REVIEW

CURRENT
OPINION Autism spectrum disorder: outcomes in adulthood
Patricia Howlin a,b and Iliana Magiati c

Purpose of review
Until recently, there has been little systematic study of adult life among individuals with autism spectrum
disorder (ASD) but recognition of the high psychological and social costs of ASD has led to an increase in
adult-focused research over the past decade. The aim of this review is to summarize recent empirical
findings on outcomes for adults with ASD.
Recent findings
Most research on adult outcomes in ASD indicates very limited social integration, poor job prospects and
high rates of mental health problems. However, studies vary widely in their methodology, choice of
measures and selection of participants. Thus, estimates of how many adults have significant social and
mental health problems are often conflicting. There is little consistent information on the individual, familial
or wider social factors that may facilitate more positive social and psychological outcomes. There is a
particular dearth of research on older individuals with ASD.
Summary
The very variable findings reported in this review reflect the problems of conducting research into lifetime
outcomes for individuals with a condition as heterogeneous as ASD. Much more systematic research is
needed to delineate different patterns of development in adulthood and to determine the factors influencing
these trajectories.
Keywords
adulthood, aging, autism, autism spectrum disorder, developmental trajectories, outcome

INTRODUCTION Articles were selected from a literature search

The estimated prevalence of autism spectrum (conducted in October 2016) of PsycINfo, Medline
disorder (ASD) in adults is high (11/1000 [1 ]),
& and Google Scholar. The search was limited to
and there are increasing concerns about the poor peer reviewed articles published in English from 1
long-term outcome for so many individuals with January 2015 to 30 September 2016. (See references
this condition. Health economists [2 ] also high-
& for details of search strategy.) Articles were included
light the high financial costs, predicting that annual in the review if: the focus was on broader social/
medical and nonmedical costs of ASD in the United lifetime outcomes in ASD (i.e. individuals with
States will be $268 billion for 2015 and $461 billion autism, Asperger syndrome/disorder and autism
&
for 2025 [2 ]. Much of this expenditure is for adults, spectrum disorder); samples were of adequate size
largely because of the costs of medical care, residen- (n >20); average cohort age was 18þ (any IQ level).
tial or supported living accommodation and pro- The search identified 1525 articles relating to adults
ductivity loss, both by caregivers and by individuals with ASD; of these, 43 met inclusion criteria for the
themselves. present review. Additional references outside the

AIMS
a
Department of Psychology, Institute of Psychiatry, Psychology and
The aim of the present article is to review data from
Neuroscience, King’s College, London, UK, bFaculty of Health Sciences,
recent studies on adults with ASD with respect to: University of Sydney, Sydney, New South Wales, Australia and
c
Department of Psychology, National University of Singapore, Singapore
(1) Social outcomes Correspondence to Patricia Howlin, PhD, MSc, BA, Department of
(2) Trajectories of change over time Psychology, Institute of Psychiatry, Psychology and Neuroscience, King’s
(3) Factors related to outcome College, London SE58AF, UK. E-mail: [email protected]
(4) Mental health Curr Opin Psychiatry 2017, 30:69–76
(5) Mortality and aging DOI:10.1097/YCO.0000000000000308

0951-7367 Copyright ß 2017 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Neurodevelopmental disorders

outcome. However, the authors also highlight the

KEY POINTS wide variability in findings. Thus, the mean esti-
 Research on outcomes in adult life for individuals with mated percentage of individuals with a ‘good out-
ASD is growing, but findings remain inconsistent and come’ was 20% but with 95% confidence intervals
often conflicting. (CIs) varying from 14 to 27% across studies;
similarly, the 95% CIs for ‘poor outcome’ ranged
 There is contradictory evidence on how many
from 37 to 59%.
individuals with ASD achieve good outcomes as adults,
and lack of agreement on how this should be measured This variability is evident in subsequent studies.
&

or how prognosis might be improved. Within a French cohort [6 ] of adults of mixed IQ

