ICES takes pulse of Ontario healthcare with SAS® Analytics

The Institute for Clinical Evaluative Sciences (ICES) produces scientific research that influences healthcare policy in Ontario. But it has 17 years’ worth of data in complex relational databases, which must be thoroughly de-identified to protect the privacy of patients. ICES is using SAS® software to manipulate large complex data sets and to evaluate chronic care and map chronic diseases to improve the lives of Ontarians.

Customer Success Video

Check out this video to learn more about ICES and its successes with SAS.

View Video

• Windows Media Player
• Real Player

(Runtime: 4 mins.)

---

Customer Viewpoint
You have questions; our customers have answers. Check out this video Q&A.

Jan Hux, MD
Chief Operating Officer
Institute for Clinical Evaluative Sciences

View Video

• Windows Media Player
• Real Player

---

(Requires Windows Media Player 6.4.7 or higher or RealPlayer 6 or higher)

Since its inception in 1992, ICES has played a key role in providing scientific insights to help policymakers, planners and healthcare practitioners shape the future direction of Ontario’s healthcare system. By examining patterns of disease, healthcare delivery and healthcare outcomes for Ontarians, ICES plays an integral role in supporting decision making, recommendations and informing changes in the province.

With upward of 150 faculty and staff, and more than 100 projects under way at any given time, ICES has huge amounts of data to sift through, including 17 years of data on every hospitalization, physician appointment and publicly funded drug claim in the province. Each month, an additional 20 million physician claims to the government for patient services are added to the system.

Under the province’s privacy legislation, ICES is required to make population-based health information anonymous to protect the confidentiality of patients and then link data from different sectors to create longitudinal profiles of utilization and outcomes. According to Dr. Jan Hux, ICES Chief Operating Officer, this linked information allows researchers to obtain a more comprehensive view of specific healthcare issues, and is used to show gaps in the system and inform health policy.

“We’ve been SAS users from the beginning,” said Hux. “SAS can handle very large terabytes of data without a hiccup. We need to manipulate complex relational databases, not just flat files, and SAS provides the advanced analytic techniques we need to help us deal with thorny methodological issues.”

Analytics staff has developed macros in SAS, so they don’t end up reinventing the wheel. They can create, for example, a standardized way to identify a breast cancer patient or identify the life expectancy of that patient. That knowledge can then be applied to subsequent research projects.

“It’s an extremely powerful and effective tool for marshalling and analyzing very large, complex data sets,” said Hux. “There’s also an advantage in that it’s the lingua franca of Canada’s top universities. We have a training program in place for PhD and master’s students, and using a commonly spoken language like SAS enables us to capitalize on their training.”

As part of a geographic expansion, ICES is creating links with universities in Ontario. The data will reside with ICES, but the universities will be able to visualize that data over a secure interface and run SAS routines. “Since they’re already using SAS, it creates an opportunity to use the same routines and data sets we’ve created in a very powerful way,” said Hux.

The data is being used for a variety of purposes, such as demonstrating equity issues, so that provincial funders can focus on increasing access to high-tech services in underprivileged areas. It can be used to evaluate quality of care, so that hospitals institute quality improvement initiatives. Or, it can be used to map chronic disease epidemics, such as diabetes, at a regional level and include factors such as socioeconomic status.

For ICES, the return on investment isn’t about dollars and cents – it’s about delivering better healthcare to Ontarians. Using statistical analysis, for example, they’ve been able to demonstrate the danger of certain prescription drugs and reported the findings to Health Canada. In some cases, certain drugs have been removed from shelves. “Because we have records on a population of more than 12 million, we can use SAS to find the needle in the haystack, to find rare occurrences and link the causal pathway,” said Hux.

The data is also being used to map disease. The provincial prevalence of diabetes, for example, is approaching 10 percent. Using SAS, ICES can map diabetes rates at a granular level, down to the postal code. It found that in outlying parts of Toronto, high immigration and low income equals high diabetes rates. Yet in the downtown core there was high immigration and low income, but surprisingly low diabetes rates; ICES was able to relate this to environmental factors such as access to public transit, walking trails and high-quality food sources.

“The ability to draw relationships at neighborhood levels using records of millions of people across a variety of databases allowed us to make inferences and demonstrate previously unappreciated associations that are of great interest to city planners,” said Hux.

With SAS, ICES has experienced increased productivity in its analytics department. “We’ve doubled our output in terms of the number of reports we produce for peer review publications, and the Ministry of Health and Long-Term Care, with only a 10 percent increase in our frontline analytic staff,” said Hux. “SAS is helping us add value to the delivery of healthcare in Ontario.”

Copyright © SAS Institute Inc. All Rights Reserved.