Toward understanding variations in price and billing in US healthcare services: A predictive analytics approach
作者：
Highlights：
• Multi-class classification proposed for understanding provider billing patterns.
• Ensemble prediction models based on gradient boosted trees perform the best.
• Socioeconomic and demographic factors of the patient base play a significant role.
• Practice excess charge ratio and procedure impact on service are key predictors.
• Results justify limiting Medicare excess charge to avoid high prices.
摘要
•Multi-class classification proposed for understanding provider billing patterns.•Ensemble prediction models based on gradient boosted trees perform the best.•Socioeconomic and demographic factors of the patient base play a significant role.•Practice excess charge ratio and procedure impact on service are key predictors.•Results justify limiting Medicare excess charge to avoid high prices.
论文关键词：Health services,Multiclass prediction,Healthcare cost,Healthcare analytics
论文评审过程：Received 28 January 2022, Revised 23 June 2022, Accepted 18 July 2022, Available online 2 August 2022, Version of Record 11 August 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.118241
