The Associations Between Clinical, Social, Financial Factors and Unmet Needs of Autistic Adults: Results from an Observational Study

Abstract

Background:

Autism is a developmental disability that affects how individuals experience the world. Each Autistic individual experiences Autism in their own way, meaning that the level and type of assistance in their everyday lives vary widely. A shortage of programs and services tailored to Autistic adults exists worldwide, and the current gap between needs and services is likely to worsen as the growing number of Autistic children being diagnosed reach adulthood. This research sought to determine priorities in terms of health and social service needs of Autistic adults and to examine factors influencing whether or not these services were being received.

Methods:

Through a multistakeholder approach, researchers and Autistic adults codeveloped the Maritime Provinces Needs Assessment Survey and collected data from August 2017 to February 2018. The research team engaged Autistic adult partners, including one that was project colead, as full partners. We recruited survey respondents from three Maritime Provinces using mainly social media and local Autism networks. We used Poisson regression analyses to identify factors most strongly associated with the number of unmet needs reported by Autistic adults.

Results:

In total, 260 respondents completed the needs assessment survey: 77 self-reporting Autistic adults (aged 19–55 years), 87 Autistic adults (aged 18–63 years) whose information was provided by a proxy respondent, and 96 professionals working in the field of Autism. Autistic adults reported a mean of 2.1 ± 1.5 (self-reported) and 2.8 ± 2.1 (proxy-reported) services wanted but not received. The number of mental health and neurodevelopmental conditions, unemployment, and perception that government support is insufficient were positively related to unmet needs.

Conclusions:

Overall, the results of this study highlight considerable gaps among the support needs of Autistic adults. The identification of factors associated with a higher number of unmet needs helps identify potential subgroups requiring more attention.

Lay summary

Why was this study done?

There is currently a shortage of programs and services for Autistic adults. Little is known about which services are most important to Autistic adults and which factors are associated with not receiving them.

What was the purpose of this study?

Our goals were (1) to determine the service need priorities of Autistic adults and (2) to examine factors associated with unmet needs.

What did the researchers do?

The researchers invited health professionals, service providers, policy makers, Autism advocates, as well as Autistic adults and caregivers of Autistic adults to participate as part of the project team. In the early phases of the project, we named an Autistic adult colead of the project. In doing so, we acknowledged that conducting relevant research on the Autistic adult community implies not only hearing their voices and concerns, but also providing an opportunity for equal say on the research itself.

We adapted previously used surveys to align them with our objectives of focusing only on Autistic adults. We launched the survey online and promoted it using various media and community channels, many of them direct suggestions from Autistic adult team members. We made paper copies available and provided a phone number for those who needed support filling out the survey.

What were the results of the study?

More than 25% of Autistic adults identified 12 different services as priority needs.

Autistic adults who (1) reported more mental health and neurodevelopmental conditions, (2) perceived their mental health as poor, or (3) felt that government support was insufficient were at greater risk of having unmet needs.

What do these findings add to what was already known?

Similar to previous studies, we found that the service needs of Autistic adults are varied, and that many are not receiving the services they consider a priority. Our study also went one step further by identifying factors that are associated with a higher number of unmet needs.

What are potential weaknesses in the study?

Our sample does not necessarily reflect the Autistic adult community as a whole, since we were unable to guarantee that hard-to-reach segments had access to our survey. In addition, we are unable to know how our results may change over time, as our survey was filled out on one occasion only. Finally, we did not measure Autistic traits and therefore are unable to estimate how different types of traits are associated with certain needs.

How will these findings help autistic adults now or in the future?

Our results may help bring attention to subgroups of Autistic adults who need more help receiving the services they require.