and autism severity (n ¼ 76; age 18–54 years), two
 In particular, data on psychiatric comorbidities in thirds had a ‘poor’ or ‘very poor’ outcome and even
adulthood vary widely, and it remains uncertain how
among those rated as having a ‘good outcome’,
many individuals do experience serious mental
health problems. none was living independently. A US-based, online
&
survey [7 ] completed by parents/carers of 143 indi-
 Nevertheless, despite lack of agreement between viduals with ASD (mean age 25 years) revealed that
studies, overall outcomes for adults with ASD in terms only 22% were in work, 7% lived independently,
of jobs, relationships, independent living and mental
whereas 87% were on benefits.
health are considerably poorer than for same
age peers. Cohorts involving more cognitively and
verbally able adults generally report more positive
 Knowledge about factors that are associated with results, although, again figures vary. Among 50
good, or poor, social and psychological outcomes adults with Asperger syndrome (mean childhood
remains very limited, with few studies considering
IQ 100þ; mean current age 30 years) living in Swe-
family, school or wider social influences. &
den [8 ], 40% were in full time education or inde-
pendent work, 62% were living independently, 48%
had two or more friends and 52% either currently or
specified date range are included in order to place in the past had a partner (14% of these were married/
current research in context. cohabiting). In a German cohort of 50 adults with
&
Asperger syndrome [9 ] (mean age 36 years), 46%
were currently employed, and 28% had a partner;
SOCIAL OUTCOMES IN ADULTHOOD however, 50% depended on their families or state
benefits for support and 28% were still living with
Assessments based on ‘normative’ measures &
their parents. The US survey [7 ] also collected data
of social functioning from 255 adults with ASD (mean age 38 years) who
Previous systematic reviews, focusing on ‘norma- were able to report on their own status. They
tive’ measures of social outcomes in ASD (i.e. objec- recorded relatively high levels of education (42%
tive measures of employment, independence, social batchelors degree or higher), employment (47%)
participation and relationships) [3] have concluded and independent living/living with a spouse or
that, even among cohorts of average intelligence, partner (67%). Employment figures in this study
most individuals remain highly dependent on are considerably higher than reported for another
others for their care, social contacts are limited &
US cohort [10 ] of individuals with ASD of average
and employment rates are extremely low. Moreover, IQ (n ¼ 73, mean age 24 years), among whom only
adults with ASD are more economically, education- one quarter was consistently employed. Neverthe-
ally and socially disadvantaged than adults with &
less, even within the former sample [7 ], over half
other developmental or intellectual disabilities [4]. were unemployed and many (36%) were dependent
&&
A recent systematic review [5 ] (12 studies pub- on federal or state benefits.
lished from 1967 to 2013; n ¼ 828) indicates little
change in these conclusions. Outcomes were rated
as in previous reviews [3] (i.e. ‘very good’ ¼ high Subjective assessments of social outcomes
independence and social functioning; ‘good’ ¼ and quality of life
some level of employment, some friends, largely The wide variability in outcomes across studies has
independent; ‘fair’ ¼ requires some support in daily led to concerns that standard concepts of what
living; ‘poor’ ¼ requires substantial support/residen- constitutes a ‘good’ social outcome may not always
tial care; and ‘very poor’ ¼ long-stay secure care). be relevant for people with ASD. For example,
Forty-eight per cent of participants were rated as higher social achievements for some individuals
having a ‘poor’/‘very poor’ outcome, and only 20% may come at the cost of higher stress and poorer
were deemed to have a ‘good’ or ‘very good’ &&
mental health [11 ]. It has been suggested [11 ]
&&

70 www.co-psychiatry.com Volume 30  Number 2  March 2017

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.

 Adult outcomes in ASD Howlin and Magiati

that when subjective factors such as satisfaction Is there ‘recovery’ from autism spectrum
with life, good physical and mental health, adequate disorder?
living conditions, supportive and fulfilling social The identification of individuals with ‘Very Positive
&&
and family relationships are taken into account, Outcomes’ [18 ] raises the question of whether
adult outcomes may be more positive than usually there can be ‘recovery’ from ASD. The Swedish
& &
reported. Thus, within a US cohort studied over follow-up [8 ,19 ] also identified a small group of
several decades [12 ] (n ¼ 60; mean age 32 years),
&&
adults (11 out of 50) who no longer met diagnostic
many participants experienced relatively good sub- criteria for ASD. All had an IQ in the average range,
jective Quality of Life (QoL) and their mean scores had friends and were living independently, and all
on the World Health Organization Quality of Life but one was employed. Nevertheless, although
measure [13] were comparable with those of the eight, either currently or in the past, had a partner
general population. In the Swedish cohort of indi- or were married, romantic relationships were lower
&
viduals with Asperger syndrome [8 ], scores on a than in the general population; three also had some
subjective quality of life measure [14] were also current psychiatric comorbidity. An earlier study
within the average normative range (albeit at the [20] described 34 children (mean age 12 years; mean
lower end of that range). Nevertheless, despite these IQ 111) who, although initially meeting ASD
more positive findings, a recent meta-analysis of criteria, were currently functioning ‘within normal
&&
studies of QoL across the lifespan [15 ] concluded limits’. Recent, more detailed data analyses, how-
that individuals with ASD, including those of higher ever, [21–23] indicate persisting subtle difficulties in
intellectual and verbal ability, have a poorer QoL social understanding, pragmatic communication,
than their non-ASD peers. attention, self-control and emotional maturity
and in psychiatric morbidity. As yet, there are no
data on this cohort in adulthood and it is uncertain
TRAJECTORIES OF CHANGE OVER TIME whether these remaining differences will abate or
Research on trajectories of development in ASD become more evident with age.
generally indicates improvements over time
& & & &&
[8 ,16 ,17 ,18 ] but, again, there is considerable
variability. An 8–10 year follow-up [16 ,17 ] in
& &
FACTORS RELATED TO SOCIAL
the United States of over 300 individuals (mean OUTCOMES IN ADULTHOOD
age at follow-up 22 years; 70% with IQ < 70) found
that one-third (35%) showed improvements in non- Intellectual and verbal functioning
verbal communication, 58% in verbal communi- Intellectual and verbal functioning in childhood are
cation, 40% in social interaction and 61% in among the strongest prognostic indicators in ASD
repetitive/stereotyped behaviors and interests. Mal- [3]. Few individuals with a childhood IQ below 70,
adaptive behaviors improved in 42%. Only 12% or who fail to develop functional speech, live inde-
showed a worsening in total autism symptoms pendently as adults and job prospects and social
and 11% in maladaptive behaviors. integation are particularly poor. In addition, they
Another longitudinal study in the United States show less improvement in cognitive or social skills,
&&
[18 ] examined progress in 85 individuals first seen and greater increases in ritualistic behaviors over
as children. In early adulthood (mean age 19 years), time than individuals with an average IQ in child-
& & &&
most (77, 91%) continued to meet ASD criteria; of hood [16 ,17 ,18 ]. Nonverbal mental age at 2 years
these, 53 had an IQ below 70 and 24 had an IQ of at is also predictive of independence in daily living
&
least 70. Eight individuals (all IQ>70) who no longer skills at age 21 [24 ]. Unsurprisingly, too, persisting
met criteria were described as having a ‘Very Positive communication and intellectual impairments in
Outcome’ (VPO). Trajectories of change (IQ, com- adulthood (especially if associated with epilepsy)
munication, social functioning and repetitive are associated with low levels of social attainments
&
behaviors) among the VPO group were significantly and independence [6 ]. Nevertheless, even among
more positive than for participants with IQ below individuals with a childhood IQ at least 70, out-
70. There were also some significant differences comes can vary widely. Some show very few autism
between the VPO group and participants of average symptoms or cognitive difficulties as young adults,
IQ who still met diagnostic criteria. The authors whereas others continue to experience significant
&&
suggest that further research focusing on different problems [18 ]. The relationship between IQ and
patterns of developmental trajectories may be outcome in higher functioning samples also
important for identifying different genetic causes depends on the variables studied. Thus, among
& &
as well as having implications for more individually the Swedish Asperger syndrome cohort [8 ,19 ],
tailored interventions. current IQ correlated highly with academic success,

