Asperger Syndrome and Autism: A Comparative Longitudinal Follow-Up Study More than 5 Years after Ori |
May 13 2010 |
| Journal of Autism and Developmental Disorders |
| © Springer Science+Business Media, LLC 2007 |
| 10.1007/s10803-007-0364-6 |
Original Paper
Mats Cederlund1, 2 
, Bibbi Hagberg1, 2, Eva Billstedt1, 2, I. Carina Gillberg1, 2 and Christopher Gillberg1, 2
| (1) | Department of Child and Adolescent Psychiatry, Institute of Neuroscience and Physiology, Göteborg University, Göteborg, Kungsgatan 12, 411 19, Sweden |
| (2) | Queen Silvia´s Child- and Adolescent Hospital, Otterhällegatan, 12 A, 411 18 Göteborg, Sweden |
| Mats Cederlund Email: mats.cederlund@vgregion.se |
Received: 15 May 2006 Accepted: 18 January 2007 Published online: 6 March 2007
Keywords Asperger syndrome - Autism - Follow-up - Intellectual ability - Outcome - DISCO
Asperger syndrome (AS), described as “autistic psychopathy” already in 1944 by the Austrian paediatrician Hans Asperger (Asperger, 1944), did not appear as a diagnostic entity until the early 1980s after being reintroduced by Lorna Wing (Wing, 1981). The first set of diagnostic criteria were formulated by Gillberg and Gillberg (1988/1989). The Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), and the ICD-10 classification of Mental and Behavioural Disorders, did not publish diagnostic criteria for AS until well into the 1990s (APA, 1994, WHO, 1993). These criteria have been widely criticized (e.g., Leekam, Libby, Wing, Gould, & Gillberg, 2000; Miller & Ozonoff, 1997), and there is still no consensus as to how AS should best be delineated. The Gillberg´s six criteria for making the diagnosis of AS are based on Hans Asperger´s original publication and require major problems with social interaction (e.g., impairing difficulties making friends and understanding social cues), narrow interests, repetitive routines, speech and language (e.g., odd ways of using expressive language, odd speech, intonation problems, and difficulties understanding others), non-verbal communication problems, and motor clumsiness. At least 9 symptoms are required for a diagnosis of AS. The DSM-IV criteria for AS specify impairments in social interaction (defined as for autism), restricted, repetitive, stereotyped behaviour (defined as for autism), and absence of clinically significant delay in language or cognitive development, including self-help skills, adaptive behaviour and curiosity about the environment, in the first 3 years of life (see Appendices). Only 3 symptoms are required for a diagnosis of AS.
Over the years, there has been much discussion about whether or not AS and so-called High Functioning Autism (HFA) are separate, similar or identical conditions (Gillberg, 1998; Mesibov, Kunce, & Schopler, 1998; Wing & Potter, 2002). Although high verbal intelligence—including better verbal than performance skills on the Wechsler scales—and earlier language development have been noted to be more common in people with AS than in classic autism, no research has so far been able to show clear-cut differences between people diagnosed with HFA as compared with those who have been given a diagnosis of AS (Eisenmajeret al., 1998; Gilchrist et al., 2001; Howlin, 2003).
Very few reports on the outcome of AS have been published. They have referred to small or highly selected clinical case samples without comparison groups and have reported low levels of employment and social functioning (Green, Gilchrist, Burton, & Cox, 2000; Tantam, 1991; Tsatsanis, 2003; Wing, 1981). In this study psychosocial functioning includes aspects of employment/studying, relationship to others, independent living, and absence of psychiatric problems. Intellectual decline over time, as measured by the Wechsler scales, was reported in one study of the intermediate term outcome of AS (Nydén, Billstedt, Hjelmqvist, & Gillberg, 2001), but has not been observed in later studies. A tendency towards closing of the gap between superior verbal and inferior non-verbal skills over time has been observed (Szatmari, 2003, personal communication). Recent studies of the short-term outcome of AS have suggested a substantially better outcome than in autism, which may have been due to earlier and more effective interventions, or to other factors (Starr, Szatmari, Bryson, & Zweigenbaum, 2003; Szatmari, Bryson, Boyle, Streiner, & Duku, 2003; Tsatsanis, 2003).
Outcome in classic cases of autism has been investigated in a number of studies in the past. The rate of poor or very poor psychosocial outcome (isolated life with high degree of dependency on others) has been around 50% (e.g., Gillberg & Steffenburg, 1987; Howlin, Mawhood, & Rutter, 2000; Howlin, Goode, Hutton, and Rutter, 2004), and IQ has decreased over time (Billstedt, Gillberg, & Gillberg, 2005).
