Background. Measles affected entire birth cohorts in the prevaccine era but was declared eliminated in the United States in 2000 because of a successful measles vaccination program.
Methods. We reviewed US surveillance data on confirmed measles cases reported to the Centers for Disease Control and Prevention and data on national measles-mumps-rubella (MMR) vaccination coverage during postelimination years 2001–2008.
Results. During 2001–2008, a total of 557 confirmed cases of measles (annual median no. of cases, 56) and 38 outbreaks (annual median no. of outbreaks, 4) were reported in the United States; 232 (42%) of the cases were imported from 44 countries, including European countries. Among case-patients who were US residents, the highest incidences of measles were among infants 6–11 months of age and children 12–15 months of age (3.5 and 2.6 cases/1 million person-years, respectively). From 2001 through 2008, national 1-dose MMR vaccine coverage among children 19–35 months of age ranged from 91% to 93%. From 2001 through 2008, a total of 285 US-resident case-patients (65%) were considered to have preventable measles (ie, the patients were eligible for vaccination but unvaccinated). During 2004–2008, a total of 68% of vaccine-eligible US-resident case-patients claimed exemptions for personal beliefs.
Conclusions. The United States maintained measles elimination from 2001 through 2008 because of sustained high vaccination coverage. Challenges to maintaining elimination include large outbreaks of measles in highly traveled developed countries, frequent international travel, and clusters of US residents who remain unvaccinated because of personal belief exemptions.
Measles is a highly infectious, acute viral disease that causes rash, respiratory symptoms, and fever. Severe complications, which may result in death, include pneumonia and encephalitis. In the decade before the national measles vaccine program was implemented in 1963, it was estimated that 3–4 million people in the United States acquired measles each year . Of the ∼500,000 measles cases reported annually, 500 resulted in death, 48,000 resulted in hospitalization, and 1000 resulted in permanent brain damage due to measles encephalitis .
Achieving a high level of population immunity is the best way to prevent measles. Accordingly, 2 doses of measles-mumps-rubella (MMR) vaccine are recommended for all US children . The first dose should be administered at 12–15 months of age and the second dose at 4–6 years of age. Laws in every state require age-appropriate vaccination of children enrolled in child care facilities and documentation of evidence of measles immunity at the time of entry into kindergarten or first grade . As the vaccinated cohorts age, all children in kindergarten through grade 12 should be covered by the requirements . For adults without evidence of measles immunity, one dose of MMR vaccine is recommended. Two doses are recommended if the adult is in a high-risk group (ie, healthcare workers, international travelers, or students at post-high school educational institutions) .
Because of the success of the measles vaccine program in achieving and maintaining high 1- and 2-dose MMR vaccine coverage in preschool and school-aged children and improved control of measles throughout Central and South America, measles was declared eliminated from the United States in 2000  and from the World Health Organization (WHO) Region of the Americas in 2002 . Elimination is defined as the absence of transmission of endemic disease (ie, no epidemiological or virological evidence that measles virus transmission is continuously occurring in a defined geographical area for ⩾12 months). However, in 2008, it was estimated that there were 20 million cases of measles worldwide and 164,000 related deaths . Importation of measles virus from abroad continues to test the status of elimination in the United States. In this report, we summarize the epidemiology of measles in the United States during measles postelimination years 2001–2008.
Reporting of measles cases to public health authorities by healthcare providers and clinical laboratories is legally mandated in all states. Measles cases are identified and classified using standard case definitions and case classifications . A confirmed case of measles is either laboratory confirmed or epidemiologically linked to a patient with laboratory-confirmed measles infection . State health departments electronically transmit data on confirmed cases of measles to the Centers for Disease Control and Prevention (CDC) via the National Notifiable Diseases Surveillance System. The CDC performs molecular typing of measles viruses according to WHO-recommended protocols [8–10]. Molecular epidemiologic data obtained from viral isolates are used to confirm the links to outbreaks occurring in other countries. Viral isolates that are genotyped are categorized by the WHO region (ie, African Region, Region of the Americas, South-East Asia Region, European Region, Eastern Mediterranean Region, or Western Pacific Region) from which the virus was imported.
In the United States, an outbreak of measles is defined as a chain of transmission with ⩾3 confirmed cases. Cases are classified either as internationally imported (ie, measles cases in which exposure to measles virus occurred outside the United States 7–21 days before the onset of rash and in which rash developed within 21 days of entrance into the United States, with no known exposure to measles occurring in the United States during that time) or acquired in the United States (USacquired cases; ie, case-patients either had not been outside the United States during the 21 days before the onset of rash or were known to have been exposed to measles within the United States). US-acquired cases were subclassified into 4 mutually exclusive groups: (1) import-linked cases (ie, any case in a chain of transmission that was epidemiologically linked to an internationally imported case), (2) imported-virus cases (ie, cases in which an epidemiologic link to an internationally imported case was not identified but viral genetic evidence indicated an imported measles genotype within the chain of transmission), (3) endemic cases (ie, cases in which transmission of measles virus was continuous for ⩾12 months within the United States), and (4) unknown source cases (ie, cases in which an epidemiological or virological link to importation or to endemic transmission within the United States could not be established after a thorough investigation) [7, 11].
We analyzed all cases of measles reported in the United States during 2001–2008. However, we differentiated between US case-patients and “foreign-visitor” case-patients (ie, foreign tourists, international students, new international adoptees, recent immigrants, refugees, and cruise ship employees) when we calculated incidence, so we could use 2008 US Census data as the denominator , as well as when we reported vaccination status, because the US vaccination recommendations apply only to its US residents. This distinction allowed the differentiation of cases that would have been preventable had the recommended vaccination policy been implemented. Non-preventable cases were defined as measles cases that occurred among US residents who either (1) had received ⩾1 dose of measles-containing vaccine, (2) were vaccinated as recommended if traveling internationally, (3) were not vaccinated but had other evidence of immunity (ie, were born before 1957 and therefore were presumed to be immune from natural disease in childhood, had laboratory evidence of immunity, or had documentation of physician-diagnosed disease), or (4) belonged to a category (such as infants <1 year of age) for whom vaccination is not routinely recommended. “Personal belief exemptors” were defined as persons who were vaccine eligible, according to recommendations of the Advisory Committee on Immunization Practices (ACIP)  or the WHO , but remained unvaccinated because of personal or parental beliefs.
We examined the rates of national 1-dose measles vaccination coverage among children 19–35 months of age from 2001 through 2008, as well as those of 2-dose coverage among adolescents from 2006 through 2008 (the only years for which data were available), using data from the National Immunization Survey, which supplies provider-verified, population-based rates of immunization with 95% confidence intervals.
In the United States, from 2001 through 2008, a total of 557 confirmed measles cases were reported from 37 states and the District of Columbia (annual median no. of cases reported, 56 [range, 37 cases in 2004 to 140 cases in 2008]), representing an annual incidence of <1 case per million population (Figure 1).
Of the 557 reported case-patients, 292 (52%) were female. The median age of the case-patients was 16 years (range, 2 weeks to 89 years of age). Ninety case-patients (16%) were <1 year of age. Of the 88 case-patients (16%) who were 1–4 years of age, 45 (51%) were 12–15 months of age. A total of 41 case-patients (7%) were 5–9 years of age, 115 (21%) were 10–19 years of age, 156 (28%) were 20–39 years of age, and 67 (12%) were ⩾40 years of age. The age groups most affected varied by year, depending on the setting of the outbreak. In 2002, the majority of cases occurred in infants <1 year of age, because of an outbreak in a child care center, compared with 2006, when a large office building was the epicenter of an outbreak and persons 20–39 years of age therefore comprised the majority of cases (Figure 2).
Of the 557 reported case-patients, 126 (23%) were hospitalized (annual median no. of hospitalized case-patients, 16 [range, 5–29 case-patients]). Of these 126 case-patients, at least 5 were admitted to an intensive care unit. Two deaths were reported, both of which occurred in 2003. One death was attributed to measles encephalitis in a 1-dose-vaccinated 13-yearold boy with chronic granulomatous disease who had received a bone marrow transplant 3 months earlier . The other death involved a 75-year-old international traveler with an unknown vaccination status and an unknown history of measles disease who developed measles pneumonitis and encephalopathy .
US residents comprised 438 case-patients (79%), and foreign visitors comprised 119 case-patients (21%) (Table 1). Among US-resident case-patients from 2001 through 2008, infants 6– 11 months of age had the highest incidence of measles of any age group (59 cases, or 3.5 cases/1 million person-years), and children 12–15 months of age had the second highest incidence (30 cases, or 2.6 cases/1 million person-years).
During all postelimination years, 287 US-resident case-patients (66%) and 77 foreign-visitor case-patients (65%) were unvaccinated, 80 US-resident case-patients (18%) and 12 foreign- visitor case-patients (10%) were vaccinated, and 71 US-resident case-patients (16%) and 30 foreign-visitor case-patients (25%) had an unknown vaccination status; most of the patients with unknown vaccination status were adults. Of the 317 index and coindex case-patients, 190 (60%) were unvaccinated, 57 (18%) were vaccinated, and 70 (22%) had an unknown vaccination status. Of the 240 nonindex (ie, secondary) case-patients, 176 (73%) were unvaccinated, 32 (13%) were vaccinated, and 32 (13%) had an unknown vaccination status. Of the 308 patients with measles related to outbreaks, 229 (74%) were unvaccinated, 43 (14%) were vaccinated, and 36 (12%) had an unknown vaccination status, whereas of the 249 patients with measles not related to outbreaks, 137 (55%) were unvaccinated, 46 (18%) were vaccinated, and 66 (27%) had an unknown vaccination status. On an annual basis, the proportion of case-patients who were unvaccinated or who had an unknown vaccination status ranged from 73% in 2001 and 2006 to 95% in 2008.
National rates of 1-dose vaccination coverage among children 19–35 months of age ranged from 91% to 93% from 2001 through 2008 , and rates of 2-dose coverage among adolescents ranged from 87% to 89% from 2006 through 2008 [14, 15].
Of the 438 measles cases that occurred among US residents, 285 (65%) were considered preventable (Table 2). More than two-thirds (196 [69%]) of the preventable cases were acquired in the United States. Of the 119 US-resident case-patients who traveled internationally, 94 (79%) were unvaccinated or had an unknown vaccination status; of these case-patients, 89 (95%) were considered to have preventable measles, including 20 (21%) who were 6–11 months of age (Table 2). During 2004– 2008, a total of 110 (68%) of 162 vaccine-eligible US-resident case-patients were known to be unvaccinated because they or their parents self-declared a personal belief exemption. (Data on reasons for remaining unvaccinated were not systematically collected until 2004.)
During 2001–2008, a total of 232 cases (42%) were imported from 44 countries (annual median no. of cases imported, 26 [range, 18 cases in 2002 to 54 cases in 2001]). Of the 123 imported cases from 2001 through 2004, the majority (68 [55%]) were fromtheWHOWestern Pacific Region (Figure 3), including 28 cases from China, 23 from Japan, and 10 from the Philippines. Whereas, of the 109 imported cases from 2005 through 2008, the WHO European Region contributed the largest number (42 cases [39%]), including 7 cases each from Italy and the United Kingdom and 6 cases imported from Ukraine. However, India was the country from which the largest number of imported cases came from during 2005–2008, with 20 cases. A median of 29 US-acquired cases were reported annually (range, 10 cases in 2004 to 115 cases in 2008). The 325 US-acquired cases reported during 2001–2008 were classified as follows: 167 (51%) were import-linked cases, 94 (29%) were imported-virus cases, 0 were endemic cases, and 64 (20%) cases were of an unknown source. The transmission setting was known for 235 (72%) of the 325 US-acquired cases. Transmission occurred in the home for 71 cases (30%), in the community for 35 (15%), at church for 26 (11%), in a healthcare facility for 23 (10%), at school/college/boarding school for 21 (9%), at a child care center for 14 (6%), at work for 14 (6%), at home school for 10 (4%), and at other locations for 21 cases (9%). The measles genotypes identified during 2001–2008 were D3-D9, H1, H2, and B3. These genotypes are found in various Asian, African, and European countries.
During 2001–2008, there were 38 outbreaks (annual median no. of outbreaks, 4 [range, 2–10 outbreaks]). Of the 557 case-patients reported from 2001 through 2008, a total of 308 (55%) were outbreak-related cases. The average outbreak size was 9 cases (median no. of outbreak cases, 5 [range, 3–34 cases]). Outbreak duration (ie, the time between the onset of rash in the first case and that in the last case) ranged from 3 to 79 days (median outbreak duration, 27 days). Of 38 outbreaks occurring from 2001 through 2008, a total of 6 (16%) lasted 1 incubation period (ie, ⩽12 days), 11 (29%) lasted 2 incubation periods (ie, 13–24 days), 12 (32%) lasted 3 incubation periods, and 6 (16%) lasted 4 incubation periods; 1 outbreak each lasted 5 (3%), 6 (3%), and 7 (3%) incubation periods, respectively. The largest outbreaks are summarized in Table 3, with the largest 3 outbreaks occurring primarily among personal belief exemptors. The index case-patient in at least 9 (41%) of the 22 outbreaks occurring from 2004 through 2008 was a personal belief exemptor.
Measles surveillance data from 2001 through 2008 show record low numbers of reported measles cases and small, short-lived outbreaks, confirming that measles elimination in the United States has been maintained. This achievement is the result of sustained high vaccine coverage among American children [14, 15]. Two doses of measles vaccine are highly effective in preventing measles; the vast majority of case patients were unvaccinated. In an era when 20 million cases of measles occur globally per year, measles virus will continue to be imported from abroad until measles is eradicated globally . These importations will likely continue to cause outbreaks in communities that have sizeable clusters of unvaccinated persons, posing a continued threat to the status of measles elimination in the United States. This report also demonstrates the severity of measles; 2 case-patients died, and 23% were hospitalized, some with severe complications.
Measles importations reflect both the incidence of measles in countries around the world and travel patterns. From 2006 through 2008, the United States experienced the arrival of an average of 10.9 million nonresidents from Western Europe annually [16–18], and 12.9 million US citizens traveled to European countries annually [19–21]; however, it experienced a much smaller number of travelers to and from Africa [16–21] and Asia [16–21], where the incidence of measles is higher . Since 2005, a majority of measles importations into the United States came from the WHO European Region, which has reported thousands of cases [22, 23], primarily among unvaccinated or partially vaccinated children . During the past decade in the United Kingdom, decreasing rates of 1-dose MMR vaccination coverage (range, 91% coverage during 1997–1998 to 80% coverage during 2003–2004)  among children 2 years of age and 2-dose coverage among children 5 years of age (range, 73%–76%)  provided inadequate population immunity to sustain measles elimination in the United Kingdom. By 2008, endemic measles transmission in the United Kingdom was reestablished.
Because of the volume of international travel and the high incidence of measles in some countries, importation of measles cases is expected to continue in the United States . Molecular epidemiology is an important tool for confirming the source of these importations, because measles genotypes are geographically distributed in regions that have not yet eliminated measles . For example, genotype D5 was circulating in large European outbreaks in 2008. Viruses with identical sequences were detected in the 2008 California and Arizona outbreaks, both of which had sources imported from Europe .
Measles outbreaks occurring during 2001–2008 were smaller than those occurring during the preelimination era. Before measles elimination occurring during 1993–2000, a total of 110 outbreaks were reported, with an average outbreak size of 16 cases , compared with the 38 outbreaks reported from 2001 through 2008, for which the average size was 9 cases. Outbreak duration during 1993–2000 ranged from 3 to 131 days (median, 29 days); in the postelimination era, outbreaks had a shorter range (from 3 to 79 days; median, 27 days). Whereas 93% of outbreaks (35 of 38) occurring in 2001–2008 lasted ⩽4 incubation periods (ie, ⩽48 days), only 73% (80 of 110) outbreaks occurring in 1993–2000 lasted ⩽4 incubation periods.
Mathematical models show that maintaining measles elimination requires that the proportion of susceptible individuals, which is estimated using rates of MMR vaccine coverage in the population and rates of effectiveness of the vaccine, is less than the epidemic threshold . The limited size and duration of recent measles outbreaks in the United States can be attributed to several factors. First, levels of national measles vaccination coverage among preschool-aged and school-aged children are high [14, 15]. Second, MMR vaccine is highly effective (ie, 2 MMR vaccine doses administered after 12 months of age are 95%–100% effective in preventing measles [31–33]). Finally, there is an aggressive and effective public health response to reported measles cases in the United States that helps limit further disease transmission.
