Effective reproductive number for measles in the United States (2002-2011)

Measles has been declared eliminated in the United States, although sporadic introductions may occasionally result in short chains of transmission. Thus, there is concern that vaccination refusal may increase the possibility of epidemic transmission. In this study, we used the chain-size distribution data for measles in the United States to estimate its transmissibility and assess the risk for epidemic spread. Transmissibility may be summarized by the effective reproduction number, R, which is the average number of new cases caused by each case. Prior analysis of 2002-2011 chain-size data based on homogeneous transmission produced an estimate for R of 0.70 (95% CI: 0.61–0.79). Application of a new method, based on a negative binomial offspring distribution, allows a flexible amount of transmission heterogeneity. The new method yields an estimate for R of 0.52 (95%: CI: 0.44-0.62) and improved agreement with the data. These new results are robust to the reality of imperfect case detection. The implications of our results are three-fold. First, we confirm the CDC's finding that endemic transmission on a national scale is not possible with the current level of vaccine coverage. Second, our finding that R may be closer to zero than prior estimates has implications for the amount of vaccine refusal that is tolerable and the relative importance of geographic importation versus local transmission. Third, our results indicate some cases cause a disproportionate amount of transmission. This heterogeneity may result from virological factors (e.g., a particular strain of virus) or demographic factors (e.g. clustering of individuals in whom vaccination coverage is low).