Improving Access to Community Occupations via Rideshare: A Pilot Study With Autistic Adults

Abstract

Importance: Adults with autism often face barriers to community access for work and leisure, yet few programs address transportation independence and safety. This study evaluated the Safe Rideshare Program (SRP), designed to teach autistic adults to use rideshare services.

Objective: To assess acceptability, appropriateness, and the potential impact of the SRP and to evaluate changes in rideshare use for community access over a 2-mo follow-up period.

Design: A one-group, pretest–posttest study.

Setting: Community.

Participants: Thirteen autistic adults.

Intervention: The SRP consisted of five virtual training modules completed synchronously with project trainers via Zoom, followed by eight ride-along sessions using either Uber or Lyft rideshare services.

Outcomes and Measures: Intervention acceptability was rated by participants using the Acceptability of Intervention Measure. Appropriateness of the intervention was assessed by trainers using the Intervention Appropriateness Measure. A 30-item Rideshare Independence Checklist was used to assess independence and safety using rideshare services.

Results: Of the 13 enrolled participants, 12 completed the program with strong acceptability and appropriateness ratings (≥75% threshold was met). Independence and safety significantly improved during the intervention, and at the 2-mo follow-up, 8 participants reported increased independent rides (p = .011).

Conclusions and Relevance: The SRP demonstrated preliminary acceptability and appropriateness, as well as the potential to improve transportation independence, safety, and community access for autistic adults.

Plain-Language Summary: Many autistic adults face challenges accessing their communities for work, shopping, or social activities because of transportation barriers. Learning how to use rideshare services, such as Uber or Lyft, can help address this issue, but few programs are designed to teach these skills safely and effectively. To fill this gap, our team developed the Safe Rideshare Program (SRP) and tested its usefulness. Thirteen autistic adults, ages 18 to 24 yr, participated in this study. Most participants completed the entire program and gave it high ratings for being easy to follow and helpful. Trainers who worked with the participants also reported that the program was useful. By the end of the training, participants showed significant improvements in their ability to use rideshare services independently and safely. Two months after finishing the program, many participants continued to use rideshare services. On the basis of our preliminary findings, we believe that the SRP shows promise in helping autistic adults overcome transportation barriers and gain more independence. With improved access to transportation, participants were able to engage in more community activities, which can improve their quality of life and open up new opportunities for engagement.

The authors developed the Safe Rideshare Program and tested its usefulness in teaching autistic adults how to use rideshare services.

Community mobility is critical for living independently and engaging in meaningful social occupations. It is especially important for people in their early adult years, as this is often a time of transition to employment and living independently (Belongia & Reynolds, 2021). Community mobility can be particularly challenging for adults with autism spectrum disorder (ASD) (hereinafter referred to as autistic adults on the basis of the preferred identity-first language of our autistic partners). Some autistic adults are unable to meet the demands of driving because of deficits in spatial perception, sustained attention, speed of information processing, social problem solving, and responding to unexpected changes (Gellar & Greenberg, 2009; Lindsay, 2017). Others are capable of driving but prefer not to because of the anxiety that driving evokes (Chee et al., 2015). Public transportation is an option for autistic adults, and autistic adults are more likely to use public transportation than their nonautistic counterparts (Lindsay, 2017). However, using public transportation may be just as challenging as driving for the autistic population. Limitations in executive functioning (planning, time management, navigation, safety) and social interaction skills (failure to initiate, preference for repetitive movement, difficulty adjusting responses) have been identified as barriers to the effective use of public transportation by this population (Radloff, Kaminski, & Dickerson, 2016). Public transportation can also be unreliable, or inaccessible, in some regions.

Another issue voiced by our autistic community partners is that public transportation, as well as community vans or rides from parents, often restrict the types of occupations they are able to pursue. Essential activities such as getting to medical appointments, work, or the supermarket are generally prioritized over going to the video game store, grabbing a hamburger, or meeting up with a friend. Transportation may need to be coordinated in advance, which limits the ability for autistic adults to be spontaneous and get where they want, when they want. Engagement in leisure, social, and entertainment activities often suffers, and, as one direct support counselor summarized, “their world may become very small” (S. Ledford, personal communication, February 26, 2020).