Despite high overall rates of measles vaccination at the national and state levels [14, 15], there are communities and counties where vaccine exemption rates are several times higher than state averages [34, 35]. Clustering of susceptible persons in such communities may result in population immunity below the herd immunity threshold of 93%–95%  and an increased risk of outbreaks [37, 38]. In the measles postelimination era, a majority of the measles outbreaks that occurred in the United States were among personal belief exemptors [28, 37, 39–41]. Salmon et al  found that the most common reason that parents claimed vaccine exemptions was fear that the vaccine might cause harm. As the incidence of a vaccine-preventable disease like measles decreases, the public perception tends to shift to a belief that the severity of the disease and susceptibility to the virus have also decreased . Concurrently, over the past decade, public concern about real or perceived adverse events associated with vaccines has increased [42, 44, 45].
Enactment and enforcement of school immunization laws has been shown to be an important factor in achieving high immunization coverage . However, in many states, homeschooled children are not covered by school-entry vaccination requirements. Between 1991 and 2004, the average proportion of schoolchildren whose parents claimed exemptions from vaccination because of personal belief exemptions increased . All 50 states allow medical exemptions from vaccination before school entry, 48 states allow religious exemptions (Mississippi and West Virginia do not), and 20 states allow philosophical/personal belief exemptions for entry into primary school . States that allow personal belief exemptions have higher non-medical exemption rates than states that only offer religious exemptions. States that easily grant exemptions have higher exemption rates than states with moderate and difficult processes for granting exemptions .
Measles virus is highly infectious and can cause severe complications or death . Susceptible individuals who remain unvaccinated may be unaware of the potential risk of acquiring measles disease. Once a person becomes infected, there is the risk of the virus being spread to other susceptible persons, including those for whom vaccination is not recommended because of medical reasons or those who are too young for vaccination. From 2001 through 2008, infants 6–11 months of age had the highest age-specific incidence (3.5 cases/1 million person-years); infants in this age group are not recommended for MMR vaccination unless they are traveling internationally. Compared with vaccinated persons, individuals who are exempt from vaccination are 22–224 times more likely to become infected with measles [34, 48, 49].
Healthcare providers play an important role in influencing parents as to whether they vaccinate their children. Nearly 40% of parents who change their minds after considering delaying or refusing a vaccine for their child credit information they receive from the child's healthcare provider as the reason they decided not to delay or refuse . Healthcare providers play an important role in sustaining elimination by maintaining a high awareness of measles in international travelers and unprotected populations and by rapidly implementing isolation precautions if measles is suspected in a patient. In addition, providers can facilitate measles prevention by encouraging patients who are planning international travel to be up to date with immunizations.
The epidemiology of measles in the United States during the postelimination era highlights the importance of implementing existing policy recommendations. Of the 119 US-resident case-patients who traveled internationally and were vaccine eligible, 89 (75%) were unvaccinated. Had existing MMR vaccine recommendations been followed, 285 US cases (65%) could have been prevented. Because the second-highest age-specific incidence among US residents in the postelimination era occurred among those 12–15 months of age, providers should be encouraged to offer vaccination as soon as possible once children reach 12 months of age. In addition, vaccination recommendations for adults should be followed, because several sizeable outbreaks primarily affected adults.
Despite its status as an eliminated disease in the United States, measles importation is continuing to occur and occasionally leads to outbreaks. Because of the widespread circulation of measles virus, especially in highly traveled European countries, the ease and volume of international travel, and the geographic clustering of unvaccinated persons in the United States, the risk for imported measles virus and subsequent spread remains. To maintain elimination, it will be necessary to sustain high rates of 2-dose measles vaccination coverage at the national and local levels, understand attitudes about vaccines and parents' rationales for exemptions, work with healthcare providers to reach out to persons who have questions about vaccines, find improved methods to communicate vaccine safety information, and continue with rapid public health containment measures when importations of measles virus occur.
We thank Cedric Brown and Claudia Chesley for their assistance with this manuscript.