Table 1.
Feature definitions.
| Extraction time window | Feature | Units | Details |
|---|---|---|---|
| Special features | Age | Years | Computed at diagnosis date, based on birth year entry |
| Gender | NA (0: male, 1: female) | No time window restrictions | |
| Ethnicity (Hispanic) | NA (binary) | No time window restrictions | |
| Ethnicity (non-Hispanic) | NA (binary) | No time window restrictions | |
| Race (African American) | NA (binary) | No time window restrictions | |
| Race (Asian) | NA (binary) | No time window restrictions | |
| Race (Caucasian) | NA (binary) | No time window restrictions | |
| Race (Multiracial) | NA (binary) | No time window restrictions | |
| Acute features | Acute bronchitis | NA (binary) | ICD-10: J20.*, J40 and ICD-9: 466.0, 490 |
| Anorexia | NA (binary) | ICD-10: R63.0, R63.8 and ICD-9: 783.0, 783.9 | |
| Body temperature | ∘C | LOINC: 8310-5 | |
| Confusion | NA (binary) | ICD-10: R41.0, R41.82 and ICD-9: 780.97 | |
| Cough | NA (binary) | ICD-10: R05 and ICD-9: 786.2 | |
| Diarrhea | NA (binary) | ICD-10: R19.7 and ICD-9: 787.91 | |
| Fatigue | NA (binary) | ICD-10: R53.1, R53.81, R53.83 and ICD-9: 780.79 | |
| Fever | NA (binary) | ICD-10: R50.9 and ICD-9: 780.60 | |
| Headache | NA (binary) | ICD-10: R51 and ICD-9: 784.0 | |
| Hemoptysis | NA (binary) | ICD-10: R04.2 and ICD-9: 786.30 | |
| Hospitalization (inpatient) | NA (binary) | Considered if reported admission–discharge period overlapping with extraction time window | |
| Myalgia | NA (binary) | ICD-10: M79.1, M79.10, M79.11, M79.12, M79.18 and ICD-9 729.1 | |
| Pneumonia | NA (binary) | ICD-10: J12.*, J13, J14, J15.*, J16.*, J17, J18.* and ICD-9: 480.*, 481, 482.*, 483.*, 484.*, 485, 486, 487.0, 488.01, 488.11, 488.81 | |
| Rhinorrhea | NA (binary) | ICD-10: J34.89 and ICD-9: 478.19 | |
| Shortness of breath | NA (binary) | ICD-10: R06.02 and ICD-9: 786.05 | |
| Sore throat | NA (binary) | ICD-10: J02.9 and ICD-9: 462 | |
| Sputum | NA (binary) | ICD-10: R09.3 and ICD-9: 786.4 | |
| Vomiting | NA (binary) | ICD-10: R11.10 and ICD-9: 536.2, 787.03 | |
| Chronic features | Active smoking | NA (binary) | Based on reported habit |
| Asthma | NA (binary) | ICD-10: J45.* and ICD-9: 493.* | |
| BMI | kg/m2 | LOINC: 39156-5, or computed from weight (29463-7) and height (8302-2) | |
| Cardiovascular disease | NA (binary) | ICD-10: I20.*, I21.*, I25.*, I48.*, I50.*, I63.*, I65.*, I67.*, I73.* and ICD-9: 410.*, 412.*, 413.*, 414.*, 427.*, 428.*, 429.*, 433.*, 434.*, 437.*, 443.* | |
| Chronic kidney disease | NA (binary) | ICD-10: E10.21, E10.22, E10.29, E11.21, E11.22, E11.29, I12.0, I12.9, I13.0, I13.10, I13.11, I13.2, N04.*, N05.*, N08, N18.*, N19, N25.9 and ICD-9: 250.40, 250.41, 250.42, 250.43, 403.*, 404.*, 581.81, 581.9, 583.89, 585.*, 588.9 | |
| Chronic obstructive pulmonary disease | NA (binary) | ICD-10: J44.* and ICD-9: 491.*, 493.2* | |
| Diabetes | NA (binary) | ICD-10: E10.*, E11.*, E13.* and ICD-9: 250.* | |
| Hypertension | NA (binary) | ICD-10: I10, I15.* and ICD-9: 401.*, 405.* | |
| Immunodeficiency | NA (binary) | ICD-10: B20, D80.*, D81.*, D82.*, D83.*, D84.*, D86.*, D89.* and ICD-9: 042, 279.* | |
| Nicotine dependence | NA (binary) | ICD-10: F17.* and ICD-9: 305.1 | |
| Obesity | NA (binary) | ICD-10: E66.0*, E66.1, E66.2, E66.8, E66.9 and ICD-9: 278.00, 278.01, 278.03 | |
| Paralytic syndromes | NA (binary) | ICD-10: G80.*, G81.*, G82.*, G83.* and ICD-9: 342.*, 343.*, 344.* |
Feature names, units and details (e.g., ICD and LOINC codes) grouped by extraction time window specifications. Binary features encode whether there is an entry in the database for the specific item of interest or not. The symbol * represents a wildcard for ICD subcategory codes.