Introduction
Adults with intellectual disability (ID) are living longer than previously. Figures from the National Intellectual Disability Database show an increasing percentage in those with moderate to profound ID aged 35 years and over, from 28.5% in 1974 to 48.7% in 2015 (Doyle & Carew, Reference Doyle and Carew2015). While adults with autism spectrum disorder (ASD) have a slightly decreased life expectancy compared with the neurotypical population (Perkins & Berkman, Reference Perkins and Berkman2012), anywhere between 8% and 40% of the adult population with ID will also have ASD (Bhaumik et al., Reference Bhaumik, Tyrer, McGrother and Ganghadaran2008; Cooper, et al., Reference Cooper, Smiley, Morrison, Williamson and Allan2007; de Bildt et al., Reference de Bildt, Systema, Kraijer and Minderaa2005; La Malfa et al., Reference La Malfa, Lassi, Bertelli, Salvini and Placidi2004; Arvio & Sillanpää, Reference Arvio and Sillanpää2003; Matson & Shoemaker, Reference Matson and Shoemaker2009). The knowledge base on life expectancy for this cohort with both ID and ASD is limited. Behavioural presentations (McCarthy et al., Reference McCarthy, Hemmings, Kravariti, Dworzynski, Holt, Bouras and Tsakanikos2010), decreased adaptive functioning (Matson et al., Reference Matson, Rivet, Fodstad, Dempsey and Boisjoli2009b), increased psychiatric (Totsika et al., Reference Totsika, Felce, Kerr, Hastings, Tricco and Lillie2010; Dunn et al., Reference Dunn, Rydzewska, Fleming and Cooper2020) and medical (Perkins & Berkman, Reference Perkins and Berkman2012; Dunn et al., Reference Dunn, Rydzewska, Fleming and Cooper2020) comorbidities are common to both disorders and have implications for multi-disciplinary care needs. Age-related risk factors within the general population are still present and some groups are at higher risk, for example those with Trisomy 21 who develop early-onset dementia, on average, in their mid-fifties (Lai & Williams, Reference Lai and Williams1989; Visser et al. Reference Visser, Aldenkamp, van Huffelen, Kuilman, Overweg and van Wijk1997; McCarron et al., Reference McCarron, McCallion, Reilly and Mulryan2014; McCarron et al., Reference McCarron, Cleary and McCallion2017). Even in those individuals with ID and without Down syndrome, dementia is more common compared with the general population (Cooper & van der Speck, Reference Cooper and van der Speck2009). This article looks at the evidence base for screening, diagnosis, management and outcomes for older adults with both ID and ASD (ID/ASD) and where future research might be directed to optimise their care.
Method
Research databases Medline, Pubmed, Embase, CINAHL and PsychInfo (inception to 4 August 2020) were searched using key words: (older adults) AND (ID OR mental retardation OR learning disability) AND (autism OR ASD).
1989 results were reduced to 1227 after duplicates were removed.
Studies or articles focusing on children, adolescents, under 35s only or their parents or caregivers (n = 467) were firstly excluded, apart from those studying children for estimates of current ASD prevalence. The following were excluded: genetic, metabolic or other aetiological studies (n = 199); studies relating to ASD (adult spinal disorder or atrial septal defect) (n = 3); studies not in English (n = 51); case reports or small case studies (<5 subjects) (n = 13) and studies or articles concerning specific syndromes only were excluded (e.g. Fragile X Syndrome, Smith Magenis Syndrome) as being outside the scope of this article, given that these studies tended to focus on anomalies and characteristics specific to these genotypes and were not relevant to a broader scoping review on ageing in ID and autism (n = 175).
The abstracts of 319 remaining results of possible interest to an adult population with ID/ASD were then screened for relevance and the full text was consulted if relevance was questionable. Ten papers concerning dental or anaesthetic techniques in this population were excluded as being outside the scope of this article. Sixty-seven results were excluded as not identifying a study group or subgroup with ASD. Forty-three results were excluded as not identifying a study group or subgroup with an ID. Nineteen results were excluded as pertaining to the family members or carers of adults with ID or ASD only, as the carer perspective is outside the scope of this article. Seventy-eight articles of possible interest were deemed irrelevant – examples include focus on metabolic markers for ID/ASD, older papers that refer to defunct diagnostic categories like ‘autistic psychosis’ or ‘idiots-savants’, validation of scales and subscales in foreign language populations, trials of medication for challenging behaviours and neuroimaging characteristics of adults with ID or ASD, among others. The full-text article was not available in ten results of possible interest.
