Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jun 1.
Published in final edited form as: Res Autism Spectr Disord. 2024 Mar 21;114:102371. doi: 10.1016/j.rasd.2024.102371

An Exploratory Study of Resilience to Stressful Life Events in Autistic Children

Jessica L Greenlee 1, Jennifer M Putney 2, Emily Hickey 3, Marcia A Winter 4, Sigan L Hartley 3
PMCID: PMC11087040  NIHMSID: NIHMS1982646  PMID: 38737198

Abstract

Background:

Autistic children experience more stressful life events (SLEs) than their neurotypical peers, which are related to poor mental health outcomes in both neurotypical and autistic individuals. However, there is a lack of longitudinal research assessing the perceived impact of stressful life events on autistic children’s mental health.

Method:

Utilizing a novel statistical technique (Ratcliff et al., 2019), called ‘area of resilience to stress events’ or ARSE in R, we aimed to quantify aspects of resilience, growth, and non-resilience for 67 autistic children (6–13 years old) enrolled in a larger longitudinal study who experienced a SLE. Parents reported demographic information (e.g., child age, biological sex, household income) as well as the child’s internalizing and externalizing symptoms and autism characteristics across multiple time points spaced one year apart (baseline, T2, T3, T4).

Results:

There was substantial variability in the resilience process within the sample. Older children exhibited a less adaptive resilience process (i.e., higher total scaled scores or arsets). Perceived stress of the disruptive event was not correlated with resilience; however, there was a significant child age x stress severity interaction, suggesting that younger children in households that perceived the disruptive event as highly stressful exhibited more efficient resilience, or lower arsets scores, compared to other children.

Conclusions:

This study introduces an innovative methodological approach to understanding the effects of stressful life events on the mental health of autistic children. Results have implications for family-based policy and practice and highlight for whom services may be most beneficial.

Keywords: autism, resilience, mental health, stressful life events, stress

Introduction

Autistic individuals are more likely to experience a stressful life event (Berg et al., 2016; Rigles, 2017) and experience more stress related to such events compared to their neurotypical peers (Bishop-Fitzpatrick et al., 2017; Taylor & Gotham, 2016). Experiencing stressful life events, or a broad array of events that confer some amount of stress through disruption of daily routines, is associated with having higher levels of emotional and behavioral challenges in autistic children (Hollocks et al., 2021). Frameworks surrounding research on stressful life events (SLEs) in autism populations tend to focus on (1) trauma and clinically relevant PTSD symptoms; (2) parent and family responses; or (3) risk factors that may predict poor outcomes after an SLE occurs (Bekhet et al., 2012; Dodds, 2020; Jacob et al., 2020). Comparatively few studies have examined autistic children’s resilience following adversity related to stressful life events.

The association between the stress of stressful life events and mental health outcomes is well established in non-autistic children (e.g., Grant et al., 2004; McLaughlin & Hatzenbuehler, 2009; Moore et al., 2017), with the negative effects of SLEs extending into adolescence and adulthood (Chapman et al., 2004; Schilling et al., 2007). Similar cross-sectional associations have been found in studies with autistic children relating SLEs to both internalizing (Kerns et al., 2017, 2020; Taylor & Gotham, 2016) and externalizing symptoms (Brenner et al., 2018; McDonnell et al., 2018); however, a recent longitudinal study found that experiencing a SLE was related to later internalizing problems only for those autistic children with challenges in cognitive shifting (Leno et al., 2022), suggesting that not all autistic children experience negative impacts on their mental health following a SLE. Overall, however, research in this area is quite limited.

A resilience framework offers an alternative approach to understanding how SLEs impact autistic individuals (Lai & Szatmari, 2019). Resilience is the capacity of an entity to adapt successfully given adversity or significant stress, and is a dynamic process that unfolds over time (Masten et al., 2021). Generally, when a person does better than expected given a set of environmental circumstances, we infer resilience. Most conceptualizations of resilience require there to be some exposure to “risk,” a biological or environmental context that confers an additive likelihood of poor outcomes (Obradovic et al., 2012), followed by “positive adaptation” either through an absence of poor outcomes or the presence of exceptionally adaptive outcomes (Wright, 2013). Resilience is the avoidance of negative outcomes in the context of adversity (Kaboski et al., 2017).

Autism research has historically focused on the autism diagnosis itself as the risk exposure and trajectories of behavioral and social outcomes are identified as “positive adaptation” (Stallworth & Masten, 2023). Outcomes are measured as changes in the behavioral phenotype of autism, such as improvements (or lack thereof) in social functioning, reduction in repetitive behaviors, or changes in adaptive functioning over time (Kaboski et al., 2017). Thus, research on resilience in the context of autism narrowly defines positive adaptation and intervention development to autism-related traits and fails to account for the increased risk of environmental stress exposure. As others have noted, however, autism is not an adversity (Lai & Szatmari, 2019) and there is a clear need for alternative approaches to resilience science in autism.

One alternative is to focus on outcomes that may hinder autistic individuals from doing well and meeting their own life goals, rather than assuming that autism itself is something to overcome. Mental health outcomes are one such example. We have ample evidence that autistic individuals are at increased risk for experiencing mental health challenges such as depression or anxiety and that these challenges pose a threat to their daily functioning and overall well-being (HeSLEy et al., 2019; Hollocks et al., 2019). Understanding when mental health challenges emerge and in response to which environmental conditions can help researchers and clinicians identify who may need help, when, and how to assist them better. Examining mental health outcomes using a resilience framework provides opportunities to develop neurodiversity-affirming care practices. It also provides key information about who does well as opposed to focusing on negative outcomes. Thus, we need research aimed at understanding autistic children’s response to “risk” that could impact their mental health, such as stressful life events.

Measuring Resilience Using Area Under the Curve

While researchers agree that understanding resilience has far-reaching implications for both prevention and intervention, measuring the resilience process has proven to be challenging. Part of the challenge lies in how scientists tap into a construct that is not easily defined, changes across time and context, and is affected by individual differences in other areas of functioning. One approach to assessing resilience is to use a statistical method that focuses on the stressor-recovery process, or how well an entity “bounces back” following stress exposure (Den Hartigh & Hill, 2022). Thus, it is not sufficient to measure personal characteristics that help an individual deal with stress; rather, the measurement of resilience should reflect the process of bouncing back. Inherit in this approach is that resilience should be measured at more than one time point, which allows for an assessment of the strength of deviation from baseline and the time it takes to “bounce back” to the original state (Den Hartigh & Hill, 2022). In other words, we cannot infer that an individual is resilient merely from the absence of depression symptoms – we need to know if depression symptoms existed before the SLE, how much symptoms change after the SLE compared to their original depressive state, and how long it takes to return to baseline depression levels.

Capturing this “bounce back” process can be done statistically, using a modified area under the curve analysis (AUC; Den Hartigh & Hill, 2022; Ratcliff et al., 2019). Importantly, an AUC analysis allows for the quantification of the magnitude and duration of deviations in an individual’s functioning compared to their baseline after exposure to a stressor. Compared to other techniques, an AUC analysis focuses on the process of change in a variable rather than the amount or rate of change. Others have used some form of AUC to assess resilience during a social stress task via cortisol reactivity (e.g., Mazurka, Wynne-Edwards, & Harkness, 2017), positive and negative affect in response to unpleasant daily events and whether it predicts later psychopathology (e.g., Kuranova et al., 2020), and in athletic performance training (e.g., Hill et al., 2020); however, this technique is underutilized in the resilience literature.

