MMR2 Immunization at 4 to 5 Years and 10 to 12 Years of Age: A Comparison of Adverse Clinical Events After Immunization in the Vaccine Safety Datalink Project
Background. The Advisory Committee on Immunization Practices recommends a second dose of measles, mumps, and rubella vaccine (MMR2) at age 4 to 5 years of age, whereas the American Academy of Pediatrics suggests MMR2 immunization at age 11 to 12 years of age. Because there is little information on whether the rate of adverse reactions to MMR2 immunization varies among these two age groups, we took advantage of differing immunization policies at two large HMOs to compare the frequency of clinical events after, and possibly related to, MMR2 immunization.
Methods. Information was collected on clinical events plausibly associated to MMR immunization (seizures, pyrexia, malaise/fatigue, nervous/musculoskeletal symptoms, rash, edema, induration/ecchymoses, lymphadenopathy, thrombocytopenia, aseptic meningitis, and joint pain) in two cohorts. At three facilities at Northern California Kaiser (Oakland, CA), 8514 children received MMR2 immunization at age 4 to 6 years of age; at Group Health Cooperative (Seattle, WA) 18 036 children received MMR2 immunization at age 10 to 12 years of age. To account for age-related differences in health care use, within each HMO, clinical events in a 30-day period after immunization were compared with a 30-day period before vaccination.
Results. Children 10 to 12 years of age were 50% more likely to have a clinical event after MMR2 immunization than in the period before immunization (odds ratio, 1.45; 95% confidence interval: 1.00,2.10). Children 4 to 6 years of age were less likely to have a visit for an event after immunization compared with the period before immunization (odds ratio, 0.64; 95% confidence interval: 0.40,1.01).
Conclusions. These results suggest that the risk for clinical events after MMR2 immunizations is greater in the 10- to 12-year age group.
The Advisory Committee on Immunization Practices currently recommends a routine second dose of measles, mumps, and rubella vaccine (MMR) at age 4 to 5 years, before school entry.1 This recommendation differs from that of the American Academy of Pediatrics Red Book Committee, which suggests that a routine second dose of MMR vaccine be given at age 11 to 12 (or earlier if required by local school entry requirements).2
Although different, both of these recommendations are based on sound rationale. Immunization at a younger age has been felt to be more easily accomplished because of greater ease of access to this age group by the health care professions. Conversely, a health supervision visit is desirable at age 11 to 12 years to address important medical, developmental, and behavioral issues of early adolescence and to administer other recommended immunizations. Additionally, MMR immunization at an older age (11 to 12 years) might minimize the impact of waning immunity from earlier immunization and ensure immunity that persists into the childbearing years.2 MMR immunization at 11 to 12 years would also have a more rapid impact on reducing the number of adolescents in the United States currently susceptible to measles as a result of primary vaccine failure. Different states have currently adopted different policies on the age for the second dose of MMR vaccine.
Side effects occur in 5% to 15% of recipients of a first MMR injection and are generally limited to susceptible vaccine recipients.3 Fever, beginning usually ∼1 week after immunization, and rash are the most common reactions after measles vaccination. Reactions temporally related to mumps vaccination are rare, but include fever, parotitis, and rash.4 Side effects after rubella vaccination are related to age of the vaccinees. Of those receiving the rubella vaccine, from 0% to 2% among females 6 to 12 years old and 13 to 16 years old, respectively, experience arthralgia and arthritis after immunization, compared with 25% of susceptible postpubertal females 20 to 24 years of age.5,6
There is little information in the literature that addresses the rates of adverse reactions to a second MMR injection in these two age groups. To study this issue, we took advantage of differing immunization policies at two large West Coast health maintenance organizations (HMOs) to compare the frequency of clinical events temporally associated with the second dose of MMR vaccine and to assess the implications for immunization policy.
