Tuesday, May 11, 2004 - 2:15 PM
5038

Recent Enhancements in the Methodology of the National Immunization Survey

Philip Smith, National Immunization Program, Centers for Disease Control and Prevention, Immunization Services Division, 1600 Clifton Road, NE, Atlanta, GA, USA, David C. Hoaglin, Abt Associates Inc, 55 Wheeler Street, Cambridge, MA, USA, Michael P. Battaglia, Abt Associates, 55 Wheeler Street, Cambridge, MA, USA, Meena Khare, Office of Research and Methodology, NCHS/CDC, 6525 Belcrest Road, # 915, Hyattsville, MD, USA, and Lawrence Barker, Immunization Services Division, Centers for Disease Control, Assessment Branch, 1600 Clifton Road, MS-E62, Atlanta, GA, USA.


BACKGROUND:
The National Immunization Survey (NIS) monitors vaccination coverage among children 19 to 35 months of age residing in 78 geographic areas that comprise the 50 states and 28 urban areas in the U.S.

OBJECTIVE:
To provide an overview the sample design, the recent enhancements to the estimation methodology, and the NIS data files that are available for public use.

METHOD:
The NIS uses a list-assisted random-digit-dialing (RDD) survey to identify households with children aged 19-35 months. An interview collects data on the child, mother, and household and requests consent to contact eligible children’s vaccination providers. With consent, vaccination providers are mailed a questionnaire to obtain the vaccination history from the child’s medical records. Recent changes in the estimation methodology account for children who have had no vaccinations and better account for households without a telephone. These enhancements are expected to improve the accuracy of vaccination coverage estimates.

RESULT:
In 2002, 0.3% of the children were entirely unvaccinated, among the children for whom completed NIS RDD interviews were obtained. Together, the new nontelephone adjustment and the refinement for unvaccinated children had very little effect on state-level vaccination coverage rates in 2002. Coverage estimates that accounted for these refinements were usually within 1.5 percentage points of the estimates that did not account for this refinement.

CONCLUSION:
Recent enhancements to the NIS methodology allow more-accurate inferences to be made about vaccination coverage levels for the total population of children aged 19-35 months, as well as for many subgroups defined by income, race, Hispanic origin, education of mother, household size, and mobility status. By using the same data collection methodology and survey instruments in all states, the NIS produces vaccination coverage levels that are comparable among states and over time.

LEARNING OBJECTIVES:
Increase understanding of the NIS and recent enhancements to the NIS estimation methodology.

See more of Current and Future Direction of the National Immunization Survey
See more of The 38th National Immunization Conference