Helping Hospitals

National University Hospital (NUH) is a specialist public hospital providing medical, surgical and dental services for patients referred there. PHOTO: ST FILE
Ang Mo Kio-Thye Hua Kwan Hospital provides specialised care to many elderly patients referred to the hospital. PHOTO: LIANHE WANBAO
In Singapore, Parkway Pantai is the largest private healthcare operator with four hospitals: Mount Elizabeth, Mount Elizabeth Novena, Gleneagles (above) and Parkway East. PHOTO: ST FILE

Data analytics in healthcare is the use of large troves of data to read and interpret trends, weaknesses and strengths in hospital services, enabling hospitals to make decisions to better allocate resources or improve services.

Dr Lim Er Luen, National University Hospital's (NUH) chief medical information officer, split it into three key stages: descriptive, predictive and prescriptive.

Descriptive analytics is the use of data to broadly answer the question, "What is happening now?" NUH does this by making charts and graphs out of many types of data, for example.

The next level, predictive analytics, combines this data with rules and algorithms to determine, "What will happen next?" With large amounts of data, machines can be "trained" to predict outcomes.

The final level is prescriptive analytics. This is focused on answering the question, "What happens if I do this?" This guides the hospital in deciding what is the best course of action for its patients.

This level of analytics, while still in the future, can provide personal information to individual patients, thereby making diagnoses and assessments much more precise and meaningful, said Dr Lim.

Raffaella Nathan Charles


NUH

Studying trends

National University Hospital (NUH) is a specialist public hospital providing medical, surgical and dental services for patients referred there.

It is the only public hospital in Singapore to offer a paediatric kidney and liver transplant programme.

The hospital hasan ongoing project to tackle data analytics.

It has come up with graphs and software to study the trends, strengths and weaknesses of its service.

Some results seen were an improvement of outcomes in the treatment of pneumonia, streamlining the care of its patients undergoing knee replacement surgery, and keeping the occupancy of hospital beds optimised to reduce patient waiting time.

Based on data, the hospital has also managed to determine whether a patient may have certain emergencies such as appendicitis; or whether a patient may have breast cancer, based on X-rays; and the likelihood that a patient may return to hospital just after being discharged.

Its data analytics project has enabled the hospital to make improvements in clinical, financial and operational domains, and reduce waste.

Dr Lim Er Luen, NUH's chief medical information officer, said that overall, Singapore's healthcare sector has made significant inroads with data analytics, while it continues to look into population health and predictive analytics, and can expect artificial intelligence to play a significant part in future.


AMK-THK Hospital

Post-stroke rehab

Ang Mo Kio-Thye Hua Kwan Hospital provides specialised care to many elderly patients referred to the hospital. This includes palliative care, dementia care and care for acute illnesses or injuries.

The hospital also collates clinical data to uncover and analyse areas of weakness.

Through this process, the hospital has developed strategies to improve current practices, said acting assistant director of nursing Stephen Tan.

Through its data records, the hospital found that half of its patients required post-stroke rehabilitation. So it set up a stroke rehabilitation unit to conduct intensified speech therapy sessions and training. Speech therapists thus spent less travelling time between wards, said Mr Tan.

Data also showed that 80 per cent of the hospital's patients referred there wore adult diapers on arrival.

The hospital thus set up a ward diaper-monitoring process to identity patients with incontinence and work on weaning them off diapers.

Mr Tan said: "There is definitely a bigger role in the future for healthcare analytics in intermediate and long-term care services, but the process of collecting, analysing the data is time-consuming, and requires a specialised skill set.

"There will be challenges to sustain such practices in the long run unless more resources are allocated to the cause."


Parkway Pantai

Pre-emptive measure

In Singapore, Parkway Pantai is the largest private healthcare operator with four hospitals: Mount Elizabeth, Mount Elizabeth Novena, Gleneagles and Parkway East.

It also owns Parkway Shenton, a large network of primary healthcare clinics and services, ParkwayHealth Radiology, ParkwayHealth Laboratory and Parkway College. It has 53 clinics islandwide.

Mr Linus Tham, group chief information officer of Parkway Pantai, said healthcare is an "information-driven industry".

"The use of data analytics will only increase as healthcare providers seek to better understand the profile and medical needs of their customers." he said.

Parkway Pantai also makes use of insights from data analytics to plan, organise and enhance services to better serve patients.

For example, primary healthcare network and third-party administration services provider Parkway Shenton utilises data analytics to better partner its patients.

It also uses data to monitor and manage the health of the staff of its corporate clients and maintain their well-being through pre-emptive measures.

This enables the healthcare provider to recognise and address potential health issues, not just when patients require hospitalisation, but throughout the year, said Mr Tham.

Join ST's WhatsApp Channel and get the latest news and must-reads.

A version of this article appeared in the print edition of The Straits Times on December 08, 2017, with the headline Helping Hospitals. Subscribe