2.3. Data Collection

All data were retrospectively compiled from the Electronic Medical Record (EMR) of the patients who satisfied the inclusion and exclusion criteria. Demographic data (age and gender) and comorbidities (DM, coronary artery disease, and hypertension, chronic kidney disease, malignancy) of all included participants were compiled. Clinical presentations were classified into groups by two authors to account for reporting bias, and in case of a dispute a third author's decision was used. Based on the primary organ affected at presentation, patient data was classified into (1) pulmonary: including pneumonia (radiological changes on X-ray or CT), pleural effusion, and lung abscess; (2) cutaneous: including soft tissue infections, and infections of nonskeletal tissue surrounding or supporting organs and other structures including subcutaneous tissue and muscle; (3) genitourinary infection of the genital organs and the urinary tract including the kidneys; (4) visceral abscesses; (5) osteomyelitis; and (6) septic arthritis, as well as splenomegaly and lymph nodal involvement. Additionally, key biochemical parameters were extracted and analysed including white blood cells (WBC) count, hemoglobin, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), albumin, and alkaline phosphatase. Mortality for each patient was also collected from death summaries from the EMR system. This study will acknowledge selection bias, as patient data is obtained from a quaternary care referral centre, but no correction is applied to address it.