Rideshare services (e.g., Uber, Lyft) are a relatively new, and widely available, form of transportation in which passengers get from Point A to Point B in private vehicles driven by their owners. A digital app, usually accessed on a smartphone, matches passengers and drivers, coordinates routes using a GPS, and facilitates payment through a linked financial account. Rideshare services have the potential to address many of the issues that autistic adults have in accessing the community. It is faster and more direct than the public train or bus, there is often limited social interaction required, and rides can be scheduled at any time. Rideshares also allow for spontaneity, because a ride can be arranged in the moment, at any time of day. A rideshare vehicle picks passengers up and drops them off at their specific destination, eliminating the need to walk to bus stops, wait at stations, or transfer to more than one bus or train. Not only does this save time, but it also may increase safety, especially at night or during inclement weather. Finally, although communicating with the driver may be a challenge, there is only one person with whom to interact versus potentially many people waiting at stations or on the bus or train. Rideshare use has the potential to improve community access and expand occupational engagement for autistic adults, but few training resources exist to introduce clients to this mode of transportation, assess readiness, or build confidence.

The purpose of this study was to pilot a rideshare training program designed to teach, and support, autistic adults to access their communities by safely using rideshare services. Our aims were to assess the acceptability, appropriateness, and potential impact of the Safe Rideshare Program (SRP) and to evaluate changes in rideshare use for community access over a 2-mo follow-up period.

Method

Study Design

To achieve our study aims, we used a one-group, pretest–posttest, repeated-measures design with both implementation and impact measures. The study procedures were approved by the Virginia Commonwealth University Institutional Review Board, and the study protocol was registered with ClinicalTrials.gov (ClinicalTrials.gov Identifier No. NCT06323850).

Sample

To be included in the study, participants had to be older than age 18 yr and living within 100 mi of Richmond, VA; this included urban, suburban, and rural regions. Participants had to have a diagnosis of ASD, given by a licensed medical professional. Participants needed to have a stable internet connection, the ability to communicate independently in English (verbally or with an alternative communication device), have the motor skills needed to independently use a smartphone (fine motor skills) and get in and out of a car (gross motor skills). Participants needed to own a smartphone and be familiar with texting, email, or both. We excluded participants who had a severe or profound intellectual disability (an IQ <35), given that the training was designed for those with the potential to travel alone safely to familiar places. We also excluded any participants who already uses rideshare services independently. Rideshare use before the start of the study and before initiation of the intervention (for those in the waitlist condition) was validated verbally or through email by the participant’s parent or legally authorized representative.

We selected a target sample of 12 participants for this pilot study. According to Moore et al. (2011), a sample size of 12 participants is ideal for preliminary studies demonstrating the ability to execute a specific research protocol and that focus on estimating average values and variability in preparation for planning larger subsequent studies. Following the “rule of 12,” this sample size makes a profound difference in the width of confidence intervals while still being practical to utilize at a single site and providing valuable information. We used convenience sampling to enroll the first 12 participants who met our inclusion criteria and consented to participate in the study. We overenrolled by an n of 1 in anticipation of possible attrition; therefore, our total participant enrollment for this study was 13.

