Authors (including presenting author) :
Chan JSP1, Chan ACM2, Chau RMW1, Wong EYW1, Cheung EYY1, Cheung LPC3, Cheng WKH3, Li KK3, Li W3
Affiliation :
1. Physiotherapy Department, Kowloon Hospital
2. Physiotherapy Department, Queen Elizabeth Hospital
3. Department of Orthopaedics and Traumatology, Queen Elizabeth Hospital
Introduction :
Length of stay (LOS) is always a seriously contemplating question and without silver bullet. Early prediction and stratification of hip fracture patients into different care trajectories with compatible rehabilitation pathway/program will definitely help. One of the major impacts on clinical practice is comorbidity. Charlson Comorbidity Index (CCI) was proven to be a valid and reliable tool of measuring comorbidity. Yet, its predictive validity to acute hospital stay was unknown.
Objectives :
The current study was to determine the predictive validity of CCI for the acute-hospital LOS.
Methodology :
Patients admitted to Department of Orthopaedics and Traumatology (O&T) of Queen Elizabeth Hospital involved in surgically repaired with diagnosis hip fracture was enrolled in the study consecutively over one-year period in 2016. For inclusion, the person had to be 65 years of age or older and should have sustained a recent low energy trauma fracture over proximal femur. The exclusion criteria were atypical fracture, pathological fracture, and discharge against medical advice condition. All patients were evaluated with CCI within first-day on admission. The LOS was determined. Receiver Operating Characteristic (ROC) Curve was undertook to assess predictive validity of CCI in respects of having the patient being discharged from the acute O&T bed more than 10 days.
Result & Outcome :
Six hundred and thirty patients (N=174, 27.6% male; N=456, 72.4% female) aged 83.76±7.22 years were included in the study. The median CCI was 6 points. LOS≤10 days was 67.3% (N=424). The predictive validity of the CCI was moderate as indicated by the area under the curve (AUC) of ROC. AUC was 0.574 (95% CI 0.525–0.622, p≤0.01). The optimal cut-off score 4 leaded to a sensitivity of 0.68 and specificity 0.47. This study demonstrated predictive validity of CCI on short or long LOS in acute hospital setting. Admission CCI score informed decision making. It provide scientific base for stratify patient with early identification of potential long or short acute hospital stay subgroups for developing targeted optimized models of care, resource allocation and service improvement .
Co-location with:
