A Cognitive Behavioral Intervention for Driving for Autistic Teens and Adults: A Pilot Study

Abstract

Background:

For many autistic individuals approaching adulthood, interventions to promote skills toward independence are lacking. Driving is an important ability to cultivate as it may be a critical step to attaining independence, securing and maintaining work, and fostering relationships. Only about one third of able autistic individuals drive independently, and fear to drive is a common reason for not driving.

Methods:

This initial pilot study was a 3-month open trial to investigate the feasibility, acceptability, and initial estimates of outcomes for the newly developed Cognitive Behavioral Intervention for Driving (CBID) intervention, a manualized curriculum to enhance executive functioning (EF) and emotional regulation (ER) skill development for driving, combined with individualized commentary-based driving simulator practice, in autistic teens and adults within a community research environment. Nineteen participants received the CBID intervention in 1.5-hour group sessions for 10 weeks, across two cohorts, with approximately five individualized driving simulator sessions. Data were collected on anxiety symptoms, driving cognitions, driving attitudes, and simulated driving performance at pre- and post-intervention assessments.

Results:

Program completion rate was 81%. Participants and parents rated both the intervention and simulator practice sessions with high satisfaction. All the participants (100%) reported both a positive attitude change (interest toward driving) and a desire to drive in the future at the post-intervention assessment. Significant changes occurred for driving cognitions, attitudes, and behaviors, and 47% of participants obtained a driver's permit or license by 2 months post-intervention.

Conclusions:

For autistic individuals, the CBID intervention appeared to directly impact the pursuit toward driving goals by both increasing driving attitudes and behaviors and reducing anxiety/apprehension. This highlights the need for driving intervention programs designed specific to autistic teens/adults that focus on EF and ER skills coupled with individualized simulator practice. CBID could be provided in community services to increase the number of autistic individuals driving.

Lay summary

Why was this study done?

Driving contributes greatly to independence in many teens and adults. Anxiety can act as a barrier to that independence by affecting driving attitudes, ability and performance. Autistic individuals are often affected by anxiety, executive functioning, and emotion regulation challenges. Previous studies show that virtual reality training and Cognitive Behavioral Therapy (CBT), separately, can reduce driving anxiety in autistic individuals and support driving skills. However, to date, no studies have developed and tested a manualized intervention specific to driving that combines such methods.

What was the purpose of this study?

The goal of this study was to develop and test a group intervention, called Cognitive Behavioral Intervention for Driving (CBID), combined with individualized, coached driving simulator practice to help reduce fears and increase cognitions, attitudes and performance towards driving. The study examined the following questions: 1) Do autistic individuals actively participate in CBID? 2) Do autistic participants like CBID? and 3) Are autistic participants more prepared to drive after taking part in CBID than they were before they participated in the program?

What did the researchers do?

After developing the CBID intervention with community member involvement, the researchers ran 2 intervention groups and individualized driving simulator sessions to a total of 19 enrolled participants. The group intervention focused on strengthening executive functioning and emotion regulation skills in 1.5 hour sessions over 10 weeks and 5, 1 hour driving simulator sessions. Researchers compared data on participant anxiety levels, driving cognitions, attitudes, and performance before and after the intervention.

What were the results of the study?

Most of the participants completed the program (81%) and all (100%) rated satisfaction with the group and simulator practice. All (100%) of the participants changed attitudes towards driving. Participants had higher levels of positive thoughts about driving, better attitudes towards wanting to drive and less driving errors (like speeding, collisions, crossing lanes) on the driving simulator after completing CBID. Almost half (47%) of participants obtained a driver's permit or license by 2 months after the intervention.

What do these findings add to what was already known?

This study adds a new option of an integrated approach, CBID, for addressing driving anxiety or apprehension. It provides initial findings of the value of incorporating executive functioning skills with traditional cognitive behavior therapy for enhancing driving readiness.

What are the potential weaknesses in the study?

This study contained a small sample size that was mostly white male participants. This limits generalizing the results to a representative diverse population. It also did not have a control group or use randomization which means that results can't be interpreted as causal at this time. There was no information about participants obtaining permit/licensure later than 2 months after CBID so it is unclear if some participants require more time to pursue a license. There was no follow up to understand if participants continued to feel comfortable driving overtime.

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

This study showed that it's possible to combine virtual reality training and Cognitive Behavioral Therapy into one driving readiness program. The new CBID program may help autistic adults by addressing multiple aspects of what they need to be ready to drive. Since the study used previously tested strategies, enrolled a community sample, manualized the program, and used a services approach, it was designed for broad distribution to other community settings.

Introduction

For many autistic individuals approaching adulthood, interventions to promote skills toward independence are lacking.¹ Driving is an important ability to cultivate as it may be a critical step to attaining independence, securing and maintaining work, and fostering relationships.² Recent literature reveals that only 24%–34% of autistic individuals without intellectual disability drive independently.^2–4 This difficulty with driving may stem from common challenges found in autism related to: (1) executive functioning (EF) deficits^5,6 (attention modulation, shifting, flexibility, multitasking), (2) delayed motor skills⁷ (coordination, motor control), emotional dysregulation⁸ (anxiety, frustrations), and (3) social processing deficits^9,10 (contextual awareness), all of which are important components of driving.¹¹

Autistic teens and adults may not pursue driving due to concerns with EF and/or motor skills, as driving requires incorporating such skills behind the wheel.^8,12 One driving study with young autistic males found less mental flexibility and greater deficits in executing movement, lowering recognition, and reaction to hazards.⁴

In addition, autistic individuals report comorbid anxiety and fear to drive as a common reason for not driving.¹³ Overall, ∼40%–60% of autistic individuals experience anxiety or have an anxiety-related disorder.^14,15 Such anxiety often manifests behind the wheel.¹³ For example, one study found that autistic individuals reported high driving anxiety and commonly requested a companion in the car while driving even after obtaining a driver's license.¹⁶ Additionally, autistic drivers reported not driving for vocational or social purposes.¹⁶

Research shows that cognitive behavioral therapy (CBT) is successful in addressing anxiety overall¹⁷ and may be particularly applicable for reducing driving anxiety.¹³ Furthermore, using virtual reality simulators to produce a realistic but safe experience can be used as exposure therapy, a core technique in CBT for treating anxiety.¹⁸ Research to date demonstrates that virtual reality exposure with CBT is an effective treatment for autistic people with phobias.¹⁹ A review focused on CBT with autistic adolescents and adults shows evidence for group CBT programs in reducing anxiety in autistic youth and adults.²⁰ When using CBT combined with driving simulator practice for nonautistic adults, individuals with driving anxiety are better able to manage their driving-related distress and reduce driving avoidance.²¹ However, no study to date has investigated group CBT with driving simulator exposure for driving anxiety in autistic populations.

Driving simulator programs not only provide a means for graded exposure but also assist in developing driving skills by producing physiological reactivity consistent with actual on-road driving.²² Using a driving simulator can allow people to practice applying EF and emotional regulation (ER) skills safely and efficiently.⁶ Parents report that their autistic teens and adults need the following: extended time to practice driving skills, immediate feedback, opportunities to correct mistakes, and taught skills in small steps,⁶ all features simulators can provide.¹¹ A recent study with transition-age (15–23 years) autistic males using a driving simulator found that autistic drivers had poor driving performance overall compared with nonautistic drivers; however, skills increased with repetitive practice.⁶ Virtual reality environments are ideal for targeting ER, as participants are able to experience anxiety-provoking situations in systematically controlled environments.¹⁸ While practicing driving with a simulator alone is beneficial, simulator training is more effective when supported by instructor-given commentary addressing errors and personalizing the application of skills targeting both motor skills and psychological aspects of driving.^13,23 Additionally, Cox et al. concluded that autistic driver training should focus on both improving driving skills and reducing driving apprehension.²⁴ Group CBT training programs that teach a variety of EF and ER skills specific to driving, with individually coached driving simulator practice, may be the ideal driving training environment for autistic individuals.

This initial pilot study was a 3-month open trial to investigate the feasibility, acceptability, and initial estimates of outcomes for the newly developed Cognitive Behavioral Intervention for Driving (CBID) intervention. CBID is a manualized curriculum to enhance EF and ER skill development for driving combined with individualized commentary-based driving simulator practice in autistic teens and young adults within a community research environment.

Methods

Study design

This quasi-experimental community services research study was developed through a research–community partnership called ACHIEVE (Active Collaborative Hub for Individuals with autism spectrum conditions to Enhance Vocation and Education) in which researchers, autistic self-advocates, family members, providers, agency administrators, and educators all come together to inform, design, develop, and test community services for autistic populations. Autistic individuals were involved in all aspects of the study, including intervention development, and were involved as members of the evaluation team in addition to the ACHIEVE group. The study was reviewed and approved by the University of California, San Diego and Rady Children's Hospital San Diego joint Social/Behavioral Science Institutional Review Board.

Intervention procedures

CBID curriculum

The CBID intervention included 10 sessions (8 sessions of content and 2 assessment/feedback sessions) offered one time per week for 1.5 hours. The number of sessions and length were determined by the ACHIEVE group based on identified content, prior experiences, and attention to feasibility factors (i.e., time available, associated costs). The curriculum combined CBT strategies to address ER skills with additional cognitive enhancement strategies targeting EF skills required for driving and adapted from the Baker-Ericzén et al. SUCCESS (Supported, Comprehensive Cognitive Enhancement and Social Skills) intervention.²⁵ The CBT components included: (1) identifying and changing anxious or negative thoughts to positive rational thoughts, (2) emotional awareness and management skills, and (3) use of practicing behavioral skills through graded step-by-step exposure. The EF skills taught included (1) contextual awareness, (2) sustaining attention, (3) shifting attention, (4) cognitive flexibility, (5) problem-solving, and (6) goal-oriented thinking and behaviors. The EF skills were taught first and then ER skills subsequently with each prior skill incorporated into the next and all skills explicitly focused on application to driving (refer to Table 1 for each session content).

Table 1.

Cognitive Behavioral Intervention for Driving Curriculum

Session numbera	Constructs	Content	Strategies
2	Cognitive skills for driving	Program overview; goal setting; identify challenges; terminology	Demonstrate value of course
			Psychoeducation on executive functioning and cognitive behavioral strategies for driving
			Explanation of skills to be developed during course
			Individualize course material by setting personal goals and a fear hierarchy
3	Contextual awareness within and outside of car: observing and identifying context clues	Identify context clues; understand internal (in car and self) and external (driving environment) context clues; graded exposure on fear hierarchy	Scanning external context clues (awareness of things outside of car as driving setting, road signs, other vehicles, other people/animals)
			Scanning internal context clues (awareness of things inside of car including self as dashboard, radio, vents, own thoughts, own emotions)
			Piecing together the big picture and determining awareness, safety, and actions
			Review fear hierarchy and pick the first fear to face, practice this
4	Sustaining and shifting attention	Sustaining attention on driving context (internal and external); boosting attention; shifting attention; multiple processing	Monitoring five senses for attention (see, hear, touch, smell, and taste)
			Boosting alertness/attention using one or more of five senses
			Eliminating distractions
			Shifting attention using a routine/system when driving and using internal and external context clue checks
			Shifting between observing and thinking
			Using “self-talk” to guide multiple processing
			Graded exposure with videos
5	Flexibility and problem solving	Flexibility and problem solving; think differently; act differently; create plans and change plans	Learning types of flexibility
			Three step flexible thinking (notice, accept, think differently, and have a flexible thought)
			Act flexibly using self-talk (taking different actions)
			Problem solving while driving (Plan A, Plan B, Plan C approach)
			Graded exposure with videos
6	Emotional training and thought changing	Understand linkages between thoughts, emotions, and actions and reactions; identify negative thoughts; learn to change those thoughts into positive	Psychoeducation on tenets of CBT (linkages between thoughts, emotions, and actions and reactions)
			Identifying negative versus positive thoughts and gaining thought awareness
			Changing thoughts and chain reactions
			Graded exposures with videos, taking action on fear hierarchy, and driving simulator practice
7	Emotion awareness and regulation	Identify emotion level; change emotion level while driving	Psychoeducation on emotions and emotional awareness
			Learning three-step emotional awareness (notice, identify, rate emotions)
			Using own emotional rating scale and activating problem solving
			Regulating emotions (breathing, sensory inputs, positive thinking, or combination)
			Graded exposures with videos and driving simulator practice
8	Goals, plans, and preparations	Goal-oriented thinking; identifying goals and plans	Psychoeducation on goal-oriented thinking
			Learning five steps to reaching goals (set, split, schedule, commit, act)
			Developing goals and plans specific to driving
			Learning driving license requirements
			Making step-by-step plan for obtaining driving license with support (family member involvement)
			Graded exposure of taking action on fear hierarchy and driving simulator practice
9	Final preparations	Prepare thoughts, emotions, and vehicle for driving; prepare for rare and common driving challenges; plan for obtaining driver's license	Summarize and review concepts
			Preparing thoughts and emotions for driving
			Preparing vehicle (including personalize with sensory needs) for driving
			Preparing for and learning how to handle rare driving challenges (accidents, traffic violations, car troubles, road trips) with checklists and video exposure
			Preparing for and learning how to handle common driving challenges (traffic, ambulance sirens, driving on curves or ramps, other enraged drivers) with practice and video exposure
			Formulating plan for taking driver's examination and graded exposures (practice examinations)
			Graded exposures with videos, taking action on fear hierarchy, mock examinations, and driving simulator practice

^a Sessions 1 and 10 were assessment and feedback sessions with no curriculum.

CBT, cognitive behavioral therapy.

One facilitator conducted both groups. The program was revised slightly to increase clarity after group 1, incorporating feedback from multiple sources (staff, participants, parents). Changes included minor language changes and adding a few visual models and videos for illustrative purposes. Each participant received a workbook with eight content sessions given to them one at a time, one per week. The participants were guided to read along and complete written activities in their workbooks, including at home practice assignments. The facilitator played an important role in delivering the curriculum in an engaging and individualized manner, monitoring comprehension, adjusting pacing of the group, and leading rich discussions targeting generalization of skills to participant driving experiences.

Participants and one parent/caregiver per participant completed paper measures at baseline (session 1) and post-intervention (session 10) (3 months after the baseline) in person during the group meeting. Group staff was available to answer questions and provide clarity as needed. Participants completed in-person interviews at post-intervention. Interviewers recorded participant open-ended responses verbatim on paper forms. Research team double entered and cleaned all data before analyses.

Driving simulator experience

The simulator consisted of a steering wheel with turn signals, a single wide screen monitor, and gas and brake foot pedals using the STISIM Drive software. Simulated driving was customized to the individual but typically targeted skills starting with mastering motor movements involved in driving (i.e., one-foot pedal use, turn signals, checking mirrors, gaging speed). Then, skills progressed to applying road rules (i.e., responding to signs, car management in turning, looking both ways at intersections). Finally, the simulated driving involved driving in complex situations and environments (i.e., rural, urban, suburban, school zones, construction zones, freeways). A driving coach provided commentary throughout simulated driving sessions matched to individual levels and course skills learned to date. Commentary provided was linked to the curriculum. After each completed drive, performance data were reviewed and individualized practice was assigned to increase skills as: motor coordination, cognitive shifting, and/or sensory/ER. Time was also spent reviewing course assignments (called “Act the Fact”) and progress toward licensure. There were three providers/staff and two interns trained in providing driving simulator coaching sessions. Coaches, after completing training, were assigned to work with the same participants throughout the program and were supervised weekly by the first author. Coaches maintained session notes recording target skills, driving goals, and progress. Coaches provided a total of 287 “drives” on the simulator to participants, totaling over 17 hours. The mean number of 45- to 60-minute driving sessions was 4.8 (standard deviation [SD] = 2.0), with mean number of total “drives” of 16.88 (SD = 11.16) per participant.

Sample

A total of 54 participants responded to the community distributed recruitment flyer. Flyers were sent via local autism association and services distribution lists. Eligibility criteria included: a prior diagnosis of autism spectrum disorder (ASD), being of California driver training age (≥15.5 years), fourth grade or above reading ability, and conversationally fluent in English. A clinical/medical diagnosis of ASD was made by a community professional (trained psychiatrist, psychologist, or school psychologist) and reported on by both the participant and parent. Reading and language levels were also provided by self-report. The first 21 autistic participants meeting criteria were enrolled to participate in the grant supported program (others were waitlisted for future funded groups). A total of 19 participants attended groups and participated in the research study; one teen never attended and one teen dropped at baseline. The participants were 95% male ranging in age from 15 to 29 years, with μ = 20.53, SD = 4.40 with 32% teenagers (<18 years). The majority (58%) reported as non-Hispanic with 11% Hispanic and 32% reported as unknown. Refer to Table 2 for additional sample demographics including service involvement.

Table 2.

Demographic Characteristics of the Sample

	Participant (n = 19)	Caregiver (n = 17)
	Mean (SD: range)/%	Mean (SD: range)/%
Gender
Male	95%	12%
Female	5%	88%
Parental role
Father	—	12%
Mother	—	82%
Other relative	—	6%
Age, years	20.53 (4.4: 15–29)	53.71 (4.9: 46–61)
<18	32%	—
Education
High school students	47%	—
High school graduates	26%	—
College attendance	26%	—
Living arrangement
Participants living at home	95%	—
Ethnicity
Hispanic	11%	18%
Non-Hispanic	58%	82%
Unknown	32%	0%
Race
White	74%	88%
American Indian/Alaskan Native	5%	6%
Asian	5%	0%
African American	5%	6%
Unknown/not reported	11%	0%
Autism spectrum diagnosis	100%	—
Mean age diagnosed, years	8 (5.37: 3–15)	—
Current services involvement
Developmental disability	79%	—
Special education	37%	—
Department of rehabilitation	27%	—
Vocational rehabilitation day program	5%	—
Current treatment involvement		—
Medication for ASD symptoms	26%	—
Counseling/therapy	21%	—
Social skill group	26%	—
Speech therapy	37%	—
Occupational therapy	16%	—
Income
Supplemental security income	21%	—
Socioeconomic status
Middle class	—	74%
Low-income poverty	—	26%

ASD, autism spectrum disorder; SD, standard deviation.

Measures

Program fidelity

CBID intervention involved following a facilitator guide with lessons and activities clearly defined. To evaluate intervention fidelity, a trained observer conducted integrity checks by completing a checklist of correctly delivered instructional content and activities during each group session. These data were summed and averaged across all items per session and reported as a percent score (total items endorsed as delivered divided by total items on the session fidelity form).

Feasibility and acceptability

Feasibility measures included a number of program adherence measurements calculated as (1) the number of sessions attended (simulator sessions and group sessions), (2) the percentage of Act the Fact practice assignments completed, and (3) the percentage of in-class activities completed.

Acceptability was assessed by using satisfaction surveys and qualitative semi-structured interview questions.

The Participant and Parent Satisfaction Surveys²⁶ were developed for this program and included five items using a 10-point Likert scale (10 = excellent or very helpful). Participants and parents rated overall satisfaction with the group (e.g., “On a scale of 1 to 10, 10 being excellent, how would you rate the CBID program overall?”), as well as specific satisfaction with the simulator practice and curriculum.

The Participant Semi-Structured Qualitative Interview had questions gathering feedback about the program (i.e., best and worst parts of the program) along with questions asking about participant changes (Participant semi-structured interview on driving-related changes). A total of seven questions asked about program acceptability probing for both positive and negative aspects of the program (e.g., Do you NOW feel like you want to be in this program? Was this program worthwhile? Is there anything you like about the simulator? Anything you don't like about it?). Responses were coded as positive or negative per response and quantified. Individual participant quotes were retained as exemplars of coded endorsements.

Participant semi-structured interview on driving-related changes

This participant semi-structured qualitative interview was developed for this study by the first author and asks about changes in driving anxiety, attitudes toward driving, driving goals, and driving-related behaviors in addition to acceptability/satisfaction (Participant Semi-Structured Qualitative Interview). Responses were written verbatim and coded as positive or negative change in each particular area (e.g., emotions, thoughts, attitudes, behaviors) based on the participant response. The code was verified during the interview with the participant (e.g., “You now have a positive attitude about driving? Yes? You now understand the complexities of driving and skills necessary to drive? A realistic understanding, Yes?). All interviewers participated in a didactic training and received practice with feedback to be consistent in data collection and coding. All responses were recorded verbatim at the time of interview. Participant quotes were retained categorized and provided as illustrative examples of positive and negative change codes.

State–Trait Anxiety Inventory

This 40-item self-report inventory assesses symptoms of anxiety.²⁷ The first 20 items target state anxiety, which relates to current anxiety in that situation, and the last 21–40 items focus on trait anxiety, which relates to anxiety in general. Responses are rated on a 1–4 scale; 1 being “not at all” and 4 being “very much so.” Total scores for each category, state and trait, range from 20 to 80. Psychometric properties are established and reported,²⁸ and the measure has been shown to remain stable when measuring stress in youth with autism.²⁹

Driving Cognitions Questionnaire

This is a 20-item self-report scale that measures driving-related concerns, rated 0 to 4: panic-related (7 items), accident-related (7 items), and social concerns (6 items), as well as a total score.³⁰ This questionnaire shows good internal consistency and convergent validity, and relevant correlations with measures of driving fear severity in nonautistic samples.³⁰

Driving Attitude Scale

This 18-item scale focuses on the participants' excitement or nervousness when talking about driving, preparing to drive or while driving, and is completed by the participant and parent.¹⁹ Item responses are on a Likert scale of “Not at all” (0) to “A Lot” (3) with nine positive and nine negative items. The parent report is 20 items long with 2 extra questions regarding their child's ability to drive safely in the upcoming months and their motivation to obtain a driver's license. The measure has established good internal consistency for parents of children with autism.⁸

Driving simulator performance

Driving simulator data include six key driving violations: (1) off-road crashes, (2) vehicle collisions, (3) pedestrian collisions, (4) tickets received, (5) speeding, and (6) crossing the centerline. The number of off-road crashes includes veering off of the road and into the scenery. The number of vehicle collisions includes hitting any vehicle parked or driving on the road. The number of pedestrian collisions includes hitting pedestrians who illegally cross the street or cross at an unprotected light, and hitting animals that cross the road. The number of tickets received includes illegal turns, running a red light or a stop sign, and speeding. The amount of time the speedometer is over the speed limit is calculated as the percentage of time speeding. A few miles over may not trigger a ticket, but it was counted for this category. Finally, the percentage of time crossing the centerline includes all left turns that crossed the centerline and crossing while on a single lane road. These six violations were assessed at the baseline “Long drive test A” and at post-intervention “Long drive test B.” These unique 20-minute driving tests are designed to be equivalent for direct comparisons.

Real-world outcomes

Two months after the completion of the program, participants were contacted by phone, e-mail, or text and asked if a driver's permit and/or driver's license had been obtained. Responses were calculated as the percentage of permits and licenses obtained.

Data analysis

Paired-samples t-tests were conducted on baseline and post-intervention scores for each domain and subscales of each measure to determine if differences were statistically significant. Within-group effect sizes (Cohen's d) were calculated for all scores. Because this was a pilot study to assess the potential value of the intervention with a small community sample, the Type I error rate was not adjusted for the number of comparisons and was kept at p < 0.05 level for each comparison. Qualitative data were coded during the interview process and later quantified as total number of participants endorsing an item and converted to percent scores.

Results

Program fidelity

The CBID curriculum was implemented with high fidelity, with the mean rating of 97% (range 92%–100%) of sessions rated as completing all content and activities across two groups.

Feasibility and acceptability

Of the 19 participants, 17 (89%) completed the entire 10-session program; 2 adults dropped after session 3. Session attendance levels were high, with 53% of the participants attending every session and another 30% missing only one session. A total of three participants missed two to three sessions. In-session assignment completion was 100% for all participants. Homework assignment completion was 89% completion overall (range = 51%–100%).

In terms of acceptability, overall satisfaction ratings were in the very high range with mean score of 8.24 (SD = 1.92) for participants and 9.13 (SD = 0.74) for parents out of 10. Regarding program helpfulness, participants rated 8.41 (SD = 1.9) and parents rated 9.07 (SD = 1.0). Participants also rated high satisfaction with both the CBID curriculum (μ = 7.65, SD = 2.15) and the coached driving simulator practice (μ = 8.18, SD = 1.94). Parents had slightly higher satisfaction ratings (μ = 9.33, SD = 0.82 for the curriculum and μ = 9.13, SD = 0.83 for simulator practice sessions).

Data quantified from qualitative interviews indicated that 82% of the participants reported that the program was worthwhile. Almost all participants (94%) indicated that the program helped them advance in driving, and the majority (82%) reported positive feelings and experiences with the driving simulator. About half of the participants (47%) indicated their favorite part of CBID was the driving simulator, and others (47%) reported their favorite part as watching the videos (driving videos incorporated into curriculum). Participants reported few negative responses about CBID even when prompted (e.g., What's your least favorite part and why?) with 30% of participants reporting only positives. Refer to Table 3 for a sample of participant's positive quotes from qualitative interviews. Negative comments were collapsed into three main categories: (1) individual group sessions too long and duration of program too short (24% of all negative comments) with one participant stating “Schedule more meetings but make them shorter. Meetings were too long,” (2) feeling bothered by other participants' behaviors in session (24%) illustrated by this quote, “distracting people in the group,” and (3) hard work (12%) by commenting “the hard work—act the fact, what do you do situations” (naming two examples of work activities).

Table 3.

Qualitative Feedback from Autistic Participants About the Cognitive Behavioral Intervention for Driving Program

Participant
“It's a safe environment when you don't have to worry about real crashes. I am not scared anymore.”
“It's a nice tool that helped me have a live example; it made me more confident about challenging driving situations.”
“My favorite part is trying to get a better chance at when I drive for real.”
“I like seeing the road through the screen and recognize the actions I am taking.”
“My favorite part was the reviewing of the simulator results because it's calm. It's important to be able to measure progress and move towards the right steps.”
“I got great improvement in reaction time over the weeks, which will help me in real life driving.”
“I now think I can drive; I feel proud of myself.”
“I've started taking actual steps towards my goal instead of just getting lost in the idea.”
“Now I am actually actively working towards learning how to drive.”
“I have less negative thoughts now that I know what kind of bad things can happen thanks to the driving simulator.”
“More positive. I don't obsess over what could happen but I focus on what I'm good at. I understand more what my real skills are and what I need to work on.”
“I am feeling more confident. I'm not letting the anxiety overwhelm me, but I now have control over it and I am thinking more objectively.”
“I went from irrationally cautious to now rationally cautious. At first, my fear was exaggerated; now I understand and know what to expect.”
“I would highly recommend the program to anyone who is interested.”
“The pack of instructional videos given at home was integral to my driving education.”

Driving-related changes in attitudes, anxiety, goals, and behaviors

Per coded qualitative interview data, 100% of the participants reported both a positive attitude change (interest toward driving) and a desire to drive in the future at post-intervention. The majority of participants (67%) reported an emotional change (less anxious/apprehensive feelings about driving). Most also reported changes in thoughts about driving from negative to positive (88%) and from idealistic to realistic (i.e., better understanding of the complexities of driving) (94%). Many participants (71%) reported positive changes in driving goals post-intervention providing examples as “more clear goals,” “taking driving tests now,” “practicing driving,” and “getting a license.” All the participants (100%) reported behavior changes by participating in one or more driving-related activity toward their driving goal (i.e., behind wheel driving, studying for driving examination, enrolling in course, reading DMV book, taking driving test), with an average of 5.4 new activities completed. When asked about what other positive changes they noticed in themselves after CBID participation, 94% endorsed one or more changes. Examples from the quotes include: “feeling more confident,” “learned to be more patient,” “more calm,” “feels a lot relaxed,” “feel proud of myself,” and “more responsible.” Coded qualitative statements indicated positive impacts as a result of driving simulator practice including: (1) increased confidence, (2) increased driving skills, and (3) appreciation for the opportunity to develop skills in a risk-free context.

Anxiety and driving cognitions

Participant anxiety levels are reflected through the State–Trait Anxiety Inventory and the Driving Cognitions Questionnaire (DCQ). As shown in Table 4, participants reported no significant difference for current (state) or general (trait) anxiety. However, participants reported fewer driving-specific concerns on the DCQ. There were significant improvements on the social concerns subscale (p = 0.01, d = 0.71) and the total score (p = 0.02, d = 0.62).

Table 4.

Univariate Effects of Cognitive Behavioral Intervention for Driving Program on Anxiety and Concerns of Autistic Participants

n = 17	Baseline		Post-intervention		Analysis
	M	SD	M	SD	t	p	d
State–Trait Anxiety Inventory
State	41.00	11.03	40.47	11.84	0.20	0.84	0.05
Trait	42.29	10.29	40.47	10.96	0.94	0.36	0.23
Driving Cognitions Questionnaire
Panic-related concerns	6.59	5.06	4.12	2.29	2.02	0.06	0.49
Accident-related concerns	9.00	7.45	6.94	6.04	1.31	0.21	0.32
Social concerns	7.65	6.44	4.29	3.41	2.94	0.01*	0.71
Total scores	23.24	16.44	15.35	10.90	2.55	0.02*	0.62

Lower scores indicate fewer concerns. Lower scores indicate less anxiety and fewer concerns. Within-group effect sizes, Cohen's d, indicate small >0.2, medium >0.5, and large >0.8.

^* p < 0.05.

Driving attitude

Both participants and parents reported increases in positive attitudes toward driving. There were significant improvements in positive attitudes when getting ready to drive (p < 0.01, d = 1.10) reported by the participants and when driving (p < 0.01, d = 1.83) reported by the parents. The positive attitude total score increased significantly for participants (p = 0.03, d = 0.74). Additionally, both participants and parents reported decreases in overall negative attitudes toward driving with small to moderate effect sizes (participants d = 0.26, parents d = 0.56). Participants reported significantly less negative attitude when talking about driving (p = 0.03, d = 0.58) and parents reported decreases without reaching significance (p = 0.08, d = 0.60) (Table 5).

Table 5.

Univariate Effects of Cognitive Behavioral Intervention for Driving Program on Autistic Participant- and Parent-Reported Driving Attitude Scale

Driving Attitude Scale	Participant							Parent
	Baseline		Post-intervention		Analysis			Baseline		Post-intervention		Analysis
	M	SD	M	SD	t	p	d	M	SD	M	SD	t	p	d
Positive attitude
When talking about driving (n = 16)	4.06	2.32	4.63	2.63	−1.21	0.25	0.30	4.47	2.97	5.53	1.69	−1.55	0.14	0.40
When getting ready to drive (n = 14)	3.57	2.31	5.21	1.97	−4.10	0.00**	1.10	3.67	2.50	4.11	2.37	−0.61	0.56	0.20
When driving (n = 12)	6.17	2.13	7.17	1.27	−1.41	0.19	0.41	4.17	1.72	6.17	1.84	−4.47	0.00**	1.83
Total positive score	14.25	4.71	17.50	3.90	−0.45	0.03*	0.74	15.33	3.08	16.00	5.51	−0.41	0.70	0.17
Negative attitude
When talking about driving (n = 16)	4.00	2.66	3.00	2.16	2.34	0.03*	0.58	4.15	2.73	2.85	1.28	1.92	0.08	0.53
When getting ready to drive (n = 14)	2.50	2.47	2.07	2.13	0.70	0.49	0.19	4.20	2.49	2.60	1.35	1.89	0.09	0.60
When driving (n = 12)	2.83	2.52	2.25	1.60	0.86	0.41	0.25	4.29	2.63	3.00	0.82	1.12	0.31	0.42
Total negative score	8.17	6.35	6.67	5.18	0.91	0.39	0.26	11.50	8.02	7.17	2.48	1.37	0.23	0.56

Higher scores on positive attitude and lower scores on negative attitude indicate improvement. Within-group effect sizes, Cohen's d, indicate small >0.2, medium >0.5, and large >0.8.

^* p < 0.05; ^**p < 0.01.

Driving behaviors

On the driving simulator, participants demonstrated meaningful improvements on driving behaviors by reducing errors across all key violations. Three of six key driving violations showed statistically significant improvements on participants' performance: pedestrian collisions (p = 0.02, d = 0.65), speeding (p = 0.03, d = 0.59), and centerline crossing (p < 0.05, d = 0.52) (Table 6).

Table 6.

Univariate Effects of Cognitive Behavioral Intervention for Driving Program on Performance-Based Driving Simulator Experience of Autistic Participants

n = 17	Baseline		Post-intervention		Analysis
	M	SD	M	SD	t	p	d
Driving simulator experience
Off-road crashes	0.35	0.10	0.00	0.00	1.46	0.16	0.35
Number of vehicle collisions	4.47	5.65	2.06	2.28	2.01	0.06	0.49
Number of pedestrian collisions	2.12	1.11	1.47	0.94	2.68	0.02*	0.65
Number of tickets	1.47	1.46	1.12	1.22	0.86	0.40	0.21
% Speeding	9.58	12.24	5.41	8.17	2.42	0.03*	0.59
% Centerline crossing	8.12	4.95	6.06	2.51	2.14	0.049*	0.52

Lower scores indicate better performance. Within-group effect sizes, Cohen's d, indicate small >0.2, medium >0.5, and large >0.8.

^* p < 0.05.

Real-world outcomes

At the start of the program, two participants (12%) had a driving permit and 0% had a driver's license. Within 2 months post-intervention, eight participants (42%) obtained a driver's permit and one (5%) obtained a driver's license for a total of 47% demonstrating positive functional outcome.

Discussion

This open trial study tested the newly developed CBID program. CBID combines CBT with graded exposures and Cognitive Enhancement Training into a group intervention with individualized driving simulator practice and coaching feedback to teach EF, ER, and motor skills necessary for driving.

The CBID curriculum taught EF and ER skills through didactic and experiential learning methods as recommended from previous research,³¹ laying the foundation for acquiring other driving skills learned during simulator practice.⁶ CBID responds to community driving instructors' previously expressed need to establish best practices for teaching autistic adolescents and adults to drive.³² The program was delivered over a 10-week period and included driving simulator practice, giving the autistic teen/adult extended time to learn and apply the material. The 5 weeks of simulator practice included in CBID far exceeds any previous study.^{6,7,11,21,22,25}

The initial results of CBID are promising, with high feasibility, acceptability, and initial estimates of significant outcomes on driving cognitions and behaviors. Positive outcomes include: reduced self-reported anxious thoughts, increased positive attitudes toward driving, and improved simulated driving performance. Real-world outcomes were achieved, with about half of the participants obtaining a driver's license or permit within 2 months post-intervention.

There is strong support in the previous literature and in the findings of this study for the use of exposure therapy through a driving simulator when treating driving-related anxiety.^6,8 Exposing participants to a controlled anxiety-provoking situation helps to eliminate danger and increase habituation to the situation, decreasing anxiety over time.⁹ Semi-structured interviews with driving instructors working with autistic individuals found that individualization of instructional strategies was key.³² Myers et al. further suggested that simulators enhance driving instruction because of the ability to individualize the experience and by allowing a means for autistic learners to safely practice challenging driving skills.³²

For autistic individuals, this intervention also appeared to elicit openness to the idea of driving and pursuing driving goals in the future, which could also be viewed as a positive outcome.

The limitations for the study include: a small sample size that is lacking in race/ethnic diversity, lack of control group, lack of measurement of EF skills, and a lack of long-term follow-up. Future studies are needed to replicate the effectiveness of CBID with a larger, more diverse sample, and to measure long-term effects of the CBID program for autistic individuals obtaining driving independence.

In conclusion, the CBID has high feasibility and acceptability as well as potential efficacy for improving driving attitudes, behaviors, and licensure rates. With increased driving comes increased self-reliance. The increased independence resulting from obtaining a driver's license could potentially have long-lasting effects on employment and social outcomes for autistic individuals. This intervention could be provided through vocational or independent living services and could potentially increase the number of autistic individuals driving substantially. Additional research and development of training programs to enhance driving outcomes in autistic teens and adults is needed.

Author Confirmation Statement

M.J.B.-E. as PI contributed to the study conception and design, wrote the curriculum, conducted the intervention groups, assisted with data collection, analysis, and interpretation of results, and prepared article for publication. L.S. contributed to interpretation of results and preparation of the article for publication. A.T. contributed to data management, analysis and interpretation of results, and preparation of the Results section and data tables for publication. K.S. contributed by providing feedback of the CBID intervention as a facilitator and preparation of the article for publication. All authors have reviewed and approved the article before submission. The article has been submitted solely to this journal and is not published, in press, or submitted elsewhere.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study could not have been possible without the generous funding support from the Doug Flutie Jr. Foundation for Autism and National Foundation for Autism Research to Dr. Mary J. Baker-Ericzén (PI).