Intervention

Rideshare Intervention Development

The SRP was developed in partnership with residents of a semi-independent living community for autistic adults (Faison Residence). Before initiation of this study, the program went through two phases of development and feasibility testing with residents at this site. (See Table A.1 in the Supplemental Material, available online with this article at https://research.aota.org/ajot.) Phase 1 of program development included a needs assessment and environmental scan; we conducted interviews with autistic adults and caregivers. On the basis of Phase 1 findings, we developed four online modules: Rideshare Basics, Starting the Ride, During the Ride, and Ending the Ride. The modules used video modeling and active tasks to educate and reinforce concepts. A research team member used Zoom to review each module with three autistic adults and get feedback about its design and usefulness. A training manual was developed and added to a website (https://sites.google.com/view/saferideshareprogram) to share resources. Phase 2 of program development included the creation of a fifth training module on money management and a tool to evaluate changes in independence and the safety of participants while completing ride-along sessions with trainers. We enrolled five autistic adults to evaluate the feasibility of the program and evaluation tool. Phase 2 outcomes suggested that the program was feasible, and by the sixth ride-along session, three participants were able to use rideshare services safely without cueing; the other two participants needed only five to six cues to complete the process (Hunter, Sedelmeyer, & Reynolds, 2023). On the basis of the outcomes of Phase 2 feasibility testing, we proposed the present pilot study to assess the acceptability, appropriateness, and potential impact of the intervention.

Intervention Overview

The SRP currently has five modules: Intro to Rideshare, Starting the Ride, During the Ride, Ending the Ride, and Money Management. During the intervention phase, participants first completed the online module with virtual support from our trainers. Successful completion of the online module was followed, usually in the same week, by an in-person, ride-along evaluation. The ride-along sessions were tailored to the participant’s communication preferences (e.g., verbal, augmentative communication) and learning pace; modules could be repeated within the 2-mo intervention timeframe as needed. The intervention was delivered by trained research staff, with a primary trainer delivering the content and a secondary support trainer collecting data and scoring sessions for fidelity. Parents or caregivers were told that they could observe sessions, but they were asked not to prompt the clients for fidelity purposes. A specification of the intervention using the Rehabilitation Treatment Specification System (Hart et al., 2024) is available (see Table A.2 in the Supplemental Material).

Procedure

Intervention Training

During the first month of the study, research staff (trainers) participated in three in-person training sessions, approximately 90 min each. Trainers were hired from a pool of graduate occupational therapy students from the College of Health Professions and employment specialists from the Rehabilitation Research and Training Center at Virginia Commonwealth University. Trainers had to have experience working with autistic adults, hold a bachelor’s degree, and be experienced with using rideshare. Trainers worked in pairs conducting the virtual and ride-along sessions. Training was led by the co–principal investigators (Stacey Reynolds and Alissa Molinelli Brooke), along with one autistic adult (Brandon Jenkins) who served as a paid research partner. Training included topics related to protocol adherence, cueing and recording cues, assessing progress, and working within the autistic community. The research team met every 2 wk during the course of the study, and additional training and support was provided on an as-needed basis.

Intervention Phase

Virtual training sessions.

Virtual sessions took place over Zoom at a time that was convenient to participants. Each session lasted approximately 20 to 30 min. Each of the five virtual sessions was aligned with one of the five online modules (https://sites.google.com/view/saferideshareprogram). For each module, exercises were built in to test comprehension. If the participant did not complete an exercise accurately, the lead trainer explained why the response was incorrect and prompted them to repeat the video modeling exercise. The support trainer took notes to support our formative assessment process and complete fidelity checks.

Ride-along training sessions.

Ride-along sessions were initiated at a location selected by the participant, usually the home. Participants were allowed to choose where they wanted to take the rideshare within a 10- to 15-mile radius of their starting location. The first ride-along session was the baseline session for assessing level of independence and safety. The participant then completed seven additional ride-along sessions, with the lead trainer providing cues as needed to complete each step of the rideshare process. The support trainer rode in the car with the participant and lead trainer to complete assessments and keep a record of any issues that arose during the ride. Only the ride to the community destination was evaluated for independence and safety; the rideshare was hailed to return to the starting location (e.g., home), providing additional practice opportunities for the participants, but no evaluation was completed.

Two-mo follow-up.

We attempted to quantify the number of times the participant used rideshare services after receiving training. Participants were instructed to either “share trip” with the research team using the rideshare app or to text the trainers when they were taking a ride. Participants were also asked to share what types of activities they were engaging in (e.g., get food, go to work). If the trainers did not hear from a participant after 1 week, they sent a text message or email asking them to state whether they had used rideshare services that week and to provide details on the activities.

