support document

Since Kanner first described the condition of autism there have been many thousands of publications dealing with this disorder.  However, relatively few of these have a focus on adults, and fewer still have systematically studied the progress of individuals with autism as they move through adolescence and into adulthood.

The earliest follow-up studies of individuals with autism were largely anecdotal and unsystematic.  However, in 1973, Kanner reported on 96 individuals, most of whom were in their 20s or 30s, whom he had initially seen as children.  The majority was highly dependent on others for their support, and most lived either with their parents, in sheltered communities, or in state institutions.  Nevertheless, amongst those who were most able, around half were functioning relatively well, with several having achieved high levels of employment, some living independently, and one man had married and had a family.

More recently there have been accounts written by high functioning individuals within the autistic spectrum themselves (e.g. Grandin, 1995; Gerland, 1996; Holliday-Willey, 1999; Jolliffe et al.,1992; Lawson, 2002; O’Neil, 1999; Williams, 1992).  These offer considerable insight into the difficulties experienced by individuals with autism, and also provide much helpful information on how lives can, given appropriate support and understanding, be significantly enhanced.

However, the most detailed information on general outcome in adulthood comes from systematic follow-up studies of individuals from childhood into adulthood.  The first prospective outcome studies were conducted by Rutter and his colleagues (Lockyer & Rutter, 1969; 1970; Rutter and Lockyer, 1967; Rutter, Greenfield and Lockyer, 1967). They followed-up 38 individuals, first diagnosed with autism at the Maudsley Hospital in 1950s or 60s, into adult life.  Subsequent follow-up studies have been conducted in a number of different countries, and include those of Lotter (1974) in the UK, Gillberg and Steffenberg in Sweden (1987), Kobayashi, Murata and Yashinaga in Japan (1992), and several US and Canadian based studies (e.g. Ballaban-Gill, Rapin, Tuchman & Shinnar, 1996; Rumsey, Rapport & Sceery, 1985, Venter, Lord & Schopler, 1992; Szatmari and colleagues, 1989).

Comparisons between studies are made difficult because of the different ages of the groups involved, their varying cognitive levels, and the wide range of measures used for assessing outcome.  Nevertheless, examination of the findings from these studies over time seems to suggest that outcome has improved over recent decades.  For example, comparing studies published prior to 1980 with those published in the last 2 decades, the average proportion of individuals in work has risen from around 5% to 25%, and the percentage living in their own homes has increased from around 2% to 4% prior to 1980  to between 10% and 15% in the last 2 decades.  There has been a major decrease, too, in the numbers of individuals living in institutional care.  In the studies reported before 1980 well over 50% of individuals were living in institutions as adults. Post 1980, this proportion had fallen to under 10%.  Outcome studies have also revealed certain factors that seem to be related to outcome.  On the whole, individuals who show greater severity of autistic symptomatology in childhood (e.g. as measured on the Autism Diagnostic Interview-Revised, Lord, Rutter and Le Couteur, 1994) have a poorer prognosis, as do those with epilepsy.  However, epilepsy also tends to be associated with lower IQ.  In most of the follow-up studies conducted women have tended to do less well than men, although on the whole the numbers of females are so small that the differences are not generally statistically significant.  The two variables that do seem to be most closely related to outcome are IQ – hardly any individuals with an IQ below 50 achieve independence in adulthood – and language.  Thus, most individuals who make good progress as adults had at least some useful language by 5 to 6 years of age.

In a recent study conducted at the Maudsley Hospital (Howlin, Goode, Hutton and Rutter, submitted) we followed-up a group of 68 individuals, all with a childhood IQ of above 50. When first seen they were aged, on average around 7 years, and at follow-up were almost 30 years old. Their average IQ on performance tests when assessed as children first seen was 80 and scores had remained remarkably stable into adulthood. Despite the fact that this was a relatively high functioning sample, only around 20% had obtained formal qualifications at school, and just over 30% were in work (either independent or supported). Two-thirds were reported by their parents as having no friends and only 10% of the group was reported as being able to live independently or even semi-independently. Overall ratings of social competence in adulthood indicated that approximately 12% were considered to have a good or fairly good outcome, just over 20% had a poor outcome and  47% were rated as having a very poor outcome. A further 8 individuals remained in hospital care as adults because no other suitable place could be found for them.

Examining the factors that appeared to predict progress from child to adulthood it was apparent, as has been found in many other studies, that initial IQ was a very powerful indicator.  Only one individual with a childhood IQ of less than 70 was rated as having a good outcome in adulthood. Nevertheless above this level, outcome remained very variable.  Even amongst individuals with an IQ above 100 when initially seen, only 40% were rated as having a good outcome, the same proportion was rated as having a moderate outcome and 20% were rated as having a poor or very poor outcome.  Early language ability was also significantly related to subsequent outcome but unlike performance IQs, which had generally remained remarkably stable over time, verbal abilities often showed much greater change.  Thus, almost half of those who had been quite unable to score on verbal tests when young went on to show significant improvements in their communication skills over time.

