Abstract
Recognizing higher rates of co-occurring health conditions in autistic adults and the frequent use of hospital-based health care services, this study compared rates of repeat emergency visits, repeat hospitalizations, and delayed discharges in autistic adults to other adults with and without developmental disabilities matched for age and sex, living in Ontario, Canada. Returning to the hospital emergency department within a month, being readmitted to hospital within a month and experiencing a delayed discharge from hospital were each more likely to occur in autistic males and females than their counterparts without developmental disabilities, with the risk ratios being the highest for delayed discharges. Males and females with other developmental disabilities were more likely to return to the emergency department within a month than their autistic counterparts, and males with other developmental disabilities were more likely than autistic males to be readmitted to hospital, but the likelihood of delayed discharge in the two groups was similar. These findings suggest that more needs to be done to both improve hospital-based experiences of autistic adults and adults with other developmental disabilities, and to strengthen community-based care to reduce the likelihood of repeat and extended stays in hospital.
Lay abstract
We know that autistic people have more health problems and are more likely to go to the emergency department and get hospitalized than other people, but we know less about the problems they have once they get to the hospital. In this study, we looked at all autistic adults in Ontario and compared them to adults who were not autistic and to adults who had other kinds of developmental disabilities to see who came back to the emergency department in the month after an emergency department visit, who got re-hospitalized in the month after being sent home from hospital, and who stayed in the hospital longer than they needed to because there was no place appropriate for them to go to. We found that both autistic males and females were more likely to have these things happen to them than the same age- and sex-matched adults who did not have developmental disabilities. We also found that adults with other kinds of developmental disabilities had similar problems to autistic people. This makes us think that we need to work harder to improve health care for autistic adults and adults with other developmental disabilities when they come to hospital. We also need to make community services work better, and work more closely with hospital services, so that people only come to hospital when they need to and that they can go home when they are ready.
Autistic adults, relative to non-autistic adults, have higher rates of medical disorders, both acute and chronic (Cashin et al., 2018; Croen et al., 2015; Forde et al., 2022), physical or sensory disabilities (Rydzewska et al., 2018), psychiatric disorders (Lai et al., 2019), and poor general health (Rydzewska et al., 2019). Autistic adults also use health services more frequently than their non-autistic peers (Weiss et al., 2018; Zerbo et al., 2019) and various qualitative studies have highlighted the stress they experience in doing so (Doherty et al., 2022; McLean et al., 2024). There are a number of reasons for this, including a shortage of more specialized and autism-informed health services, a lack of training of healthcare providers, and ableism more generally in healthcare (Mazurek et al., 2023; McLean et al., 2024).
A recent population-based study (Tint et al., 2023) investigated health conditions and healthcare use for autistic adults relative to age- and sex-matched adults with and without developmental disabilities (DDs). The study found that the prevalence of health conditions and healthcare use was higher for autistic adults than age-matched individuals without any DD. It also found that psychiatric diagnoses and psychiatric services were more frequently used by autistic adults than age-matched adults with other DD.
Given the high rates of health issues present in many autistic adults, it is not surprising and may in fact be appropriate that this group uses healthcare more frequently than non-autistic adults. However, there are other reasons why acute care is relied upon which go beyond the actual health conditions, including inaccessible community-based health services, and gaps in social care, housing, employment, and family supports. In the general population, certain patterns of health service use have been identified as potentially avoidable and are sometimes monitored as indicators of the effectiveness of healthcare delivery. These include having repeated emergency visits and hospitalizations within a short timeframe, and being deemed ready for hospital discharge but remaining in hospital because of a lack of appropriate supports in the community. While it has been well documented that emergency department (ED) use is more common in autistic adults compared with other adults (Lunsky et al., 2015; Vohra et al., 2016; Weiss et al., 2018), how likely autistic adults are to return within a month of a prior ED visit, a commonly used effectiveness indicator of healthcare, is not known. This question is important to answer, as disproportionate returns to the ED may indicate either that the underlying issue was not addressed at the first visit or that the appropriate follow-up in community did not occur. In the general population, repeat visits are costly and stressful, but to an autistic person, the physical, social, and sensory environment of the ED can be extra traumatic. In a similar vein, we know that hospitalizations are also more likely for autistic adults (Tint et al., 2023; Zerbo et al., 2019), but the likelihood of being hospitalized repeatedly within a short timeframe in autistic adults, which can be both costly and traumatizing, has not been studied. Finally, there has been significant public attention paid to the phenomenon of some autistic youth and adults being hospitalized for unnecessarily long periods, not simply because of the severity of their health concern but because of the gaps in community-based supports (Ince et al., 2022). This phenomenon of delayed discharges accounts for high healthcare costs generally (Bai et al., 2019) and in those with DD (Lunsky et al., 2019), and is a target of healthcare improvement, but has yet to be studied at the population level for autistic adults specifically.
The purpose of this study is to examine the likelihood of repeated and prolonged hospital use in the cohort of autistic males and females in the Tint et al.’s (2023) study, relative to their same-sex (assigned at birth) non-autistic counterparts, matched by age. To address whether issues encountered by autistic adults in hospital are unique to this group, a second set of comparisons was made to age-matched males and females with other DD.
Methods
This population-based cohort study is part of a larger project involving all Ontario adults with DD, including autism, aged 19–65 years as of 1 April 2010 and a random sample of 20% of Ontario adults without these disabilities, followed from 2010 until 2016, as part of the Health Care Access Research and Developmental Disabilities (www.hcardd.ca) program. This age group was studied because of their eligibility for government-funded adult-specific disability support services at the time of cohort creation. Diagnostic information was obtained by linking administrative data from provincial health and social services at ICES, an independent, non-profit research institute whose legal status under Ontario’s health information privacy law allows it to collect and analyze healthcare and demographic data, without consent, for health system evaluation and improvement. Individual-level data were linked using a unique encoded identifier and analyzed at ICES. Individuals were considered to have DD on the basis of (1) the presence of diagnostic codes in two or more physician claims in the Ontario Health Insurance Plan (OHIP) since database inception in 1991; (2) the presence of diagnostic codes on one or more hospital claims through the Canadian Institute for Health Information’s (CIHI) Discharge Abstract Database (DAD), the Ontario Mental Health Reporting System (OMHRS), or through National Ambulatory Care Reporting System (NACRS) since database inception in 1988; or (3) DD recorded as a diagnosis leading to eligibility for disability benefits from social services in 2008. DD codes included intellectual disability, autism, fetal alcohol syndrome, and genetic conditions associated with intellectual disability. The algorithm was developed as part of a larger study and is described elsewhere (Lin et al., 2014, 2019).
Individuals from the DD cohort were further categorized as autistic if they had an autism diagnosis in two or more physician claims since database inception in 1991, in one or more hospital claim since database inception in 1988, or autism recorded as a diagnosis leading to eligibility for disability benefits from social services. If they did not meet the autism criteria, they were considered to have DD without autism (see also Tint et al., 2023). Given that the main outpatient administrative dataset (i.e. OHIP) only allowed for recording a single diagnosis, it was not possible to determine which autistic adults had a co-occurring intellectual disability. Autistic adults were matched one-to-three with adults of the same age without DD and one-to-one with adults with other DD.
The use of data for this project was authorized under Section 45 of Ontario’s Personal Health Information Protection Act, which does not require review by a Research Ethics Board.
Outcomes
All outcomes occurred between 1 April 2010 and 31 March 2016 and were binary in nature, capturing whether they occurred at least once within this timeframe.
Repeat ED visits: We used the NACRS to identify ED visits. Repeat visits were defined as a second visit to the ED within 30 days following an ED episode.
Repeat admissions: We used the CIHI DAD and OMHRS to identify hospital admissions and discharges. Repeat admissions were defined as a subsequent hospitalization episode within 30 days following a hospital discharge (not including hospital transfers as part of the initial episode of care).
Delayed discharge: We used CIHI DAD and OMHRS to identify hospital admissions and discharges, and the “alternate level of care” (ALC) designation. ALC designation was recorded in the CIHI DAD at the time of discharge for non–mental health-related hospitalizations and was recorded through the OMHRS for mental health–related hospitalizations.