The present study is probably the first ever to present a long-term perspective on the natural outcome of a reasonably large group of males with AS. The background and associated factors of the AS group have been outlined in two previous publications (Cederlund & Gillberg, 2004; Gillberg & Cederlund,2005). The AS group was contrasted with a similarly aged group of males with classic autism/atypical autism followed up using identical/similar instruments.
One aim of the study was to examine the diagnostic stability over time within the autism spectrum, in this case the stability of AS as diagnosed on the basis of the criteria formulated by Gillberg and Gillberg (1989) and Gillberg (1991), and of autistic disorder and atypical autism as diagnosed according to the DSM-IV. Another aim was to estimate the frequency of severe psychiatric disorder (psychosis) in the investigated groups, given earlier reports of a high frequency of such disorders within the autism spectrum (e.g., Wozniak et al., 1997; Wolff, 1995). Finally, there has been controversy about the rate of criminal offence within the AS/HFA group particularly after the publication of reports documenting a high rate of these diagnoses among highly selected samples of offending psychiatric patients (e.g., Siponmaa, Kristiansson, Jonsson, Nydén, & Gillberg, 2001; Scragg & Shah, 1994), and so our third aim was to examine the prevalence of such offence committed by individuals with AS and autism in early adult life.
Some individuals with a diagnosis of Asperger syndrome in childhood appear to do well in adult life, and this has led some people to question the appropriateness of making the diagnosis at all. Could the diagnosis of Asperger syndrome actually contribute to problems (such as the supposed stigma of having a diagnosis at all) in certain cases? We consider that, if diagnoses are only made in individuals who have clinical impairment, this would not be an issue. In the present study we used the Global Assessment of Functioning (GAF) scale to score psychosocial functioning, so as to determine whether or not an individual, who fulfils symptom criteria for AS should also be given the diagnosis of AS.
The following hypotheses were tested: (i) diagnoses within the autism spectrum are stable over time, (ii) AS has a psychosocially better outcome than autism, (iii) better outcome in AS is attributable to higher IQ, (iv) intellectual ability declines over time in these disorders, (v) individuals with higher verbal IQ have the best psychosocial outcome, (vi) those with an earlier diagnosis have fewer problems in early adult life, (vii) there is a higher frequency of severe psychiatric disorders (psychosis), than in the general population, (viii) involvement with the police and the law is at the same rate as in the general population.
One-hundred males with AS, 16–36 years old, diagnosed at the Child Neuropsychiatric Clinic (CNC) in Göteborg, Sweden, in the years 1985–1999, were approached for inclusion in the follow-up study. They were born 1967–1988, and had been diagnosed with AS at ages 5.5–24.5 years. Mean age at diagnosis was 11.3 years (SD 3.8 years), which is in keeping with previous findings (Howlin & Ashgarian, 1999). They all fulfilled the Gillberg and Gillberg criteria for AS (Gillberg 1991; Gillberg & Gillberg, 1989), which had been in use at the clinic since 1985.
Criteria for inclusion in the follow-up study were (i) normal intelligence (IQ > 70) at diagnosis on the Wechsler scales (WISC-R (n = 40), WISC-III (n = 52) or WAIS-R (n = 8)), (ii) AS diagnosis made ≥5 years prior to follow-up, and (iii) age ≥16 years at the census date 30/06/2004. We included males only, since no more than 7 females meeting inclusion criteria were found. The AS group had been assessed on the Wechsler scales (Wechsler, 1974, 1992a, 1992b) at original diagnosis, and had had the following mean results: Full Scale Intelligence Quotient (FSIQ) 101.4 (SD 18.3), Verbal Intelligence Quotient (VIQ) 107.2 (SD 18.6) and Performance Intelligence Quotient (PIQ) 94.6 (SD 18.7) (VIQ > PIQ, p < 0.01).
Sixty-one individuals in the AS group had had no further contact with the CNC after the diagnosis had been made and information about results had been shared with the family. A further 8 had had no contact with the CNC for the past 5 years, but 31 had had one or several contacts in the recent past (i.e., 2001 or later). We have argued elsewhere (Cederlund & Gillberg, 2004) that the 100 individuals approached are fairly representative of all males with AS, whose parents applied for clinical assessment and help for them in the late 1980s and throughout the 1990s.
A group of 16 to 38-year-old males from a group with autistic disorder (AD) diagnosed with autism (n = 62) or atypical autism (n = 22, including 2 with disintegrative disorder) using the DSM-III (APA, 1980), and DSM-III-R (APA, 1987) criteria, before 10 years of age and followed-up with a similar protocol as the AS group at the CNC 1999–2002 (Billstedt, Gillberg, & Gillberg, 2005; Billstedt, 2006, personal communication), were used as a contrast group. These 84 individuals were all the males from a community-based study of autism, who were approached for inclusion in a longitudinal prospective follow-up study designed and carried out by the last three authors. We consider this autism group to be representative of all individuals with autism diagnosed in the 1960s, 1970s and 1980s in the community. The group was strikingly different from the AS group in terms of IQ, but they still represented a reasonable contrast group vis-à-vis our AS group as regards their social impairment. The Autism group had been evaluated in childhood with the following results: Severe Mental Retardation (SMR, IQ < 50) (n = 39), Mild Mental Retardation (MMR, IQ 50–69) (n = 31), Near Average Intelligence (NA, IQ 70–84) (n = 10) and Average Intelligence (A, IQ 85–114) (n = 4).
