Dept. of Child and Adolescent Psychiatry, University Hospital
Evald Saemundsen
State Diagnostic and Counselling
Centre
Introduction
This is a study of the prevalence
of autism comparing a case register with a previous study in Iceland
and two recent population based studies in Canada and Sweden.
The objective is to estimate the prevalence of autism in Iceland
given the evolving diagnostic practices and services over the
last decades. This is done by means of a case register based
on the records of two institutions. One or both of the authors
have been involved in the diagnosis and/or the assessrnent of
all the children diagnosed with autism in the age-groups reported.
Thus the present study is based on an ongoing registration of
cases by the authors.
In Iceland several institutions
and professionals are involved in the early detection of developmental
disorders. These include hospitals, primary health care centres
and private practices. In the case of suspected serious developmental
disorder a referral is made to the State Diagnostic and Counselling
Centre (SDCC). Primary health care centres have the legal responsibility
of developmental surveillance of all children between the ages
0-4 years. The system provides a comprehensive capture of preschool
children at risk for developmental disorders.
Until approximately 1988 suspected
cases of autism were referred directly to the Department of Child
and Adolescent Psychiatry (DCAP) of the University Hospital in
Reykjavik. Since 1988, most referrals have been made to the SDCC
where the diagnosis of autism is made in close cooperation with
the DCAP that supervises the treatment of autistic individuals
in the whole country.
The diagnostic systems used
to diagnose autism and other PDD disorders in Iceland have evolved
from a loose set of non-operationalized criteria of Kanner's infantile
autism (Kanner, 1943), Creak's 9 points criteria (Creak, 1963),
Rutter's criteria (Rutter, 1978), ICD-9 (WHO, 1978) and ICD- 10
(WHO, 1993). At the same time we have been moving away from
a purely categorical notion of autism to the concept of autistic
spectrum disorders or PDD.
In a recent review, Lorna
Wing (1993) examied sixteen multinational studies of the prevalence
of autism published in the period 1966-1991. Age specific rates
of autism varied from 3.3 to 16.0 per 10.000 in the studies examined.
There seems to be a trend for the more recent studies, published
after 1985, to yield higher rates than the earlier studies. The
rnost likely explanations are increased awareness of the manifestations
of autistic conditions and changes in diagnostic practice (Wing,
1993).
It is not known whether this
applies to Iceland. The present study is intended to test the
hypothesis of a similar trend towards higher rates of autism in
Iceland. Three studies have been chosen to serve as points of
comparison: one Icelandic study (Magnusson, 1977) and two population
based studies, one in Canada (Bryson et al., 1988) and one in
Sweden (Gillberg et al., 1991).
Methods
The accumulation of data was
based on the ongoing registration of children receiving diagnoses
of PDD in Iceland. All the children in this group were diagnosed
at either the Department of Child and Adolescent Psychiatry or
the State Diagnostic and Counselling Centre. In each case the
earliest PDD-type diagnosis was selected as the case diagnosis.
Diagnostic classification
was made either according to ICD-9 or ICD-10. This list includes
autism, atypical autism 2, atypical autism 1, disintegrative disorder,
Asperger syndrome (and Rett syndrome). For all of the children
born between 1984-1992, the definitions of atypical autism were
based on the ICD- 10 specifications and operationalized through
the Autism Diagnostic Interview (Le Couteur et al., 1989). The
category referred to as Atypical autism 2 includes individuals
who rnet diagnostic criteria for autism in two out of three areas
of abnormality, that is impairment of social interaction, communication
or restricted, repetitive or stereotyped patterns of behaviour
and interests. The term Atypical autism 1, here, refers to individuals
who meet diagnostic criteria in one out of three areas of abnormality.
In both cases the criterion of age of onset hefore 36 months
is met.
The case register study chosen
for comparison (Magnusson, 1977) is the only previous estimate
of the prevalence of autism in Iceland. Using Creak's 9 points
(Creak, 1963) as diagnostic criteria it comprised all children
diagnosed with autism or disintegrative disorder born in Iceland
in the 10 year period 1964-1973. Hence it was decided to use
a similar approach in the present study, dividing cases into two
groups according to year of birth, 1974-1983 and 1984-1992 respectively.
The younger group (1984-1992) was considered a sample reflecting
more recent diagnostic practices in Iceland.
The two other studies (Bryson
et al. 1988- Gillberg et al. 1991) were both chosen for comparison
because of their proximity in time, their population based approach,
similar diagnostic guidelines, and the high rates of autism found.
The Swedish study used DSM-IIIR criteria but the Canadian study
employed Denckla's (1986) criteria which for the present purposes
may be considered equivalent to DSM-IIIR criteria (Wing, 1993).
The prevalence rates given
are period prevalences for individuals 3-12 years in 1976, individuals
13-22 years in 1996, and individuals 4-12 years in 1996. According
to the Icelandic Bureau of Statistics the total population of
the country was 267.809 on 1 Dec 1995. Children born in the two
periods numbered 42.276 and 38.746 respectively on 1 Dec. 1995.
Results
Comparison within Iceland
Rates. In
order to make the data comparable to the results of the earlier
Icelandic study (Magnusson, 1977), the categories of Childhood
autism and Disintegrative disorder are presented both conjointly
and separately (Table l). Furthermore, figures of estimated prevalence
are presented for other PDD categories: Atypical autism 2, Atypical
autism 1 and Asperger syndrome. For the categories of Childhood
autism and Disintegrative disorder the estimated prevalence was
considerably higher in the last period (1 984-1992) than the rates
found in the two previous periods. This trend was still apparent
when further PDD categories were added to the comparison.
Table 1. Estimated prevalence of autism and PDD in three age groups in Iceland.
Age specific rates
per 10.000
| Diagnostic categories: |
|
|
|
| Childhood autism | |||
| Childhood autism Disintegrative disorder | |||
| Childhood autism Atypical autism 2 | |||
| Childhood autism
Disintegrative disorder Atypical autism 2 Asperger syndrome | |||
| Childhood autism Disintegrative disorder Atypical autism 2 and 1 Asperger syndrome | |||
Male-female ratio. A
clear trend towards higher male-female ratio appeared when the
last period was compared with the two previous ones (Table 2).
This trend was apparent throughout the PDD range.
Table 2. Male-female
ratio of three age groups in Iceland
| Diagnostic categories: |
|
|
|
| Childhood autism | |||
| Childhood autism Disintegrative disorder | |||
| Childhood autism Atypical autism 2 | |||
| Childhood autism
Disintegrative disorder Atypical autism 2 Asperger syndrome | |||
| Childhood autism Disintegrative disorder Atypical autism 2 and 1 Asperger syndrome | |||
Level of intelligence.
The comparison
with the previous study in Iceland was possible only for the categories
of Childhood autism and Disintegrative disorder (Table 3). In
the last period the distribution of IQ/DQ was different from the
two previous ones, with fewer individuals in the most severe range
and in the normal or near normal range.
Table 3. Level of intelligence of children with Childhood autism and
Disintegrative disorder
in Iceland in three age groups (percentages)
Year of survey: 1976
1996 1996
Age and birth year
3-12 years 13-22 years 4-12 years
(1964-1973) (1974-1983)
(1984-1992)
IQ/DQ n= 19 n=
16 n = 25
<50 47 50 36
50-69 37 25 56
>70 16 25 8
|
|
| |
Comparison between Iceland
and poputalion based studies in Sweden and Canada Rates.
For comparison
with the studies based on DSM-IIIR criteria a conservative approach
was adopted as to the inclusion of PDD categories. Gillberg (1991)
explicitly excludes Asperger children from his autistic group
but includes children with atypical autism while Bryson (1988)
makes no such differentiation. For this reason children with
the diagnosis of Asperger syndrome were excluded from the Icelandic
group in this comparison as well as children who met diagnostic
criteria in only one area of abnormality (Atypical autism l).
In this comparison estimated prevalence of autism in Iceland
was lower than the rates found in Sweden and Canada.
Male-femate ratio. The
male-female ratio found in the present study was higher than in
the studies selected for comparison and higher than the ratios
found in studies published after 1985 and reviewed bv Wing (1993).
Level of intelligence. The present study includes fewer individuals in the severely
retarded range (IQ/DQ <50)
than the comparison studies and a higher percentage in the mildly
retarded range (IQ/DQ 50-69). This difference was, however, not
significant.
