Tropical Medicine and International Health

doi:10.1111/j.1365-3156.2008.02214.x

volume 14 no 2 pp 167173 february 2009

Immune response to measles vaccine in 6 month old infants in

Papua New Guinea
Jonah Kurubi1, John Vince1, Paulus Ripa1, Nakapi Tefuarani1, Michaela Riddell2 and Trevor Duke1,3
1 School of Medicine, University of Papua New Guinea, Boroko, Port Moresby, Papua New Guinea
2 Victoria Infectious Disease Reference Laboratory, Melbourne, Vic., Australia
3 Centre for International Child Health, University of Melbourne, MCRI, Melbourne, Vic., Australia

Summary

objective To assess the efficacy of the current measles immunization schedule in Papua New Guinea,
which is to give the first dose at 6 months of age and the second at 9 months.
methods Humoral immune response study of 140 Papua New Guinean infants at 6 months of age,
measuring measles IgG antibodies by enzyme immunoassay before and 85 days after the 6-month dose
of measles vaccine.
results After vaccination at 6 months, 35.7% of infants developed a level of measles antibodies
consistent with protection (IgG >330 IU ml); 17.7% had an antibody response (150330 IU ml) that is
likely to afford some protection; 46.8% had no detectable antibody response (IgG <150 IU ml). Among
53 infants with no antibody response, 37 (69.5%) developed an antibody response, while 42.4%
(37 87) of those with maternal antibodies sero-converted (P = 0.002).
conclusions Antibody response to measles vaccine was lower than expected at 6 months. While the
presence of maternally derived antibodies accounted for some of the limited seroconversion in young
infants, other factors are involved. Issues to be considered in determining the value of the first dose of
measles vaccination in mid infancy in poor countries are complex and antibody responses are only one
factor. Others, such as cell mediated immune responses, the non-specific protective effect of measles
vaccine in preventing illness and death and the practicalities of uptake of vaccines at different ages, are
also important.
keywords measles, immunization, vaccination schedule, infants, Papua New Guinea

Introduction
Measles vaccine was first introduced in Papua New Guinea
(PNG) in 1982 for infants at 9 months of age. In 1992,
because of concerns that infants younger than 9 months
were at the highest risk of dying from measles, a 6-month
dose was added to the schedule. Despite this, major
measles epidemics still occurred. The latest epidemic
started in 1998 and continued until late 2002 (Mgone et al.
2000). During that time more than 30 000 children
developed measles, and in 2001 alone 292 children died
from measles in health facilities in PNG (Duke 2003).
More than half of the hospitalized measles cases were
infants <9 months old and fatal cases were reported in
infants as young as 4 months (Mgone et al. 2000).
High vaccine coverage has been difficult to achieve and
sustain in PNG. In 2001, the national 9 month measles
vaccine coverage was 43%. In response to this PNG began
regular three to four yearly supplemental immunization,

2009 Blackwell Publishing Ltd

bringing coverage to over 80% of the population as at

9 months. This approach, with mass campaigns complimenting routine services, follows current WHO UNICEF
strategy for the global elimination of measles.
In industrialized countries the first dose of measles
vaccine is generally given at 12 months or 15 months and
the second dose at 4 years. The probabilities of adequate
antibody responses are >95% if the first dose is given at 12
or 15 months, and >99% after the second dose. PNG has a
different epidemiology of measles to industrialized countries, where measles affects children older than 1 year and
the disease is generally mild. In PNG measles has a high
attack rate and high mortality in infants (Mgone et al.
2000). Although the 6 month dose of measles vaccine in
PNG was introduced to protect infants younger than
9 months, there has been uncertainty about the efficacy
and effectiveness of such an approach. A small study in the
late 1980s using standard titre measles vaccine in infants
aged 67 months in the Southern Highlands of PNG
167

