Evolution of Surveillance of Measles, Mumps, and Rubella in England and Wales: Providing the Platform for Evidence-based Vaccination Policy

Received for publication May 20, 2002; accepted for publication September 25, 2002. Vaccination is one of the most cost-effective measures for preventing morbidity and mortality that modern medicine has to offer (1). Measles, mumps, and rubella are three viral infections causing significant morbidity for which an effective vaccine is available. Otitis media (5 percent of cases), pneumonia or bronchitis (4 percent), and neurologic complications (1 percent), including subacute sclerosing panencephalitis, are associated with cases of measles (2), while mumps is recognized as a common cause of aseptic meningitis and can cause orchitis in adult males (3). Rubella infection in pregnancy, especially during the first trimester, can cause miscarriage or congenital rubella syndrome, which is characterized by a pattern of congenital abnormalities including nerve deafness, cataracts, cardiac abnormalities, and mental retardation (4, 5). Before vaccines became available, immunity to measles, mumps, and rubella was obtained through acquisition of the wild-type virus. In 1968, a monovalent measles vaccine was introduced for infants in England and Wales (6), and it was followed in 1970 by rubella vaccine for schoolgirls and susceptible women (7). Monovalent measles vaccine was replaced by the combined measles-mumps-rubella (MMR) vaccine in 1988, with the aim of eliminating all three diseases. Monovalent rubella vaccine remained available to susceptible adult women. For the first 3 years of the program, a catch-up dose was also offered to children at age 4 years (8). In 1994, a combined measles-rubella vaccine was offered to all schoolchildren aged 5–16 years in a national campaign lasting 6 weeks (9, 10). Since 1996, a two-dose schedule of MMR has been routinely offered to all children (11). This review describes how and why these decisions on vaccine policy for measles, mumps, and rubella have been made in England and Wales over the last three decades. It demonstrates how the national surveillance system for measles, mumps, and rubella has evolved and illustrates the way in which it is used in conjunction with mathematical models to inform national policy. The situation in England and Wales is used as an example because the surveillance methods employed there can serve as a model for other countries with an interest in controlling these infections, particularly given the commitment by the World Health Organization to eliminate global morbidity and mortality resulting from measles (12). Four surveillance methods are used in England and Wales to monitor the impact of the MMR vaccination program. These methods provide data on vaccination coverage, clinical notifications, laboratory-confirmed cases, and antibody prevalence. Since the mid-1960s, data for estimating coverage in 2-year-old children in England and Wales have been collected annually by the Department of Health and the Welsh Office from health authorities. Estimates initially used the number of livebirths in each district health authority as the denominator, moving in 1988 to computerized child health registers holding vaccination details for all children resident in the district (Körner returns). These computerized records, which cover the entire population, have been used to generate more timely quarterly information on primary immunization coverage for England and Wales (and later Northern Ireland). The system was developed by the Communicable Disease Surveillance Centre and is termed the Cover of Vaccination Evaluated Rapidly (COVER) Program (13). Its purpose is twofold: first, to improve vaccination coverage through regular and accurate feedback to district immunization coordinators, thereby stimulating local interest in vaccination coverage and investigation of reasons for poor performance; and second, to enable changes in vaccination coverage to be detected quickly. The COVER Program calculates quarterly vaccine coverage data for sentinel antigens. These included initially measles vaccine and then subsequently MMR vaccine. Data are provided on children who reached their second or fifth birthday in the evaluation quarter—that is, on coverage of the first dose of measles/MMR vaccine at age 14 months and coverage of the first and second doses at age 5 years. These data are requested from, and fed back to, district immunization coordinators, as well as published quarterly in the Communicable Disease Report. Measles (since 1940), mumps (since 1988), and rubella (since 1988) are statutory notifiable infectious diseases, and clinicians are legally required to report any cases diagnosed to the appropriate officer of the local government authority. This is usually the responsibility of the consultant in communicable disease control for the area where the patient resides, who reports weekly to the Office for National Statistics. There are no formal case definitions; notifications are made on suspicion or diagnosis of clinical disease (14). When an infection is common, the positive