Introduction

Autism is a developmental disability that affects how individuals experience the world. Each Autistic individual experiences Autism in their own way, although they most often differ from non-Autistic individuals on themes such as the way they think, process their senses, move, communicate, and socialize.¹ Interacting in a world built for non-Autistic individuals means that some may require more help than others in their day-to-day lives. While many programs and supports have been developed for Autistic children at the preschool and school age, access to Autism-specific services decreases as Autistic individuals become adults.^2,3 Autistic adults present a wide range of medical, psychological, social, and community needs.⁴ Various supports and services are necessary to assist with various aspects pertaining to quality of life, such as behavior issues,^5,6 social skills,⁷ mental health,^8–10 and vocation.¹¹

In response to the desire expressed by the Autism community that research focus more concretely on improving the daily lives of those on the Autism spectrum,¹² a growing number of needs assessment surveys have taken place.^13–16 Some Canadian provinces, including Nova Scotia,¹³ Ontario,¹⁴ and Newfoundland and Labrador,¹⁵ have produced individual snapshots of the life span needs of Autistic individuals, with the Canadian Autism Spectrum Disorders Alliance (CASDA) producing the most comprehensive of such reports at the national level.¹⁶ These reports document a high overall use of services, but also a large range in service use among Autistic individuals with some receiving none or few services. All age groups indicated that the amount of services received was insufficient, with adults attributing the greatest weight to employment services, social skills programs, life-skills training, postsecondary education, housing options, and mental health treatment.¹⁶ The national report acknowledges gaps, which this study seeks to address, including the small number of Autistic adults who self-reported and of Adult respondents from the Maritime provinces.¹⁶ Other studies have begun to look at factors that impact whether or not services are being received by Autistic adults.^3,17–19 Age was found to be negatively associated with receiving services deemed a priority,³ depression in adolescents and young adults was associated with greater reception of transition services,¹⁷ women and lower income were associated with a higher likelihood of having unmet sensory/motor services in young adults,¹⁸ and African Americans and young adults from lower income households had a higher likelihood of not receiving any services at all.¹⁹ Considering these limited studies, our study sought to (1) identify priority service needs for Autistic adults, and (2) determine which factors are associated with unmet needs in Autistic adults living in the Maritime Provinces.

Methods

Study design and survey

We developed the Maritime Provinces Needs Assessment Survey to study the needs of Autistic adults living in the three Maritime Provinces: New Brunswick, Nova Scotia, and Prince Edward Island. We framed this survey based on the CASDA National Autism Needs Assessment Survey¹⁶ and the Ontario Adults Needs Assessment,¹⁴ with permission from the authors. We made modifications to the surveys mentioned above to better represent the needs and experiences of Autistic adults and their Maritime-specific reality. We accomplished this through an iterative consultation process involving Autistic adults, Autism advocates, professionals working with Autistic adults, and researchers. The iterative process included an in-person roundtable discussion with a question-by-question breakdown of the survey, followed by an opportunity to review and submit feedback individually. To identify needs from another perspective, we also used data obtained from a separate and distinctly different questionnaire administered to professionals working with Autistic adults. Although the professionals' questionnaire was developed for project objectives that fall outside the scope of this analysis, items pertaining to the identification of the top 5 priority service needs for Autistic adults were the same. For the purposes of this study, we only used the portion of questionnaires relevant to the objectives.

Using an observational design, participant recruitment took place from August 2017 to February 2018. Our recruitment strategy for Autistic adults and professionals working with Autistic adults relied on the experience and knowledge of team members who were imbedded within the Autistic community. Autistic adults on the team helped create an animated promotional video for the project and shared it via social media platforms, gave interviews to traditional media outlets, reached out to others within their networks and communities, and promoted the project during an Autistic adults Summit held in New Brunswick in 2017. Organizations from across the three Maritime Provinces, such as Autism resources centers, also shared the promotional video and links to the online survey (available in English and French) through their networks. We made paper copies available and provided a phone number for those who needed support in filling out the survey.

Ethics approval

We obtained informed consent before filling out the survey as approved by the Université de Moncton's Research Ethics Board, file number 1617-068.