Group Health Cooperative of Puget Sound (GHC) is a staff model HMO based in King County, WA, serving > 635 500 members in the state of Washington. Northern California Kaiser (NCK) is based in Oakland, CA, and serves > 2.2 million members. GHC routinely gives the second dose of MMR vaccine at age 10 to 12 years, whereas NCK routinely administers this dose at age 4 to 6 years. Both HMOs are members of the Centers for Disease Control and Prevention Vaccine Safety Datalink study, where automated vaccination and medical records are linked for epidemiologic studies of vaccine adverse events.7
For this study, we identified two cohorts of children, one from each HMO. The two cohorts were explicitly selected so that the type and amount of available information was equivalent between the two HMOs. To accomplish this, at GHC the cohort was composed of children who received an MMR immunization between the ages of 10 and 12 years, whereas at Northern California Kaiser the cohort was composed of children who received an MMR immunization between the ages of 4 and 6 years at one of three facilities. For all GHC children, and for children at the three NCK facilities, information was available on subsequent visits to outpatient clinics or emergency departments, or for hospitalizations. We excluded from study all children who received immunization at Northern California Kaiser facilities not able to track information (at the time of this study) on medical visits after vaccination. At each site, children identified as having received an MMR immunization and who were enrolled at least 3 months before and 3 months after vaccination were included in the study. The resulting cohorts of children provided the population for this study. Within these cohorts, we studied immunizations given during the period extending from March 1991 through December 1994. In the present study, we use the term MMR2 to refer to the MMR injection number received by children within each of these two cohorts, not to any specific products used formerly or currently.
The analysis was performed in several steps. First, to identify potential vaccine-related clinical events, within each cohort, information was collected on the number and types of visits to health care providers in the 30 days after MMR immunization. As noted above, at GHC all visits for children in the cooperative to outpatient clinics and emergency departments and hospitalizations were identified, and at NCK all visits to outpatient clinics and emergency departments and hospitalizations were identified for children who attended three separate outpatient clinics. Next, to account for age-related differences in overall health care use rates between the two cohorts of children, the number of visits was counted for two separate periods: one period before and one period after immunization with MMR2. The first period began 3 months before immunization and lasted 30 days, ending 2 months before immunization. The second period began the day after immunization and lasted 30 days. All visits to any site (ie, emergency department, outpatient clinic, hospitalization) were counted. The period immediately before immunization was purposely not selected as the comparison period to avoid studying children at a point where they are most likely to be healthy.8
Then we progressed to study common side effects of MMR immunization, which include rash, fever, malaise, and induration at the injection site. Other more rare but biologically plausible adverse events that may be causally associated with MMR immunization include aseptic meningitis, seizures, lymphadenopathy, thrombocytopenia, and joint pain. Therefore, to compare rates of these events after MMR immunization by age, we collected information for the following set of diagnoses: seizures, pyrexia, malaise/fatigue, nervous/musculoskeletal symptoms, rash, edema, induration/ecchymoses, lymphadenopathy, thrombocytopenia, aseptic meningitis, and joint pain.
Rates of all visits and rates of visits for the specific diagnoses of interest, listed above, were calculated at both sites for the periods before and after vaccination. Unconditional and conditional logistic regressions were used to calculate the odds ratio (OR) to compare the risk for events after vaccination relative to the risk in the period before vaccination. Adjustment was made within these calculations for gender and season of vaccination. Subsequent analyses calculated the OR after excluding children who received a pertussis-containing immunization (primarily NCK members) or after excluding children receiving hepatitis B (primarily GHC members).
In the first set of analyses, the first visit for any specific diagnosis (for example, the first visit for rash) within each time frame was counted. Subsequent visits for the same diagnosis were not included in the analysis. However, because the specific diagnoses examined were possibly not independent events (ie, fever and rash occurring in the same person), subsequent analyses were also performed that counted each person only once within each period (regardless of the number of different diagnoses made for that person within each period).
Finally, for the clinical events found most commonly after vaccination (rash, seizures, and joint pain), patient charts were reviewed to confirm the presence of the condition and that the visit was for an event of acute onset. For cases that met these criteria (a new event of acute onset either before or after vaccination), the analysis was repeated in the same manner.
We identified a total of 18 036 children receiving MMR vaccine between the ages of 10 and 12 years at GHC, whereas 8514 children received MMR vaccine between 4 and 6 years of age at NCK in the three facilities of interest (Table 1).
At GHC, 2101 patients were seen by health care providers in the 30-day period beginning 3 months before MMR immunization. The vast majority of these 2082 visits were in the outpatient setting. In the month after MMR immunization, a total of 2907 children were seen by health care providers; again, the majority of these 2891 visits were in the outpatient clinic. At NCK, 874 patients were seen in the 30 days beginning 3 months before MMR immunization. Here also the majority (687) were seen as outpatients. In the month after MMR immunization, 523 children were seen by health care providers; again, most (350) in clinics.