predictive value of a clinical diagnosis is sufficiently high to base surveillance on notifications. However, as the infection becomes less common because of childhood vaccination, it is to be expected that an increasing proportion of notified cases will be due to other infections that have a similar clinical presentation. For example, cases of parvovirus B19, human herpes virus 6 (roseola infantum), human herpes virus 7, and group A streptococcus all involve symptoms of rash and fever and may be misdiagnosed as measles or rubella (15, 16). Laboratory confirmation by the detection of specific immunoglobulin M (IgM) in an appropriate clinical specimen can be used in association with notification to improve the accuracy of the data. Traditionally, the specimen of choice for such antibody testing has been serum. Saliva (oral fluid) is an alternative noninvasive type of clinical specimen that contains immunoglobulins and has considerable compliance advantages. Saliva is safe and easy to collect (17–20), which makes it attractive for surveillance. A 1991–1993 pilot study designed to validate oral fluid diagnosis of measles demonstrated the high specificity of oral fluid IgM testing and showed that less than 40 percent of clinically diagnosed cases were confirmed by laboratory testing (21, 22). This demonstrated that reliance on notification alone was not sufficient to monitor progress towards elimination of measles and that identification of the true incidence of infection could only be achieved through laboratory investigation of suspected cases. Prior to 1994, surveillance of laboratory-confirmed cases of measles was based on cases investigated by public health and National Health Service laboratories and reported to the Public Health Laboratory Service Communicable Disease Surveillance Centre. Since November 1994, the Public Health Laboratory Service has enhanced surveillance of measles, mumps, and rubella by offering to test oral fluid specimens from all notified cases for virus-specific IgM. Samples collected between 1 and 6 weeks after the onset of symptoms are sent to the Enteric, Respiratory, and Neurological Virus Laboratory at the Public Health Laboratory Service Central Public Health Laboratory, where they are tested for specific antibody (22) to confirm the diagnosis (9, 23–25). Such laboratory confirmation is currently not obligatory, but every effort is made to facilitate this service (9). Between January 1995 and December 2001, 65.3 percent of measles notifications, 60.9 percent of mumps notifications, and 54.6 percent of rubella notifications had subsequent oral fluid samples collected and sent for laboratory confirmation, with the diagnosis being confirmed in 2.4 percent, 19.6 percent, and 12.7 percent, respectively (http://www.phls.co.uk). Genotyping of measles, mumps, and rubella is an additional laboratory technique that is now being used to supplement surveillance data (26–29). Measles, mumps, and rubella virus genome can be extracted from suitable clinical specimens, including oral fluid (30), collected soon after the onset of symptoms in suspected cases. Specific regions are then amplified by reverse transcription polymerase chain reaction, and the amplicons are directly sequenced (26). In addition to assisting in laboratory confirmation, analysis of genomic variation (genotyping) allows investigation of transmission pathways and sources of infection (31). This is particularly useful for epidemiologically linking cases and investigating unusual cases, and genotyping was used to investigate and link outbreaks of measles that occurred within Rudolph Steiner communities in England in 1997 (32, 33). Genotyping is also a useful tool for obtaining information with which to document progress towards disease elimination (34). Since 1986, the prevalence of antibodies to measles, mumps, and rubella in the population has been monitored using anonymized residues of specimens submitted for microbiologic or biochemical testing to public health laboratories in England and Wales. Only age, the and the of are of samples have been collected each Samples from age years are collected as a every fifth samples are collected the age This is the of the population of England and Wales currently available. The first was in the of MMR and it using laboratory that detected specific immunoglobulin was used to while and was detected by is now used as the of choice for all three infections These data on the population are by investigation of rubella in women. of the of rubella infection in by age and has been by sentinel which one of the in England and Wales The sources of surveillance data may be with information obtained from sent to the Office for National sentinel surveillance in from the of and congenital rubella surveillance When the of measles, mumps, and rubella infection in England and Wales in conjunction with vaccination and surveillance to be These for the infection was the subsequent of and the later in vaccine policy. Before measles vaccine was available, England and Wales regular of measles, with of of notifications each In 1968, a measles vaccine was coverage was