0951-7367 Copyright ß 2017 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com 71

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Neurodevelopmental disorders

but not with independent living, friendships or also explored the effects of stress in two groups of
marital status. adults with ASD (n ¼ 38, 40; mean ages 23–24 years;
rates of employment 37–47%; and living independ-
ently 17–21%). These adults experienced signifi-
Autism symptom severity cantly more stressful life events and stress than
Another established predictor of adult outcome is community controls, and levels of global stress sig-
& &
autism severity [3]. In the Swedish study [8 ,19 ], nificantly predicted overall social functioning and
higher levels of autism symptomatology, both in social disability. In a long-term follow-up in the
&&
childhood and adulthood, were associated with United States [12 ], current levels of perceived
reduced independence, lower rates of employment stress, together with frequency of bullying in child-
and more limited social relationships. Similarly, hood, were the two factors consistently linked
&
within the French cohort [6 ], no adults with a with poorer adult self-reported quality of life. In
childhood diagnosis of ‘severe autism’ were judged turn, good quality of life was positively correlated
to have a ‘good’ outcome; in contrast, 40% of those with better-developed daily living skills and good
with ‘moderate autism’ and all those with a child- physical health.
hood diagnosis of Asperger syndrome were rated as
having a ‘good’ outcome.
Mental health
Data on rates of psychiatric disorders in ASD are
Gender &
highly variable. In a UK longitudinal study [30 ] of
The role of gender remains uncertain as most studies 58 adults originally diagnosed as children (mean IQ
&
involve so few women [25 ]. There are some sugges- 69; mean age 44 years), 28% had at some time
tions that women with ASD have poorer social out- experienced mild-to-moderate mental health prob-
&
comes, especially with respect to employment [10 ] lems, and 28% had severe or very severe difficulties.
&& &
and quality of life [11 ] than men. However, other A retrospective case review [31 ] of 474 adults
& &&
studies [16 ,26 ] report no significant impact of attending an ASD diagnostic clinic in the UK found
gender on autism symptoms, behavior problems that around half (57%) had a comorbid psychiatric
or social outcomes. disorder. A similar figure (54%) was reported in a
large US database study [32 ] (n ¼ 1507 adults with
&&

ASD). In the Swedish cohort with Asperger syn-

Family/environmental factors &
drome [19 ], 54% had a current diagnosis but almost
In the community sample followed up over 8–10 all (94%) met lifetime criteria for a comorbid psy-
& &
years in the United States [16 ,17 ], higher levels of chiatric/neurodevelopmental disorder. In a Dutch
cohort [33 ] including older adults (n ¼ 172, age
&&
inclusion in social and academic activities in school
were associated with more positive outcomes, over 19–79 years, IQ > 80), 79% met criteria for a psy-
and above individual characteristics such as age, IQ chiatric disorder at least once in their lives. Among
and sex. Greater maternal praise/positivity in child- participants with Asperger syndrome in the German
&
hood also predicted higher levels of nonverbal com- cohort [8 ], 70% had at one or more psychiatric
&
munication and social reciprocity in adulthood; comorbidities. In the online US survey [7 ], 86%
increases in maternal praise over time were associ- of the self-report group and 73% of the proxy-report
ated with decreases in externalizing behaviors. In group had at least one mental/behavioral comor-
addition, quality of mother–child relationship was bidity.
significantly associated with fewer maladaptive The majority of diagnoses/symptoms identified
behaviors at follow-up; improvements in mother– in these studies relate to anxiety and/or depression,
child relationships were related to a decline in but again there are many inconsistencies. Estimates
behavioral and social problems. Other research of depressive disorders range from 20 to 58% and
&& &&
[11 ,12 ] has also highlighted the positive impact anxiety disorders from 22 to 39%; other commonly
of maternal warmth on adult quality of life, along reported difficulties include Attention Deficit Hyper-
with factors such as physical health, greater inde- activity Disorder (ADHD) (10–28%); tic disorders
pendence in daily living skills and better executive (1–50%); Obsessive Compulsive Disorder (OCD)
function. (8–28%) and somatoform and eating disorders
&
More recently, the specific impact of stress has (6–17%). Two studies [34,35 ] report high rates of
received attention. Among 25 Swedish adults (mean social anxiety (50–52%). Estimates of psychotic
&
age 34 years) [27 ], greater perceived stress was disorders tend to be relatively low (usually around
associated with more severe autism symptoms and 2–4%), but again there is wide variability, with a
& & &
poorer coping in adulthood. A US study [28 ,29 ] has recent review recording figures from 0 to 35% [36 ].

