Dispelling Myths about Data Analytics

Health care analytics initiatives often rely on claims data, some clinical data and other traditional data sources. But according to Jason Burke, managing director and chief strategist at the Center for Health Analytics and Insights at business software/analytics vendor SAS, don’t look past non-traditional sources.


Health care analytics initiatives often rely on claims data, some clinical data and other traditional data sources. But according to Jason Burke, managing director and chief strategist at the Center for Health Analytics and Insights at business software/analytics vendor SAS, don’t look past non-traditional sources.

The National Collaborative for Bio-Preparedness, a public-private partnership in North Carolina, discovered it is possible to predict certain events, such as propagation of disease, with non-traditional data. A Norovirus outbreak, for instance, could have been predicted six to eight weeks ahead of time by analyzing more obscure sources as ambulance call data and, remarkably, people reporting their gastro-intestinal problems on Twitter, Burke said.

Speaking at Health Data Management’s Healthcare Analytics Symposium in Chicago, Burke dispelled another myth: Big Data. “It doesn’t actually matter how much data we have,” he contended. “Amassing greater and greater amounts of data will do one thing--drive up your data management costs.”

Some clinical data is proving to be useful for analytics, such as CPOE, lab results and medical device data, but in general the utility of using electronic health records data is challenging, Burke said. “Technology can change unstructured data to structured data. But what I don’t have is a means of looking across the continuum of care. This is not a, ‘We need a data warehouse’ thing. It’s more about getting into the semantics of data.”

More for you

Loading data for hdm_tax_topic #better-outcomes...