but it to a of percent in 1988 The of vaccine had an on the incidence of The resulting in notifications with the of vaccine and increasing vaccine coverage over is in In 1988, the MMR vaccine was and were made to improve coverage (8). The of high vaccine coverage and an MMR catch-up for children aged years at the the vaccine was introduced had a impact in the transmission of measles, and infection was in the However, surveillance showed that while had in the in MMR vaccine coverage, to measles children aged years had from 6 percent in to percent in in conjunction with mathematical models that by the the in schoolchildren be sufficient for an to in these age was due to a of coverage in the a proportion of primary vaccine and transmission of the virus in the This that an vaccination campaign children was required such an were to be a national measles and rubella vaccine campaign for the was The in November children aged 5–16 years were offered the combined measles-rubella vaccine. the children in the percent were during the which 6 weeks 10). monitor the impact of this campaign on measles in the age the Public Health Laboratory Service collected samples in from children aged years and for by The of antibody to be required for clinical is The proportion of susceptible to from percent in to percent in the of the measles-rubella However, antibody children aged who were not included in the were similar in and and percent had antibody of less than These in conjunction with mathematical that immunity in the population was now sufficient to However, local outbreaks were expected to to as a of from cases (9). The of the national measles-rubella campaign in the for of a second dose of MMR children to provide an for poor vaccine or who their first This to the in of a second dose of offered to all children to as a for of (11). The of enhanced surveillance using oral fluid IgM detection between January 1995 and December are in In the January 2001, percent of notified cases of measles were for with antibody being detected in percent notifications during this the true incidence of measles, only a proportion of cases were confirmed by laboratory of oral fluid The number of confirmed measles cases has remained the measles-rubella and most cases in children or However, local outbreaks of laboratory-confirmed measles occurred in 1996, and as a of from cases. This is with from the of vaccine coverage and surveillance data. The in 1997 and children in Steiner which not in measles vaccination for The from one to was confirmed by The of cases were in children years of age, and in only confirmed cases had the child been measles A similar situation occurred in in an in which vaccination coverage was The complications of measles infection pneumonia and subacute sclerosing these complications are they are A in the number of cases of subacute sclerosing has been a of the of measles vaccine the of an effective vaccination program. Prior to 1988, mumps was not a notifiable and surveillance of and data. In the of an immunization program, of mumps occurred at for mumps in England and Wales due to aseptic over the In a of samples from aged years were for The by age the acquisition of immunity through infection the high number of cases of mumps mortality and were mumps vaccination was only a cost-effective the combined MMR vaccine became available mumps was introduced in England and Wales in 1988 as a of the MMR vaccine and is offered routinely to all children aged In MMR was offered at age 4 years in a catch-up program. for mumps in England and Wales to in the to and notification data showed that the incidence of mumps reached after the of This was due to the high of vaccine in the age the of mumps virus was used in MMR vaccine. However, subsequent demonstrated that cases of aseptic meningitis were associated with of this The of cases of meningitis associated with an alternative mumps to the in to the with in MMR vaccine used in the The of mumps vaccination on was by the of collected in from aged for The of susceptible to mumps in with the susceptible the of vaccine is in The immunity of children aged years in can be to MMR In the proportion of aged years who were for mumps was in than in this was due to to infection in this age group high of MMR vaccine. the of these surveillance the proportion of children aged years with no mumps antibody was expected to at percent in 1997 A mumps was not included with measles and rubella in the measles-rubella This was because there was no suitable mumps available at the which from sufficient of a vaccine that included a mumps in addition to measles and models the for a of percent immunity within the population of England and Wales mumps were to be This was not with the vaccine coverage for MMR in England and Wales using a not susceptible adult the of complications from mumps infection is of a second dose of mumps vaccine was This was in by a second dose of MMR for children (11). the of a second dose of mumps vaccine to available after November 1994, the Public Health Laboratory Service enhanced surveillance of mumps by offering a oral fluid test for all cases