Sample inclusion criteria

As indicated by our Autistic adult partners and a recent study on participatory research with Autistic adults,²⁰ inclusion of participants was based solely on formal diagnosis risks leaving out Autistic adults unable to access or afford private psychiatric/psychological assessments. Therefore, participants reported having received one of the following: (1) a formal Autism diagnosis (Autism, Asperger syndrome, pervasive developmental disorder-not otherwise specified [PDD-NOS], PDD, childhood disintegrative disorder, or Autism spectrum disorder) by a licensed professional, (2) an informal diagnosis suggested by a professional, or (3) a self-diagnosis. We invited caregivers and/or family members to answer the survey only if an Autistic adult could not complete it. In this case, respondents received the instruction that answers to all following questions must relate to the Autistic adult, not to the caregiver (proxy respondent). As such, each questionnaire should represent one known individual and assessment of their specific needs.¹⁴ Duplicates, nine in total, were removed based on respondent first name or initials, postal code, province of residence, date of birth, and IP address. We excluded respondents if they lived outside of New Brunswick, Nova Scotia, or Prince Edward Island over the previous 12 months, or if they were younger than 18 years. We also excluded professionals working with Autistic adults if they lived outside those provinces over the previous 12 months, or if they had no experience working with Autistic adults.

Autistic adult engagement

The study, which went under the name CONNECT (CONtiNuity of carE and support for autistiC adulTs), was conducted at all stages through a collaborative, multistakeholder approach. Autistic adults recruited to the research team played the role of patient partners in adherence to the values and principles set forth by Canada's national Strategy for Patient-Oriented Research (SPOR).²¹ While acknowledging that the term patient may not be fitting for all contexts, the CIHR defined it to be inclusive and to encompass individuals with personal experience of a health issue, which qualifies them as essential in identifying research topics and questions most relevant to the needs of the community they represent.²¹ As a result, the research team relied heavily on, and learned extensively from, the perspectives and lived experiences of Autistic adults. The roles and responsibilities of the Autistic adult partners were varied and included involvement in designing the study, recruitment of survey participants, promotion of the project, interpretation of data, development/revision of knowledge translation material, as well as presented posters and discussed the patient engagement dimensions of the project at regional,²² national,²³ and international Autism conferences.^24,25

Data analyses

Respondents indicated current services wanted but not received in the last 12 months as well as their current top 5 priority service needs. The first question read “In the last twelve months, did you receive or want to receive a specific service? Check all that apply” and the response options read “Received; Not wanted; Didn't know about this service; Wanted but not received.” The second question read “What are the five services you need most (whether they are available or not)?” Professionals working with Autistic adults indicated top 5 priority service needs for Autistic adults using the question “According to your experience working with autistic adults, what are the five services they need more (whether they are available or not)?” We used the same list of 23 services, which include support and social services, health services, and professional and personal development services, for all 3 questions. Similar to the approach used by Lai and Weiss,³ we calculated the main dependant variable at the individual level, for self-reporting and proxy respondents only, as a proportion between the number of services listed as wanted but not received among those identified as their top 5 service needs. In addition, when respondents indicated that a service was wanted but not received, they were prompted to answer the question “why not received?” by selecting one of four response options: (1) not affordable, (2) not available, (3) did not meet criteria, (4) other.

Sociodemographic measures included the gender of the Autistic adult, ethnicity, marital status, housing, employment, and annual income. We measured the overall physical health status by asking individuals to rate their current perceived health on a 5-point Likert scale, dichotomized as poor (poor, not very good, or fair health) or good (good or very good), which has been shown to be a valid indicator of morbidity across various populations.²⁶ Overall dental and mental health status was likewise dichotomized and measured using the same response options. The number of physical health concerns was the sum of conditions identified from a list of 14 physical health conditions. We calculated the total number of mental health and neurodevelopmental conditions by summing those identified from a different checklist, including 18 different conditions. Intellectual disability was not included in this total, and was instead reported separately since it is linked to another stream of access to government support and services.²⁷ We derived government support received from the question “Did you receive support from the government (federal, provincial, and/or municipal) to assist with accessing services in the last 12 months?” and government support sufficient from the question “Overall, were the supports from the different levels of government sufficient to access all services needed in the last 12 months?”