Measures

On the basis of Proctor et al.’s (2011) implementation outcomes framework, implementation outcomes often preceded both service outcomes and client outcomes. In this study, we selected implementation outcomes that we felt were the best indicators of the implementation success and could be proximal indicators of implementation processes. Specifically, for Aim 1, the primary implementation outcome measures were acceptability of the intervention and the appropriateness of the intervention. We also assessed the fidelity of the virtual modules using an investigator-developed fidelity checklist; this was part of a formative evaluation process. Our primary service and client measures for this aim were functional independence and safety using rideshare services. We assessed independence and safety using the Rideshare Independence Checklist (RIC). See Figure 1 for an overview of the study outcomes mapped onto Proctor et al.’s model.

Figure 1.

Overview of the study outcomes mapped onto Proctor et al.’s (2011) implementation outcomes framework.

Note. Text in red borders indicate the outcomes measured in this study. The client outcome of function was indicated by changes in rideshare independence scores over time. Adapted from “Outcomes for Implementation Research: Conceptual Distinctions, Measurement Challenges, and Research Agenda,” by E. Proctor, H. Silmere, R. Raghavan, P. Hovmand, G. Aarons, A. Bunger … M. Hensley, 2011, Administration and Policy in Mental Health Services Research, 38, p. 66. Copyright 2011 by Springer. Adapted with permission.

Acceptability of Intervention Measure

We assessed the acceptability of the intervention using the Acceptability of Intervention Measure (AIM; Proctor et al., 2011; Weiner et al., 2017). On the AIM, participants rated four prompts on a response scale ranging from 1 to 5 (1 = completely disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = completely agree). The four prompts on the AIM were as follows (prompts were tailored for the SRP):

• The Safe Rideshare Program meets my approval.

• The Safe Rideshare Program is appealing to me.

• I like the Safe Rideshare Program.

• I welcome the Safe Rideshare Program.

Intervention Appropriateness Measure

We assessed the appropriateness of the intervention using the Intervention Appropriateness Measure (IAM), which was completed by our trainers (Weiner et al., 2017). For each participant they worked with, each trainer rated four prompts on a response scale ranging from 1 to 5 (1 = completely disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = completely agree). The four prompts on the IAM were as follows (prompts were tailored for the SRP):

• The Safe Rideshare Program seems fitting for [participant’s name].

• The Safe Rideshare Program seems suitable for [participant’s name].

• The Safe Rideshare Program seems applicable for [participant’s name].

• The Safe Rideshare Program seems like a good match for [participant’s name].

Both the AIM and the IAM have demonstrated strong psychometric properties, including content validity, structural validity, discriminant validity, and reliability in a series of studies conducted by Weiner et al. (2017).

Fidelity Checklist

Our team developed a fidelity checklist (Table A.3) to assess the components of the SRP virtual training. Each virtual session was assessed for fidelity; any fidelity scores below 100% were reviewed with the team, and necessary adjustments were made.

RIC

Our team developed a 30-item RIC (Table A.4) to assess independence and safety using rideshare services. Trainers were instructed to check “yes,” “no,” or “with cues” to indicate how the participant completed each action. “Yes” was selected if the participant performed the action from start to finish independently. “With cues” was selected if the participant was able to complete the action but required cues to complete it. “No” was selected if the participant was unable to complete the action, even with cues. The total RIC Independence score reflected all 30 items on the tool. Ten items on the RIC related specifically to safety, and we examined them separately to generate a total safety score:

• Item 9: Shares ride information with someone else

• Item 10: Waits in safe and appropriate location until ride arrives

• Item 11: Checks vehicle license plate number

• Item 14: Double checks name with driver

• Item 17: Buckles seatbelt

• Item 19: Does not share any personal information with the driver

• Item 20: Checks map during ride

• Item 21: Ensures they have arrived at the correct destination

• Item 22: Waits for the car to stop to exit

• Item 28: Enters the desired location safely

Statistical Analysis Plan

Our first hypothesis for Aim 1 was that autistic adults participating in the SRP would rate 75% of the items on the AIM as “agree” or “completely agree.” We used descriptive statistics (percentages) to test this hypothesis. Our second hypothesis for Aim 1 was that trainers conducting the SRP intervention would rate 75% of the items on the IAM as “agree” or “completely agree.” Again, we used descriptive statistics to test this hypothesis.

