  |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
09-August-2002, Vol 115 No 1159
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
The beneficial impact of Hib vaccine on disease rates in New
Zealand children
Nick Wilson, Jay Wenger, Osman Mansoor, Michael Baker and
Diana Martin
Haemophilus
influenzae type b (Hib) was the most common cause of life-threatening
bacterial infection (usually meningitis) in New Zealand children up until the
early 1990s.1 Maori and Pacific children were
at greatest risk.2 The substantial disease
burden led to demands for using the newly-available Hib conjugate
vaccine.2,3 Hib vaccine was added to the
childhood immunisation schedule in January 1994 with DTP-Hib (Tetramune)
replacing DTP at six weeks, three months and five months, and monovalent Hib
being given at eighteen months. Parents with children aged under five years
(under-5s) were also encouraged to take them to their general practitioner for
immunisations. Coverage has since probably exceeded 80% for the third dose and a
decline in laboratory reported Hib invasive disease
followed.4,5 This decline has subsequently
continued in recent years.6 Hospital data for
Christchurch also suggest that a decline in epiglottitis is associated with the
introduction of Hib vaccine.7 This report uses
routinely collected hospital and mortality data along with surveillance data to
examine in more detail the impact of the introduction of Hib immunisation on the
health of New Zealand children.
MethodsNational mortality data were
obtained for the years 1988 to 1998 and national hospitalisation data were
obtained for 1988 to 2000 (from the New Zealand Ministry of Health). ICD9 codes
covered meningitis (320-323 + 047) and epiglottitis (464.3). Some of the codes
referred to conditions caused by Haemophilus influenzae (Hi) – but the
disease type is not distinguished by ICD9 coding. Hence this data source
includes a small number of cases that are due to strains of Hi other than type b
(ie that are not Hib). Population data for calculating rates were also obtained
from the Ministry of Health (based on 1991 and 1996 population censuses with
inter-census estimates).
The analysis was confined to the under-5s since this age group had the greatest burden of Hib disease in the pre-vaccine era and is the age group targeted for immunisation. Sub-group analysis also included the population aged 6–17 months (since this age group has historically tended to have the highest rate of Hib disease of all). Data were analysed with EpiInfo version 6.04 and relative risk calculations were made comparing the post-vaccine period (1995–2000) with the pre-vaccine period (1988–1993). This division ignores some private use of monovalent Hib vaccine during 1993 (as reported by Teele8). National changes in the coding of ethnicity by hospital clerical staff were introduced in September 1995. Coincident with this change, the proportion of children hospitalised with meningitis and pneumonia who were classified as being of Maori and Pacific ethnicity dropped in 1995 (eg for the years 1992 to 1998 the proportions of children hospitalised with pneumonia who were classified as Maori were: 33%, 35%, 33%, 20%, 32%, 35%, 34% for each year). However, the proportions of Maori and Pacific children reverted to the pre-1995 pattern in the following years even though there were no more changes to the coding system. The reason for this shift is unclear, but may be due to transient problems with introduction of the new system that were subsequently overcome. National reference laboratory data on Hib were collected by ESR, Wellington. Isolates are obtained from blood, CSF or aspirated joints but no ethnicity data were collected with the isolates. It is considered that virtually 100% of isolates were recovered from Hib disease cases prior to 1994, but since then some less complete reporting from regional laboratories might possibly have occurred. ESR also collects national notification data, however Hib only became a notifiable disease in June 1996. Consequently, notification data prior to 1996 are incomplete. Since 1996, ESR has combined notification data with isolate data to improve the sensitivity of surveillance and also to provide more complete information on each case, notably ethnicity and immunisation status. ResultsMeningitis
In the post-vaccine period (1995–2000) relative to the pre-vaccine
period (1988–1993), the rate of meningitis from all causes in under-5s
declined by 13%, and by 39% in those aged 6–17 months (ie relative risk
(RR) = 0.87 and RR = 0.61). Furthermore, the hospitalisation rate for meningitis
due to Hi in children in these two age groups declined by 92% and 93%
respectively (ie RR = 0.08 and RR = 0.07) (Table 1). The annual hospitalisation
rate for meningitis due to Hi declined from 27 to 2 per 100 000 population (in
under-5s). For Maori children the equivalent declines were similarly large at
87% and 91% for the under-5s and 6–17 month age groups respectively (ie RR
= 0.13 and RR = 0.09). For Pacific children the respective reduction was 91% in
both age groups. Yet there were statistically significant
increases for rates of
“meningitis – unspecified cause” (RR = 1.85) and
“meningitis due to unspecified bacteria” (RR = 1.54) among under-5s,
but the absolute number involved in these increases was smaller than the
reduction in overall meningitis, or Hi meningitis hospitalisations.
There was a 67% decline in the death rate from
“meningitis due to Hi” in under-5s (RR = 0.33) but this was not at a
statistically significant level. (The detailed mortality results and the more
detailed hospitalisation results for the 6–17 month age group are not
tabulated but are available on request from the authors).
Table 1. Hospitalisations for meningitis in children
under five years of age
*=p<0.05;
†=p<0.01;
‡=p<0.001;
§=Rate is per 100 000 population;
¶ =This category does not include
meningitis cases associated with meningococcal disease
Epiglottitis There
was a significant 94% decrease in the rate of epiglottitis hospitalisations in
under-5s (RR = 0.06) (Table 2). The decline was also significant among Maori and
Pacific children with no occurrence of the disease in these populations in the
post-vaccine period. Also there were no deaths from epiglottitis in the
post-vaccine period, compared to four deaths during the pre-vaccine
period.
Table 2. Hospitalisations for epiglottitis in children
under five years of age (ICD9 code = 464.3)
*=p<0.05
(Fischer exact test);
†=p<0.01
(Fischer exact test);
‡=p<0.001;
§=Rate is per 100 000
population
Disease rates in
1–4 year olds In the
1–4 year old age group (who were eligible for “catch-up” Hib
immunisations), there was a significant 54% decline in the hospitalisation rate
for “meningitis due to Hi” from 13.5 to 6.3 per 100 000 population
in the immediate post-vaccine period of 1994–1997 (RR = 0.46, 95%CI = 0.31
– 0.68, p = 0.00007). This decline suggests that approximately 42
hospitalised cases of meningitis in this group of “catch-up
eligible” children were prevented over this post-vaccine period.
Similarly, for epiglottitis in the “catch-up eligible” group in the
immediate post-vaccine period, there was a significant decline in the rate from
12.3 to 2.0 per 100 000 population (RR = 0.16, 95%CI = 0.09 – 0.30,
p<0.0000001). This decline suggests that 60 hospitalised cases of
epiglottitis in “catch-up eligible” children were prevented over
this period.
Equity issues The
rate of “meningitis due to Hi” decreased substantially in all ethnic
groups in the post-vaccine period (Table 1). However, the proportion of all
cases in the post-vaccine period who were classified as Maori increased from
23.2% (108/465), to 39.5% (15/38) (p = 0.03). The equivalent increase for
Pacific children was from 6.9% (32/465), to 7.9% (3/38), but this was not
statistically significant. Analysis of post-vaccine notification data found no
significant difference by ethnic group in the proportion of Hib cases who were
“not immunised” (but the numbers involved were relatively
small).
Laboratory and notification
data The number of Hib isolates obtained from under-5s (from CSF, blood
and other clinical specimens) submitted to the national reference laboratory at
ESR is detailed in Table 3.
Table 3. Number of hospitalised cases of meningitis and
epiglottitis compared to numbers of laboratory confirmed cases of Hib disease
and of Hib disease notifications in children aged under five years
* Some private administration of Hib vaccine began in
1993.
†
Hib vaccine introduced to the routine childhood immunisation schedule in January
1994.
The trend in laboratory diagnosed Hib also showed a marked
reduction after the introduction of Hib vaccine (Figure 1). There also appeared
to be a fairly close correspondence between the hospitalisation data and the
national laboratory data. In the post-vaccine period (1995–2000) relative
to the pre-vaccine period (1988–1993), there was a significant overall
reduction in submitted Hib isolates for children under five years (RR = 0.07,
95%CI = 0.05 – 0.09, p<0.0000001).
Figure 1. Hospitalisations and laboratory reported
cases attributed to Hi disease
 DiscussionHospitalisation data in general are
known to have quality limitations, but for Hib disease the New Zealand data
appear to be remarkably robust when compared with reported laboratory data
(Table 3). The ethnicity-specific data are of particular interest, but should be
interpreted cautiously in light of the changes in the coding system made during
the study period (as detailed in the Methods section). The current study clearly
shows that routinely collected data can be used to monitor major changes in
disease trends, and also identify gaps in the impact of the Hib immunisation
programme for specific ethnic groups.
The hospitalisation data suggest that the introduction of
Hib vaccine is very significantly associated in time with a major reduction in
hospitalisation rates for “all meningitis” and “meningitis due
to Hi” in under-5s. Indeed, the rate for meningitis due to Hi decreased by
92%, which is equivalent to around 79 fewer hospitalisations per year for this
serious disease. The pattern seen for meningitis hospitalisation data is very
similar to that seen in the declining number of Hib isolates identified by the
national reference laboratory. However, the laboratory data would suggest that
over 100 cases of Hib disease (including meningitis) in under-5s are being
prevented each year.
At the same time as this major decrease in all meningitis
and meningitis due to Hi was noted, there was an increase in the rate of
hospitalisation attributed to “meningitis – unspecified cause”
and for “meningitis due to unspecified bacteria”. Although the
reason for this is not clear, several possible contributing factors may be
hypothesised. The increase could have been driven by the meningococcal disease
epidemic that began in 1991 and reached a sustained high level by
1995.9 About a third of these cases are
diagnosed clinically without laboratory confirmation and it is likely some are
coded to these “unspecified” diagnostic categories. It is also
possible that in the setting of the introduction of a new vaccine for
meningitis, the clinical suspicion of, and admission to hospital for, suspected
meningitis increased during this time period. Thus, a larger proportion of
clinical syndromes previously not reported as meningitis may have been reported.
Another possible contributing factor may have been the greater use of antibiotic
therapy on children with suspected meningitis seen in primary care settings as a
result of provider response to the meningococcal disease epidemic. Such
antibiotic use before hospitalisation would tend to make a final diagnosis more
difficult. To some extent this trend in unspecified meningitis might mean that
the magnitude of the decline in “meningitis due to Hi” is partly a
diagnostic artefact. Nevertheless, the overall significant decline in meningitis
from all causes strongly supports a key role for the introduction of Hib
vaccine. Further evidence comes from the sudden decline in disease rates in the
immediate post-vaccine years (Table 3). There is even some suggestion that for
both meningitis and epiglottitis the decline began in 1993, which was the year
in which the private purchase and use of Hib vaccine first began (but probably
for only a minority of this age group).
The 94% decline in the epiglottitis hospitalisation rate in
the post-vaccine period was even more sudden and substantial than for
“meningitis due to Hi”. Given the important role of
Haemophilus influenzae infection in the
causation of epiglottitis, this substantial and sudden decline (Table 3)
provides very strong evidence for a causative role for Hib vaccine in the
decline of this condition. This decline is equivalent to around 29 fewer
hospitalisations for epiglottitis per year and so this condition can now be
considered to be extremely rare in New Zealand (at around only one to two cases
per year).
There was a significant decline in the rates of meningitis
and epiglottitis in the 1–4 year old children who were eligible for
“catch-up” immunisations in the immediate post-vaccine period. This
is likely to reflect a mix of direct immunisation (for children who were given
catch-up Hib immunisations), the private use of Hib vaccine in 1993, and also a
herd immunity effect (as detailed for Hib immunisations programmes elsewhere in
the world10–12). Detailed data on
“catch-up” immunisation coverage are not readily available so the
contribution of each of these factors is not identifiable.
It is apparent that Maori and Pacific children have
benefited substantially from the sudden and marked decline of “meningitis
due to Hi” and epiglottitis. However, the proportion of all cases of
“meningitis due to Hi” in the post-vaccine period that were
classified as Maori actually increased significantly from 23% to 40%. One
possible reason for this may be the less effective delivery of immunisation to
Maori relative to the rest of the population. Indeed, the relatively poor
delivery of immunisation to Maori (and Pacific children) has been documented
extensively in the New Zealand setting in the
past.13,14 It appears that the opportunity
provided by the addition of this new vaccine to the immunisation schedule was
not successfully used to fully address inequalities in disease burden by ethnic
group. This suggests that a prudent course for current policy makers and health
workers would be to further support immunisation activities designed to reach
Maori and Pacific children in New Zealand. Furthermore, this finding highlights
the value of routinely collected data in identifying populations poorly served
by the existing immunisation system and focusing new efforts to address these
deficiencies. A national immunisation surveillance system would help to expand
these findings and monitor programmatic efforts to ensure that these
deficiencies are appropriately
addressed.15
Author information:
Nick Wilson, Public Health Physician, Wellington; Jay Wenger, Medical
Officer, World Health Organization, Geneva; Osman Mansoor, Scientist, World
Health Organization, Manila; Michael Baker, Public Health Physician, ESR,
Porirua; Diana Martin, Principal Scientist, ESR, Porirua
Acknowledgements: We
thank Martin Bonné and Durga Rauniyar of the New Zealand Ministry of
Health, who assisted with data extraction. We also thank Elizabeth Sneyd (ESR)
for providing the notification data. Initial aspects of this work were funded by
the World Health Organization (Western Pacific Region).
Correspondence: Dr
Nick Wilson, 367A Karori Road, Karori, Wellington. Fax: (04) 476 3646; email: nwilson@actrix.gen.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|