The distribution of MMR immunization by season was similar at the two sites, although proportionately more MMR immunizations were given in the summer to 10- to 12-year-old children. Among the 10- to 12-year-old children, hepatitis B was given concurrently to 5412 (30%) undergoing MMR immunization, whereas only 10% of the 4- to 6-year-olds received hepatitis B concurrent with their MMR immunization. As expected, more children in the younger age group received DTaP (or a DT or Td formulation), DTP, and oral polio virus vaccine at the same time as the MMR immunization.
In the month after MMR2 immunization, there were a total of 31 visits at NCK and 68 visits at GHC for the conditions of interest, including the most common events of seizures, rash, and joint pain (Table2). Children at NCK were less likely to have a visit for one of the diagnoses in the month after vaccination relative to the rate in the month that was 3 months before vaccination (OR, 0.64; 95% confidence interval [CI]: 0.40,1.01, conditional logistic regression [CLR]). However, children at GHC were almost 50% more likely to have a visit for one of the diagnoses in the month after vaccination compared with the period before vaccination (OR, 1.45; 95% CI: 1.00,2.10, CLR). These results did not change appreciably when we counted each person only once (regardless of the number of different diagnoses that occurred within each independent period) or when analyses were adjusted for season and gender. Results were similarly unchanged when we excluded from the analysis children who received a whole-cell pertussis formulation at the time of MMR immunization or when we excluded children who received hepatitis B immunization.
When cases were reviewed, we found that the majority of visits for rash or joint pain after MMR immunization were for acute events, whereas most diagnoses of seizures were, in fact, visits for other reasons (such as follow-up visits for previous seizures or well-child care visits among children with seizures). When the analysis was limited to medical record-confirmed specific diagnoses, the direction of the associations found previously did not change. Children 10 to 12 years of age had a significant increase in visits for rash, seizures, and joint pain after MMR2 immunization (OR, 1.78; 95% CI: 1.07,2.97, CLR). Younger children undergoing MMR2 immunization did not show this increase and, in fact, had statistically significant fewer visits after MMR2 immunization (OR, 0.52; 95% CI: 0.28,0.96, CLR) (Table3).
Finally, among the reviewed cases, we found that risks for arthropathy and rash after MMR2 immunization were more pronounced among older females (OR, 1.83; 95% CI: 1.01,3.30, CLR) than among older males (OR, 1.23; 95% CI: 0.76,2.00, CLR).
Although the automated databases identified 17 seizures after immunization within the two cohorts, only 3 were confirmed by chart review as being acute seizure events (as opposed to check-back visits, follow-up care, or visits for seizure medication). These 3 confirmed seizures occurred in the month after MMR immunization among the older group, whereas none occurred after MMR immunization in the younger group. Of the three children with confirmed seizures that occurred after MMR immunization, one had a grand mal seizure (with a subsequent diagnosis of partial seizures with secondary generalization), one had a syncopal seizure, and one had a partial complex seizure. These latter two children had had similar seizures, before the seizure that occurred after MMR2 immunization. The child with the grand mal seizure had been evaluated by a pediatric neurologist for a tic disorder but had not had a diagnosis of a seizure disorder before undergoing MMR2 immunization. According to information available in the medical records, all three children had been seizure-free for at least 2 years after their last documented seizure episode.
We found a greater risk for clinical events temporally related to MMR2 immunization in the 10- to 12-year age group than in the 4- to 6-year age group. Even though the risk for such events appears to decrease after MMR immunization in the younger age group (an observed absolute decrease of 2.0 visits per 1000 person-months), it is unlikely that MMR immunization is associated with a protective effect against the advent of seizures, rash, and the other conditions listed in Table2. Rather, it is probable that the decreased rate of adverse conditions in the month after MMR immunization is a manifestation of the healthy vaccinee effect. This phenomenon, noted previously, is the propensity for children to be immunized when they are in a relatively healthy, disease-free state.8 Hence, the relative paucity of disease at the time of immunization is also associated with a decreased rate of disease in the period directly after vaccination. In this view, the increased risk for clinical events found in the month after the immunization of 10- to 12-year-old children may, in fact, have underestimated the true increased rate of side effects after MMR immunization (an observed absolute increase of 1.7 visits per 1000 person-months). Regardless, immunization was significantly associated with a decreased risk of clinical events such as rash, seizures, and joint pain in the younger age group and a significantly increased risk of clinical events in the older age group. Within our study, not only were such events more common in the month after MMR immunization among the older age group, but the increase was particularly marked among females. The fact that this association was strongest in adolescent females lends biologic plausibility to our findings, because arthralgia and arthritis are known to be more common after immunization of older females.5,6 It is worth noting that the absolute number of medical visits for possible MMR-related adverse events was small, and the majority of these were outpatient visits for relatively mild conditions likely to be self-limited.