72 www.co-psychiatry.com Volume 30  Number 2  March 2017

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.

 Adult outcomes in ASD Howlin and Magiati

Conflicting data on rates of mental health prob- (53 vs. 12%) and more mood disorders than women
lems in ASD are due to many factors, including (32 vs. 6%).
differences in sampling, the range of conditions Data on substance abuse are, yet again, contra-
&
selected for study and the diagnostic procedures dictory. In the German cohort [8 ], more women
and measures used. Although, overall, the data than men had drug abuse problems (19 vs. 9%) but
indicate that psychiatric morbidity is higher in fewer had alcohol abuse/dependence (12 vs. 20%).
&& &&
ASD that in the general population [32 ], until In contrast, in the US sample [32 ], drug/alcohol
there is greater methodological consistency across problems were more frequent in men (drugs 5 vs.
studies, it remains impossible accurately to estimate 3%; alcohol 4 vs. 2%).
the true risk. Findings on the relationship between mental
Estimates of substance abuse disorders are also health and age are equally confusing. The German
&
contradictory. A Swedish epidemiological study study [8 ] found that young adults (<40 years)
[37 ] (n ¼ 26 986 individuals with ASD and 96 557
&
showed less psychopathology than those over 40.
&
controls) reported a substantially increased risk of In the US longitudinal study [39 ], women were
drug [odds ratio (OR) 8.5] and alcohol abuse/ more likely to show greater increases in anxiety
dependence (OR 4.0); risks of substance-related and depressive symptoms over time, whereas ado-
crime (OR 1.4), or deaths linked to abuse were also lescent males had more depressive symptoms that
&
high (OR 3.0). A German study [8 ] recorded sim- were maintained into young adulthood. In contrast,
&&
ilarly high rates of drug abuse (12%) or alcohol- a study including a much wider age range [33 ]
related problems (18%). In contrast, in the US data- (n ¼ 344, age 19–79 years) concluded that psycho-
&&
base [32 ], alcohol abuse/dependency was ident- pathology declined with age, with fewer adults in
ified in 3% and drug abuse/dependency in 4%. the older age group (55–79 years) meeting criteria
&
Combined rates of these problems in the UK [31 ] for any psychiatric diagnosis and particularly social
&
and Swedish [19 ] samples were also low (2 and 4%, phobia. Discrepancies here are likely because of the
respectively). This inconsistency is highlighted in a very small number of older adults in the two
recent systematic review of substance abuse in ASD former samples.
[38 ] (n ¼ 18 studies; 11 epidemiological). Although
&&
Findings concerning the relationship with other
estimates were generally low, figures ranged from variables that are frequently associated with mental
0.7 to 36%, making it impossible to establish a health in the general population (i.e. cognitive
reliable prevalence figure. The only consistent find- functioning, social economic status and living situ-
ing noted in the review was the lack of knowledge ations, as well as life events and family factors)
among professionals on how to treat this group of remain inconclusive and inconsistent.
patients, and the dearth of intervention research.

MORTALITY AND SUICIDE IN ADULTHOOD

Variables associated with adult mental Many major chronic medical conditions occur sig-
health nificantly more frequently in adults with ASD than
&&
Although there has been little systematic explora- in the general population [32 ], and mortality risks
tion of factors related to mental health in adults are also higher. In a Danish epidemiological study
[41 ] (total n ¼ 1,912 904; n ASD ¼ 20 492), mortality
&
with ASD, there is some evidence of an association
between poor mental health and poorer social func- rates for young adults with ASD were double those in
& &
tioning [30 ], lower life satisfaction [39 ] and higher the general population, and similar to the risk for
& & & &&
levels of autism symptoms [7 ,19 ,30 ,33 ,40]. individuals with neurological or mental/behavioral
The relation between mental health and gender disorders. Comparable rates were identified in
is inconsistent, probably because of the small Sweden [42] (27 122 individuals with ASD;
numbers of women in most studies. Most research 2,672 185 matched general population controls).
suggests that, compared with men, women are at Mortality was over twice as high in the ASD group
greater risk of anxiety and mood disorders (OR 2.56) and mean age of death was much lower
&& &&
[32 ,33 ] and of conditions such as dementia, [controls 70.2 years; ASD 53.9 years (39.5 years in
&&
schizophrenia and bipolar disorder [32 ]; men tend those with Intellectual Disability (ID); 58.4 years in
&&
to have higher rates of OCD and ADHD [32 ]. those without ID)]. The most frequent causes of
& &
However, two studies [30 ,31 ] identified no signifi- death were nervous, circulatory, respiratory or diges-
cant sex differences in overall rates or types of tive disorders and congenital malformations.
mental health problems and in the German cohort Overall, death rates in men and women were
&
[8 ], whereas there were no differences in rates of similar, but women were more likely to die from
major depression, men had more anxiety symptoms endocrine disease, congenital malformations or

