Autism Incidence In UK

Local Monitoring By Health Authorities Of Childhood Autism Incidence In The UK

David Thrower, © February 2000

Monitoring of Autism


There has been considerable debate recently over whether there has been an increase in the levels of autism in children under 16, and whether there is an underlying upward trend.

To answer these questions obviously requires an accurate monitoring system in each Health Authority area, together with joint arrangements with education and social services, to produce a register of cases. This is in addition to the application of consistent diagnostic criteria, and other considerations.

In 1998, the then Junior Health Minister for England, Paul Boateng, stated in response to a Parliamentary written question by Helen Southworth MP, that:

………..commissioners are advised to have arrangements in place to use both local and national information to build up a picture of the health needs of people…………with autism in their locality. We also encourage health authorities to work together closely with local authorities in profiling both the volume, and needs, of…………people with autism (Parliamentary Written Answer of 9th June 1998)

To verify what the actual situation was, in the light of the then Ministers comments, at Health Authority/Board level, every Health Authorities in England and Wales, the Health Boards of Scotland, the Education and Library Boards of Northern Ireland and also the Social Services Departments of London were approached. Each body was asked to give a brief indication of how it assessed the incidence of child autism in its area.

This paper summarises the responses received. A relatively good response rate of 52% was obtained (considering that this work was unfunded, and was informal research from an individual parent), particularly from Health Authorities in England and Wales and from the Education and Library Boards of Northern Ireland. Only the response rate from London Social Services Departments was notably poor. Responses are still being received as at February 2000.

The total numbers of English & Welsh Health Authorities, Scottish Health Boards and Northern Ireland Education & Library Boards approached was 128. Of these, 77 responded. A further 33 London Social Services Departments were also approached, of which 7 responded.

Summary Of Results

  1. Of the total of 84 Health Authorities, Boards, Education & Library Boards and Social Services departments that responded:
  2. Just 14 (17% of respondents) have detailed data on autism in all or part of their area
  3. A further 32 (38% of respondents) have some very limited data and/or are taking active steps to establish monitoring systems
  4. The remaining 38 (45% of respondents have no data whatsoever, and (at the time of their response during 1999) had no firm plans to establish any systems to gather data on autism numbers.
  5. Overall, this response confirms that Paul Boatengs 1998 statement of Government expectations, insofar as it relates to monitoring of incidence, is not being borne out by on-the-ground practice, though a number of authorities are taking measures to establish or improve databases.

If only the 72 responses from the health authorities and boards of England, Scotland and Wales are considered (i.e. excluding London Social Services departments and Northern Ireland Education & Library Boards), then the situation within the responding health authorities/boards is as follows:

  1. Just 12 authorities/boards (17% of respondents) have detailed data
  2. A further 24 (33% of respondents) have some data, or are taking active steps to obtain it
  3. The remaining 36 (50% of respondents) have no data, and no firm plans to obtain any. (By pure chance, the above results offer an easy-to-remember 1-2-3 ratio).

Other Comments

The following additional observations are made, based upon the responses received:

The range of variation between estimates in some health areas and actual measured incidence in some other areas is immense, varying at the extremes by up to several hundred-fold the authorities with the most proactive monitoring are those with the highest rates several sources have reported a bulge in the figures for pre-school years one health authority has reported an extremely high rate of prevalence, of 1 in 69, amongst three year old boys, the result of a targeted audit of boys of this age across an entire health district over a two year period several other health boards/authorities have reported very high rates of prevalence, with rates of well over 1/200 for boys a number are uncertain as to whether it is a matter of better diagnosis or increased prevalence. A significant number believe there is increased prevalence evidence of very high rates matches evidence found by other researchers (e.g. through LEA research), and points to rapid increases in numbers diagnosed in recent times, though apportioning this apparent increase accurately between greater incidence and improved diagnosis would require a detailed study

Detailed Responses From Health Authorities, Boards, Etc.