AS Group
There was no mortality in the original AS group of 100 males. A letter of information about the AS study was sent to all 100 individuals or their parents (in those under 18 years of age). The study was referred to as “a follow-up study of Asperger syndrome”. Of the 100 males with AS and/or their parents targeted for inclusion, 24 refused or failed all participation in the follow-up study. Nineteen of these declined participation over the telephone, and three did so in writing. In two cases there was no response in spite of reminders, and so we have no clue as to who was “responsible” for the failure to participate. (Of the 24 who refused, five had never been informed about their condition. Parents had never told their child about the diagnosis, and they did not want for their sons to find out about it now. In the other 19 cases, we only have limited information about the reason for refusal/failure to participate). There was no significant difference in mean FSIQ at original diagnosis between the participating (n = 76), and the non-participating (n = 24) AS groups.
Autism Group
In the original autism group of 84 males, three (4%) had died before follow-up. The causes of death were: death in an accidental fire (boy with Fragile X syndrome and autism at 11 years of age); death in complications after heart surgery (boy with trisomy 13 and a severe heart malformation and autism, age at death unknown); unknown date and cause of death (boy with atypical autism, age at death unknown). An additional 4 declined participation, leaving 77 for possible inclusion in the Autism study group. Of those who declined participation two had MMR and two SMR.
AS Study Group
Of the 76 cases participating in the follow-up study, seventy had had a complete DISCO interview performed, and these were included for further study, since we intended to match clinical and DISCO-diagnoses. These 70 cases constituted the “AS study group”.
Among these 70 cases there were 4 parents who participated without their son (in one of these cases the son was unaware of his condition), and test information about intellectual ability at follow-up is missing for these four individuals.
Autism Study Group
Of the 77 cases participating in the follow-up study, 75 had had a DISCO-interview completed, and of these, seventy males fell in approximately the same age-range as those 70 males who were chosen for the “AS study group”. These 70 autism cases constituted the “Autism study group”.
AS and Autism groups
Both study groups had been assessed by experts in the field of autism/AS, working at the CNC, with “autism spectrum instruments” that were state-of-the-art at the time of the diagnostic evaluations, e.g., in-depth clinical interview, the Handicaps, Behaviours, and Skills Schedule (Wing 1980), the Childhood Autism Rating Scale (Schopler, Reichler, DeVellis, & Daly, 1980), the Autism Behaviour Checklist (Krug, Arick, & Almond, 1980) and the Asperger Syndrome Diagnostic Interview (ASDI) (Gillberg, Gillberg, Rastam, & Wentz, 2001). The Autism Diagnostic Interview (ADI) (LeCouteur et al., 1989), and the Diagnostic Interview for Social and Communication Disorders (DISCO) (Wing, Leekam, Libby, Gould, & Larcombe, 2002), were not available in Swedish at the time of original diagnosis except for a small number of cases. About 50% of all AS cases had also received medical assessments including one or more of the following; karyotyping, neuroimaging, EEG, auditory brainstem response examination, and a variety of urine, blood and cerebrospinal fluid examinations. All the follow-up assessments were administered, co-ordinated and carried out by the first four authors (authors one and two in the AS-part, and authors three and four in the autism-part). The first author had not been involved at all, and the second author had only been involved in a few cases in the original AS diagnostic process. Authors three and four had not been involved in the original Autism diagnostic process. However, none of the investigators were completely “blind” to the original diagnoses, in their respective parts of the study. The fifth author, who is the head of the Child and Adolescent Psychiatric Department at Göteborg University, initiated and supervised both studies, but did not see the individuals at follow-up. All the follow-up assessments in the AS-part took place at the CNC, whereas those in the autism part were carried out at the CNC or in the homes/workplaces of the participants. All four assessors have trained together in the use of the DISCO and have been working together in an autism spectrum disorder/child neuropsychiatric team at the CNC for several years.
