We examined rubella vaccination trends, rubella surveillance, and disease patterns for the Americas, Mexico, and the United States, to evaluate the impact of hemispheric rubella control on rubella elimination in the United States during 1997–2004. In 1997, 130,375 rubella cases were reported in the Americas, with 38,042 reported in Mexico. Over the next 7 years, a rubella control initiative resulted in the administration of ∼110 million rubella-containing vaccine doses in Latin America, with 77.7 million doses administered within Mexico. By 2004, the number of reported rubella cases had declined to 3103 in the Americas and 698 in Mexico. Concurrently, the number of rubella cases in the United States fell from 817 during 1997–1999 to <25 cases/year from 2001 onward, with loss of seasonality and geographic clustering, despite no change in vaccination rates. Implementation of rubella control strategies in the Americas, particularly in Mexico, appears to have facilitated rubella elimination in the United States.
Rubella vaccine was licensed in the United States in 1969, and the ensuing 2 decades saw a dramatic decline in disease rates , although viral transmission continued. In 1989, a goal to eliminate indigenous rubella transmission and congenital rubella syndrome (CRS) was established for the year 2000.
In pursuit of the elimination goal, childhood rubella vaccination coverage was increased to>90%  during the 1990s, and, by the end of the decade, overall population immunity levels surpassed the threshold thought to ensure interruption of viral transmission . Nevertheless, hundreds of cases of rubella continued to be reported every year in the United States. The demographics of rubella disease in the 1990s revealed a problem difficult to solve solely within the United States. For instance, in 1999, 65% of cases with known country of origin were among persons born outside the United States and, of these, 98% were from the Western Hemisphere, with 81% from Mexico . The year 2000 saw 176 confirmed cases of rubella and 9 cases of CRS, and the goal of rubella elimination was deferred to 2010.
The World Health Organization (WHO) had not included rubella vaccine in the Expanded Program on Immunization (EPI) in 1974 , and, until the late 1990s, no formal rubella vaccination program existed within the region of the Americas. However, in 1997, a regionwide control initiative was adopted for the Americas, and in 2000 the WHO recommended use of rubella vaccine in all countries with well-functioning childhood immunization programs where prevention of CRS or elimination of rubella was considered to be a public health priority and where resources could be mobilized to ensure implementation of an appropriate strategy .
Beginning in 2001, the epidemiological profile of rubella in the United States appeared to change: the number of cases markedly diminished, the seasonal peak was lost, and individual cases were isolated temporally and geographically . This occurred in the absence of any marked alteration in vaccination coverage within the United States  but in a context of an increasingly vigorous rubella control program in the Americas. This suggested that hemispheric rubella vaccination control may have played an important role in helping to reduce rubella disease to elimination levels within the United States. To address this question, we comparatively examined the history of rubella vaccination programs, vaccination trends, rubella surveillance, and rubella disease patterns for the WHO Region of the Americas, Mexico, and the United States.
We compared the rubella vaccination programs and temporal patterns of disease in the United States with those in the Western Hemisphere and Mexico, focusing on 1997–2004, the period during which the epidemiological profile of rubella disease in the United States changed to an elimination pattern.
Region of the Americas.We obtained information from the records of the Pan American Health Organization (PAHO), the regional body of the WHO responsible for health issues of the 35 nations and 9 territories in the Western Hemisphere. To examine the history of vaccination policies and programs, we reviewed minutes of conferences dealing with rubella issues, recommendations of technical advisory bodies, reports published in PAHO bulletins, and guidelines developed by working groups. To examine vaccination trends, we reviewed rubella vaccine doses administered as reported by Latin American countries (i.e., all countries in the Region of the Americas, excluding Canada, the United States, and the English-speaking Caribbean) to the Immunization Unit of the PAHO. These reports were divided into first doses administered as part of routine vaccination programs and doses administered as part of national campaigns, which may target both adults and children without regard to previous vaccination status. To examine rubella surveillance and disease trends, we reviewed the reports of rubella disease and CRS by participating nations to the PAHO.
Mexico.For the history of vaccination policies and programs and for patterns of rubella disease and congenital rubella syndrome, we obtained information from the National Center for the Health of Infants and Adolescents, Secretary of Health, Mexico, where we also obtained data concerning doses of rubella vaccine administered.
United States.By methods described in detail elsewhere, we examined reports of rubella disease  and rubella vaccine doses administered and childhood coverage . We also examined patterns of rubella disease imported into the United States from Mexico and other nations in the Americas.
Region of the Americas
Rubella vaccination programs.Before 1990, only 6 of the 44 nations/territories in the region included rubella vaccine in their routine childhood vaccination programs. In 1997, the PAHO Technical Advisory Group on Vaccine Preventable Diseases recommended the implementation of a regional initiative to enhance prevention of rubella and CRS. A strategy for accelerated rubella control and CRS prevention was developed . This strategy included the introduction of rubella-containing vaccines into the routine childhood immunization programs administered at 12 months of age and a 1-time measles-rubella (MR) vaccination campaign for adults to interrupt rubella virus circulation among men and women of childbearing age. Additional vaccination strategies initially used to achieve measles elimination included achievement and maintenance of high routine vaccination coverage (⩾95%) with measles-mumps-rubella (MMR) vaccine (i.e., ``keep up" vaccination) and “follow-up” campaigns with MR vaccine. By 2004, all but 1 nation (Haiti) in the region had incorporated rubella-containing vaccine into their national vaccination program . Of the 44 nations/territories, 24 had also conducted rubella vaccination campaigns for adults , with varying target groups based on patterns of susceptibility and fertility. Most of the countries in the English-speaking Caribbean targeted both men and women 20–40 years of age  in vaccination campaigns conducted during 1997–2001. Chile organized a vaccination campaign targeting only females 10–29 years of age in 1999 . Costa Rica targeted individuals of both sexes 15–39 years of age in 2001 . Brazil conducted a vaccination campaign in 2001–2002, targeting women 12–39 years of age; however, some states adjusted the target age group . Vaccination campaigns targeting both men and women were conducted in Honduras in 2002, in El Salvador in 2004, and in Ecuador in 2004 . In September 2003, a goal of hemispheric rubella eradication was established for 2010 [13, 14].
Indicators of rubella vaccination.The birth cohort for Latin America is estimated to be 11.45 million, and the total population is estimated to be ∼543 million . The annual number of first doses administered routinely to children increased from 5.4 million in 1997 to 9.6 million in 2004 (figure 1). These data do not include doses delivered during the same period as part of mass campaigns, which are estimated at a total of 46.16 million. Therefore, ∼110 million doses of rubella-containing vaccine were administered in Latin America during the period 1997–2004.
Rubella/CRS surveillance.Before 1992, only 4 nations/territories reported rubella cases to PAHO. In 1994, a measles surveillance system was implemented  as part of the measles eradication strategy in the Americas , and overall viral diagnostic capacity was enhanced. In 1996, rubella testing was added for all suspected measles cases with negative laboratory results for measles; additionally, guidelines for CRS surveillance were developed to detect infants <12 months of age with suspected CRS. By 1998, all 44 nations/territories were reporting rubella cases . In 1999, measles and rubella surveillance systems were fully integrated , permitting simultaneous laboratory testing for all suspected measles and rubella cases. The number of nations/territories in the Americas reporting CRS cases increased from 18 in 1998 to all 44 in 2003.
Rubella/CRS disease patterns.With increasing surveillance, the number of reported rubella cases in the Americas increased from 7640 in 1990 to a peak of 130,375 in 1997. After 1998, the number of nations/territories reporting rubella remained stable, but—with increasing implementation of rubella vaccination strategies—the number of reported rubella cases decreased to 3103 in 2004 (figure 1). The number of reported CRS cases fell from a peak of 90 in 2000 to 27 in 2004.
Rubella vaccination programs.Rubella vaccine was introduced into the childhood vaccination schedule in 1998, when measles monovalent vaccine was replaced with MMR in a 2-dose schedule directed at children 1 and 6 years of age. In 2000, to protect older age groups not reached by the routine childhood vaccination program, a strategy of MR vaccination of selected high-risk groups was initiated, including health care workers, teachers, military and police personnel, and tourist and migrant workers. In 2001, a vaccination campaign was conducted to deliver 1 dose of MR to junior and senior high school students, and, in 2002, a permanent immunization program for adults was adopted. In 2004, a mass vaccination campaign of adolescents and adults 13–39 years of age began.