In 2019, Ratcliff et al. published a new statistical technique called ‘area of resilience to stress events’ or ‘ARSE.’ ARSE is an applied extension of the area under the curve analysis, a technique that has been suggested as useful in resilience science by others (Den Hartigh & Hill, 2022) and addresses some of the limitations in the resilience literature broadly and in the resilience literature specifically in autistic children. ARSE attempts to quantify the stressor-recovery process (see Figure 1). A strength of the ARSE approach is that it accounts for both the degree of departure from baseline (robustness) as well as how quickly an individual returns to baselines (rapidity) in one quantitative metric (Ratcliff et al., 2019). The ARSE technique requires a baseline assessment of an outcome of interest before a SLE occurs and then repeated measurement of that outcome following the SLE. ARSE then calculates the deviations from the baseline score for each individual, which includes the quantification of resilience (a return to baseline functioning), growth (when the end state exceeds baseline functioning), and non-resilience (when there is no return to baseline functioning) (Ratcliff et al., 2019). The result of the ARSE analysis is a numeric score representing the overall efficiency of the resilience process.

Figure 1.

Figure 1

Open in a new tab

The Resilience Process as Conceptualized via ARSE

Note. A visual representation of the resilience process as captured by ARSE, which includes an assessment of both the rapidity and robustness of deviations from baseline. Figure adapted from Ratcliff et al., 2019.

There are several benefits to this approach when used in a resilience framework. Compared to techniques like trajectory-based models that describe the course of a variable across time, often to examine how covariates affect the shape of that trajectory, ARSE provides a quantitative metric unique to each individual specifically assessing the robustness and rapidity of change in a variable. Thus, ARSE scores represent a two-dimensional characterization of resilience that other statistical techniques fail to capture that can be used in further testing (i.e., predictor in a regression model). In addition, each individual acts as their baseline measure rather than using the sample average (i.e., intercept value) which can obscure individual differences and variability in the resilience process. Finally, the theoretical orientation of the ARSE technique focuses on resilience as a process rather than a static outcome.

Risk and Protective Factors in the Resilience Process

Key to understanding the resilience process and developing effective preventative measures is identifying what factors help children do well in the face of adversity (protective factors) and what factors increase the likelihood of poor outcomes (risk factors). Research in non-autistic children from the past few decades has identified a “shortlist” of common protective factors at the individual (e.g., demographic characteristics, self-regulation, problem-solving), family (e.g., sensitive caregiving, family routines), and community (e.g., effective teachers, social support) levels (Masten et al., 2021). The lack of research on the resilience process in autistic children has resulted in limited information about potential protective factors unique to this population (Rigles, 2017). A preliminary step is to identify who may be at risk for a less efficient resilience process. Generally, research suggests that factors such as lack of co-occurring intellectual disability, fewer or less severe autism traits, and receiving intervention services are associated with positive outcomes in the context of autism (see Kaboski, McDonnell, & Valentino, 2017 for a review); however, to our knowledge, no research has examined whether such factors moderate the association between SLEs and the resilience process.

Current Study

Despite evidence that autistic children are at increased risk for SLEs, the resilience process has not been well-studied in this population (Elsabbagh, 2020). Therefore, the purpose of this study was to utilize a novel statistical method, area of resilience to stress events (‘arse’; Ratcliff et al., 2019), to examine the resilience process in response to stressful life events, as well as to examine characteristics that are related to resilience in autistic children. In particular, we examined whether characteristics such as child sex, age, co-occurring intellectual disability, autism characteristics, and participation in intervention services change how SLEs impact the resilience process in autistic children. Due to the limited research in this area, the current study is exploratory, and hypotheses were not made a priori.

Method

Participants

Data for this project comes from a larger study examining longitudinal outcomes of 188 families of autistic children. Participants were recruited from local schools, autism clinics, community centers, and research registries in the Mid-Western United States. Children who had received a documented diagnosis of autism spectrum condition from a medical or educational professional were eligible to participate in the larger study. Couples did not have to be married nor did they have to be the biological parent of the autistic child. Diagnostic documentation had to include a score above the Autism Spectrum cutoff (69% reported scores above the Autism cutoff) on the Autism Diagnostic Observation Schedule (about half were assessed using Module 2; ADOS-2; Lord et al., 2012). The average age of diagnosis was 48.17 months and ranged from 18 to 140 months. If more than one child in the family had been diagnosed with autism, the oldest autistic child was chosen as the target child.

The current study includes the 67 (35%) autistic children who experienced a stressful life event in the year since the baseline visit (i.e., between T1 and T2). All children were 5–12 years at baseline (M = 9.32, SD = 2.23; 88.1% male; 35.8% with co-occurring intellectual disability). A full description of the sociodemographic characteristics of the study sample can be found in Table 1. In this study, we followed a definition of resilience that necessitates some exposure to adversity to demonstrate resilience. Given that the primary analysis is a quantification of the resilience process, we chose to exclude children who did not experience a SLE from the analysis.

Table 1.

Child, parent, and family demographics

Child
Age, M years (SD, range) 9.37 (2.21, 6–13)
Sex (male), n (%) 58 (87.9)
ID status (yes), n (%) 23 (34.8)
 Mild ID 9 (38.3)
 Moderate ID 8 (36.6)
 Severe or Profound ID 5 (23.3)
 Missing 1 (1.8)
Life events, n (%)
 1 Life event 31 (47)
 2 Life events 19 (28.8)
 3 life events 14 (21.2)
Parent Mother Father
Age, M years (SD, range) 39.88 (6.24, 24–54) 41.20 (6.82, 27–60)
 Ethnicity
  White non-Hispanic 62 (93.9) 62 (93.9)
  African-American 1 (1.5) 1 (1.5)
  Asian or Pacific Islander 2 (3.0) 3 (4.5)
  White Hispanic 1 (1.5)
Family
 Household Income, n (%)
  < $19,999 2 (3)
  $20,000–39,999 3 (4.5)
  $40,000 –,59,999 14 (21.2)
  $60,000 – 79,999 12 (18.2)
  $80,000 – 99,999 7 (10.6)
  > $ 100,000 23 (34.8)
Open in a new tab

Materials & Design

Stressful life events

At the first follow-up visit (T2), parents reported whether their family had experienced any of five common SLEs during the year since their previous visit: moved residences, parent job loss or change, death in the family, major illness or injury in the family, and birth in the family. These events were chosen to represent common disruptions or transitions experienced across the family unit in three broad categories: interpersonal problems, loss of social status, and employment difficulties (Cohen, Murphy, & Prather, 2019). Additionally, parents were able to write in additional SLEs at their discretion. Each SLE was coded as 1 ‘present’ or 0 ‘absent’. Parents also provided a score indicating the severity of stress felt by each endorsed SLE from 0 (‘no stress’) to 5 (‘extremely stressful’). A SLE stress score was created by averaging the stress ratings for any SLE exposure per family, including any write-in SLEs.