Analysis
We calculated risk ratios (RRs) and neighborhood income quintile adjusted prevalence and risk ratios (aRRs) with 95% confidence intervals (95% CI) of the three outcomes using log-binomial regression and accounting for matching. All analyses excluded those with missing neighborhood income quintile information (approximately 0.46% of autistic adults, 0.75% of adults with other DD, and 0.40% of adults without DD). Analyses were conducted at ICES using SAS version 9.4.
Community involvement
Members of the autistic and autism communities were not directly involved in this project although general findings from the study were presented to advisories which included autistic people and organizations supporting autistic people at an earlier stage of this project.
Results
Table 1 presents sex-stratified demographic information (age, rurality, and neighborhood income) and the percentage of individuals who experienced each of the three outcomes at least once over the 6-year period studied according to sex and diagnostic group. We had no available data on race or ethnicity. Repeat ED visits occurred for 24.5% of autistic males and 31.9% of autistic females. Repeat hospital admissions occurred for 4.7% of autistic males and 6.7% of autistic females, and delayed discharges occurred for 2.5% of autistic males and 2.8% of autistic females. Lower rates were observed for people without DD, and similar or higher rates were observed for people with other DD.
Participant demographic profiles and health outcomes of males and females (autistic, without DD (No DD), and with other DD).
In the first set of comparisons, each of the outcomes were more likely to occur in autistic males and females than their counterparts without DD, with the aRRs being the highest for delayed discharges (aRR (95% CI) 8.00 (6.19–10.33) in males and 5.58 (4.08–7.63) in females). When considering the likelihood of occurrence of the three outcomes for autistic adults relative to adults with other DD, the pattern was very different; in males and females, delayed discharge rates were similar in autistic people and those with other DD (1.02 (0.83–1.25) in males and (0.81 (0.61–1.06) in females, as were repeat hospitalizations for females (0.93 (0.78–1.11)). Repeat admissions for males with other DD (0.84 (0.73–0.97)), and repeat ED visits for males (0.78 (0.74–0.82)) and females (0.88 (0.82–0.94)) with other DD were more likely when compared to autistic people (Table 2).
aRRs of health outcomes between 2010/2011 and 2015/2016 in autistic females and males compared to those with No DD and those with Other DD.
Discussion
This study found that over a 6-year period, autistic adults relative to an age-matched cohort of non-autistic adults had higher use of possibly unnecessary and costly hospital-based care, including repeat ED visits, repeat hospital admissions, and delayed hospital discharges. These outcomes were not more common, however, when compared with age-matched males and females with other DD. Studying these health outcomes with a comparison group of adults with DD is important, as it reminds us that some of the hospital use patterns (which could be targets for intervention) by autistic adults are not unique to them, but occur in DD more broadly. Thus, intersectoral policies and practices to improve healthcare for autistic adults should be extended to include other related disability groups.
This study shows the greater use of hospital care by autistic people and those with other DD relative to the general population. The literature suggests at least four possible explanations for these healthcare patterns. Greater complexity of healthcare needs due to a range of co-occurring conditions can require more frequent and intensive healthcare. A prior study with these cohorts demonstrated higher rates of medical and psychiatric conditions in autistic adults than their age- and sex-matched counterparts without disabilities (Tint et al., 2023). At the same time, inadequate health supports and social services in the community, and various social determinants of health which occur more frequently and with detrimental impact in people with disabilities can result in greater reliance on hospital-based services (Doherty et al., 2020). Ableism (Campbell et al., 2023; Iezzoni et al., 2021; Weir et al., 2022) limited the training of mainstream healthcare providers (Hemm et al., 2015), and inadequate resources in hospitals for this population (Tuffrey-Wijne et al., 2014) can impact the quality of care received in hospital, which can also contribute to repeated or extended use. Finally, fragmented care processes between hospital and community can lead to communication breakdowns and poor transitions (Gilmore et al., 2022; Shady et al., 2022; Tuffrey-Wijne et al., 2014).