Indicators of rubella vaccination.The Mexican birth cohort is ∼2.2 million, and the total population is ∼106 million. During the period 1998–2004, 77.7 million doses of rubella-containing vaccine were administered, which suggests that most of the population <40 years of age may have been vaccinated (figure 1). During the same period, the reported vaccination coverage with the 1 dose of MMR among children 1–4 years of age was>95% .
Rubella/CRS surveillance.Since 1978, clinically diagnosed rubella cases have been reported to the Single Epidemiological Surveillance System (SUIVE). With the implementation of measles elimination in 1993, laboratory-confirmed cases have been reported through the Febrile Exanthematic Disease Surveillance System (FEDSS). Since 1997, the FEDDS has also been following women infected during pregnancy, to detect potential cases of CRS.
Rubella/CRS disease patterns.From 1978 through 1999, rubella disease followed a recurrent pattern, with peaks occurring every 3–5 years. In 1997, 38,042 rubella cases were reported, and, in 1998 (the year in which the national rubella vaccination program was initiated), 51,846 cases were reported. Thereafter, the number of reported rubella cases fell steadily, reaching a low of 698 in 2004 (figure 1). For the period 1997–1999, 110 infants with congenital malformations consistent with CRS were reported, 50 cases of which were laboratory confirmed . In 2004, 5 CRS cases were reported, and 1 was laboratory confirmed .
Rubella vaccination programs.Rubella vaccine was introduced in the United States in 1969, targeting children 1 year of age to puberty . In the 1970s, outbreaks were reported among adolescents and young adults, and additional recommendations targeting susceptible postpubertal females and persons in high-risk settings were established in 1978. The resurgence of measles in 1989 prompted the implementation of a 2-dose MMR schedule in 1989, and a goal of rubella elimination by 2000 was set the same year . In the 1990s, rates of rubella disease declined, but a relative upward shift in the age distribution of rubella disease prompted renewed efforts to vaccinate adults missed by childhood vaccination programs, particularly women of childbearing age and persons born outside the United States. However, in contrast to the rest of the Americas, no mass campaigns were conducted among these groups or others, and the rubella elimination program continued to rely on ensuring high childhood coverage through routine vaccination. With evidence of continued rubella circulation in 2000, the goal of rubella elimination was reset to 2010, but the primary strategy of routine childhood vaccination remained unchanged [1, 6].
Indicators of rubella vaccination.The US birth cohort is ∼4 million, and the total population is ∼300 million . During the period 1997–2004, when the epidemiological profile of rubella in the United States changed markedly, the annual number of doses administered remained stable, ranging from 10.8 to 14.2 million (figure 1). National surveys suggested that childhood rubella vaccination coverage also remained stable at>90% during the same period .
Rubella/CRS surveillance.Rubella has been a reportable disease in the United States since 1966. By protocol, suspected cases are laboratory tested in public or commercial laboratories. As is reported in detail elsewhere, for the period 1997–2004, the surveillance system appeared to be both sensitive and specific for detection of rubella disease .
Rubella/CRS disease patterns.During 1997–1999, when the last major peak occurred, a total of 817 confirmed rubella cases were reported; during 2001–2004, the annual number of cases was never>25 and was a median of 13 (figure 1). In 2001, for the first time, the rubella seasonal peak was lost and never regained, with cases occurring sporadically in time and place. In 1998, of 291 cases with known country of origin, 231 (79%) were among persons born outside the United States, of whom 210 (91%) were from Latin American countries, with 99 (43%) from Mexico alone. During 2001–2004, of the 45 cases with known country of origin, 10 (22%) were from Latin American countries, and 4 (9%) were from Mexico. A total of 24 infants with laboratory-confirmed CRS were born in the United States during 1997–1999. Of the 23 mothers with known country of birth, 21 (91%) were born outside the United States, with 13 (62%) of these born in Mexico. Moreover, 10 (42%) of the 24 CRS cases were imported; 6 of the mothers (60%) were exposed in Mexico, and another 2 (20%) were exposed elsewhere in Latin America . After 1999, the number of reported CRS cases declined steadily, reaching a low of 0 in 2002.