of mumps reported to the Office for National Data on notified cases and laboratory-confirmed cases for January are in this percent of notified mumps cases were for with antibody being detected in percent increasing from percent in 1995 to percent in The outbreaks occurred in children from 1997 to These children were to have MMR it was introduced in 1988 and are to have mumps vaccine. This is with the based on vaccine coverage and surveillance data. The of mumps in has also been in other developed countries of the of a second dose of vaccine is provided by a analysis of MMR vaccination and of in laboratory-confirmed mumps cases with onset between 1995 and This analysis that the of who had a mumps infection during this were 1988 and had no vaccine or a dose of mumps vaccine The of with an mumps vaccination had been to the of mumps vaccine in England and Wales. These may be to have doses of a but they could have a dose Only percent of cases had a second dose of vaccine. These data confirm that a second dose of mumps vaccine is in for a vaccination policy to this disease within the Rubella immunization are designed to rubella infections and subsequent congenital rubella The to rubella vaccination on the the incidence of infection in the proportion of women susceptible and the of infection to which susceptible women are A to the proportion of women who are susceptible to rubella infection by and women for A is designed to transmission of the of susceptible women being Prior to rubella vaccine was not available in the and immunity was through When rubella vaccine was introduced in the in a was in which and women or after were given one dose of monovalent rubella vaccine. This was because of the of immunity and because measles vaccine coverage at that was percent), that a for rubella vaccination not be (7). vaccination for aged years achieved coverage of percent between 1970 and of vaccination for women to be susceptible during was not monitored but it from 5 percent to more than percent in one study The of these and the of women of The first study showed that only percent of aged years rubella a percent than that for males of the age confirmed that by only percent of women and percent of women were these to percent and percent, by However, women who remained susceptible to be at of infection during In of women as susceptible at the of that were during the of their The of infection was more than in women than in that women being by their children (7). The impact of the on the incidence of rubella infection in can be by reports on congenital rubella (since of to rubella infection (since and laboratory-confirmed rubella infection during (since showed significant the from an of congenital rubella cases, and infections during annually in to and infections during in In 1988, coverage and of coverage had and surveillance had been the rubella vaccination policy was by vaccination of children with rubella vaccine as of the MMR vaccine. This was designed to eliminate such a program, high vaccination coverage of children of in the second of with catch-up vaccination for the vaccine is is coverage there is a of increasing the age of infection and the number of congenital rubella cases, because an in the proportion of women susceptible may the in their of infection there was a in the number of cases of rubella between and A of rubella infection was in because of an that adult in the was children aged years. These children were to have MMR vaccine during the catch-up for children Between and 1994, an of associated with congenital rubella and were reported annually in the The of of children with reported congenital rubella were from countries rubella vaccination A pilot study of notified cases of rubella in England and Wales from was to investigate the accuracy of rubella surveillance based on clinical reports and to validate an oral fluid Only percent of notified cases were confirmed by the detection of IgM in serum. The and specificity of the oral fluid IgM test were to be percent and percent, the of rubella in the Public Health Laboratory Service enhanced surveillance program, which in In November 1994, combined measles-rubella vaccine was offered to all children aged 5–16 years in a national vaccination The for the campaign was to a measles but it also provided an to improve rubella control in schoolchildren by their of particularly in the percent of susceptible males aged years who had not been offered rubella vaccine the measles-rubella these were expected to have increasing of outbreaks in over the years. such outbreaks the to rubella elimination The prevalence of and after the measles-rubella campaign was using samples obtained from children aged 5–16 years in and 1995 The proportion of susceptible to from percent in to percent in to the proportion antibody from percent to percent in and from percent to percent in The in in this age group immunity from the vaccination policy that was in Data on notified cases of rubella that were reported and laboratory-confirmed between January 1995 and December are in 1996, an occurred that particularly adult males at and in the in England