0951-7367 Copyright ß 2017 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com 73

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Neurodevelopmental disorders

suicide, and men from diseases of the nervous and how many experience a good psychological and
circulatory systems. The most common cause of physical quality of life. More importantly, we are
death in individuals with ID and ASD was epilepsy. a long way from identifying the individual, family or
This study [40] also found that death by suicide environmental factors that enhance resilience and
was significantly elevated among cognitively able ensure social and psychological well being in adult-
individuals with ASD (OR 9.4). However, there is hood. High-quality adult outcome research must be
little consistent information on suicidal behaviors a priority if we are to meet the needs of current and
or ideation, or how frequently these result in serious future generations of adults with ASD.
or fatal suicide attempts, and reported rates of
suicide in ASD vary widely (from <10 to >50%) Acknowledgements
&
[41 ]. For example, suicide attempts occurred in None.
&&
only 2% of the US database cohort [32 ]. In the
Swedish Asperger study [19 ] (n ¼ 50), 13 individuals
&

Financial support and sponsorship

(26%) exhibited ongoing suicidal behavior, but only
None.
one was considered to be at high risk. This conflicts
with an earlier study [42] in which 66% of 367
Conflicts of interest
individuals with a diagnosis of Asperger syndrome
reported suicidal ideation; 35% reported plans or There are no conflicts of interest.
attempts at suicide. Recent studies in this area
&&
[43 ,44] note the importance of more research into REFERENCES AND RECOMMENDED
the factors associated with suicide, especially in READING
more able individuals, and the need for more Papers of particular interest, published within the annual period of review, have
been highlighted as:
reliable ways of identifying at-risk adults with ASD. & of special interest
&& of outstanding interest

1. Brugha TS, Spiers N, Bankart J, Cooper SA. Epidemiology of autism in adults

AGING & across age groups and ability levels. Br J Psychiatry 2016; 209:498–503.
Data from the recent UK Adult Psychiatric Morbidity Survey confirm that pre-
Although there has been very little investigation of valence of autism in adults in England is 11/1000 (95% CI 3–19/1000).
the impact of aging in ASD, current research 2. Leigh JP, Du J. Brief report: forecasting the economic burden of autism in
& && 2015 and 2025 in the United States. J Autism Dev Disord 2015; 45:4135–
[45 ,46 ] suggests that many cognitive skills (proc- &

4139.
essing speed, attention, verbal memory, cognitive Authors highlight the huge costs of ASD in the United States and predict that this
will continue to rise over the next decade.
flexibility and planning, and theory of mind) show 3. Magiati I, Tay XW, Howlin P. Cognitive, language, social and behavioural
similar patterns of decline as in typical aging. Con- outcomes in adults with autism spectrum disorders: a systematic review of
longitudinal follow-up studies in adulthood. Clin Psychol Rev 2014; 34:73–86.
versely, individuals with ASD may be less prone to 4. Roux AM, Shattuck PT, Cooper BP, et al. Postsecondary employment
decline in visual and working memory than elderly experiences among young adults with an autism spectrum disorder. J Am
&& Acad Child Adolesc Psychiatry 2013; 52:931–939.
in the general population [46 ]. Quality of life in 5. Steinhausen HC, Mohr Jensen C, Lauritsen MB. A systematic review and
ASD also seems to be less affected by age than in the && meta-analysis of the long-term overall outcome of autism spectrum disorders
in adolescence and adulthood. Acta Psychiatrica Scandinavica 2016;
general population, but to date studies on aging in 133:445–452.
ASD are small scale, mainly cross-sectional and con- This article describes the very variable findings from studies of prognosis in
individuals with ASD. It is the first to provide CIs for estimates of good/fair/poor
tain few participants aged over 60. More extensive outcomes and concludes that the prognosis for almost 50% of individuals is poor.
research in this area is crucially needed. The authors highlight the need for research on pathways and predictors of
outcome.
6. Chamak B, Bonniau B. Trajectories, long-term outcomes and family experi-
& ences of 76 adults with autism spectrum disorder. J Autism Dev Disord 2016;
CONCLUSION 46:1084–1095.
Data are based on parental reports of social functioning in adults (age 18–54) with
It is evident from this review that individuals with ASD. Negative prognostic variables include greater autism severity, low IQ, poor
language and the presence of epilepsy.
ASD continue to face many challenges throughout 7. Gotham K, Marvin AR, Taylor JL, et al. Characterizing the daily life, needs, and
adulthood. Current social and health services for & priorities of adults with autism spectrum disorder from Interactive Autism
& && Network data. Autism 2015; 19:794–804.
adults [47,48 ,49 ,50–52] are often inadequate, This is an internet-based survey completed by individuals with autism or parents/
resulting in high levels of stress for both individuals carers. It describes the situation for adults in the United States and highlights the
& need to improve healthcare and public services.
themselves and their families [53 ]. To date, how- 8. Helles A, Gillberg IC, Gillberg C, Billstedt E. Asperger syndrome in males over
ever, methodological issues particularly related to & two decades: quality of life in relation to diagnostic stability and psychiatric
comorbidity. Autism 2016; 1362361316650090 [Epub ahead of print]
wide heterogeneity in the cohorts studied and Analysis of factors related to outcome in adulthood indicates the importance of
variability in the measures used, have resulted in examining subjective aspects of quality of life (physical and emotional health) in
addition to objective measures (jobs, relationships and so on).
inconsistent and sometimes contradictory research 9. Roy M, Prox-Vagedes V, Ohlmeier MD, Dillo W. Beyond childhood: psychia-
findings. We still do not know with any certainly & tric comorbidities and social background of adults with Asperger syndrome.
Psychiatria Danubina 2015; 27:50–59.
what proportion of individuals manage to attain This German study identifies high rates of mental health problems in 50 adults (age
adequate levels of social integration as adults or 20–62). Rates of employment and close relationships were also limited.

