Predicting Patients at Risk for 3-Day Postdischarge Readmissions, ED Visits, and Deaths

Abstract

The 3-day REDD model predicts high-risk patients with fair discriminative power. The discriminative power of the 30-day REDD model is also better than the previously reported models under similar settings. The 3-day REDD model has been implemented and is being used to identify patients at risk for AEs.

Social and Behavioral Determinants of Health in the Era of Artificial Intelligence with Electronic Health Records: A Scoping Review

“…Overall, 29.5% of the studies reviewed focused on SBDH factors associated with healthcare access and quality, 24.7% focused on economic stability, 20% focused on social and community context, 16.3% focused on neighborhood and built environment, and 9.5% focused on education access and quality. Widely studied SBDH factors include substance use/abuse (9%) [ 14 , 16 , 20 - 35 ], education (7.3%) [ 16 , 21 , 36 - 46 ], employment status (6.3%) [ 16 , 20 , 21 , 23 , 31 , 36 , 38 , 39 , 45 - 48 ], socioeconomic status (6.3%) [ 29 , 39 , 44 , 46 , 49 - 56 ], lifestyle (5.8%) [ 22 , 29 , 32 , 43 , 57 - 63 ], socioeconomic factors (5.3%) [ 28 , 38 , 39 , 46 , 58 , 63 - 67 ], diet (5.3%) [ 21 , 22 , 26 , 43 , 58 , 68 - 72 ], housing status (5.3%) [ 16 , 21 , 33 , 35 , 38 , 46 , 73 - 76 ], social support (5.3%) [ 14 , 15 , 35 , 74 , 77 - 82 ], physical activity (4.8%) [ 20 , 31 , 40 , 58 , ...…”

“…Twelve studies [ 24 , 28 , 49 , 54 , 58 , 60 , 64 , 70 , 84 , 92 , 93 , 95 ] used tree-based machine learning (ML) algorithms, including random forest and decision trees. Nau et al utilized nonparametric machine learning methods such as Conditional Random Forest (CRF) to identify the combination of community features that are most important for the prediction of obesogenic and obesoprotective environments for children [ 70 ], whereas Agrawal et al also used random forests along with gradient boosting methods and stacked generalization methods to attain their outcome using structured data [ 28 ]. Davoudi et al [ 24 ], Walsh et al [ 64 ], Grinspan et al [ 88 ], Feller et al [ 95 ], and Erickson et al [ 16 ] all made use of random forest variants.…”

“…Overall, 29.5% of the studies reviewed focused on SBDH factors associated with healthcare access and quality, 24.7% focused on economic stability, 20% focused on social and community context, 16.3% focused on neighborhood and built environment, and 9.5% focused on education access and quality. Widely studied SBDH factors include substance use/abuse (9%) [14,16,[20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35], education (7.3%) [16,21,[36][37][38][39][40][41][42][43][44][45][46], employment status (6.3%) [16,20,21,23,31,36,38,39,[45][46][47][48], socioeconomic status (6.3%) [29,39,44,46,[49]…”

“…Overall, 30% of the studies reviewed focused on SDOH factors associated with healthcare access and quality, 25% focused on economic stability, 21% focused on social and community context, 15% focused on neighborhood and built environment, and 9% focused on education access and quality. Widely studied SDOH factors include substance use/abuse (9%) [14], [16], [20]- [35], education (8%) [16], [21], [36]- [46], lifestyle (7%) [22], [29], [32], [43], [47]- [53], employment status (6%) [16], [20], [21], [23], [31], [36], [38], [39], [45], [46], [54], [55], socioeconomic status (6%) [29], [39], [44], [46], [56]- [63], socioeconomic factors (5%) [28], [38], [39], [46], [48], [53], [64]- [67], diet (5%) [21], [22], [26], [43], [48],…”

An interpretable machine learning model for predicting 28-day mortality in patients with sepsis-associated liver injury

“…The Global Burden of Disease Study, published in 2020, analyzed global, regional and national sepsis incidence and mortality rates from 1990 to 2017 and reported that there were approximately 48.9 million cases of sepsis in 2017, with about 11 million sepsis-related deaths, accounting for 19.7% of all deaths worldwide [ 2 ]. High risk of rehospitalization and high cost of treatment for sepsis [ 3 , 4 ]. In the United States, sepsis was the most expensive condition treated, amounting to $38.2 billion or 8.8% of aggregate costs for all hospital stays in 2017 [ 5 ].…”