Full text of DOI:10.1111/j.1467-842X.2001.tb00318.x

Article
Controlling Infection
A serosurvey evaluation of the school-based
measles ‘catch-up’ immunisation
campaign in Victorian school-aged children
Michaela A. Riddell
Abstract
Victorian Infectious Diseases Reference Laboratory and Department of Paediatrics,
University of Melbourne, Victoria
Objective: To determine the proportion of
Victorian primary school students protected
against measles infection one year after the
completion of the measles ‘catch-up’
immunisation campaign of 1998 and to
compare this with the proportion of year 9
and 10 (aged 14-16 years) students.
Design & setting: Three-stage random
cluster survey in Victorian primary and
secondary schools.
Jennie A. Leydon, Antony Ugoni and Heath A. Kelly
Victorian Infectious Diseases Reference Laboratory
n 1968 the live attenuated measles
We aimed to determine the proportion of
vaccine was licensed in Australia. primary school students who were protected
I
However, the vaccine was not included
against measles infection one year after the
completion of the campaign. We compared
this with the proportion of year 9 and 10 sec-
Main outcome measures: Proportion of
primary and year 9 and 10 secondary
school students protected against measles
infection one year after the completion of
the mass ‘catch-up’ immunisation
in the national immunisation schedule for
infants until 1975.1 The introduction of this
,2
vaccine, and subsequent revisions to the ondary school students (aged 14-16 years)
immunisation schedule, reduced the circu- protected against measles, as these students
lation of wild type virus and correspondingly were not specifically targeted by the vaccina-
campaign. Secondary outcomes: the
proportion of both primary and year 9 and
10 secondary school students protected
against both mumps and rubella.
reduced the incidence of measles-related ill-
tion campaign. As a secondary outcome we
assessed the susceptibility in both groups to
3
ness. However, inadequate vaccine uptake
allowed measles transmission to continue mumps and rubella virus infection.
throughout Australia.4
Results: Of 1,037 Victorian primary and
2
,357 years 9 and 10 secondary school
students invited to participate in this study,
03 (39%) and 752 (32%) respectively
In 1988, the first national measles campaign
was conducted. Vaccine coverage, however, Methods
failed to achieve sufficient protection against
4
Ethics approval for the study was obtained
from the Department of Human Services
(Victoria) Ethics Committee. Before recruit-
ment of schools commenced, permission to
contact school principals directly was sought
from the Department of Education (School
Community Support Branch), the Catholic
Education Office and the Association of
Independent Schools Victoria.
provided a blood specimen for serological
testing for antibodies against measles,
mumps and rubella. 94.8% (95%
measles infection in the community and in
1993/94 a number of Australian States faced
major measles epidemics, resulting in two re-
ported deaths.5-8 In 1998 the Commonwealth
Department of Health and Family Services
implemented the Measles Control Campaign
confidence interval, 91.5, 96.9) of primary
school and 93.1% (90.9, 94.8) of year 9
and 10 students were protected against
measles infection.
Conclusion: One year after the completion
of the school-based measles ‘catch-up’
immunisation campaign the level of
protection in Victorian primary school aged
students is sufficient to prevent the
continuing circulation of measles virus
within this age group. The proportion of year
(MCC). The cornerstone of the MCC was a
school-based national ‘catch-up’ vaccination
program of all primary school aged students
with the measles, mumps and rubella (MMR) Study design and sample
10
vaccine.
Victoria had the highest participation rate
in the school-based campaign, with more
We used a three-stage cluster sample and
based our sample size estimates on an ex-
pected 95% of primary school students and
9
and 10 secondary school students
than 84% of all eligible children vaccinated 85% of secondary school aged students be-
protected against measles is also probably
sufficient to prevent continuing circulation of
wild type virus in Victoria, even though this
age group was not specifically targeted by
the ‘catch-up’campaign.
ing protected from measles infection.11 We
estimated that 264 primary and 707 year 9
and 10 students were needed to provide a
sample of blood for the measurement of
at school. Immunisation providers, other than
the school campaign, accounted for a
further 3% of vaccinations recorded for Vic-
torian children.9
(Aust N Z J Public Health 2001; 25: 529-33)
Correspondence to:
Submitted: March 2001
M. Riddell, Epidemiology Division, Victorian Infectious Diseases Reference Laboratory, Locked
Bag 815, Carlton South 3053. Fax: (03) 9342 2665; e-mail: michaela.riddell@mh.org.au
Revision requested: August 2001
Accepted: September 2001
2
001 VOL. 25 NO. 6
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
529

Riddell et al.
