In southern Ontario a number of local health authorities are pooling data on millions of patients to create a huge repository of information that can be queried to help doctors and nurses being deliver better medical care.
It’s the kind of project that Canada needs to take maximum advantage of data analytics, say experts.
But our medical system is still a long way from taking from that, a Toronto health care analytics conference for clinicians, medical researchers and business intelligence staff was told Tuesday.
“We’re still in our teenage years in analytics,” said Bill Pascal, an Ottawa-based IT health care consultant and former chief technology officer of the Canadian Medical Association.
The goal is to get to predictive and prescriptive analytics, he said, where IT systems can help predict patient outcomes and point clinicians in the right direction for care.
He admitted that analytic tools today are better than they were five years ago.
Still, he said “it will be a 10 year journey. This isn’t going to happen quickly.”
One problem is “we have a surfeit of (patient) information and a lacuna of understanding.” IT solutions for collecting data from a myriad of electronic health records are “clunky,” he complained.
Yet technology is only 10 per cent of the solution, he added, with altering business processes another 30 per cent. The largest part is getting clinicians to change.
He also said that health care staff shouldn’t be swept up in the rush to big data. Keep projects simple, he said.
That was echoed by Dennis Giokas, chief technology officer of Canada Health Infoway, a federal agency that works with the provinces to improve health information systems. Take “cautious first steps,” he told the conference.
In an interview Pascal said hospitals and clinicians don’t need sophisticated business intelligence tools to reap a lot of benefit.
He pointed to one project presented to the conference, an administrative integrated decision support (IDS), which originated at the Hamilton (Ont.) Health Sciences Centre and has expanded.
The IDS is a data repository of patient data from Hamilton and Niagara-area health institutions built around Microsoft SharePoint and a set of Microsoft business intelligence tools to query it. As Wendy Gerrie, the centre’s director of decision support explained, there are nine data sources (such as data from emergency rooms and acute care patients) that tracks a patient’s journey where ever they go — down to did they see a community health care worker.
It also links to the Clinical Connect Web portal, an integrated electronic health record system used in several southern Ontario regions, so medical staff treating a patient can immediately see the history. Approved users can also use the data for some predictive analysis.
At the moment it holds records on four million patients and 35 million patient encounters with clinical staff. But that will grow to 45 million encounters when three health care regions join by the end of March, 2014. About 650 analysts. clinicians and administrators currently use the system.
Gerrie said users can, for example, delve into how many people in a given area have been admitted to hospital, identify the five per cent of users that today typically cost the health care system the most money and then look for patterns — sex, age, diagnosis, were they discharged to home or a secondary care facility etc.
The findings can help improve patient’s care coordination, understand demographic and clinical patterns and trends right down to a patient level, and improve system-wide performance for patients, she said.
What’s important is that three other area health districts, including Toronto, are joining the IDS — which operates on a software-as-a-service (SaaS) model — which will expand the data repository. Almost every Ontario health district (or local health integration network) wants in, Gerrie said.
What the IDS does not yet have is family doctor records. Nor is it a real-time support system.
But in an interview Pascal said it is a good example of getting a lot through basic patient analytics.
There are some institutions around the world that are leaders in health care analytics, he added, but Canada is neither behind nor ahead of others.
This is the second data analytics for health care conference organized by the Strategy Institute.
A number of case studies are being presented during the two-day conference, including how an Eastern Ontario group analyses patient data collected by paramedics; how the Ontario Brain Institute uses a data to drive research; how Swiss hospitals use analytics to fight infectious diseases; and how a Texas children’s hospital uses it to improve outcomes.
Cost Benefit Case for Pure Data System for Analytics
Comparing Costs and Time to Value with Oracle Exadata Database Machine