Health Authorities South East England

These responded as follows (text is close to verbatim as possible, but summarised):

Berkshire: has funded project with Berkshire Autistic Society to examine incidence. Has no data, so cannot comment if has been increase.

Buckinghamshire: recently produced strategy on Mental Health of Children & Young People, 1998-2002. This revealed that local data on prevalence of many different mental health problems, including autism, was not available Currently uses published national studies to estimate local prevalence.

East Kent: no response.

East Surrey: alone amongst all the health authorities contacted, East Surrey has undertaken a positive audit (trawl) of all three year old children, to find out how many are ASD (see later). In former Mid Surrey area, team identified 23 children age between 0-5, in 1996, and 52 in 1998. Background population of 8,000 equates to rate of incidence of 1/154, boys plus girls combined. In remainder of area (East Surrey district), estimate of at least 50 cases aged under 5. This equates to rate of at least 1/250. Special needs audit of all children aged 3 by community paediatrician has resulted in 36 children being identified during 1998-99, of which 29 were between 2-3 and the remainder just over three. Background population is about 2,500 children born per year. Prevalence rate therefore equates, according to East Surrey Health Authority, to 1/139 for boys plus girls combined, but is 1/69 for boys only of this age. View of local professionals is that incidence is increasing.

East Sussex Brighton & Hove: uses epidemiological data and applies this to local population. Data suggests incidence of up to 1/500. Likely to be between 715 and 3,526 ASD children in HA area. NHS Community Trust maintains National Child Health Computing Systems which incorporates a special needs module to record details of those children. Education Department has data on numbers of children meeting criteria for special schools. Implication was that neither of these gives complete picture.

Isle of Wight: no response.

North & Mid Hampshire: no response.

Northamptonshire: see Anglia & Oxford section, later.

Oxfordshire: see Anglia & Oxford section, later..

Portsmouth & South East Hampshire: uses published studies to assess likely numbers of cases. Has local information on new referrals, for example, 40 referrals in six months following 11/98, though not all will confirm as ASD. This 40 comprises 32 boys, 8 girls.

Southampton & South West Hampshire: no response.

West Kent: work currently taking place on disabilities, including autism. Community paediatricians being asked to estimate autism numbers in each locality. Estimates 500 cases in West Kent. Estimated prevalence of 1/333 to 1/500.

West Surrey: does not collate information at health authority level. Collected data during 1998 on all children seen during one week by child and adolescent mental health services (485 cases), but none were autistic. Confirms view that condition remains rare.

West Sussex: uses national research to estimate numbers. Would expect to have 370 cases in child/teenage population. Checks this by collecting data from variety of sources, including own local register of children with special conditions. Identifies 102 cases this way. West Sussex Education Department has identified 427 cases aged 5-19 (equates to 1/288 rate against background population). Has agreed joint working definition and approach to assessing needs.

Health Authorities South Thames

These are all listed under South East, above, or London, below.

Health Authorities North Thames

Most of these are listed under London, below. The remainder are:

East & North Hertfordshire: did not respond to original inquiry.

North Essex: no response.

South Essex: no response.

West Hertfordshire: does not hold details on children with autism. Will be reviewing services in light of National Service Framework.

Health Authorities London

These responded as follows:

Barking & Havering: replies failed to address questions. Appears not to have data or monitoring systems.

Barnet: no response.

Bexley & Greenwich: have impression they are seeing more referrals. Have good data for Borough of Bexley through local paediatrician in Oxleas NHS Trust. District assessment service has resulted in large increase in referrals. In three years 9/96 to 4/99, 109 cases diagnosed plus 50 more awaiting assessment. The incidence rate in Bexley equates to 1/333, but is presumably much higher for boys, probably 1/200.