Both groups were followed-up with in-depth examinations performed in 1999–2005. The following instruments were used:
Diagnostic Interview for Social and Communicative Disorders (DISCO-10: A semi-structured interview intended for interview with a person (often a parent), who knew the individual well. The DISCO-10 has excellent inter-rater and test-retest reliability, and is highly valid for assigning diagnoses in the autism spectrum (Wing et al., 2002). It also includes sections on common associated problems in autism such as psychiatric disorder, including “psychosis”, has a developmental perspective, and is designed for use throughout the lifespan (Wing et al., 2002). The DISCO was chosen in favour of the ADI, (LeCouteur et al.,1989) because the latter was designed for use in the diagnosis of classic autism, whereas the DISCO includes a range of items intended to detect milder forms of autism spectrum disorders. In addition, the DISCO has a developmental perspective and is designed for use from early childhood into adult life (Wing et al.,2002).
Wechsler Adult Intelligence Scale-Third edition (WAIS-III) (Wechsler, 2003): This well-established IQ-test, including Full Scale IQ (FSIQ), and subtests for Verbal IQ (VIQ) and Performance IQ (PIQ), was used with all participants in the AS study group. In the Autism study group only a minority could be tested on the Wechsler scales ((Wechsler Adult Intelligence Scale-Revised (WAIS-R) (Wechsler, 1981) or (Wechsler Intelligence Scale for Children-Third Edition (WISC-III) (Wechsler, 1999)), and the majority were categorized in terms of IQ/DQ/SQ-band using the Vineland Adaptive Behaviour Scales (see below).
Global Assessment of Functioning scale (GAF) (APA, 1994): GAF yields scores from 0 to 100, where a score of ≥70 indicates good or only mildly abnormal psychosocial functioning. GAF was scored conjointly by the first/second, and the third/fourth authors in all cases in the respective studies. Most studies performed on the GAF have found it to be a reliable and useful instrument in measuring a persons psychosocial functioning, requiring only minor pre-scoring information (e.g., Billstedt et al., 2005; Hilsenroth et al., 2000; Startup, Jackson, & Bendix, 2002).
Vineland Adaptive Behavior Scales (VABS) (Sparrow, Balla, & Cicchetti, 1984): a semi-structured interview with a parent/caregiver that offers a comprehensive assessment of adaptive behaviour. Cases were categorized in DQ/SQ bands (Developmental Quotient/Social Quotient) on the basis of the results on the VABS. The VABS has been reported to be a valid instrument in establishing the cognitive level for an individual functioning at an IQ-level below 70–75 (APA, 1994; Luckasson et al., 1992). The VABS has also been widely used to map the overall functioning of an individual in socialization, communication, and daily living skills, regardless of IQ in order to be used as a prognostic and intervention tool for habilitation (Balboni, Pedrabissi, Molteni, & Villa, 2001; Gilotti, Kenworthy, Sirian, Black, & Wagner, 2002; Rhea et al., 2004).
A structured neuropsychiatric assessment, performed by the first and fourth authors respectively.
Clinical Diagnosis
A clinical diagnosis of autistic disorder, atypical autism, or AS, at follow-up of the AS study group was made systematically—including all available information, except the information generated at DISCO-10 interview—using the Gillberg (1991) criteria for AS (except the motor clumsiness criterion, which was not universally fulfilled) or DSM-IV/ICD-10 criteria for autistic disorder/atypical autism. In the Autism study group a clinical autism spectrum diagnosis was made using the same criteria as mentioned above for the AS part of the study.
DISCO-10 diagnosis
Research diagnoses of autism spectrum disorders were also made according to the algorithm of the DISCO-10 (DISCO-algorithm Gillberg criteria for AS and the DISCO-algorithm DSM-IV/ICD-10 criteria for autism/atypical autism). Diagnoses were generated by computer on the basis of the results obtained at the DISCO-interview.
The criteria used for the classification of outcomes, were similar to those employed in an earlier study of autism in our centre (Gillberg & Steffenburg, 1987), which was based on the outcome criteria published by Lotter (Lotter, 1978). Reliability studies—to our knowledge—have not been performed on the use of these criteria. The classifications were based on all available information (including the DISCO) at the time of examination.
The Outcome Criteria Were
Good outcome: (a) being employed or in “higher” (age and IQ-appropriate (“normal”)) education or vocational training, and, (b) if 23 years of age or older, living independently, or if 22 years or younger, having two or more friends/a steady relationship;
Fair outcome: either (a) or (b), but not both, under good outcome;
Restricted outcome: neither (a) nor (b) under good outcome, and not meeting criteria for a major psychiatric disorder other than autistic disorder or another autism ASD. This category refers to a group of people with the characteristics of poor outcome, but who have been accepted by a group of peers or personnel to such an extent that their handicaps are not so readily obvious;
Poor outcome: Obvious severe handicap, with either of, no independent social progress or presence of a major psychiatric disorder, but with some clear verbal or non-verbal communicative skills;
Very poor outcome: Obvious very severe handicap, unable to lead any kind of independent existence, no clear verbal or non-verbal communication.