74 www.co-psychiatry.com Volume 30  Number 2  March 2017

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.

 Adult outcomes in ASD Howlin and Magiati

10. Taylor JL, Henninger NA, Mailick MR. Longitudinal patterns of employment 26. Kirby AV, Baranek GT, Fox L. Longitudinal predictors of outcomes for adults
& and postsecondary education for adults with autism and average-range IQ. && with autism spectrum disorder systematic review. OTJR: Occup Particip
Autism 2015; 19:785–793. Health 2016; 36:55–64.
This article focuses on the poor employment prospects of young adults with ASD. Systematic review of research on predictors of outcome in ASD. Most studies
Outcomes for women were found to be even less positive than for men. were found to be of poor quality methodologically. The impact of most factors was
11. Bishop-Fitzpatrick L, Hong J, Smith LE, et al. Characterizing objective quality inconsistent, but individual variables (IQ and so on) appear to be most important.
&& of life and normative outcomes in adults with autism spectrum disorder: an 27. Hirvikoski T, Blomqvist M. High self-perceived stress and poor coping in
exploratory latent class analysis. J Autism Dev Disord 2016; 46:2707– & intellectually able adults with autism spectrum disorder. Autism 2015;
2719. 19:752–757.
The authors draw attention to the need for broader measures of outcome in ASD, Adults with ASD found to have significantly higher subjective stress and
rather than focusing on ‘normative’ measures such as rates of employment/ poorer ability to cope with stress compared with typical adults. Higher levels
independent living and so on. More positive outcomes are associated with better of autistic traits were associated with higher levels of stress and poorer coping
daily living skills, better executive function and more maternal warmth. ability.
12. Hong J, Bishop-Fitzpatrick L, Smith LE, et al. Factors associated with sub- 28. Bishop-Fitzpatrick L, Mazefsky CA, Minshew NJ, Eack SM. The relationship
&& jective quality of life of adults with autism spectrum disorder: self-report versus & between stress and social functioning in adults with autism spectrum disorder
maternal reports. J Autism Dev Disord 2016; 46:1368–1378. and without intellectual disability. Autism Res 2015; 8:164–173.
This article compared self-rated and parent-rated quality of life among adults with See note to 29
autism. The agreement between self-report and proxy report was good. Factors 29. Bishop-Fitzpatrick L, Minshew NJ, Mazefsky CA, Eack SM. Perception of life
related to better quality of life included independence in daily activities and physical & as stressful, not biological response to stress, is associated with greater
health. Higher levels of perceived stress and bullying were associated with poorer social disability in adults with autism spectrum disorder. J Autism Dev Disord
quality of life. 2016; DOI: 10.1007/s10803-016-2910-6.
13. Skevington SM, Lotfy M, O’Connell KA. The World Health Organization’s This and reference [28] report on high levels of stress in adults with ASD and their
WHOQOL-BREF quality of life assessment: psychometric properties and greater exposure to stressful events. In turn, these are significantly associated with
results of the international field trial. A report from the WHOQOL group. Qual social disability. The authors highlight the need for more interventions targeting
Life Res 2004; 13:299–310. stress management.
14. Pallant JF, Lae L. Sense of coherence, well being, coping and personality 30. Moss P, Howlin P, Savage S, et al. Self and informant reports of mental health
factors: further evaluation of the sense of coherence scale. Pers Individ Dif & difficulties among adults with autism findings from a long-term follow-up study.
2002; 33:39–48. Autism 2015; 19:832–841.
15. van Heijst BF, Geurts HM. Quality of life in autism across the lifespan: a meta- Self and informant reports of psychiatric comorbidities in adults with ASD in their
&& analysis. Autism 2015; 19:158–167. mid-40s indicate that around half have some mental health problems. However,
Meta-analysis of 10 studies (2004–2012; n ¼ 486 people with autism and 17 776 over 40% had experienced no psychiatric problems in adulthood.
controls). Data indicated that, across the lifespan, people with autism experience a 31. Russell AJ, Murphy CM, Wilson E, et al. The mental health of individuals
much lower quality of life compared with people without autism. However, age did & referred for assessment of autism spectrum disorder in adulthood: a clinic
not have an effect on quality of life. report. Autism 2016; 20:623–627.
16. Woodman AC, Smith LE, Greenberg JS, Mailick MR. Change in autism Clinical case note study of 474 adults with ASD. Just over half (57%) had a
& symptoms and maladaptive behaviors in adolescence and adulthood: the history of psychiatric problems. As in the previous study [30], this rate is higher
role of positive family processes. J Autism Dev Disord 2015; 45:111–126. than in the general population, but considerably less than reported in some other
The authors report improvements in autism symptoms and maladaptive behaviors studies.
over 8.5 years in a large, community-based sample of adolescents and adults. 32. Croen LA, Zerbo O, Qian Y, et al. The health status of adults on the autism
17. Woodman AC, Smith LE, Greenberg JS, Mailick MR. Contextual factors && spectrum. Autism 2015; 19:814–823.
& predict patterns of change in functioning over 10 years among adolescents Very informative article on physical and mental health problems in a large cohort of
and adults with autism spectrum disorders. J Autism Dev Disord 2016; adults with ASD identified via an insurance database. Although such sampling is