Child Emotional and Behavioral Functioning

Child Emotional and Behavioral Functioning was measured at baseline, the first follow-up visit (T2), and two additional follow-up visits (T3 and T4). Parents completed the Child Behavior Checklist (CBCL; Achenbach & Rescorla, 2001) at each of the four visits. The CBCL is a widely used measure of child emotional and behavioral symptoms measured via 140 questions rated on a scale from 0 (not true) to 2 (very true or often true). Two subscales were created following the author’s recommendations: internalizing and externalizing symptoms. These subscales are the primary outcome measures in the current study. The CBCL is highly reliable within the ASD population (Pandolfi et al., 2014), and had a high internal consistency across all timepoints (internalizing: Cronbach’s α = .84 to .85; externalizing: Cronbach’s α = .90 to .92)

Child Autistic Traits

Parents also completed the Social Responsiveness Scale (SRS-2; Constantino & Gruber, 2005), a 65-item questionnaire used to assess the severity of a child’s autism traits, at each of the four visits. Parents rated items on a 4-point Likert scale (1 = ‘not true’; 4 = ‘almost always true’) to create a total score. Higher scores reflect more intense autism traits or characteristics (Cronbach’s α = .93).

Service Use

Parents reported whether or not their child currently received any of the following intervention services: occupational therapy, physical therapy, speech therapy, respite care, behavioral training, counseling, or Head Start, coded as 1= ‘yes’ and 0 = ‘no’.

Sociodemographics

Demographic information about the child, parents, and the family was provided by the parents. This included child age, sex, intellectual disability status (coded as 1= ‘yes’ and 0 = ‘no’ and parents reported the level of intellectual disability as mild, moderate, or severe/profound), family income, parent education, etc.

Procedures & Analysis

The larger study was approved by the Institutional Review Board at a large university in the Midwestern United States. All participants provided informed consent as well as permission for their child to participate in the study. Following the consent process, families completed a 2-hour baseline visit in the lab or their home according to their preference. Families were then followed for three years, completing a total of four research sessions between 2014 and 2019 each spaced a year apart (M = 11.66 months; SD = 3.70 months). Baseline and follow-up visits included a series of caregiver questionnaires and video-recorded observational tasks. Only survey responses are used in the current study.

ARSE Analysis Description

Data were analyzed in R using the ‘arse’ package (Ratcliff et al., 2019) to characterize the area created from deviations from baseline child mental health scores (internalizing and externalizing symptoms) following a SLE. This technique calculates the area created from the magnitude and speed at which an individual returns to their baseline functioning following a stressful life event via a series of x-y coordinates. The region beneath the curve, or area of resilience, represents the efficiency of the resilience process – a smaller area reflects a more efficient process (i.e., lower magnitude and faster return to baseline) while a larger area reflects a less efficient resilience process (i.e., higher magnitude and slower return to baseline). This provides a scaled numeric value (arsets or the total scaled score for arse) that represents the resilience process while accounting for end-state growth and/or non-resilience that can then be used in subsequent analyses identifying factors associated with the resilience process. The arsets score is calculated by multiplying the arset value (which is the raw arse score – the area of growth) by the quotient of the baseline and endstate score (Ratcliff et al., 2019). Lower arsets values indicate a more efficient resilience process while larger values indicate a high degree of departure from baseline and/or a slow return to baseline.

Data-Preprocessing & Analytic Plan

We used multiple imputation to account for missing data using the ‘MICE’ package in R (van Buuren & Groothuis-Oudshoorn, 2011). Participants needed at least two CBCL data points to be included in the analysis. Participants with missing data did not differ on child age, child sex, intellectual disability status, or parent education compared to those with complete data. We examined associations between arsets and socio-demographic characteristics (e.g., child sex, age, intellectual disability status, household income, autism symptom severity, therapy services) and characteristics of the SLE such as perceived severity of the SLE and total number of SLEs using independent samples t-tests and correlations as appropriate for categorical vs. continuous variables. Finally, exploratory analyses using regression-based models in R investigated potential interactions between socio-demographic characteristics and perceived severity of the SLE to answer questions related to whom disruptive events may have particularly negative effects within the autistic population.

Results

Descriptive Statistics

Preliminary data analysis examined potential socio-demographic differences of those autistic children who did experience an SLE between T1 and T2 (39% of the sample) and those in the sample who did not. Independent samples t-tests for continuous variables (age, baseline mental health) and chi-square tests of independence for categorical variables (ID status, sex, income) were used to test for differences. There was no baseline difference between those children who experienced a SLE and those who did not in age [t(187) = 1.449, p = .149; MSLE = 8.23, SD = 2.72; MNOSLE = 7.73, SD = 2.23] or mental health [internalizing symptoms: t(186) = 0.587, p = .558; MSLE = 14.796, SD = 7.744; MNOSLE = 14.082, SD = .731; externalizing symptoms: t(186) = 0.242, p = .809; MSLE = 15.402, SD = 1.048; MNOSLE = 15.062, SD = .864]. Similarly, chi-square tests revealed no significant differences in child sex [X2 = 0.388, p = .53], intellectual disability status [X2 = 0.175, p = .68], or family income [X2 = 5.67, p = .894] based on whether the child had experienced a SLE or not between T1 and T2.

Describing Stressful Life Events

Frequency information regarding individual SLEs can be found in Table 2. Parent job loss or change was the most common SLE and no family reported a birth. Most families reported a single SLE (49.2%) but about a third reported experiencing two SLEs (28.6%) and 22.2% reported three SLEs. Twenty-six families also chose to describe at least one additional SLE not captured by the 5 larger categories. Common write-in SLEs included issues with or transition to school or another school, divorce, pet adoption or death, marriage, other children receiving an autism diagnosis, and natural disasters. There were no statistically significant differences in resilience scores based on the type of SLE, the perceived stress of the SLE, or the number of SLEs reported (see Table 2). However, children who did not experience a change in residence had lower arsets scores (i.e., a more efficient resilience process) for externalizing symptoms compared to children who did experience a change in residence. In other words, children who did not change residence showed more resilient outcomes compared to children who did change residence. Children of families who experienced more than one SLE did not differ in their resilience scores compared to children of families who reported experiencing a single SLE (ARSEts externalizing: t(47.683)=1.556, p =.126; ARSEts internalizing: t(61) = −.019, p = .985).

Table 2.