The third hypothesis for Aim 1 was that autistic adults who participated in the SRP would display greater independence over time. We modeled the total number of items that participants could perform independently using Poisson regression with a fixed session effect and a random participant effect. Session was treated as a continuous measure and modeled using a restricted cubic spline with three knots to capture a potential nonlinear relationship between time and independence score. In addition to testing coefficient estimates from the model, we used the model to predict the independence score for a new participant at each session. We used bootstrapping methods to obtain 95% prediction intervals.

The fourth hypothesis for Aim 1 was that autistic adults who participated in the SRP would display greater safety over time. The safety score was modeled using the same methods used for the independence score.

For Aim 2, we hypothesized that we would see a statistically significant change in rideshare use between the number of independent rideshare rides taken in the 2-mo before enrolling in the study (which was 0 for all participants) and the number of independent rideshare rides taken during the 2-mo follow-up period. We used a nonparametric Wilcoxon signed-rank test to test this hypothesis comparing the number of rideshares taken in sum during each phase.

Results

Participants

Thirteen participants were enrolled, with an average age of 21 yr (minimum = 18, maximum = 24). Eight of the participants identified as male, two as female, and one as “other” (nonbinary, transgender–transitioning, and prefer not to disclose were also options on the demographic form). The sample was primarily Black or African American (n = 6), with the remainder of the sample identifying as White (n = 5), Asian (n = 1), or Hispanic (n = 1). Most of our participants lived in suburban areas (92%). Before enrollment in this study, none of the participants had ever used, or attempted to use, rideshare services independently. Co-occurring diagnoses were recorded, the most frequently reported being anxiety (n = 6), attention deficit hyperactivity disorder (n = 5), and intellectual disability (n = 2). Sixty-nine percent of the sample reported spending at least some time in the community every week, and 38% either worked or volunteered. A full summary of participant demographics is presented in Table 1.

Table 1.

Sociodemographic Characteristics of Participants

Baseline Characteristic	n	%
Gender
Female	4	31
Male	8	61
Other	1	8
Race
American Indian/Alaska Native	0	0
Asian	1	8
Black/African American	6	46
Hispanic/Latino	1	8
Native Hawaiian or Pacific Islander	0	0
White	5	38
Co-occurring diagnoses
Anxiety	6	46
Depression	0	0
Bipolar disorder	0	0
ADHD	5	38
Learning disability	0	0
Intellectual disability	2	15
Seizure disorder	1	8
Sleep disorder	0	0
Sensory processing disorder	0	0
Cerebral palsy	0	0
Down syndrome	1	8
Rett syndrome	0	0
Fragile X syndrome	0	0
Eating disorder	0	0
Feeding–eating challenges	1	8
Gastrointestinal issues	1	8
Other	2	15
None	2	15
Hours per week spent in the community (alone or with others)
1–5	5	38
6–15	3	23
16–25	1	8
≥25	0	0
I do not do things in the community very often	4	31
Hours per week spent working or volunteering
1–5	1	8
6–15	3	23
16–25	1	8
≥25	0	0
I do not work or volunteer	8	61
Residence location
Urban	0	0
Suburban	12	92
Rural	1	8

Note. All data were self-reported by participants. ADHD = attention deficit hyperactivity disorder.

Of the 13 participants who were enrolled in the study, 12 completed all training components and one was withdrawn by his caregiver after completing four training modules and five ride-along sessions. The reason for withdrawal was initially reported as illness, but the reason that was later given was that participation was inconvenient for the family. Data from this participant were kept in the analysis; however, not all assessment measures were completed (i.e., the AIM, IAM, and 2-mo follow-up ride data).