For the first objective, to identify priority service needs, we examined the proportion of services most often included in the top 5 most needed services for Autistic adults. Thereafter, high-frequency priority needs were defined as services endorsed by over 50% of the sample as a top 5 service need, and moderate-frequency priority needs were those endorsed by 25%–49%.³ For the second objective, to determine which factors are associated with unmet needs in Autistic adults, we used Spearman's p correlations to identify variables most strongly associated with the number of unmet needs. We then entered variables that had a correlation ≥0.25 and an associated p < 0.05 in a subsequent regression model. We used Poisson regression to identify correlates of number of unmet needs. Results from the regression model are presented as adjusted rate ratios (RRs) and 95% confidence intervals (CIs). We adjusted regression models for intellectual disability and diagnosis type to account for the possibility that these variables affect access to services. Also, for the second objective, we used frequency distributions to examine reasons cited for wanting but not receiving services.

Results

Sample characteristics

In total, 260 respondents from the 3 Maritime Provinces completed the needs assessment survey. Respondents included 77 self-reporting Autistic adults (aged 19–55 years), 87 proxy respondents reporting for Autistic adults (aged 18–63 years), and 96 professionals working with Autistic adults (aged 18–64 years). Demographic, clinical, and systemic characteristics for both self-reported and proxy-reported Autistic adults are found in Table 1. As for professionals, they reported working with a median of 6 (range: 1–100) Autistic adults in the past year and having served an Autistic adult clientele for a median of 5 years (range: 1–34). Most professionals worked for a nonprofit organization (62.5%) or the public sector (20.8%), while a minority were self-employed (8.3%) or worked in the private sector (4.2%).

Table 1.

Characteristics of Study Participants

	Self-reported (n = 77), n (%) or mean ± SD	Proxy-reported (n = 87), n (%) or mean ± SD
Demographic variables
Age	33 ± 9	26 ± 9
Gender (male)	31 (41.3)	70 (82.3)
Ethnicity (Caucasian)	72 (94.7)	78 (89.7)
Marital status (married or common-law)	18 (24.3)	1 (1.2)
Employment status (full, part, or self-employed)	50 (64.9)	26 (29.9)
Annual income	18,750 ± 19,563	9448 ± 6931
Housing (lives with family)	41 (53.3)	63 (72.4)
Urban	46 (63.0)	54 (68.4)
Clinical variables
Formal diagnosisa	62 (83.8)	80 (91.9)
Informal diagnosisb	8 (10.4)	5 (5.8)
Self-diagnosis	2 (2.6)	0 (0)
Intellectual disability	3 (3.9)	31 (35.6)
Physical health statusc	46 (61.3)	70 (81.4)
Dental health statusc	40 (54.8)	60 (71.4)
Mental health statusc	29 (38.2)	40 (47.1)
No. of physical health conditions	1.5 ± 1.1	1.2 ± 1.0
No. of mental health and neurodevelopmental conditions	2.4 ± 1.9	2.0 ± 1.64
Systemic variables
Government support receivedd	17 (34.7)	44 (57.9)
Government support sufficiente	15 (36.6)	9 (16.1)
No. of services received	4.3 ± 3.7	5.3 ± 4.6
No. of services wanted but not received	2.1 ± 1.5	2.8 ± 2.1
One unmet needf	21 (27.3)	23 (26.4)
Two unmet needsf	17 (22.1)	12 (13.8)
Three or more unmet needsf	11 (14.3)	19 (21.8)
Barriers to health care (yes, at least one barrier encountered)	32 (52.5)	37 (50.0)

^a Autism, Asperger syndrome, PDD-NOS, PDD, childhood disintegrative disorder, or Autism spectrum disorder diagnosed by a licensed professional.

^b Diagnosis suggested by a professional.

^c Health status is perceived as good to very good.

^d Any support received from any level of government to access services over the last 12 months.

^e Supports from government sufficient to access all services needed over the last 12 months.

^f Services that were wanted but not received from among individually identified top 5 priority services need.

PDD-NOS, pervasive developmental disorder-not otherwise specified; SD, standard deviation.

