The Elimination of Indigenous Measles, Mumps, and Rubella from Finland by a 12-Year, Two-Dose Vaccination Program

The only MMR vaccine used in Finland during the study period was a live-virus vaccine (M-M-R II or Virivac, Merck, Sharp & Dohme, Rahway, N.J.) consisting of the Moraten (Enders-Edmonston B) strain of measles virus, the Jeryl Lynn B strain of mumps virus, and the Wistar RA 27/3 strain of rubella virus. The vaccine dose of 0.5 ml contained 25 μg of neomycin and less than 20 μg of phenol red indicator. It was injected subcutaneously into the upper arm or buttock.

Children 14 to 18 months and 6 years of age were vaccinated beginning in November 1982. In order to accelerate the immunization program, children who were between 18 months and 6 years old in November 1982 were vaccinated in 1983 through 1986. The program was later extended to military conscripts, children 11 to 13 years old, mothers who were seronegative for rubella after delivery, and students at various schools of nursing.

All childhood vaccinations were voluntary, free of charge, and performed by nurses after oral consent was obtained from parents.

As described in detail elsewhere,² the MMR vaccination program took place under the auspices of the National Board of Health and was organized by the National Public Health Institute in Helsinki. Briefly, Finland is divided into 1036 child health areas, each of which is covered by a child health center run by a physician and public health nurses. The nurses were trained in the program procedures by means of local teaching seminars and printed material.

Follow-up Studies

Special forms and instructions were designed for several substudies, which focused on vaccination rates,² common adverse reactions to the vaccine,¹¹ and tolerability of the vaccine in those with a history of severe allergy¹². Arrangements were made to identify the unvaccinated children¹³ and to detect any complications too rare for detection by a controlled trial in 1162 twins¹¹. Serologic confirmation was undertaken for suspected cases of disease. Every vaccinated person was identified by social security number. Unvaccinated children were identified by linkage of health center records with the national population register. In surveillance for rare adverse events, true complications were defined as those considered life-threatening (for example, anaphylaxis) and those that resulted in hospitalization or permanent sequelae.

Serum was obtained from a sample of 300 children from three areas at intervals of one to three years in order to monitor antibody levels. Measles antibodies were measured by the classic hemagglutination-inhibition technique,¹⁴ with antibody titers of 1:20 or higher considered protective. Rubella antibodies were determined with the hemolysis-in-gel technique¹⁵ (Orivir Rubella, Orion Diagnostica, Espoo, Finland). An area of hemolysis with a diameter of 6 mm or more was considered to indicate clinical protection. Measurement of mumps antibodies was performed by enzyme immunoassay and the plaque-inhibition technique¹⁶.

During the first five years of the study the incidence of the three diseases was monitored by clinical reports. In 1987 virologic confirmation was required for all suspected cases. Tubes for the collection of paired serum samples with prepaid mailing envelopes were distributed to child health centers.

Vaccination Coverage

The target population of the program was 563,000 children, of whom 86 percent were vaccinated in 1982 through 1986. A special campaign was then undertaken,¹³ beginning with reports in the national media and the local newspapers. Lists of unvaccinated children were then sent to the public health nurses, who personally contacted the parents. Finally, personal letters were sent to the nonparticipating families. These efforts resulted in the vaccination of 62,000 additional children, increasing participation to 97 percent. The hard-to-reach group was characterized by a survey and has since received special attention.

Vaccine Reactions and Complications

Table 1. Table 1. Symptoms and Signs Attributed to the MMR Vaccine in Five Studies.

A large double-blind crossover trial in twins showed only a few relatively common adverse reactions attributable to the MMR vaccine¹¹. In all, local or systemic reactions were observed in 4 to 5 percent of the vaccinated children. Our findings challenge those of previous uncontrolled studies (Table 1) that reported much higher rates of reactions^17-19. In one double-blind, placebo-controlled study,²⁰ the incidence of respiratory symptoms was virtually identical among the vaccinated and unvaccinated groups (72 and 74 percent, respectively).

