viernes, 31 de agosto de 2012

Announcement: Addition of Households with Only Cellular Telephone Service to the National Immunization Survey, 2011

Weekly
August 31, 2012 / 61(34);685

Before 2011, the National Immunization Survey (NIS) used a random-digit–dialed, list-assisted landline telephone sample of households to monitor national, state, and selected local area vaccination coverage among noninstitutionalized children aged 19–35 months and 13–17 years (NIS-Teen) in the United States. Since NIS was begun in 1994, landline telephone use has decreased while cellular telephone use has increased. By the second half of 2011, the proportion of children in the United States living in households with only cellular telephone service was 38.1% (1). At least one factor, poverty, has been associated both with having only cellular telephone service and lower vaccination coverage, increasing the potential for bias in landline telephone surveys because of a lack of a representative sampling frame (1–3).
Beginning in 2011, the NIS sampling frame was expanded from a single landline frame to dual landline and cellular telephone sampling frames. This change increased the representativeness of the sample characteristics but had little effect on the final 2011 NIS and NIS-Teen national estimates of vaccination coverage overall and when stratified by poverty status (4,5).
Public health surveillance systems must occasionally change methods, and telephone surveys particularly need to include households with only cellular telephone service (6). The impact of this change on the validity of NIS estimates will be monitored annually. Further information, including a description of the dual landline and cellular sampling frames, specific weighting methods, and detailed national, state, and local area tables comparing estimates from the landline and dual frames by poverty level, is available at http://www.cdc.gov/vaccines/stats-surv/nis/dual-frame-sampling-08282012.htm.