The Origin and Spread of a Mumps Epidemic: United Kingdom, 2003–2006 : Epidemiology

Secondary Logo

Journal Logo

Infectious Diseases: Original Article

The Origin and Spread of a Mumps Epidemic

United Kingdom, 2003–2006

Jick, Hershel; Chamberlin, Douglas P.; Hagberg, Katrina Wilcox

Author Information
doi: 10.1097/EDE.0b013e3181b056f5

Abstract

An epidemic of mumps began in the United Kingdom (UK) in 2003 and continued through at least 2005.1–3 It was by far the largest recorded epidemic in almost 4 decades in the United States (US) or the UK. In this paper, we describe the origin, geographic spread, and scope of the 2003–2006 epidemic. In addition, we evaluate the effectiveness of the mumps, measles, and rubella (MMR) vaccine. The findings are based on standardized information derived from a comprehensive, validated, computerized UK general practice database.

METHODS

Data Source

This study was based on information derived from the General Practice Research Database, a large automated UK medical record database. The source population for this study comprises all people for whom data are recorded in the Database which is maintained by the Boston Collaborative Drug Surveillance Program. Some 3 million people annually have been continuously enrolled in 350 general practices and followed through 2006. The size and distribution of the population by age and region has remained steady during this time.

The comprehensive nature of the information on clinical diagnoses and drug exposure recorded in the Database has been repeatedly validated and found to be of high quality for the purpose of conducting epidemiologic research.4–8 The general practitioners (GPs) who contribute data use office computers in their routine practice, and have agreed to provide the information for research purposes. The GPs have been trained to record information (including demographic data, medical diagnoses, and deaths) in a standard format. The GPs routinely record the administration of vaccines. Diagnoses, physical findings, and administrative events are recorded using Oxford Medical Information System or Read codes. All information is anonymized.

The distribution of age, sex, and geography in the practices that contribute to the Database is closely similar to that of the UK except that London is somewhat underrepresented.9 The UK population has been divided into 11 regions by the Medicines and Healthcare products Regulatory Agency, which is responsible for the data collection and storage. The regions are further subdivided into postal zones (using the first 2 digits of the postal code), each containing no fewer than 3 practices to avoid comprising confidentiality. In this paper, we identify the postal zones by the city, town, or county upon which it is based.

The protocol for this study was reviewed and approved by the Independent Scientific Advisory Committee of the Medicines and Healthcare products Regulatory Agency.

Case Definition

We considered a person to be an incident case of mumps if the patient had a first-time clinical diagnosis recorded in the computerized medical record between 1 January 2003 and 31 December 2006, without any prior recorded mumps diagnosis. We could not obtain laboratory results to validate the mumps diagnoses; however, surveillance reports have indicated that accuracy of mumps diagnosis in the UK is high, especially during outbreaks. These studies indicate that more than 75% of clinically reported cases of mumps in the UK were confirmed by serum or oral fluid IgM laboratory testing during the 2004 outbreak.3,10

Incidence Estimation and Source Population

We used a cohort design to estimate 3-monthly regional and GP-specific cumulative incidences of clinically diagnosed mumps by age and calendar year for the years 2003–2006. We identified all people for whom data were recorded in the Database through 31 December 2006.

Evaluation of Vaccine Status and Effectiveness

We used a nested case-control design to estimate the extent to which mumps vaccination protected from the risk of infection. Cases and controls were 1–24 years old at the time of the mumps diagnosis and were born in or after 1982. Up to 4 controls who had no diagnosis of mumps were identified from the study population. Controls were matched to each case by age, sex, general practice, calendar time (the date of the case's mumps diagnosis), and the duration of time the patient had been registered in the Database.

Data Analyses

The nationwide incidence of mumps rose from early 2003 to the middle of 2005, when it reached a peak and then fell sharply until the end of 2006. To clarify the degree of change over time, we computed 3 month cumulative incidences by determining the number of cases in every 3-month period and then dividing by the number of people who were active in the practices during the same 3-month period (and who, therefore, would have been recorded as a case of mumps if mumps developed). Region- and practice-specific cumulative incidences were estimated according to age and calendar time.

We used conditional logistic regression in the case-control analysis to estimate the relative risk of mumps infection in those who received the mumps vaccination compared with those who did not. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. The statistical analysis was conducted using the software program SAS, version 9.1 (SAS Institute Inc, Cary, NC).

RESULTS

A total of 5683 cases of mumps were recorded by participating GPs for all ages during the period 2003–2006. Fifty-three percent were male and 90% were age 30 or younger. Based on our estimate that the Database represents about 5% of the UK population, we infer that there were more than 100,000 cases of mumps diagnosed by GPs during the 2003–2006 epidemic.

