  |
|
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
15-July-2005, Vol 118 No 1218
|
||||||||||||||||||||||||||
|
Prevalence of Type 1 diabetes in New Zealanders aged
0–24 years
Denise Wu, Deborah Kendall, Helen Lunt, Jinny Willis, Brian
Darlow, Christopher Frampton
For an optimal long-term outcome, Type 1 diabetes in
children and young adults requires intensive management by the patient, their
family, and also their health professional team. The incidence of Type 1
diabetes has doubled in New Zealand in the last 15 years, mirroring the
international trend towards an increasing incidence of Type 1
diabetes.1,2 Prevalence of Type 1 diabetes will
also therefore be increasing, which has implications for service delivery and
planning. There has been a recent estimate of diabetes prevalence in New
Zealanders aged 25 and more,3 but there are no
current estimates of the prevalence of Type 1 diabetes in New Zealand in
children and young adults.
MethodsThe number of children and
young adults with diabetes, aged 24 years or less and residing in the CDHB
(Canterbury District Health Board) catchment area on 1 November 2003, was
estimated through review of current databases and medical records.
Diabetes type (i.e. Type 1, Type 2, and other types)
was determined by the attending clinician, using immunogenetic testing where
appropriate.4 Currently, young patients with
newly diagnosed diabetes are routinely tested for the diabetes-related
antibodies IA2 and GAD, as an aid to diagnostic classification. Selected
families may also undergo genetic testing for monogenic forms of diabetes.
Patients with transient steroid-induced diabetes were excluded from analysis.
Ethnicity was either self-defined or, for younger children, defined by the main
carer(s). Locally, younger patients who regularly attend secondary care diabetes
services are not routinely included in the free Ministry of Health primary care
diabetes annual check, thus no attempt was made to use primary care records as a
method of case identification.
Christchurch Diabetes Services have a database of
children and young adults with Type 1 diabetes, assembled from both clinical and
research records, as previously described.5 The
database appears to be near complete, and includes patients not currently
attending secondary care services. As a further check of completeness of
clinical records, all Christchurch hospital diabetes discharges were reviewed
for the 2½ year period prior to undertaking this study.
Community-focussed diabetes healthcare workers were
also asked if they knew of anyone with diabetes who was not currently in contact
with any hospital service. Some patients attend both Christchurch Hospital and
also the Ashburton Diabetes Clinic for their specialist diabetes care, and
Ashburton clinical records were also reviewed.
The domicile of people who attended school or
university in one location but were resident during holiday periods in one or
more different locations, was arbitrarily defined by the location which
delivered the majority of their diabetes healthcare.
The prevalence of diabetes in the CDHB catchment area,
which includes the city of Christchurch, the town of Ashburton, and rural areas
around North Canterbury, was estimated using the latest available census
statistics, from 2001.6 There are known ethnic
differences in the incidence and prevalence of Type 1 diabetes—with Maori,
Pacific peoples, and Asians having a lower incidence than those of European
descent.2,7,8 Also, the age distribution of the
CDHB population is skewed towards the elderly when compared to New Zealand as a
whole.6 When extrapolating local results to the
whole of New Zealand, data was therefore stratified by ethnic group and by age
bands 0–14 and 15–24 years of age.
The confidence interval on this estimate was derived
from the pooled error from each of the age-ethnic group estimates. This audit
had local ethics committee approval.
ResultsThere were 353 children and young
people residing in the CDHB catchment area in 2003 with known diabetes. Of
these, 330 had Type 1 diabetes and 51% were male. Seventeen of the remaining 23
people had Type 2 diabetes. At the latest census, there was an estimated 145,164
people aged 24 years or younger, residing in the CDHB catchment area. The
prevalence of Type 1 diabetes in children and young people was therefore 227 per
100,000 (95% confidence interval 203 to 252). In the age range 0–9 years,
prevalence was 99 per 100,000. In the 10–19 age range, it was 261 per
100,000.