Frequency of stressful life events and difference in ARSEts scores based on SLE

Event Occurred Event did not Occur
Type of SLE N (%) Mean Stress Rating (SD) M SD M SD t value p value Cohen’s d
ARSEts Scores – Externalizing Symptoms
Moved Residence 19 (28.8) 4.37 (.831) .298 15.689 −9.671 18.694 2.032 .046* .556
ARSEts Scores – Internalizing Symptoms
−7.771 24.602 −9.191 35.320 .159 .874 .044
ARSEts Scores – Externalizing Symptoms
Parent job loss or change 33 (50.0) 3.88 (.960) −5.081 16.453 −8.466 20.232 .736 .464 .184
ARSEts Scores – Internalizing Symptoms
−6.656 39.007 −11.019 23.678 .537 .593 .134
ARSEts Scores – Externalizing Symptoms
Death in family 28 (42.4) 4.04 (1.071) −6.511 14.728 −6.883 20.886 .08 .937 .020
ARSEts Scores – Internalizing Symptoms
−11.037 42.359 −7.006 22.111 −.492 .625 .124
ARSEts Scores – Externalizing Symptoms
Major illness or injury 31 (47.0) 4.16 (.934) −9.285 19.332 −4.311 17.248 −1.087 .281 .272
ARSEts Scores – Internalizing Symptoms
−11.728 41.121 −5.990 21.330 −.707 .482 .177
Birth in family 0 (0.0) 1.00 (0)
Open in a new tab

Describing ARSE

Figure 2 shows the sample average resilience process for both internalizing and externalizing symptoms. On average, autistic children showed a growth process in externalizing symptoms (i.e., the average level of externalizing symptoms was lower than baseline functioning) across time following a SLE that occurred between Time 1 and Time 2 (the greyed area in Figure 2). For internalizing symptoms, children tended to show a mixed pattern of resilience and non-resilience such that initially, there was some evidence of growth (i.e., the average level of internalizing symptoms was lower than baseline functioning) but by year 3 (Time 4), the average internalizing score was above the average baseline score (i.e., non-resilience). Thus, on average, autistic children in this sample showed a pattern of growth, or a pattern such that the end state at Time 4 indicated a trend of improving symptoms, for both externalizing (M = −9.248; SD = 27.300; range = 191.935) and internalizing symptoms (M = −8.895; SD = 31.934; range = 227.074), although it is unclear whether the rate of change is statistically significant as we did not test the rate of change of symptoms across the three years (i.e., slope). Rather, the ARSE score provides a metric of both the speed and the magnitude of change from baseline by assessing the area above and below the curve created from the baseline score and subsequent scores across each the three years.

Figure 2.

Figure 2

Open in a new tab

Average Resilience Process for Internalizing and Externalizing Symptoms

The large standard deviations in arsets scores, or total scaled arse scores, suggest that there is much variability or spread to the scores and that the average arsets score does not represent the data well. This variability in the resilience process is exemplified in the arse plots of individual children in Figure 3. The focus of these plots is not on the slope of the line but rather on the area created by the baseline score and the symptom score at each timepoint (greyed area on each plot). For example, the child in Figure 3a represents a pattern of resilience - externalizing symptoms were higher than their baseline score following the disruptive event between baseline and T1 but eventually returned to baseline across the three years (arsets = 9.50). The child in Figure 3b exemplifies growth given that their end state showed a lower externalizing score compared to their baseline score (arsets = −2.51). Figure 3c represents non-resilience. Although the child maintained their baseline symptoms immediately following the SLE, their end state showed a higher externalizing symptom score compared to their baseline (arsets = 3.348). The child in the final graph (Figure 3d) exhibited a mixed pattern of growth and non-resilience, with their externalizing scores fluctuating from below to above baseline levels (arsets = 2.250). These children’s resilience processes look markedly different from the average resilience process for externalizing symptoms shown in Figure 2, but provide a unique view of the variety of experiences following a SLE in the current sample.

Figure 3.

Figure 3

Open in a new tab

Plots exemplifying individual differences in arse trajectories using externalizing symptoms as an example

Note. The dotted line represents each child’s baseline externalizing symptom score. Figure 3a provides an example of an area of resilience in which the child’s symptoms are higher than their baseline score but return to baseline eventually - the process is slow (rapidity) and the deviation from the baseline score is high (magnitude), yielding a high area of resilience score (arsets = 9.50). Figure 3b is an example of growth – the end externalizing score at the final timepoint is lower than the baseline score (arsets = −2.51). The child in Figure 3c represents non-resilience – the child’s score at the final timepoint is higher than their baseline externalizing score (arsets = 3.34). Finally, figure 3d exhibited a mixed pattern of growth/resilience and non-resilience (arsets = 2.250).

Primary Analysis

Resilience Process (ARSE) and Mental Health

Externalizing Symptoms.

There were no statistically significant associations between externalizing arsets scores and the SLE severity [r(62) = −.136, p = .290], or the number of SLEs reported [r(63) = .092, p = .473]. Arsets scores were correlated with child age [r(65) = .394, p = .001] – older children had higher arse scores, or a less efficient resilience process. There was no difference in arsets scores based on whether or not the child had a co-occurring intellectual disability [t(63) = .324, p =.754], child sex assigned at birth [t(63) = .760, p =.45], or household income [t(59) = −1.129, p =.263]. Arsets scores were also not correlated with the child’s level of autistic traits at baseline [r(65) = −.173, p = .17] or at the final visit three years later [r(43) = −.065, p = .68]. We also examined whether participation in any services (e.g., occupational therapy, speech therapy, counseling, etc.) was related to arsets scores and found that children participating in behavioral intervention had lower arsets scores (M = −9.917, SD = 19.981), or a more efficient resilience process than children not participating in behavioral intervention (M = −0.528, SD = 9.661; t(56.91) = −2.474, p = .008, Cohen’s d = .559). There were no other statistically significant differences based on the type of service participation.

Internalizing Symptoms.

Similarly, there were no statistically significant associations between arsets scores and the SLE stress severity [r(62) = −.242, p = .058], or the number of SLEs reported [r(63) = −.036, p = .779]. Arsets scores for internalizing symptoms were correlated with child age [r(65)=.322, p =.009; older children had a less efficient resilience process] but not the child’s level of autistic traits at baseline [r(65) = −.178, p = .15] or at the final visit [r(44) = −.047, p = .76]. There was also no difference in resilience scores based on ID status, child sex assigned at birth, or household income. Children participating in behavioral therapy also exhibited lower arsets scores (M=−9.087, SD = 22.446), or a more efficient resilience process, compared to children not participating in behavioral therapy (M = −0.431, SD = 12.440; t(59.210) = −1.950, p = .028, Cohen’s d = .45).

Child Age as a Moderator.

Given the correlation between child age and resilience scores, we tested whether child age moderated the association between perceived stress of the SLE and resilience score using the ‘lm’ package in R. Perceived stress of the SLE was mean-centered before analysis. For both externalizing [F(3,58) = 7.041, p < .001] and internalizing [F(3,58) = 5.599, p = .002] symptoms, the interaction between perceived stress and child age was statistically significant (see Table 3). In both cases, there was a significant negative relationship between perceived stress of the SLE and arsets resilience scores, such that higher perceived stress was associated with lower arsets scores (i.e., more efficient resilience process) for younger children (−1 SD) (externalizing: B = −.45, p =.004; internalizing: B = −.51, p =.002) but not older children (mean and +1 SD; see Figure 4).

Table 3.

Exploratory regression model for the interaction of perceived stress and child age

Dependent Variable Predictor df β t p sr 2 95% CI
ARSE Score (Externalizing) Stress 58 −0.10 −0.89 .379 .01 [0.00, 0.05]
Age 58 0.32 2.83 .006** .10 [0.00, 0.23]
Stress × Age 58 0.35 3.26 .002** .13 [0.00, 0.28]
ARSE Score (Internalizing) Stress 58 −0.21 −1.83 .072 .04 [0.00, 0.14]
Age 58 0.24 2.07 .043* .06 [0.00, 0.16]
Stress × Age 58 0.30 2.74 .008** .10 [0.00, 0.23]
Open in a new tab
Figure 4.