In summary, our data suggest that rubella control efforts in the Americas generally and in Mexico in particular may have had a significant positive influence on the success of rubella elimination efforts in the United States. After 2 decades of intensive vaccination efforts focusing on children, the United States had achieved, by the 1990s, high coverage and immunity levels, levels consistent with elimination of endemic transmission of rubella disease [2, 3]. Yet hundreds of cases continued to be reported each year, with large outbreaks occurring periodically . In the 1990s, in contrast to previous decades, rubella occurred predominantly among adults and persons born in Latin America, especially Mexico. Then, in the late 1990s, Latin American nations and Mexico specifically began major rubella control efforts, with a marked increase in the number of rubella vaccine doses administered and a marked decrease in the number of rubella cases. Simultaneously, the epidemiological profile of rubella in the United States changed: the number of cases diminished to <25 annually, and the recurrent seasonal peak was lost. The shift to a pattern of isolated and sporadic disease suggested that endemic transmission had ended. This abrupt change in patterns of disease was not preceded by any change in vaccination strategy, by any increase in the number of rubella doses administered, or by any dramatic improvements in population-based immunity in the United States, but it was accompanied by a dramatic decrease in the number and proportion of cases in individuals born in Latin America. This suggests that the implementation of rubella control strategies in the Americas, particularly in Mexico, accelerated the decline of rubella disease to elimination levels in the United States.
As an ecological and observational examination, our study has numerous limitations, and our findings should be interpreted with caution. Simultaneity and causation are not the same. One could argue that hemispheric rubella control may have been irrelevant—decades of sustained high childhood vaccination coverage and the aging of a well-vaccinated population into adulthood may be sufficient to account for rubella elimination in the United States. Similarly, the relative suddenness in the shift of disease to an apparent elimination pattern in 2001 is difficult to explain on the basis of hemispheric disease trends. Furthermore, even in good surveillance systems, the number of reported cases generally underestimates true disease rates, particularly because half of all rubella cases are asymptomatic. Thus, the decrease in the number of US cases exceeds the reduction attributable to the decrease in documented importations from Latin America
Elimination is conventionally defined as the end of ongoing transmission of a disease within a defined area, but infectious diseases do not necessarily respect national borders. Our study suggests that, although conditions to eliminate an infectious disease can be created within a single nation, actual elimination may be difficult to achieve without the help of a regional strategy and ongoing international cooperation.
Global efforts at rubella control and elimination are still in their infancy. Although the proportion of the world's population living in countries with national rubella vaccination programs has doubled from 12% in 1996 to 25% in 2003 (figure 2), the world's most populous nations and the world's poorest nations have still not begun efforts at rubella control, much less elimination. The global burden of CRS remains high, with an estimated 100,000 infants born annually with blindness, deafness, heart defects, or other disabilities attributable to rubella virus infection . The impact of this disability burden may be highest in the least-developed nations.
Rubella shares many clinical characteristics with measles, making joint surveillance efforts synergistic. In 2000, the WHO recommended that countries undertaking measles elimination should take the opportunity to eliminate rubella, through use of MR or MMR vaccine . This approach has been undertaken by the Region of the Americas and countries with strong routine immunization programs in other regions, such as Europe and the Western Pacific . Combining rubella and measles surveillance and vaccination activities will increase their cost-effectiveness . We believe that the benefits of such a joint measles-rubella strategy have been demonstrated in the Americas .
The experience of rubella and CRS elimination in the United States demonstrates that endemic rubella transmission can be interrupted in a large geographic area with diverse populations. This experience also highlights the importance of regional control efforts and provides a well-documented example of the feasibility of rubella elimination in a country in the context of a region with established and successful elimination goals.
We acknowledge Charles LeBaron and Mary McCauley, for their assistance in editing this article.
Financial support.Centers for Disease Control and Prevention.
Supplement sponsorship.This article was published as part of a supplement entitled “The Evidence for the Elimination of Rubella and Congenital Rubella Syndrome in the United States: A Public Health Achievement,” sponsored by the Centers for Disease Control and Prevention.
Potential conflicts of interest.All authors: no conflicts.