What It Is

The Multiple Readmissions Predictive Model uses a combination of clinical theory and machine learning to automatically identify patients who are likely to be readmitted multiple times over the next year.​

The model supports the Hospital to Home programme through identifying high-risk patients for enrolment into the programme. Through the programme, nurses visit patients and their caregivers at their homes, educate them on caring for themselves, and help them with their needs such as arranging for meals, change of wound dressings or follow-up appointments. The programme is targeted to benefit 19,000 patients over the next year.​


​The Multiple Readmissions Predictive Model was developed by our IHiS team, together with National Healthcare Group (NHG), National University Health System (NUHS) and Singapore Health Services (SingHealth). It is the first predictive model in Singapore that is used across all public healthcare clusters.​

How It Works
Traditionally, care support programmes identify patients who are at risk of multiple hospital readmissions, through their historical data. These risk scoring methods have limitations in handling patients with multiple co-morbidities.

The Multiple Readmissions Predictive Model is able to analyse multi-dimensional facets of the patient, ranging from multiple co-morbidities, chronic diseases to end of life conditions, and has an accuracy of seven in 10 patients predicted. Using over a thousand indicators, which include patient age, the number of inpatient admissions and total length of stay in the past two years, the model automatically identifies patients who have a history or are at risk of multiple readmissions. The team is looking at moving the Multiple Readmissions Predictive Model further upstream to primary care settings, to enable early intervention and delay the progression of their conditions.

How We Implemented Predictive Analytics

1. Business Integration Approach
Develop as part of H2H service involving clinicians, care team, policy maker, administrator which give clear value & accelerate the adoption.

2. Existent of Analytics Infrastructure
Business Research Analytics Insights Network (BRAIN) deployed as national/common analytics platform to support Ministry, agencies, Public Health Institutions (PHIs) self-help analytics by:

i. Bring data together
ii. Patient linked
iii. Harmonise
iv. Anonymise
v. Common toolsets and capabilities
vi. Collaborate in secured environment

3. Multi-discipline Skill Sets
• Respiratory and critical care
• Family Medicine and continuing care
• Bio-informatics
• Health economics
• Health delivery
• Health research
• Data science and analytics

4. Agile & Scientific Methodology
Scientific & systematic development approach that engage to gain trust of user.

5. Volume & Veracity Data
1.4 billion data points   |   7 million records   |   200+ variables   |   3 years data Automated daily generation


To Patients & Caregivers
• Improved care support and potential reduction of length of stays in the hospital

With the early identification of high-risk patients, nurses are able to follow up with phone calls or arrange for home visits if necessary. This allows for timelier intervention to help reduce patients’ average length of stay in the hospitals and healthcare utilisation.

To Staff
• Productivity gains with automatic identification of patients

Previously nurses had to spend almost half a day manually screening through the entire inpatient list and going round the wards, speaking to newly admitted patients and their care team to identify those at risk of multiple readmissions.

The predictive model automatically flags out high-risk patients from the hospital’s daily inpatient admission list. This reduces nurses’ assessment workload by up to 90%, and they can spend more time focusing on direct patient care

Supporting Partners