Article
measles-specific immunoglobulin class G (IgG).This calculation
incorporated a design effect of 1.3 to allow for cluster sampling,10
based on similarly designed surveys of immunisation coverage12,13
to obtain a confidence interval (CI) around our estimates of the
proportion of children protected from measles infection of ±3%.
Victorian Local Government Areas (LGAs) were used as the
primary sampling unit for analysis since the ‘catch-up’ campaign
in Victorian schools utilised existing immunisation services pro-
vided by LGAs. For the first stage, 22 of 78 LGAs were selected,
with probability of selection proportional to the size of the pri-
mary school population in the LGA.
manufacturer’s instructions. Mumps-specific IgG was determined
using the Enzygnost anti-parotitis-virus IgG immunoassay (Dade
Behring, Marburg, Germany) in accordance with the manufac-
turer’s instructions. Rubella-specific IgG was determined using
the BeckmanAccess immunoassay system (Beckman Instruments,
Chaska, MN, USA). For all three test antigens, initially equivocal
specimens were retested. Students were considered protected from
measles if their measles IgG was reported as positive. Students
with an equivocal measles result were considered susceptible to
modified measles. Protection from mumps was based on either a
positive or equivocal test result as recommended by the manufac-
turer and antibody titres were not calculated. Students were con-
sidered protected from rubella virus infection if the
rubella-specific IgG concentration, calculated from the standard
curve of the internal standard of the assay, was determined to be
>15 IU/mL.14 Results were reported back to the parents or
guardian(s) of students who provided a specimen with a recom-
mendation for further vaccination if appropriate.
The second stage required a complete list of all schools, both
primary and secondary, from all selected LGAs and subsequent
selection of one primary and one secondary school (with probabil-
ity of selection proportional to size of the whole school popula-
tion). However, if an insufficient number of children agreed to
participate in one school, the next randomly selected school in the
same LGA was invited to participate, in addition to the first school.
A third sampling stage was required for the primary school sam-
ple since the level of immunisation coverage during the school-
based ‘catch-up’ campaign increased with age (Virginia Kaltzis,
Department Human Services, Victoria, personal communication).
To ensure equal representation of the seven primary year levels
throughout the 22 LGAs, each year level was randomly assigned to
three LGAs. In some schools composite grades and, in one case, a
small school population made it difficult to recruit from only one
grade level. From each LGA, a minimum of 12 primary and 33
year 9 and 10 students were required to provide a blood specimen.
Data analysis
Data were analysed using the survey commands (svymean,
svytab) of Stata 6.0 with LGA as the primary sampling unit.15
This analysis allows adjustment of the standard errors based on
the sampling strategy and the confidence intervals reflect the
effects of clustering in the study population.
Results
Participation rates
Recruitment
Forty-seven primary and 40 secondary schools were approached
to successfully recruit 25 (53%) primary and 25 (62.5%) second-
ary schools. Univariate analysis revealed no significant associa-
tion of region (metro/non metro) or school governing body
(government/non-government religious/non-government other
independent) with school participation.
Schools were invited to participate by mail and by follow-up
telephone call. Visits were made to schools that agreed to partici-
pate. Information seminars were given to the year 9 and 10 stu-
dents and were offered to the parents of primary school students.
Individual information packs containing a detailed information sheet
for the parents, a consent form and a short questionnaire were given
to each student. Information packs and questionnaires were avail-
able in English, Italian, Greek and Vietnamese. The questionnaire
collected information on age, sex, language spoken at home, coun-
try of birth of student and parents, disease and vaccination history.
Vaccination details were considered to be valid if a complete date
Participating primary schools reported 82.8% of their students
vaccinated at school with MMR during the ‘catch-up’ campaign,
which was not significantly different from the reported rate for
9
Victoria (83.4%). However, primary schools who refused to par-
ticipate reported significantly lower MMR vaccination rates at
school during the ‘catch-up’ campaign relative to participating
schools (78.8%, Fisher’s exact p<0.001).