In all, 1.5 million children and young adults, about a third of the Finnish population, were immunized in 1982 through 1993. No deaths or permanent sequelae were reported. There were more reports of rare complications early in the study than later, despite the use of the same surveillance system. Febrile convulsions attributable to vaccination had an incidence of approximately 7 per 100,000 vaccinated children, and 5 children had urticaria (0.6 per 100,000). Several children received epinephrine because of a suspected general allergic reaction; only two had probable vaccination anaphylaxis, and both recovered uneventfully. Diabetes was diagnosed in one child two weeks after vaccination, which is less than the expected incidence²¹. A girl who had earlier received a measles vaccine (Schwarz strain) was immunized with the MMR vaccine at seven years of age, and acute lymphatic leukemia was diagnosed 23 days later. Severe encephalopathy developed, and measles virus was isolated from the cerebrospinal fluid on two occasions. She recovered uneventfully²².

The most common complication clearly attributable to the MMR vaccine was acute thrombocytopenic purpura, detected in 23 children²³. The estimated incidence was 3.3 cases per 100,000 vaccinated children. Characteristically, petechiae or ecchymosis developed within 3 weeks after vaccination (median, 17 days). Circulating antiplatelet autoantibodies suggesting an autoimmune mechanism were found in 33 percent of the patients. Full recovery within six months with or without intravenous immune globulin therapy was the rule.

The vaccine was well tolerated even by those with a history of severe allergy; only 7 of 135 severely allergic children were not vaccinated because of a positive reaction to the MMR vaccine on a skin-prick test¹².

Serologic Findings

The measles component of the vaccine induced seroconversion in 99 percent of the 159 seronegative children from whom paired serum samples were available. The single exception was a girl who responded adequately to the rubella and mumps antigens. For the rubella component, the rate of seroconversion in 167 previously seronegative children was 100 percent. The antibody response to the mumps component of the vaccine was less clear because of the lack of a suitable laboratory method of assessing seroconversion. Antibodies were detected one year after vaccination in 94 percent of 170 previously seronegative children; this finding suggests that good protection was induced.

The measles and rubella antibodies persisted at high levels for at least four years after vaccination, and few children did not have protective levels of antibodies. The second dose of the vaccine given to children at six years of age also induced a good response against all three components, with the highest antibody levels against rubella and mumps (data not shown). The slightly weaker response to the measles-virus component (data not shown) might be explained by the very high antibody concentrations prevalent before the second dose was given, since the greatest response occurred among the children who had the lowest antibody levels after the first dose.

Clinical versus Virologic Diagnosis

Clinical suspicion of measles, mumps, or rubella was rarely confirmed serologically after the program had been in effect for several years. In a series of 655 consecutive vaccinated children in whom one of the three diseases was clinically diagnosed through 1986, measles was confirmed serologically in only 0.8 percent (5 cases), mumps in 2.0 percent (13 cases), and rubella in 1.2 percent (8 cases).

Incidence and Outbreaks

Figure 1. Figure 1. Annual Number of Cases of Measles, Mumps, and Rubella in Finland.

Reported cases are included through 1986, but only serologically confirmed cases thereafter. The arrows indicate the first receipt of inactivated mumps vaccine by conscripts in 1960 (this vaccine was administered through 1986), the first receipt of the Schwarz-strain measles vaccine by children and of the live-virus rubella vaccine by 11-to-13-year-old girls in 1975 (both were administered through 1982), the beginning of the national MMR vaccination program in 1982, and the beginning of the requirement for serologic confirmation of suspected cases of measles, mumps, or rubella in 1987. The vertical axis shows values on a logarithmic scale.