Tropical Medicine and International Health

volume 14 no 2 pp 167173 february 2009

J. Kurubi et al. Immune response to measles vaccine PNG

showed that none of 12 had residual maternal antibodies

and all 12 infants (100%, 95% CI 77100%) achieved
titers of measles specific IgG consistent with protection
(Rogers et al. 1991). Data from other countries are
variable. A study from Guinea in Africa found that 91% of
71 infants vaccinated against measles at 6 months had
seroconverted 1 year later (Kourouma et al. 1992) and a
study from Ghana showed 92% seroconversion in infants
vaccinated at 7 months (Sakatoku et al. 1994) However
other studies from USA (Johnson et al. 1994; Gans et al.
2001), Latin America (Anonymous 2005), India (Vidvashankar 2002), Haiti (Halsey et al. 1985) and South Africa
(Dick et al. 1975) have found much lower rates of
seroconversion in infants at 67 months. A large, recent
study from Malawi showed serocoversion of 5968% at
6 months, but that the 6 month dose had an important
priming effect on the seroconversion rate when a second
dose of measles vaccine was given at 9 months (Helfand
et al. 2008). We aimed to determine the serological
response of PNG infants at 6 months of age to measles
vaccine, to provide evidence for a review of the current
national schedule.

doubt whether they could return for the 9 month dose

received their first measles vaccine and were advised to
return in 3 months, or to visit the nearest functioning
health facility for the second dose.
At follow-up we recorded weight, clinical signs of
malnutrition and clinical history of measles during the
intervening period. A second blood sample was taken, after
which the second dose of measles vaccine was given.
Sample size calculation
We calculated that if 70% of 140 67 month old infants
seroconverted, the 95% CI for probability of seroconversion would be 6277%. We considered even the lower
limit of 62% would probably be sufficient to recommend a
continuation of the 6 month dose, as well as a dose in later
infancy. With coverage of measles vaccine in PNG being
6070% we assumed many mothers would not return for
second vaccination and follow-up serological testing, so we
aimed to reach a sample of 200 enrolled infants to ensure
that 140 infants had paired serological results.
Sample collection and specimen handling

Materials and methods

We assessed the humoral immune response to measles
vaccine among infants presenting for their first measles
vaccine at 6 months of age. The infants were recruited at
Port Moresby General Hospitals Well Baby Clinic
between May and December 2006. Infants were eligible for
inclusion if they were presenting for their 6 month vaccines
and were younger than 7 months, lived close to the health
facility and had a health book to document the vaccines
administered and other health events. Infants born prematurely were only recruited if the corrected age was
6 months. Infants were not recruited if they had previously
had an illness requiring hospital admission; if the parents
were in doubt whether they would return at 9 months; and
if parents declined to participate.
The demographic data recorded during the first contact
included the following from the infant: date of birth,
history of clinical measles and other infections, weight and
clinical signs of malnutrition. The following information
was recorded from the mothers: weight, height, highest
level of education, the familys source of income and the
type of housing the family was living in.
Initially, parents were asked to bring their child for the
second blood sample 1 month after the 6 month measles
vaccine. As this involved an additional visit, the time
interval was changed to 3 months, when the child was
brought for the second measles vaccine. Infants of parents
who declined to be enrolled in the study or who were in
168

Two blood samples were collected from each child; one

prior to the 6-month measles vaccine and one either at
around 1 month later or just prior to the 9 month vaccine.
The measles vaccine used was the Edmonston-Zagreb
strain. Mothers were requested to breastfeed their infants
23 min before 2 ml of venous blood were drawn from a
vein in the hand or the foot. A sterile technique was
followed to obtain all blood specimens.
The blood specimens were taken to the Port Moresby
Central Public Hospital Laboratory (CPHL) and centrifuged for 20 min. The serum was then pipetted into a
sterile tube, which was then labelled with a unique
identifying number. The specimens were stored at )20C.
On the day of transportation, serum specimens were
packed in three layers into the three carriers that had jelly
ice frozen at )20C. An express freight courier transported
the serum specimens from CPHL to Victoria Infectious
Disease Reference Laboratory in Melbourne for immunoassays.
Serum samples were evaluated for the presence of
measles IgG antibodies by commercial indirect enzyme
immunoassay (Enzygost, Dade Behring, Marburg, Germany) in accordance with the manufacturers instructions.
We defined protective immunity as a measles IgG level of
>330 mIU ml [equivalent to >0.2 optical density (OD)
value]; and an equivocal antibody response as 150
330 mIU ml (0.10.2 OD value). Specimens which tested
equivocal initially (OD value 0.10.2) were retested and