(dd/mm/yy) was recorded and invalid if a history of vaccination
was given without a date or the date was incomplete.
Student response rates for the study are detailed in Table 1.
Students were encouraged to return the questionnaire even if they
did not wish to provide a blood specimen. For analysis of
measles, mumps and rubella antibodies, 379 students aged 6-12
years (subsequently referred to as ‘primary’) and 739 students
aged 14-16 years (subsequently referred to as ‘secondary’) pro-
vided a blood specimen. A further 90 primary and 161 secondary
students returned the questionnaire but did not provide a blood
specimen. Excluded from the analysis were 23 students aged five
who provided a specimen but were not eligible for immunisation
at school during the 1998 ‘catch-up’ campaign and a further 14
students aged 13, 17, 18 and 19 who provided specimens.
Specimen collection and testing
An experienced paediatric phlebotomist collected up to 10 mL
of blood from each consenting student 30-45 minutes after treat-
ment of the venepuncture site with anaesthetic cream (Ameth-
ocaine gel 4% ‘Angel’ Royal Children’s Hospital Pharmacy,
Melbourne, Australia). Specimens collected at the school were
transported to the laboratory and stored at 4ºC within 48 hours.
Serum was tested for measles specific IgG at VIDRL using
an Enzygnost anti-measles-virus/IgG enzyme immunoassay
(
Dade Behring, Marburg, Germany) in accordance with the
530
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
2001 VOL. 25 NO. 6

Controlling Infection
School-based measles ‘catch-up’immunisation campaign
Table 1: Student participant response rates.
63% reported vaccination with MMR in school vaccination pro-
grams prior to 1997.
Number
Number
Number who
provided
students questionnaires
invited
returned
% invited)
blood specimen
(% quest.
Prevalence of antibodies to measles, mumps
and rubella
The empirical design effect for this study with respect to
(
returned)
5
-13 year olds
1,037
502 (48%)
913 (39%)
403 (80.3%)
752 (82.4%)
seroprevalence of measles IgG antibodies in primary school stu-
dents was 1.24.10
13-19 year olds 2,357
Figure 1 shows the immunoassay result for measles IgG by year
of age with 95% CI for the positive test results. There was no
statistically significant difference in the proportion of students
positive for measles IgG for each year of age. When considering
protection against clinical measles infection by age, the point es-
timates indicated more than 92% of students at each age, except
nine-year-olds, were protected. However, since the sample number
for each age group is small, the 95% CI are wide. There were no
statistically significant differences in the geometric mean titre
Primary school students who returned the questionnaire but
did not provide a specimen were not significantly different from
those students who provided a specimen with respect to school
region, school governing body, age, vaccination history, language
spoken at home and country of birth. The age and sex distribu-
tion of primary students who completed the questionnaire and
gave a specimen was not significantly different from those pri-
mary students who completed the questionnaire but did not give
a specimen. This was not true for female secondary school stu-
dents, however, due to a much higher proportion of 14, compared
to 15 and 16-year-old girls who returned the questionnaire and
provided a sample for testing (p=0.008, data not shown). There
was no significant difference in the proportion of students who
provided a specimen, compared with those who completed the
questionnaire but did not provide a specimen, based on their re-
ported vaccination status or history of illness from measles, mumps
or rubella.
(
mIU/mL) by age despite the majority of secondary students
receiving a second dose of MMR more than two years prior to the
serosurvey (data not shown).