Figure 4

Open in a new tab

Plotting the interaction between the perceived stress of the SLE and child age.

Note. The star indicates the presence of a statistically significant slope. On the x-axis, higher scores indicate higher levels of perceived stress of the SLE. This rating was mean-centered prior to analysis. On the y-axis, higher scores indicate a less efficient resilience process. In both graphs, there was a statistically significant negative relationship between perceived stress of the SLE and arse scores for younger kids. The other slopes were not statistically significant.

Behavioral Therapy as a Moderator.

We also explored whether participating in behavioral therapy moderated the relationship between the perceived stress of the SLE and resilience scores after controlling for child age. For both externalizing arsets scores [F(4,57) = 4.304, p = .004] and internalizing arsets scores [F(4, 57) = 3.579, p = .011], the interaction between perceived stress of the SLE and behavioral therapy was statistically significant above and beyond the effect of child age (Table 4). There was a negative relationship between arsets resilience scores and perceived stress for children participating in behavioral therapy (B = −7.50, SE = 3.46, p =.03) but the relationship was non-significant for children not in behavioral therapy (B = 5.22, SE = 4.99, p =.30; see Figure 5). A similar pattern was found for internalizing arsets scores such that the negative relationship between stress and resilience scores was statistically significant for those children participating in behavioral therapy (B = −10.32, SE = 3.64, p =.01) but not for children not in behavioral therapy (B = 2.75, SE = 5.26, p =.60).

Table 4.

Exploratory regression model for the interaction of perceived stress and behavioral training controlling for child age

Dependent Variable Predictor df β t p sr 2 95% CI
ARSE Score (Externalizing) Stress 57 −0.11 −0.90 .374 .01 [0.00, 0.06]
Behavioral Training 57 −0.19 −1.62 .111 .04 [0.00, 0.12]
Age 57 0.30 2.53 .014* .09 [0.00, 0.21]
Stress × Behavioral Training 57 −0.26 −2.09 .041* .06 [0.00, 0.16]
ARSE Score (Internalizing) Stress 57 −0.21 −1.74 .087 .04 [0.00, 0.13]
Behavioral Training 57 −0.12 −1.02 .311 .01 [0.00, 0.07]
Age 57 0.23 1.92 .059 .05 [0.00, 0.15]
Stress × Behavioral Training 57 −0.25 −2.05 .045* .06 [0.00, 0.16]
Open in a new tab
Figure 5.

Figure 5

Open in a new tab

Plotting the interaction between the perceived stress of the SLE and the presence of behavioral training while controlling for child age.

Note. The star indicates the presence of a statistically significant slope. On the x-axis, higher scores indicate higher levels of perceived stress of the SLE. This rating was mean-centered prior to analysis. On the y-axis, higher scores indicate a less efficient resilience process. In both graphs, there was a statistically significant negative relationship between perceived stress of the SLE and arse scores for children who were receiving behavioral training after controlling for child age. The other slope was not statistically significant.

Discussion

This study aimed to use a novel statistical technique, ARSE (Ratliff et al., 2019), to quantify the resilience process in autistic children responding to common stressful life events. We chose to focus on mental health as a proxy for resilience given the high prevalence of mental health challenges experienced by autistic children and evidence that stressful and stressful life events can harm children’s mental health (Dodd, 2022). Rather than relying on self- or proxy reports of resilience measured via a questionnaire at a single time point, the ARSE technique takes into account an individual’s baseline mental health and measures the magnitude of the timing of deviations from that baseline over several years. Thus, rather than calculating the rate of change in symptoms over time (slope), ARSE provides a quantitative metric of both the robustness and rapidity of symptom change (the greyed area in Figure 1) via a modified area under the curve analysis.

This allows for a process-oriented assessment of resilience that is unique to each individual.

On average, experiencing a SLE was not associated with mental health symptoms in this sample of autistic children. In contrast, most research on SLEs and autism has found a relationship between SLE exposure and mental health symptoms (Brenner et al., 2018; McDonnell et al., 2019; Taylor & Gotham, 2019), although other longitudinal work has found similar results to the current study (e.g., Leno et al., 2022). The average arsets score for both externalizing and internalizing symptoms showed slight growth or a pattern in which the average externalizing score at year 3 (Time 4) was lower than the average baseline score; however, focusing on the sample average obscures individual differences in the resilience process of autistic children as was evidence by the large standard deviation of arsets scores and exemplified in Figure 2. This suggests that the average arsets score may not be a reliable indicator of the resilience process in this population. Thus, an individual differences approach to understanding resilience in autistic children is warranted. Some autistic children exhibited no long-term impairment in mental health (resilience), and others experienced significant challenges in externalizing and/or internalizing symptoms across three years (non-resilience). The timing of mental health challenges also varied by individual as seen in the plotted resilience processes (Figure 3). For example, one child showed a delayed response (i.e., symptom change occurred later; Figure 3d) while another child exhibited higher externalizing symptom scores right away but the externalizing symptom score ended below baseline (Figure 3a). Together, average scores and individual plots help identify who is doing well in the face of adversity and for whom intervention might be necessary. Such findings also highlight that the timing of intervention may also be important. For example, mental health screening following an SLE may initially overlook the child in Figure 3b. By taking a process-oriented approach to measuring resilience, we can identify these individual differences and apply more targeted and tailored interventions. Additional research is warranted aimed at understanding exactly for whom SLEs pose a risk and under what contexts that risk manifests specifically in autistic populations.

Child age was significantly negatively correlated with both internalizing and externalizing resilience scores, with younger children exhibiting a more efficient resilience process in the presence of more stressful SLEs as reported in the first moderation analysis. This was an unexpected finding and we have two primary interpretations. First, it is worth considering that it could be a spurious interaction effect. This is a small sample resulting in limited statistical power to detect interactions and the potential for an inflated Type I error rate. The interaction could also be picking up general developmental trends (e.g., externalizing challenges lessen as children age), or what parents are reporting as “high stress” is actually relatively low/moderate stress.

Second, it could be a true effect, suggesting that there is something about early childhood (6–7 years old) compared to middle childhood and early adolescence (8–13 years old) that changes the relationship between the stress of a SLE and resilience. The social environments of these two developmental periods are different and may be related to age differences in the resilience process identified in this study. Parenting a 6-year-old through a change in residence will likely look different from parenting an 11-year-old through a similar disruption. For example, parents may be more likely to co-regulate with younger children in the face of stress (Rosanbalm & Murray, 2017) compared to older children or parents may more explicitly use exposure to an SLE as an opportunity to facilitate regulatory skills, an important protective or resilience factor for non-autistic children (Masten & Barnes, 2018).

Resilience research in non-autistic groups has consistently found that dose exposure to adversity matters such that more severe or more frequent exposures to stress have a larger impact on brain and physiological development, which are then associated with poorer mental and physical health outcomes (Masten & Barnes, 2018). Thus, the present interaction may represent a dose effect where older children have had more opportunity for exposure to various types of stressful events and have a less efficient resilience processes (i.e., worse arsets scores) compared to younger children who have had fewer exposures. The timing of stress exposure is also important. While exposure to stress during early childhood can have long-term developmental effects, it is also a time when protective factors can have a significant impact. Preventative interventions in non-autistic children at risk for adversity often target early childhood given the rapid physiological, social, and emotional development happening during that developmental period (Masten & Barnes, 2018). It may be that the interaction reported in the present study, that more stressful SLE exposure is associated with better resilience scores in younger autistic children, reflects this developmental window of opportunity. In other words, if there are protective factors in place (e.g., positive relationship with an adult, skilled parenting, coregulation with an adult, well-established family routines, etc.; Masten & Barnes, 2018), they may be more impactful for younger children, resulting in a more efficient resilience process. Another possibility is that the interaction between the stress rating of the SLE and child age is a proxy for family-level factors. There may be something about families of older autistic children that undermines the resilience process. For example, families of older children may be more likely to have additional younger children in the household or additional children with special healthcare needs. These families may also be dealing with additional stressors that families of younger children are not. The transition to school, for instance, can be a particularly stressful time for autistic children and their families (Nuske et al., 2018; Marsh et al., 2017). The additional demands of a SLE may diminish children’s ability to handle stress or it could impede typical protective processes associated with positive family relationships. Parents who are rating the SLE as more stressful may also be compensating in their parenting specifically for younger children. There are, of course, other potential explanations, and these interpretations are speculative. Additional research is needed to replicate and extend this finding. Moving forward, research that examines environmental factors specific to autistic children will be important for understanding how autistic children respond to SLEs, who may be most impacted by SLEs, and how clinicians can help mitigate the negative impact of SLEs specific to the needs of autistic children.

The interaction between behavioral therapy participation and the perceived stress of the SLE was also not expected; however, conclusions based on these results are limited because there is no data available that contextualizes the behavioral therapy children were receiving. Children’s resilience process was better (i.e., lower arse scores) as the stress of the SLE increased specifically for those receiving behavioral therapy, suggesting that children may be learning skills that help them adapt and cope in the face of stress. We do not have any context about the types of specific services this sample received but we wonder whether behavioral therapy in particular is more likely to teach skills that help autistic children deal with stress compared to other therapies listed (e.g., receiving counseling services). For example, Scarpa and colleagues (2021) propose a model that targets flexibility as a key mechanism of intervention to build resilience in autistic children. They specifically highlight skills commonly taught in behavioral therapy such as relaxation techniques, biofeedback, mindfulness, and problem-solving as ways to foster resilience. An alternative explanation is that behavioral therapy may be acting as a proxy for something else. For example, it could be that these moderating effects have more to do with being a child of a parent who seeks out (and is granted) or has the means to acquire therapy rather than the therapy itself. In other words, being in behavioral therapy has more to do with some characteristic of the parent or family as opposed to the therapeutic environment. This is all speculative, and additional research is needed to understand the relationship between SLEs, behavioral therapy, and mental health outcomes in autistic children.

Resilience scores were not correlated with whether the child had a co-occurring intellectual disability or the severity of their autism traits. This may mean that the biology of autism is not, in and of itself, putting children at risk in the context of SLEs. Rather it may be that the environmental conditions in which autism traits are situated tip the scale toward resilience (or not). This does not necessarily mean that, compared to neurotypical or non-autistic children, we would not find differences in the resilience process of autistic children, and we cannot conclude that autism is or is not a risk factor for more negative outcomes in the face of SLEs. It does provide preliminary evidence, however, that environmental factors are important and additional research should explore environmental factors that contribute to resilience in autistic populations. Alternatively, the lack of a correlation between autism traits and arse scores could also be methodological. Our sample size was relatively small and possibly underpowered, which could be why the correlation coefficient was not significant. There may also be a significant overlap between what is measured by the CBCL (mental health measure) and the SRS-2 (autism trait measure), limiting the unique variance explainable by arse scores.

Much of the research to date on stressful life events has focused on adverse childhood experiences (ACEs), which are more severe events known to have long-term physical and mental health impacts. This body of research suggests that the number and types of ACEs a child experiences matter in determining outcomes. In contrast, in the current study, there was no correlation between the type of SLE and resilience scores or the number of SLEs and resilience scores. One reason for this discrepancy might be that SLEs “disrupt,” but are typical life experiences that are not predicted to have long-term consequences; however, ACEs adversely impact the nervous system, are not typical life experiences, and have known long-term consequences. It is also possible that facing some adversity may be adaptive (Liu, 2015). Thus, the processes through which ACEs and SLEs impact children are different and should be explored in subsequent research.

Future research should also address process-oriented questions to understand how SLEs impact the mental health of autistic children. One possibility is through parents. It may be that SLEs are associated with child mental health symptoms via exacerbation of parent symptoms of stress or mental health and there is empirical evidence in non-autism populations to support this hypothesis (e.g., Platt, Williams, & Ginsburg, 2016). This is in line with the interpersonal theory of stress that suggests that stress impacts children’s mental health through the disruption of interpersonal relationships and interactions (Grant et al., 2006; Hammen & Rudolph, 1996). Examining parent stress or mental health as the mechanism through which SLEs impact autistic children’s outcomes, to our knowledge, has not explicitly been tested. Still, there is some evidence that parent stress impacts children’s internalizing and externalizing symptoms in the context of autism (e.g., Rodriguez et al., 2019). Understanding whether parent stress acts as a mechanism linking SLEs and child mental health outcomes would provide important information for intervention development.

While this longitudinal study provides evidence of individual differences in the resilience process of autistic children responding to SLEs, results should be considered in light of study limitations. Although a wide range of the autism spectrum, including those with co-occurring intellectual disabilities, is represented in this study, the sample is small and includes predominately White males from well-educated, middle-class families. Thus, the generalizability of our findings is limited. Given the small sample size, this study also lacks the statistical power to robustly test interaction terms. In addition, the impact of SLEs may not be evenly distributed among demographic groups, and more research is needed to address the impact of intersectional identities and experiencing an SLE as it relates to resilience in autistic children. A strength of the ARSE method is that each participant acts as their own baseline but it is not possible to determine that a causal relationship exists between SLEs and changes in mental health symptoms. Many other factors may be contributing to changes in mental health symptoms in this sample and should be studied in the context of experiencing an SLE. We advise caution in interpreting any causal relations between the SLE and mental health outcomes. Ultimately, the extent to which changes in internalizing and externalizing symptoms can be directly attributed to SLEs remains unknown; however, this is not unlike other research examining the impact of life events on mental health given that this question cannot be answered using experimental designs for ethical reasons. Another limitation is that there are likely normative developmental changes in internalizing and externalizing symptoms that the ARSE analysis is picking up on. This is a limitation of the analysis generally and future researchers should examine ways to account for potential confounds in this analysis like developmental trends. We also chose not to include children who did not experience a SLE in the ARSE analysis. While this was a theoretical choice reflecting our working definition of resilience that requires some experience of adversity, it does limit our ability to make comparative conclusions. It should also be noted that parent reports of SLEs and child mental health in the current study could be impacted by parent’s mental state. Finally, this was also an exploratory study without a priori hypotheses and results need to be replicated in larger, more diverse samples.