2009 Blackwell Publishing Ltd

Tropical Medicine and International Health

volume 14 no 2 pp 167173 february 2009

J. Kurubi et al. Immune response to measles vaccine PNG

the second test result recorded, according to standard

procedures in the reference laboratory. A negative response
to vaccination was defined as IgG level <150 mIU ml
(Chen et al. 1990; Tischer et al. 2007). For the purposes of
evaluating the effect of the presence of measles antibodies
prior to vaccination, we defined a trace of antibodies as an
OD which was >0 but <0.1 OD value.
Ethical approval and consent
Permission for this project was obtained from the Medical
Research Advisory Committee of PNG. Before enrolment
time was taken to explain to the parents the nature and
significance of the study, basic information about measles
and immunization, and the requirement to take two blood
samples. Consent forms used were written in Motu, Tok
Pisin and English; the three languages understood and
spoken by most people in PNG.
Statistical analysis
Stata, version 9 (Stata Corp Texas, College Station, TX,
USA) was used to analyze the data. Data are presented as
mean (SD) if normally distributed; or median (interquartile
range) in case of marked skewness of distribution. Chisquared tests were used to determine differences in
proportions. Logistic regression analysis was used test the
associated between seroconversion and demographic or
other factors.
Results
The 202 infants were recruited at presentation for their
6 month dose of measles vaccine and had a first blood
sample collected; 141 of these returned after the 6 month
vaccination dose and had a second blood sample collected.
The interval between first vaccination and repeat sampling
of blood for anti-measles IgG was a mean of 85 days (SD
20, range 21123).

The 141 infants were a mean age of 189 (SD 11) days.
The mean weight was 7.7 (SD 1.1) kg, and 8 (5.7%) had
moderate or severe malnutrition; 71 (50%) were male. One
infant (who had a protective level of measles IgG after
immunization) had an inadequate pre-vaccination volume
of serum obtained, so was excluded from further analyses.
Prior to vaccination no infant had more than
150 mIU ml antibodies, but 87 infants had a trace of
measles antibodies (Table 1 and Figure 1). After vaccination at 6 months of age 50 141 (35.7%) infants had a level
of measles IgG >330 mIU ml. An additional 25 (17.7%)
infants had a measles IgG level between 150 and
330 mIU ml; 66 (46.8%) had no detectable antibody
response (IgG <150 mIU ml). Among 53 infants with no
antibody response, 37 (69.5%) developed an anitbody
response, while 42.4% (37 87) of those with maternal
antibodies sero-converted (chi-square P = 0.002)
(Table 1). None of the infants had a history of measles or
suspected measles prior to or throughout the study period;
26 of 91 (29%) who failed to seroconvert did so despite
having no pre-existing serum antibodies.
Using logistic regression analysis, factors that were
independently associated with seroresponse (>150 IU ml)
after vaccination were the absence of pre-existing antibodies in serum at 6 months (odds ratio for seroconversion
if maternal antibodies present: 0.41, 95% CI 0.190.87,
P = 0.021), and a shorter time interval between vaccination and the second serum IgG estimation (odds ratio for
seroconversion if sample taken within 60 days of measles
vaccine being given: 4.1, 95% CI 1.0415.00, P = 0.043)
(Figure 2).
Mothers of participating infants came from each of the
20 provinces in PNG. The probability of seroconversion
varied markedly between the province of origin (from 14%
to 80%, although the numbers of mothers from some
provinces were small), but this was not significant
(KruskallWallis Chi-square P = 0.169).
No other factors recorded, including infants anthropomorphic characteristics, age or maternal physical or

Table 1 Relationship between presence of any maternal antibodies and lack of antibody response

Antibodies present in
infants serum prior to
first vaccination
No antibodies present prior to first vaccination
Trace (OD >0, <0.1) of antibodies present prior
to first vaccination
Total

2009 Blackwell Publishing Ltd

No detectable
response to
6 month dose
(IgG <150 IU ml)

Equivocal
antibody
response
(150330 IU)

Definite
protective
response to
6 month dose
(IgG >330 IU ml)

16
50

10
15

27
22

53
87

66

25

49

140

Total

169

Tropical Medicine and International Health

volume 14 no 2 pp 167173 february 2009

3000
2000
1000
0

Measles lgG response after 6 month vaccine (IU/ml)

J. Kurubi et al. Immune response to measles vaccine PNG

0.05
0.1
0.15
0.2
Measles antibodies in infant serum before first vaccination (optical density)

Figure 1 The relationship of any antibodies present in the serum to response to

vaccination.

Measles lgG response after 6 month vaccine (IU/ml)

0
1000
2000
3000

20

40
60
80
100
Time between vaccination and sampling (days)

sociodemographic factors, were significantly associated

with antibody response.
Discussion
Since 1992 PNG infants have been given measles vaccine in
the two doses schedules at 6 and 9 months. This study
found that 52% of Papua New Guinean infants at
6 months of age had an antibody response to measles
170

120

Figure 2 Time from first vaccination and

measles antibody response.

vaccine, and only 36% developed humeral immunity

consistent with well accepted definitions of protection
against infection (Tischer et al. 2007) However this finding
bares scrutiny. We used a conservative definition of
humoral protection. Other authors have described antibody levels as low as 125 mIU ml as indicting protection.
There is no clear consensus as to what constitutes a
protective level, and this probably varies according to the
level of exposure. In a study from Senegal 50% of children

2009 Blackwell Publishing Ltd

Tropical Medicine and International Health

volume 14 no 2 pp 167173 february 2009

J. Kurubi et al. Immune response to measles vaccine PNG

with levels below 125 mIU ml were protected (Samb et al.

1995) Furthermore, just because a child develops measles
after vaccination this does not mean that the vaccination
failed to provide any protection. The same study found
that infants with antibody responses <125 mIU ml after
vaccination who developed measles had mild disease, and
studies from Guinea Bissau have shown considerable
protection against measles when measles vaccine is given in
young infancy. Only an efficacy trial can clearly answer if
the vaccine gives protection. Given this, it is likely that at
least 50% of the infants vaccinated were protected against
measles.
Although we do not know what proportion of infants
given their first measles vaccine at 9 months of age develop
a protective antibody response, a study large recently from
Malawi showed higher seroconversion rates at 12 months
after receiving a two-dose schedule of 6 and 9 months,
than infants who received their first dose at 9 months
(Helfand et al. 2008). This study did not report the measles
antibody titre that they defined as positive, but reclassified
indeterminate results as positive after validation with some
samples using plaque neutralization assays. Therefore their
6-month seroconversion rate of 5968% (the lower range
occurring in HIV-infected infants) is a little higher that
what we observed in PNG. However their definitions of
protection may have been different. If PNG were to change
to a single dose of measles vaccine at 9 months, it seems
unlikely that higher levels of protection for the population
would result at 12 months, compared with continuing the
current two-dose schedule of 6 and 9 months.
Our study could not measure cell-mediated or mucosal
immunity, both of which are important in the protection
against measles. A T-cell mediated immune response has
been found to be as effective at 6 months as a 9 months
(Gans et al. 2001) and studies differ on whether the
development of T-cell mediated immunity is independent
of the presence of passively acquired antibodies or may be
inhibited by them.
Enzyme immunoassays assays, such as the ones we used,
may underestimate the antibody response. They measure
antibody to the nuclear rather than haemaglutinin protein,
which is the target of the plaque neutralization test (PNT)
and the haemagglutinin inhibition test. Enzyme immunoassays assays are less sensitive than PNT, but PNT are very
expensive and time consuming, and we did not have the
resources to use these tests in our study.
In our study 97 140 infants (69%) had residual maternal antibodies at 6 months, as defined by presence of OD
below the test cut-off of 150 mIU ml; 71% of infants who
had either no or an equivocal serological response to
measles vaccine at 6 months had traces of measles IgG
prior to immunization. As none of the participating infants

2009 Blackwell Publishing Ltd

had a history of measles or suspected measles prior to or

throughout the study period, these antibodies are most
likely to have been maternally derived. The presence of
residual maternally derived antibodies may partly explain
the lack of response to vaccination. However it is important to note that enzyme immunoassay tests are particularly poor in measuring the level and functional effect of
maternal antibody in infant serum (Goncalves et al. 1999).
The proportion of children with measles antibodies at
6 months of age varies between populations studied; 26%
of Turkish infants of seropositive mothers remain seropositive for measles IgG antibodies between 4 and
9 months; only 17% of African infants of HIV-positive
mothers have detectable antibodies at 6 months (Perry
et al. 2000; Metintas et al. 2002).
Although we could not measure other factors than the
presence of maternal antibodies must be important in the
response to measles vaccine at 6 months; 19% of infants in
our study failed to seroconvert despite having no preexisting serum antibodies. A similar finding was reported in
nine 6-month old infants of seronegative mothers in the
USA (Gans et al. 1998; Kumar et al. 1998).
Serum was collected at varying times after immunization
(mean 85 days, SD 20, range 21123). For the first 19 cases
the instructions to parents were to return for repeat blood
sampling 1 month after the 6 month dose. Because many
mothers had transport difficulties, we asked them to return
when their infant was 9 months of age, so that the second
blood sampling could be combined with receiving the
routine second scheduled dose of measles vaccine. Curiously, infants whose IgG was measured within 60 days of
vaccination were more likely to seroconvert than infants
whose serological response was measured later (Figure 2).
This effect on the variability of antibody response
remained even after accounting for the presence of IgG
prior to immunization. It is possible that this points to
early secondary vaccine failure, with loss of acquired IgG
within months of seroconversion. To our knowledge a
rapid loss of antibodies has not been reported before, and
unfortunately the methodology did not allow us to explore
this unexpected finding any further. However we feel that
the low antibody response to the 6 months vaccine is more
likely to reflect primary lack of seroconversion. Besides the
presence of residual maternal antibodies there are other
possible reasons for primary vaccine failure; including
interruption of the cold chain, errors in administration,
host factors (such as malnutrition and HIV AIDS).
While it is possible that the low antibody response in our
study was due to administration errors or cold chain
problems, all possible measures were taken to reduce this
likelihood. The vaccines were regularly checked for their
expiry date, and the refrigerator in the clinic had a
171

Tropical Medicine and International Health

volume 14 no 2 pp 167173 february 2009

J. Kurubi et al. Immune response to measles vaccine PNG

temperature monitor, which during the study never indicated that the vaccines had been exposed to inadequate
refrigeration.
PNG has used standard-titre Edmonston-Zagreb (EZ)
vaccine in recent years, and this was the vaccine used in the
current study. Under 9 months of age EZ vaccine produces
higher rates of seroconversion than Schwarz vaccine, but
the response is variable between populations studied (Cutts
et al. 1995). Among Bangladeshi infants given standard
dose EZ vaccine there was 70% seroconversion at
6 months and 95% seroconversion with vaccination at
9 months (Schnorr et al. 2001). The previous study in PNG
showed higher rates of seroconversion at 67 months than
we found in this present study, but the numbers were small
(Rogers et al. 1991). This previous PNG study involved
infants of mothers who would not have been vaccinated
(the study was done in the late 1980s and measles vaccine
was only introduced in 1982) therefore they may have been
less likely to have measles antibodies, unless previously
exposed to infection. The variation in seroresponse to
Edmonston-Zagreb vaccine in 6-month old infants found
in other studies is partly because of differences in maternal
antibody profiles of the populations studied, methods of
vaccine titre measurement, serological assays and
definitions of seroresponse (Cutts et al. 1995). We used a
level of IgG of >330 IU ml as indicating definite protective
immunity, which is supported by other studies (Chen et al.
1990; Tischer et al. 2007). But given that it is likely that
protection occurs at lower levels of antibody responses, the
proportion of infants protected by a dose of measles
vaccine at 6 months may be at least 50%.
Port Moresby, the capital of PNG, is a multiethnic
community where the residents come from all regions in
the country, so to at least some extent the results are likely
to be representative of the national population.
This study has enabled a better understanding of the
serological response to measles vaccine in mid-infancy in
PNG. Based on this there will be careful consideration of
the national immunization schedule. The current
WHO UNICEF plan of action for global measles mortality
reduction includes the use of mass campaigns to provide
indirect protection to unvaccinated persons, e.g. young
infants, as well as to increase coverage. Achieving very high
coverage at 912 months of age should be the goal of every
country. However in some countries there may still be
considerable advantages in scheduling an additional earlier
dose of measles vaccine in mid-infancy. These advantages
are the likely non-specific effects of measles vaccine on
mortality (Aaby et al. 1995; Shann 2002); lower overall
mortality when measles vaccine is given before 9 months of
age (Aaby et al. 1993; Martins et al. 2008) and the priming
in young infants due to stimulation of cell mediated
172

immunity, and a later booster response (Helfand et al.

2008). In a country like PNG, where access to health
services is difficult, immunization coverage scheduled for
later infancy is lower because of waning parental interest in
bringing children to preventive health clinics as they get
older. Recently a study showed that the average age of
infants receiving their scheduled 6-month dose of measles
vaccine was 8.8 months, and their average age when
receiving their 9-month dose was over 12 months. Therefore, apart from the potential protective efficacy of a young
infant dose (Martins et al. 2008), keeping a mid-infancy
dose as well as a late infancy dose is also a practical way of
ensuring most infants are not too late in receiving their first
dose of measles vaccine.
Acknowledgements
We thank the members of the Paediatric Society of PNG
for endorsing this research idea. Particular thanks to Dr
William Lagani and Dr David Mokela. We are grateful to
Anthony Gomes and Richard Duncan, World Health
Organization for assistance with funding and logistics for
transporting serum specimens. We thank Dr Heath Kelly
from VIDRL for his enthusiastic commitment to addressing
this issue. We thank the RE Ross Trust, Victoria for
generous assistance with funding through the PNG Child
Health Research Fellowship.
References
Aaby P, Anderson M, Sodemann M et al. (1993) Reduced childhood mortality after standard measles vaccination at 48 months compared with 9-11 months of age. BMJ 307, 1308
1311.
Aaby P, Samb B, Simondon F et al. (1995) Non-specific beneficial
effect of measles immunisation: analysis of mortality studies
from developing countries. BMJ 311, 481485.
Anonymous (2005) Seroconversion rates and measles antibody
titres induced by measles vaccination in Latin American children
six to 12 months of age. Reviews of Infectious Diseases 5, 596
605.
Chen RT, Markowitz L, Albrecht P, Stewart JA & Mofenson LM
(1990) Measles antibody: re-evaluation of protective titres.
Journal of Infectious Diseases 162, 10361042.
Cutts FT, Grabowsky M & Markowitz LE (1995) The effect of
dose and strain of live attenuated measles vaccines on serological responses in young infants. Biologicals 23, 95106.
Dick B, Smith T & Kipps A (1975) A minimum age for measles
vaccination to coloured children. South African Medical Journal
49, 19511954.
Duke T (2003) The crisis of measles and the need to expand the
ways of delivering the EPI in PNG. Papua New Guinea Medical
Journal 46, 17.

2009 Blackwell Publishing Ltd

Tropical Medicine and International Health

volume 14 no 2 pp 167173 february 2009

J. Kurubi et al. Immune response to measles vaccine PNG

Gans HA, Arvin AM, Galinus J et al. (1998) Deficiency of the

humoral immune response to measles vaccine in infants immunized at 6 months. JAMA 280, 527532.
Gans H, Yasukawa L, Rinki M et al. (2001) Immune responses to
measles and mumps vaccination of infants at 6, 9 and
12 months. Journal of Infectious Diseases 184, 817826.
Goncalves G, Cutts F, Forsey T & Andrade HR (1999) Comparison of a commercial enzyme immunoassay with plaque reduction neutralization for maternal and infant measles antibody
measurement. Revista do Instituto de Medicina Tropical de Sao
Paulo 41, 2126.
Halsey NA, Boulos R, Mode F et al. (1985) Responses to measles
vaccination in Haitian infants 6 to 12 months old. Influence of
maternal antibodies, malnutrition, and concurrent illnesses.
New England Journal of Medicine 29, 544549.
Helfand RF, Witte D, Fowlkes A et al. (2008) Evaluation of the
immune response to a 2-dose measles vaccination schedule
administered at 6 and 9 months of age to HIV-infected and HIVuninfected children in Malawi. Journal of Infectious Diseases
198, 19.
Johnson CE, Nalin DR, Chui LW et al. (1994) Measles immunogenicity in 6- versus 15-month-old infants born to mothers in the
measles vaccine era. Pediatria 93, 939944.
Kourouma K, Konde MK, Diallo MP et al. (1992) Vaccination
against measles at 6 months of age. Annales de Pediatrie (Paris)
39, 566571.
Kumar ML, Johnson CE, Chui LW et al. (1998) Immune response
to measles vaccine in 6-month infants of measles seronegative
mothers. Vaccine 16, 20472051.
Martins CL, Garly M-L, Bale C et al. (2008) Protective efficacy of
standard Edmonston-Zagreb measles vaccination in infants aged
4.5 months: interim analysis of a randomised clinical trial.
British Medical Journal 337, 339343.

Metintas S, Akgum Y, Arslantas D, Kalyoncu C & Ucar B (2002)

Decay of maternally derived measles antibodies in central Turkey. Public Health 116, 5054.
Mgone J, Mgone C, Duke T, Frank D & Yeka W (2000) Control
measures and the outcome of the measles epidemic of 1999 in
the Eastern Highlands Province. Papua New Guinea Medical
Journal 43, 9097.
Perry R, Mmiro F, Ndugwa C & Semba RD (2000) Measles
infection in HIV-infected African infants. Annals of the New
York Academy of Sciences 918, 377380.
Rogers S, Sanders RC & Alpers MP (1991) Immunogenicity of
standard dose Edmonston-Zagreb measles vaccine in highland
Papua New Guinean children from four months of age. Journal
of Tropical Medicine and Hygiene 94, 8891.
Sakatoku H, Nakano T, Arai S & Afari EA (1994) Antibody
responses to measles vaccination in rural Ghanaian infants.
Journal of Tropical Paediatrics 40, 291293.
Samb B, Aaby P & Whittle H (1995) Serological status and measles
attack rates among vaccinated and unvaccinated children in rural
Senegal. Pediatric Infectious Disease Journal 14, 203209.
Schnorr JJ, Cutts FT, Wheeler JG et al. (2001) Immune modulation after measles vaccination of 6-9 months old Bangladeshi
infants. Vaccine 19, 15031510.
Shann F (2002) Non-specific effects of vaccines in developing
countries: we need evidence about the effects of vaccines on
mortality from all causes. British Medical Journal 321, 1423
1424.
Tischer A, Gassner M, Richard J-L et al. (2007) Vaccinated students
with negative enzyme immunoassay results show positive measles
virus-specific antibody levels by immunoflourescence and plaque
neutralization. Journal of Clinical Virology 38, 204209.
Vidvashankar C (2002) Optimal age for measles vaccination.
Journal of the Indian Medical Association 100, 2426.

Corresponding Author Trevor Duke, Centre for International Child Health, Department of Paediatrics, University of Melbourne,
Royal Childrens Hospital, Flemington Road, Parkville, Melbourne, Vic. 3052, Australia. Tel.: +613 9345 5968;
E-mail: trevor.duke@rch.org.au

2009 Blackwell Publishing Ltd

173