The incidence of mumps began to rise in early 2003. It progressively increased over time, peaking in the first and second quarters of 2005 and falling to pre-epidemic levels by the end of 2006. Overall, the incidence was highest among young adults age 18–24 years (Fig. 1).

F1-7
FIGURE 1.:
Three-month cumulative incidence of mumps by year and age group.

In 2005 alone (the peak year of the epidemic) there were 3553 cases of mumps recorded in the Database. Thus, we estimate that there were about 70,000 cases in the entire UK in that year.

Region-Specific Incidence of Mumps

We estimated region-specific 3-month cumulative incidences of mumps by calendar time and age (Fig. 2). The incidence began to rise in the first 6 months of 2003 in one practice in Wales where 36 cases were diagnosed. Most were 10–14 years old. The outbreak quickly spread to other parts of Wales and western Britain. By the middle of 2004, outbreaks had spread to the midlands and reached the East in mid 2006. Except for Wales and Scotland, outbreaks in a particular city or region tended to last for less than 9 months and tended to occur in larger urban areas with large populations of college students.

F2-7
FIGURE 2.:
Three-month cumulative incidence of mumps by calendar time and region.

The details of the outbreaks according to region and postal codes are provided in an online appendix (eAppendix, https://links.lww.com/A1447).

MMR Use and Effectiveness

The MMR vaccine was introduced in the UK in October 1988. In 1989, about 95,000 children age 1–19 years had a recorded MMR vaccination in their computer record, from which we estimate that about 1.9 million children of this age group were vaccinated nationwide during 1989. The distribution of the number the people who received a first MMR vaccination according to age and calendar year is provided in Figure 3. The number of doses of MMR given each year by age is provided in Figure 4. In the later years most were receiving a second dose. The number of children age 1–2 years who received a first vaccination remained stable until 1999, when it fell in connection with the reported association between MMR vaccination and autism.6,11 Vaccination of children age 3–9 years diminished from 1989 to 1992 and remained low, except for an elevation in 1994 in association with a measles outbreak.1,3 Vaccination in children age 10–19 years showed a similar pattern. In 1996, a large nationwide campaign to provide a second MMR dose was initiated. We estimate that 3.7 million children were vaccinated nationwide in 1996–1997 (Fig. 4). Few, if any, young adults age 20 or more were vaccinated until 2004 when the mumps outbreaks led to a 3-year vaccination program. Increased vaccination of young adults is demonstrated in Figure 4.

F3-7
FIGURE 3.:
Distribution of first MMR vaccination by year and age group.
F4-7
FIGURE 4.:
Number of people who had first and/or second vaccination by age and year.

We evaluated the effectiveness of the MMR vaccine by comparing vaccination status in cases of mumps to that in matched controls (Table). Vaccinations reduced the odds of mumps for children of all ages; it was slightly more effective for those younger than 10. Two doses of MMR vaccine were more effective than one dose, with the strongest protection occurring in children younger than 10 with 2 vaccinations (OR = 0.16, 0.12–0.22). We conducted a similar analysis evaluating MMR vaccine effectiveness in the single practice in Wales where the earliest cases of mumps were recorded in the first 6 months of 2003. For the 36 cases and 135 matched controls aged 10–17, the ORs for the risk of mumps in those who received one and 2 vaccination were 0.24 (18 cases and 91 controls; 95% CI = 0.09–0.65) and 0.32 (8 cases and 30 controls; 0.09–1.09), respectively, compared with those who were not vaccinated (10 cases and 14 controls).

T1-7
TABLE:
Odds Ratios for Mumps Infection by MMR Vaccination Status and Age Group

DISCUSSION

The medical information recorded in the Database since 1989 contains a detailed description of the UK epidemic of mumps in 2003–2006. It provides a context for recorded mumps outbreaks before and after the introduction of the MMR vaccine.

The first vaccine to prevent mumps was licensed in the US in 1967 after an epidemic that numbered more than 140,000 cases.12 In 1977, a policy to implement a one-dose mumps vaccine (primarily MMR) for children was initiated in the US. Subsequently, the annual number of reported mumps cases remained low until 1986–1987 when a modest increase in the number of mumps cases was reported. The official recommendations were changed in 1989 to 2 doses of MMR, and the incidence of mumps in the US had remained low until 2006.12

Routine immunization against mumps was not introduced in the UK until October 1988 provided as a combined measles, mumps, and rubella vaccine (MMR). This was recommended for children age 12–15 months.1,2,13 For 22 years prior to 1988, the annual incidence rate of mumps in the UK reported by GPs to the Communicable Disease Surveillance Centre had ranged from about 250 to 1000 cases per 100,000 population.1 In 1994, an outbreak of measles prompted a national catch-up campaign for 5–16 year olds with a combined measles and rubella vaccine (MR). At that time, a global shortage of MMR prevented full offering of the combined vaccine that also included mumps.1 In 1996–1997, a campaign introducing a second dose was initiated since most children had previously received only one dose of the MMR vaccine (Fig. 3).1,2

After the introduction of the MMR vaccine and the catch-up vaccination campaigns, the reported annual incidence of mumps decreased to fewer than 150 cases per 100,000 people through the 1990s.1

After recognition of the substantial increase in the incidence of mumps starting in early 2003, a catch-up MMR vaccination program was started in the beginning of 2004, primarily in those older than 19 years of age (Fig. 3). Within 2 years, the outbreak spread slowly to all regions of England, Wales, Scotland, and Northern Ireland and reached an epidemic peak in the first 6 months of 2005 (Fig. 1).

