Older Autistic Adults’ Experiences, Attitudes and Support Needs

Abstract

The paper presents new empirical data on the living and employment situations, aspirations and barriers to improving these situations of 78 autistic people aged 50-79. It uses a mixed methods approach with both qualitative and quantitative data collection from a single survey instrument. The work makes a significant contribution to the scarce literature on older autistic people, which is rarely concerned with their lived experiences. It provides both statistical data and gives older autistic people a voice through direct quotes from them which show them as actors and agents in their own lives. The analysis shows the importance of appropriate support as a potential enabler and contributor to quality of life, but more often its lack acting as a barrier to high quality living and occupational situations. It also highlights both some broad similarities between older autistic people’s living and employment situations and aspirations to those of the general older population, and some important differences. These differences included a greater need for quiet environments and their own space, sometimes even separate from partners, concerns about being accepted for who they are and a greater likelihood of living along and less likelihood of living with a partner.

Keywords: older autistic, experiences, living situations, occupational situations.

ACM CCS: user studies, people with disabilities, user characteristics,

1. Introduction and Literature Review

There are a significant number of older autistic people in the population, but a variety of factors, including underdiagnosis and autistic people’s raised mortality rates from all causes [1] make calculation of accurate percentages and numbers at different ages difficult. Calculations based on available data [2, 3] give between 2.96 and 7.63 million autistic people aged 50 and over in the European Union on 1 January 2024. This very wide range indicates the need for a better methodological approach and additional data collection to obtain more accurate values.

Although autism is a lifelong condition, research and practice have generally been on autistic children rather than adults, though this is gradually starting to change. However, there has been minimal work on the experiences of older autistic people. The term autistic people will be used for everyone on the autistic spectrum including those with Asperger’s syndrome, as the term preferred by many autistic people [4]. The aim of this study is to explore the living and occupational situations, barriers to improving them and aspirations of older autistic adults. Our research draws on the social model of disability [5] and the compatible neurodiversity model [6] according to which autistic people experience social, attitudinal and infrastructural barriers and frequently also social exclusion e.g. [7]. This is often compounded by non-autistic people’s lack of understanding and acceptance of ‘autistic’ differences in thinking, moving, interacting, and sensory and cognitive processing, and deficit-based ‘labelling’ and terminology and a frequent focus on the need to change autistic people’s ‘deficits’ [8].

Healthy ageing involves participation in activities, which may include education, employment and civic engagement, as well as leisure, and daily living activities, and not just accessing services. However, existing research shows that the life experiences of autistic people are often negative with regard to mental and physical health, life expectancy, education and employment and discrimination and stereotyping. For instance, autistic people have high rates of mental and physical health conditions [9], including anxiety and depression [10], very high suicide rates and higher mortality rates from all causes than non-autistic people [1] and significantly reduced life expectancies. They also frequently experience social isolation, including from family members, negative attitudes and discrimination, including risk of job loss due to being autistic [11].

Much of the research is comparative, either of the experiences of older autistic and non-autistic people or of those of older and younger autistic adults. There is very limited work on older autistic people’s living situations or their experiences of employment or sports or leisure activities, though some literature on instrumental activities of daily living (IADLs). Sensory issues and difficulties with uncertainty and in planning and carrying out tasks largely due to their cognitive rather than physical requirements have been found to make carrying out IADLs challenging to autistic adults aged 50 to 73 (n=15) without cognitive impairments [12]. Partners were the preferred source of support, as participants found that support workers could be intrusive and they had to negotiate their needs with them.

The limited research on older autistic people means that there is little evidence available on the ways in which their experiences are similar and those in which they are different from those of younger autistic adults. The fact that many older autistic people have not been diagnosed [13] or are diagnosed late and therefore have not received appropriate, if any, support for much of their lives may have a significant impact on their experiences. This often leads to older autistic people facing a confusing, illogical and often hostile world without understanding the reasons for this. They may blame themselves for the difficulties they frequently encounter in line with the common focus on the need to change the ‘deficits’ of autistic people [8]. In addition, older autistic people have a combination of the experiences of autistic and older people, including the negative stereotyping of older people, negative images of ageing e.g. [14, 15] and potential for age related impairments.

Research shows that autistic people have low education and employment rates, which are lower than those of other disabled people [16,17], Therefore, poverty is often a problem for autistic people throughout life and continuing into old age. They often experience social, attitudinal and infrastructural barriers and social exclusion e.g. [7]. Various factors, including sensory issues, differences in social interaction and different design requirements [18] need to be taken into account in service provision. However, specialised service provision has generally focused on autistic children rather than adults and personnel are unlikely to have appropriate training for working with older autistic people [18].

The few solutions aimed specifically at older autistic people include a study of the requirements of older autistic adults for residential care [18]. While taking a positive approach to including older autistics, this project did not involve any residents of care centres and lacked discussion of when residential care was appropriate for older autistic people. In addition, older adults have a strong preference for staying in their own homes [19]. This may be particularly important for autistic people due to sensory issues, often needing to control or restrict their interactions with other people and a strong preference for a familiar and stable environment. Some architects have recognised the importance of a quiet room or calm space autistic people could retreat to in homes and public buildings [20].

The paper is set out as follows. The methodology is presented in section 2. This is followed by the results in section 3 and discussion in section 4. Conclusions are presented in section 5. Brief overviews of the literature on older autistic people’s mental and physical health and cognitive impairments and quality of life are presented in Supplementary Materials.

2. Methodology

2.1 Research Questions

The research was carried out to answer the following two research questions:

1. What are the living situations of older autistic people, how do they compare to their desired living situations and those of older non-autistic people and what barriers do they experience to a better living situation?

2. What are the occupational situations of older autistic people, how do they compare to their desired occupational situations and those of older non-autistic people and what barriers do they experience to a better occupational situation?

2.2 Procedures and Ethical Approval

A mixed methods approach was used with both qualitative and quantitative data collection [21] from a single survey instrument. Data analysis was also carried out in parallel with the qualitative data used to expand on and further explain the quantitative results and, in particular, to obtain information about desired living and employment situations and barriers to achieving them. We developed the study with input and support from an Advisory Board comprising three older autistic people. Ethical approval was obtained from the Ethics Committee of the University of Glasgow College of Science and Engineering (reference number 300210009). The questionnaire was made available on the JISC online survey site (https://www.onlinesurveys.ac.uk/) between 1 November and 21 December 2021. This site was chosen as it is available through Glasgow University, meets data protection and privacy requirements and is reasonably accessible to autistic people, including those with other impairments. The information sheet was presented on the first page of the questionnaire. Written consent was obtained from a consent question which was the only compulsory one on the questionnaire. It had yes/no answer options rather than requiring a signature to maintain anonymity. This was approved by the Ethics Committee.

