Vaccination coverage survey versus administrative data in the assessment of mass yellow fever immunization in internally displaced persons--Liberia, 2004.

Yellow fever (YF) is a mosquito-borne vaccine-preventable disease with high mortality. In West Africa, low population immunity increases the risk of epidemic transmission. A cluster survey was conducted to determine the effectiveness of a mass immunization campaign using 17D YF vaccine in internally displaced person (IDP) camps following a reported outbreak of YF in Liberia in February 2004. Administrative data of vaccination coverage were reviewed. A cluster sample size was determined among 17,384 shelters using an 80% vaccination coverage threshold. A questionnaire eliciting demographic information, household size, and vaccination status was distributed to randomly selected IDPs. Data were analyzed to compare vaccination coverage rates of administrative versus survey data. Among 87,000 persons estimated living in IDP camps, administrative data recorded 49,395 (57%) YF vaccinated persons. A total of 237 IDPs were surveyed. Of survey respondents, 215 (91.9%, 95% CI 88.4-95.4) reported being vaccinated during the campaign and 196 (83.5%, 95% CI 78.6-88.5) possessed a valid campaign vaccination card. The median number of IDPs living in a shelter was 4 (range, 1-8) and 69,536 persons overall were estimated to be living in IDP camps. Coverage rates from a rapid survey exceeded 90% by self-report and 80% by evidence of a vaccination card, indicating that the YF immunization campaign was effective. Survey results suggested that administrative data overestimated the camp population by at least 20%. An emergency, mop-up vaccination campaign was avoided. Coverage surveys can be vital in the evaluation of emergency vaccination campaigns by influencing both imminent and future immunization strategies.