A particular strength of this study was that by using each age cohort as its own comparison group, we were able to control not only for age-specific differences in medical use, but also for other more general differences in medical care between the two HMOs. However, there were some inherent limitations to this study. First, because of the timing of the start of the database, we were not able to identify specific children who had not undergone MMR immunization previously. However, both Washington state and California have school entry requirements, and it is likely that the majority of children in both age groups had undergone MMR immunization previously. In Washington state, among children at school entry, 98.0% have been immunized against MMR. However, if the MMR immunization received by the study cohort members during this period was, in fact, the first MMR rather than the MMR2 immunization, such vaccinees were probably more likely to be in the younger cohort. Because most side effects are more common with the first MMR immunization (ie, among susceptible children), we would have expected to see an increased risk of clinical events among the younger age group, not the older group as observed. Such a misidentification of MMR dose in the younger age group, if it occurred, would most likely have lessened the adverse event rate difference that was found between the two age groups.
A second limitation is that this study probably underestimated the true rate of clinical events that follow MMR immunization. Patients undergoing MMR immunization are routinely informed of expected side effects. Many children with mild symptoms (such as fever or rash) who did not seek care from health care providers would have been missed by our study. Although a prospective study might be preferable to assess adverse reactions that otherwise would have not come to medical attention, such a study would be very large and extremely expensive to assess minor or small differences between these two age groups.
Finally, it may be that 10- to 12-year-old children were more likely than 4- to 6-year-old children to seek medical care after immunization. Such questions, which focus on behavior after vaccination, are difficult to prove or disprove, and our data cannot fully answer this question. However, it is equally plausible that parents of 4- to 6-year-old children would be more likely to bring their children in for a rash or fever after immunization than would parents of 10- to 12-year-old children.
We attempted to discover why the overall visit rate increased after vaccination among the older adolescents. We reviewed a random sample (n = 82) of charts among 10- to 12-year-old children with a visit within the month after immunization for a condition other than the selected diagnoses of interest. Of these 82 visits, 11 were check-back visits (visits to recheck a finding found previously at the time of immunization); 11 were referrals to optometry; and 3 were for hepatitis B immunization. The remaining visits were for acute or chronic problems presumably unrelated to immunization. Therefore, ∼30% (25/82) of visits were for follow-up of findings discovered at the time of immunization or for a subsequent dose of hepatitis B in the three-dose series. These findings suggest that the increased rates of visits after immunization among older children arise from follow-up and referral activities as a result of routine health care at this age. In comparison, younger children seldom receive hepatitis B vaccine at 4 to 6 years, have their vision routinely screened in office, and may receive fewer follow-up visits—after any given visit—because of more frequent overall medical care use.
In this study, it was not possible to differentiate whether the clinical events found were attributable to an increasing proportion of susceptible children in the older age group (as a result of waning immunity) or a greater propensity for side effects among older susceptible children. Rash, joint pain, seizure, lymphadenopathy, and fever are well documented adverse events associated with the measles and rubella components of MMR vaccine, and such sequelae appear to occur primarily among susceptible children. Joint-related complaints after MMR immunization are more frequent among susceptible adolescents and adults, particularly females, than among preschoolers. After primary immunization with MMR, a proportion of children manifest mild, attenuated illness, and it is recognized that after natural measles infection, older children are more likely to have serious illness than are younger children.9-12 These findings, along with our results, would argue for earlier administration of a second dose of MMR2 vaccine to reduce the number of age-related side-effects among susceptible vaccinees. Conversely, if waning immunity after vaccination contributed substantially to the numbers of susceptible 10- to 12-year-olds, the excess occurrence of adverse events in this group supports rather than argues against administering the second dose of MMR vaccine at 11 to 12 years of age. To date, however, most available information suggests that immunity to measles is lifelong and that waning immunity does not occur to a meaningful degree.13 A recent metaanalysis suggests that the rate of waning immunity after measles vaccination is probably < 0.2%, countering this rationale for delayed administration of MMR2 vaccine.14 The majority of epidemiologic evidence from outbreaks in the United States in general has suggested that waning immunity does not play an important role and that the predominant reasons for susceptibility result from either nonvaccination or primary vaccine failure.
