Racial/ethnic disparities in vaccination coverage by 19 months of age: An evaluation of the impact of missing data resulting from record scattering

Abstract

We describe how trends in the vaccination coverage at 19 months of age vary by race/ethnicity; explore the extent to which data required to evaluate a child's up-to-date vaccination status is missing as a result of the scattering of vaccination records among many vaccination providers; evaluate how the prevalence of that missing data varies by race/ethnicity; and evaluate the impact that the missing data has on estimated race/ethnic disparities in vaccination coverage. We analyzed data from 255,043 children sampled between 1995 and 2006 by the National Immunization Survey (NIS). Among children who had 2+ vaccination providers reporting, estimated vaccination coverage was significantly lower by approximately 15 per cent among children who did not have all of their providers reporting to the NIS compared with children who had all of their vaccination providers reporting to the NIS. By comparing coverage estimates that were adjusted for missing data to unadjusted estimates, we found that unadjusted estimates consistently underestimated vaccination coverage by as much as 4.9 per cent for Asians, 4.8 per cent for Hispanics, 4.1 per cent for American Indian/Alaska Natives, 3.3 per cent for non-Hispanic blacks, and 2.8 per cent for non-Hispanic white children. Estimates of disparities in estimated vaccination coverage did not depend on whether coverage estimates were adjusted for missing data. Hispanic and non-Hispanic black children had estimated coverage rates that were significantly less than that of non-Hispanic white children, with median disparities of 4 and 9 per cent, respectively. Regardless of whether estimates are adjusted, data from the NIS show that disparities in vaccination coverage that existed in the early 1990s persist.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 23 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.