Eighty-five articles of interest were included for appraisal and the data are presented covering domains of diagnosis, prevalence, psychiatric comorbidities and functional outcomes in adults with ID/ASD. Seven articles were also included as relating to current childhood prevalence of ASD, in order to estimate future generations of an ageing population with ASD.
PRISMA Guidelines were followed in the conduct of this review (Tricco, Reference Tricco, Lillie, Zarin, O'Brien, Colquhoun and Levac2018). Additional sources of information were identified from websites of governmental and non-governmental organisations, e.g. Irish Department of Health and National Institute for Health and Care Excellence (U.K.), where relevant.
Sixteen further papers of interest were identified from references of those included within the study. Six articles of interest to the Irish population were personally identified by the authors of this study. One article of interest regarding ageing in autism was personally identified by an author of this study.
Results
On examining extant literature, the data were collated and are presented covering domains of diagnosis, prevalence, psychiatric comorbidities and functional outcomes.
Diagnosis
Older people are less likely to have been diagnosed with ASD for various reasons, mainly that ‘childhood autism’ (Kanner, Reference Kanner1943) was largely ignored in adulthood for decades. Diagnostically, DSM-IV required that symptoms were present before 3 years which presented some difficulty in establishing a reliable history for adult subjects, although this requirement has since been dispensed with in DSM-V. Public awareness of ASD has grown, but much of the research continues to focus on children. Yet, the estimated prevalence of 1% in the community has been found across all age groups (Brugha et al., Reference Brugha, McManus, Bankart, Scott, Purdon, Smith, Bebbington, Jenkins and Meltzer2011), so autism is no longer only a disorder of childhood.
Defining ‘ageing’ in ID and ASD is somewhat arbitrary, given that the evidence base largely emerged in the last decade. A wide range of cutoffs has been used in different studies on ageing in this context, from 30 years (Wise et al., Reference Wise, Smith and Rabins2019) to 65 years (Cooper, Reference Cooper1997).
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20220815011048878-0739:S0790966721000653:S0790966721000653_fig1.png?pub-status=live)
Fig. 1. Study flow chart.
This review selected 35 years as a cut-off for selecting studies of interest, given that national figures in this jurisdiction (Doyle & Carew, Reference Doyle and Carew2015) demonstrate an increase in adults with ID living over this age in recent decades and that a mean age of death of 39.5 years in ‘low-functioning ASD’ was noted in one Swedish study (Hirvikoski et al., Reference Hirvikoski, Mittendorfer-Rutz, Boman, Larsson, Lichtenstein and Bölte2016).
Overall, mortality is increased in ASD (Hirvikoski et al., Reference Hirvikoski, Mittendorfer-Rutz, Boman, Larsson, Lichtenstein and Bölte2016; Hwang et al., Reference Hwang, Srasuebkul, Foley, Arnold and Trollor2019), probably owing to medical comorbidity associated with ID and genetic syndromes. However, it has not been proven that autism per se reduces life expectancy (Gillberg et al., Reference Gillberg, Billstedt, Sundh and Gillberg2010). One review notes that low case identification in an adult population and a relatively poor research focus on ageing in autism means that it is unclear whether autism in itself can predict certain trajectories of mortality (Janicki et al., Reference Janicki, Henderson and Rubin2008).
There is a lack of consensus on what ASD screening, diagnostic and neuropsychological evaluations to use in an adult cohort (Roestorf et al., Reference Roestorf, Bowler, Deserno, Howlin, Klinger, McConachie, Parr, Powell, Van Heijst and Geurts2019). Table 1 summarises findings in this regard.
