Table 2.

Characteristics of the included predictive modeling studies.

Study, year	Country	Procedure	Total patients	Data source	Outcome(s) of interest	Number of input features
Navarro et al., 2018	USA, UK	TKA	141,446	SPARCS	LOS	8
Ramkumar et al., 2019	USA, UK	TKA	170,766	NIS, OrthoMiDaS OME	LOS	15
Ramkumar et al., 2019	USA, UK	THA	79,226	NIS, OrthoMiDaS OME	LOS	15
Ramkumar et al., 2019	USA, UK	THA	122,334	SPARCS	LOS	8
Lee et al., 2019	USA	Both	525	Local (USA)	90-d readmission	33
Gabriel et al., 2019	USA	THA	960	Local (USA)	LOS	9
Wei et al., 2021	USA	TKA	25,115	NSQIP	LOS	11
Han et al., 2021	China	TKA	1298	Local (China)	LOS	27
Kugelman et al., 2021	USA	THA	1409	Local (USA)	LOS	15
Yeo et al., 2022	USA	TKA	10,021	Local (USA)	DOS	15
Klemt et al., 2022	USA	rTKA	2588	Local (USA)	LOS	26
Lopez et al., 2022	USA	Both	424,443	NSQIP	LOS	20
Abbas et al., 2022	Canada	TKA	302,300	NSQIP	LOS, DOS	29
Motesharei et al., 2022	France	TKA	1061	Local (USA)	DOS	17
Zalikha et al., 2022	USA	TKA	305,577	NIS	LOS	15
Johannesdottir et al., 2022	Denmark	Both	9512	Local (Denmark)	LOS	22
Klemt et al., 2022	USA	TKA	10,021	Local (USA)	90-d readmission	24
Li et al., 2022	China	TKA	1590	Local (Singapore)	LOS	14
Kugelman et al., 2022	USA	TKA	899	Local (USA)	LOS	15
Trunfio et al., 2022	Italy	THA	2515	Local (Italy)	LOS	15

NIS, National Inpatient Sample; NSQIP, National Surgical Quality Improvement Program; OrthoMiDaS OME, Orthopedic Minimal Data Set Episode of Care; rTKA, revision total knee arthroplasty; SPARCS, New York State Department of Health’s Statewide Planning and Research Cooperative System.