AS Study Group
Fifty-nine individuals in the AS study group (84%) still met clinical (Gillberg, 1991) diagnostic criteria for AS, three (4%) met criteria for atypical autism, and 8 (12%) no longer met criteria for a clinical diagnosis in the autism spectrum. Seven (10%) of the males meeting clinical criteria for AS also met clinical criteria for autistic disorder (DSM-IV) and were clinically judged to better fit the latter diagnosis.
The DISCO-classification (Gillberg, DISCO-algorithm criteria) concurred with that of the clinical assessment in 55 of the 59 (93%) cases with a clinical diagnosis of AS at follow-up.
Clinical/DISCO diagnoses | AS (%) | Autism (%) |
|---|---|---|
Clinical diagnosis of AS | 52 (75%) | 0 (5)** (0%) (7%) |
Clinical diagnosis of autistic disorder | 7* (10%) | 58 (83%) |
Clinical diagnosis of atypical autism | 3 (4%) | 11 (16%) |
No clinical diagnosis of any autism spectrum disorder | 8 (11%) | 1 (1%) |
DISCO algorithm diagnosis of AS | 59 | 11 |
DISCO algorithm diagnosis of autistic disorder | 55 | 56 |
DISCO algorithm diagnosis of atypical autism | 10 | 14 |
No DISCO algorithm diagnosis of an autism spectrum disorder | 1 | 0 |
Autism Study Group
In the Autism study group, 43 out of 53 (81%) originally diagnosed AD still met clinical criteria for AD, 9 (17%) had a clinical diagnosis of atypical autism, and one individual did not have a clinical autism spectrum diagnosis at follow-up.
Only 2 out of 17 individuals with atypical autism at original diagnosis fulfilled criteria for the same diagnosis at follow-up. All remaining 15 individuals fulfilled criteria for AD at follow-up. Fifty-four out of 58 (93%) with a clinical AD diagnosis at follow-up also had a DISCO-10 diagnosis of AD and the other four had atypical autism. Nine out of eleven individuals with a clinical diagnosis of atypical autism had the same diagnosis according to DISCO-10 algorithm, and the other two had an autism diagnosis according to the same algorithm. Five males in the Autism study group fulfilled criteria for an AS diagnosis (Gillberg, 1991), but were clinically judged better to fit a diagnosis of autism. The only male in the Autism study group without a clinical diagnosis at follow-up had an atypical autism diagnosis on the DISCO-10 (Table 1).
AS Study Group
IQ/DQ/SQ-band | AS original diagnosis (n = 70) (%) | AS follow-up (n = 66) (%) | Autism original diagnosis (n = 70) (%) | Autism follow-up (n = 70) (%) |
|---|---|---|---|---|
≤49 | 0 (0%) | 0 (0%) | 33 (47%) | 50 (72%) |
50–69 | 0 (0%) | 1 (2%) | 24 (35%) | 15 (21%) |
70–84 | 15 (21%) | 4 (6%) | 10 (14%) | 2 (3%) |
85–114 | 41 (59%) | 45 (68%) | 3 (4%) | 3 (4%) |
115–129 | 10 (14%) | 14 (21%) | 0 (0%) | 0 (0%) |
≥130 | 4 (6%) | 2 (3%) | 0 (0%) | 0 (0%) |
Autism Study Group
In the Autism study group, Wechsler scale testing (WAIS-R or WISC-III) was only possible in 16 individuals. The mean FSIQ in this small tested group was 59.6 (SD 17.9, n = 16), VIQ 63.2 (SD 19.8, n = 15) and PIQ 58.9 (SD 12.3, n = 15). The results from the overall intellectual assessment of the Autism study group (including those tested on the Wechsler scales) are presented in Table 2. Those individuals who could not be tested on the Wechsler scales were categorized according to results on the VABS. The results in the Autism study group were significantly lower than at original evaluation, and contrasted to those of the AS group, in which mean FSIQ had not changed over time (Table 2).