This study introduces an innovative methodological approach to understanding the effects of stressful life events on the mental health of autistic children. Results suggest significant variability in the resilience process in autistic children, with some evidence that younger children for whom the life events are perceived as a significant family stressor have less resilient mental health outcomes. Given that autistic children are at increased risk for experiencing stressful life events (Berg et al., 2016; Rigles, 2017) as well as the lack of research examining the impact of stressful life events on the mental health of autistic children over time, this study fills a significant gap in the literature. The current study provides initial empirical, longitudinal evidence on the individual differences present in the resilience process amongst autistic children. Additional research that focuses on factors related to autistic children’s positive adaptation to stress has the potential to inform person-specific intervention strategies.

Acknowledgments

Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health [T32HD007489; U54HD090256], the National Institute of Mental Health [R01MH099190 to S. Hartley], and the University of Wisconsin-Madison. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Funding sources had no involvement in the study design, the collection, analysis, or interpretation of the data, the writing of the report, or the decision to submit the study for publication.

We would like to thank the autistic children and their families for their willingness to participate in this study. We would also like to thank members of the Hartley Lab and Neurodiversity Lab for their thoughtful suggestions and recommendations.

We have no conflicts of interest to disclose.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper

References

  1. Achenbach TM, & Rescorla LA (2001). Manual for the ASEBA school-age forms & profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, and Families. [Google Scholar]
  2. Anderson LK (2023). Autistic experiences of applied behavior analysis. Autism, 27(3), 737–750. [DOI] [PubMed] [Google Scholar]
  3. Berg KL, Shiu C-S, Acharya K, Stolbach BC, Msall ME (2016). Disparities in adversity among children with autism spectrum disorder: A population-based study. Developmental Medicine & Child Neurology, 58(11), 1124–1131. 10.1111/dmcn.13161 [DOI] [PubMed] [Google Scholar]
  4. Bekhet AK, Johnson NL, & Zauszniewski JA (2012). Effects on resilience of caregivers of persons with autism spectrum disorder: The role of positive cognitions. Journal of the American Psychiatric Nurses Association, 18(6), 337–344. 10.1177/10783903124670 [DOI] [PubMed] [Google Scholar]
  5. Bishop-Fitzpatrick L, Minshew NJ, Mazefsky CA, & Eack SM (2017). Perception of life as stressful, not biological response to stress, is associated with greater social disability in adults with autism spectrum disorder. Journal of Autism and Developmental Disorders, 47(1), 1–16. 10.1007/s10803-016-2910-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Brenner J, Pan Z, Mazefsky C, Smith KA, Gabriels R (2018). Autism and developmental disorders inpatient research collaborative: Behavioral symptoms of reported abuse in children and adolescents with autism spectrum disorder in inpatient settings. Journal of Autism and Developmental Disorders, 48(11), 3727–3735. 10.1007/s10803-017-3183-4. [DOI] [PubMed] [Google Scholar]
  7. Chapman DP, Whitfield CL, Felitti VJ, Dube SR, Edwards VJ, Anda RF (2004). Adverse childhood experiences and the risk of depressive disorders in adulthood. Journal of Affect Disorders. 82(2), 217–25. 10.1016/j.jad.2003.12.013. [DOI] [PubMed] [Google Scholar]
  8. Cohen S, Murphy MLM, & Prather AA (2019). Ten surprising facts about stressful life events and disease risk. Annual Review of Psychology, 70, 577–597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Constantino JN, & Gruber CP (2012). Social Responsiveness Scale, Second Edition (SRS-2). Torrance, CA: Western Psychological Services. [Google Scholar]
  10. Den Hartigh RJR, & Hill Y (2022). Conceptualizing and measuring psychological resilience: What can we learn from physics? New Ideas in Psychology, 66, 100934. 10.1016/j.newideapsych.2022.100934 [DOI] [Google Scholar]
  11. Dodds RL (2020). An exploratory review of the associations between adverse experiences and autism. Journal of Aggression Maltreatment & Trauma, 30(2), 1–20. 10.1080/10926771.2020.1783736 [DOI] [Google Scholar]
  12. Ellis BJ, & Boyce WT (2008). Biological sensitivity to context. Current Directions in Psychological Science, 17(3), 183–187. [Google Scholar]
  13. Elsabbagh M (2020). Linking risk factors and outcomes in autism spectrum disorder: Is there evidence for resilience? BMJ, l6880. 10.1136/bmj.l6880 [DOI] [PubMed] [Google Scholar]
  14. Grant KE, Compas BE, Thurm AE, McMahon SD, & Gipson PY (2004). Stressors and child and adolescent psychopathology: Measurement issues and prospective effects. Journal of Clinical Child & Adolescent Psychology, 33(2), 412–425. 10.1207/s15374424jccp3302_23 [DOI] [PubMed] [Google Scholar]
  15. Grant KE, Compas BE, Thurm AE, McMahon SD, Gipson PY, Campbell AJ, … & Westerholm RI (2006). Stressors and child and adolescent psychopathology: Evidence of moderating and mediating effects. Clinical psychology review, 26(3), 257–283. [DOI] [PubMed] [Google Scholar]
  16. Hammen C, & Rudolph KD (1996). Childhood depression. In Mash EJ, & Barkley RA (Eds.), Child psychopathology (pp. 153 – 195). New York, NY; Guilford Press. [Google Scholar]
  17. Hedley D, Uljarević M, Bury SM, & Dissanayake C (2019). Predictors of mental health and well-being in employed adults with autism spectrum disorder at 12-month follow-up. Autism Research, 12(3), 482–494. 10.1002/aur.2064 [DOI] [PubMed] [Google Scholar]
  18. Hollocks MJ, Lerh JW, Magiati I, Meiser-Stedman R, & Brugha TS (2019). Anxiety and depression in adults with autism spectrum disorder: A systematic review and meta-analysis. Psychological Medicine, 49(4), 559–572. 10.1017/s0033291718002283 [DOI] [PubMed] [Google Scholar]
  19. Hollocks MJ, Meiser-Stedman R, Kent R, Lukito S, Briskman J, Stringer D, Lord C, Pickles A, Baird G, Charman T, & Simonoff E (2021). The association of adverse life events and parental mental health with emotional and behavioral outcomes in young adults with autism spectrum disorder. Autism Research, 14(8), 1724–1735. https://doi.org/ 10.1002/aur.2548 [DOI] [PubMed] [Google Scholar]
  20. Jacobs D, Steyaert J, Dierickx K, & Hens K (2020). Parents’ views and experiences of the autism spectrum disorder diagnosis of their young child: A longitudinal interview study. European Child & Adolescent Psychiatry, 29(8), 1143–1154. 10.1007/s00787-019-01431-4 [DOI] [PubMed] [Google Scholar]
  21. Kaboski J, McDonnell CG, Valentino K (2017). Resilience and autism spectrum disorder: Applying developmental psychopathology to optimal outcome. Review Journal of Autism and Developmental Disorders, 4(3), 175–189. 10.1007/s40489-017-0106-4 [DOI] [Google Scholar]