Brent & Harrow: incidence of autism is not routinely or comprehensively monitored locally. National Information Strategy will make monitoring possible in next five years or so. Authority has commissioned a report on needs of ASD children, through NAS, which includes detailed estimates of incidence. The Harrow study, A Report On The Needs Of Autistic Spectrum Children and Adults and Their Families In Harrow, Simon Jarrett, National Autistic Society 1998, acknowledged that until a detailed counting exercise was undertaken, it was impossible to confirm nationally-based prevalence rates. Harrow Social Services has a learning disability register, but Jarrett noted that this did not include the under-fives, children with Aspergers (unless they also had significant other problems), children with only a mild degree of autism, children placed in mainstream schools and any children whose referral to the learning disability team had been turned down. The report commented that parents and professionals were virtually unanimous in their view of there being high numbers of ASD cases.

Bromley: is in process of collating information on numbers with disabilities, including autism. Data sources include health trusts, health visitors, local schools, local authority and voluntary groups (see also Bromley Social Services, below)

Camden & Islington: no response.

Croydon: no response.

Ealing, Hammersmith & Hounslow: has no current mechanisms to monitor incidence. Data only comes from hospital admissions, which does not provide a good estimate. Diagnostic criteria for autism are unclear, so difficult to determine whether increasing or not.

East London & City: no response.

Enfield & Haringey: has attempted to assess numbers of ASD children. Gathered data from local authority special needs registers. Neither authority feels these are accurate, but these give an estimate of 75 ASD young persons (this would give incidence, boys plus girls, 0-19, of 1/1372). Feel there has been increase in recent years. Haringey Council points out that they have high numbers of pre-school children registered as having autistic tendencies, compared with the number of school age children.

Hillingdon: has detailed data on ASD children, based on data from education and health, that identifies numbers in contact with local services. Does not regard this as robust estimate of incidence, especially for milder disorders.

Kensington & Chelsea & Westminster: no response.

Kingston & Richmond: no response.

Lambeth, Southwark & Lewisham: no response.

Merton, Sutton & Wandsworth: no response.

Redbridge & Waltham: works closely with partners, but only expects to be contacted when there was a health need, rather than notified of each case. Does not appear to monitor, nor have data.

Social Services Departments, London

These social services responses are listed here for geographical reasons:

Barking & Dagenham: no response.

Barnet: no response.

Bexley: no response.

Brent: no response.

Bromley: measures incidence through a voluntary agency, Bromley Autistic Trust. BATs estimate of ASD numbers is 185, out of child population 0-19 years of 69,000, which equates to overall rate of 1/373. BAT figures for each year of birth are 3 (1980), 2 (1981), 5 (1982), 6 (1983), 7 (1984), 9 (1985), 7 (1986), 17 (1987), 13 (1988), 4 (1989), 17 (1990), 19 (1991), 16 (1992), 20 (1993), 16 (1994), 10 (not complete total)(1995). If these years are grouped together into blocks of five, then the numbers for 1985-89 show a 120% increase over 1980-84, and the numbers for 1990-94 show a 280% increase over 1980-84.

Camden: no response.

City of London: has very small population, and does not consider its local statistics to offer valid comparisons. Not clear whether monitors systematically or not.

Croydon: no response.

Ealing, Enfield, Greenwich, Hackney, Hammersmith & Fulham, Haringey, Harrow, Havering, Hillingdon, Hounslow, Islington, Kensington & Chelsea, Kingston-upon-Thames, Lambeth: no response.

Lewisham: developing strategy for disabled children, including reliable and accurate data on nature and extent of disabilities. Data held by community health service and social services disabilities register does not include autism. Children with autism may be classified under a number of headings. This makes it difficult to provide accurate statistics. This is a source of concern, because of apparent growth in numbers of autism cases. Consultant paediatrician is currently collecting and analysing data. Analysis of social services data suggests that autism cases feature significantly on waiting lists for services.

Merton: no response.

Newham: feels that autism is increasing. Is currently working with Department of Health on methods of calculating prevalence of child disabilities. Education authority has provided data on 143 autism cases. Five-year group aged 13-17 totals 27 cases. Five year age group aged 8-12 totals 38 cases. Five-year age group aged 3-7 totals 73 cases, or nearly three times higher than oldest five-year group.