AS study Group
Outcome categories/Independent living | AS | Autism |
|---|---|---|
Good outcome | 19 (27%) | 0 (0%) |
Fair outcome | 33 (47%) | 5 (7%) |
Restricted outcome | 16 (23%) | 12 (17%) |
Poor outcome | 2 (3%) | 14 (20%) |
Very poor outcome | 0 (0%) | 39 (56%) |
Independent living (23 years of age or older) | 14/22 (64%) | 3/40 (8%) |
Outcome AS | FSIQ (n = 66) | VIQ (n = 66) | Age at follow-up (SD) years |
|---|---|---|---|
Good 19 (27%) | 107.8 (SD 14.3) | 109.3 (SD 17.5) | 21.5 (SD 3.7) |
Fair 33 (47%) | 105.0 (SD 14.8) * | 104.3 (SD 14.2) * | 20.7 (SD 4.5) |
Restricted 16 (23%) | 97.4 (SD 13.0) | 98.6 (SD 15.1) | 23.5 (SD 4.8) |
Poor 2 (3%) | 82.0 (SD 4.2) | 92.0 (SD 12.7) | 18.5 (SD 2.4) |
Very poor 0 (0%) |
When participants in the AS study group were divided into groups according to age at original diagnosis, good outcome was seen in 9/26 (35%) in the youngest group (5.5–9.5 years at diagnosis), in 9/35 (26%) of the “in-between-group” (10.0–15.5 years at diagnosis), and in 2/9 (22%) of the oldest group (16.0– 24.5 years at diagnosis) (n.s.).
Autism Study Group
Outcome Autism | Intellectual level | Mean age at follow-up (SD) years | |||
|---|---|---|---|---|---|
A | NA | MMR | SMR | ||
Good n = 0 | |||||
Fair n = 5 (7%) | 1 | 2 | 2 | 0 | 26.1 (5.9) |
Restricted n = 12 (17%) | 2 | 0 | 7 | 3 | 25.9 (7.1) |
Poor n = 14 (20%) | 0 | 0 | 5 | 9 | 25.5 (5.1) |
Very poor n = 39 (56%) | 0 | 0 | 1 | 38* | 23.5 (4.9) |
AS Study Group
Eight of the AS males did university studies, and a further two had a university degree (computer science, civil engineering). Thirty-three males in the AS study group were in, or had currently finished, high school. Only 21 of these 33 (64%) were, or had been, following ordinary study programs. Three males in the AS group were in special AS classrooms, four had special individually developed study-programs, and five had some kind of special education regularly. Two AS males were studying at a Folk High School.2 Eight males with AS were (or had been) in a school for adolescents with mild learning disabilities. Three males with AS were in special schools away from home, where their training also included training in social and daily living skills. Two males had finished school after nine years of compulsory school.
Autism Study Group
None of the males in the Autism group did, or had done, university studies. Six males in the Autism study group were in, or had currently finished, high school (4 of whom had IQ >70). One of the males in the autism group was studying at a Folk High School. Twenty-three males with autism were (or had been) in a school for adolescents with mild learning disabilities. An additional 40 males had or had had their training in special training schools.
AS Study Group
Seven men in the AS group held ordinary jobs, and a further six individuals had “daily occupational activities” in a group centre. Twelve males with AS (17%) had no organized daily activity at all and were dependent on social services and/or the Swedish insurance system for their welfare.
Autism Study Group
In the Autism group one man held an ordinary job, and four individuals had “daily occupational activities” in a group centre. Thirty-three males in the Autism study group had regular individually tailored daily activities. Thirteen males with (19%) had no organized daily activity at all and were dependent on social services and/or the Swedish insurance system for their welfare.
AS Study Group
In the AS study group 14/22 (64%) of those who were ≥23 years, were living independently. In addition, 5 males with AS ≤ 22 years of age were living independently. Although living away from their parents, they were all dependent upon them for support. Three males with AS were living in a long-term relationship, and a further 10 had had relationships for varying periods of time in the past.
Autism Study Group
Of the males who were ≥23 years in the Autism group (n = 40), only 3 males (8%), one each with IQ-level A, NA and MMR), were living independently. Although living independently, they were all in need of their parents for support. One male with Autism was living in a long-term relationship, and yet another one had recently had a relationship for a longer duration.
AS Study Group
GAF-scores | AS | FSIQ (SD) | Autism | Intellectual ability |
|---|---|---|---|---|
Mean GAF-score (SD) | 58.9 (9.4) | 22.2 (16.5) | ||
GAF-score 70- | 12 | 109.7 (15.2) | 0 | – |
GAF-score 50–69 | 51 | 102.7 (14.2) * | 8 | A (3) NA (2) MMR (3) |
GAF-score 31–49 | 7 | 91.9 (13.7) | 13 | MMR (5) SMR (8) |
GAF-score -30 | 0 | – | 49 | MMR (4) SMR (45) |
There was no difference at follow-up in GAF-scores between the AS-age-at-diagnosis-groups, 5–9 years 59.5 (SD 7.1, n = 26), 10–15 years 58.4 (SD 9.7, n = 35) and 16+ years 59.0 (SD 14.3, n = 9). The individuals with AS, who had been followed-up at the CNC on a fairly regular basis after diagnosis (n = 29) had a mean GAF score of 57.6 (SD 6.8), similar to those who had not been followed there (n = 41) 59.8 (SD 10.8).
Autism Study Group
The mean GAF-score was 22.2 (SD 16.5 range 4–67, p < 0.001) in the Autism study group. No individual in the Autism study group had a GAF score of 70 or above (Table 6).
