Development of vaccine risk communication messages using risk comparisons and mathematical modeling.

It is important to systematically assess the vaccine information needs of parents in order to maintain or improve childhood immunization coverage. Our objectives were to obtain suggestions for the optimal presentation of vaccine-related information and to determine if an educational intervention affected mothers' vaccine safety attitudes. Focus groups were used to develop messages that then were tested through a randomized, pre- and post-test mail survey of non-Hispanic White mothers who reported vaccine safety concerns (n = 927). Focus groups were analyzed using text analysis software. Increases in attitude scores between the pre- and post-test surveys were calculated, and logistic regression was used to compare intervention groups with a control group. Of survey participants who recalled the test messages, 50% (85/171) who received a "consequences of reduced coverage" message reported an improved opinion of vaccines. A greater proportion of participants receiving one or more intervention messages reported an improved attitude score from pre-to post-test compared with the control group for four of the five variables measured; however, differences were small and none were statistically significant. A mixed method approach was used to develop and test vaccine messages. The message describing potential consequences of reduced vaccination coverage had the greatest impact on improving concerned mothers' opinions of childhood vaccines.