for chronic liver disease in Chinese adults.
Supporting Evidence
- 57 articles and 114 models were identified in the systematic review.
- 28.4% of the models had undergone external validation.
- Models with high discrimination (C-index ≥ 0.70) were identified for both general and HBV-infected populations.
- Calibration plots indicated that models tended to overestimate risk in the validation cohort.
- Models incorporating non-laboratory parameters showed good performance for risk stratification.
Takeaway
This study looked at different ways to predict who might get chronic liver disease in China, finding some methods that work well.
Methodology
The study involved a systematic review of existing models, meta-analysis of their performance, and external validation using data from the China Kadoorie Biobank.
Potential Biases
There may be selection bias due to the nature of the cohorts used for model development and validation.
Limitations
The study faced limitations such as variability in eligibility criteria across original cohorts and potential biases in model performance estimates.
Participant Demographics
The study involved a diverse population of Chinese adults, including those with various risk factors for chronic liver disease.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website