46:176–189. prone to a number of biases, the large size of the sample and the detailed
Exploration of factors associated with positive trajectories over time identified the information included provide valuable insight into the frequency and range of
importance of educational inclusion and maternal praise/warmth. problems experienced by individuals with ASD.
18. Lord C, Bishop S, Anderson D. Developmental trajectories as autism phe- 33. Lever AG, Geurts HM. Psychiatric co-occurring symptoms and disorders in
&& notypes. Am J Med Genet Part C: Semin Med Genet 2015; 169:198–208. && young, middle-aged, and older adults with autism spectrum disorder. J Autism
This article examined trajectories of development in 85 individuals with autism Dev Dis 2016; 46:1916–1930.
followed into young adulthood. Three distinct groups were identified: adults with In this study of 344 adults with ASD aged 19–79 years, 79% had a history of
intellectual disability and persistent ASD; adults with IQs in the average range who psychiatric disorder (mainly depression and anxiety). However, older adults were
continued to show ASD impairments and a small group (n ¼ 8) with IQs in the less likely to experience mental health problems than younger individuals. The
average range who were judged to be functioning at age appropriate levels at age discrepancies in reported rates of psychiatric difficulties across studies (e.g. 19,
19 years, despite a previous childhood diagnosis of ASD. The findings have 30, 31, 32 and 37), and disagreements concerning the association with age,
potentially important implications for understanding subtypes of autism and highlight the need for more systematic research in this area.
developing more individually tailored interventions. 34. Maddox BB, White SW. Comorbid social anxiety disorder in adults with
19. Gillberg IC, Helles A, Billstedt E, Gillberg C. Boys with Asperger syndrome autism spectrum disorder. J Autism Dev Disord 2015; 45:3949–3960.
& grow up: psychiatric and neurodevelopmental disorders 20 years after initial 35. Spain D, Happé F, Johnston P, et al. Social anxiety in adult males with autism
diagnosis. J Autism Dev Disord 2016; 46:74–82. & spectrum disorders. Res Autism Spect Disord 2016; 32:13–23.
The authors report very high rates (94%) of psychiatric comorbidities in 50 adult As in reference [33], the authors report high rates of social anxiety (52%) in
males diagnosed with Asperger syndrome. These rates are considerably higher adults with ASD. The article concludes that more research is needed to ascertain
than in several other adult outcome studies. the prevalence social anxiety in ASD and to identify triggers and maintaining
20. Fein D, Barton M, Eigsti IM, et al. Optimal outcome in individuals with a history factors.
of autism. J Child Psychol Psychiatry 2013; 54:195–205. 36. Chisholm K, Lin A, Abu-Akel A, Wood SJ. The association between autism
21. Orinstein A, Tyson KE, Suh J, et al. Psychiatric symptoms in youth with a & and schizophrenia spectrum disorders: a review of eight alternate models of
history of autism and optimal outcome. J Autism Dev Disord 2015; 45:3703– co-occurrence. Neurosci Biobehav Rev 2015; 55:173–183.
3714. Very useful review of studies on the association between autism and schizophre-
22. Orinstein AJ, Suh J, Porter K, et al. Social function and communication in nia. The article highlights the large discrepancies between many studies in this
optimal outcome children and adolescents with an autism history on struc- area and explores possible reasons for these inconsistencies.
tured test measures. J Autism Dev Disord 2015; 45:2443–2463. 37. Butwicka A, Långström N, Larsson H, et al. Increased risk for substance
23. Suh J, Orinstein A, Barton M, et al. Ratings of broader autism phenotype and & use-related problems in autism spectrum disorders: a population-based
personality traits in optimal outcomes from autism spectrum disorder. J Autism cohort study. J Autism Dev Disord 2016; 1–0. doi:10.1007/s10803-016-
Dev Disord 2016; 46:3505–3518. 2914-2.
24. Bal VH, Kim SH, Cheong D, Lord C. Daily living skills in individuals with autism These authors identify very high rates of substance abuse in ASD and in their
& spectrum disorder from 2 to 21 years of age. Autism 2015; 19:774–784. siblings and parents. They conclude that ASD is a risk factor for substance abuse-
Authors found that nonverbal mental age, receptive language and social-commu- related problems. However, other studies (e.g. [36]) have not identified this as a
nication impairment in early childhood predicted levels of daily living skills in young significant problem in ASD; reasons for these conflicting findings remain unclear.
adults with ASD. They suggest that daily living skills should be a focus of treatment 38. Arnevik EA, Helverschou SB. Autism spectrum disorder and co-occurring
particularly for adolescents transitioning to adulthood. && substance use disorder: a systematic review. Substance Abuse: Res Treat
25. Halladay AK, Bishop S, Constantino JN, et al. Sex and gender differences in 2016; 10:69–75.
& autism spectrum disorder: summarizing evidence gaps and identifying emer- A systematic review (18 articles) of substance abuse in ASD. The authors
ging areas of priority. Mol Autism 2015; 6:1. conclude that overall comorbidity rates appear to be low, but lack of agreement
Discussion article highlighting how little is known about the role of sex and gender across studies makes difficult to establish a general prevalence rate. The authors
in ASD. highlight the lack of high-quality research in this area.