The vaccination program has had a substantial effect on the incidence of the three diseases. The lowest number of reported clinical cases of measles was 344 in 1985 (Figure 1). This implies a reduction of 87 percent from the five-year period before the vaccination program began (1977 through 1981) when there were, on average, 2704 cases reported per year. Actual protection rates were much higher, since unvaccinated age groups were also included in the reports. Serologic data indicate a decline from 317 new cases of measles in 1987 to 13 in 1993 (Figure 1).

Figure 2. Figure 2. Distribution of Cases during Outbreaks of Measles, Mumps, and Rubella.

The age groups covered by the MMR vaccination program are indicated by asterisks.

Small outbreaks of measles occurred in different regions, almost exclusively among the unvaccinated age groups. Often the index patient in such outbreaks had recently traveled abroad. Each year about 1 million Finns, one fifth of the population, visit another country. In an outbreak in 1988 and 1989, when there were 1748 confirmed cases of measles, there were far fewer among the 2-to-12-year-olds then covered by the program than among the unvaccinated older children (Figure 2).

Verified cases of mumps declined to less than 20 per year in 1990 through 1993 (Figure 1) -- a decrease of more than 99 percent from the average of 9366 cases per year in 1977 through 1981. A local outbreak of 77 proved cases occurred in central Finland in 1987 and 1988 (Figure 2). Again, these cases occurred almost exclusively in the unvaccinated age groups. The changing epidemiologic features of mumps were reflected clinically in the fact that the most common form of encephalitis, that caused by the mumps virus, disappeared from the country¹.

The difficulty of making the diagnosis of rubella on clinical grounds renders traditional reporting unreliable. Nevertheless, no vaccinated patient was found among 423 with laboratory-confirmed rubella in 1984 through 1986, including some children too old (11 years or more) to have been vaccinated during the program²⁴. The lowest number of rubella cases was in 1993, when there were 25 documented cases overall (Figure 1). The efficacy of the vaccine in a rubella outbreak is indicated by the age distribution of the 336 cases confirmed in 1991 (Figure 2).

In slightly more than a decade a program of vaccination against measles, mumps, and rubella virtually eliminated these diseases from Finland. Careful planning, efficient organization, and an effective system of primary health care for children ensured high rates of vaccination coverage throughout the country. The two-dose vaccination policy was an important factor in the program's sustained success. The rationale for this policy was threefold. First, up to 5 percent of recipients were expected to have no response to all three vaccine components (primary failures)²⁵. Second, a two-dose regimen was considered likely to reach more children than a single-dose schedule. Finally, the second dose was expected to boost declining antibody concentrations.

The decision to immunize children at 14 to 18 months and 6 years of age was made for several reasons. There was no reason for initiating immunization with the MMR vaccine before the first birthday, as there is in developing countries²⁶. The system of primary health care for children required a visit to the center after the first birthday, when no other immunization was scheduled. Because most Finnish children enter school at seven years of age, the second dose was received before school entry, thus avoiding the need to involve the school health system in the program. The two-dose vaccination policy generated higher costs than a single-dose policy but had the advantage of compensating for more violations of the vaccination schedule.

The vaccine proved to be safe. For common adverse reactions (occurring within 21 days after vaccination), the data could hardly be more reliable than those from the double-blind, placebo-controlled crossover study performed in more than 1000 twins¹¹. In that study, the great majority of symptoms and signs earlier attributed to the vaccine were found to have other causes. The incidence of adverse reactions was substantially lower among these children than among children in four studies, three of which were uncontrolled (Table 1). Except for mild arthralgia in about 1 percent of those vaccinated,¹¹ arthropathy or other adverse effects have not occurred, perhaps because we were immunizing children rather than adults, in whom acute arthritis more often develops²⁷.

Among rare adverse reactions, nationwide surveillance revealed that acute thrombocytopenic purpura, clinically indistinguishable from childhood idiopathic thrombocytopenic purpura, was the most common true complication of the MMR vaccine. This is not surprising, since low platelet counts also occur after natural cases of measles,²⁸ mumps,²⁹ and especially rubella^30,31. Moreover, thrombocytopenia has been described after vaccinations against measles³² and rubella,³³ although the incidence was lower than that following natural infection³⁰. In our program, no permanent sequelae of post-vaccination thrombocytopenia have been reported²³. Mumps-vaccine-related meningitis is rare³⁴.