Intervention Fidelity and Descriptives

Twelve of the subjects completed all five online modules; one participant only completed four modules before withdrawing from the study. Two participants repeated one module, and one participant repeated two modules. Fidelity to the online intervention was met for each training module conducted, with 100% “yes” responses for each question, for each module.

Of the 101 ride-along sessions conducted over the course of this study, 45 were to get food or groceries, 45 were to stores or shopping centers, five were to libraries or bookstores, three were to the participant’s workplace, and one ride each was taken to a school (community college), entertainment venue (arcade), and other (army base).

Implementation Outcomes

Acceptability of Intervention

Ten participants completed all four questions on the AIM; there were no incomplete forms. Of the 40 total responses gathered, 39 fell into the “agree” or “completely agree” categories (97.5%), meeting our 75% threshold for acceptability. The participants were also invited to provide narrative comments to explain their AIM scores. Several comments indicated that having a trusting and positive relationship with their trainers was very important to them and that the trainers were patient and understanding of their specific needs. Two participants also noted that they wished the program could be extended.

Appropriateness of Intervention

The trainers completed all four items on the IAM for each participant they worked with, resulting in a total of 104 responses: (2 trainers × 13 participants) × 4 items on the scale. Of the 104 total responses gathered, 79 fell into the “agree” or “completely agree” categories (75.9%), meeting our 75% threshold for appropriateness.

Trainers were also invited to provide narrative comments explaining their AIM scores. Both trainer groups described an incident in which a rideshare driver spoke rudely or displayed behaviors that were suggestive of a clear negative bias toward the participant. In one instance, trainers asked the driver to stop the car mid-ride because they felt unsafe. Additional comments addressed the social customs surrounding rating and tipping drivers at the end of a ride. Several trainers noted that participants were unfamiliar with the unwritten expectation to tip or give a high rating, even for standard service. Because the appropriate amount to tip and the criteria for a “five-star” rating are not explicitly stated, participants found these decisions ambiguous and challenging. Trainers further reported that participants’ tipping and rating decisions often reflected their sensory experiences during the ride. For example, some participants gave lower ratings if they were bothered by the car’s smell or the music being played. One participant assigned a one-star rating because she perceived the driver as going too fast, even though trainers reported that the driver did not exceed the speed limit.

Preliminary Impact Outcomes

Independence

Both the linear (p < .001) and nonlinear (p < 0.001) session terms in the Poisson regression model for independence were statistically significant. This suggests that participant independence changed over the course of the intervention at a rate that also changed over time. The predicted independence score for a new participant was highest at Week 6 (26.08; 95% prediction interval [PI] [23.66, 28.75]), and the predicted independence score at Session 8 was 25.26; 95% PI [23.24, 27.25]. Model results for the Poisson regression are provided in Table 2, and predictions with 95% PIs are provided in Table 3 and Figure 2A.

Table 2.

Poisson Regression Coefficient Estimates by Score

Score and Term	Estimate	95% CI	p
Independence
Session	.16	[0.10, 0.22]	<.001
Nonlinear session	−.14	[−0.22, −0.07]	<.001
Safety
Session	.11	[0.02, 0.20]	.023
Nonlinear session	−.09	[−0.21, 0.04]	0.167

Note. Session was included in the model as a restricted cubic spline to capture any potential nonlinear relationship between session and the outcome. CI = confidence interval.

Table 3.

Predicted Score and 95% Prediction Intervals by Session and Score

Session	Predicted Score [95% Prediction Interval]
	Independence	Safety
1	16.18 [14.17, 18.45]	6.39 [5.87, 7.00]
2	18.95 [16.80, 21.48]	7.11 [6.56, 7.72]
3	21.80 [19.37, 24.65]	7.82 [7.15, 8.55]
4	24.21 [21.52, 27.03]	8.42 [7.62, 9.26]
5	25.63 [22.90, 28.55]	8.81 [7.96, 9.56]
6	26.08 [23.66, 28.75]	8.98 [8.23, 9.52]
7	25.84 [23.74, 28.11]	9.02 [8.44, 9.32]
8	25.26 [23.24, 27.25]	8.98 [8.32, 9.26]

Figure 2.

