App-based meditation habits maintain reductions in depression symptoms among autistic adults

Abstract

Mobile app-based meditation is an effective self-care solution for depression, but limited evidence exists for the long-term benefits among autistic adults, who are at increased risk of experiencing depression. The primary goal of this study was to examine the feasibility and effectiveness of incorporating habit training into an app-based meditation intervention to maintain reductions in depressive symptoms among autistic adults. Participants were randomized to an App Only (who only received access to the meditation app), App + Habit Training (who received the meditation app and instructions for anchoring their app-based meditation to an existing routine), or a waitlist control group. All participants completed questionnaires at baseline, post-intervention, and 6 months post-intervention, and responded to SMS ecological momentary assessments regarding their meditation practice during the 8-week intervention and an 8-week follow-up period. The results show that the App + Habit Training group reported significantly more days with meditation per week during and after the intervention (p < 0.05) and also experienced a larger decrease in depression symptoms post-intervention and 6 months later (p < 0.05) relative to the control group. These findings demonstrate that app-based meditation habits are an effective self-care solution for autistic adults with depression, and anchoring is a promising strategy for establishing other healthy habits among autistic adults.

Lay Abstract

Existing research has identified an increased risk of depression among autistic adults, which can negatively impact their adaptive functioning abilities and socioeconomic outcomes. Mobile app-based meditation is a feasible, accessible, and effective self-care solution for depression among neurotypical adults, but there is limited evidence for the long-term benefits of app-based meditation among autistic adults. Habits are a key behavioral strategy for maintaining behavior change, and anchoring is one effective habit formation intervention that has yet to be tested among autistic adults. This study demonstrates that it is both feasible and effective to integrate the anchoring habit formation strategy into an app-based meditation intervention for establishing meditation habits among autistic adults. In addition, the study shows that app-based meditation habits were successful at maintaining reductions in depressive symptoms over 6 months. These results demonstrate the power of anchoring-based habit formation interventions for establishing healthy habits among autistic adults, which offers a promising behavioral intervention technique for establishing other healthy habits among autistic adults. The study also shows that app-based meditation habits are an effective long-term self-care solution for managing depressive symptoms among autistic adults that should be used by mental health providers and policymakers. Future research should test this combined anchoring and app-based meditation intervention technique among larger samples of autistic adults and over longer durations to better understand the mechanisms underlying the success of this intervention.

Keywords

anchoring depression meditation mindfulness mobile health

Adults with autism spectrum disorder (ASD) have an increased risk of experiencing depression, with lifetime prevalence rates that are three to four times higher than typically developing adults (Hudson et al., 2019). Comorbid depression in autistic adults can exacerbate existing difficulties with adaptive functioning including socialization, emotion regulation, and executing daily tasks (Kraper et al., 2017; Smith & White, 2020; Wallace et al., 2016). The compounding effects of depression and ASD on adaptive functioning can have significant socioeconomic consequences, including lower rates of post-secondary education, employment, and independent living (Kraper et al., 2017; Levy & Perry, 2011). Despite increased efforts to address these disparities (Bishop-Fitzpatrick et al., 2013), there is still limited accessibility to effective mental health services for autistic adults (Adams & Young, 2021), highlighting the need to develop accessible and cost-effective interventions for reducing depressive symptoms among autistic adults that can complement in-person mental health services.

Emerging research has highlighted the potential of mindfulness meditation for improving comorbid depression and mental health quality of life in autistic adults (Braden et al., 2022; Kiep et al., 2015; Pagni et al., 2020; Sizoo & Kuiper, 2017; Spek et al., 2013; White et al., 2018). Mindfulness meditation is a practice defined as a conscious awareness and nonjudgmental acceptance of the present moment (Kabat-Zinn, 1994), and has been linked to improvements in depressive symptoms and reduced risk of depression recurrence in both clinical and nonclinical populations (Gu et al., 2015; Querstret et al., 2020). Recently, there has been increased interest in the use of mobile applications (i.e. apps) to deliver mindfulness meditation interventions. Several randomized controlled trials and meta-analyses have demonstrated the efficacy of mindfulness meditation apps for improving a wide range of mental health outcomes in both clinical and nonclinical populations, and these tools have demonstrated similar effect sizes to other face-to-face or app-delivered psychosocial interventions (Champion et al., 2018; Economides et al., 2018; Gál et al., 2021; Huberty et al., 2021). Meditation apps offer several advantages over traditional in-person interventions, such as ease of use, affordability, and accessibility, making them a feasible and cost-effective alternative to in-person interventions (Gál et al., 2021; Spijkerman et al., 2016). For autistic adults, the convenience of meditation apps may be particularly helpful for overcoming commonly faced barriers such as financial costs, insurance coverage, social challenges, and transportation (K. B. Beck et al., 2020; Zheng et al., 2021). Furthermore, meditation apps can easily be tailored to fit an individual’s specific needs (e.g. duration or frequency of sessions), offering greater flexibility than existing mindfulness-based interventions for autistic adults (K. B. Beck et al., 2020; Lunsky et al., 2022).

While app-based mindfulness meditation interventions offer a potentially effective and accessible mental health self-care solution for autistic adults with depression, the existing research examining the feasibility and long-term effectiveness of mobile apps for reducing depression among autistic adults is limited (Kim et al., 2018). Furthermore, adherence and retention appear to be major limiting factors in app-based studies targeting autistic populations. A recent study by Hartley et al. (2022) reported several significant challenges when delivering a commercially available mindfulness meditation app in autistic adults. Specifically, participants reported difficulties with concentration and motivation during meditation sessions, as well as problems with scheduling a regular time for their meditation practice. Importantly, poor adherence is a common issue in app-based mindfulness meditation interventions across all populations, with one meta-analysis of 34 randomized controlled trials reporting average adherence and attrition rates of 43% and 31%, respectively (Gál et al., 2021). Even outside of intervention settings, many existing commercial mobile health apps report a sharp decline in user engagement following initiation, with one study reporting median 15- and 30-day app retention rates of 3.9% and 3.3%, respectively (Baumel et al., 2019; Vaghefi & Tulu, 2019). Similar to other healthy behaviors, the benefits of mindfulness meditation require long-term, persistent practice (Athanas et al., 2019; Carmody & Baer, 2008; Engen et al., 2018; Lardone et al., 2018; Linardon & Fuller-Tyszkiewicz, 2020), emphasizing the need for novel strategies to promote both intervention adherence and long-term engagement with meditation apps, particularly among autistic adults.

One potential strategy for maintaining meditation app use is through the development of a meditation habit. Habits are characterized as automatic behavioral responses that are formed gradually over time as a result of repeatedly and consistently performing a desired behavior in response to the same contextual cue (Gardner, 2015; Wood & Neal, 2016). Importantly, habits have been shown to persist despite declining motivation or distractions (Galla & Duckworth, 2015; Gardner, Lally, & Wardle, 2012; Neal et al., 2013). One commonly used habit formation strategy is to pair the desired behavior to an existing daily routine, called “anchoring” (Gardner, Lally, & Wardle, 2012; Lally et al., 2011; Lally & Gardner, 2013). For example, “After I get dressed and wash my face in the morning, I will practice meditation for 10 minutes.” Anchoring has been used to successfully establish habits for various health-promoting behaviors including flossing (Judah et al., 2013), nutrition behavior (Keller et al., 2021), physical activity (Fleig et al., 2017; Keller et al., 2017; Pimm et al., 2016), and medication adherence (Brooks et al., 2014; Stecher, Mukasa, & Linnemayr, 2021). Moreover, a recent study of mindfulness meditation app users found that those who were instructed to anchor their mindfulness meditation practice to an existing behavior maintained greater meditation with the app over time, compared with those given no anchoring instruction (Stecher, Sullivan, & Linnemayr, 2021).

Anchoring a meditation practice to an existing routine may be particularly effective for autistic adults compared with other habit formation strategies for several reasons. First, restricted, repetitive patterns of behavior (RRB) are diagnostic criteria of ASD and in some cases manifest as an insistence on sameness, preference for routines, and ritualized behaviors (Cooper et al., 2022). Although RRB can make it difficult to add a new behavior to a routine, over time, the need to maintain a routine may also facilitate long-term behavior maintenance (Nichols et al., 2019). We hypothesize that autistic adults with these types of RRB may be more likely to form a meditation habit through anchoring, because routines are known to underlie the maintenance of health behaviors in other settings (Brooks et al., 2014; Stawarz et al., 2020). In addition, habits are triggered unconsciously, which reduces the cognitive demand of performing a behavior (Gardner, Lally, & Wardle, 2012). This may be especially helpful for overcoming difficulties with executive functioning that are associated with both autism and depression (Smith & White, 2020).

Given the limited evidence for the long-term benefits of app-based mindfulness meditation among autistic adults, the primary goal of the present study was to examine the feasibility and effectiveness of incorporating habit training into an app-based mindfulness meditation intervention to reduce depressive symptoms among autistic adults. We hypothesized that the addition of anchoring-based habit training would increase adherence to the prescribed 10 min per day mindfulness meditation practice using the Ten Percent Happier app, a commercially available meditation app, during an 8-week intervention that also included daily SMS text reminder messages. A secondary goal of this study was to investigate whether meditation app use supported by habit training would enhance the beneficial effects of mindfulness meditation (hereafter referred to as meditation) on depression symptoms in autistic adults over time, relative to those who did not receive habit training. We hypothesized that those who received anchoring-based habit training would experience greater improvements in depressive symptoms following the 8-week intervention, and that these greater improvements would be maintained over the following 6 months through stronger meditation app habits.

Method

Participants and procedure

The Arizona State University Institutional Review Board reviewed and approved all aspects of this study (STUDY00012092) and all participants provided electronic consent prior to participation. Participants were recruited primarily through the Simons Foundation Powering Autism Research for Knowledge (SPARK) Research Match program but also through flyers and online advertisements distributed by local autism research centers and support groups in the United States of America. Interested individuals were contacted by a member of the study team via telephone to complete an initial eligibility screening, which included questions about demographics (i.e. age, sex, race/ethnicity, education level), ASD diagnosis, medical history (i.e. additional diagnoses, psychotropic and supplemental drug use, history of seizures and/or head trauma), and a general health screening. During the phone interview, a member of the study team administered the Wide Range Achievement Test (WRAT) (Wilkinson & Robertson, 2017) and the Social Responsiveness Scale, Second Edition (SRS-2) (Constantino & Gruber, 2012) to estimate reading ability and autism symptom severity, respectively. To participate in the study, individuals had to be at least 18 years old, able to read and speak English, reported an ASD diagnosis, and met the ASD threshold on the SRS-2 (T-score > 59). Nonverbal individuals were excluded from the study to ensure test compliance and increase sample homogeneity. Individuals with reading scores <70 (WRAT < 70) were excluded to minimize variability due to general cognitive functioning. Individuals were also excluded if they reported any major medical illnesses, history of seizures, or head trauma with loss of consciousness, as these health factors can confound behavioral assessments. Participants were offered monetary compensation for their participation.

Out of the 224 individuals screened for the study, 125 were eligible and agreed to participate in the study. Once enrolled, participants were randomly assigned to one of three intervention groups, which are described in more detail below: (1) the Ten Percent Happier App group (App Only; n = 42); (2) the Ten Percent Happier App plus habit training (App + Habit Training; n = 41); or (3) a waitlist control (WLC) group (n = 42). Self-report questionnaire data were collected via online Qualtrics surveys at baseline, immediately following the 8-week intervention, and 6 months after the intervention. All participants also completed SMS ecological momentary assessments (EMAs) at the end of every week of the 8-week intervention and an 8-week follow-up period that assessed both the number of days with any meditation and the average duration of meditation over the past week.

Intervention

This study tested an app-based meditation intervention using the Ten Percent Happier app, a commercially available mindfulness meditation app. This app was chosen on the basis of (1) greater concrete language usage relative to other applications (which may rely more heavily on abstract/metaphorical language) as originally adapted for in-person meditation training in adults with ASD by Spek et al. (2013), (2) the practices predominantly originated from the Vipassana tradition, which has demonstrated efficacy in ASD (Pagni et al., 2020, 2023; Sizoo & Kuiper, 2017; Spek et al., 2013), and (3) technical and research support provided by Ten Percent Happier. All participants received free access to the Ten Percent Happier app for 14 months (8-week intervention period plus 12-month follow-up period). Participants randomized to the Ten Percent Happier (App Only) group were immediately given access to the app and asked to meditate for at least 10 min every day during the 8-week intervention period. Prior to the start of the intervention, participants in the App Only group watched a brief video presentation which provided education regarding the habit formation process and benefits of meditation, without instructions on how to develop a habit. The App Only group also received daily SMS reminders to meditate for at least 10 min per day. Participants randomized to the Ten Percent Happier plus habit training (App + Habit Training) group were also given access to the app at the start of the study and asked to meditate for at least 10 min every day during the 8-week intervention period. Prior to the start of the intervention period, participants in the App + Habit Training group also watched a brief video presentation which provided education on the benefits of habits and the habit formation process, as well as instructions on how to anchor their daily meditation practice to an existing routine. Specifically, the App + Habit Training group was instructed to complete their new meditation habit after leaving the bathroom in the morning upon waking. To reinforce this anchoring plan, the App + Habit Training group received daily SMS reminders to meditate for at least 10 min after their existing routine (cue) during the 8-week intervention that read: “Remember to meditate for at least 10 min as your first activity after going to the bathroom in the morning tomorrow.” Finally, participants in the WLC group were given access to the Ten Percent Happier app after the 8-week wait period. In addition to receiving app access after 8 weeks, the WLC group was randomized to either the App Only or App + Habit Training interventions and received 8 weeks of SMS reminder messages. In addition, at the beginning of the intervention period all groups completed weekly EMAs sent via SMS message for 16 weeks to assess meditation performance during the intervention and maintenance after the intervention. Those in the WLC group received an additional 8 weeks of SMS EMAs that corresponded to the start of their 8-week intervention period.

Measures

Depressive symptoms

Depressive symptoms were assessed using the Beck Depression Inventory-II (BDI-II) at baseline, 8 weeks (post-intervention), and 6 months (follow-up). Developed based on the Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV), BDI-II is a 21-item self-report measure of depressive symptoms over the past 2 weeks (A. T. Beck et al., 1996). BDI-II has demonstrated high reliability and validity in previous research and has been used in both autistic and typically developed populations (Wang & Gorenstein, 2013; Williams et al., 2021). Scores on BDI-II can range from 0 to 63, with higher scores indicating greater symptom severity. BDI-II scores of 0–10 indicate typical ups and downs in mood; 11–16 = mild mood disturbance; 17–20 = borderline clinical depression; 21–30 = moderate depression; 31–40 = severe depression; and 40+ = extreme depression.

Health status

At baseline, participants were asked to self-report any diagnoses provided by a medical professional (i.e. psychiatrist, psychologist, or counselor). Those who reported receiving either an anxiety or depression diagnosis were noted as having anxiety and/or depression. Participants also reported on current prescription use, the number of nutritional supplements they were taking, and whether they were suffering from any physical health conditions at the time of each survey assessment.

Habit strength

Habit strength was assessed using a modified version of the Self-Report Behavior Automaticity Index (SRBAI) at baseline, 8 weeks (post-intervention), and 6 months (follow-up). The SRBAI is a validated and reliable four-item self-report measure of behavior automaticity frequently used to assess habit strength (Gardner, Abraham, et al., 2012). Participants are asked to rate their degree of agreement with four statements starting with “Daily meditation is something . . .” (e.g. “I do automatically”; “I do without thinking”) using a 5-point Likert-type scale, with higher scores indicating higher levels of behavioral automaticity.

Meditation app use and maintenance

Participants’ app-based meditation during the intervention was measured using EMA questions that assessed (1) the number of days with any meditation and (2) the average duration of meditation sessions in minutes per week. Participants received mobile EMA questions once per week (Sundays at 6:00 p.m. in their respective time zone) and provided numerical SMS responses for question (1) with acceptable responses of 0–7 and question (2) with acceptable responses of 0–120. The number of days of meditation per week was our primary measure of meditation behavior, and we multiplied the number of days with meditation by the average duration of meditation sessions to estimate the total minutes of meditation per week. These two measures, that is, the number of days with any meditation and the total minutes of meditation per week, were also averaged over the 8-week intervention and 8 weeks post-intervention.

Statistical analyses

Sample characteristics were described through the mean and standard deviation (for continuous variables) and percentage and count (for binary variables). To assess the statistical significance of differences in each characteristic between study groups, analysis of variance (ANOVA) tests were used for continuous variables and the nonparametric Kruskal–Wallis equality-of-populations rank test was used for binary variables. The weekly measures of participants’ meditation behavior were analyzed using panel regression models that were estimated using heteroskedasticity-robust standard errors and participant-level random effects. Specifically, we estimated Poisson panel regression models for the number of days with any meditation per week separately over the 8-week intervention and the 8-week post-intervention period. We estimated ordinary least squares (OLS) panel regression models for the total minutes of meditation per week, which were also separately estimated over the 8-week intervention and the 8-week post-intervention period. Study groups are included in each of these models through binary variables that equal one if the participant was randomized to the indicated group; the control group is typically excluded from the model as the reference group. To analyze how the interventions were differentially experienced by those with an anxiety and/or depression diagnosis at the start of the study, an additional set of interaction terms between the study group binary variables and a binary variable equal to one for those with an anxiety and/or depression diagnosis were included in the regression models.

Finally, to analyze changes in depressive symptoms and habit strength over time, we applied the logarithm transformation to the BDI-II and SRBAI scores so they would have an approximately normal distribution. Then separate OLS regressions were estimated on the logarithm of BDI-II and SRBAI at 8 weeks (post-intervention) and at 6 months (follow-up), and both models included the logarithm of participants’ baseline BDI-II and SRBAI scores and the study group binary variables as covariates. All regression models were first estimated without additional measures of participants’ demographics and health status and then estimated after including these additional covariates to test the sensitivity of our results. Although our study groups were balanced on observable characteristics at baseline, including these additional covariates in the model helps to increase our statistical power for detecting study group-level differences, so the full models are our preferred specifications. The estimated coefficients and 95% confidence intervals are reported to assist with interpreting statistical significance, and all statistical analyses were performed using Stata/MP 16.1.

Community involvement

As a part of the SPARK Research Match program, our project was approved by the SPARK Participant Access Committee, which included autistic adults and parents of children with autism. One autistic graduate student contributed to the implementation of the study and is thanked in the acknowledgments. One community autism center leader was involved in recruitment and is thanked in the acknowledgements.

Results

Participant characteristics for the full sample and within each study group are presented in Table 1. These results show that the sample was majority female (83/125; 66.4%), under the age of 40 years (80/125; 64%), White (95/125; 76%), reported at least one physical health condition (98/125; 81.7%), and reported a diagnosis of anxiety and/or depression (68/125; 53.1%). Our measure of current prescriptions suffered from item nonresponse and is not included in our analyses. There were no significant group differences in demographic characteristics or medical history across the three study groups (WLC, App Only, and App + Habit Training), as evidenced by results from the nonparametric Kruskal–Wallis equality-of-populations rank tests.

Table 1.

Sample characteristics.

Variable	Full sample		Waitlist control		App Only		App + Habits		All groups equal(p values)^a
Variable	n	%	n	%	n	%	n	%	All groups equal(p values)^a
Sex
Female	83	66.4	25	59.5	33	78.6	25	61.0	0.189
Age (years)
18–29	47	37.6	16	38.1	17	40.5	14	34.2	0.746
30–39	33	26.4	11	26.2	10	23.8	12	29.3	0.911
40–49	25	20.0	11	26.2	4	9.5	10	24.4	0.132
50+	20	16.0	4	9.5	10	23.8	6	14.6	0.172
Race
Asian	8	6.6	3	7.1	5	11.9	0	0.0	0.081
White	95	76.0	32	76.2	27	64.3	36	87.8	0.054
Black	2	1.6	0	0.0	2	4.8	0	0.0	0.136
Hispanic	6	4.8	1	2.4	3	7.1	2	4.9	0.608
Other race(s)	11	8.8	4	9.5	4	9.5	3	7.3	0.897
Education
⩽High school	18	14.4	9	21.4	4	9.5	5	12.2	0.268
Some college	33	26.4	11	26.2	9	21.4	13	31.7	0.571
Bachelor’s degree	36	28.8	11	26.2	14	33.3	11	26.8	0.729
Graduate degree	38	30.4	11	26.2	15	35.7	12	29.3	0.628
Medical history
Takes any supplement(s)	70	57.9	25	59.5	25	59.5	20	48.8	0.434
Any physical health condition(s)	98	81.7	37	88.1	34	80.9	27	65.9	0.145
Anxiety and/or depression	68	53.1	27	64.3	21	50.0	20	48.8	0.298
Sample size	125		42		42		41

The nonparametric Kruskal–Wallis equality-of-populations rank test was used to test for group-level differences in each characteristic.

Table 2 compares our main outcomes between each study group at baseline and over time. Specifically, we conducted an ANOVA to examine group-level differences in BDI-II and SRBAI scores at baseline, post-intervention (week 8), and follow-up (6 months post-intervention). The results show that there were no significant differences in the BDI-II or SRBAI scores between groups at baseline (p > 0.05). However, there were significant differences in the BDI-II and SRBAI scores between groups at week 8 (p < 0.05), with the App + Habit Training group reporting the lowest BDI-II (M = 7.6, SD = 8.7) and the highest SRBAI scores (M = 11.1, SD = 4.1) as compared with the WLC and App Only groups. The BDI-II scores remained the lowest in the App + Habit Training group and SRBAI remained the highest in the App + Habit Training group after 6 months, but the differences in SRBAI between all three study groups were not statistically significant.

Table 2.

Group-level differences in outcome variables over time.

Variable	Full sample		Waitlist control		App Only		App + habit training		All groups equal (p values)^a
Variable	M	SD	M	SD	M	SD	M	SD	All groups equal (p values)^a
BDI-II
Baseline	16.8	11.8	17.9	11.3	14.6	11.8	17.8	12.3	0.369
Week 8	11.0	11.1	14.6	12.1	10.3	11.3	7.6	8.7	0.046
Month 6	11.4	9.6	13.7	10.3	11.9	9.7	8.8	8.7	0.041
SRBAI
Baseline	9.9	4.6	10.6	4.2	10.1	5.2	9.1	3.7	0.404
Week 8	10.1	4.4	10.4	4.7	8.9	4.5	11.1	4.1	0.025
Month 6	11.5	4.4	10.9	3.5	10.2	5.1	12.5	5.0	0.053
Days with any meditation
Intervention	4.7	0.8	4.1	0.3	4.5	0.8	5.7	0.2	<0.001
Follow-up	4.7	0.8	4.1	0.1	4.1	0.8	5.7	0.2	<0.001
Weekly meditation minutes
Intervention	66.7	47.4	57.1	64.0	66.6	33.4	76.1	36.0	<0.001
Follow-up	66.4	71.1	69.1	105.3	49.8	41.1	77.4	43.7	<0.001

BDI-II: Beck Depression Inventory-II; SRBAI: Self-Report Behavioral Automaticity Index.

An ANOVA was used to test for group-level differences in depressive symptoms and habit strength at baseline, 8 weeks (post-intervention), and 6 months (follow-up).

Statistical significance of the differences in outcome measures across all three study groups.

We also examined group-level differences in reported meditation behavior (days with any meditation and weekly meditation minutes) during the 8-week intervention and 8-week follow-up period using an ANOVA. There were significant group-level differences in average number of days with any meditation (p < 0.001) and the average weekly meditation minutes (p < 0.001) during both the 8-week intervention and 8-week follow-up periods. During the intervention period, the App + Habit Training group reported the highest average days with any meditation (M = 5.7, SD = 0.2) and average weekly meditation minutes (M = 76.1, SD = 36.0), relative to the WLC and App Only groups. We observed a similar pattern during the 8-week follow-up period, with the App + Habit Training group reporting the highest average number of days with any meditation (M = 5.7, SD = 0.2) and average weekly meditation minutes (M = 77.4, SD = 43.7) relative to the WLC and App Only groups. The results of these ANOVA tests as well as means and standard deviations for all outcomes variables are summarized in Table 2.

Estimated treatment effects on frequency of meditation sessions

Figure 1 displays the average number of days with any meditation per week between the three study groups during the 8-week intervention and 8-week follow-up periods. This figure shows that the App + Habit Training group had the highest average number of days with meditation during the intervention. Importantly, the App + Habit Training group was able to maintain that level of meditation during the 8-week follow-up period, while the App Only group saw a large decline in their average number of days with any meditation.

Figure 1.

Mediation days/week over time by group.

To examine study group differences in meditation behavior more rigorously, Poisson regression models were used to estimate the treatment effects on weekly measures of the number of days with any meditation during the 8-week intervention and 8-week follow-up periods while controlling for participant-level characteristics. For both sets of analyses, we first regressed the number of days with meditation per week on just the treatment group identifier variables (Model 1 in Tables 3 and 4), and then we added sociodemographic characteristics (sex, age, education, race/ethnicity, medical history) as additional control variables (Model 2 in Tables 3 and 4). All models also estimated participant-level random effects to capture the influence of both time-varying and time invariant participant characteristics.

Table 3.

Days with meditation per week during the intervention period.

Variable	Days of meditation per week
	(1)		(2)
	IRR	95% CI	IRR	95% CI
App Only	1.388**	[1.088, 1.771]	1.313*	[1.027, 1.680]
App + Habit Training	1.566***	[1.238, 1.981]	1.493***	[1.191, 1.872]
Sex
Female			1.024	[0.858, 1.222]
Age (years)
18–29	Reference
30–39			0.987	[0.798, 1.222]
40–49			0.949	[0.745, 1.208]
50+			1.121	[0.901, 1.395]
Education
⩽High school			0.827	[0.627, 1.091]
Some college			0.958	[0.785, 1.170]
BA/BS			1.003	[0.834, 1.205]
Graduate	Reference
Race/ethnicity
Asian			0.565	[0.208, 1.533]
White			1.058	[0.756, 1.480]
Black			0.900	[0.492, 1.647]
Hispanic			0.835	[0.558, 1.248]
Other race(s)	Reference
Medical history
Takes any supplements			0.899	[0.775, 1.043]
Any physical health conditions			0.929	[0.771, 1.119]
Anxiety and/or depression			0.806**	[0.699, 0.929]
Observations	740		719

Exponentiated Poisson regression model coefficients indicate the incident rate ratio for each outcome. All models estimated participant-level random effects and standard errors were clustered at the participant level. Column (1) displays exponentiated coefficients from models using only treatment group as the predictors. Column (2) displays exponentiated coefficients from models that also included sociodemographic covariates. 95% confidence intervals are listed in brackets. IRR: incident rate ratio.

p < 0.05; **p < 0.01; ***p < 0.001.

Table 4.

Days with meditation per week during the follow-up period.

Variable	Days of meditation per week
	(1)		(2)
	IRR	95% CI	IRR	95% CI
App Only	Reference
App + Habit Training	1.507**	[1.105, 2.055]	1.818***	[1.329, 2.487]
Observations	379		333

IRR: incident rate ratio.

Exponentiated Poisson regression model coefficients indicate the incident rate ratio for each outcome. All models estimated participant-level random effects and standard errors were clustered at the participant level. Column (1) displays exponentiated coefficients from models using only treatment group as the predictors. Column (2) displays exponentiated coefficients from models that included covariates. For brevity, all covariates are not listed. 95% confidence intervals are listed in brackets.

p < 0.05; **p < 0.01; ***p < 0.001.

Table 3 displays the results of the two Poisson regression models predicting the number of days with any meditation per week during the 8-week intervention. In Model 1, participants in the App Only group reported 1.39 times more days with meditation per week (p < 0.01) and the App + Habit Training group reported 1.57 times more days with meditation per week (p < 0.001) compared with the WLC group. After adjusting for sociodemographic variables in Model 2, both the App Only (1.31) and App + Habit Training groups (1.49) reported significantly more days with meditation per week compared with the WLC group (p < 0.05). In addition, there were few significant sociodemographic predictors of the number of days with any meditation, except the participants who reported an anxiety and/or depression diagnosis had significantly lower days with meditation per week during the 8-week intervention (p < 0.05).

Table 4 displays the same set of Poisson regression models predicting the number of days with any meditation per week during the 8-week follow-up period. In Model 1, which only included treatment groups as the predictors, the App + Habit Training group reported 1.51 more days with meditation per week relative to the App Only group (p < 0.01). After adjusting for sociodemographic variables in Model 2, the App + Habit Training group reported 1.82 times more days with meditation per week relative to the App Only group (p < 0.001).

Estimated treatment effects on depressive symptoms

Figure 2 displays the average depressive symptoms in each study group at baseline, following the 8-week intervention, and 6 months after the intervention ended. Depressive symptoms were measured using the BDI-II, and we used a logarithmic transformation so that the distribution of scores would be approximately normal and to improve the interpretability of our regression results. Figure 2 shows that depressive symptoms reduced for all study groups, including the WLC group, during the 8-week intervention. After the intervention ended, the App + Habit Training group maintained their reductions in depressive symptoms, while the App Only group saw an increase in their depressive symptoms over the following 6 months.

Figure 2.

Change in depression over time.

To rigorously investigate the short- and long-term impact of these treatments on depressive symptoms, we ran a series of Poisson regression models to estimate the effects of treatment group on changes in depression following the 8-week intervention and again 6 months later. For both sets of analyses, we first regressed BDI-II scores at week 8 and month 6 on participants’ baseline BDI-II score and treatment groups only (Model 1 in Tables 5 and 6), and we then added sociodemographic characteristics (sex, age, education, race/ethnicity, medical history) as additional control variables (Model 2 in Tables 5 and 6). BDI-II scores were log transformed such that a one-unit change in the predictor variables indicate a percentage change in BDI-II scores.

Table 5.

Changes in depressive symptoms following the 8-week intervention period.

Variable	BDI-II scores
	(1)		(2)
	IRR	95% CI	IRR	95% CI
Baseline BDI–II	0.738***	[0.511, 0.965]	0.614***	[0.417, 0.811]
App Only	−0.131	[−0.539, 0.277]	−0.033	[−0.523, 0.438]
App + Habit Training	−0.409**	[−0.799, −0.020]	−0.534*	[−1.082, 0.014]
Sex
Female			−0.366**	[−0.710, −0.023]
Age
18–29	Reference
30–39			0.569***	[0.177, 0.960]
40–49			0.321	[−0.330, 0.972]
50+			0.542	[−0.121, 1.205]
Education
⩽High school			0.647*	[−0.068, 1.361]
Some college			0.406	[−0.181, 0.994]
BA/BS			0.554*	[−0.025, 1.133]
Graduate	Reference
Race/ethnicity
Asian			−0.202	[−0.938, 0.534]
White			−0.141	[−0.621, 0.339]
Black			−1.369***	[−2.051, −0.687]
Hispanic			−1.401***	[−2.237, −0.565]
Other race(s)	Reference
Medical history
Takes any supplements			−0.283	[−0.725, 0.158]
Any physical health conditions			0.354	[−0.678, 1.387]
Anxiety and/or depression			0.019	[−0.411, 0.448]
Observations	80		70

BDI-II: Beck Depression Inventory-II; IRR: incident rate ratio.

Exponentiated Poisson regression coefficients indicate the incident rate ratio for each outcome. All models estimated participant-level random effects and standard errors were clustered at the participant level. Column (1) displays exponentiated coefficients from models using only baseline depression symptoms and treatment group as the predictors. Column (2) displays exponentiated coefficients from models that also included additional covariates. 95% confidence intervals are listed in brackets.

p < 0.05; **p < 0.01; ***p < 0.001.

Table 6.

Changes in depressive symptoms 6 months after the intervention.

Variable	BDI-II scores
	(1)		(2)
	IRR	95% CI	IRR	95% CI
Baseline BDI-II	0.692***	[0.490, 0.895]	0.504***	[0.191, 0.818]
App Only	Reference
App + Habit Training	−0.382*	[−0.822, 0.059]	−0.437*	−0.915, 0.040]
Observations	51		48

BDI-II: Beck Depression Inventory-II; IRR: incident rate ratio.

Exponentiated Poisson regression coefficients indicate the incident rate ratio for each outcome. All models estimated participant-level random effects and standard errors were clustered at the participant level. Column (1) displays exponentiated coefficients from models using only baseline depression symptoms and treatment group as the predictors. Columns (2) displays exponentiated coefficients from models that also included additional covariates. For brevity, covariates are not listed. 95% confidence intervals are listed in brackets.

p < 0.05; **p < 0.01; ***p < 0.001.

Table 5 shows the results from Poisson regression models predicting changes in BDI-II scores following the intervention period. Baseline BDI-II scores were associated with depressive symptoms following the intervention period in all four models (p < 0.001), such that higher baseline BDI-II scores were associated with significantly higher BDI-II scores after the intervention period. In Model 1, participants in the App + Habit Training group experienced a 40.9% decline in BDI-II scores relative to the WLC group (p < 0.01), and this change remained significant even after adjusting for sociodemographic variables in Model 2 (−0.534, p < 0.05).

Table 6 shows the results of Poisson regression models predicting changes in BDI-II scores 6 months after the intervention ended between the App Only and App + Habit Training groups. First, baseline BDI-II scores were significantly associated with BDI-II scores at 6 months post-intervention in both models, such that higher baseline BDI-II scores were associated with higher BDI-II scores 6 months later. In Model 1, the App + Habit Training group had significantly greater decline in BDI-II scores relative to the App Only group (−0.382, p < 0.05). After adjusting for sociodemographic variables in Model 2, the App + Habit Training group experienced a larger decrease in depression symptoms relative to the App Only group 6 months post-intervention (−0.437, p < 0.05).

Estimated treatment effects on self-reported habit strength

Poisson regression models were also used to estimate the effects of each treatment on changes in self-reported meditation habit strength following the 8-week intervention and then at follow-up 6 months later. In Table 7, Models 1 and 2 estimated changes in SRBAI scores from baseline to week 8 (post-intervention), and Models 3 and 4 estimate changes in SRBAI scores from baseline to month 6. For both sets of analyses, we regressed SRBAI scores at week 8 and month 6 on baseline SRBAI score and treatment groups only in Models 1 and 3. In Models 2 and 4, we added sociodemographic characteristics (sex, age, education, race/ethnicity, medical history) as additional control variables. SRBAI scores were log transformed such that a one-unit change in the predictor variables reflects a percent change in SRBAI scores.

Table 7.

Changes in self-reported habit strength at post-intervention and 6-month follow-up.

Variable	SRBAI scores
	Post-intervention				Follow-up
	(1)		(2)		(3)		(4)
	IRR	95% CI	IRR	95% CI	IRR	95% CI	IRR	95% CI
Baseline SRBAI	0.281***	[0.105, 0.458]	0.361***	[0.186, 0.536]	0.088	[−0.120, 0.296]	0.180	[−0.141, 0.501]
App Only	Reference
App + Habit Training	2.467**	[0.544, 4.390]	3.564***	[0.911, 6.216]	1.696	[−1.408, 4.800]	3.036	−0.724, 6.795]
Sex
Female			2.459*	[−0.368, 5.286]			2.717	[−1.659, 7.094]
Age
18–29	References
30–39			0.162	[−2.654, 2.978]			−0.950	[−5.435, 3.534]
40–49			1.183	[−2.625, 4.990]			−3.196	[−9.633, 3.241]
50+			0.889	[−2.697, 4.475]			−2.648	[−8.048, 2.752]
Education
⩽High school			−0.649	[−4.807, 3.508]			0.880	[−5.189, 6.949]
Some college			−1.967	[−5.821, 1.888]			0.740	[−3.824, 5.305]
BA/BS			−1.991	[−6.073, 2.091]			1.590	[−4.228, 7.409]
Graduate	References
Race/ethnicity
Asian			1.231	[−2.782, 5.245]			0.574	[−12.77, 13.92]
White			−2.301	[−6.188, 1.585]			−1.285	[−3.886, 1.317]
Black			−4.128*	[−9.001, 0.745]			−5.336**	[−10.17, −0.507]
Hispanic			−6.411***	[−10.66, −2.163]			4.574	[−8.77, 17.92]
Other race(s)	References
Medical history
Takes any supplements			0.350	[−2.206, 2.906]			0.692	[−3.493, 4.876]
Any physical health conditions			−0.598	[−4.436, 3.241]			0.404	[−9.382, 10.19]
Anxiety and/or depression			0.902	[−1.547, 3.350]			0.301	[−3.300, 3.901]
Observations	72		65		45		40

SRBAI: Self-Reported Behavioral Automaticity Index; IRR: incident rate ratio.

Exponentiated Poisson regression coefficients indicate the incident rate ratio for each outcome. All models estimated participant-level random effects and standard errors were clustered at the participant level. Columns (1) and (2) show changes in SRBAI scores from baseline to post-intervention (Week 8), while Columns (3) and (4) show changes in SRBAI scores from post-intervention (week 8) to follow-up (month 6). Columns (1) and (3) display exponentiated coefficients from models using only baseline habit strength and treatment group as the predictors. Columns (2) and (4) display exponentiated coefficients from models that also included additional covariates. 95% confidence intervals are listed in brackets.

p < 0.05; **p < 0.01; ***p < 0.001.

Table 7 shows the results from the four Poisson regression models predicting changes in SRBAI scores following the intervention, and the App Only group was the reference variable in all four models. In Model 1, participants in the App + Habit Training group experienced a 246.7% increase in SRBAI scores relative to the App Only group (p < 0.01) following the intervention, and this change remained statistically significant after adjusting for sociodemographic variables in Model 2 (3.564, p < 0.001). In Models 3 and 4, the App + Habit Training group no longer experienced an increase in SRBAI scores at month 6 relative to the App Only group.

Discussion

To our knowledge, this study is the first to use anchoring-based habit formation strategies to promote the mindfulness meditation app use among autistic adults. We hypothesized that those who were instructed to anchor their meditation app use with an existing routine would report increased meditation practice and in turn, report greater improvements in depressive symptoms as compared with those who only received access to the app and a WLC group. Relative to the WLC group, those who received access to the Ten Percent Happier meditation app reported significantly increased meditation practice during the 8-week intervention, and this effect was stronger for those who also received the supplementary anchoring-based habit training. Furthermore, those who received the habit training in addition to the Ten Percent Happier app were able to maintain their meditation practice during the 8 weeks following the intervention period, reporting significantly more days with any meditation (and mediation minutes, see Supplemental Table S1) than those who only received access to the app. Crucially, those who received the habit training reported a significantly greater decline in depressive symptoms following the 8-week intervention and they maintained a significantly lower level of depressive symptoms over the subsequent 6 months. These findings suggest that anchoring is a viable strategy for helping adults with autism establish and maintain app-based mindfulness meditation practice that leads to persistent depressive symptom reductions.

Our first hypothesis related to the effects of gaining access to a mindfulness meditation app on meditation practice. We expected participants who were given access to Ten Percent Happier app would practice meditation more than those in a WLC group. Relative to the WLC, those who only received access to the app (App Only) meditated over 1.3 times more days per week during the intervention period, suggesting that app access alone helped autistic adults engage in more frequent meditation practice. Mindfulness apps offer several advantages that complement in-person mental health supports, including increased accessibility and flexibility (Gál et al., 2021; Spijkerman et al., 2016). This is crucial, as autistic adults often face financial and logistical barriers when accessing mental healthcare, highlighting the need for more accessible treatment options such as mobile apps (K. B. Beck et al., 2020; Zheng et al., 2021). However, head-to-head efficacy studies examining in-person versus app-delivered mindfulness training in adults with ASD are still needed to determine whether app-based programs are best used as supplementary or as standalone treatments. First, the limitations of app-based programs must be addressed before such head-to-head studies are possible.

A significant limitation of previous mindfulness meditation app studies, particularly in autistic populations, has been poor intervention adherence and high attrition rates (Gál et al., 2021; Hartley et al., 2022). Thus, one of the main goals of this study was to test the effectiveness of anchoring-based habit training on long-term meditation app practice among autistic adults. Anchoring is a behavioral strategy used to form habits in which an individual pairs or “anchors” a new behavior to an existing routine (Gardner, 2015; Wood & Neal, 2016). Habit formation strategies like anchoring have been used in other populations to promote long-term performance of health behaviors, including meditation app use (Stecher, Sullivan, & Huberty, 2021). During the intervention period, those who received anchoring-based habit training in addition to the meditation app reported nearly 1.5 times more days with meditation per week relative to the WLC group. Importantly, our results show that those who received the habit training reported significantly more days with meditation per week during the 8-week follow-up period when compared with the App Only group. These findings indicate that the addition of habit training led to greater maintenance of app-based meditation over time. In addition, these effect sizes on the maintenance of app-based meditation are larger than those observed in similar anchoring-based habit training interventions among typically developed populations (Stecher, Sullivan, & Huberty, 2021), which suggests that anchoring is both a feasible and effective strategy for promoting health behavior maintenance among autistic adults. This is in line with past research on RRB and the importance of routines for some autistic individuals (Cooper et al., 2022). Furthermore, we observed a larger treatment effect from the App + Habit Training intervention among the participants who reported either an anxiety or depression diagnosis (see Supplemental Table S2), which suggests that this is a feasible and effective intervention for improving mental health among those most in-need of an effective self-care solution.

A secondary goal of this study was to determine the effects of meditation app use and habit training on depressive symptoms over time. After adjusting for sociodemographic characteristics, those who received the anchoring-based habit training reported a 53.4% decline in depressive symptoms following the 8-week intervention, which was a significantly greater decline relative to the WLC group. After month 6, those who received habit training reported a 43.7% decline in depressive symptoms, which was a significantly greater decline than the App Only group. In addition, there were no significant changes in depressive symptoms among participants in the App Only group relative to the WLC following the 8-week intervention, suggesting that having access to the app alone was not enough to have a meaningful impact on depressive symptoms. Meditation apps have been shown to improve depressive symptoms in other populations (Champion et al., 2018; Economides et al., 2018; Gál et al., 2021; Huberty et al., 2021); however, these effects have yet to be demonstrated in an autistic population. This may be because the mental health benefits of meditation app use often takes time and consistent practice to be attained (Athanas et al., 2019; Carmody & Baer, 2008; Engen et al., 2018; Lardone et al., 2018; Linardon & Fuller-Tyszkiewicz, 2020), and autistic adults may face additional obstacles to maintaining their meditation practice as compared with neurotypical populations (Hartley et al., 2022; Smith & White, 2020). The significant short- and long-term decline in depressive symptoms seen in the study group receiving anchoring-based habit training is promising, as it suggests that anchoring may have helped these participants establish meditation habits and in turn, experience greater mental health benefits. These findings warrant additional research to determine the long-term effects of habitual meditation app use on other challenges among autistic adults.

To further investigate how the treatments impacted meditation habits, we also examined self-reported habit strength following the intervention and 6 months later. Relative to those who only received the Ten Percent Happier app, those who additionally received the habit training reported significantly stronger habits following the intervention period; however, this result was not statistically significant after 6 months. These findings suggest that the habit training was initially successful at establishing app-based meditation habits, but these may have been weak habits. Existing research suggests that habits can take anywhere from 18 to 254 days to form, depending on the frequency and complexity of the behavior (Lally & Gardner, 2013). Thus, our results imply that a 56-day (8-week) intervention may not be sufficiently long for everyone to establish strong app-based meditation habits, and future research should consider testing longer anchoring-based interventions among autistic adults. Alternatively, recent research has begun to question the validity of the self-reported habit strength measure (Hagger et al., 2015), therefore we suggest interpreting our findings on self-reported habits strength with some caution.

Limitations and future directions

This study benefited from the use of a successful commercial mindfulness meditation app and a randomized controlled trial study design; however, it is not without limitations. First, we had to rely on weekly EMAs to measure meditation behavior in all study groups, which means that those in the WLC group were also receiving weekly messages asking about their meditation practice. These weekly messages likely served as meditation reminders for all participants, which can be seen by the high average number of days with meditation among the WLC group during the intervention (M = 4.1 days) and during the follow-up period (M = 4.1 days). Accordingly, our results show the effects of receiving access to a meditation app and receiving habit training relative to just receiving reminders, which means our estimates are likely smaller than the true treatment effects when compared with receiving no treatment. In addition, since the WLC was receiving weekly messages during the first 8 weeks of the study, it is likely that some of these participants had started to form their own meditation habits. This helps to explain why our models of the self-reported habit strength measure are not perfectly aligned with the meditation behavior results, and again suggests that the real effects of our treatments may be larger when compared with a true control group. Supplemental Tables S3 to S6 show the regression results for models that include the WLC group in either the App Only or App + Habit Training groups they were assigned after the initial 8-week intervention, in order to better understand how our results vary among these different analytical approaches. Second, our measures of meditation app use were derived from self-reported assessments as opposed to objective meditation app usage data. Self-reported behavioral data is known to suffer from several biases such as expectation and social desirability effects (Althubaiti, 2016; Parry et al., 2021), so future research should demonstrate the efficacy of meditation apps and anchoring-based habit training among adults with autism using objective outcome measures. Third, our participants were primarily female (66.4%), White (76%), <50 years old (84%), and college educated (85.6%). Therefore, generalizability is limited in terms of biological sex, race, age, reading ability, and education. Furthermore, these demographic factors were significant in some of the models that were tested, warranting their study in future investigations. We speculate that a greater portion of females in the study may be attributed to recruitment biases or sex-biased interest in meditation- or app-based interventions advertised for depression in ASD, especially given the strong male sex bias in ASD diagnoses. Moreover, our sample may be generally younger, reflecting generational differences in the adoption of technology and engagement with research and meditation practices. The inclusion criterion of a reading ability >70 (WRAT) and being verbal was to enhance participant’s compliance with the study requirements and comprehension of the guided meditation instructions. Therefore, the findings cannot be generalized to those with lower reading abilities or who are nonverbal. Finally, our measures of meditation behavior were recorded over the first 16 weeks of the study, while the self-report questionnaire data was collected at baseline, week 8 and 6 months later. These different assessment timepoints make it difficult to associate long-term changes in mental health with short-term behavioral changes, and additional work should assess the ability of habit formation interventions to maintain health behaviors over longer durations of time among autistic adults.

Conclusion

These results demonstrate that an anchoring-based habit formation training established and maintained app-based meditation habits over an 8-week follow-up period among autistic adults. Importantly, those who received the habit training also reported significantly lower depressive symptoms at the end of the intervention and 6 months later, which suggests that app-based meditation habits are an effective long-term mental health solution for autistic adults who are experiencing depression. In addition, this anchoring-based habit training had a larger effect on meditation behavior than previously seen in typically developed adult populations, which highlights the potential of anchoring for establishing other healthy habits among autistic adults.

Supplemental Material

sj-docx-1-aut-10.1177_13623613231200679 – Supplemental material for App-based meditation habits maintain reductions in depression symptoms among autistic adults

Supplemental material, sj-docx-1-aut-10.1177_13623613231200679 for App-based meditation habits maintain reductions in depression symptoms among autistic adults by Chad Stecher, Broc A Pagni, Sara Cloonan, Schuyler Vink, Ethan Hill, Destiny Ogbeama, Shanna Delaney and B Blair Braden in Autism

Footnotes

Acknowledgements

The authors thank the SPARK Research Match program and Nicole Matthews for their assistance with recruitment, and Stephen Gallegos and Melony Valdez for their assistance with data collection.

Authors’ note

De-identified data from this study are not available in a public archive. De-identified data from this study will be made available (as allowable according to institutional IRB standards) by emailing the corresponding author.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by Arizona State University’s College of Health Solutions.

ORCID iDs

Chad Stecher

B. Blair Braden

Supplemental material

Supplemental material for this article is available online.

References

Adams

Young

(2021). A systematic review of the perceived barriers and facilitators to accessing psychological treatment for mental health problems in individuals on the autism spectrum. Review Journal of Autism and Developmental Disorders, 8(4), 436–453. https://doi.org/10.1007/s40489-020-00226-7

Althubaiti

(2016). Information bias in health research: Definition, pitfalls, and adjustment methods. Journal of Multidisciplinary Healthcare, 9, 211–217. https://doi.org/10.2147/JMDH.S104807

Athanas

A. J.

McCorrison

J. M.

Smalley

Price

Grady

Wehner

Campistron

Schork

N. J.

(2019). Association between improvement in baseline mood and long-term use of a mindfulness and meditation app: Observational study. JMIR Mental Health, 6(5), Article e12617. https://doi.org/10.2196/12617

Baumel

Muench

Edan

Kane

J. M.

(2019). Objective user engagement with mental health apps: Systematic search and panel-based usage analysis. Journal of Medical Internet Research, 21(9), Article e14567. https://doi.org/10.2196/14567

Beck

A. T.

Steer

R. A.

Brown

(1996). Manual for the Beck Depression Inventory-II. Psychological Corporation.

Beck

K. B.

Greco

C. M.

Terhorst

L. A.

Skidmore

E. R.

Kulzer

J. L.

McCue

M. P.

(2020). Mindfulness-based stress reduction for adults with autism spectrum disorder: Feasibility and estimated effects. Mindfulness, 11, 1286–1297. https://doi.org/10.1007/s12671-020-01345-2

Bishop-Fitzpatrick

Minshew

N. J.

Eack

S. M.

(2013). A systematic review of psychosocial interventions for adults with autism spectrum disorders. Journal of Autism and Developmental Disorders, 43, 687–694. https://doi.org/10.1007/s10803-012-1615-8

Braden

B. B.

Pagni

B. A.

Monahan

Walsh

M. J. M.

Dixon

M. V.

Delaney

Ballard

Ware

J. E

Jr . (2022). Quality of life in adults with autism spectrum disorder: Influence of age, sex, and a controlled, randomized mindfulness-based stress reduction pilot intervention. Quality of Life Research, 31, 1427–1440. https://doi.org/10.1007/s11136-021-03013-x

Brooks

T. L.

Leventhal

Wolf

M. S.

O’Conor

Morillo

Martynenko

Wisnivesky

J. P.

Federman

A. D.

(2014). Strategies used by older adults with asthma for adherence to inhaled corticosteroids. Journal of General Internal Medicine, 29, 1506–1512. https://doi.org/10.1007/s11606-014-2940-8

10.

Carmody

Baer

R. A.

(2008). Relationships between mindfulness practice and levels of mindfulness, medical and psychological symptoms and well-being in a mindfulness-based stress reduction program. Journal of Behavioral Medicine, 31, 23–33. https://doi.org/10.1007/s10865-007-9130-7

11.

Champion

Economides

Chandler

(2018). The efficacy of a brief app-based mindfulness intervention on psychosocial outcomes in healthy adults: A pilot randomised controlled trial. PLOS ONE, 13(12), Article e0209482. https://doi.org/10.1371/journal.pone.0209482

12.

Constantino

J. N.

Gruber

C. P.

(2012). Social responsiveness scale, second edition (srs-2). Western Psychological Services.

13.

Cooper

Russell

Calley

Chen

Kramer

Verplanken

(2022). Cognitive processes in autism: Repetitive thinking in autistic versus non-autistic adults. Autism, 26(4), 849–858. https://doi.org/10.1177%2F13623613211034380

14.

Economides

Martman

Bell

M. J.

Sanderson

(2018). Improvements in stress, affect, and irritability following brief use of a mindfulness-based smartphone app: A randomized controlled trial. Mindfulness, 9, 1584–1593. https://doi.org/10.1007/s12671-018-0905-4

15.

Engen

H. G.

Bernhardt

B. C.

Skottnik

Ricard

Singer

(2018). Structural changes in socio-affective networks: Multi-modal mri findings in long-term meditation practitioners. Neuropsychologia, 116, 26–33. https://doi.org/10.1016/j.neuropsychologia.2017.08.024

16.

Fleig

Gardner

Keller

Lippke

Pomp

Wiedemann

A. U.

(2017). What contributes to action plan enactment? Examining characteristics of physical activity plans. British Journal of Health Psychology, 22(4), 940–957. https://doi.org/10.1111/bjhp.12263

17.

Gál

Stefan

Cristea

I. A.

(2021). The efficacy of mindfulness meditation apps in enhancing users’ well-being and mental health related outcomes: A meta-analysis of randomized controlled trials. Journal of Affective Disorders, 279, 131–142. https://doi.org/10.1016/j.jad.2020.09.134

18.

Galla

B. M.

Duckworth

A. L.

(2015). More than resisting temptation: Beneficial habits mediate the relationship between self-control and positive life outcomes. Journal of Personality and Social Psychology, 109(3), 508–525. https://doi.org/10.1037/pspp0000026

19.

Gardner

(2015). A review and analysis of the use of “habit” in understanding, predicting and influencing health-related behaviour. Health Psychology Review, 9(3), 277–295. https://doi.org/10.1080%2F17437199.2013.876238

20.

Gardner

Abraham

Lally

de Bruijn

G.-J.

(2012). Towards parsimony in habit measurement: Testing the convergent and predictive validity of an automaticity subscale of the self-report habit index. International Journal of Behavioral Nutrition and Physical Activity, 9, Article 102. https://doi.org/10.1186/1479-5868-9-102

21.

Gardner

Lally

Wardle

(2012). Making health habitual: The psychology of “habit-formation” and general practice. British Journal of General Practice, 62(605), 664–666. https://doi.org/10.3399%2Fbjgp12X659466

22.

Strauss

Bond

Cavanagh

(2015). How do mindfulness-based cognitive therapy and mindfulness-based stress reduction improve mental health and wellbeing? A systematic review and meta-analysis of meditation studies. Clinical Psychology Review, 37, 1–12. https://doi.org/10.1016/j.cpr.2015.01.006

23.

Hagger

M. S.

Rebar

A. L.

Mullan

Lipp

O. V.

Chatzisarantis

N. L.

(2015). The subjective experience of habit captured by self-report indexes may Lead to inaccuracies in the measurement of habitual action. Health Psychology Review, 9(3), 296–302. https://doi.org/10.1080/17437199.2014.959728

24.

Hartley

Dorstyn

Due

(2022). Challenges encountered with a mindfulnes app: Lessons learnt from a pilot randomized trial involving caregivers and individuals with autism. Research in Autism Spectrum Disorders, 96, Article 101991. https://doi.org/10.1016/j.rasd.2022.101991

25.

Huberty

Puzia

M. E.

Green

Vlisides-Henry

R. D.

Larkey

Irwin

M. R.

Vranceanu

A. M.

(2021). A mindfulness meditation mobile app improves depression and anxiety in adults with sleep disturbance: Analysis from a randomized controlled trial. General Hospital Psychiatry, 73, 30–37. https://doi.org/10.1016/J.GENHOSPPSYCH.2021.09.004

26.

Hudson

C. C.

Hall

Harkness

K. L.

(2019). Prevalence of depressive disorders in individuals with autism spectrum disorder: A meta-analysis. J Abnorm Child Psychol, 47(1), 165–175. https://doi.org/10.1007/s10802-018-0402-1

27.

Judah

Gardner

Aunger

(2013). Forming a flossing habit: An exploratory study of the psychological determinants of habit formation. British Journal of Health Psychology, 18(2), 338–353. https://doi.org/10.1111/j.2044-8287.2012.02086.x

28.

Kabat-Zinn

(1994). Wherever you go, there you are: Mindfulness meditation in everyday life. Hyperion.

29.

Keller

Fleig

Hohl

D. H.

Wiedemann

A. U.

Burkert

Luszcynska

Knoll

(2017). Which characteristics of planning matter? Individual and dyadic physical activity plans and their effects on plan enactment. Social Science & Medicine, 189, 53–62. https://doi.org/10.1016/j.socscimed.2017.07.025

30.

Keller

Kwasnicka

Klaiber

Sichert

Lally

Fleig

(2021). Habit formation following routine-based versus time-based cue planning: A randomized controlled trial. British Journal of Health Psychology, 26(3), 807–824. https://doi.org/10.1111/bjhp.12504

31.

Kiep

Spek

A. A.

Hoeben

(2015). Mindfulness-based therapy in adults with an autism spectrum disorder: Do treatment effecrs last? Mindfulness, 6, 637–644. https://doi.org/10.1007/s12671-014-0299-x

32.

Kim

J. W.

Nguyen

T.-Q.

Gipson

S. Y. M. T.

Shin

A. L.

Torous

(2018). Smartphone apps for autism spectrum disorder—understanding the evidence. Journal of Technology in Behavioral Science, 3, 1–4. https://doi.org/10.1007/s41347-017-0040-4

33.

Kraper

C. K.

Kenworthy

Popal

Martin

Wallace

G. L.

(2017). The gap between adaptive behavior and intelligence in autism persists into young adulthood and is linked to psychiatric co-morbidities. J Autism Dev Disord, 47(10), 3007–3017. https://doi.org/10.1007/s10803-017-3213-2

34.

Lally

Gardner

(2013). Promoting habit formation. Health Psychology Review, 7, S137–S158. https://doi.org/10.1080/17437199.2011.603640

35.

Lally

Wardle

Gardner

(2011). Experiences of habit formation: A qualitative study. Psychology, Health & Medicine, 16(4), 484–489. https://doi.org/10.1080/13548506.2011.555774

36.

Lardone

Liparoti

Sorrentino

Rucco

Jacini

Polverino

Minino

Pesoli

Baselice

Sorriso

Ferraioli

Sorrentino

Mandolesi

(2018). Mindfulness meditation is related to long-lasting changes in hippocampal functional topology during resting state: A magnetoencephalography study. Neural Plasticity, 2018, Article 5340717. https://doi.org/10.1155/2018/5340717

37.

Levy

Perry

(2011). Outcomes in adolescents and adults with autism: A review of the literature. Research in Autism Spectrum Disorders, 5(4), 1271–1282. https://doi.org/10.1016/j.rasd.2011.01.023

38.

Linardon

Fuller-Tyszkiewicz

(2020). Attrition and adherence in smartphone-delivered interventions for mental health problems: A systematic and meta-analytic review. Journal of Consulting and Clinical Psychology, 88(1), 1–13. https://doi.org/10.1037/ccp0000459

39.

Lunsky

Redquest

Albaum

Hutton

Share-Strom

Weiss

(2022). Virtual group-based mindfulness intervention for autistic adults: A feasibility study. Mindfulness, 13, 1706–1718. https://doi.org/10.1007/s12671-022-01909-4

40.

Neal

D. T.

Wood

Drolet

(2013). How do people adhere to goals when willpower is low? The profits (and pitfalls) of strong habits. Journal of Personality and Social Psychology, 104(6), 959–975. https://doi.org/10.1037/a0032626

41.

Nichols

Block

M. E.

Bishop

J. C.

McIntire

(2019). Physical activity in young adults with autism spectrum disorder: Parental perceptions of barriers and facilitators. Autism, 23(6), 1398–1407. https://doi.org/10.1177%2F1362361318810221

42.

Pagni

B. A.

Hill

Walsh

M. J. M.

Delaney

Ogbeama

Monahan

Cook

J. R.

Guerithault

Dixon

M. V.

Ballard

Braden

. (2023). Distinct and shared therapeutic neural mechanisms of mindfulness-based and social support stress reduction groups in adults with autism spectrum disorder. Journal of Psychiatry and Neuroscience, 48(2), 102–114. https://doi.org/10.1503/jpn.220159

43.

Pagni

B. A.

Walsh

M. J. M.

Foldes

Sebren

Dixon

M. V.

Guerithault

Braden

B. B.

(2020). The neural correlates of mindfulness-induced depression reduction in adults with autism spectrum disorder: A pilot study. Journal of Neuroscience Research, 98(6), 1150–1161. https://doi.org/10.1002/jnr.24600

44.

Parry

D. A.

Davidson

B. I.

Sewall

C. J. R.

Fisher

J. T.

Mieczkowski

Quintana

D. S.

(2021). A systematic review and meta-analysis of discrepancies between logged and self-reported digital media use. Nature Human Behaviour, 5, 1535–1547. https://doi.org/10.1038/s41562-021-01117-5

45.

Pimm

Vandelanotte

Rhodes

R. E.

Short

Duncan

M. J.

Rebar

A. L.

(2016). Cue consistency associated with physical activity automaticity and behavior. Behavioral Medicine, 42(4), 248–253. https://doi.org/10.1080/08964289.2015.1017549

46.

Querstret

Morison

Dickinson

Cropley

John

(2020). Mindfulness-based stress reduction and mindfulness-based cognitive therapy for psychological health and well-being in nonclinical samples: A systematic review and meta-analysis. International Journal of Stress Management, 27(4), 394–411. https://doi.org/10.1037/str0000165

47.

Sizoo

B. B.

Kuiper

(2017). Cognitive behavioural therapy and mindfulness based stress reduction may be equally effective in reducing anxiety and depression in adults with autism spectrum disorders. Research in Developmental Disabilities, 64, 47–55. https://doi.org/10.1016/j.ridd.2017.03.004

48.

Smith

I. C.

White

S. W.

(2020). Socio-emotional determinants of depressive symptoms in adolescents and adults with autism spectrum disorder: A systematic review. Autism, 24(4), 995–1010. https://doi.org/10.1177/1362361320908101

49.

Spek

A. A.

van Ham

N. C.

Nyklíček

(2013). Mindfulness-based therapy in adults with autism spectrum disorder: A randomized controlled trial. Research in Developmental Disabilities, 34(1), 246–253. https://doi.org/10.1016/j.ridd.2012.08.009

50.

Spijkerman

M. P. J.

Pots

W. T. M.

Bohlmeijer

E. T.

(2016). Effectiveness of online mindfulness-based interventions in improving mental health: A review and meta-analysis of randomised controlled trials. Clinical Psychology Review, 45, 102–114. https://doi.org/10.1016/j.cpr.2016.03.009

51.

Stawarz

Gardner

Cox

Blandford

(2020). What influences the selection of contextual cues when starting a new routine behaviour? An exploratory study. BMC Psychology, 8, Article 29. https://doi.org/10.1186/s40359-020-0394-9

52.

Stecher

Mukasa

Linnemayr

(2021). Uncovering a behavioral strategy for establishing new habits: Evidence from incentives for medication adherence in uganda. Journal of Health Economics, 77, Article 102443. https://doi.org/10.1016/j.jhealeco.2021.102443

53.

Stecher

Sullivan

Huberty

(2021). Using personalized anchors to establish routine medication practice with a mobile app: Randomized controlled trial. JMIR Mhealth and Uhealth, 9(12), Article e32794. https://doi.org/10.2196/32794

54.

Vaghefi

Tulu

(2019). The continued use of mobile health apps: Insights from a longitudinal study. JMIR Mhealth and Uhealth, 7(8), Article e12983. https://doi.org/10.2196/12983

55.

Wallace

G. L.

Kenworthy

Pugliese

C. E.

Popal

H. S.

White

E. I.

Brodsky

Martin

(2016). Real-world executive functions in adults with autism spectrum disorder: Profiles of impairment and associations with adaptive functioning and co-morbid anxiety and depression. Journal of Autism and Developmental Disorders, 46(3), 1071–1083. https://doi.org/10.1007/s10803-015-2655-7

56.

Wang

Y.-P.

Gorenstein

(2013). Psychometric properties of the Beck Depression Inventory-II: A comprehensive review. Brazilian Journal of Psychiatry, 35(4), 416–431. https://doi.org/10.1590/1516-4446-2012-1048

57.

White

S. W.

Simmons

G. L.

Gotham

K. O.

Conner

C. M.

Smith

I. C.

Beck

K. B.

Mazefsky

C. A.

(2018). Psychosocial treatments targeting anxiety and depression in adolescents and adults on the autism spectrum: Review of the latest research and recommended future directions. Current Psychiatry Reports, 20(10), Article 82. https://doi.org/10.1007/s11920-018-0949-0

58.

Wilkinson

G. S.

Robertson

G. J.

(2017). Wide range achievement test, fifth edition. Pearson Corporation.

59.

Williams

Z. J.

Everaert

Gotham

K. O.

(2021). Measuring depression in autistic adults: Psychometric validation of the Beck Depression Inventory-II. Assessment, 28(3), 858–876. https://doi.org/10.1177/1073191120952889

60.

Wood

Neal

D. T.

(2016). Healthy through habit: Interventions for initiating & maintaining health behavior change. Behavioral Science & Policy, 2(1), 71–83.

61.

Zheng

Adams

Taylor

J. L.

Pezzimenti

Bishop

S. L.

(2021). Depression in independent young adults on the autism spectrum: Demographic characteristics, service use, and barriers. Autism, 25(7), 1960–1972. https://doi.org/10.1177%2F13623613211008276

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.06 MB