Data on DrugsBy Tony Kontzer | Posted 2009-09-29 Email Print
Re-Thinking HR: What Every CIO Needs to Know About Tomorrow's Workforce
Companies in the life sciences industry are adapting key technologies to fit their very specific business process environments.
Data on Drugs
Once a drug has hit the market, the pharmaceutical company that’s producing it needs data. Lots of it. This “postmarket surveillance,” as it’s known, includes data on how and when drugs are being prescribed, what diagnoses are being matched to those prescriptions and what the lab results show after the drug has taken effect. This information enables the company to determine the impact its drugs are having on the patients taking them.
The informatics division of Premier, a Charlotte, N.C.-based research firm owned by some 200 hospitals and health systems across the United States, provides just such data to a host of pharmaceutical companies through its ClinicalAdvisor product, which provides users with a Web-based interface that allows them to slice and dice the information being collected and disseminated by Premier’s research team.
The only problem was, as the quantity of data exposed via ClinicalAdvisor grew, it became increasingly challenging to organize and present it in the way pharmaceutical companies needed. Researchers at Premier had to spend more time entering, analyzing and manipulating the data, and that meant slower responses to customer queries.
“We found that people want to [get and use] the data the way they want to,” says Chris Stewart, senior architect for the informatics division. “We couldn’t structure the data in a way that would [provide] it in all the ways our customers wanted to get it.”
Whatever Premier initially did to address the situation didn’t seem to help. Buying faster servers, deploying faster storage systems, hiring more database administrators—nothing worked, because Premier’s researchers and their pharmaceutical customers aren’t always sure exactly what they’re looking for, so their queries can be worded in unusual ways.
The answer, Premier found, was to deploy a more effective data warehouse and layer a more powerful business intelligence engine over it. The company chose a data warehousing appliance from Netezza that was designed for rapid analysis of large data volumes, and it coupled that with a front-end reporting tool from MicroStrategy.
The result, Stewart says, “was like night and day.” But even that wasn’t enough. Because the amount of data continues to grow exponentially each year, Premier has had to upgrade its Netezza appliance at every opportunity, with the most recent upgrade implemented in August.
The research team has seen a surge in productivity, and customers are getting better results, no matter how they word their queries. The company hasn’t measured the impact lately, but Stewart says even the first iteration of the Netezza appliance resulted in slashing the time needed to answer several common queries from as long as five hours to three minutes or less.
“We’re letting the data drive our analyses, instead of needing to [know beforehand which] data we need,” says Stewart. “It’s really helped our analysts get the data back to customers faster, with no indexing or tuning of queries.”