Table 1. Summary of findings regarding screening, diagnostic and evaluation tools in an older population with ASD/ID
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20220815011048878-0739:S0790966721000653:S0790966721000653_tab1.png?pub-status=live)
Some self-report questionnaires may not be appropriate for those with an ID due to cognitive ability (Roestorf et al., Reference Roestorf, Bowler, Deserno, Howlin, Klinger, McConachie, Parr, Powell, Van Heijst and Geurts2019). The prevailing need for a reliable informant history to make an accurate diagnosis of ASD in those with ID (Bhaumik et al., Reference Bhaumik, Tyrer, Barrett, Tin, McGrother and Kiani2010; Matson et al., Reference Matson, Hess, Mahan, Fodstad and Neal2012) is also recommended in an older population (van Niekerk et al., Reference van Niekerk, Groen, Vissers, van Driel-de Jong, Kan and Oude Voshaar2011). The latter authors warn that the probable under-identification of ASD in an older population may lead to iatrogenic harm, if incorrectly treated as psychiatric disorder.
Prevalence
In the last two decades, with more comprehensive assessments and increased public awareness, the estimated prevalence of ASD has increased. While an earlier study (Brugha et al., Reference Brugha, McManus, Bankart, Scott, Purdon, Smith, Bebbington, Jenkins and Meltzer2011) found no evidence of a statistically significant reduction in the prevalence of ASD as a function of age, another noted a small decrease in ASD prevalence with age, perhaps due to decreased longevity in those with moderate to profound ID (Wise, Reference Wise2020).
Table 2 summarises recent findings regarding estimates of prevalence of ASD, ID and both.
Table 2. Estimates of prevalence
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20220815011048878-0739:S0790966721000653:S0790966721000653_tab2.png?pub-status=live)
The prevalence of ID in adults with ASD is being revised. Recent data from the US SPARK cohort (Fombonne et al., Reference Fombonne, Green Snyder, Daniels, Feliciano and Chung2020) shows that most of those with ASD aged 30 and over received their diagnosis in adulthood. Changes in diagnostic criteria (e.g. the removal of the ‘early onset’ requirement in DSM-V) and overall increased public awareness of ASD have reduced the proportion of those with ASD who are believed to also have an ID (Table 2).
From the perspective of those with ID, the estimated prevalence of ASD in adults with ID varies (Table 2), depending on the screening tools or diagnostic criteria used. Innovative tools are emerging, such as music-based screening tool MUSAD (Bergmann et al., Reference Bergmann, Sappok, Diefenbacher, Dames, Heinrich, Ziegler and Dziobek2015). One study identified a doubling of the previously recognised prevalence of ASD when 256 adults with severe ID were recruited for screening and diagnosis, suggesting a tendency towards underestimation (Saemundsen et al., Reference Saemundsen, Juliusson, Hjaltested, Gunnarsdottir, Halldorsdottir, Hreidarsson and Magnusson2010). The known prevalence of ID in Ireland is 5.96/1,000 (or 3.49/1,000 with moderate to profound ID) (Hourigan et al., Reference Hourigan, Fanagan and Kelly2018). Unfortunately, the National Intellectual Disability Database (NIDD) in Ireland does not capture data on ASD at present (Dept. of Health, 2018).
Finally, autistic traits may be present in those with ID who do not meet the diagnostic criteria for ASD. One study (Bhaumik et al., Reference Bhaumik, Tyrer, Barrett, Tin, McGrother and Kiani2010) found that 69% of an adult population with ID displayed at least one autistic trait and that therefore autistic traits in people with ID were not specific to ASD. Those with two or more autistic traits without an ASD diagnosis tended to be older, suggesting that some traits e.g. elaborate routines or reduced social interaction may develop or become more prominent with age. A previous study (Shah et al., Reference Shah, Holmes and Wing1982) found strongly autistic traits in 38% of adult inpatients with ID. Matson et al. (Reference Matson, Dempsey, Lovullo and Wilkins2008) assessed IQ level as a moderator for expression of the ASD triad of impairments in adults with ID. ASD traits in those with ID only were affected by IQ level, but symptoms were not significantly moderated by IQ level in the ID/ASD group. They also noted that the domain of repetitive behaviours caused the most functional impairment for those with ID/ASD, followed by social interaction, then communication.
