Synthesis of Evidence to Characterize National Measles and Rubella Exposure and Immunization Histories

Abstract

Population immunity depends on the dynamic levels of immunization coverage that countries achieve over time and any transmission of viruses that occur within the population that induce immunity. In the context of developing a dynamic transmission model for measles and rubella to support analyses of future immunization policy options, we assessed the model inputs required to reproduce past behavior and to provide some confidence about model performance at the national level. We reviewed the data available from the World Health Organization (WHO) and existing measles and rubella literature for evidence of historical reported routine and supplemental immunization activities and reported cases and outbreaks. We constructed model input profiles for 180 WHO member states and three other areas to support disease transmission model development and calibration. The profiles demonstrate the significant variability in immunization strategies used historically by regions and member states and the epidemiological implications of these historical choices. The profiles provide a historical perspective on measles and rubella immunization globally at the national level, and they may help immunization program managers identify existing immunity and/or knowledge gaps.

Modeling the Transmission of Measles and Rubella to Support Global Management Policy Analyses and Eradication Investment Cases

“…Recognizing the variability in current national measles and rubella immunization programs, experience with historical immunization, population structures, and viral transmission conditions, we model each area separately and then aggregate the results to regional and global levels. The model includes 180 WHO member states and three other areas (i.e., Puerto Rico, Hong Kong, and Macao) for which we found sufficient demographic and immunization data . The 2013 population in the model represents >99.5% of the estimated global population of 7.16 billion people .…”

“…We used these data in the model to stratify the population for each area by sex and into 35 age groups: 0 to <6, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 to <18, 18 to <21, 21 to <24 months, and 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15–19, 20–24, 25–29, 30–34, 35–39, 40–44, 45–49, and 50+ years. The model tracks females and males separately because the immunity of pregnant women impacts the maternal antibodies that infants passively receive for both measles and rubella, and some countries historically used selective immunization of adolescent and/or adult women for rubella . The model adds infants into the first age group using a daily rate estimated from the number of surviving infants for each calendar year (i.e., the number of live births minus the number of infants who do not survive the first year of life).…”

“…We ignore the reality that some individuals actually receive their scheduled dose at a younger or older age than implied by the RI schedule. We used reconstructed historical RI schedules for all of the 183 areas modeled using available data from the WHO and UNICEF and information obtained from the published literature . The historical RI data should account for actual reductions in coverage that occurred due to disruptions in program performance and/or vaccine supply (e.g., due to insecurity, instability, humanitarian and natural disasters, and/or other challenges).…”

“…For countries for which RI second doses occur, but the country does not report coverage estimates to the WHO (e.g., the United States), we estimated first and second dose RI coverage based on the best available information. Given significant revisions in coverage estimates in the WHO–UNICEF estimates that occurred, we updated our coverage estimates from those previously reported . We conceive of the model as a living tool such that the model inputs can reflect updates over time and include future projections, which will allow it to support prospective analyses.…”

“…We explore the behavior of the model in the context of available epidemiological data and the results of serological surveys that provide a snapshot of the dynamic population immunity at the time of data collection. For measles, we could potentially observe most of the cases (i.e., relatively few asymptomatic infections), while for rubella relatively more asymptomatic infections occur .…”