Rideshare safety and independence scores over time.

Safety

The linear session term in the Poisson regression model for safety was statistically significant (p = .023), whereas the nonlinear term was not significant (p = .167). This suggests that participant safety changed over the course of the intervention at a constant rate. The predicted safety score for a new participant was highest at Session 7 (9.02; 95% PI [8.44, 9.32]), and the predicted safety score at Session 8 was 8.98; 95% PI [8.32, 9.26]. Model results for the Poisson regression are provided in Table 2, and predictions with 95% prediction intervals are provided in Table 3 and Figure 2B.

2-Mo Follow-Up Rideshare Use

Two of the thirteen participants did not provide follow up data; one was withdrawn before completing the intervention, and one did not respond to our requests for follow-up data. Of the 11 participants who completed the study and reported follow-up data, all of them opted to text the trainers during or after their rides; none of the participants used the “share ride” feature on the rideshare apps.

On the basis of the responses received, the SRP elicited an increase in the number of independent rides taken by eight of the participants, with a total of 40 rides taken in the 8 wk after our intervention (minimum = 0 rides, maximum = 15 rides). Rides were taken to work or volunteering (14 rides), therapy (4 rides), the library (3 rides), stores or shopping (3 rides), school (2 rides), to get food (3 rides), and to self-care venues (e.g., nails, barber; 3 rides); eight rides were reported but with undisclosed destinations. A Wilcoxon signed-rank test determined that there was a statistically significant median increase in the number of rides taken after the participants took part in the SRP (Mdn = 3.5) compared with before they participated in the program (Mdn = 0), z = 2.53, p = .011.

Discussion

This study presents the findings from a pilot study of the SRP that was developed in conjunction with autistic community partners. Overall, the results of this study suggest that the SRP is an acceptable and appropriate training program for young autistic adults who already possess the skills to use a smartphone. In this study, participants showed improvements in independence and safety for using rideshare services. Peak independence and safety scores were achieved on Rides 6 and 7, respectively, suggesting that this may be the optimal range for future studies.

Research to date supports the fact that community mobility is an important factor that can affect community participation in autistic people. Lack of transportation may lead to lack of community participation in autistic young adults, which may subsequently affect health outcomes and overall quality of life (Brusilovskiy et al., 2024). Transportation independence is also a key factor in supporting employment in autistic adults; as noted by Pfeiffer et al. (2024), employed autistic adults exercise more transportation independence, whereas unemployed autistic adults report a lower ability to use transportation services independently. Given the challenges autistic adults face with community access, the SRP has excellent potential to expand community mobility options for this population and may be well suited for widespread use given that training elements of the program are already freely accessible and available in a virtual format. An increase in transportation options for autistic adults has implications that go far beyond that of simply being able to move from one location to another. Access to reliable and safe transportation could allow autistic adults to be more independent; hold a wider variety of jobs; access more opportunities for education, socializing, self-care, and health care; and improve their quality of life overall.

We were surprised at the variability in how frequently rideshare services were used after participation in the SRP. Some participants appeared to use rideshare at least weekly to get to work or therapy appointments, whereas others only used the skill occasionally to go to the store or to get food. Although examining factors that predict postintervention performance was beyond the scope of this study, this remains an important direction for future research. Financial considerations are also critical, particularly given the costs associated with rideshare use. In this study, all rides taken during the intervention period were covered by study funding; however, participants were responsible for out-of-pocket costs for any rides taken during the 2-mo follow-up. Although rideshare-based transportation services are increasingly used to support health care–related transportation (Bell-Brown et al., 2022; Chen et al., 2023; Fraade-Blanar, Koo, & Whaley, 2022), funding to support rideshare use for community participation and employment activities remains limited. Ideally, sustained evidence of effectiveness and positive outcomes in this area could inform policy changes and expand funding mechanisms to make rideshare use a viable, reimbursable option for enhancing community mobility for autistic people.

