Big Data to help in Malaysian healthcare policy and planning

Socso is interested in utilising big data to help provide compensation to injured workers faster, research accident causes and non-communicable diseases prevention, assessing invalid claims, and determining the extent of disabilities in the local context. Photo: Filepic

In her opening address to the Big Data in Healthcare Forum held recently at Taylor’s University’s Lakeside Campus in Subang Jaya, Selangor, Clinical Research Centre director Dr Goh Pik Pin noted that the Health Ministry (MOH) has a lot of data from the nation’s public healthcare system, but it cannot really be used.

“Last year, when we were doing the healthcare information system, we found a lot of data couldn’t be used because it was stored in various places, in different formats, and with the recent Personal Data Protection Act, it is considered protected and not even available to MOH,” she said.

However, she stressed that the Govern-ment is very committed to ICT, which is one of the strategic pillars of the 11th Malaysia Plan (2016-2020).

For healthcare, one of the practical manifestations of this is the Malaysia Health Data Warehouse (MyHDW).

A project in the making since 2011, its first form is due to be rolled out this August, focusing on the cardiology field.

Said MOH Planning Division deputy director and Health Informatics Centre head Dr Md Khadzir Sheikh Ahmad: “This is the first time we are running a project involving local technology, which is from Mimos.

“One of the main reasons we chose Mimos was the security-layered component in our Malaysia Health Data Warehouse.”

Dr Khadzir and his team are preparing to roll out the Malaysia Health Data Warehouse this August. — Taylor’s University
Dr Khadzir and his team are preparing to roll out the Malaysia Health Data Warehouse this August. Photo: Taylor’s University

As Mimos is developing the software, it would be patented and owned by the Malaysian Government, not by a foreign provider.

Dr Khadzir shared that MyHDW would be managing both structured and semi-structured data, using Snomed-CT (Systematized Nomenclature of Medicine - Clinical Terms) and MiHarmony.

MiHarmony is the tool that will be used to code and map the data from various healthcare databases into MyHDW using the international standardised vocabulary of clinical terminology in Snomed-CT.

“The first dataset that will go into the Malaysia Health Data Warehouse is what we call Sistem Maklumat Rawatan Perubatan (SMRP).

“This is the system that will collect a patient’s every visit to the in-patient (ward) and day-care, and later to include the outpatient and every other visit,” said Dr Khadzir.

He added that one of the main problems in the public healthcare system is that patients’ data are in silos. This means that every visit to a public hospital requires their information to be keyed in again.

“We are trying to bring the data flow from every patient visit to the hospital in SMRP automatically to the Patient Registry Information System (PRIS),” he said.

Both SMRP and PRIS are databases that will go live in August with the MyHDW.

He also explained that not all patient information would be collected by the MyHDW; only that which can be used for public health planning and research.

“We are responsible for secondary-use data, where data is collected from all facilities, and it is meant for planning and what-not. We don’t collect everything that is collected at the primary facility,” he said.

According to Dr Khadzir, MyHDW is meant to be a trusted, comprehensive source of healthcare data, which must come from every healthcare visit, including those to allied health professionals, traditional and complementary medicine practitioners, and maternal health.

Related story: The benefits of big data in healthcare

“In the data warehouse, we should be able to link them all up,” he said, giving the example of being able to track how many patients diagnosed with stroke go on to visit physiotherapists, occupational therapists and traditional medicine practitioners.

“So, this is the intention: to have a snapshot of every visit to the healthcare system and to provide value-added data to the people in the Ministry of Health to make policy.”

Prevention and compensation

MOH is not the only public body in Malaysia interested in utilising big data to help its planning.

Socso Prevention and Health Promotion Unit manager Dr Azlan Darus also spoke at the forum on the potential use of such data to the Social Security Organisation.

Currently covering around 15.5 million workers in Malaysia, and managing just under 112,000 new claims a year, Dr Azlan noted that there are a few things Socso would like to do that can probably only be accomplished with big data.

In terms of operations, he said that the idea is to decrease the time it takes to initiate and process accident claims.

“For example, we tried to link up with the police database; we didn’t want to wait for people to come to us, we wanted to get real-time information from police reports and go to them and help them with their claims,” he said.

Another example was access and integration of medical reports to see whether the person is still able to work after their accident. This would be for both the practical purpose of giving out invalid grants, and the information-gathering purpose of seeing how long rehabilitation really takes for future planning.

“We can get it from the person themselves, but would like to get it more quickly, in order to provide compensation faster,” said Dr Azlan.

Socso is also interested in research into areas like accident prevention and evidence-based impairment determination.

Dr Azlan gave the example of how a worker who loses his entire right, dominant hand is considered to be 60% impaired.

“Where do we get this number from? We follow the American Medical Society guidelines, which is based on Americans,” he said.

But in order to find out what percentage of impairment really applies to Malaysians, Dr Azlan noted that they would need the impairment level, type of job and education level that will help determining the worker’s ability in finding another job, among others.

“How can we do this?

“Big Data will help,” he said.

Article type: metered
User Type: anonymous web
User Status:
Campaign ID: 46
Cxense type: free
User access status: 3
Join our Telegram channel to get our Evening Alerts and breaking news highlights

Stories You'll Enjoy