Journals of Gerontology: Psychological Sciences

cite as: J Gerontol B Psychol Sci Soc Sci, 2021, Vol. 76, No. 9, 1738–1744
https://doi.org/10.1093/geronb/gbab136
Advance Access publication July 19, 2021

Research Report

Autism Characteristics and Self-Reported Health in Older

Adulthood

Downloaded from https://academic.oup.com/psychsocgerontology/article/76/9/1738/6323676 by guest on 03 April 2022

Jennifer  Lodi-Smith, PhD,*, Jonathan  D.  Rodgers, PhD, Karl  Kozlowski, PhD,
Sarah Khan, BS, Valeria Marquez Luna, BS, Caleb J. Long, BA, James P. Donnelly, PhD,
Christopher Lopata, PsyD, and Marcus L. Thomeer, PhD
Department of Psychology/Institute for Autism Research, Canisius College, Buffalo, New York, USA.
*Address correspondence to: Jennifer Lodi-Smith, PhD, Department of Psychology/Institute for Autism Research, Canisius College, 2001 Main
Street, Buffalo, NY 14028, USA. E-mail: [email protected]

Received: February 4, 2021; Editorial Decision Date: July 13, 2021

Decision Editor: Derek M. Isaacowitz, PhD, FGSA

Abstract
Objectives:  The present research used a continuous measurement approach to extend the evidence that autism is associated
with significant struggles in physical health as well as mental health and psychological well-being.
Methods:  The relationship of autism characteristics to physical health and psychological well-being was examined in
294 individuals (M age = 70.51, SD age = 8.17, age range = 53–96). The sample is 57.4% female (n = 166) and prima-
rily White (n = 270, 96.8%). The majority of the participants did not identify as having an autism diagnosis (n = 284,
96.6%). Participants completed the Autism-Spectrum Quotient Scale alongside self-report measures of physical health,
mental health, and psychological well-being.
Results:  Autism characteristics correlated strongly with challenges in social engagement due to poor health (r = 0.46), de-
pression (r = 0.39) and anxiety (r = 0.47), limitations due to poor mental health (r = 0.41), satisfaction with life (r = −0.47),
and psychological well-being (r = −0.62).
Discussion:  These findings help shed light on the challenges experienced by individuals aging with elevated autism charac-
teristics. The limitations of this study and prior work on this topic help identify important avenues for future research in
this area.
Keywords:  Autism, Autism-spectrum quotient scale, Autism spectrum disorder, Healthy aging, Mental health, Physical health,
Psychological well-being
  

Evidence suggests that autism is associated with challenges Rubenstein, 2019; Croen et  al., 2015; Hand et  al., 2020;
in physical health (Bishop-Fitzpatrick & Rubenstein, 2019; Rydzewska et al., 2018, 2019). These studies are based on
Croen et  al., 2015; Hand et  al., 2020; Rydzewska et  al., medical records review or census reports and, while they
2018, 2019; Stewart et al., 2020) as well as mental health provide important insights into the challenges of aging with
and psychological well-being (Barneveld et al., 2014; Hand autism, an epidemiological approach to understanding au-
et al., 2020; Rydzewska et al., 2018; Stewart et al., 2020; tism and aging is potentially problematic given the diag-
Wallace et al., 2016) in later life. nostic history of autism where many older adults are part
The majority of work to date uses an epidemiological of a “lost generation” (Lai & Baron-Cohen, 2015) who
approach to understand the association between autism grew up in a society that did not have an understanding of
and health outcomes in later life (Bishop-Fitzpatrick & autism when they were children. Given the lack of validated

© The Author(s) 2021. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. 1738
For permissions, please e-mail: [email protected].
Journals of Gerontology: PSYCHOLOGICAL SCIENCES, 2021, Vol. 76, No. 9 1739

diagnostic tools available for older adults and that existing assessment for both autism and health-related variables be-
gold standards in diagnostics are either not calibrated or yond what is present in the existing literature.
feasible for older adults (Heijnen-Kohl et al., 2017), an epi-
demiological approach likely underestimates the landscape
of autism in later life. Method
Recent research on autism and health recognizes Participants
this limitation by leveraging self-report measures of au-
tism characteristics to operationalize autism rather than The relationship of autism characteristics to physical health
relying on diagnostic reports. In a sample of 66 older and psychological well-being was examined in 294 (M
adults aged 61–88, the 20 participants who self-reported age = 70.51, SD age = 8.17, age range = 53–96) individuals
elevated Broad Autism Phenotype Questionnaire scores who participated in a study of autism and aging funded

Downloaded from https://academic.oup.com/psychsocgerontology/article/76/9/1738/6323676 by guest on 03 April 2022

also reported greater challenges in depression and anx- by NIH Grant #R21 AG059051-01. This sample size was
iety (Wallace et al., 2016). Furthermore, in a large study based on the grant’s goal to recruit 250 adults older than the
of 20,220 midlife and older adults, a group of 276 par- age of 65 to be able to detect small to moderate correlation
ticipants who self-reported elevated autism character- effects of r = 0.20 with a power of 0.90. Additional partici-
istics had higher rates of physical health and mental pants younger than the age of 65 participated in the project
health conditions as well as higher anxiety and depres- as part of age-inclusive recruitment efforts. The sample was
sion (Stewart et al., 2020). The results of these self-report 56.7% female and 97.5% White. Ten (3.4%) participants
studies echo the findings of the epidemiological studies. reported having an autism diagnosis, 24 (8.2%) partici-
However, further echoing the epidemiological studies, pants reported having an autistic first-degree relative, 59
these studies rely on the categorization of both autism (20.1%) participants reported elevated Autism-Spectrum
characteristics as well as health-related measures. Quotient (AQ) scores above the standard 26 threshold
Dichotomization of the continuous variables of autism (Booth et al., 2013; Woodbury-Smith et al., 2005), and 111
characteristics and health-related measures may lead to (37.8%) scored at elevated levels on the measure of au-
known challenges including underestimation of effect sizes tism characteristics described below. In the context of the
and the potential for increased Type I error rates (Cohen, Broader Autism Phenotype literature, this means the ma-
1983; MacCallum et  al., 2002). In addition to the statis- jority of participants did not fall into either the medium
tical costs of dichotomous approaches, categorization (n = 17, 5.8%) or narrow (n = 15, 5.1%) autism phenotype
comes at a cost of external validity by artificially binning range (Wheelwright et al., 2010). Participants were gener-
understanding the natural variability inherent in contin- ally physically healthy with a low disease burden (Table
uous constructs. Thus, a continuous operationalization of 1 and Supplementary Files). Eighty-seven (26.4%) partici-
both autism characteristics and health-related measures pants reported a history of depression.
will allow for potentially stronger estimation of effect sizes
as well as more accurately represent the individual differ-
Procedure
ences being studied and the important variability in these
constructs. Sample participants were recruited via press releases,
Furthermore, a continuous approach to studying autism mailers to community partners, flyers in the community,
characteristics allows for the assessment of two distinct email recruitment via community partners and institu-
aspects of autism characteristics: social challenges and in- tional mailing lists, posting on autism-focused websites
terest in numbers and patterns. This allows for the poten- and discussion forums, paid and unpaid social media posts,
tial to gain a better understanding of what components of mailers with paid addresses, paid advertisements, and
autism may be associated with the health challenges seen in events in the local community. All participants were entered
the literature to date. into a monthly drawing for $100. There were no screening
or exclusion criteria for the study. All data were collected
with the approval of our Institutional Review Board and
met American Psychological Association ethical guidelines
Summary for human subjects research.
This study tests the general hypothesis that the direction All data were recorded and are stored online via
of effects from prior categorical research will conceptually REDCap (Research Electronic Data Capture) electronic
replicate and potentially be more robust with the contin- data capture tools hosted at Canisius College (Harris et al.,
uous assessments used in this sample. Though we refrain 2009). All participants had the option to take surveys via
from making more specific hypotheses given the explora- paper-and-pencil format rather than as an online survey.
tory nature of this research and cannot test causality given These responses were manually entered into REDCap by
the cross-sectional design of this preliminary research, a research assistant. All data were collected between Fall
testing this general hypothesis will extend our knowledge 2018 and Fall 2019 and, therefore, were not affected by
about autism and aging by increasing the granularity of coronavirus disease 2019.
1740 Journals of Gerontology: PSYCHOLOGICAL SCIENCES, 2021, Vol. 76, No. 9

Table 1.  Descriptive Statistics for Autism Characteristics, Physical Health, and Mental Health and Well-Being Measures

Variable N Mean (SD) Minimum Maximum Alpha reliability

Autism characteristics
  Social behavior 330 2.26 (0.52) 1.06 3.84 0.88
 Numbers/patterns 330 2.27 (0.75) 1.00 4.00 0.76
Physical health
  Body mass index 217 29.84 (6.88) 17.07 61.46 NA
 Fatigue 330 2.30 (0.96) 1.00 5.00 0.94
 Pain 330 1.86 (1.01) 1.00 5.00 0.95
  Physical function 330 4.44 (0.85) 1.00 5.00 0.90

Downloaded from https://academic.oup.com/psychsocgerontology/article/76/9/1738/6323676 by guest on 03 April 2022

  Sleep disruption 330 2.45 (0.90) 1.00 5.00 0.87
  Social participation 329 3.89 (0.92) 1.00 5.00 0.92
  General health 329 3.42 (0.92) 1.00 5.00 NA
  Days of poor physical health 282 4.80 (8.14) .00 3.00 NA
  Days of limitations 287 3.46 (7.03) .00 3.00 NA
  Total disease burden 330 1.87 (1.40) .00 7.00 NA
Mental health and well-being
 Depression 330 1.58 (0.72) 1.00 4.50 0.88
 Anxiety 330 1.80 (0.79) 1.00 5.00 0.89
  Days of poor mental health 285 5.13 (8.06) .00 3.00 NA
  Satisfaction with life 321 4.84 (1.40) 1.00 7.00 0.89
  Psychological well-being 324 4.77 (0.77) 1.50 6.00 0.84

Deidentified data, data scoring codes, analysis code, and and Prevention, 2018) captured participants’ self-reports of
R Markdown output for this manuscript are available at general health on a 1 (Poor) to 5 (Excellent) scale, number
https://osf.io/dejgn. The full Aging and Autism Study data of days in the past month health was poor because of
repository and codebook are available online at https://osf. physical health, number of days in the past month experi-
io/mwszy/ (Lodi-Smith et al., 2021). enced limitations due to health, total disease burden from
self-reported diagnoses on a checklist of 12 possible diag-
noses, and number of days in the past month health was
Measures poor because of mental health. Participants completed the
Table 1 provides the descriptive and reliability statistics Satisfaction with Life Scale (Diener et al., 1985) to assess
for each continuous measure. All measures are scored so life satisfaction on a 1–7 scale and a 14-item measure as-
that a higher score indicates a higher rating on the con- sessing their psychological well-being on a 1–6 scale (Ryff
struct. All scale scores are computed as averages of the & Keyes, 1995).
scale items.

Autism characteristics Analyses

The Autism-Spectrum Quotient scale (Baron-Cohen et al., All analyses were preregistered at https://osf.io/nc3m7 fol-
2001) was scored using AQ-Short scoring (Hoekstra et al., lowing guidelines for open science with preregistration of
2011) with two subscales, social behavior and an interest planned secondary analyses (Weston et  al., 2019). This
in numbers/patterns. Each item was rated from 1 indicating preregistration provides additional information about the
low agreement with an item related to autism characteris- study, measures, and scoring as well as researchers’ knowl-
tics to 4 indicating high agreement. Item-level data for the edge of the data prior to conducting analyses and experi-
full 50-item AQ are included in the data set available on the ence with the data sets. Deviations from the preregistration
Open Science Framework. and the rationale for these deviations can be viewed at
https://osf.io/dejgn.
Physical health, mental health, and well-being All analyses were conducted in R version 4.0.2 (R Core
Participants’ self-reported height and weight were used to Team, 2019). Item-level missing data were excluded in cal-
compute their body mass index (BMI). Participants also culated and averaged scale scores. The few cases of missing
completed the PROMIS (Ader, 2007) to self-report on fa- data at the scale level were excluded pairwise and sample
tigue, pain, physical function, sleep disruption, and ability size for specific analyses is reported. Correlations were the
for social participation as well as depression and anxiety. primary statistical analysis and were conducted with the
All scales were scored on a 1–5 scale. The Behavioral Risk apaTables package (Stanley, 2018). To test age-moderation
Factor Surveillance System (Centers for Disease Control effects, we standardized age and autism characteristic
Journals of Gerontology: PSYCHOLOGICAL SCIENCES, 2021, Vol. 76, No. 9 1741

scores, computed interaction terms, and regressed each as limitations. Medium correlations were estimated between
indicators of the target measure. the social behavior subscale of the AQ-Short and physical
Confidence intervals and statistical significance are re- function, general health, and days of poor physical health
ported for all effect sizes, but no explicit inferences are based and sleep disruption. The correlations between the social
on these values given the multiple analyses and sample size. behavior subscale of the AQ-Short and BMI, pain, and total
Instead, judgments about correlations are based on the ef- disease burden were negligible to small as were correlations
fect size estimates themselves using recent guidelines such between the numbers/patterns subscale of the AQ-Short
that r = 0.10 is small, r = 0.20 is medium, r = 0.30 is large, and physical health measures. In every instance where me-
and r = 0.40 is very large (Funder & Ozer, 2019). dium or stronger effects were present, estimates indicate
Because prior work on autism and physical health that elevated autism characteristics were associated with
largely relies on group-level comparisons, an anticipated greater physical health challenges.

Downloaded from https://academic.oup.com/psychsocgerontology/article/76/9/1738/6323676 by guest on 03 April 2022

correlation effect size was not available for power analyses. Elevated autism characteristics also were associated
Our prior work on the relationship between the AQ and with greater mental health and well-being challenges
well-being (Rodgers et  al., 2018) suggests a potential ef- with more robust effects for the social behavior subscale.
fect size range from r = −0.30 to r = −0.48. General power Specifically, there were very large correlations between the
computed with the pwr package in R (Champely, 2020) social behavior subscale of the AQ-Short and depression,
indicates the present sample should be adequately powered anxiety, days of poor mental health, poor satisfaction with
to detect these large effects as it is powered to detect a cor- life, and poor psychological well-being. These effect sizes
relation of 0.16 with a power of 0.90 and a significance were negligible to small for the numbers/patterns subscale
level of 0.05. of the AQ-Short. Exploratory nonlinear tests did not reveal
curvilinear or cubic relationships between autism charac-
teristics and health-related measures. Results of these ana-
Results lyses and scatterplots of the association between autism
The full zero-order correlation matrix of all study variables characteristics and health-related measures are available in
is available in Supplementary File 1. Table 2 reports zero- Supplementary File 2.
order correlations of autism characteristics with physical Given the age-graded nature of health and the need for
health, mental health, and well-being measures. The cor- research on age variability of the association between au-
relations of the social behavior subscale of the AQ-Short to tism characteristics and health, we conducted exploratory
the following physical health measures were large to very analyses to test the extent to which these variables mod-
large: fatigue, social participation, and number of days of erated the association between autism characteristics and

Table 2.  Correlations of Autism Characteristics With Physical Health, Mental Health, and Well-Being

Autism characteristics

Variable Social behavior Numbers/patterns

r [95% CI] p r [95% CI] p

Physical health
  Body mass index 0.04 [−0.11, 0.18] .626 0.01 [−0.14, 0.15] .941
 Fatigue 0.30 [0.19, 0.40] <.001 0.07 [−0.05, 0.18] .254
 Pain 0.16 [0.05, 0.27] .005 0.12 [0.01, 0.23] .037
  Physical function −0.21 [−0.32, −0.10] <.001 −0.12 [−0.23, 0.00] .045
  Sleep disruption 0.19 [0.08, 0.30] <.001 0.07 [−0.04, 0.19] .215
  Social participation −0.46 [−0.54, −0.36] <.001 −0.17 [−0.28, −0.06] .004
  General health −0.25 [−0.36, −0.14] <.001 −0.09 [−0.20, 0.03] .145
  Days of poor physical health 0.22 [0.10, 0.34] <.001 0.03 [−0.10, 0.15] .662
  Days of limitations 0.29 [0.18, 0.40] <.001 0.04 [−0.08, 0.16] .532
  Total disease burden 0.12 [0.01, 0.23] .036 0.03 [−0.08, 0.15] .559
Mental health and well-being
 Depression 0.39 [0.29, 0.49] <.001 0.12 [0.01, 0.24] .034
 Anxiety 0.47 [0.38, 0.56] <.001 0.13 [0.02, 0.24] .026
  Days of poor mental health 0.41 [0.30, 0.51] <.001 0.05 [−0.08, 0.17] .464
  Satisfaction with life −0.47 [−0.56, −0.38] <.001 −0.15 [−0.26, −0.03] .011
  Psychological well-being −0.62 [−0.69, −0.55] <.001 −0.18 [−0.29, −0.07] .002

Note: Brackets indicate 95% confidence intervals around the correlation.

1742 Journals of Gerontology: PSYCHOLOGICAL SCIENCES, 2021, Vol. 76, No. 9

health measures. These effects were negligible to small. engage a sample that is representative of our aging society.
Furthermore, controlling for age in the analyses did not at- Furthermore, participants were physically and cognitively
tenuate the zero-order effects. Results of these analyses are healthy enough to participate in our research. Survival bias
given in Supplementary File 3. may even be exacerbated for participants with elevated
autism characteristics. The present findings may underesti-
mate the burden of autism for healthy aging as the present
Discussion sample may comprise those individuals who have been able
to find ways to negotiate a world that was not and is not
The present research took a continuous measurement ap-
supportive of the challenges they face.
proach to conceptually replicate and extend prior work based
on categorical approaches to autism and health measures
(Bishop-Fitzpatrick & Rubenstein, 2019; Croen et al., 2015;

Downloaded from https://academic.oup.com/psychsocgerontology/article/76/9/1738/6323676 by guest on 03 April 2022

Supplementary Material
Hand et  al., 2020; Rydzewska et  al., 2018, 2019; Stewart
et  al., 2020; Wallace et  al., 2016). The association between Supplementary data are available at The Journals of
social behavior autism characteristics and self-reports of poor Gerontology, Series B: Psychological Sciences and Social
physical health as well as mental health and psychological Sciences online.
well-being paralleled the challenges shown in prior catego-
rical work. In many instances, these effects were quite large.
These effects remained robust when controlling for age. The
Funding
second autism characteristic from the AQ-Short, interest in This work was supported by the National Institute on
numbers and patterns, had much smaller and even negligible Aging at the National Institutes of Health (grant number
associations to health-related measures. Together, these find- 1R21AG059051-01).
ings suggest that the challenges evidenced in prior research
on autism and health may be due to the social challenges that
characterize autism. This echoes research that suggests that Conflict of Interest
social factors such as loneliness (Ong et al., 2016) are impor- None declared.
tant for healthy aging.

Acknowledgments
Limitations and Future Directions The authors would like to thank Mary Ann Langlois for
Longitudinal studies are critically needed in this area to be her substantial support of this program of research. The au-
able to build conclusions as to the directionality and/or re- thors would like to thank the participants of the Aging and
ciprocal effects of autism characteristics on healthy aging. Autism Study and our community partners for their sus-
Such studies can also measure key aspects of healthy aging tained engagement in our growing understanding of autism
in research on normative aging such as loneliness, social in later life. The preregistration for this project can be ac-
support, health behaviors, and personality traits to better cessed at https://osf.io/nc3m7. The data, code, and output
understand the processes at play in these patterns. For ex- for this manuscript can be found at https://osf.io/dejgn. The
ample, recent research in younger samples suggests that full Aging and Autism Study data repository and codebook
lifestyle factors may be an important component of poor are available online at https://osf.io/mwszy/ (Lodi-Smith
health outcomes for autistic adults (Weir et al., 2021). et al., 2021).
The AQ-Short does not capture the full landscape of
autism characteristics and, of course, cannot capture the
clinical significance of reported autism characteristics. It Author Contributions
focuses on social behavior problems associated with au- J. Lodi-Smith, J.  D. Rodgers, K.  Kozlowski, S.  Khan,
tism and interest in numbers and patterns versus the actual V.  Marquez Luna, and C.  J. Long prepared the primary
presence of restricted, repetitive, or stereotyped behaviors. drafts of the preregistration and manuscript. J. Lodi-Smith,
Furthermore, all of the measures in this sample relied on S.  Khan, V.  Marquez Luna, and J.  P. Donnelly conducted
self-report. Non-self-report and objective measure will be the analyses. C.  J. Long, J.  P. Donnelly, C.  Lopata, and
critical for future research and tests of measurement invar- M.  L. Thomeer provided substantial input into the study
iance will be essential moving forward to validate health design, preregistration, subsequent drafts, and final man-
measures for use in diagnosed autistic samples. uscript. All authors approved the final preregistration and
Finally, as with the majority of aging research, the submission. All coauthors have reviewed and approved the
present sample has both selection biases and survival biases manuscript prior to submission. This manuscript has been
at play. While efforts were made to engage participants submitted solely to this journal and is not published, in
from lower-income and more racially and ethnically di- press, or submitted elsewhere. The results of this work were
verse backgrounds, future work must do more to effectively presented at the 2020 Geneva Centre for Autism Virtual
Journals of Gerontology: PSYCHOLOGICAL SCIENCES, 2021, Vol. 76, No. 9 1743

Symposium in November 2020 and the Canisius College Journal of Biomedical Informatics, 42, 377–381. doi:10.1016/j.
Ignatian Scholarship Day in April 2021. jbi.2008.08.010
Heijnen-Kohl, S. M. J., Kok, R. M., Wilting, R. M. H. J., Rossi, G.,
& van  Alphen,  S.  P.  J. (2017). Screening of autism spec-
References trum disorders in geriatric psychiatry. Journal of Autism and
Ader,  D.  N. (2007). Developing the Patient-Reported Outcomes Developmental Disorders, 47(9), 2679–2689. doi:10.1007/
Measurement Information System (PROMIS). Medical Care, 45, s10803-017-3185-2
S1–S2. doi:10.1097/01.mlr.0000260537.45076.74 Hoekstra,  R.  A., Vinkhuyzen,  A.  A., Wheelwright,  S., Bartels,  M.,
Barneveld,  P.  S., Swaab,  H., Fagel,  S., van  Engeland,  H., Boomsma,  D.  I., Baron-Cohen,  S., Posthuma,  D., &
& de  Sonneville,  L.  M. (2014). Quality of life: A  case- van der Sluis, S. (2011). The construction and validation of an
controlled long-term follow-up study, comparing young high- abridged version of the autism-spectrum quotient (AQ-Short).
Journal of Autism and Developmental Disorders, 41(5), 589–

Downloaded from https://academic.oup.com/psychsocgerontology/article/76/9/1738/6323676 by guest on 03 April 2022

functioning adults with autism spectrum disorders with adults
with other psychiatric disorders diagnosed in childhood. 596. doi:10.1007/s10803-010-1073-0
Comprehensive Psychiatry, 55(2), 302–310. doi:10.1016/j. Lai,  M.  C., & Baron-Cohen,  S. (2015). Identifying the lost gener-
comppsych.2013.08.001 ation of adults with autism spectrum conditions. The Lancet
Baron-Cohen,  S., Wheelwright,  S., Skinner,  R., Martin,  J., & Psychiatry, 2, 1013–1027. doi:10.1016/S2215-0366(15)00277-1
Clubley,  E. (2001). The autism-spectrum quotient (AQ): Lodi-Smith,  J., Rodgers,  J., Luna,  M.  V.  M., Lundy,  Z., Long,  C.,
Evidence from Asperger syndrome/high-functioning autism, & Khan, S. (2021). Aging and autism study data repository &
males and females, scientists and mathematicians. Journal of codebook. doi:10.17605/OSF.IO/MWSZY
Autism and Developmental Disorders, 31(1), 5–17. doi:10.102 MacCallum,  R.  C., Zhang,  S., Preacher,  K.  J., & Rucker,  D.  D.
3/a:1005653411471 (2002). On the practice of dichotomization of quan-
Bishop-Fitzpatrick,  L., & Rubenstein,  E. (2019). The physical and titative variables. Psychological Methods, 7, 19–40.
mental health of middle aged and older adults on the autism doi:10.1037/1082-989X.7.1.19
spectrum and the impact of intellectual disability. Research Ong, A. D., Uchino, B. N., & Wethington, E. (2016). Loneliness and
in Autism Spectrum Disorders, 63, 34–41. doi:10.1016/j. health in older adults: A mini-review and synthesis. Gerontology,
rasd.2019.01.001 62, 443–449. doi:10.1159/000441651
Booth,  T., Murray,  A.  L., McKenzie,  K., Kuenssberg,  R., R Core Team. (2019). R: A language and environment for statistical
O’Donnell,  M., & Burnett,  H. (2013). Brief report: An evalu- computing. R Foundation for Statistical Computing.
ation of the AQ-10 as a brief screening instrument for ASD in Rodgers, J. D., Lodi-Smith, J., Hill, P. H., Spain, S. M., Lopata, C., &
adults. Journal of Autism and Developmental Disorders, 43(12), Thomeer, M. (2018). Personality traits and self-concept clarity
2997–3000. doi:10.1007/s10803-013-1844-5 mediate the relationship between autism spectrum disorder char-
Centers for Disease Control and Prevention. (2018). Behavioral Risk acteristics and well-being. Journal of Autism and Developmental
Factor Surveillance System. http://www.cdc.gov/brfss/ Disorders, 48, 307–315. doi:10.1007/s10803-017-3290-2
Champely,  S. (2020). pwr: Basic functions for power anal- Rydzewska,  E., Hughes-McCormack,  L.  A., Gillberg,  C.,
ysis. R package version 1.3-0. https://CRAN.R-project.org/ Henderson,  A., MacIntyre,  C., Rintoul,  J., & Cooper,  S.  A.
package=pwr (2018). Prevalence of long-term health conditions in adults
Cohen,  J. (1983). The cost of dichotomization. Applied with autism: Observational study of a whole country popula-
Psychological Measurement, 7, 249–253. doi:10.1177 tion. British Medical Journal Open, 8, e023945. doi:10.1136/
%2F014662168300700301 bmjopen-2018-023945
Croen,  L.  A., Zerbo,  O., Qian,  Y., Massolo,  M.  L., Rich,  S., Rydzewska,  E., Hughes-McCormack,  L.  A., Gillberg,  C.,
Sidney,  S., & Kripke,  C. (2015). The health status of Henderson, A., MacIntyre, C., Rintoul, J., & Cooper, S. A. (2019).
adults on the autism spectrum. Autism, 19, 814–823. General health of adults with autism spectrum disorders—A whole
doi:10.1177/1362361315577517 country population cross-sectional study. Research in Autism
Diener, E. D., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The Spectrum Disorders, 60, 59–66. doi:10.1016/j.rasd.2019.01.004
satisfaction with life scale. Journal of Personality Assessment, Ryff,  C.  D., & Keyes,  C.  L.  M. (1995). The structure of psycho-
49, 71–75. doi:10.1207/s15327752jpa4901_13 logical well-being revisited. Journal of Personality and Social
Funder,  D.  C., & Ozer,  D.  J. (2019). Evaluating effect size in psy- Psychology, 69(4), 719–727. doi:10.1037/0022-3514.69.4.719
chological research: Sense and nonsense. Advances in Methods Stanley,  D. (2018). apaTables: Create American Psychological
and Practices in Psychological Science, 2, 156–168. doi:10.1177 Association (APA) style tables. R package version 2.0.5. https://
%2F2515245919847202 CRAN.R-project.org/package=apaTables
Hand, B. N., Angell, A. M., Harris, L., & Carpenter, L. A. (2020). Stewart,  G.  R., Corbett,  A., Ballard,  C., Creese,  B., Aarsland,  D.,
Prevalence of physical and mental health conditions in Medicare- Hampshire,  A., Charlton,  R.  A., & Happé,  F. (2020). The
enrolled, autistic older adults. Autism, 24, 755–764. doi:10.117 mental and physical health profiles of older adults who endorse
7/2F1362361319890793 elevated autistic traits. The Journals of Gerontology, Series B:
Harris,  P.  A., Taylor,  R., Thielke,  R., Payne,  J., Gonzalez,  N., Psychological Sciences and Social Sciences, 76(9), 1726–1737.
& Conde,  J.  G. (2009). Research electronic data capture doi:10.1093/geronb/gbaa112
(REDCap): A  metadata-driven methodology and workflow Wallace,  G.  L., Budgett,  J., & Charlton,  R.  A. (2016). Aging and
process for providing translational research informatics support. autism spectrum disorder: Evidence from the broad autism
1744 Journals of Gerontology: PSYCHOLOGICAL SCIENCES, 2021, Vol. 76, No. 9

phenotype. Autism Research, 9, 1294–1303. doi:10.1002/ Wheelwright,  S., Auyeung,  B., Allison,  C., & Baron-Cohen,  S.
aur.1620 (2010). Defining the broader, medium and narrow au-
Weir,  E., Allison,  C., Ong,  K.  K., & Baron-Cohen,  S. (2021). An tism phenotype among parents using the Autism
investigation of the diet, exercise, sleep, BMI, and health out- Spectrum Quotient (AQ). Molecular Autism, 1, 1–9.
comes of autistic adults. Molecular Autism, 12, 31. doi:10.1186/ doi:10.1186/2040-2392-1-10
s13229-021-00441-x Woodbury-Smith,  M.  R., Robinson,  J., Wheelwright,  S., &
Weston,  S.  J., Ritchie,  S.  J., Rohrer,  J.  M., & Przybylski,  A.  K. Baron-Cohen,  S. (2005). Screening adults for Asperger
(2019). Recommendations for increasing the transparency syndrome using the AQ: A  preliminary study of its diag-
of analysis of preexisting data sets. Advances in Methods and nostic validity in clinical practice. Journal of Autism and
Practices in Psychological Science, 2, 214–227. doi:10.1177 Developmental Disorders, 35(3), 331–335. doi:10.1007/
%2F2515245919848684 s10803-005-3300-7

Downloaded from https://academic.oup.com/psychsocgerontology/article/76/9/1738/6323676 by guest on 03 April 2022