AS Study Group
The vast majority in the AS study group were considered very law–abiding. However, according to parent report, seven males (10%) with AS had been involved with police and the law for different reasons, (fraud (1), harassment of police officer (1), harassment of young woman (1), stealing (1), assault (1), sexual abuse (1), and unknown in 1 case).
AS Study Group
Three individuals in the AS study group had been diagnosed as suffering from psychosis by independent psychiatrists. One of these males had received a diagnosis of bipolar disorder, and there was suspicion of such disorder in at least one further case. No individual had been diagnosed with schizophrenia in the AS group. All three individuals with a psychosis diagnosis in the AS group were on current anti-psychotic medication. In terms of intellectual ability two of them (the third one was not assessed) had had significant drop (≥20 IQ-points) in their FSIQ as measured on the WAIS-III. However, all 3 individuals with psychosis were reported to have shown a significant drop in intellectual ability between original diagnosis and follow-up.
Autism Study Group
In the Autism study group four individuals had been diagnosed as suffering from psychosis, and in none of those cases had a diagnosis of bipolar disorder been made. No individual had been diagnosed with schizophrenia in the Autism group. Two out of four individuals with a psychosis diagnosis in the Autism group were on current anti-psychotic medication.
This, to our knowledge, is the largest prospective comparative follow-up study ever published of a cohort of individuals with AS, and autism, followed over a period of more than 5 years. We believe that the individuals included are representative of AS, and autism/atypical autism, as diagnosed 10 years ago or more. The results are of particular interest, given that most previous studies have related to smaller, and/or possibly highly selected samples.
Were our hypotheses supported by the findings? The diagnosis of AS was still clinically valid in the vast majority of AS cases (84%), but 11 individuals did not meet clinical criteria for this particular diagnosis at follow-up. Three of these 11 had a clinical diagnosis of atypical autism, but eight did not have sufficient clinical impairment to warrant a clinical diagnosis of an autism spectrum disorder. However, this was not equivalent to lack of impairment from autistic type problems. In the Autism study group 81% originally diagnosed with AD still met clinical criteria for AD at follow-up. All the remaining males but one had a clinical diagnosis of atypical autism. Of the 17 individuals with atypical autism at original diagnosis, 15 fulfilled criteria for AD at follow-up. The DISCO-10 identified the vast majority of individuals in both study groups in the same categories as the clinicians.
“Restricted” and “poor” outcome affected more AS individuals (26%) than expected, considering the average intellectual level for this group as a whole, and the outcome criteria used. However, the Autism study group had a much worse outcome, with the vast majority of individuals (76%) belonging in the poor or very poor outcome groups. The intellectual level was much lower in the Autism study group, where only five (7%) individuals had a normal intellectual ability at follow-up.
There was a tendency for FSIQ, and particularly, as hypothesized, for VIQ to be correlated with better outcome within the AS group, even though those with better outcomes had not been followed up for as long as those with restricted outcome making definite conclusions about ultimate prognosis somewhat less certain. In contrast, in the Autism study group the individuals in the “very poor” outcome group were the youngest (i.e. mean age was lowest in this group).
There was no decline in FSIQ in the AS group over time. However, the difference between verbal and non-verbal ability of 15 IQ-points or more that applied in 45% at original diagnosis was now present only in 19% of the individuals. Performance results tended to improve over time and the mean VIQ > PIQ difference of 11 IQ-points or more that was present at original diagnosis was no longer at hand at follow-up. This might reflect an improvement in visual-spatial ability over time, but the subtests included in the Perceptual Organization index had changed over time, meaning that definite conclusions in this respect will have to await further studies. The stable overall IQ in the AS group contrasted with the decline of intellectual ability in the Autism study group, where there had been a considerable drop in intellectual ability over the years.
A small, but not insignificant, group of individuals in both the AS and Autism study groups had been diagnosed by independent psychiatrists as having “psychosis”. “Schizophrenia” had not been diagnosed in a single individual, and even though cases of this condition might well appear with time, it seems clear that neither AS, nor autism is associated with a much increased risk for schizophrenia in early adult life. At least two of the three males in the AS group, who had been diagnosed with psychosis had had severe decline in intellectual functioning between original diagnosis and follow-up (the third male was not tested at follow-up). They all fulfilled criteria for AS at follow-up.
There was a (non-significant) young age at diagnosis by better outcome trend in the AS group. Given that young age at diagnosis probably also indicates a more difficult child, this trend would seem to indicate that an early diagnosis in itself possibly would contribute to a better outcome. However no definite conclusions can be drawn in this respect on the basis of the current results.
