Skip to main content

Advertisement

Table 3 Characteristics of the highest risk groups (top 10%), according to each model separately (PREADM-H vs. PREADM)

From: Identifying patients at highest-risk: the best timing to apply a readmission predictive model

Characteristics Total PREADM-H high risk; PREADM non high risk PREADM-H non high risk; PREADM high risk P value
N = 1057 n = 229 n = 218
Age, y, mean (SD) 78.9 (7.8) 79.2 (7.6) 80.0 (8.3) 0.272
Male, n (%) 529 (50.0%) 112 (48.9%) 129 (59.2%) 0.037
Socioeconomic status, n (%)
 Low 306 (29.3%) 56 (24.8%) 46 (21.3%)  
 Medium 446 (42.6%) 95 (42.0%) 104 (48.1%)  
 High 294 (28.1%) 75 (33.2%) 66 (30.6%) 0.420
Ethnicity, n (%)
 Jewish 861 (81.5%) 205 (89.5%) 200 (91.7%)  
 Arab 196 (18.5%) 24 (10.5%) 28 (12.8%) 0.518
Before index admission
 CHF, n (%) 649 (61.4%) 113 (49.3%) 128 (58.7%) 0.059
 COPD, n (%) 458 (43.3%) 74 (32.3%) 88 (40.4%) 0.095
 CRF, n (%) 645 (61.0%) 116 (50.7%) 142 (65.1%) 0.003
 Malignancy, n (%) 438 (41.4%) 89 (38.9%) 85 (39.0%) 1.000
 Arrhythmia, n (%) 620 (58.7%) 129 (56.3%) 130 (59.6%) 0.541
 Disability, n (%) 752 (71.1%) 127 (55.5%) 156 (71.6%) 0.001
 Oncology (treatment phase), n (%) 290 (27.4%) 72 (31.4%) 28 (12.8%) < 0.001
 Body mass index, mean (SD) 27.1 (6.2) 27.8 (7.1) 27.1 (5.1) 0.187
 No. hospital admissions in the past year, mean (SD) 5.7 (3.3) 3.4 (1.7) 3.9 (1.8) 0.003
 No. primary care and specialist visits in the past year, mean (SD) 20.9 (16.2) 18.6 (14.6) 16.9 (14.2) 0.213
 No. days from last hospitalization, mean (SD) 39.3 (39.4) 54.0 (55.0) 39.2 (33.4) 0.001
 Residing in hospitals’ catchment area, n (%) 1–190 (0.1–18%) 1–48 (0.4–21%) 0–45 (0–21%) 0.696
During index admission
 Index admission LOS > =5 days 622 (58.8%) 200 (87.3%) 34 (15.6%) < 0.001
 Procedure, n (%) 26 (2.5%) 8 (3.5%) 15 (6.9%) 0.160
 Index admission type: urgent, n (%) 1054 (99.7%) 229 (100.0%) 195 (89.4%) < 0.001
 Hemoglobin level (last) < 12 g/dL, n (%) 933 (88.3%) 226 (98.7%) 107 (49.1%) < 0.001
 Sodium level (last) < 135 mEq/L, n (%) 304 (28.8%) 112 (48.9%) 10 (4.6%) < 0.001
  1. Abbreviations: y years, SD Standard deviation, CHF Congestive heart failure, COPD Chronic obstructive pulmonary disease, CRF Chronic renal failure, No Number, LOS Length of stay, PREADM Preadmission Readmission Detection Model, PREADM-H Preadmission Readmission Detection Model + Hospital model