Table 1.
List of variables considered for predictive modeling and clustering analysis.a
| Variables | Description | |
| Sociodemographic data | ||
|
|
1: Age (100%) | Patient’s age (numerical) |
|
|
1: Gender (100%) | Patient’s sex (binary, male/female) |
|
|
1: HBA (100%) | Health Basic Area (categorical, 105 levels) |
| Medical complexity | ||
|
|
3: AMG score [18-21] (100%) | Adjusted Morbidity Groups score (numerical) |
|
|
2: SIIP Plan [24] (90.45%) | Shared Individual Intervention Plan (binary, yes/no) |
|
|
2: CCP [24] (90.45%) | Complex chronic patient (binary, yes/no) |
|
|
2: ACP [24] (90.45%) | Advanced chronic patient (binary, yes/no) |
| Patient’s functional capacity | ||
|
|
2: Barthel [25] (90.45%) | Barthel Index (numerical, 0-100) |
|
|
2: Lawton Brody [26] (82.97%) | Lawton Brody Index (numerical, 0-8) |
|
|
2: Pfeiffer [27] (90.58%) | Pfeiffer Index (numerical, 0-10) |
|
|
2: Braden [28] (78.71%) | Braden Index (numerical, 0-23) |
|
|
2: Geriatric syndrome [29] (100%) | Geriatric syndrome label (binary, yes/no) |
| Social frailty indicators | ||
|
|
2: MNA [30] (71.87%) | Mini Nutritional Assessment Index (numerical, 0-30) |
|
|
2: TSRI [31] (100%) | Table of Social Risk Indicators (numerical, 0-6) |
|
|
2: Barber [32] (81.9%) | Barber Index (numerical, 0-9) |
|
|
2: Dependence (100%) | Dependence label (binary, yes/no) |
| Unhealthy lifestyle habits | ||
|
|
2: BMI (74.19%) | BMI (numerical) |
|
|
2: Physical activity (75.61%) | Patient’s physical activity (categorical, 3 levels) |
|
|
2: Alcohol intake (73.94%) | Patient’s alcohol intake (categorical, 3 levels) |
|
|
2: Smoking (73.03%) | Patient’s smoking habits (categorical, 3 levels) |
| Use of health care resources | ||
|
|
3: Hospital admissions (100%) | Number of admissions during the previous 12 months (numerical) |
|
|
3: Emergency room visits (100%) | Number of emergency room visits during the previous 12 months (numerical) |
|
|
3: Primary care encounters (100%) | Number of encounters with primary care professionals during the previous 12 months (numerical) |
|
|
3: Outpatient visits (100%) | Number of specialized care outpatient visits during the previous 12 months (numerical) |
|
|
3: Medication (100%) | Number of drugs prescribed during the previous 12 months (numerical) |
|
|
3: Health care expenditure (100%) | Total health care expenses of the previous 12 months in euros (numerical) |
| Acute episode complexity | ||
|
|
1: Composite Queralt Index [22,23] (100%) | Composite Queralt Index (numerical) |
|
|
1: Type of hospitalization (100%) | Type of hospitalization (binary, hospital avoidance/usual care) |
|
|
1: Length of stay (100%) | Total hospitalization days (numerical) |
|
|
1: Number of active diagnoses (100%) | Number of active diagnoses at admission (numerical) |
|
|
1: Leukocytea (87.33%) | Leukocyte count (numerical) |
|
|
1: Lymphocytesa (87.33%) | Percentage of lymphocytes (numerical) |
|
|
1: Hemoglobina (87.33%) | Hemoglobin concentration (numerical) |
|
|
1: RDWa (87.33%) | Red blood cell distribution width (numerical) |
|
|
1: Glucosea (87.33%) | Glucose concentration (numerical) |
|
|
1: Creatininea (87.33%) | Creatinine concentration (numerical) |
|
|
1: Sodiuma (87.33%) | Sodium concentration (numerical) |
|
|
1: Potassiuma (87.33%) | Potassium concentration (numerical) |
| Immunization records | ||
|
|
2: Vaccination, flua (100%) | Flu vaccine administration (binary, yes/no) |
|
|
2: Vaccination, pneumococcal 13a (100%) | Pneumococcal 13 vaccine administration (binary, yes/no) |
|
|
2: Vaccination, pneumococcal 23a (100%) | Pneumococcal 23 vaccine administration (binary, yes/no) |
aVariables used only for predictive modeling purposes. The number before the variable’s name indicates the database from which the information was retrieved: (1) HCB’s electronic medical records; (2) primary care’s electronic medical records; and (3) the Catalan Health Surveillance System. The percentages of data availability are displayed after variable’s name.