Redbridge: not prepared to provide detailed response.

Richmond-upon-Thames: no response.

Southwark: no information gathered on incidence. Not possible to give accurate indication of any increases.

Sutton: referred correspondence to SW London Community NHS Trust, which provided details of 30 cases (28 boys, 2 girls). Distribution of cases was aged 15-19 (2), aged 10-14 (13), aged 5-9 (14), implying sharp rise in incidence. Sutton Social Services has contracted-out running of Children Act register to Trust. Data position likely to improve.

Tower Hamlets: no response.

Waltham Forest: no detailed response.

Wandsworth: no response.

City of Westminster: has 35 children with statements primarily relating to autism (as at 7/99), whereas as at 12/98, only 24 children were identified.

Health Authorities South & West England

These responded as follows:

Avon: has no data on numbers with autism. Need for better data gathering is acknowledged. Local education authorities indicate increase in numbers. Has estimated up to 383 cases of ASD, based on national data.

Cornwall & Isles of Scilly: health, education and social services are working together to establish user profile. Autism is included. Joint agency work commenced 4/99. No data at present.

Dorset: does not have accurate data. Is setting up inter-agency group to assess scale of autism. Certainly the view that demand for support services for autistic children has increased.

East Somerset: incidence has increased enormously because diagnosis criteria have been widened. Has 44 cases that fit the four essential diagnostic criteria. The figure of 50-60 children in special education, plus 200 in mainstream education, total 260, would be accurate, plus additional Asperger cases. Offers comment that, although Health Minister has stated we also encourage (authorities) to work together, there is no evidence of this encouragement.

Gloucestershire: no response.

Isle of Wight: no response.

North & East Devon: undertook full survey of Aspergers in 1997, using combined data from health, education, social services and voluntary sector, but only as one-off exercise. Suggested numbers in range of 150-200.

North & Mid Hampshire: no response.

Somerset (see also East Somerset): most severe cases recorded through multi-agency list. Of 399 cases of children with severe disabilities, 28 are autism, 11 are autistic tendencies, 6 Aspergers, total 45 cases. Figures later updated as 43 autism, 19 autistic tendencies, 10 Aspergers, total 72 (reason for higher numbers not clear). There was also a 1996 report from the Childrens Strategic Planning Team in Somerset, which indicated 50-60 cases in special schools plus 200 in mainstream schools, with ratios of 4 boys to 1 girl. The Somerset figures would equate to a prevalence rate of 1/388, but higher for young boys.

South & West Devon: no response.

Southampton & South West Hampshire: no response.

Wiltshire: numbers not being monitored, though does appear to be a rise.

Health Authorities West Midlands

These responded as follows:

Birmingham: Birmingham Childrens Hospital maintains data on prevalence. Study looked at incidence/prevalence in two areas, for children under age 5 who had received diagnosis before 5th birthday. Shows large increase in the incidence of autistic spectrum disorders. Incidence rose from 1/2857 in 1991-92 to 1/763 in 1995-96. The incidence of ASD increased on average by 37% per year. The cumulative incidence to the 5th birthday for the 1991 birth cohort was 1/238. Study does not say this, but if there were 5 boys to 1 girl in this, boy rate would be 1/144.

Coventry: has established assessment service which has assessed 90 children in first 11 months. Is receiving 10 referrals per month.

Dudley: no response.

Herefordshire: no response.

North Staffordshire: no response.

Sandwell: does not have data. Is attempting to collect data from other agencies.

Shropshire: has been looking with education and social services at incidence of ASD and comparing it with previous exercise in 1991. Numbers are small, but growth in incidence from 1-2 new cases per year in 1991 to 4-5 per year in 1999.

Solihull: no response.

South Staffordshire: does not have data. Has apparent incidence of 1/250, according to local data within HA area. Has noted in local school a bulge of younger cases. Speech therapists have also identified this. Underlines need for register. The 1999 Review of the Childrens Services Plan includes target of improving knowledge of incidence of autism.

Walsall: currently 95 children in Borough of Walsall (not known if boundary coincides precisely with Health Authority) with ASD, plus 8 under investigation, total 103. There are 50,029 pupils, giving an estimated prevalence of 1/526. Pre-school children are monitored, but this reduces at school age, though school doctors are invited to annual multi-disciplinary school reviews.

Warwickshire: reviewed incidence in mid-1990s. Further monitoring being undertaken. Has hitherto used estimates based on published prevalence rates.

Wolverhampton: has not formally assessed numbers. LEA assesses between 50-60 school age children, out of school population age 5-15 of 34,045, which would give rate of 1/567. Has no system of monitoring incidence. Is intending to develop inter-agency approach to meet needs, but no timescale.

Worcestershire: LEA informs health authority it has 1,987 special needs places in mainstream (note) schools for children classified as having ASD. School age population age 5-16 is 80,100, which equates to 1/40, an extremely high rate (special needs schools places would raise this rate even higher?)

Health Authorities Anglia & Oxford

These responded as follows:

Bedfordshire: no response.

Berkshire: has funded project with Berkshire Autistic Society to look at prevalence. Does not have process at present for counting actual numbers. Relies on provider trusts to highlight any change in incidence. Data also may be with Unitary Authority (education data). Working closely with partners to identify needs.

Buckinghamshire: local data not available. Uses published rates to estimate incidence.

Norfolk: authority works with other sectors to develop annual childrens services plan. Latest plan has specific commitment to review and improve services for ASD children. Consultant appointed to quantify level; of needs. Work continuing.

North West Anglia: no response.

Northamptonshire: data on significant developmental disorders is collected as part of routine child health surveillance. Data was not supplied for Northants North, but Northants South indicated 95 ASD cases, giving a prevalence of 1/666.

Oxfordshire: no response.

Suffolk: no response to initial inquiry.

Health Authorities Trent

These responded as follows:

Barnsley: no monitoring undertaken. Does not know if there has been an increase. Has a number of local registers which include autism, and would ()if assessing incidence) check these against published studies.

Doncaster: has no reliable estimate. Notes that many milder cases do not attend special schools and do not present themselves to health services.

Leicestershire: local learning disability register only includes adults. No data on children. Aware of significant increase in Aspergers cases.

Lincolnshire: does not have monitoring systems in place, but states that Childrens Act Register and revised Learning Disability Register should now begin to address this. Unable to assess whether increase, due to lack of data. Aware of clear rise in demand for specialist services for autistic people.

North Derbyshire: no response.

North Nottinghamshire: recent review group looking at various services in response to autism and other disorders. Monitoring of incidence being undertaken by Central Nottinghamshire part of HA. From information obtained, annual incidence will be 1/10,000. This represents an increase over last five years. Is continuing to monitor incidence.

Nottingham: no response.

Rotherham: does not have data readily available. Not aware of monitoring system. Sources that would be used to assess incidence are hospital admissions data and local authority learning disability register. Any more detailed assessment would require a special study.

Sheffield: no response.

South Humber: refused to respond.

Southern Derbyshire: no response.

Health Authorities North West England

These responded as follows:

Bury & Rochdale: has no data. Uses two published rates of incidence, then applies these to the total local population. This implies that there are between 8 and 348 cases in Bury, and between 10 and 434 cases in Rochdale (no typing errors here).

East Lancashire: does not have data on incidence of autism. Aware of increase in Aspergers, but ascribes this to better assessment.

Liverpool: discussions with this HA are continuing.

Manchester: does not have database. Relies on national prevalence statistics. Aware of concerns in education and social services about increasing numbers.

Morecambe Bay: no response.