It is evident from this and other studies of outcome that although some individuals do do remarkably well as adults, the majority does not.  Whilst progress is affected by the degree of cognitive impairment, this is not the only factor affecting outcome, and indeed work by other groups (e.g. Venter et al., 1992) suggests that one of the most crucial factors influencing outcome is the degree of support that individuals receive as they make the transition from child to adulthood.  How then might we improve this support in ways that have a marked impact on individual progress?

One way, of course, is to improve educational facilities generally.  Much is known about the teaching strategies that are most helpful for children with autism (cf Schopler, 1997) and these are now widely employed in schools with a specific focus on children with autism.  With the increasing move to inclusive education, however, many children will not have access to specialist provision. It is clear that if children’s progress is to be enhanced, much more training and support for teachers is needed, both in order to increase their understanding of the needs of pupils with autism and to help them put in place the support structures that are known to minimise behavioural difficulties and increase learning (Howlin, 2002; Jordan and Jones, 1999).

Whether or not they have autism, if students fall behind in school, or fail to obtain qualifications commensurate with their intellectual level, their chances of success in adult life are likely to be very limited.  In the absence of formal qualifications, job prospects are extremely limited; without a job social opportunities are greatly reduced and in the absence of a regular daily routine the structure and predictability that seem essential for autistic individuals to function well are removed. All this results in a life of emptiness and boredom, which in turn increases the risk of ritualistic, stereotyped and non-functional behaviours. Similarly, isolation from the peer group can frequently lead to low self-esteem, and sometimes to severe depression and even suicide.

Moreover without adequate support and guidance for parents in the early years, behaviours that were not particularly difficult to cope with in childhood may present major problems in adult life.  Thus, although levels of offending amongst individuals with autism or Asperger in adult life appear to be relatively low (Ghaziuddin, Tsai and Ghaziuddin, 1991) if difficulties do occur these can often be traced back to early childhood.  For example, an obsessional interest in woman’s earrings, or people’s feet, may be quite acceptable in a 3-year-old.  If a 20 or 30 year old is still approaching any woman whom he sees wearing earrings, or freely removes people’s shoes in order to feel their feet, the problems become to be viewed very differently by society as a whole.  Fascinations with trains, washing machines or lighting systems on public transport may cause only mild difficulties for a young child who is closely supervised by his or her parents.  However, for older individuals who insist on pursuing these interests whenever they have the opportunity very serious difficulties can arise.

Problems can occur, too, because of the lack of social understanding of individuals with autism.  Often, as they grow older they become very aware that other people of their own age have friendships or close sexual relationships, and understandably they wish to be the same.  However, without understanding the highly complex, and unwritten rules governing social interactions they may easily overstep the boundaries of acceptable behaviours, and again, on occasion, find themselves in serious difficulties.  This wish to be accepted by their peer group can also result in their being very vulnerable to unwanted social pressures.  For example, it is not uncommon for young people trying to socialise with their peers, to be inveigled into activities such as lending them large amounts of money (which are never repaid), giving away expensive belongings, or even occasionally, being tricked into carrying out illegal acts for them, such as shop lifting or taking and driving away cars.

Although it is never possible entirely to prevent the occurrence of social difficulties in adulthood, much can be done to prevent major difficulties if, from an early age, parents recognise that behaviours that may be viewed as clever, charming or “cute” at 3, may result in major difficulties when individuals reach their teens or adulthood.  Thus, it is much safer to establish firm rules in early childhood, about when, where, and with whom certain activities may take place, than it is to have a very liberal approach to behaviours in the early years and then have to severely limit these behaviours as time goes on.  Few individuals with autism are able to understand why a behaviour that was tolerated, or even encouraged at 3 years of age suddenly becomes totally unacceptable at the age of 13, and most, quite understandably, will be extremely reluctant suddenly to modify their patterns of behaviour for no apparent reason.

The difficulties in social awareness and social understanding associated with autism are often so profound that interventions such as social skills training or programmes to enhance theory of mind, can have only limited effects.  These strategies may be important in teaching individuals very basic skills, but they should not be expected to counteract the fundamental deficits.  Nevertheless, one potentially effective way of enhancing social interaction is to focus, not so much on individuals’ deficits in social understanding, but on any particular skills or abilities that they possess. If systematically encouraged such skills can, in time, help the individuals concerned  to make social contacts with others, which, in turn,  can enhance social status and  self-esteem.  Special interests or knowledge in topics such as the weather, transport, music, insects or even more arcane topics, such as television test cards, have all proved a means of meeting with other like minded individuals, and often prove a vital source of social interaction for older more able people.

For those who are less able, a focus on modifying the environment rather than the individual him/herself, may also prove more productive.  Educational and psychological research with individuals with autism demonstrates clearly that they function best in environments that are predictable, consistent, and at least to some degree under their own control (Schreibman, 2000).  Co-operation and progress are also likely to be enhanced if verbal messages are augmented by visual or other cues (Durand & Merges, 2001; Green, 2001).   Thus, those working with adults with autism need to be aware of these individuals’  need for routine and predictability, of their difficulties in coping with change, or even making choices of their own.  Problems in understanding language or social situations also mean that they are unlikely to enjoy reciprocal social contacts in the same way that non-autistic individuals with learning disabilities are able to do.  Thus, their care environment may need to be very different from that which is usually effective with other individuals with learning disabilities, and education for staff working with autistic clients is essential if they are to be provided with the type of environment they need.