Psychiatric comorbidity
People with an ID have a significantly higher rate of comborbid psychopathology than the general population (Matson & Shoemaker, Reference Matson and Shoemaker2009; Horovitz et al., Reference Horovitz, Matson, Sipes, Shoemaker, Belva and Bamburg2011). Difficulties in identifying psychiatric disorder in subjects with ASD include the diagnostic overlap between the triad of impairments in autism and symptoms of major psychiatric disorders (Bakken et al., Reference Bakken, Helverschou, Eilertsen, Heggelund, Myrbakk and Martinsen2010). The presence of complex behavioural presentations, or ‘challenging behaviours’, may also affect assessment for psychiatric disorder (O’Dwyer et al., Reference O'Dwyer, McCallion, Burke, Carroll, O'Dwyer and McCarron2018). Table 3 summarises the literature relating to ID/ASD and these comorbidities.
Table 3. Summary of findings regarding ID/ASD and psychiatric comorbidity/challenging behaviours
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20220815011048878-0739:S0790966721000653:S0790966721000653_tab3.png?pub-status=live)
Prescribing for adults with ID/ASD
Concerns have been raised about the inappropriate prescribing (Matson & Neal, Reference Matson and Neal2009; O’Dwyer et al., Reference O'Dwyer, McCallion, Henman, McCarron, O'Leary, Burke, O'Connell and O'Dwyer2019) and monitoring of side effects (Paton et al., Reference Paton, Bhatti, Purandare, Roy and Barnes2016) of psychotropics in a population with ID. There is a lack of quality evidence regarding the prescription of psychotropics in a population with ASD, as much of the research focuses on children and ethical issues arise in relation to carrying out randomised controlled trials. The importance of correctly differentiating psychiatric disorder from ASD impairment or behavioural problems lies in the risk of inappropriate prescribing (e.g. benzodiazepines, antipsychotics) and iatrogenic harm in a population already at risk of polypharmacy (O’Dwyer et al., Reference O'Dwyer, Peklar, McCallion, McCarron and Henman2016; Schoufour et al., Reference Schoufour, Oppewal, van der Maarl, Hermans, Evenhuis, Hilgenkamp and Festen2018). Table 4 summarises the literature in this regard.
Table 4. Summary of literature regarding prescribing of psychotropics in ID/ASD
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20220815011048878-0739:S0790966721000653:S0790966721000653_tab4.png?pub-status=live)
Functional outcomes
All older adults have the added complications of age-related cognitive decline, frailty and medical comorbidity to contend with. One of the hopes for older adults with ID is that functional independence is maintained as much as possible. Objective measurement of functioning might be used to assess activity limitations, such as the Waisman Activities of Daily Living (W-ADL) scale, validated in a longitudinal study for adults with autism and ID (Maenner et al., Reference Maenner, Smith, Hong, Makuch, Greenberg and Mailick2013). Functional outcomes may be affected by medical comorbidity and care setting factors (Table 5).
Table 5. Summary of findings regarding physical and mental health outcomes in adults with ID/ASD
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20220815011048878-0739:S0790966721000653:S0790966721000653_tab5.png?pub-status=live)
Discussion
Research on the ageing of people with ID in general has been challenging given that ageing with a known diagnosis of ASD has been a relatively recent phenomenon, ageing services and policies have not included people with ID, and the prevalent services philosophy in ID services has been focused upon building independence rather than maintaining function (McCallion et al., Reference McCallion, Jokinen, Janicki, Wehmeyer, Brown, Percey, Shogren and Fung2017). Available research confirms patterns of comorbid conditions are often different, onset of conditions may be at earlier ages, there are fewer natural supports such as spouses and one’s own children in older years, and access to needed healthcare is often difficult and not suitable for needs (Burke et al., Reference Burke, O’Dwyer, McGlinchey, Foran, MacGiolla Phadraig, Carroll, McCallion, McCarron, Prasher and Janicki2019; McCarron et al., Reference McCarron, Swinburne, Burke, McGlinchey, Carroll and McCallion2013,Reference McCarron, McCallion, Reilly, Dunne, Carroll and Mulryan2017; McCausland et al., Reference McCausland, McCallion, McCarron, Putnam and Bigby2021). Yet there are reports of people with ID who are living fulfilling lives as they age maintaining valued relationships and community connections (McCausland et al., Reference McCausland, McCallion, Brennan and McCarron2018). There are also examples of ageing in place and of housing and programming options that have lessons for the general population (Jokinen et al., Reference Jokinen, Janicki, Keller, McCallion and Force2013; McCallion et al., Reference McCallion, Jokinen, Janicki, Wehmeyer, Brown, Percey, Shogren and Fung2017). As research develops on the ageing of people with ASD, there must be attention to understand the similarities to and differences from people with ID and without ASD, as much as attention to those issues compared to the general population.