and Wales who had been offered vaccine and This the proportion of susceptible aged which from percent to percent after the of MMR and the that the of rubella is as vaccination policy has been to offer to women of age There have been no outbreaks of rubella in England and Wales as the measles-rubella campaign have reached Between January 1995 and December 2001, percent of with notified cases of rubella were for IgM in the oral fluid surveillance program, with antibody being detected in percent The proportion of notified cases confirmed by the detection of IgM in oral fluid from 5 percent in November and December to more than 40 percent in the first 6 months of Since November 1994, more than percent of cases confirmed by the detection of IgM in oral fluid were in males over age years with the of any rubella in England and Wales the of the measles-rubella campaign the in the incidence of rubella in the of 1996, there were cases of congenital rubella during the women in the than congenital rubella is to be this the to investigate all women for rubella antibody and to offer vaccination Rubella is now as an infection in England and with most cases as a of or surveillance of rubella and congenital rubella is with rubella immunization being for from countries with no immunization between and to The of infection susceptible women has been in women than in in to the situation MMR vaccine was and this is with the in the age group in which transmission is Rubella has been women of age in with particularly women from countries where rubella vaccination is not available. Between and percent of women were susceptible to However, this to a of percent between and the being given to this group The purpose of a second dose of rubella vaccine was to the proportion of susceptible children who be to transmission and susceptible to of this proportion not be to rubella transmission The second dose of MMR is offered at and was in (11). a vaccination the of the infection and the of infection in such an an to monitor the impact of vaccination through effective surveillance that accurate data from a of sources and When in conjunction with mathematical such surveillance data can also be used to the to which a disease will be and how may to be This has been in England and Wales to control measles, mumps, and rubella infection and was to the to MMR vaccine in 1988, the measles-rubella campaign in November 1994, and a two-dose for MMR in For vaccination to be it is that high vaccine coverage be reached and coverage will the age of infection and can to as in the case of where coverage in an in congenital rubella high coverage was achieved for the MMR vaccine in the but coverage subsequently to less than percent The in MMR vaccine coverage the the vaccine has in the media because of between the vaccine and and disease MMR vaccine coverage is to and vaccine significant in vaccine coverage in the it that public in the of MMR vaccine as an effective measles, mumps, and rubella be In addition to surveillance methods designed to the of diseases, it is to enhanced surveillance for the of vaccine vaccination This will that and will be to the decisions vaccination to control and eliminate vaccination for measles, mumps, and rubella have been in England and the surveillance has the value of including a second dose of vaccine in the which is now routinely for It has been that as the vaccination and the incidence of disease effective surveillance of measles, mumps, and rubella is suitable microbiologic vaccination is in only a proportion of notified cases are confirmed by laboratory that a surveillance system which on clinical diagnosis the for effective disease of a and collected type of clinical as oral fluid) with high compliance for laboratory confirmation a The of surveillance developed in England and Wales illustrates how surveillance can be used to provide the the of measles, mumps, and rubella to decisions on national vaccine policy. The number of confirmed cases of these is a to the of these surveillance methods are only being in England and Wales at they serve as a model for other countries with a commitment to controlling these it may not be to such an surveillance system in may be of particularly to Communicable Disease Surveillance measles notifications and vaccine coverage in England and of samples for measles antibody in England and Wales in using collected in and of notified and laboratory-confirmed cases of measles in England and January of samples for mumps antibody in England and Wales in using collected in and of notified and laboratory-confirmed cases of mumps in England and January of laboratory-confirmed mumps cases in England and Wales between 1995 and by measles-mumps-rubella (MMR) vaccination and of after of notified and confirmed cases of rubella in England and January of the population of England and Wales that was susceptible to rubella antibodies to between and by age group and of women of age to rubella in England and Wales between and by 1 Communicable Disease Surveillance Enteric, and Neurological Virus Laboratory, Central Public Health Laboratory, 3 Public Health Laboratory,