CMS data may predict flu outbreaks

CMS claims data might be useful to determine influenza activity in particular areas before the disease becomes widespread, according to findings presented at the AACC Annual Meeting & Clinical Lab Expo.

The findings could spark interest in improved or new surveillance that could assist health agencies in their responses in influenza emergencies, according to a press release.

Earlier this year, a report issued by Trust for America’s Health suggested that if a large health emergency occurred in the United States without warning, much of the country would be ill-equipped to sufficiently combat it.

In the present study, Lin Fan, PhD of the CDC division of laboratory systems, and colleagues looked at CMS Medicare outpatient claims data with positive test volumes using rapid influenza tests from the National Respiratory and Enteric Virus Surveillance System (NREVSS) for one state in each of HHS’s 10 regions. Researchers then created spline models and employed them to match the time series of weekly volumes of influenza, influenza rapid tests and NREVSS test positives for the 2007 to 2012 influenza seasons.

According to Fan and colleagues, preliminary data showed 67 influenza episodes using NREVSS data. The predictive models using influenza tests caught 64 of the 67 NREVSS confirmed episodes and the model using influenza diagnoses caught 60 of them.

Researchers also discovered that the CMS claims data determined increases in influenza activity on average 3 weeks sooner than data from the NREVSS.

“We hope that this approach can be tried by other health care systems with large administrative data sets because it doesn't require knowledge of the test results, allowing the use of various indicators for disease activity,” Rex Astles, PhD, of the CDC division of laboratory systems, said in the press release. “This could be especially helpful for other diseases that have less effective surveillance systems.”

Astles warned that the results did not use real-time Medicare data. He also suggested that their findings could not substitute for traditional influenza surveillance and could be used to expand on the more established methods. - by Janel Miller

Reference:

Fan L, et al. Evaluation of the utility of CMS claim data for early detection of increasing influenza activity. Presented at: AACC Annual Scientific Meeting & Clinical Lab Expo; July 30 to Aug. 3, 2017; San Diego.

Disclosure: Healio Family Medicine was unable to determine researchers’ relevant financial disclosures prior to publication.