North Cheshire: has jointly commissioned (with South Cheshire) the Cheshire Autism Project, through the NAS, to gain a clearer picture of actual numbers of ASD cases, and other key objectives, such as implications for service delivery. Study to report late 2000.

North West Lancashire: no response.

Salford & Trafford: does not have register of people with autism. Has disability register for Salford. Knows from studies that prevalence of autism disorder is around 1/25,000 (no misprint here). Felt to be an increase.

Sefton: does not have strong epidemiologically validated data on autism. Does not have system for monitoring.

South Cheshire: is participating in joint study, with North Cheshire (see above).

South Lancashire: has no specific routine monitoring system. Information on learning disability cases would include autism cases. Data on children under 5 is gathered by Child Development Centres. These indicate 12 children under 5 with recognised autism. Within the school population, rely on statementing process. This indicates that there are 43 cases across Chorley, South Ribble, West Lancs and Preston (background population not supplied, so incidence rate not available).

St Helens & Knowsley: does not have a database on ASD incidence. Informal discussions indicate increase in either diagnosis or prevalence. Agrees as result of correspondence that a co-ordinated look at services for children with autism would be useful.

Stockport: does not routinely monitor the incidence of autism.

West Pennine: monitoring is undertaken through Child Handicap Register. Has reviewed data as result of correspondence. Data to confirm whether increase or not is not available. May attempt more sophisticated data collection in the future.

Wigan & Bolton: no routine statistics collected. Could extrapolate from national data to suggest what might be expected locally.

Wirral: have not carried out specific local needs assessment for autism. Aware of local increase in referrals for specific autistic problems.

Health Authorities Northern & Yorkshire

These responded as follows:

Bradford: no response.

Calderdale & Kirklees: does not monitor incidence of autism. If it did, it would seek information from local hospitals or GP practices.

County Durham: assesses incidence by monitoring use of speech/language therapy services, school health services, social service agencies, and setting these against national rates. Recognises this is not accurate.

East Riding: uses national rates of between 4 and 18 cases per 10,000 (1/2,500 to 1/555), and applies this to local population. Knows that about 6 new cases present in Hull each year.

Gateshead & South Tyneside: no response.

Leeds: no response.

Newcastle & North Tyneside: no response.

North Cumbria: there is no routine collection of data. Impression of increase in awareness, and definite increase in referrals. Education Department in Cumbria also reporting increased referrals.

North Yorkshire: no formal mechanism for routinely collecting data on numbers of ASD children. Specific joint agency project was carried out in mid-1990s (date not given) with local partners, in North Yorkshire County Council part of HA area (excludes City of York). This produced a figure of 254 cases against a background child population of 143,200, giving an incidence rate of 1/564. Of 202 of these, in a sub-area, 169 were male and 33 female (ratio of 5:1). Of the larger total of 254, 61 were autistic, 46 Aspergers, 44 atypical autism, 103 not specifically classified into any particular ASD category. Within the group, there were 71 cases aged 11-16, but 104 cases aged 5-10.

Northumberland: no response.

Sunderland: no response.

Tees: no response.

Wakefield: autism is not routinely monitored. Issue being considered as result of recent education department work (LEA reports increase from 5 to 111). Has identified autism as one of a number of conditions that has apparently increased.

Health Boards Scotland

These responded as follows:

Argyll & Clyde: no response.

Ayrshire & Arran: has very limited data, from Continuous Morbidity Recording System. Only 3 GP practices (out of 62) feed into this. This indicates a prevalence rate of 1.6/1000 (1/625) for age group 0-4, no cases for 5-14 and 0.5/1000 (1/2000) for age 0-14 overall.

Borders: assessment of numbers of ASD children is currently (9/99) being undertaken. Reports that Scottish Office has funded the health service in the Borders to undertake some specific work on the multi-agency approach to the treatment of autism in children. This work will be evaluated in late 2000.

Dumfries & Galloway: currently does not assess incidence. Estimated that there are 50-60 children with autism diagnosis. Population age 0-18 in HA area is 35,000, which would indicate rate of 1/583. A number of initiatives are under way to identify incidence more accurately. These include multi-agency/multi-disciplinary group set up to report on developing a register on the incidence of child autism.

Fife: no response.

Forth: no response.

Grampian: does not have a system for assessing the incidence of child autism. Autism and related disorders have recently been producing increased demand for referrals and services. Believes this to be due to heightened awareness and raised parental expectations.

Greater Glasgow: has not performed any specific surveys and does not have any time-trend data. Uses published rates to estimate number of cases. Also monitors activity at child development centres, mental health services and the Scottish Centre for Autism, and this gives some indication of numbers of ASD children using services. Is about to undertake (as at 9/99) joint agency work to assess numbers of special needs children known to health, education and social services. This will provide some indication of prevalence of autism, by age and sex.

Highlands: no response.

Lanarkshire: no response.

Lothian: has special needs register, which aims to include all ASD cases, though some of the less severe cases are thought to be missing. Currently there are 89 cases with autism plus 68 cases with other ASD disorders, total 157, against a background population of 143,379, giving an incidence rate of 1/913. There has been a recent increase in referrals.

Tayside: no response.

Orkney: no response.

Shetland: there are 13 autistic children, born in the years 1987 to 1996, but drawn from records covering children born in the years 1980-99. All the cases come from this 1987-96 subset of the years. No cases are listed as being born 1997 onwards. No children appear as being born 1980-86. The children are aged 12 (two cases), 11 (one case), 9 (one case), 8 (two cases), 7 (two cases), 5 (two cases), and 3 (three cases). These divide up amongst the Shetland child population data (1996 figures are the latest available) as age 0-4, three cases, including one female, plus any more that may be on the way, amongst 1,621 population: age 5-9, seven cases, all male, amongst 1,722: age 10-14, three cases, including one female, amongst 1,572: age 15-19, no cases, amongst 1,596. I have re-calculated the figures so as to set boys-only autism numbers against one-half of the child population for each age band. This results as follows: age 3-4 (forget ages 0-2, as it is too early for diagnosis), you have 2 boys against a population of 324: age 5-9, you have seven boys against a population of 861, giving a rate of 1/123: age 10-14, it works out as 1/393 for boys: age 15-19, it is nothing, as there are no cases.

Western Isles: no response received. However, there are understood to be 9 boys (plus one girl) cases of autism, all under age 12. There are no cases at all in the 13-19 age group, which matches the Shetland experience. The WI boys rate works out at 9/1983, or 1/220, but this includes boys from age 0, so the rate for ages 3-12 will be higher, probably 1/200.

In addition to the above, the September 1998 Scottish Schools Census included questions on numbers of pupils (male/female, primary/secondary categories) with a primary diagnosis of autism. This listed 323 Recorded (= Statemented in England & Wales) autistic pupils in primary, 71 in secondary and 378 in special schools, total 772. As a percentage of total schools population, this was 0.1%, or 1/1000. Non-recorded (non-statemented) autistic pupils numbered a further 172 in special schools, grand total 944. Note that these figures will exclude children with only a secondary diagnosis of autism, that they relate to 9/98, and that they of course exclude pre-school children.

Health Authorities Wales

These responded as follows:

Bro Taf (HQ Cardiff): details awaited.

Dyfed Powys: no response.

Gwent (HQ Pontypool): Welsh health authorities have received no specific guidance from the National Assembly for Wales on conducting the health needs assessment for people with autism. HA is developing a strategy for autistic children. A surveillance system to monitor the incidence of autism will be considered as part of this strategic planning work, but does not exist at present (10/99).

Morgannwg (HQ Swansea): information is held on child health registers for each of the three unitary authorities in the HA area (Bridgend, Swansea, Neath & Port Talbot). A 1996 investigation in Swansea and Neath & Port Talbot areas found 40 cases registered, out of a child population age 0-19 of 92,876, equating to 1/2,322. There is no trend data.