Services needed

Only two services were considered high-frequency priority needs, defined as those endorsed by over 50%, among the proxy-reported Autistic adults' and professionals' group, and none among the self-reported Autistic adults' groups. They included housing/residential options and life skills training services. Some services were reported among the top 5 most needed services more frequently than others. These included 10 services that were endorsed by over 25% of either the self-reported or proxy-reported Autistic adult sample, meeting the moderate-frequency priority definition (Fig. 1). There was agreement from over 25% of professionals on 9 out of the 12 services, with only dental care, behavior therapy, and postsecondary education program not meeting the threshold for moderate priority among professionals. As for convergence among all three groups, only the provincial disability program, employment counseling, and social skills training simultaneously met the moderate-frequency priority definition.

FIG. 1.

Distribution of services selected among the top 5 needs of Autistic adults.

Unmet needs

In univariate analyses, the following variables emerged as being correlated with unmet needs with a correlation ≥0.25 and an associated p < 0.05 in the self-reported group: housing, number of physical health conditions, number of mental health and neurodevelopmental conditions, annual income, gender, marital status, sufficient government support, barriers to health care, perceived mental health, perceived dental health, and perceived physical health. In the proxy-reported group, the following variables were correlated with unmet needs: employment, number of mental health and neurodevelopmental conditions, sufficient government support, and barriers to health care. When included in the regression models, we found that the likelihood of having unmet needs was greater for the proxy-reported Autistic adults who reported not receiving sufficient government support (RR = 9.6, CI: 1.26–73.41), who were unemployed (RR = 5.2, CI: 1.58–17.21), and for each increase in the number of mental health and neurodevelopmental conditions (RR = 1.14, CI: 1.02–1.27). None of the variables included in the self-reported model was significantly associated with having unmet needs. In addition, for 9 of the 12 priority services identified as part of the first objective, the most common reason cited for wanting but not receiving services was perceived unavailability (Fig. 2).

FIG. 2.

Reasons given for service wanted but not received by Autistic adults.

Discussion

Despite being a lifelong condition, existing supports and services for Autistic individuals are generally more accessible and tailored to the needs of children.^3,28 The results of this study demonstrate that Autistic adults perceive they are not receiving many of the varied services they desire and that several factors are linked with a higher likelihood of having unmet needs.

Twelve different services were identified as top 5 service needs by more than a quarter of Autistic adults. These findings are similar to those of Lai and Weiss,³ who found that Autistic adults report a broad set of priority service needs. Other studies examining service usage among Autistic adults have reported moderate usage across most of the different types of services queried.^19,29 These findings point to the heterogeneous nature of Autism and the need for coordination across multiple sectors. There was convergence among professionals and self-reported and proxy-reported Autistic adults on moderate-level importance of three services, including the provincial disability program, employment counseling, and social skills training. Employment and social skills training are reported among the most needed services in several other studies and reports, both in Canada^16,30 and elsewhere.²⁹ Conversely, the proxy-reported group did not as highly endorse mental health services, which is a departure from much of the literature. Mental health services have routinely been reported among the most needed and most often unmet services for Autistic adults.^3,18,31,32 Given the frequency of co-occurring mental health and neurodevelopmental conditions in Autistic adults,^33–35 it may be an underestimation of the need for mental health services on the part of proxy-respondents.

Autistic adults in our study had a high number of service priorities that went unsatisfied. While other studies also demonstrate that few individuals receive all the priority services they need,^3,29 our results show that many Autistic adults are unable to receive the services that they want and deem to be important. Our results suggest that the perception of lack of government support is most strongly linked with unmet needs. While there has been increasing recognition on the part of governments to better support Autistic adults,^36,37 policies and programs have generally focused on children and youth.^38–40 Gaps remain such as in career planning and adult support services, which have contributed to poor outcomes in adulthood.^41–43 Our results expand upon these observations by showing that insufficient government support to help access needed services puts Autistic adults at greater risk of having unmet needs. The decentralized nature of Autism-related services means that greater efforts in assisting Autistic adults to find and obtain services are critical.