0951-7367 Copyright ß 2017 Wolters Kluwer Health, Inc. All rights reserved. www.co-psychiatry.com 75

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Neurodevelopmental disorders

39. Gotham K, Brunwasser SM, Lord C. Depressive and anxiety symptom 46. Lever AG, Geurts HM. Age-related differences in cognition across the adult
& trajectories from school age through young adulthood in samples with autism && lifespan in autism spectrum disorder. Autism Res 2016; 9:666–676.
spectrum disorder and developmental delay. J Am Acad Child Adolesc These authors [44], [45] are among the very few who have explored cognitive and
Psychiatry 2015; 54:369–376. psychological changes in ASD in later adulthood. They have identified certain
Study of trajectories of psychiatric disorders in 109 individuals with ASD. Rates of areas of cognitive difficulty (e.g. in Theory of Mind) that seem to become less
affective and anxiety disorders were high, but men and women showed different apparent in old age in ASD; other skills (e.g. visual memory) show less decline with
patterns of change over time. age than in the general population. The authors conclude that age-related cognitive
40. Garcı́a-Villamisar D, Rojahn J. Comorbid psychopathology and stress mediate difficulties in ASD are mainly parallel to those found in typical aging, although
the relationship between autistic traits and repetitive behaviours in adults with deterioration in some areas may actually be reduced. They suggest that ASD could
autism. J Intell Disabil Res 2015; 59:116–124. partially protect against an age-related decrease in cognitive functioning.
41. Schendel DE, Overgaard M, Christensen J, et al. Association of psychiatric 47. Havlicek J, Bilaver L, Beldon M. Barriers and facilitators of the transition to
& and neurologic comorbidity with mortality among persons with autism adulthood for foster youth with autism spectrum disorder: perspectives of
spectrum disorder in a Danish population. JAMA Pediatr 2016; service providers in Illinois. Child Youth Serv Rev 2016; 60:119–128.
170:243–250. 48. Turcotte P, Mathew M, Shea LL, et al. Service needs across the lifespan for
Large epidemiological study identified high risks of early mortality in individuals & individuals with autism. J Autism Dev Disord 2016; 46:2480–2489.
with ASD. The presence of comorbid mental/behavioral or neurologic conditions US study indicating that adults with ASD were in greater need of services but had less
significantly increased the risk of early death (2.6-fold to 7.6-fold) compared with access to a range of different services than individuals in the general population.
individuals without ASD. 49. Murphy CM, Wilson CE, Robertson DM, et al. Autism spectrum disorder in
42. Cassidy S, Bradley P, Robinson J, et al. Suicidal ideation and suicide
&& adults: diagnosis, management, and health services development. Neurop-
plans or attempts in adults with Asperger’s syndrome attending a specialist sychiatric Dis Treat 2016; 12:1669.
diagnostic clinic: a clinical cohort study. Lancet Psychiatry 2014; 1:142– This article highlights the lack of health services research for adults with ASD. The
147. authors focus on the need for more rigorous pharmacological and pharmacological
trials, and the importance of taking into account the views of individuals with ASD
43. Hirvikoski T, Mittendorfer-Rutz E, Boman M, et al. Premature
themselves.
&& mortality in autism spectrum disorder. Br J Psychiatry 2016; 208:232–
50. Vohra R, Madhavan S, Sambamoorthi U. Comorbidity prevalence, healthcare
238.
utilization, and expenditures of Medicaid enrolled adults with autism spectrum
Important epidemiological study highlighting greatly increased risks of premature
disorders. Autism 2016; pii: 1362361316665222. [Epub ahead of print]
mortality in ASD. The study also identifies factors that appear to be associated with
51. Vohra R, Madhavan S, Sambamoorthi U. Emergency department use among
a greater risk of early death. The presence of a range of different comorbid medical/
adults with autism spectrum disorders (ASD). J Autism Dev Disord 2016;
neurological conditions was associated with a higher risk of death, especially
46:1441–1454; 50.
among women with intellectual disability. Individuals of average intellectual ability
52. Raymaker DM, McDonald KE, Ashkenazy E, et al. Barriers to healthcare:
had a high risk of suicide. The authors stress the need for more research into
instrument development and comparison between autistic adults and adults
factors related to early death in ASD.
with and without other disabilities. Autism 2016; pii: 1362361316661261.
44. Salvatore T, Brown J, Hastings B, et al. Suicide risk in adults with [Epub ahead of print]
autism spectrum disorder: an exploratory discussion. J Special Popul 53. McKenzie K, Ouellette-Kuntz H, Blinkhorn A, Démoré A. Out of school and into
2016; 1:1–11. & distress: families of young adults with intellectual and developmental dis-
45. Lever AG, Werkle-Bergner M, Brandmaier AM, et al. Atypical working memory abilities in transition. J Appl Res Intell Disabil 2016; doi: 10.1111/jar.12264.
& decline across the adult lifespan in autism spectrum disorder? J Abnormal [Epub ahead of print]
Psychol 2015; 124:1014. This study describes the high levels of stress experienced by many families during
See note to [46]. their sons’/daughters’ transition from school.

76 www.co-psychiatry.com Volume 30  Number 2  March 2017

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.