Of the 330 people with Type 1 diabetes, 307 were European
New Zealanders, 13 were Maori, 3 were Pacific peoples, and 7 were Asian. The
estimated prevalence per 100,000 population was therefore 274 for Europeans, 81
for Maori, 77 for Pacific peoples, and 52 for Asians. There were sufficient
numbers of Europeans to stratify prevalence data by 5-year age bands. Results
are shown in Table 1.
Table 1. Number and prevalence of European New
Zealanders with Type 1 diabetes (stratified by age) residing in the Canterbury
District Health Board Catchment Area in 2003
*Age stratified population
data obtained from 2001 census.
Extrapolating the above (Christchurch) figures to the whole
of the New Zealand population thus produced an estimate of 2,158 Europeans, 231
Maori, 74 Pacific peoples, and 77 Asians with Type 1 diabetes—giving a
total number with Type 1 diabetes of 2,540 (95% confidence interval 2250 to
2830).
There were an estimated 1,118 people with Type 1 diabetes in
the age range 5 to 14 years, thus approximately 1:516 New Zealand schoolchildren
in this age range have Type 1 diabetes. An estimate of the number of young
people with Type 1 diabetes in each District Health Board is given in Table
2.
Table 2. Estimated number of young people with Type 1
diabetes by DHB (District Health Board)
*Population data from 2001
Census; **Measured rather than estimated number with Type 1
diabetes.
DiscussionThe incidence and prevalence of
Type 1 diabetes in young people is increasing. The prevalence of Type 1 diabetes
in Canterbury in 1988 was measured as 30 per 100,000 in the age range 0–9
years and 180 per 100,000 in the age range 10–19
years.9 In contrast, the corresponding 2003
estimates (which included Ashburton cases) are 99 and 261 per 100,000, in the
respective age ranges. The reason(s) for this temporal increase in prevalence in
the younger age group in particular is unclear.
It has been suggested that the age of presentation of Type 1
diabetes may be falling, in association with increased childhood
obesity.10 Between 1970 and 1999, there were
474 incident cases presenting in the region but the age of presentation did not
change significantly with time.1 The absolute
numbers of children developing diabetes in Christchurch is small however. Also,
a formal patient tracking study would be required to determine whether temporal
change in prevalence data was in part due to disproportionate movement
(geographical relocation) by a subgroup of patients, into or out of the CDHB
catchment area.
Currently, there are approximately 2,540 children and young
people with Type 1 diabetes in New Zealand. These findings have service delivery
implications. For example, if around 1:500 schoolchildren have Type 1 diabetes
then this has implications for health policy planning in schools, as most
secondary schools are likely to have at least one pupil with diabetes. If the
rate of change in prevalence continues at its current pace, it is likely to
place sufficient burden on both paediatric and adult secondary care services to
require new models of service delivery.
The method used to estimate of the number of young people
with Type 1 diabetes has several limitations, but the magnitude of any
associated bias is likely to be small. For example, the local estimate of the
number of people with Type 1 diabetes may not have captured every individual
with this condition, but no additional patients were identified either from
patient discharge data or by diabetes community workers, who were not already on
the clinic database. This suggests that the database was near complete.
It can be hypothesised that people with Type 1 diabetes may
preferentially choose to reside in areas where they have ready access to
emergency and specialist services, thus they may choose to move from a rural to
an urban area, following the diagnosis of diabetes. Undertaking an estimate of
Type 1 diabetes prevalence in a DHB with a large rural population may in theory
result in an underestimation of the number of people with Type 1 diabetes. The
urban:rural split in the CDHB catchment area population is, however, similar to
that of New Zealand as a whole. For example, one in seven Cantabrians live
rurally, a figure that is identical to that of the total New Zealand
population.6
There has been a small (3%) increase in the New Zealand
population between the last census in 2001 and the time of the study in
2003,6 thus the current study is likely to
result in a minor overestimate of prevalence per 100,000 population. This slight
shift in demographics will not, however, affect the estimate of the absolute
number of people with Type 1 diabetes in New Zealand.
Internationally, there are large geographic variations in
the incidence and prevalence of Type 1
diabetes.2 New Zealand data from the previous
three decades suggests that the incidence of Type 1 diabetes is higher in the
South Island compared to the North
Island,2,7,11 but these studies made no or
limited attempts to stratify by ethnicity. In contrast, a comparison of
incidence data over a 20-year period (1977 to 1996) from Christchurch and
Auckland showed mean incidence rates for childhood onset Type 1 diabetes of
16.44 and 11.61 per 100 000, per annum, for each respective city. However this
study did not detect any difference in incidence between these South and North
Island cities, when results were stratified by ethnicity (Jinny Willis, personal
communication, 2004).12
Our study found the prevalence of Type 1 diabetes in
Europeans to be three to four times higher than non-European New Zealanders.
This ratio is similar to the ethnic differences reported in the more recent
incidence study.12 Extrapolation of local
prevalence data, stratified by ethnicity, to the North Island population would
therefore seem reasonable. The ideal approach to estimating the prevalence of
Type 1 diabetes in New Zealand would of course be for each DHB to undertake its
own local study of prevalence, using predefined methodology, then collate
results with those from other centres.
This study estimated the prevalence of Type 1 diabetes only,
in part because the number of young people with Type 2 diabetes in our local
population was low. In regions where there is a high percentage of Maori in the
population, Type 2 diabetes in young adults is of increasing
importance,13 and may represent a major
clinical burden in that area.14 Estimating Type
2 prevalence in young people is, however, likely to be more difficult than
estimating the prevalence of Type 1 diabetes. Onset of Type 2 diabetes is
usually slower, with a gradual progression from impaired glucose tolerance to
frank diabetes, thus there is no clearly defined symptomatic onset of disease
requiring the patient to seek medical attention. Also, at least in our local
area, not all young people identified as having Type 2 diabetes are referred to
specialist services, so case identification for research and audit purposes is
more difficult.
In summary, this report gives an up-to-date population-based
estimate of the number of children and young people with Type 1 diabetes in New
Zealand. Young New Zealanders are from an increasingly multiethnic background
and this is reflected in the increasing numbers of non-European New Zealanders
estimated to have Type 1 diabetes. Specialist diabetes services, in particular,
therefore need to address cultural issues in relation to Type 1 diabetes in
ethnic groups, whilst at the same time accommodating the health needs of an
increasing number of young people with Type 2 diabetes.
Author information:
Denise Wu, Medical Student, Christchurch School of Medicine and Health Sciences,
University of Otago, Christchurch; Deborah Kendall, Research Nurse, Diabetes
Centre, Christchurch Hospital, Christchurch; Helen Lunt, Physician, Department
of Medicine, Christchurch Hospital, Christchurch; Jinny Willis, Scientist, Lipid
and Diabetes Research Group, Christchurch Hospital, Christchurch; Brian Darlow,
Professor of Paediatrics, Department of Paediatrics, Christchurch Hospital,
Christchurch; Christopher Frampton, Associate Professor of Statistics,
Department of Medicine, Christchurch Hospital, Christchurch
Acknowledgements:
Denise Wu was awarded a Christchurch School of Medicine and Health Sciences
summer Studentship to undertake this study. The Studentship was funded by the
National Diabetes Training and Research Centre Trust. We thank Trudy Brown, Neil
Owens, Dr Obefami, and the staff at Christchurch Diabetes Centre and Ashburton
Diabetes Centre for their help.
Correspondence: Dr
Helen Lunt, Christchurch Hospital, Private Bag 4710, Christchurch. Fax: (03) 364
0171; email: helen.lunt@cdhb.govt.nz
References:
|
|
||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|