Current Status of Prediction Models of Traditional Chinese Medicine for Emerging Acute Infectious Diseases
10.13422/j.cnki.syfjx.20251626
- VernacularTitle:突发急性传染病中医药预测模型研究的现况分析
- Author:
Jiaxin ZUO
1
;
Chengyang JING
2
;
Tiantian ZHOU
1
;
Fang HAN
3
;
Di WANG
4
;
Xing LIAO
2
;
Jing HU
1
Author Information
1. Center for Evidence-based Chinese Medicine， Beijing Hospital of Traditional Chinese Medicine（TCM）， Capital Medical University/Beijing Institute of TCM， Beijing 100010，China
2. Institute of Basic Research in Clinical Medicine， China Academy of Chinese Medical Sciences， Beijing 100700，China
3. Beijing Hospital of TCM， Capital Medical University， Beijing 100010，China
4. Beijing Anding Hospital， Capital Medical University， Beijing 100088，China
- Publication Type:Journal Article
- Keywords:
emerging acute infectious diseases;
traditional Chinese medicine;
prediction model research;
methodological quality;
current analysis
- From:
Chinese Journal of Experimental Traditional Medical Formulae
2026;32(22):247-254
- CountryChina
- Language:Chinese
-
Abstract:
ObjectiveTo analyze the current status of traditional Chinese medicine （TCM） prediction model research for emerging acute infectious diseases， in order to provide reference for subsequent model construction and optimization. MethodsEight databases， including PubMed， Cochrane Library， and Embase， were systematically searched to comprehensively collect Chinese and English literature on TCM prediction models for emerging acute infectious diseases. The search period ranged from database inception to July 5， 2024. Reporting quality compliance was evaluated using the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis—Traditional Chinese Medicine （TRIPOD-TCM） guideline. Risk of bias and applicability were assessed using the Prediction Model Risk of Bias Assessment Tool （PROBAST）. ResultsA total of 21 studies were included， comprising 11 early warning model studies， 5 diagnostic model studies， 3 prognostic model studies， and 2 prescription prediction model studies， involving 84 models in total. The diseases involved were mainly viral infectious diseases such as coronavirus disease 2019 （COVID-19） and hepatitis B. The most frequently applied TCM theory was the "five movements and six climates" theory. The most commonly used modeling method was neural network-based approaches. The most frequently reported predictive factors were meteorological variables， including daily mean temperature， wind speed， precipitation， relative humidity， and vapor pressure. In terms of reporting quality， the included studies showed poor item-level compliance， with only 13 sub-items achieving compliance rates above 50%. Overall compliance ranged from 27% to 59% across studies. Although the included studies showed relatively good applicability， all were assessed as having a high risk of bias. The main sources of bias included inappropriate handling of missing data， absence of calibration methods， and unreasonable validation strategies， indicating deficiencies across multiple stages of model development. ConclusionCurrent research on TCM prediction models for emerging acute infectious diseases exhibits distinctive modeling characteristics. These models are based on traditional TCM theories such as the "five movements and six climates"， incorporate TCM-specific features as predictive variables， and use TCM syndromes as outcomes. However， they generally suffer from low reporting compliance and a high risk of bias. Future studies should systematically design model development and validation processes， adhere to TRIPOD-TCM reporting standards， and apply PROBAST or develop dedicated bias risk assessment tools for TCM prediction models to improve methodological quality and promote their translation into clinical practice.