A small, slow decline in antibody levels occurred after the first dose of the vaccine. Although a loss of antibodies to levels below the detection limit of available assays does not necessarily imply the loss of clinical protection,¹⁶ it is plausible that boosted levels predict long-lasting protection³⁵.

Only 0.8 percent of the clinically diagnosed cases of measles were verifiable serologically, and the disease gradually disappeared from Finland. When measles was still rampant, cases of measles-like diseases (whose incidence may not have changed much since the vaccination program) represented a tiny fraction of the true measles cases. Now the ratio of measles-mimicking diseases to true measles is dramatically higher. Hence, the true clinical efficacy of the MMR vaccine is far greater than it appears to be when cases without serologic confirmation are included. This difference in apparent incidence prompted a new approach after five years of the program. From 1987 on, only virologically proved cases of measles, mumps, or rubella were counted. Serologic testing, which was encouraged and provided free of charge, was the only method of confirming that the true incidence was approaching zero, even if some cases were missed in the process. Our studies in progress indicate that many suspected cases of these three diseases were in fact caused by other viruses. A reduction of more than 90 percent in the incidence of all three diseases occurred within a few years, as had been expected on the basis of vaccination programs in the United States³⁶ and Sweden¹⁸. Further declines to almost zero occurred over the 12 years of the program.

In Finland, the use of the MMR vaccine has wiped out one third of all childhood encephalitis, of which 16, 13, and 2 percent of the cases were once caused by the mumps, measles, and rubella viruses, respectively¹. The rising incidence of type 1 diabetes, especially among five-to-nine-year-old children, has reached a plateau,³⁷ a fact that may be due to the vaccination program. Vaccination against rubella is undertaken to eliminate the associated embryopathy. Declining numbers of verified cases of rubella (Figure 1), with no vaccinated persons among 423 with proved cases in 1986,²⁴ indicate that the MMR vaccine is successful in preventing this disease.

Maintaining good immunity in the population is the cornerstone of the control of epidemics of measles, mumps, and rubella. In Finland, where the primary health care system is well organized and efficient and the two-dose vaccination regimen is routine, it is feasible to immunize the population with a further dose, should the need arise. In 1987 and 1988, a mumps outbreak (Figure 2) was successfully extinguished by the immunization of 500 previously unvaccinated schoolchildren.

In the United States, where, in addition to the vaccine failures in young adults, many cases of measles occur in preschool children, those living in the inner city, those who are indigent, and unimmunized infants,³⁸ the recommended 70 to 90 percent vaccination coverage^39,40 is generally not achieved before school age⁴¹. A second dose is likely to prevent primary and secondary vaccine failures,⁴² and in fact, this policy was adopted in 1989 in the United States and in 1991 in New Zealand⁴³; it is under discussion in Japan⁴⁰ and the United Kingdom⁴⁴. However, a two-dose policy alone does not guarantee the protection of young children if vaccination is not ensured before the age of school entry⁴¹. A policy such as ours, which includes tracing unvaccinated children, has proved effective, feasible, and cost effective,¹³ and such a policy might be adopted in the United States⁴¹. Hong Kong, on the other hand, has chosen another strategy⁴⁵. Every contact of a child with the health care system is regarded as an opportunity for immunization; this policy has increased vaccination rates to 97 percent. The same strategy is now being used in the United States⁴⁶.

In Finland, the annual number of cases of measles, mumps, and rubella has been reduced to a few dozen, and the circulation of the three viruses has been interrupted. Indigenous cases of measles, mumps, and rubella have been virtually eliminated. Since international tourism continues to lead to the importation of these viruses, however, it is necessary to maintain high rates of vaccination coverage if future epidemics are to be prevented.