Table 2 shows the proportion of students protected against
measles, mumps and rubella and the difference in the proportions
for each infection by sex and school level (primary vs. second-
ary). While it appears that higher proportions of females were
protected against measles, mumps and rubella, these differences
were not statistically significant. Higher proportions of primary
school students were protected against measles, mumps and ru-
bella than were secondary school students, but this difference was
only statistically significant when considering the proportion of
students protected against mumps infection. There was no statis-
tically significant difference in the proportion of students pro-
tected against measles, mumps or rubella when considering region
(metro/non-metro) and school governing body. Nor was there any
significant difference in the proportion of students protected
against measles or rubella, who were born overseas, compared
Vaccination providers
Of 297 primary students (63% of sample) who nominated the
provider of their second MMR, 205 (69%) reported vaccination
at school during the 1998 school campaign, 35 (11.8%) reported
vaccination at local council immunisation services and 49 (16.5%)
reported vaccination at their GP or hospital clinic. Eight students
reported vaccination in ‘other school’ campaigns. Fewer than 5%
of secondary students reported vaccination in 1998 or 1999, while
Figure 1: Measles IgG
5
7
37
50
68
72
52
43
21 1
37 7
15 1
S a m ple s ize (n)
0 0%
1
immunoassay result
95% CI for positive
(
9
8
7
6
5
4
3
2
1
0 %
0 %
0 %
0 %
0 %
0 %
0 %
0 %
0 %
result) in Victorian
school students, by
age.
N e g a t i v e
E q u iv o c a l
P o s it iv e
0
%
6
7
8
9
10
11
12
14
15
16
P o s it iv e
( 9 5 % C I)
A ge (year )
2
001 VOL. 25 NO. 6
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
531

Riddell et al.
Article
Table 2: Proportion of students protected against measles, mumps and rubella and difference in proportions by sex
and school level.
n
Measles
Mumps
Rubella
%
(95% CI)
% (95% CI)
% (95% CI)
Primary (6-12yo)
Male
Female
All
189
190
379
94.2 (89.3–96.9)
95.3 (91.3–97.5)
94.7 (91.3–96.8)
89.9 (84.2–93.8)
94.2 (90.5–96.5)
92.1 (88.9–94.4)
97.9 (94.6–99.2)
99.5 (96.5–99.9)
98.7 (96.5–99.5)
Difference in proportion
Secondary (14-16yo)
Female vs. Male
1.1% (-3.3–5.5)
4.3% (-0.9–9.5)
1.6% (-0.2–3.4)
Male
Female
All
330
409
739
92.7 (89.2–95.2)
93.2 (90.7–95.0)
93.0 (90.8–94.7)
84.2 (79.7–87.9)
88.5 (85.0–91.3)
86.6 (83.6–89.1)
97.6 (95.3–98.8)
99.0 (97.6–99.6)
98.4 (97.2–99.0)
Difference in proportion
Difference in proportion
Female vs. Male
0.4% (-3.0–3.8)
1.8% (-1.3–4.8)
4.3% (-0.6–9.2)
5.5% (1.8–9.2)
1.4% (-0.7–3.5)
0.3% (-1.2–1.8)
Primary vs.
Secondary
with Australian-born students. However, the proportions of
overseas-born primary (79.2%) and secondary (75.7%) school-
aged students protected against mumps were significantly less
than the proportions of Australian-born students (92.9% primary
p=0.0005, 87.7% secondary p=0.003).
three-stage randomised cluster survey ofVictorian primary and sec-
ondary school students collected one year after the completion of
the 1998 school-based measles ‘catch-up’vaccination campaign.
The level of immunity required in a community to prevent trans-
mission of measles virus is thought to be between 92-95%.16
A
Victorian study conducted in 1994 found approximately 84% of
Reported immunisation and presence of antibodies
Table 3 shows the proportion of students protected against
measles, mumps and rubella according to self-report of a sec-
ond dose of MMR. The proportion of students protected against
measles was not significantly different, whether the parental
report of immunisation was valid (full date provided), invalid
year 2 (7-8 year olds) and 86% of year 7 (12-13 year olds) stu-
11
dents were protected against measles. These data suggested that
prior to the ‘catch-up’campaign measles immunisation coverage
of Victorian school students might have been insufficient to pre-
vent circulation of wild-type measles virus. However, our point
estimate of 94.7% students protected indicates an adequate level
of immunity has been achieved among primary school students
following the immunisation program.
(
incomplete or no date) or no/unsure. When assessing mumps
protection, a validated report of MMR was significantly more
likely to indicate protection. When considering rubella protec-
tion, parental report of uncertain or no vaccination was signifi-
cantly more likely to indicate susceptibility, but there was no
significant difference in the proportion of students protected
against rubella when comparing students with a valid or invalid
report of vaccination.