The inclusion criteria for the study reported in this paper were: 1) being autistic, whether through a medical diagnosis of an autism spectrum condition or self-diagnosis (and possibly waiting for a medical diagnosis); 2) being age 50 or over. . We used a lower age limit of 50, as it is used in several previous studies e.g. [22, 23] and our funders use it in their definition of ‘older’. In addition, it gives us a potentially large study population, for instance more than 50% in the European Union [2].

We used a variety of approaches to publicise the questionnaire, including posting on lists aimed at autistic or disabled people, professionals working with them and researchers, social media and asking our contacts and those of the Advisory Board to circulate information about the study. We also contacted a number of organisations asking them to circulate information and offering to send them hard copies of the questionnaire. In addition to contacting organisations of autistic people, including black autistic people, we also contacted some diagnosis centres, health centres/doctors’ surgeries and churches. We also used a flyer to support our publicity.

The answers to the research questions were obtained from the personal information section of this survey. It was placed last rather than, as more commonly, first, as some autistic people could find requests for personal information difficult or off-putting. However, after answering other questions which they found relevant and interesting they would be more likely to answer these questions. The text of this section is presented in Supplementary Materials.

The other three sections covered attitudes to and experiences of technology, websites and app functions; features of interest in a potential solutions; and attitudes to different payment and funding models. The first draft of the questionnaire was produced by a research team of three older autistic people, comprising an academic researcher and two peer researchers. Two revisions were carried out in response to comments from the Advisory Board comprising three older autistic people. It was then piloted with five older autistic people. The results of all the feedback were used to make a few changes in question wording and response options.

2.3 Analysis

For the purpose of analysis we split the participants into two groups: people aged 50-59 and people aged 60-79. The closed questions were analysed using statistical techniques, including calculation of percentages followed by calculation of statistical significance where there were difference between the responses of the 50-59 and 60-79 age groups. Data was rounded to the nearest integer, as, although there were a good number of participants, there were insufficient to justify greater precision. As a result the data may not sum to 100%. Statistical significance at the 0.05 level was calculated using the Fisher exact test for variables with only two values and c² for variables with more than two values for which the Fisher exact test cannot be used. Software developed by Langsrud [24] was used for the Fisher exact test and by SocSciStatistics [25] or Preacher [26] for c².

Analysis of the qualitative data in the open questions involved a thematic analysis [27] carried out in several stages. The first stage was the identification of first-order themes without a reference framework to obtain an overall understanding of the data. This involved inductive coding of the responses of each participants aged 50 and over to each of the open questions. Each separate concept in each response was given its own code. Splicing was then carried out to combine codes to produce a small number of themes which encapsulated the data. This also involved the physical movement of text to different sections of the paper. A further analysis stage was carried out to determine whether the themes linked to the topics of the closed questions. This was followed by triangulation to investigate the relationship between the qualitative and quantitative data, including any apparent inconsistencies and where the qualitative data explained the responses to the closed questions.

3. Results

The results section is divided into three subsections. The first presents an overview of the participants. Sections 3.2 and 3.3 present the results related to the two research questions on living and employment situations respectively. Three main themes were identified from the participant comments. They are lack of support, desire for changes, and managing different patterns of employment and retirement, and a minor theme: satisfaction with the existing situation.

The lack of support theme was the most prominent with regards to the number of comments. It was relevant to both participants’ living and employment situations. We have therefore decided to present this theme in both sections 3.2 and 3.3. While other features, such as independence and interdependence were also salient in this theme, we considered that the expression ‘lack of support’ best described it. We consider that this theme acted as a unifying factor that brought the data from the two research questions together. The same was true, but to a lesser extent as it was a minor theme, of the ‘satisfaction’ theme. We included it despite the small number of positive comments to show that some participants did comment about things they were happy with. The ‘lack of support’ theme also fits well with the social model of disability and the associated neurodiversity model. It relates to the barriers (older) autistic people experience to taking control of their lives and achieving what they want. The other two themes, ‘desire for changes; and ‘managing different patterns of employment and retirement’ are only commented on in the contexts of living situations and employment situations respectively. They are therefore presented in sections 3.2 and 3.3 respectively.

3.1 Participant Overview and Demographic Information

113 responses were received from autistic people aged 21 to 79. They included 33 people aged 50-59, 40 aged 60-69 and 5 aged 70-79. Data is provided separately for the 50-59 (average age 55, standard deviation three years, subsequently referred to as 50*) and 60-79 (average age 64, standard deviation five years, subsequently referred to as 60+) age groups to allow investigation of the impact of increasing age. As there were unfortunately only five participants 70 and over, insufficient to analyse separately and compare with other groups, their data and that of the 60-69 year old participants has been grouped together in the 60+ age group.

The majority of participants had a medical diagnosis of an autistic spectrum condition (see table 1). However, a much lower percentage of 60+ participants (64%) than 50* ones (79%) had a medical diagnosis and over twice the percentage of 60+ (23%) compared to 50* (6%) were self-diagnosed, with another 3% of 50* indicating both self- and medical diagnosis. However, the difference in the percentages of participants in the two age groups who had a medical diagnosis and self-diagnosed was not statistically significant (c² = 3.75, p = 0.05284). Therefore, further research with a larger number of participants would be required to investigate the effect and possible causes.

Female participants were the majority at about 60% for both age groups. The percentage of male participants was much higher for 60+ than 50*. A small percentage of 60+ participants (2%) and a moderately significant minority of 50* participants (12%) were non-binary, but the difference was not statistically significant (c² = 3.11, p = 0.07783). Just over half (55%) of the 50* group and just over three quarters of the 60+ group (78%) were from England with just over a fifth (21%) and just over a tenth (11%) from Scotland for 50* and 60+ respectively. Other countries represented were Ireland and the USA [60+] and Wales, Northern Ireland, Germany and Norway [50*].

	Medical diagnosis	Seeking diagnosis	Self and medical diagnosis	Self- diagnosed	Self diagnosed & seeking one	Unknown[1]	Male	Female	Non-binary
50* no.	25	5	1	2	0	0	9	20	4
50* %	76	15	3	6	0	0	27	61	12
60+ no.	28	5	0	10	1	1	18	26	1
60+ %	64	11	0	23	2	2	40	58	2

Table 1 Gender and autistic status

3.2 Living Situations

This section is divided into four subsections. Section 3.2.1 presents the quantitative data. Sections 3.2.2, 3.2.3 and 3.3.4 present the qualitative data under the two main themes and minor theme relevant to living situations. These themes are ‘lack of support’, ‘desired changes’ and ‘satisfaction with the existing living situation’.

