AUTHOR=Jing Cheng-yang , Zhang Le , Feng Lin , Li Jia-chen , Liang Li-rong , Hu Jing , Liao Xing 

TITLE=Recommendations for prediction models in clinical practice guidelines for cardiovascular diseases are over-optimistic: a global survey utilizing a systematic literature search

JOURNAL=Frontiers in Cardiovascular Medicine

VOLUME=Volume 11 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2024.1449058

DOI=10.3389/fcvm.2024.1449058

ISSN=2297-055X

ABSTRACT=Background: To synthesize the recommendations on prediction models in cardiovascular clinical practice guidelines (CPGs) and to assess the methodological quality of relevant primary modeling studies. Methods: We performed a systematic literature search of all available cardiovascular CPGs published between 2018 and 2023 that presented the specific recommendations (whether support or nonsupport) on at least one multivariable clinical prediction model. For guideline-recommended models, the assessment of methodological quality for their primary modeling studies was further conducted by Prediction model Risk Of Bias ASsessment Tool (PROBAST). Results: In total, 46 qualified cardiovascular CPGs were included, with 69 prediction models and 80 specific recommendations accordingly. Of the 80 specific recommendations, 74 had moderate to strong strength supporting 57 models (53 for fully recommended and 4 for conditional recommended) into cardiovascular practice. Most of guideline-recommended models were devoted to predicting prognosis outcomes (53/57, 93%) in primary prevention and tertiary prevention, focusing primarily on long-term risk stratification and prognosis management. A total of 10 conditions and 7 types of target population were involved in 57 models, while heart failure (HF) (14/57, 25%) and general population with or without cardiovascular risk factor(s) (12/57, 21%) had received the most attention from the guidelines.The assessment of methodological quality of 57 primary studies on development of the guidelinerecommended models revealed that only 40% of modeling studies were at low risk of bias (ROB). The concerns of high ROB were mainly from the analysis domain and participants domain. Conclusions: Global cardiovascular CPGs presented an unduly positive appraisal of the existing prediction models in terms of ROB, leading to stronger recommendations than were warranted. Future cardiovascular practice may benefit from the well-established clinical prediction models with better methodological quality and extensive external validation.