Topic 1:

Leveraging Data Analytics to Develop and Implement a Real-time Frequent Hospital Admitter Risk Stratification Model (FAM-FACE-SG score) to Predict “Familiar Faces in Singapore”

 Speaker: Dr Low Lian Leng

Frequent hospital admitters or “familiar faces” to the healthcare system are high-cost users. Early identification of these high risk patients allow for intervention to prevent avoidable readmissions. Leveraging on SingHealth’s established enterprise data repository eHINTs, we developed and validated an implementable real-time Frequent Admitter risk stratification tool. Possible predictors were extracted near real-time from eHINTs and the outcome was frequent hospital admissions, defined as three or more inpatient admissions within 12 months of discharge. Multivariable logistic regression was used to derive the model for the final risk score. The final model contained nine significant predictors (Furosemide intravenous 40 milligrams and above; Admissions in the past one year; Medifund; Frequent Emergency Department use ≥3 in the past six months; Anti-depressants in past one year; Charlson Comorbidity Index; End stage renal failure on dialysis; Subsidised ward stay; and Geriatric patient) and termed as the FAM-FACE-SG score. The area under receiver operating characteristic curve (AUROC) achieved 0.839 (95% CI: 0.825-0.853), outperforming established risk scores. The FAM-FACE-SG score has the advantage of being generated in near real-time and its impact on improving readmission rates in addition to a decision making algorithm is being studied in a pragmatic cluster randomised controlled trial.



Topic 2:

Operating Theatre Dashboard Initiative: The KKH Team Experience

 Speaker: Dr Ong Lin Yin

The Operating Theatre (OT) dashboard was created as part of a SingHealth initiative to optimise OT utilisation. In KK Women’s and Children’s Hospital (KKH), the dashboard went live in February 2016 and selected reports are distributed at either the division or departmental levels on a monthly basis. The benefits and pitfalls of the OT dashboard will be discussed, as well as its potential for future development and how this information will aid clinicians and administrators in providing better care for our patients.

Topic 3:

Does OBIEE Together With REDCap Enhance Hospital-based Registries?

 Speaker: Dr Huang Honghong

The Renal Cell Carcinoma Registry (RCCR) and Benign Prostatic Hyperplasia Registry (BPHR) were developed in 2012 on Research Electronic Data Capture (REDCap) that interacts with Oracle Business Intelligence Enterprise Edition (OBIEE). We aim to highlight the strengths and challenges of using OBIEE and REDCap in the automation of RCCR and BPHR. 
 

A committee was formed consisting of experts from the Department of Urology, REDCap team and OBIEE team to critically evaluate the feasibility and efficacy of the system. Since its implementation in 2012, 1,177 and 1,082 cases have been recorded in RCCR and BPHR respectively. The strengths included: 1) accessibility to the registries via the hospital web with security features; 2) direct access via Sunrise Clinical Manager (SCM); 3) automatic and timely data extraction; 4) visual presentation of data. The challenges faced: 1) IT support needed; 2) expertise required to match data variables; 3) limited benefit for BPHR. The shortcomings could be potentially solved with technical training and careful selection of suitable diseases. 
 

OBIEE together with REDCap partially automated RCCR, greatly enhancing efficiency and security. However, BPHR saw less benefits as most of the BPH data is not available electronically.