3.2.1 Quantitative Data

A surprisingly high proportion of the 60+ participants (40%) lived in a rural area or village compared to only 15% of 50* participants (see table 2). This difference is statistically significant (p = 0.0235892).

The percentages of 60+ living in other types of location was correspondingly smaller than that of 50*, particularly for small towns or cities and the percentage of 50* in large cities nearly equalled that of 60+ in rural areas or villages, indicating a possible population movement with age. The differences between the distributions of participants in the two age groups between different types of towns and cities and villages was statistically significant (c²= 8.231, p = 0.0414712).

	60-79		50-59
	No.	%	No.	%
Large city	11	24	12	36
Medium sized town or city	10	22	7	21
Small town or city	5	11	9	27
Small town or city/rural area or village*	1	2	0	0
Rural area or village	18	40	5	15
Total	45		33

*One participant marked both these options.

Table 2 Type of area participants lived in

The majority of 60+ participants were equally split between those living alone (42%) and those with a partner (40%). Most of the remainder lived with their children (see table 3). Two 60+ participants lived in sheltered or retirement housing. Most 50* participants fitted into four groups, with a third living alone, 27% with a partner and 36% with a partner and children, and 12% with children. One 50* participant lived with a close friend. The differences in living situation between the two age groups were not statistically significant (c²= 6.262, p = 0.0995363).

Living situation	60-79		50-59
	No.	%	No.	%
Alone	19	42	11	33
With partner(s) or spouse	17	40	9	27
With children	5	11	4	12
With partner(s) and children	1	2	8	24
With close friend(s)	0	0	1	3
Sheltered or retirement housing	2	4	0	0
Total	45		33

Table 3 Living arrangements

3.2.2 Lack of Support: Living Situation

Lack of support affected a number of participants’ ability to live a dignified and quality life. For instance, a wheelchair using 60+ participant was ‘housebound’ as they lived alone with ‘no one to push my wheelchair to take me out, which would be pleasant’.

The two 60+ people living in supported or seniors’ housing highlighted major problems with lack of respect and appropriate support. One lived in a ‘non … regulated … individual living flats … and I have to pay to have a second proper firm of carers come in for an hour, 5 days a week … moral support, to counteract or mitigate the abuse and criminality’ of the charity supposed to be doing this. The other ‘would really appreciate it if the council could ‘mend what is broken’ and found it ‘an ongoing struggle to get help’. Further related comments included wanting a ‘cleaner to help’ and a carer ‘instead of everyone telling us my husband is my carer when he can’t meet my needs’ [50*].

The lack of support theme also included concerns about suitable accommodation in the future expressed by two of the 50* group. One lived ‘alone in a housing coop’ and ‘fear[ed] for the future … as this government may try to destroy social housing, and I have no assets or inheritance’. The other would ‘need to live with others when older’ and hoped ‘to have other autistic people renting there’.

A further concern about being accepted and valued rather than marginalised and excluded was expressed by one of the 60+ participants ‘when I need elder care it might be in place where people who think differently (me) might feel welcome because the staff understand the full range of neurodiversity’.

3.2.3 Desire for Changes

The desire for changes theme focused on the need for their own space, including through adult children moving out or becoming more independent. Comments included ‘eldest daughter living with us out of work and would love her to find a job’ and ‘I have my adult son (asd) here I prefer to live alone’ [60+] and ‘In a rental home with my adult autistic child. … I would prefer to live alone.’ and wanting ‘both boys to become [financially] independent but we don’t yet know if this is possible [50*]. Other comments about their own space included ‘live alone in someone else’s home, but would like my own place’ and related to managing their own space while having a partner: ‘I’d like to find a partner but would probably still prefer to live apart so I can control my living space’.

Other desired changes related to making their accommodation better meet their needs, including through noise reduction. ‘I would prefer to live in a quieter place’ [50*] and I would like my flat to be sound-proofed. The insulation is very poor and noise makes me ill [60+].’ There was also an interest in moving house, for instance ‘to live nearer to my loved ones’ [60+]. One participant wanted to move, but found this difficult: ‘need to find the energy to deal with it, and decide where’ [50*]. Another participant desired a better working situation, as they were working from home and their partner was using the PC and printer: ‘office space or equipment at home or nearby’.

3.2.4 Satisfaction with the Existing Living Situation

A few participants made positive comments about their living situation, housing and/or independence, sometimes without providing details, for instance ‘I am happy with my living arrangements’. Comments from 50* participants included ‘I’m happily married, I live in a nice council flat with my wife’ and ‘I live in a house myself and my partner are purchasing. I have adapted the place to be more friendly to my living requirements’. Comments from 60+ included the only positive comment about living alone: ‘I live on my own, but live independently’, a well as positive comments about living with family: ‘I am lucky. Live with husband in our own house’ and ‘happy living with husband and … our 23-year-old daughter’.

3.3 Occupational Status

This section is divided into four subsections. Section 3.2.1 presents the quantitative data. Sections 3.2.2, 3.2.3 and 3.3.4 present the qualitative data under the two main themes and minor theme relevant to occupational situations. These themes are ‘lack of support’, ‘managing different patterns of employment and retirement’ and ‘satisfaction with the existing occupational situation’.

3.3.1 Quantitative Data

Probably unsurprisingly, nearly a third (31%) of the 60+ group were retired, but only 6% of the 50* group (see table 4). Another 9% of the 60+ group combined retirement with some form of employment, such as working part time or sessionally or being self-employed. Self-employment was the most significant occupation for 50* (33%), but only half that for 60+ (16%). Full time employment was more common for both groups than part time at just under a quarter compared to about 5%.

The difference in the percentages of participants in the two age groups who were fully retired was statistically significant (c² = 7.32, p = 0.006792), as were the overall differences in occupational status (c² = 14.049, p = 0.0070010), but the difference between the percentages of self-employed participants was not (c² = 3.39, p = 0.065608).

	60-79		50-59
	No.	%	No.	%
Retired	14	31	2	6
Retired & part time, sessional or self employment	4	9	0	0
Unemployed	5	11	4	12
Part time employment	2	4	2	6
Full time employment	10	22	8	24
Full time employment & own business	1	2	0	0
Self-employed	7	16	11	33
Carer	1	2	2	6
Student	0	0	2	6
Student and carer	0	0	1	3
Disability benefit	0	0	1	3
Freelance and writing but not earning	1	2	0	0
Total	45		33

Table 4 Occupational status

3.3.2 Lack of Support: Occupational Situation

Participant comments under the lack of support theme related to barriers to working, though participants often considered this due to health and disability issues rather than a lack of support affecting their ability to work. Several 60+ participants mentioned ‘disabilities’. In at least one case this was caused by work-related stress, ‘burnout … after 20 years in IT earning £100+’, now ‘signed off sick’ as postman ‘as I cannot cope with the workload’ [50*]. Another 60+ participant who considered themselves ‘very qualified’ was experiencing ‘exhaustion from decades masking, … undiagnosed autism and EDS’ (Ehlers-Danlos syndrome).

A 50* participant would ‘love help to get better and go back to work after my degree’, but did not specify the nature of this help. A 60+ participant who had not worked for 25 years was more explicit about support in terms of ‘trauma therapy … and diagnosis earlier’ in making them ‘able to work before now.’ However, support in terms of ‘people who did the domestic work etc’ had enabled them to work outside the UK until having a child.

However, this does not mean that employment is necessarily the best option for everyone. For instance, one 50* participant was unsure whether they could ‘work with enough support’ and was ‘look[ing] forward to being officially retired’. The impact of pain was mentioned as something making retirement desirable rather than giving rise to a need for support or being a barrier to working. For instance, a 50* participant was wanting to retire ‘so not in pain’, but worried about not being able to afford to. The lack of social care had reduced some participants’ choices. One 50* participant had given up work to become a ‘full time unpaid carer’ and several 60+ participants were full time (unpaid) carers.

A lack of support could result in (early) retirement. For instance, the 60+ group included participants who had retired ‘medically’ and who were forced to retire due to age discrimination, including a 60+ participant due to ‘attitudes and abilities associated with a ‘certain age’.

3.3.3 Managing Different Patterns of Employment and Retirement

This theme included a few 60+ participants with multiple (part time) jobs. One of them had ‘expertise in two careers’ which they worked ‘concurrently’ together with ‘some part time additional work’. They would like to be able to ‘afford to retire from some of my work’, but recognised that they had ‘never been able to do just one job full time as I get bored’. Some participants, including those with multiple jobs, were underemployed, as their jobs were not at an appropriate level. For instance, a 60+ participant had ‘multiple part time jobs’ and was ‘not using much of the skills, knowledge and experience I have’. Participants included retirees who were still working, sometimes to earn money for a specific purpose, such as ‘to save money for my daughter’s solicitor’s exams’; or part time, for instance at a local university which they ‘love’ and where they found ‘students warm to my different ways of thinking’.

Included in this theme were participants who wanted to retire partially or reduce workloads, but could not afford to. For instance, two 60+ participants wanted to retire, one ‘to do only the work I like doing but can’t afford to’ and another ‘a lower workload and better life-work balance, but not a part-time salary’. The other side of this was a few 50* participants wanting to increase their hours, pay and job security and career prospects, for instance ‘better pay and a chance for progression and support to do so’ and ‘studying to get a better job’. One with ‘very part time hours’ would ‘like to manage more’ and another had spent ‘too many years of earning almost nothing, or claiming benefits, for mental health reasons’ which they ‘recently realised were contributed to heavily by autism’.

3.3.4 Satisfaction with the Existing Occupational Situation

A few participants made positive comments about their work or being retired. Being positive about work included ’illustrator alongside … Autism-based projects … not a full time professional Autistic yet but getting there’ and ‘my employment is not high level but suits my situation … some adaptations’ [50*] and ‘found my niche in work and enjoy what I do’ and ‘courier driver … I love it’ [60+]. A few 50* participants indicated that they were ‘happy to be retired’.

4. Discussion and Answers to the Research Questions

The research has investigated older autistic people’s living and occupational situations, their aspirations for improving them, the barriers to improvement and made comparisons of these situations for the non-autistic older population.

4.1 Participant Profiles

It has been suggested that older people should be divided into ‘younger old’ and ‘older old’, with the former active and the latter increasingly frail and from the age of 80 or 85 [28, 29]. Thus, age-wise our participants are in the ‘younger old’ group. However, there are questions as to whether the age boundary between the two groups should be lower for older autistic people and problems with associating the second group with ‘disability’.

The much higher percentage of self-diagnosed 60+ than 50* participants and lower percentage of medically diagnosed ones may reflect the lack of options for obtaining a diagnosis when participants were younger in line with the idea of the ‘lost generation’ of undiagnosed or late diagnosed autistic adults [13]. It may also indicate that a number of older participants had decided to self-diagnose rather than seek a medical diagnosis, possibly due to (perceived) difficulties in obtaining a diagnosis and feelings by some 60+ participants that it was no longer worthwhile.

The higher percentage of female participants, despite the much higher percentage of diagnosed males, is in line with other research on autistic people e.g. [30]. The relatively high percentage of non-binary people is in line with the literature on the high percentage of autistic people who are trans and/or reject binary approaches to gender [31]. However, there does not seem to be any research on how the age of autistic people affects their gender identity and it would be interesting to investigate this, particularly as the non-binary (agender and agnostic) identification was higher for participants aged under 50, at 14%.

4.2 Living Situations

Examining the quantitative and qualitative data raises the questions of how older autistic people can have the space and quiet they need while avoiding social isolation and obtaining appropriate support to meet their needs.

4.2.1 Current and Comparative Situations

Approximately equal numbers of 60+ participants lived on their own (42%) and with partners or spouse (40%), whereas just over half (51%) of 60+ participants lived with partners or spouse, and in many cases also with children. A higher percentage of the 60+ participants lived alone than in the general 65+ population in England and Wales (36% of women and 23% of men) and a considerably lower percentage lived with a partner or spouse than those in this group who are legally married or in a civil partnership, including those separated (58%). However, the data is sufficient to counter still-prevalent stereotypes that autistic people are incapable of having relationships e.g. [32].

4.2.1 Current and Comparative Situations

As well as a higher percentage of 50* than 60+ participants living with partners, a much higher percentage of them were living with children, 36% compared to 13%, in most cases also with partner(s) or spouse. This higher percentage was probably due to 50* participants having younger, though possibly adult, children and at least some of these adult children becoming financially independent and moving out over time. However, this was not universally the case, as both age groups commented on wanting their children to become more independent and/or move out. These comments raise the issue of whether (older) autistic people are more likely to have autistic or other disabled children who experience greater barriers to independence. This relates back to the need for additional or more appropriate support, in this case to enable these (young) adults to be more autonomous. The comments also indicate that (older) autistic parents both shared the concerns of other parents about their children’s futures and may also have had a greater need for their own space and therefore for their adult children to move out.

Social isolation has been found to be a problem for both autistic adults e.g. [11] and older people in general. About a third of older people experience social isolation and up to half of the over 50 population are at risk of getting further socially isolated [33]. While their comments indicate that at least some participants preferred living alone even if they had a partner, the higher percentage of 60+ than 50* participants doing this indicates the risk of increasing loneliness with age for autistic people. It also increases the risk of lack of support, particularly as the literature has found a preference for support from partners [12].

The high proportion (40%) of 60+ participants living in rural areas is paralleled by the higher percentage of older people than the general population of England in them. For instance, 25% of the population of rural areas are over 65, but only 17% of the population of urban areas [34]. This is often due to a combination of lower birth rates, young people moving to towns and cities for education and employment opportunities and older people moving to rural areas to retire [35]. Further research would be useful to investigate whether autistic people are ageing in place or moving to rural areas and villages as they get older and, if the latter, why these areas are particularly attractive to 60+ autistic people. The overwhelming majority of participants seemed to have their own accommodation, in line with the literature on 97.3% of people over 60 living in private accommodation [36]. More 50* than 60+ participants lived in both small towns and cities (27% compared to just under half that) and large cities (a third compared to a quarter).

4.2.2 Desired Living Situations

Having their own space was important to many participants, as shown by comments about preferring their own space and independence, a quiet environment and a ‘nice’ home, possibly with adaptations to meet their needs. Although not stated explicitly, the comments indicated that in some cases independence required the support of carers or other forms of assistance.

A number of participants valued having or wanted a partner, but not necessarily to live with them. They also valued their children becoming (financially) independent and/or wanted them to move out. However, a few participants were happy with their children living with them and another one wanted to move nearer to family.

Participants wanted quiet environments to live in and were interested in noise reduction through soundproofing or moving to a quieter place. This aligns with the fact that sensory issues, particularly noise, are a significant concern for autistic people e.g. [37]. The desire for a quiet environment may have motivated the surprisingly high percentage of 60+ participants (40%) living in rural areas with a lower percentage of 15% of 50* ones. The difference was statistically significant and compares with 25% of the rural population being over 65.

4.2.3 Barriers to Desired Living Situation

The main barriers to participants achieving their desired living situations were lack of money and lack of (appropriate) support. Appropriate support would have improved their living environments and quality of life through providing appropriate care, sound proofing and quieter rented accommodation, space and equipment at or close to home for home working, and support for their disabled adult children to become more autonomous. It could also have been used to overcome the barriers one participant was experiencing to moving and a wheelchair using participant to having someone to push their wheelchair so they could go out of the house. (Older) autistic people may have a smaller or no network of people they can contact to do this and may find it very difficult to contact other people to ask for assistance. It would have also enabled two participants to resolve the problems they were experiencing in supported or seniors’ housing accommodation and considerably improved their quality of life.

These negative experiences illustrate some of the ways in which housing and support provision can break down. The risk of break down has been increased by the significant reduction in the availability of social housing in the UK from a peak of 33% in 1981 to 18% in 2017 [38]. Participants expressed concerns about having suitable accommodation in the future, including through ‘this government … try[ing] … to destroy social housing’ and receiving ‘elder care’ in a place where they were ‘welcome’ as ‘the staff understand the full range of neurodiversity’. Concerns about suitable accommodation in the future and the need for appropriate adaptations are shared by many older adults, though this is generally expressed in terms of appropriate adaptations to allow ageing in place e.g. [39]. However, in the case of older autistic people there are additional dimensions of being welcomed for who they are rather than having to worry about whether they will be expected to change and needing their sensory, communication and other needs as older autistic adults met. For instance, the need for a quiet room or sensory free space for autistic people to retreat to has been recognised by some architects [20].

The lack of access to appropriate support could be experienced by other people, but may be more difficult to resolve for older autistic people due to the barriers they experience to communication, interaction and networking and executive functioning difficulties. Good maintenance and being treated with respect by staff have been found to be important for older people’s housing satisfaction, but not always to be the case e.g. [40]. However, older autistic people may find it more difficult to impossible to carry out repairs and decorate themselves and negotiate or put pressure on the renting organisation to do this and organise better carers. The difficulties reported by 50-74 year old autistic people in activities of daily living, largely due to their cognitive requirements [12] imply that many of them will experience even greater difficulties in carrying out home adaptations.

A participant commented on adapting their home to better meet their needs, but unfortunately did not provide details of the adaptations, which might have been useful to other autistic people. The issues of the types of home adaptations that autistic people require and support for carrying them out seem not to have been researched. They may require financial and other support to do this, as they are generally on low incomes and are less likely to have access to other people who are able to do help them adapt their homes. There are also issues of the availability of appropriately adapted social housing which need investigation.

Assumptions about the division of responsibility for providing support between society and the individual or their family could also act as barriers. For instance, one of the participants complained about the assumption ‘my husband is my carer when he can’t meet my needs’. This comment is in line with assumptions that the family is responsible for caring for its older members, though this responsibility generally falls on women [41]. There is clearly a need for societal provision of support, particularly where support from family and friends is not available or unsuitable. While other autistic people may prefer support from partners, as found by [12], partners and other family members may need training to provide appropriate support. There is also a need for both trained, including in working with autistic people, and appropriately vetted volunteers to, for instance to push (autistic) wheelchair users and a wider range of socially provided services with workers trained to support autistic people. In relatively rich countries such as the UK, the costs of doing this should not be prohibitive, raising issues of society’s priorities, which are beyond the scope of this paper.

4.3 Occupational Situations

The data raises the questions for what support and resources participants require to empower them to make and implement good choices about employment and retirement.

4.3.1 Current and Comparative Situations

The results illustrate older autistic people’s varied experiences of employment and retirement. This included retired participants working part-time, having multiple jobs, being underemployed, wanting to increase hours, pay and job security, and experiencing barriers to working. A quarter of both groups were working full time and about a third of the 50* and just under half that percentage of 60+ were self-employed, with the difference not statistically significant. The percentage of participants in employment is higher than generally quoted in the literature e.g. [16, 17]. This suggests a need for further investigation of the employment rates of autistic people, including older ones, and whether there are specific factors or groups of older autistic people with higher employment rates. Since diagnosis rates decrease with age, it would also be useful to investigate any relationship between having a diagnosis and being in employment. A considerably higher percentage of the older age group were retired (31% compared to 6%) with the difference statistically significant.

Participants’ patterns of employment and retirement paralleled those in the general population. This includes total retirement being replaced by partial retirement with reduced hours and changed working patterns [42], early retirement followed by re-entry, often in a different form of work [43] and bridge employment [44], including as a self-employed person. This has led to suggestions that it is ‘the process of disengagement from the labor force’ that is important and when a person is defined as retired is ‘somewhat arbitrary’ [45]. The results to some extent also parallel a study of the general population of England which found that almost a fifth were employed after state pension age with two thirds of them continuing work because they ‘enjoyed’ it or to ‘keep active and fit’ and a third for ‘financial’ reasons [46]. Just over 30% of retirees in this study had done so involuntarily and just over a third each voluntarily or due to reaching state pension age.

The high percentage of self-employed participants is consistent with findings in the literature that self-employment of the general population over 50 or 55 is significant and/or increasing in many countries including the UK [47]. Many older self-employed people became self-employed later in life with a third of self-employed people over 50 making this transition at 50+ [48]. However, the much higher percentage of self-employed 50* than 60+ participants is out of line with the literature. For instance, several studies reported in [43] showed a self-employment rate of about 20% from ages 50 to 59 or 60 and then increasing to peak at 39% at age 66. However, this finding was not universal and, for instance, a study of a similar age group (50-75 year olds), found that only 7% were self employed and 57% were retired or not in paid employment [49].

Self-employment is an interesting potential solution for some older autistic people, as it gives greater control over working conditions, making it easier to manage health (or other issues) and work [48]. There is also evidence, though it is dated, of the success of self-employed older people. For instance, 70% of business start-ups with owner-managers over 55 survived more than six years compared to 19% of all business start-ups [50]. However, in addition to the risks of failure, current administrative and other requirements for self-employment may pose executive functioning barriers to many older autistic people.

Several 60+ participants were full time (unpaid) carers, but did not comment on whether they had left employment or not tried to obtain it as a result of being carers. There was also a lack of comments from these participants and another one who had left work to become a carer on the choices they would have made if high quality affordable social care had been available to the people they were caring for.

The results also illustrate the multi-dimensional nature of autistic people’s lives and support needs. Juggling the demands of paid employment and domestic work and possibly also child or other dependent care are an issue for many workers, but may put significant additional pressures on autistic people due to a combination of the time involvement, executive functioning issues and cognitive demands [12], as well as a lack of support.

Desired Occupational Situations

Participants wanted to be in a position to make good choices about their employment and retirement without barriers to implementing them. Many participants wanted to be able to work at well or at least reasonably paid employment with appropriate hours and an appropriate level of challenge without (excessive) stress and which was suited to their skills and interests. Others wanted to be able to afford to retire with a reasonable income or move between employment and retirement or combine them. More 50* participants were interested in better employment opportunities and more 60* ones in being able to afford to retire, though this was not universal. This would need further research, but it may be that many 50* autistic people considered they could still overcome the barriers they had experienced to a satisfying career, whereas many 60+ ones had given up on this and would therefore prefer to be able to retire.

4.3.3 Barriers to Desired Occupational Situations

Many of the barriers to appropriate employment and retirement choices arose from the nature of many work places and the resulting excessive stress, overwork and/or low incomes. Other barriers included the lack of support for disabled people and those with long-term health conditions to work, the lack of early diagnosis and treatment, and age and possibly also disability discrimination which had led to some participants being forced out of employment and probably made it more difficult for others to obtain it.

Participants generally considered the difficulties they experienced to be related to health and disability issues rather than barriers and discrimination and several participants commented on the impact of disability and health conditions affecting their ability to work. However, as indicated by a comment about ‘burnout … after 20 years in IT earning £100+’, the issue may often not be managing health conditions and work, but managing excessive workloads and expectations and high stress. There is a body of literature on work-related stress e.g. [51], but less discussion about how both specific additional stresses and general workplace stress factors affect autistic people. A comment on ‘exhaustion from decades masking’ indicates the impact of masking in the workplace, in line with the literature on its impacts on autistic people e.g. [12]. The further comment about undiagnosed conditions implies that it is difficult to obtain both a diagnosis of autism and other conditions quickly and appropriate support subsequently. Some participants considered that they would be unable to work, even with appropriate support. The main barriers to participants being able to retire (fully) when they chose to were financial. Although participants did not discuss this, the solution is probably a significant increase in the state pension and being able to claim it earlier.

The comments about barriers to employment are in line with a Polish study [52] which found that the most important factor enabling older people to work was workplace adaptations to meet their needs. There is direct mention of age discrimination, including through stereotypical assumptions about abilities (or their lack), but not of discrimination or stereotypes on account of being autistic, though the lack of support or adaptations in many cases could be considered discriminatory.

The different perspectives on barriers to work being due to health and disability issues or a lack of support and expectations of behaviour align with the medical [53], and the social [5] and compatible neurodiversity [6] models of disability. A change in attitudes in society as a whole to disabled people and those with long-term health conditions to be more in line with the social model of disability [5] would facilitate greater understanding of the needs of older autistic people and possibly also greater prioritisation of available resources. The latter should include higher state pensions and financial support for those unable to work. However, any costs of supporting (older) autistic people into and in work are likely to be outweighed by the benefits to society.

5. Conclusions

The paper has presented new empirical data on the living and occupational situations of 78 older autistic people to discuss their living and occupational situations. The data was used to answer research questions about these situations, how they changed with age and compared to those of non-autistic people, the barriers to improving them and their desired living and employment situations. The responses to these questions are presented in subsection 6.1 and 6.2.

The combination of 33 responses from age 50-59 participants and 45 from 60-79 ones allowed us to compare the experiences of these age groups and partially answer the research questions on changes with age, as well as comment on the experiences of over-50 people overall. However, we only obtained five responses from age 70-79 participants, so did not analyse them separately. The majority of participants had a medical diagnosis of an autistic spectrum condition, but a much higher percentage of the 60+ than 50* participants had self-diagnosed. There was a reasonable gender distribution, with more women than men and a number of non-binary participants, particularly in the 50* group.

5.1 Strengths and Weaknesses

The greatest strength of the research was that it was carried out by a team of older autistic people supported by an advisory committee of older autistics and piloted by older autistic individuals so that it was informed throughout by older autistic perspectives and experiences. A further strength was the choice of topic and the move away from a health and deficit perspective. This is again linked to the co-production approach and underlying older autistic perspective focussing on opening up opportunities and quality of life. Another important strength was considerably extending the literature on older autistic people, which is currently very sparse, both in general and on the topics of their experiences and living and occupational situations in particular.

The main weakness was the fact that the participants were not fully representative of the older autistic population. In particular, there was low representation of ethnic minority older autistic people, though older black and ethnic minority people are underrepresented in the population as a whole, for instance only 7.3% in the 65-74 year old population of England and Wales [36] and the oldest age groups, with no participants over 80. It is also unlikely that older autistic people with learning disabilities were involved to any significant extent due to the cognitive demands of the questionnaire. In addition, the comparisons with the general non-autistic population were based on existing data from large scale studies rather than a control group of non-autistic people. We do not consider the fact that participants came from several different countries in Europe and North America a limitation.

5.2 Further Work

While the paper has contributed to filling some knowledge gaps, there is still a dearth of research on older autistic people and consequently a need for research in all areas, including their experiences, aspirations and support needs. Studies should preferably involve (older) autistic researchers, both peer researchers and academic ones to give an (older) autistic perspective; They should cover the whole autistic community, including ethnic minority autistic people, autistic people over 80 and older autistic people both with and without cognitive impairments. To date there has been a divide between research with autistic participants who generally do not have cognitive impairments and the use of carers, support people and others to provide information about autistic people, including those with cognitive impairments. There is a need for the development and wide dissemination of methodologies enabling the direct participation of older autistic participants with cognitive impairments in research, possibly with support, rather than the use of proxy informants.

The paper identified a number of specific questions where further research is required, possibly as part of a larger study. This includes the following:

The numbers and percentages of older autistic people (over several different age ranges) in the population in different countries and regions through direct research and developing a robust methodology for doing this.
How the relative percentages of older autistics who obtain a medical diagnosis and self-diagnose vary with age and potential explanations.
How the percentage of older autistics living in villages and rural areas varies with age, comparison with the percentage of non-autistic people in villages and rural areas and potential explanations.
How identification of autistic people as trans, non-binary, agender etc varies with age and potential explanations.
The extent to which self-employment is an appropriate solution for at least some older autistic people and whether and, if so, what factors determine its suitability for them.
Determining the employment rates of older autistic people, any relationship between diagnosis and employment and the reasons for it, if it exists.
Investigating whether younger groups of older autistic people are more interested in support to continue in employment and older groups in being able to afford to retire and the reasons for this.
How national cultural factors and attitudes to ageing affect the living and occupational situations of older autistic people.

5.3 Final Comments

The results show older autistic people as actors and agents in their own lives, as well as experiencing a range of barriers which reduced their options. There were both some broad similarities between their living and employment situations and aspirations and those of the general population, and some important differences. The latter included the higher percentage of older autistic people living alone and in rural areas or villages, and a higher percentage of self-employed people. In the workplace they were disadvantaged by the lack of support and reasonable adjustments to meet their requirements as older autistic people.

Important differences in aspirations included particular stress on and a greater need for quiet (often visually as well as in terms of sound) and their own space, sometimes not shared with even a partner. They also had a particular need for support for their (adult disabled) children to become more autonomous and move out, showing both care for them and a need for their own space. They probably also had a greater need for access to volunteers and other forms of support to ensure, for instance, repairs were carried. They also had particular concerns about being treated with respect and understanding for their differences if they required care later in life.

Acknowledgements

We would like to very gratefully thank all the autistic people who completed our questionnaire and shared their experiences and expertise with us, our Advisory Board, UKRI for funding and ZINC, particularly Tim Shakespeare, for their support.

Funding

UKRI Healthy Ageing Catalyst Award, Grant reference ES/W006588/1

Data

Anonymised data for which the participant gave permission has been deposited in the University of Glasgow data repository with an embargo date of 8/1/2024. It can be accessed at http://dx.doi.org/10.5525/gla.researchdata.1386

Conflicts of Interest

There are no conflicts of interest

References

Hirvikoski T, Mittendorfer-Rutz E, Boman M, Larsson H, Lichtenstein P, Bölte S. Premature mortality in autism spectrum disorder. The British Journal of Psychiatry. 2016 Mar;208(3):232-8.
Eurostat S. Population on 1 January by age group and sex. 2023. https://ec.europa.eu/eurostat/databrowser/view/DEMO_PJANGROUP__custom_1269323/bookmark/table?lang=en&bookmarkId=66c88f91-bde0-4606-abbd-bd1a65c8224a

Sacco R, Camilleri N, Eberhardt J, Umla-Runge K, Newbury-Birch D. The prevalence of autism spectrum disorder in Europe. Autism Spectrum Disorders-Recent Advances and New Perspectives. 2022 Oct 25.
Kenny L, Hattersley C, Molins B, Buckley C, Povey C, Pellicano E. Which terms should be used to describe autism? Perspectives from the UK autism community. Autism. 2016 May;20(4):442-62.
Barnes C. Disabled people in Britain and discrimination: A case for anti-discrimination legislation. London.; Hurst Co.; 1991.
Davidson J, Orsini M, editors. Worlds of autism: Across the spectrum of neurological difference. U of Minnesota Press; 2013 Nov 1.
Humphrey N, Lewis S. Make me normal’ The views and experiences of pupils on the autistic spectrum in mainstream secondary schools. Autism. 2008 Jan;12(1):23-46.
Dinishak J. The deficit view and its critics. Disability Studies Quarterly. 2022 Dec 25;36(4).
Miot S, Akbaraly T, Michelon C, Couderc S, Crepiat S, Loubersac J, Picot MC, Pernon É, Gonnier V, Jeandel C, Blain H. Comorbidity burden in adults with autism spectrum disorders and intellectual disabilities—a report from the EFAAR (Frailty Assessment in Ageing Adults With Autism Spectrum and Intellectual Disabilities) study. Frontiers in psychiatry. 2019 Sep 19;
Gillott A, Standen PJ. Levels of anxiety and sources of stress in adults with autism. Journal of intellectual disabilities. 2007 Dec;11(4):359-70.
Jones SC, Gordon CS, Akram M, Murphy N, Sharkie F. Inclusion, exclusion and isolation of autistic people: Community attitudes and autistic people’s experiences. Journal of Autism and Developmental Disorders. 2022 Mar;52(3):1131-42.
Hwang YI, Foley KR, Elley K, Brown S, Joy-Leong D, Li X, Grove R, Trollor J, Pellicano E, Zheng L. Experiences of Performing Daily Activities in Middle-Aged and Older Autistic Adults: A Qualitative Study. Journal of Autism and Developmental Disorders. 2023 May;53(5):2037-49.
Lai MC, Baron-Cohen S. Identifying the lost generation of adults with autism spectrum conditions. The Lancet Psychiatry. 2015 Nov 1;2(11):1013-27.
Neville S, Napier S, Adams J, Shannon K, Wright-St Clair V. Older people’s views about ageing well in a rural community. Ageing & Society. 2021 Nov;41(11):2540-57.
World Health Organization. World report on ageing and health. World Health Organization; 2015 Oct 22.
Putz C, Sparkes I, Foubert J. Outcomes for disabled people in the UK: 2020. Office for National Statistics. 2021 Feb 18:1-27.
Paradiz V. Autism, Identity and Employment, Transcending the Obstacles of the Working World. Autism Advocate, London. 2009.
Crompton CJ, Dawson M, Fletcher-Watson S. Residential care for older autistic adults: Insights from three multiexpert summits. Autism in Adulthood. 2020 Jun 1;2(2):121-7.
Wagner SL, Shubair MM, Michalos AC. Surveying older adults’ opinions on housing: Recommendations for policy. Social indicators research. 2010 Dec;99:405-12.
Nagib W, Williams A. Toward an autism-friendly home environment. Housing Studies. 201;32(2):140-167.
Shorten A, Smith J. Mixed methods research: expanding the evidence base. Evidence-based nursing. 2017 Jun 14.
Davids RC, Groen Y, Berg IJ, Tucha O, van Balkom ID. Local-global processing approaches in older autistic adults: A matched control study using RCFT and WAIS-IV. Research in Autism Spectrum Disorders. 2020 Oct 1;78:101655.
Tse VW, Crabtree J, Islam S, Stott J. Comparing intellectual and memory abilities of older autistic adults with typically developing older adults using WAIS-IV and WMS-IV. Journal of Autism and Developmental Disorders. 2019 Oct;49:4123-33.
Langsrud, Ø Fisher’s exact test. http://www.langsrud.com/fisher.htm, accessed 13.7.2023.
ScoSciStatistics. Chi-square test calculator. https://www.socscistatistics.com/tests/chisquare2/default2.aspx, 2023. accessed 3.5.2023.
Preacher, K. J. Calculation for the chi-square test: An interactive calculation tool for chi-square tests of goodness of fit and independence [Computer software]. http://www.quantpsy.org/chisq/chisq.htm. 2001, April. accessed 13.7.2023.
Joffe H, Yardley L. Chapter AO. Content and thematic analysis. Research Methods for Clinical and Health Psychology. 4;56. Marks D, Yardley L (ed): Sage Publications, London.
Rolls L, Seymour JE, Froggatt KA, Hanratty B. Older people living alone at the end of life in the UK: research and policy challenges. Palliative Medicine. 2011 Sep;25(6):650-7.