North Wales (HQ Caernarfon): has set up strategy to look at autism in North Wales, including establishing monitoring systems and co-ordinating agencies. A recent local study has suggested prevalence rates similar to national rates (unspecified).

Education & Library Boards, Northern Ireland

To offer a variation from health and social services departments, the Education and Library Boards were approached in Northern Ireland:

Belfast: at present, no data on diagnostic basis. However, records contain information on the medical conditions of each child who is referred for statutory assessment. Board conducted limited exercise in 12/98, when ASD numbers were extracted from educational psychologists caseloads. Ample anecdotal evidence of increased numbers. There has also been a University of Ulster report on autism diagnosis, no further details provided.

North Eastern: assesses incidence through Stages 3-5 of Code of Practice on Identification and Assessment of Special Educational Needs. View that there is increase in identification of special needs, but this does not necessarily imply actual increase in numbers.

Southern: do not have children categorised as autism, though some statements recognise autistic tendencies. Some children have autistic tendencies but do not have statements. Has a (statemented) figure of 100, against a background of 70,000 children, equating to 1/700. E&LB is currently (4/99) discussing with Health Board the possibility of setting up a diagnostic/development unit for autism, which would provide more definite statistics.

South Eastern: management information system not sufficiently sensitive to provide accurate information on incidence or prevalence of autism. Could produce some information by circuitous means, but not possible to guarantee accuracy. Computer system being updated so that data on autism and other conditions can be extracted more easily.

Western: has data on children with autism (or other conditions) as primary disability. Does not include cases where autism is secondary disability. Impression is that numbers with ADS or Aspergers has increased noticeably within past decade, particularly milder-condition cases. Data includes 41 cases of ASD and 43 of Aspergers, total 84 cases, divided up amongst different categories of school. Background population not supplied, so incidence rate not able to be calculated).

The Department for Education, NI, was also approached. During the period 1990-97, the numbers of children with statements of special educational needs increased from 5504 to 8806, (= approximately 60%). The Department does not hold a breakdown of statement numbers by category of disability, so (as with its Whitehall counterpart) has no information on the number with ASD.
Top of Page

Data In Context Of Published Studies

The above data can be set against the following prevalence rates quoted by Lorna Wing in The Definition and Prevalence of Autism: A Review, published in European Child and Adult Psychiatry, Volume 2, issue 2, April 1993, pages 61-74

(The four with the lowest rate of incidence are marked * and the four with the highest rate are marked **):

Kanners Criteria

Lotter, 1966, 1967, Middlesex – 1 in 2222

Brask, 1972, Aarhus, Denmark – 1 in 2326*

Wing & Gould, 1979, Camberwell, London – 1 in 2041

Hoshino et al, !982, Fukushima-ken, Japan – 1 in 2000

Rutters Criteria

Bohman et al, 1983, Vasterbotten, Sweden – 1 in 1786

Cialdella & Mamelle, 1989, Rhone – 1 in 926

DSM-III Criteria

Matsuishi et al, 1987, Kurume, Japan – 1 in 645**

Tanoue et al, 1988, Ibaraki, Japan – 1 in 725**

Sugiyania & Abe, 1989, Nagoya, Japan – 1 in 769**

Burd et al, 1987, North Dakota – 1 in 3030*

Ritvo at al, 1989, Utah – 1 in 2500*

Gillberg, 1984, Goteborg region, Sweden – 1 in 2500*

Steffenburg & Gillberg, 1986, Goteborg city – 1 in 1333

DSM-III-R Criteria

Bryson et al, 1988, Nova Scotia – 1 in 990

Gillberg et al, 1991, Goteborg city – 1 in 869

Other Criteria

Ishii & Takahashi, 1983, Toyota, Japan – 1 in 625**

David Thrower ©

49, Ackers Road, Stockton Heath,



England WA4 2DZ


Telephone (+44) (0)1925-264156

February 2000