30367 Comparisons of Trends In National Estimates of Number of Influenza Vaccinations In Physician Offices From Medical Insurance Claims Data and Survey Data

Monday, March 26, 2012
Poster Hall

Background: CDC monitors influenza vaccinations in-season using telephone surveys. The National Flu Survey (NFS) is administered twice yearly providing child/adult/combined population data. The Behavioral Risk Factor Surveillance System (BRFSS) is state-based and provides monthly adult data. The National Immunization Survey (NIS) provides weekly child data. From 2009-10 through 2011-12 seasons, CDC contracted with IMS to provide count estimates of vaccinations in private physician offices from medical claims.

Objectives: 1) Compare national coverage trends from NFS to claims as of November 2010 to same in 2011; and 2) Compare national/state monthly vaccination trends within each season from claims to NIS/BRFSS.

Methods: Medical claims from >60% of all private providers were used to project national/state levels weekly. To compare coverage trends 2010 to 2011 between claims and NFS estimates, Census denominators were used to convert claims counts by first week of November to cumulative coverage.  To compare trends between medical claims and NIS/BRFSS estimates, population denominators from U.S. census were used to convert monthly coverage to numbers of persons vaccinated.

Results: Coverage estimated in November 2011 from medical claims increased  0.8 points from 2010. NFS reported a 1.1 point increase over the same interval. Adult coverage increased in 2011 in both (claims +0.3; NFS +1.1) as did pediatric coverage (claims +2.7; NFS +1.1).[1] Estimates from claims were similar than those by survey and will be discussed in the poster. 

Conclusions: We found that although claims and NFS methodologies differ, their national trends 2010-2012 were generally consistent. Further analysis is required to determine if these sources provide similar trends by month nationally and by state, if so, claims could be used to estimate early season trends and patterns of vaccination uptake before survey estimates become available.


[1]Additional comparisons of medical claims data and NIS/BRFSS by state will be included in the final presentation.