  22. Kerns CM, Renno P, Storch EA, Kendall PC (2017). Anxiety in children and adolescents with autism spectrum disorder: Evidence based Assessment and Treatment. Academic Press. [Google Scholar]
  23. Lai M-C, & Szatmari P (2019). Resilience in autism: Research and practice prospects. Autism, 23(3), 539–541. 10.1177/1362361319842964 [DOI] [PubMed] [Google Scholar]
  24. Leno CV, Wright N, Pickles A, Bedford R, Zaidman-Zait A, Kerns C, … & P., Elsabbagh M (2022). Exposure to family stressful life events in autistic children: Longitudinal associations with mental health and the moderating role of cognitive flexibility. Autism, 26(7), 1656–1667. 10.1177/13623613211061932 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu RT (2015). A developmentally informed perspective on the relation between stress and psychopathology: When the problem with stress is that there is not enough. Journal of Abnormal Psychology, 124(1), 80–92. 10.1037/abn0000043 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Makin C, Hill V, & Pellicano E (2017). The primary-to-secondary school transition for children on the autism spectrum: A multi-informant mixed-methods study. Autism & Developmental Language Impairments, 2, 2396941516684834. [Google Scholar]
  27. Masten AS, & Barnes AJ (2018). Resilience in children: Developmental perspectives. Children, 5, 98–112. 10.3390/children5070098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Masten AS, Lucke CM, Nelson KM, & Stallworthy IC (2021). Resilience in development and psychopathology: Multisystem Perspectives. Annual Review of Clinical Psychology, 17(1), 521–549. 10.1146/annurev-clinpsy-081219-120307 [DOI] [PubMed] [Google Scholar]
  29. Mazurka R, Wynne-Edwards KE, & Harkness KL, (2017). Sex differences in the cortisol response to the trier social stress test in depressed and nondepressed adolescents. Clinical Psychological Science, 6, 301–314. [Google Scholar]
  30. McDonnell CG, Boan AD, BraSLEy CC, Seay KD (2018). Child maltreatment in autism spectrum disorder and intellectual disability: Results from a population based sample. Journal of Child Psychology and Psychiatry, 60(5). 10.1111/jcpp.12993 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. McLaughlin KA, Hatzenbuehler ML, & Phil M (2009). Mechanisms linking stressful life events and mental health problems in a prospective, community-based sample of adolescents. Journal of Adolescent Health, 44(2), 153–160. 10.1016/j.jadohealth.2008.06.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Moore SE, Norman RE, Suetani S, Thomas HJ, Sly PD, & Scott JG (2017). Consequences of bullying victimization in childhood and adolescence: A systematic review and meta-analysis. World Journal of Psychiatry, 7(1), 60–76. 10.5498/wjp.v7.i1.60 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Nuske HJ, McGhee Hassrick E, Bronstein B, Hauptman L, Aponte C, Levato L, … & Smith T (2019). Broken bridges—new school transitions for students with autism spectrum disorder: A systematic review on difficulties and strategies for success. Autism, 23(2), 306–325. [DOI] [PubMed] [Google Scholar]
  34. Obradović J, Shaffer A, & Masten AS (2012). Risk and adversity in developmental psychopathology. The Cambridge Handbook of Environment in Human Development, 35–57. 10.1017/cbo9781139016827.004 [DOI] [Google Scholar]
  35. Pandolfi V, Magyar CI, & Norris M (2014). Validity study of the CBCL 6–18 for the assessment of emotional problems in youth with ASD. Journal of Mental Health Research in Intellectual Disabilities, 7(4), 306–322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Platt R, Williams SR, & Ginsburg GS (2016). Stressful life events and child anxiety: Examining parent and child mediators. Child Psychiatry & Human Development, 47, 23–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Ratcliff NJ, Mahoney-Nair DT, & Goldstein JR (2019). The area of resilience to stress event ({arse}): A new method for quantifying the process of resilience. The Quantitative Methods for Psychology, 15(2), 148–173. 10.20982/tqmp.15.2.p148 [DOI] [Google Scholar]
  38. Rigles B (2017). The relationship between adverse childhood events, resiliency and health among children with autism. Journal of Autism and Developmental Disorders, 47(1), 187–202. 10.1007/s10803-016-2905-3 [DOI] [PubMed] [Google Scholar]
  39. Richter M, Flavier E, Popa-Roch M, & Clément C (2020). Perceptions on the primary-secondary school transition from French students with Autism Spectrum Disorder and their parents. European Journal of Special Needs Education, 35(2), 171–187. [Google Scholar]
  40. Rosanbalm KD, & Murray DW (2017). Caregiver Co-regulation Across Development: A Practice Brief. OPRE Brief #2017–80. Washington, DC: Office of Planning, Research, and Evaluation, Administration for Children and Families, US. Department of Health and Human Services. [Google Scholar]
  41. Scarpa A, Swain DM, Factor RS, Dahiya AV, & Bertollo JR (2021). Enhancing flexibility: A biosocial model for resilience to adversity in youth with autism. SAGE Open, 11(3), 21582440211037997. [Google Scholar]
  42. Schilling EA, Aseltine RH, Gore S (2007). Adverse childhood experiences and mental health in young adults: a longitudinal survey. BMC Public Health, 7,30. 10.1186/1471-2458-7-30 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Shapero BG, Hamilton JL, Stange JP, Liu RT, Abramson LY, & Alloy LB (2015). Moderate childhood stress buffers against depressive response to proximal stressors: A multi-wave prospective study of early adolescents. Journal of Abnormal Child Psychology, 43, 1403–1413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Stallworthy IC, & Masten AS (2023). Advancing Research on early autism through an integrated risk and Resilience Perspective. Development and Psychopathology, 35(1), 44–61. 10.1017/S0954579421001437 [DOI] [PubMed] [Google Scholar]
  45. Szatmari P, Georgiades S, Duku E, Bennett TA, Bryson S, Fombonne E, … & Thompson A (2015). Developmental trajectories of symptom severity and adaptive functioning in an inception cohort of preschool children with autism spectrum disorder. JAMA Psychiatry, 72(3), 276. 10.1001/jamapsychiatry.2014.2463 [DOI] [PubMed] [Google Scholar]
  46. Taylor JL, & Gotham KO (2016). Cumulative life events, traumatic experiences, and psychiatric symptomatology in transition-aged youth with autism spectrum disorder. Journal of Neurodevelopmental Disorders, 8. 10.1186/s11689-016-9160-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. van Buuren S, & Groothuis-Oudshoorn K (2011). mice: Multivariate Imputation by Chained Equations in R. Journal of Statistical Software, 45(3), 1–67. 10.18637/jss.v045.i03 [DOI] [Google Scholar]
  48. Wright MO, Masten AS, & Narayan AJ (2013). Resilience processes in development: Four waves of research on positive adaptation in the context of adversity. In Goldstein S & Brooks RB (Eds.), Handbook of resilience in children (pp. 15–37). Springer Science + Business Media. 10.1007/978-1-4614-3661-4_2 [DOI] [Google Scholar]

RESOURCES