We estimate that more than 100,000 people developed mumps nationwide during the UK epidemic from 2003 through 2006, with more than 70,000 cases nationwide in 2005. This estimate is somewhat higher than the 56,395 cases that were reported for 2005 in England and Wales by the UK Centers for Disease Control and Prevention.3

The current study provides compelling evidence that the virulent 4-year UK epidemic of mumps in 2003–2006 originated in the Cardiff/Newport postal zone of Wales in the first 6 months of 2003, primarily in children 10–17 years of age, followed by a more robust outbreak that spread to 2 practices in the adjacent zone of Swansea in the last 3 months of 2003 among those aged 10–24. The overall cumulative incidence of mumps in Wales was more than twice the average of the other 10 regions.

The epidemic initially spread from Wales, northeastward to the Bolton zone, and further north to Glasgow in the west of Scotland beginning in the first quarter of 2004. It spread to Wigan, adjacent to Bolton, in the third quarter of 2004. A large outbreak occurred in 3 of 9 practices around Coventry to the east starting in the fourth quarter of 2004.

By the first quarter of 2005, the epidemic had spread to one practice in Liverpool and one in Warrington zones. The largest outbreak in the UK occurred in 5 of 11 practices in Gloucester in the first quarter of 2005. A small outbreak in the Trent region to the northeast occurred in Stockport in the first quarter of 2005, primarily in 18–24 year olds.

The incidence of mumps was lowest in London and the east of England and Scotland, where it started in the latter part of 2005 and continued into 2006. A small outbreak was recorded in one practice in Dundee in the fourth quarter of 2005 and in Edinburgh in the first quarter of 2006.

The epidemic was characterized by clusters of cases primarily in certain urban areas, such as Cardiff/Newport, Swansea, Coventry, and Gloucester. We do not have further detail on the precise location of the general practices that recorded the data, but it seems reasonable to conclude that the outbreaks occurred primarily in boarding schools, colleges and universities, military bases, and prisons, as well as other locations that have previously been frequently reported to be the source of localized outbreaks of mumps during this outbreak.2,14–17 A previously described yearly third-quarter decrease in the number of cases coinciding with the annual summer school holiday was also demonstrated in these data.3 Another consequence of the epidemic in the UK was an outbreak of 39 cases of mumps in Crete during the summer of 2004 and the spring of 2005. Thirty-eight of the cases were young adults from the UK on holiday in Crete, and who were hospitalized shortly after they arrived.18

In 2005, Bloom and Wharton reviewed the experience with MMR vaccine and the risk of mumps in the US in the context of the major UK epidemic of mumps that started in 2003.19 Based on the UK experience, they raised the question as to whether the US was at risk for a mumps outbreak. The answer came in 2006 when 6584 cases nationwide were reported in the US, primarily in college students age 18–24 in Midwest states.12 The US outbreak lasted only a year and the number of reported cases was far lower than the number of cases in the UK epidemic. This may be at least in part explained by the fact that the initial mumps vaccination campaign in the US was started more than 10 years before that in the UK and was more comprehensive.

The evaluation of the effectiveness of the MMR vaccine in the UK over the 17 years since it was introduced in October 1988 is complex. First, the majority of children initially vaccinated were below the age of 10 years. Second, a large update program of the vaccination was initiated in 1996–1997, primarily in children age 3–9 years. Finally, the outbreak in 2003 led to a 2004–2006 campaign to provide MMR vaccination particularly to young adults. Children and young adults who were at risk for mumps during the 2003–2006 epidemic could have been vaccinated during the early 1990s, the mid-1990s, or after 2003. It may be noted that the proportion of children age 1–2 years who were routinely vaccinated with MMR in the earlier years fell about 25% after 1998 in association with the widely publicized purported association between the vaccine and autism.6,11

To explore the effectiveness of the MMR vaccine we conducted a case-control evaluation that controlled for age, sex, general practice, and date of diagnosis of mumps in the cases. This design provided for considerable control of the timing of the various vaccination programs. We do not have direct information on socioeconomic status and therefore we could not match on socioeconomic status, which may be associated with other variables that could increase the risk of developing mumps (ie, over-crowding, lack of hygiene). However, due to the nationalized healthcare system in the UK, all patients in the Database should have access to a GP and the opportunity to be vaccinated regardless of their socioeconomic status. Overall, our results indicate that the MMR vaccine provided substantial protection against mumps for each age group up to 24 years old. Two doses consistently provided more protection than one (Table).

Cohen et al20 evaluated vaccine effectiveness for a recent outbreak in 312 cases of mumps in children eligible to have received 2 doses of MMR vaccine. They estimated the vaccine effectiveness was 88% for one dose and 95% for 2 doses in 2 year olds, but the estimate of effectiveness for one or 2 doses fell to 66% in 11–12 year olds. In the current study, substantial protection against mumps for all ages up to age 24 was consistent with that reported by Cohen et al in younger children.

In summary, we have provided documentation that an epidemic of mumps occurred in the UK during 2003–2006. It started in Wales in the first 6 months of 2003 and gradually spread to the north and west of England and Scotland and then to the Midlands, finally spreading to the east in the last 6 months of 2005. MMR vaccination provided substantial protection. A reduction in the number of children who were vaccinated with MMR during the late 1990s as a result of concern about its possible connection with autism is likely to have increased the number of cases of mumps in 2003–2006 and contributed to the magnitude of the epidemic.

REFERENCES

1. Gupta RK, Best J, MacMahon E. Mumps and the UK epidemic 2005. BMJ. 2005;330:1132–1135.
2. Savage E, Ramsay M, White J, et al. Mumps outbreaks across England and Wales in 2004: observational study. BMJ. 2005;330:1119–1120.
3. Centers for Disease Control and Prevention. Mumps epidemic-United Kingdom, 2004–2005. Morb Mortal Wkly Rep. 2006;55:173–175.
4. Jick H, Terris BZ, Derby LE, Jick SS. Further validation of information recorded on a general practitioner based computerized data resource in the UK. Pharmacoepidemiol Drug Safety. 1992;1:347–349.
5. Jick SS, Kaye JA, Vasilakis-Scaramozza C, et al. Validity of the general practice research database. Pharmacotherapy. 2003;23:686–689.
6. Kaye JA, Melero-Montes MM, Jick H. Mumps, measles, and rubella vaccine and the incidence of autism recorded by general practitioners: a time-trend analysis. BMJ. 2001;322:460–463.
7. Meier CR, Napolkov PN, Wegmueller Y, Jefferson T, Jick H. Population-based study on the incidence, risk factors, clinical complications, and drug utilization associated with influenza in the United Kingdom. Europ J Clin Microbiol & Infect Dis. 2000;19:834–42.
8. Kaye JA, Derby LE, Melero-Montes MM, Quinn M, Jick H. The incidence of breast cancer in the General Practice Research Database compared with national cancer registration data. Br J Cancer. 2000;83:1556–1558.
9. Hollowell J. General Practice Research Database (GPRD): Scope and Quality of Data. London: Office of Population Censuses and Statistics; 1994:1–20.
10. Communicable Disease Report Weekly. Changes in laboratory testing as the increase in mumps cases in England and Wales continues. CDR Wkly Rep. 2005;15:2–3.
11. Godlee F. Think mumps. BMJ. 2005;330:0.
12. Dayan GH, Quinlisk MP, Parker AA, et al. Recent resurgence of mumps in the United States. N Engl J Med. 2008;358:1580–1589.
13. Dobson R. Mumps cases rise among teenagers and young adults. BMJ. 2004;329:132.
14. Mackenzie DG, Hallam N, Stevenson J. Younger teenagers are also at risk of mumps outbreaks. BMJ. 2005;330:1509.
15. Dobson R. Mumps cases on the risk in England and Wales. BMJ. 2005;330:324.
16. Mackenzie DG, Craig G, Moore J. Mumps in a boarding school: description of an outbreak and control measures. Br J Gen Pract. 2006;56:526–529.
17. Donaghy M, Cameron JC, Friederichs V. Increasing incidence of mumps in Scotland: options for reducing transmission. J Clin Virol. 2006;35:121–129.
18. Spanaki A, Hajiioannou J, Varkarakis G, Antonakis T, Krymizakis DE. Mumps epidemic among young British citizens on the island of Crete. Infection. 2007;35:104–106.
19. Bloom S, Wharton M. Mumps outbreak among young adults in UK. Is the US at risk? BMJ USA. 2005;5:E363–E364.
20. Cohen C, White JM, Savage EJ, et al. Vaccine effectiveness estimates, 2004–2005 mumps outbreak, England. Emerg Infect Dis. 2007;13:12–17.

Supplemental Digital Content

© 2009 Lippincott Williams & Wilkins, Inc.