The differing rationales for MMR immunization at 4 to 6 years of age or 10 to 12 years of age remain valid. As noted previously, these reasons relate to the ease and cost of vaccinating a large percentage of the target population and the need for a health care visit in early adolescence. In this study, we found a greater risk for adverse clinical events after MMR2 immunization among 10- to 12-year-olds than among 4- to 6-year-olds. The results of our study may be of interest to the major advisory groups (ACIP and AAP) and others for considering the optimal age for administration of the second MMR vaccine.
- Received December 17, 1996.
- Accepted April 7, 1997.
Reprint requests to (R.L.D.) Group Health Cooperative Center for Health Studies, Immunization Studies Program, 1730 Minor Ave, Suite 1600, Seattle, WA 98101-1448.
Other Vaccine Safety Datalink Team members, by site. National Immunization Program, Centers for Disease Control and Prevention: John W. Glasser, PhD, MPH; Philip H. Rhodes, PhD; Emmett Swint, MA. Group Health Cooperative of Puget Sound: Lisa A. Jackson, MD, MPH; William E. Barlow, PhD; Virginia H. Immanuel, MPH; Patti J. Benson, MPH. Northwest Kaiser Permanente: John P. Mullooly, PhD; Lois Drew, BA; Barbara Mendius, MS. Kaiser Permanente of Northern California: Ned Lewis, MPH; Bruce H. Fireman, MA. Center for Vaccine Research, Harbor-UCLA Medical Center: Joel I. Ward, MD; Connie M. Vadheim, PhD; S. Michael Marcy, MD; Jennie Jing, MA; Michael Wulfson MD, PhD. Kaiser Permanente of Southern California: Marlene Lugg, DrPH; Patricia Osborne, MS. Center for Biologics Evaluation and Research, Food and Drug Administration: Robert P. Wise, MD, MPH; Suresh Rastogi, PhD; Peter Patriarca, MD. Division of Vaccine Injury Compensation, Health Resources and Services Administration: Vito Caserta, MD, MPH.
- MMR =
- measles, mumps, and rubella vaccine •
- HMO =
- health maintenance organization (HMO) •
- GHC =
- Group Health Cooperative of Puget Sound •
- NCK =
- Northern California Kaiser •
- OR =
- odds ratio •
- CI =
- confidence interval •
- CLR =
- conditional logistic regression
- ↵American Academy of Pediatrics. Measles. In: Peter G, 1994 Red Book. Report of the Committee on Infectious Diseases. 23rd ed. Elk Grove Village, IL: American Academy of Pediatrics; 1994:308–322
- ↵Markowitz LE, Katz SL. Measles vaccine. In: Plotkin SA, Mortimer EA Jr, eds. Vaccines. Philadelphia, PA: WB Saunders; 1994:229–276
- ↵Cochi SL, Wharton M, Plotkin SA. Mumps vaccine. In: Plotkin SA, Mortimer EA Jr, eds. Vaccines. Philadelphia, PA: WB Saunders; 1994:277–301
- Swartz TA,
- Klingber W,
- Goldwasser RA,
- Klingber MA,
- Goldblum N,
- Hilleman MR
- Weibel RE,
- Stokes J,
- Buynak EB,
- Hilleman MR
- Chen RT,
- Glasser JW,
- Rhodes PH,
- et al.
- Fine PE,
- Chen RT
- ↵Black FL. Measles. In: Evans AS, ed. Viral Infections of Humans. Epidemiology and Control. 3rd ed. New York, NY: Plenum; 1989:451–465
- Barkin RM
- Barkin RM
- Copyright © 1997 American Academy of Pediatrics