The National Centre for Immunisation Research and Surveil-
lance (NCIRS) has evaluated population immunity in Australia
using residual sera collected from diagnostic laboratories na-
9
tionwide before and after the immunisation campaign. The
NCIRS estimated that 94.4% (95% CI, 90.9-96.8) of 267 Victo-
rian primary school children were protected from measles in-
fection (unpublished data, NCIRS), consistent with our results
of 94.7% (91.3-96.8) of 379 students, despite the different
sampling frames of the two studies. We found that 93.0% (95%
CI, 90.8-94.7) of 739 secondary school students were protected
against measles infection. This is again consistent with the
Discussion
Seroprevalence
We report here the seroprevalence of measles antibodies from a
Table 3: Proportion of Victorian primary and secondary school students protected against measles, mumps and rubella
according to reported vaccination with second dose of MMR.
Report of immunisation
with second dose MMR
Number of students
reporting immunisation
with second dose MMR (n)
Number (%) students protected
Measles
n (%)
Mumps
n (%)
Rubella
n (%)
Yes (valid date)a
Yes (invalid date)b
No/Unsure
p-valuec
436
346
230
411 (94.3%)
325 (93.9%)
213 (92.6%)
0.62
405 (92.9%)
299 (86.4%)
200 (87.0%)
0.02
435 (99.8%)
342 (98.8%)
220 (95.7%)
0.0003
Notes:
(
(
(
a) Day/month/year provided.
b) Date incomplete or no date provided.
c) Comparison of all three report categories.
532
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
2001 VOL. 25 NO. 6

Controlling Infection
School-based measles ‘catch-up’immunisation campaign
results from the national evaluation where 91.7% (95% CI, 84.9-
in study design. Thanks to Debbie Gercovich for specimen col-
lection, Graham Byrnes for statistical advice and helpful discus-
sions, Pamela Christofis, Trang Vu and Renato Raimondi for
translation of information packs, Chrishni Karunakaran, Rosa Billi
and staff at VIDRL for technical assistance and sample process-
ing. Heather Gidding, NCIRS, kindly providedVictorian data from
the national evaluation. The authors thank the principals, staff,
parents and students of the schools that participated in the study.
Financial support: MR is supported by an NHMRC Public
Health Research Scholarship. Gifts to children who provided a
specimen were donated by BioMerieux Vitek, Crown Scientific,
Interpath Services, Medos Company, Pan Bio Pty Ltd, Proscience,
Selby Biolab andTrace Scientific. Dade Behring Diagnostics and
Beckman Coulter provided the enzyme immunoassays used for
this study to VIDRL at a reduced cost.
9
6.2) of 109 Victorian students of the same age as those in our
study were found to be protected from measles infection (un-
published data, NCIRS).
Participation bias
The design employed in this study is subject to participation
bias. In our sample we believe this bias has occurred at two stages.
First, there is bias at the school level with respect to vaccination
rates at school during the ‘catch-up’ campaign. Schools that did
not participate in this study were less likely to appreciate the
social value and benefits of a study such as ours.17 This attitude
may have extended to participation in the ‘catch-up’ campaign,
explaining the lower proportion of children vaccinated at school
during the ‘catch-up’ campaign in non-participating schools.
Second, there is evidence to suggest that our sample may be
more likely to contain students who received an MMR vaccine.
Over 28% of the primary school students who nominated a vac-
cination provider indicated they had been vaccinated outside the
school-based campaign, a much higher proportion than the 3%
vaccinated by other providers, reported for the whole of Victoria,
References
1
.
Lambert SB, Morgan ML, Riddell MA, Andrews RM, et al. Measles outbreak
in young adults in Victoria, 1999. Med J Aust 2000;173:467-71.
Gidding HF, Burgess MA, Kempe AE. A short history of vaccination in Aus-
tralia. Med J Aust 2001;174(1):37-40.
2
.
9
during the campaign. Our overall response rate indicates that
3. Tobin S, Kelly H. Measles encephalitis in Victoria, 1962-96: Down but not
out [letter]. Aust N Z J Public Health 1999;23(4):443.
less than 50% of the school population approached was repre-
sented in this sample. It would appear that this sample contains a
high proportion of children of motivated parents who obtained
vaccinations outside the school-based campaign and who thus
may have been more likely to be protected against measles.