Although rideshare services show promise as a transportation option for autistic adults, our findings highlight important challenges that affected the program’s acceptability and appropriateness. Many of these challenges—largely unanticipated during program design—centered on navigating social interactions and managing sensory experiences during rides. For example, trainers reported instances of inappropriate driver behavior, which led our team to begin developing a new training module focused on recognizing, navigating, and reporting such situations. Sensory differences, which are highly prevalent in the autistic community (Ausderau et al., 2016), also emerged as a significant factor influencing participants’ perceptions and ratings of rideshare experiences; yet, to our knowledge, no studies have examined how sensory factors intersect with transportation use for autistic adults.

Social conventions around tipping and rating drivers presented additional barriers. Our research team, including an autistic research partner, observed that these decisions were often highly personal, shaped by cultural norms, and further influenced by sensory experiences during the ride (e.g., discomfort with odors, music, or perceived driving speed). These findings suggest that both sensory and social dimensions are integral to successful rideshare use and other community mobility options and that they should be explicitly addressed in future program adaptations.

It is important to note that the design of the SRP was based on the desires expressed by autistic members of our community and that this study was conducted with ongoing input from an autistic adult who served as a part of our research team. Any changes to this program or efforts to validate program impact in the future should continue to involve members of the autistic community to ensure that the program remains aligned with the needs, preferences, and lived experiences of this community.

Limitations

In this article, we presented the outcomes of a pilot study that evaluated a select number of outcomes from Proctor et al.’s (2011) implementation outcomes framework. Although our study results are promising, several limitations need to be considered. First, the study used a mix of validated tools and original assessments. Specifically, the RIC, our primary impact measure for safety and independence, has not undergone any formal testing for reliability or validity. Given that there are no existing tools specifically designed to assess rideshare use, we felt that our checklist, which had been previously used to demonstrate change in a feasibility study, was the best option for the present project. Future assessment of the checklist is warranted before completing larger studies, particularly determining whether safety and independence can or should be evaluated as two distinct constructs. Our fidelity tool was also developed by our research team and has not undergone reliability or validity testing. Although this is a limitation of the study, fidelity was enhanced by always having two trainers implement the intervention and having twice-monthly team meetings to discuss implementation challenges. In the future, fidelity will also need to be assessed during the ride-along sessions in addition to the virtual training component.

To further advance the implementation-science foundation of the SRP, future studies should evaluate additional implementation, service, and client outcomes beyond those examined in this pilot study. Although the present study assessed acceptability, appropriateness, fidelity, safety, and independence, subsequent research should use more rigorous hybrid designs and theory-driven implementation strategies and incorporate outcomes such as cost, feasibility, penetration, effectiveness, and long-term sustainability.

Finally, we had hoped for this study to attract a diverse geographic sample (urban, suburban, and rural), given our wide radius for recruitment. However, we identified early on that this would not be possible, because the vast majority of adults who inquired about the program identified lived in suburban areas. It is likely that using rideshare services in urban and rural areas may present challenges that are different from those identified in this study. It will be important that future projects identify those challenges so that additional supports and resources can be put in place.

Implications for Occupational Therapy and Practice

The study results have the following implications for occupational therapy practice:

• Occupational therapists can use structured training (e.g., video modules and supervised ride-along sessions) to teach autistic adults to use rideshare services safely and independently.

• Community mobility interventions should address both functional app-based skills (e.g., requesting rides, navigation, payment) and safety behaviors (e.g., verifying driver identity, monitoring routes).

• Rideshare training can expand participation in work, education, health care, and leisure for autistic adults who face barriers to driving or public transportation.

• Interventions should address sensory experiences and social conventions during rides (e.g., communication with drivers, tipping, and rating) to support successful community use.

Conclusion

To our knowledge, this is the first known study that demonstrates the appropriateness, acceptability, and positive effects of a training program specifically designed to teach autistic adults how to use rideshare services. Use of this program could increase access to community settings and have a widespread impact on socialization, job opportunities, and quality of life.