Firstly, research is needed to assess the prevalence of adults with ID/ASD in Ireland and in other jurisdictions. The effect of ageing on ASD prevalence should be observed, in both the neurotypical population and in those with ID, especially as there is some evidence for a decreased life expectancy with those with ASD. The benefits of attentive medical and nursing care in residential care facilities may not be as easily accessible to those who are able to live independently, but who may be isolated or socially avoidant in the context of autistic traits.
Any research in the area of ID/ASD, psychiatric comorbidity and behavioural problems is hampered by the overlap between the symptoms of many major mental illnesses, the decreased cognitive and adaptive functioning associated with an ID and the triad of impairments classically observed in ASD. Studying these symptoms in an ageing population may be further complicated by age-related cognitive impairment, a lack of reliable informants (particularly regarding early developmental history) and medical comorbidity. Awareness of these difficulties, which are demonstrated consistently in the literature to date, may better inform future study design. The drive to validate specific screening and diagnostic tools for psychiatric and neuropsychological comorbidity within a population with ID/ASD is one positive trend which may contribute to the growing evidence base. Longitudinal observation of the effect of ageing on cognitive ability and mental health conditions in ID and ASD has been deemed the foremost priority in data collection by many researchers.
Appropriate screening and diagnostic tools should be validated for those with ID and used as a matter of best practice. Ideally, screening for ASD would be carried out on all adults registered with the NIDD, although it is acknowledged that this would be an onerous task. The correct identification of ASD, or even autistic traits, in an ID population may save some patients from being labelled unfairly as having ‘challenging behaviours’ or misdiagnosed with major mental illness. Psychotropic medication may be inappropriately prescribed in these situations. If increased medical comorbidity cannot be avoided (e.g. epilepsy, age-related cardiovascular changes), a prudent diagnostic approach for psychiatric symptoms might reduce the drug burden in an ageing population.
Currently, many older adults with ID live in residential or supported care facilities and State funding is allocated for their social, occupational and healthcare needs. The increasing prevalence of ASD in children is of relevance for future resource planning, as ageing may lead from a family-based care model in childhood/early adulthood to residential care in later years. However, those who continue to live with family or who live independently must also have their comorbid medical and psychiatric conditions appropriately managed, potentially allied to day centre or work placement attendance. Some ID service providers have developed specific day and residential services for people with ASD, but not always in a purpose-built fashion. We need to optimise independence and acquisition of daily living skills in adults with ID/ASD. There is a challenge in improving and maintaining adaptive functioning in additional ways to that of the neurotypical ageing population.
Limitations
This review leads to a discussion on what we know, and what is as yet unknown, about an older population with ID/ASD. Most of the evidence is drawn from cross-sectional study designs. A longitudinal study measuring social and occupational functioning, residential and care needs and healthcare data is indicated to provide for future generations with ID/ASD.This review is best categorised as a scoping review, a type of systematic knowledge synthesis which identifies the main concepts, theories, sources and knowledge gaps within the evidence base for a particular topic (Tricco et al., Reference Tricco, Lillie, Zarin, O'Brien, Colquhoun and Levac2018). It is therefore a brief and broad overview which the reader may use to identify topics of more specific interest for systematic appraisal or meta-analysis. It is possible, for example, that the planner of a residential facility for adults with ID/ASD may wish to consult the view of family or caregivers, the evidence base on which has been excluded from the scope of this paper for expedience. Different studies use different cutoffs to define ‘ageing’ or ‘older’ and employ different stratifications of age groups which limit direct comparability of subjects. Ten papers of possible interest were not included as the full text could not be retrieved, and the 51 papers not in English may have added a different perspective, given the global burden of ID and ASD and the cultural and socioeconomic differences between international healthcare systems.
Conflict of interest
Author [EM] has no conflicts of interest to disclose.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committee on human experimentation with the Helsinki Declaration of 1975, as revised in 2008. The authors assert that ethical approval for publication of this review was not required by their local Ethics Committee.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.