Despite this acknowledged bias, our estimate of the proportions
of both primary and secondary school students protected against
measles infection is not significantly different from the estimates
for Victorian students obtained from the national evaluation.
4
5
.
.
Hanna JN, Kass RB. Immunization status of Aboriginal children in Central
Australia. Med J Aust 1985;143:S56-S57.
Lambert S. Measles in Victoria 1992 to 1996: The importance of laboratory
confirmation. Commun Dis Intell 1998;22(2):17-22.
6. Lush D, Maloney M, Merianos A. Measles Outbreak in the Alice Springs
region, Northern Territory, June to October 1994. Commun Dis Intell
1
994;18(23):597-8.
7. Jeremijenko AM, Kelly H, Patel M. The high morbidity associated with a
measles outbreak in a west Australian town. J Paediatr Child Health
1
996;32(5):382-5.
8. Herceg A, Passaris I, Mead C. An outbreak of measles in a highly immunised
population: Immunisation status and vaccine efficacy. Aust J Public Health
1
994;18(3):249-52.
9
1
.
National Centre for Immunisation Research and Surveillance of Vaccine Pre-
ventable Diseases. Australian Measles Control Campaign 1998: Evaluation
Report. Sydney: University of Sydney, Royal Alexandra Hospital for Chil-
dren, 1999 June 30. http://www.health.gov.au/pubhlth/immunise/metadata/
measeval.htm
Validity of parental report of immunisation
Another interesting finding in our study is that, at high levels
of community-wide protection and vaccination coverage, as would
occur following a successful mass immunisation campaign, the
validation of parental report of immunisation records may not be
necessary. This is in contrast to the current practice of only
accepting validated parental records as evidence of immunisa-
tion.18 For measles and rubella, even those participants who said
they were not vaccinated or unsure of vaccination status were
estimated to be sufficiently immune to prevent circulation of wild-
type virus. For mumps, however, this was not the case.
0. Carlin JB, Hocking J. Design of cross-sectional surveys using cluster sam-
pling: An overview with Australian case studies. Aust N Z J Public Health
1
999;23(5):546-51.
11. Lester R, Hogg G, Murphy M. Prevalence of antibody to measles and diph-
theria in Victorian school children. Abstract. In: 5th National Immunisation
Conference; 1996; Sydney, New South Wales.
12. Herceg A, Daley C, Schubert P, Hall R, et al. A population-based survey of
immunisation coverage in two-year-old children. Aust J Public Health
1
995;19(5):465-70.
13. Bond L, Nolan T, Lester R. Immunisation uptake, services required and gov-
ernment incentives for users of formal day care. Aust N Z J Public Health
1
999;23(4):368-76.
Our results confirm that, one year after the measles ‘catch-up’
immunisation campaign, theVictorian primary school-aged com-
munity is protected from measles infection. The same is true of
protection against rubella infection, although protection against
mumps infection is less certain.
14. Banatvala JE, Best JM, O’Shea S, Dudgeon JA. Persistence of rubella anti-
bodies after vaccination: Detection after experimental challenge. Rev Infect
Dis 1985;7 Suppl 1:S86-90.
15. StataCorp: statistical software [computer program]. Version 6.0. College
Station(TX): Stata Corporation, 1999.
1
6. Anderson RM, May RM. Immunisation and herd immunity. Lancet
990;335(8690):641-5.
1
1
7. Pound B, Riddell M, Byrnes G, Kelly H. Perception of social value predicts
participation in school-based research. Aust N Z J Public Health
2
000;24(5):543-5.
Acknowledgements
1
8. Hawe P, Wilson A, Fahey P, Field P, et al. The validity of parental report of
vaccination as a measure of a child’s measles immunisation status. Med J
Aust 1991;155(10):681,684-6.
We are grateful to Professor Terry Nolan for helpful discussion
in study design, analysis and critical reading of the manuscript.
We thank Dr Rosemary Lester, Ross Andrews, Stephen Pellisier,
Dr Mike Catton and Dr Stephen Lambert for helpful discussion
2
001 VOL. 25 NO. 6
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
533