Abstract Objective: To summarize the risk prediction models of chronic
disease in Chinese medicine, describe their performance, and assess
suitability of clinical or administrative use. Methods: The China
National Knowledge Infrastructure and Wanfang Data were searched through
February 2021, and hand searches were performed of the retrieved
reference lists. Dual review was conducted to identify studies of
prediction models of chronic disease in Chinese medicine. Results: From
399 citations reviewed, 17 studies were included in the analysis. Most
of the studies were from single-centers (50%) or did not external
validated (81.25%). The sample sizes were smaller and the models’
discrimination were larger compared with studies in fully western
medicine. All the models used both laboratory findings and subjective
judgements from doctors or patients. 9 models concentrated on diabetes
mellitus or cardiovascular disease, and showed better performance and
clinical application. Conclusions: The prediction models of chronic
disease in Chinese medicine have unique advantages due to their
considerations of doctors’ and patients’ subjective judgement. Diabetes
mellitus and cardiovascular disease prediction models were in higher
quality and clinical usability. Efforts to improve their quality are
needed as use becomes more widespread.
The uncertainty with using risk prediction models for individual decision making: an exemplar cohort study examining the prediction of cardiovascular disease in English primary care