Estimated prevalence
of autism, male-female ratios and level of intelligence of autistic
children in three countries
| Rates | M/F
ratio | ||||
| Canada (Bryson et al., 1988). Age group 6-14 years (n = 21) | |||||
| Sweden (Gillberg et al., 1991). Age group 4-13 years (n = 47) | |||||
| Iceland (Magnusgon & Saemundsen, 1996)
Age group 4-12 (n = 34) | |||||
Discussion
The estimated prevalence of
autism found in this study is higher than previously described
in Iceland and closer to the higher rates found in some of the
more recent epidemiological studies than to those found in earlier
studies (Wing, 1993). i.e. before 1985.
The possibility of a true
increase in prevalence of autism cannot be excluded. In the studies
reviewed by Wing, the only evidence for a true increase in the
prevalence of autism comes from Gillberg's study (1991) the increase
was considered, at least partly, derived from children of first
generation inmigrants. All families in the present study were
native Icelanders except for one immigrant father. Increased
awareness of PDD conditions and/or related changes in diagnostic
practice seem to be more feasible explanations.
The possibilitv of false positives
is present but this risk is, however, minimized by the fact that
one of the authors and his coworkers are involved in follow-up
of all individuals who have received diagnoses of autism.
The current methods of identification
of autistic individuals in Iceland may not lead to 100% capture
of all cases. The male-female ratio in the present study is higher
than that found in the Canadian and Swedish studies. This may
indicate an underrepresentation of severely cognitively impaired
individuals in our autistic group, since some studies report that
girls with autism tend to be more severely impaired than autistic
boys (Wing, 1981; Gillberg, 199l).
The case register method is
conventionally considered likely to yield underestimates even
when entire birth cohorts are followed up (Bryson. 1988). However,
this study has produced data that can be considered estimates
of the true prevalence of autism in Iceland and an indication
of changes in diagnostic practice. Furthermore the results have
direct implications for the orranization of services for individuals
with PDD in Iceland and suggest a line of further research.
The results raise the question
whether more attention should be paid to the severely retarded
group in Iceland to ensure that the possibilitv of autistic symptomatology
is considered in each individual case. In the severely
retarded preschool child, the symptoms of autism may not be apparent
upon detection of the retardation but may become a serious handicap
at a later date. In this context it might be feasible to define
"at risk for autism" criteria that would ensure a follow-up
with reassessment.
Further research focusing
on severely retarded children born in the period 1984-1992, screening
for autistic symptoms and making a careful diagnostic evaluation
of suspect cases might provide answers to some of the questions
raised by this study.
References
Bryson, S., Clark, B., S.,
Smith, I., M. (1988). First report of a Canadian epidemiological
study of autistic syndromes. Journal of Child Psychology and
Psychiatry, 29, 43 3-445.
Creak, E. M. et al. (1 96
l). Schizophrenic syndrome in childhood. Report of a working
party. British Medical Journal, 2, 889.
Denckla, M. B. (1986). Editorial:
New diagnostic criteria for autism and related behavior disorders:
Guidelines for research protocols. Journal of the American
Academy of Child and Adolescent Psychiatry, 25, 221-224.
Gillberg, C., Steffenburg,
S. & Schaumann, H. (199l). Is autism more common now than
ten years ago? British Journal of Psychiatry, 158, 403-409.
Kanner, L. (1943). Autistic
disturbances of affective contact. Nervous Child, 2, 217-250.
LeCouteur, A., Rutter, M.,
Lord, C., Rios, P., Robertson, S., Holdgrafer, M., & McLennan,
J. D. (1989). Autism Diagnostic interview. Journal of autism
and developmental Disorders, 19, 363-387.
Magnusson, G. T. (1977). Athugun a gedveikum boernum a Islandi: Boern faedd 1964-1973.
Laeknabladid, 63, 237-243.
Rutter, M. (1978). Diagnosis
and definition of childhood autism. Journal of Autism and
Childhood Schizophrenia, 8, 139-161.
Wing, L. (1993). The definition
and prevalence of autism: A Review. European Child and Adolescent
Psychiatry, 2, 61-74.
World Health Organization
(1978). International classification of diseases (ICD-9) (9th
edition). Geneva: World Health Organization.
World Health Organization (1993). Mental disorders: The ICD-10 classifícation of mental and behavioural disorders. Diagnostic criteria for research. Geneva: World Health Organization.