One major factor preventing many individuals with autism interacting with their peer group, or achieving any form of financial independence, is their lack of employment opportunities.  Job prospects for any group of individuals with a disability are almost always poor, and unfortunately autism is no exception.  However, it has become apparent, particularly from supported employment schemes in the United States (McGaughey, Kiernan, McNally, Gilmore & Keith, 1994) that with appropriate help the job market can be successfully opened up to people with autism.   For example, recent research by Mesibov and colleagues (Keel, Mesibov & Woods, 1997) indicated that the majority of clients supported by the TEACCH organisation were able to find work, some in independent settings, some supported by a job coach in a small group enclave, and many, even of their most disabled clients could work on an outreach basis.  Similar findings were reported by Smith and her colleagues in Maryland (Smith, Belcher & Juhrs, 1995). Unfortunately, supported job schemes in the UK have been slower to follow suit, and those that do exist have generally focused on individuals with more general learning disabilities.  In recent years, a pilot scheme involving the National Autistic Society has demonstrated that, with appropriate support, individuals with autism can be helped into stimulating, productive, and sometimes high level jobs. Mawhood and Howlin (1999) found that amongst a group of high functioning individuals (average age 30 years; IQ and language abilities within the normal range) who received specialist support, over 60% were able to find work during a 2 year pilot project. Moreover, most were employed in “White collar” jobs such as computing, banking, or office administration.  In a comparison group of individuals who did not have access to such support, only around 20% found work, and the jobs that did exist were mainly low paid manual jobs, such as stacking supermarket shelves.  Since the completion of the pilot scheme, in the last 6-7 years over 90 new jobs have been found, with 81% of these being in clerical administrative or computing work.  In addition, at a time when short-term fixed contacts are becoming the norm, over half of the jobs offered have been on permanent contracts.

Because of the careful matching in this scheme between clients’ skills and the job vacancies available, very few difficulties arose in coping with the work requirements themselves and if difficulties did occur these generally concerned social interactions.  The solutions to such problems were based on the principles that have been demonstrated to be effective for individuals with autism of all ages i.e. a reliance on non-verbal cues and instructions (written lists, pictures, checklists etc) and a high level of structure, routine and predictability.  Clear and explicit guidelines, concerning work standards and behavioural and social requirements are also essential.  Finally, and possibly most importantly it is crucial that immediate feedback is given by line managers if behaviours or work output are, for any reason, unacceptable.

With support from job coaches an increasing number of companies are now employing individuals with autism, and the original London based scheme has recently extended to Scotland and the North of England.  Encouragingly, once a firm has taken on one person with autism, that company is often more than willing to employ other individuals with this condition, because of the benefits they can bring to the work place.  Autistic individuals do not waste time in idle gossip, do not spend hours on the phone talking to friends, are often willing to do the sorts of mundane routine jobs that other workers avoid and they generally prove highly reliable, honest, and hard working.

In summary, there is much that can be done to ease the transition from child to adulthood for individuals with autism.  However, there is still room for considerable  improvement.  Firstly, although the age at which diagnosis is being made appears to be decreasing, many children, particularly those who are more able, may not receive a diagnosis until well into childhood (Howlin and Asgharian, 1999).  Moreover, early diagnosis, of itself, is not enough and needs to be coupled with practical advice for parents on how to deal with behaviour problems, enhance communication skills, and reduce the risk of early obsessional or ritualistic behaviours leading to major difficulties as the child gets older.  Advice on how to develop children’s particular skills in ways that may ultimately promote social interactions is also essential.  In addition it is crucial to “educate the educators” concerning the needs of people with autism and the best ways of catering for these at pre-school, primary, secondary  and higher education levels. There must be much wider recognition, too, among social, health, and employment services of the needs of adults with autism, and progress towards the development of a much greater range of options for supported and semi-independent living and productive employment. More effective ways of improving access to social interactions, by means of social support or social skills groups, befriending schemes and the like, also need to be explored. Finally, the risk of mental health problems in this group must be better recognised, and more effective ways of avoiding psychiatric problems, or of dealing with these should they arise need to be developed.

Summary
Although admissions to hospital care have fallen, and expectations about the future for people with disabilities generally have risen over the years, specialist provision for adults with autism is still far less well developed than services for children. Overall, it is evident that the majority of individual with autism, regardless of their intellectual level, continues to experience many problems in adult life. However, it is also clear that outcome can depend crucially on the degree and appropriateness of support that is provided beyond the school years and into adulthood. Thus, although the focus of much recent research has been on the importance of early intervention programmes (National Research Council, 2001) true social inclusion will only be possible if the long-term needs of adults, as well as children with autism, are fully recognized and adequately supported.