Variable name | Definition and Measure Used | Variable classification in analysis |
---|---|---|
Frailty | A clinical syndrome characterized by increased vulnerability to adverse health outcomes when exposed to acute stressors. Assessed using Clinical Frailty Scale (CFS, range 1–9) | Analysed as a categorical variable: non-frail (CFS1-4), mild frailty (CFS 5), moderate frailty (CFS 6), severe frailty (CFS ≥ 7). |
Depression | Mood assessed using Patient Health Questionnaire-2 (PHQ-2) responses and medical records for prior history of depression | PHQ-2 and history combined to derive 4 categories: PHQ-/ History-, PHQ-/History+, PHQ+/History±, uncommunicative |
Malnutrition | Nutritional status assessed using 3-minute Nutritional Screen (3MNS, range 0–9) | Binary: malnutrition (3MNS ≥ 3) vs. non-malnourished |
Cognitive impairment | Operationalized as impaired performance on Abbreviated Mental Test (AMT, range 0–10), or established diagnosis of dementia from clinical records. | Binary variable: AMT and history dichotomized as cognitive impaired (AMT < 8 or history of dementia) vs. unimpaired (AMT ≥ 8, no history) |
Delirium | An acute neuropsychiatric disorder characterized by inattention, global cognitive dysfunction, disturbance in consciousness, assessed using Confusion Assessment Method (CAM) with 2 core features and at least 1 of 2 other supportive features diagnostic of delirium; any 2 features not meeting diagnostic algorithm classified as subsyndromal | Analysed as a binary variable: delirium (including subsyndromal) vs. no delirium |
Functional decline | Operationalized as the need for incremental assistance in activities of daily living (ADLs: feeding, toileting, dressing, bathing, walking) at discharge compared with patient’s baseline. Each ADL rated as independent, needing assistance or dependent | Analysed as a binary variable: functional decline (any ADL registering higher level of assistance relative to baseline) versus no functional decline (stable or improved) |
Oral intake | Optimal intake was classified as consuming at least ¾ share of each meal ≥ 50% of all provided meals during the admission based on daily review of intake-output charts | Analysed as a binary variable: poor oral intake versus optimal intake |
Comorbidity burden | Assessed using Charlson’s Comorbidity Index (CCI) with weighted CCI - low (0 points), medium (1–2 points), high (3–4 points) and very high (≥ 5 points) | Analysed as a categorical variable: low, medium, high, very high |
Severity of illness | Severity of Illness Index (SII) with 4 levels provide a measure of the burden of illness and how sick a patient is while in hospital, to allow meaningful comparison across diagnostic groups. | Binary variable: mild (Level 1 or 2) versus severe (Level 3) as patients in intensive care or high dependency (Level 4) excluded from study |
Geriatric syndromes | A range of multifactorial conditions that do not fit into discrete disease categories, share risk factors and commonly co-exist in older adults. Admission diagnoses categorized as geriatric syndromes include falls, delirium or cognitive impairment, and functional decline. Frailty and poor oral intake considered geriatric syndromes. | Each syndrome analysed as a binary variable (present or absent) |