Unemployment was also associated with greater vulnerability to having unmet needs. This is particularly distressing, as employment rates for Autistic adults are generally lower than those of the general population.^44,45 Only 46% of Autistic adults in our sample had some form of employment, with less than one-fifth holding full-time employment. These rates are in line with other studies documenting employment rates that range from 6% to 54% among Autistic adults.^46–49 The challenges in obtaining and maintaining employment are many and varied. They include, among others, understanding complex job application materials, quick thinking during the interview process, adapting to new routines and procedures, responding to unexpected situations, planning and managing multiple tasks, communicating effectively with coworkers, social interactions, and managing sensory-sensitive environments.^50–53 However, with proper support, Autistic adults are more likely to experience positive employment outcomes,¹¹ which in turn can lead to better quality of life and increased economic stability.²⁹

Mental health was another important factor, as the number of co-occurring conditions was associated with unmet needs. Autistic individuals are at greater risk of co-occurring mental health conditions compared with their neurotypically developing counterparts.^31,54 Previous studies have shown that between 40% and 70% of Autistic individuals live with at least one co-occurring mental health condition.^33–35 In our sample, 80% of Autistic adults reported at least one co-occurring mental health or neurodevelopmental condition. Although it is now widely recognized in the medical community that mental health conditions are more common in Autistic individuals,^34,55 overlapping traits and non-Autistic-specific diagnostic tools pose a barrier to receiving appropriate service and treatment.⁵⁶ Moreover, challenges in accessing or engaging with services may stem from stressors embedded in the health care setting or past negative interactions with the health care system.^31,57,58 While mental health services were identified as an important priority need, our results suggest that when considered among multiple factors, mental health is linked with individualized unmet needs. The presence of a mental health condition may trigger mental health support services, but not the larger coordination of other priority services.^3,59 Unmet needs in Autistic adults with co-occurring mental health conditions can have a negative impact on personal well-being,⁶⁰ and can increase the burden on both individuals and caregivers.⁶¹

Limitations

As most of the Autism literature relies on proxy respondents, a major strength of this study was including the first-person perspective of Autistic adults, when possible, and professionals working with Autistic adults. Our study did, however, have limitations. First, our sample is based on convenience sampling, and data collection used primarily online methods. While the internet has proven an important communication tool for Autistic adults,⁶² some may not have had access to it. Second, engagement of Autistic adults and Autism advocates as members of the research team helped reach out to the Autistic population in general, but could have biased the mean respondent profile toward adults with similar social behavior and interests. For methodology and practical reasons, we could not reach out to all Autistic adults, because they have no internet access, or no interaction with the local Autism Resource Centers, or because they were institutionalized in medical, community, or penal settings. Therefore, we cannot guarantee the representativity of service users in the geographic areas surveyed and the generalizability of our results. Third, the observational design does not allow assessing the predictive nature of factors, which may change over time. Fourth, we did not measure Autistic traits and therefore are unable to estimate how different traits are associated with certain needs. Finally, we collected the data before the COVID-19 pandemic and our results reflect that reality. Our findings may understate the unmet service needs of Autistic adults in the post-COVID-19 landscape.

Implications

This study sought to address the needs and concerns of the Autistic adult community by conducting research in line with their priorities. The results highlight that Autistic adults lack the supports they desire, and that mental health and neurodevelopmental conditions, self-perception of poor mental health, and the perception that government support is insufficient are associated with a greater risk of having unmet needs. Furthermore, our findings highlight the importance for policy makers to consider the wide array of service needs and to provide more support in navigating and obtaining available services. Offering more government support to assist with accessing services is an achievable change and could meaningfully improve the outlook for Autistic adults seeking to live full and independent lives. Further research is needed to examine the impact of Autistic traits, to reach the underrepresented segments of the Autistic adult community, and to assess the long-term associations affecting reception of priority service needs.

Authorship Confirmation Statement

None of the authors has conflicts of interest and all authors meet the standard authorship criteria. All authors participated in the conception of the objectives of the study. J.G. analyzed the data. J.G., C.J., P.G.-Z., L.T., D.P., and A.B. contributed to the interpretation of the data. J.G. and C.J. wrote the article, while M.R., M.B., L.T., and A.B. provided support. All the authors have reviewed and approved the article before submission. The material in this article has not been published in whole or in part elsewhere, and this article is not currently being considered for publication elsewhere.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

The CONNECT project received funding from the Maritime SPOR SUPPORT Unit (MSSU) and the Consortium national de formation en santé (CNFS).