Zeng Y, Gu D, George LK. Association of religious participation with mortality among Chinese old adults. Research on aging. 2011 Jan;33(1):51-83.
Livingston LA, Shah P, Happé F. Compensatory strategies below the behavioural surface in autism: A qualitative study. The Lancet Psychiatry. 2019 Sep 1;6(9):766-77.
Walsh RJ, Krabbendam L, Dewinter J, Begeer S. Brief report: Gender identity differences in autistic adults: Associations with perceptual and socio-cognitive profiles. Journal of Autism and Developmental Disorders. 2018 Dec;48:4070-8.
MacKenzie, A. Prejudicial stereotypes and testimonial injustice: Autism, sexuality and sex education. International Journal of Educational Research. 2018; 89, 110-118.
Landeiro F, Barrows P, Musson EN, Gray AM, Leal J. Reducing social isolation and loneliness in older people: a systematic review protocol. BMJ open. 2017;7(5).
Anon. Rural population and migration. 2021. https://www.gov.uk/government/statistics/rural-population-and-migration/rural-population-and-migration. Accessed 02.07.2023.
Berry EH. Demographic ageing and rural population change. In Rural Gerontology 2020 Dec 28 (pp. 17-28). Routledge.
Office for National Statistics (ONS) released 3 April 2023, ONS website, article, Profile of the older population living in England and Wales in 2021 and changes since 2011. accessed 14.8.23
Landon J, Shepherd D, Lodhia V. A qualitative study of noise sensitivity in adults with autism spectrum disorder. Research in Autism Spectrum Disorders. 2016 Dec 1;32:43-52.
Bailey N. Poverty and the re-growth of private renting in the UK, 1994-2018. PloS one. 2020 Feb 5;15(2):e0228273.
Dalistan R, George S, Laver K. Considering the home environment and planning for the future: A qualitative exploration of the views of older adults and individuals with older relatives. Scandinavian Journal of Occupational Therapy. 2023 Mar 25:1-3.
Sheppard CL, Pattni N, Gunasegaran T, Austen A, Hitzig SL. Housing Satisfaction Among Older Adults Living in Low-Income Seniors’ Housing. Journal of Gerontological Social Work. 2023 Jan 2;66(1):134-51.
Walker A. The relationship between the family and the state in the care of older people. Canadian Journal on Aging/La Revue canadienne du vieillissement. 1991;10(2):94-112.
Maestas N, Zissimopoulos J. How longer work lives ease the crunch of population aging. Journal of Economic Perspectives. 2010 Feb 1;24(1):139-60.
Zissimopoulos JM, Karoly LA. Labor-force dynamics at older ages: Movements into self-employment for workers and nonworkers. Research on Aging. 2009 Jan;31(1):89-111.
Mazumdar B, Warren AM, Brown TC. Bridge employment: Understanding the expectations and experiences of bridge employees. Human Resource Management Journal. 2021 Apr;31(2):575-91.
Cahill KE, Giandrea MD, Quinn JF. Retirement patterns and the macroeconomy, 1992–2010: The prevalence and determinants of bridge jobs, phased retirement, and reentry among three recent cohorts of older Americans. The Gerontologist. 2015 Jun 1;55(3):384-403.
Di Gessa G, Corna L, Price D, Glaser K. The decision to work after state pension age and how it affects quality of life: evidence from a 6-year English panel study. Age and Ageing. 2018 May 1;47(3):450-7.
Lewis K, Walker EA. Self-employment: Policy panacea for an ageing population?. Small Enterprise Research. 2011 Dec 1;18(2):143-51.
Zissimopoulos JM, Karoly LA. Transitions to self-employment at older ages: The role of wealth, health, health insurance and other factors. Labour economics. 2007 Apr 1;14(2):269-95.
Curran J, Blackburn RA. Older people and the enterprise society: Age and self-employment propensities. Work, employment and society. 2001 Dec;15(4):889-902.
Cressy, R. and Storey, D. (1995). New Firms and their Banks. Centre of Small and Medium Sized Enterprise, Warwick University Business School and Nat West Bank, Warwick and London.
Johnson S, Cooper C, Cartwright S, Donald I, Taylor P, Millet C. The experience of work‐related stress across occupations. Journal of managerial psychology. 2005 Mar 1;20(2):178-87.
Ćwirlej-Sozańska A, Widelak M, Wiernasz M, Wawrzykowska I, Turkosz N. An assessment of the work ability, disability and quality of life of working people of pre-retirement and retirement age in Poland–a cross-sectional pilot study. International Journal of Occupational Medicine and Environmental Health. 2021;34(1):69-85.
Armstrong HL. International classification of diseases, eleventh revision (ICD-11).

[1] One 60+ participant provided information about their age, but not their autistic status. After discussion the team decided to include this participant, as the information sheet indicates that the survey was aimed at (older) autistic people, they provided consent, their responses were consistent with being autistic and they did not state they are not autistic.