Seven individuals with AS (10%) were reported to have been involved with the police and the law for various acts of crime. In this age group in Sweden, a rate of 10% criminality is not surprisingly high (National Council for Crime Prevention, Sweden, 2005). Nevertheless, the nature of the acts of crime performed by the young men with AS was rather “severe” in several cases, and visualized difficulties with perspective-taking, and difficulties in appreciating the consequences of their actions, which was also reported in an earlier study (Murphy, 2003).
In summary, we found a diagnosis in the autism spectrum to be stable over time in the vast majority of cases. Outcome in the AS study group was worse than expected, taken the level of intelligence and outcome criteria used into account. However, it was dramatically better than in the Autism study group. The better outcome could probably be attributed to the much higher FSIQ in the former group. We also found FSIQ to be stable over time in the AS study group, which was in contrast to the Autism study group where there was an intellectual decline over time. In a small AS subgroup there was psychotic disorder which, in turn, was associated with decline in intellectual ability, and hence in outcome. Criminal acts were not reported at very high rates, but the acts of crime were sometimes odd and reflecting the lack of common sense that is one of the key issues in AS.
Of the 100 males approached for inclusion in the AS follow-up study, only 76 participated, and 70 of these were included in the present study. We know that the non-participants did not differ from the participants in terms of overall intelligence and there are no obvious indications that the examined group differs in any major way from the larger group originally targeted. The autism contrast group may not be regarded as ideal because of its much lower IQ. Nevertheless, it would be unrealistic (and, clinically, probably impossible) to recruit an Autism study group matched for IQ, particularly given speculation (e.g., Gillberg, 1998) that the main difference across cases clinically diagnosed as AS and autism is the much higher IQ in the former group, and that there are no clinically diagnosed cases of autism who have IQ in the superior range. We believe that, as regards the social deficits, the two groups were roughly comparable in childhood and/or adolescence, and that they therefore constituted reasonable contrast groups for the purpose of follow-up of psychosocial adjustment and general outcome in adult life.
Males with AS diagnosed in childhood–young adult age have outcomes that are very much better than those found in males diagnosed in childhood as suffering from autism/atypical autism. Nevertheless, given their good intellectual capacity, the outcomes must be regarded as sub-optimal. Medical, social and occupational services must find ways to achieve more individually adjusted solutions so as to be more successful in meeting the needs of individuals with autism spectrum disorders.
Gillberg and Gillberg (1989) diagnostic criteria elaborated (Gillberg 1991)
| 1. | Social impairment (extreme egocentricity) (at least two of the following)
| ||||||||||
| 2. | Narrow interest (at least one of the following)
| ||||||||||
| 3. | Repetitive routines (at least one of the following)
| ||||||||||
| 4. | Speech and language peculiarities (at least three of the following)
| ||||||||||
| 5. | Non-verbal communication problems (at least one of the following):
| ||||||||||
| 6. | Motor clumsiness
|
Qualitative impairment in social interaction, as manifested by at least two of the following:
| (a) | marked impairment in the use of multiple non-verbal behaviours such as eye-to-eye gaze, facial expression, body postures, and gesture to regulate social interaction |
| (b) | failure to develop peer relationships appropriate for developmental level |
| (c) | a lack of spontaneous seeking to share enjoyment, interests, or achievements with other people (e.g., by lack of showing, bringing, or pointing out objects of interests to other people) |
| (d) | lack of social or emotional reciprocity |
Restricted repetitive and stereotyped patterns of behaviour, interests, and activities, as manifested by at least one of the following:
| (a) | encompassing preoccupation with one or more stereotyped and restricted patterns of interest that is abnormal either in intensity or focus |
| (b) | apparently inflexible adherence to specific, non-functional routines or rituals. |
| (c) | stereotyped and repetitive motor-mannerisms (hand- or finger-flapping or twisting or complex whole-body movements) |
| (d) | persistent preoccupation with parts of objects |
The disturbance causes clinically significant impairment in social, occupational, or other important areas of functioning:
The is no clinically significant general delay in language (e.g., single words used by age 2 years, communicative phrases used by age 3 years).
There is no clinically significant delay in cognitive development or in the development of age-appropriate self-help skills, adaptive behaviour (other than in social interaction), and curiosity about the environment in childhood.
Criteria are not met for another Pervasive Developmental Disorder or Schizophrenia.
References
Footnotes
News from the Swamp.....
Bits & Pieces (Autism Science & News) AUGUST 2009
Anxiety, Social Deficits, and Loneliness in Youth ...
Twenty-Year Outcome for Individuals With Autism an...
Crazy Talk: Jenny McCarthy, Suzanne Somers & Oprah
Maternal Intimacies: Discussing Autism Diagnosis
Bits & Pieces...(Autism Science & News) JULY 2009 ...
Autism & Anorexia (article)
Members who read this post also read: