Optimizing health organization performance and patient care outcomes via predictive analytics: implications for data science and artificial intelligence research

Purpose This commentary presents the analytic development of patient classification, health resource use and outcome research and identifies opportunities to perform longitudinal research. Design/methodology/approach The authors use a transdisciplinary framework to formulate multilevel models for ascertaining the causal relationship between hospital efficiency and effectiveness in panel data analysis. Findings The longitudinal design of organization research enables to delineate the relationship between hospital performance and quality of care in future research. Research limitations/implications The inclusion of multivariates in health organization research and modeling is pivotal to the identification of a comprehensive set of predictor variables. The authors signify the need to build a systems-oriented theoretical framework to integrate micro- and macro-level predictor variables in conducting data analysis. Practical implications The authors signify the need to build a theoretical framework to integrate micro- and macro-level predictor variables in conducting data analysis. Social implications Health organization research is essential to broaden the scope of health services research and policy development, particularly related to global health as noted in the promotion of sustainable development and health goals. Originality/value Health organization research should include a complex set of exogenous and endogenous variables in designing and modeling the determinants of hospital performance and patient care outcomes.