Promising solutions within hospital include enhancing education and training of all staff, implementing alerts with information on appropriate accommodations in each patient’s hospital record, and specialist staff in the hospital who are familiar with autism and other DD so that appropriate healthcare adjustments can be made. In the United Kingdom, for example, mandatory training in autism and intellectual disability is now part of the Health and Care Act (2022), and the standardized training package includes teachers with lived experience. Patient-provider communication tools, such as “hospital passports,” can play an important role, if they are codesigned with the people who would use them within a system that recognizes their importance and provides necessary training (Ellis et al., 2023). The role of the intellectual disability consultation-liaison nurse in hospital has shown some benefits (Bur et al., 2020; Doody et al., 2023) and is a common practice in the United Kingdom. These specialist roles can be valuable across disabilities.
Equally important to improving hospital-based care is investment in community-based intersectoral interventions to divert unnecessary hospital use, along with investments focused on transitions out of these settings. At the community-level, autistic-informed patient-centered medical home models can support primary care delivery and proactive healthcare (Hart et al., 2022). There has been a recent push to promote health checks for all autistic adults, specifically in the United Kingdom (Mason et al., 2022), building off of prior work demonstrating the value of health checks for adults with intellectual disabilities. Importantly, there are several clinical tools to aid with organizing and communicating healthcare issues, codesigned by health service researchers and autistic adults, such as the AASPIRE toolkit in the United States (Nicolaidis et al., 2016) that can help improve healthcare encounters, reducing the need for unnecessarily repeated hospital use. These general resources cannot address the specific challenges autistic people face when transitioning out of hospital, however. Findings here point to the importance of planning for transitions while in hospital and appropriate community supports outside of hospital available to the individual and to their carers. For individuals who cannot find appropriate supports outside of hospital, collaborative efforts are needed and planning must start as soon as possible following hospital admissions (Quinn et al., 2023; Selick et al., 2024).
This study has several limitations including its reliance on diagnostic information recorded in health record for adults prior to 2010. This means some autistic individuals were misclassified as non-autistic, particularly individuals more likely to receive a late diagnosis, which tend to be people without intellectual disabilities, and people assigned female at birth (May & Williams, 2018). The information available in the dataset did not include reasons for hospitalization or quality of care received inside or outside of the hospital. It is important to understand how these healthcare patterns are in fact appropriate for the individual and their health conditions, and the specific challenges during hospital encounters. Current results, stratified by sex, did not provide information by age group, or other relevant characteristics, including individual-level income, race, or ethnicity. Furthermore, due to dataset limitations, we were not able to explore the likelihood of these outcomes in autistic adults separately for those with and without an intellectual disability. It is possible that the lack of differences between the autistic individuals and those with other DD may be because of the proportion of people in both groups with intellectual disabilities. This population-based research could be complemented by in-depth studies of repeat visits and delayed discharges from the perspectives of healthcare providers, patients, and family members. More research is needed to understand these healthcare service use patterns from multiple perspectives, while also exploring and evaluating efforts involving health and social care together to prevent and improve hospital care use for autistic people and those with other types of DD.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is part of the Health care access research and DD program (
). It was funded by the Canadian Institutes of Health Research Partnerships for Health Systems Improvement (grant no. PHE 103973). This study was also supported by ICES, an independent, non-profit research institute funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). Parts of this material are based on data and information complied and provided by MOH and MLTC, the Ontario Ministry of Children, Community and Social Services (MCCSS). This document used data adapted from the Statistics Canada Postal Code Conversion File, which is based on data licensed from Canada Post Corporation, and data adapted from the Ontario Ministry of Health Postal Code Conversion File, which contains data copied under license from ©Canada Post Corporation and Statistics Canada. As a prescribed entity under Ontario’s privacy legislation, ICES is authorized to collect and use health care data for the purposes of health system analysis, evaluation, and decision support. Secure access to these data is governed by policies and procedures that are approved by the Information and Privacy Commissioner of Ontario. The opinions, results, and conclusions are those of the authors and are independent from the funding sources. No endorsement by CIHR, ICES, or the Ontario MOH or MLTC or MCCSS is intended or should be inferred. Parts of this material are based on data and information compiled and provided by CIHI. However, the analyses, conclusions, opinions, and statements expressed in the material are those of the author(s), and not necessarily those of CIHI. These datasets were linked using unique encoded identifiers and analyzed at ICES.