Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles (2023)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

P A R T I I Conduct of Research

15   This research was conducted to provide insight into the use of child restraint systems (CRS) in rideshare vehicles and taxis. The research responds directly to specific tasks prescribed by the BTSCRP addressing the following areas: 1. Annotate current state and local policies for ensuring child passenger safety in for-hire motor vehicles. 2. Describe current company practices, both national and international, for providing CRS in for-hire motor vehicles. 3. Identify noteworthy international policies and practices for CRS use applicable to the United States. 4. Determine the CRS experiences and attitudes of parents, other caregivers, and for-hire drivers. 5. Document the barriers and facilitators to CRS use with a focus on companies, drivers, and caregivers. 6. Identify objective techniques to measure the use of CRS in for-hire motor vehicles. 7. Describe the roles of stakeholders who can influence the use of CRS. 8. Develop targeted behavioral change strategies and messages to promote child passenger safety, including at-risk and hard-to-reach populations. 9. Propose strategies to evaluate the guidance. The research approach incorporated document reviews, surveys of parents and caregivers and drivers of rideshare vehicles and taxis, and expert opinions from a highly qualified advisory group. The research identified a gap between levels of child seat use reported in personal vehicles and in rideshare vehicles or taxis. In a nationally representative survey of parents of children below the age of 13, most parents reported that they consistently use child restraints whether in their personal vehicle or in rideshare vehicles or taxis. Of parents for whom complete data on this question were available, 86% reported consistent use. The remainder reported either less frequent use in rideshare vehicles or taxis than in their personal vehicle, or inconsistent use of child restraints in any type of vehicle. Despite this gap, the fact that a small proportion of trips with children are conducted in for-hire vehicles indicates that potential benefits to addressing this problem are incremental at best. While additional research is needed to refine approaches and prioritize interventions, the findings point to a few areas of potential. Educational campaigns aimed at parents and Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles: Conduct of Research S U M M A R Y

16 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles caregivers should explain that the correct selection and use of car seats are essential for their child’s safety, not only in the family vehicle, but also in rideshare vehicles and taxis. Addi- tional state and/or city laws could be constructive, both through direct effects on compliance by parents, caregivers, and rideshare and taxi corporations and through their indirect influence in communicating safety expectations to these groups and other stakeholders. Exemptions for rideshare vehicles in current child passenger protection laws could be removed in seven states and exemptions for taxis could be removed in 28 states. Considering the strong effect that use laws have had on child restraint use in private vehicles, these legislative amendments could improve child passenger safety. In addition, innovative new legislation that places more responsibility on rideshare and taxi companies for providing CRS for customers with children and/or facilitating their use may be an effective means for increasing use rates. Such legislation could be pilot tested and evaluated to assess effectiveness and encourage replication. Similarly, advocacy from respected safety leaders may be effective in highlighting the potential role of corporations in reducing barriers to child seat use in rideshare vehicles and encouraging these businesses to act even without legislative mandates. Findings from the surveys of parents and rideshare and taxi drivers, and lack of responses from rideshare and taxi companies, indicate that responsibility for child safety is focused on parents and caregivers and reflect the challenges faced by rideshare customers in providing the best protection for their children (National Center for Statistics and Analysis, 2022). CRS can be bulky, heavy, and inconvenient for parents and caregivers to carry on trips—short or long—that involve use of taxis or rideshare vehicles. Reducing these convenience barriers through engineering innovation holds promise for increasing use. The introduction of CRS that are easier to carry or vehicle seating systems with integrated child passenger protection is likely to lead to increased use. These types of engineering solutions could be stimulated by regulation or incentivized through consumer demand, perhaps with the help of vehicle safety ratings and awareness campaigns.

17   Background Rideshare services provided by companies like Uber and Lyft have become more popular over the last decade. In Q4 2021, Uber reported an average of 19.5 million trips a day worldwide (Uber Technologies, Inc., 2021), and Lyft reported 18.7 million active riders (defined as riders who take at least one trip during the quarter) (Lyft. Inc., 2022). As personal vehicles become less desirable and common among urban residents (Polzin et al., 2014), rideshare and taxi services will likely continue to become more popular. Unfortunately, growing evidence suggests that use of child restraint systems (CRS) in ride- share (Owens et al., 2019) and taxi vehicles (Prince et al., 2019) is much lower than usage rates in personal vehicles (Li and Pickrell, 2018). Because CRS are a crucial tool in decreasing the risk of crash injury for children (Kahane, 2015), this research aimed to ensure that advances in personal mobility in the United States are not accompanied by setbacks in child safety. To address this challenge, the research team completed the following tasks, as requested by the BTSCRP, to improve understanding of factors that may affect child seat use in for-hire vehicles and inform the development of strategies to increase use: 1. Annotate current state and local policies for ensuring child passenger safety in for-hire motor vehicles. 2. Describe current company practices, both national and international, for providing CRS in for-hire motor vehicles. 3. Identify noteworthy international policies and practices for CRS use applicable to the United States. 4. Determine the CRS experiences and attitudes of parents, other caregivers, and for-hire drivers. 5. Document the barriers and facilitators to CRS use with a focus on companies, drivers, and caregivers. 6. Identify objective techniques to measure the use of CRS in for-hire motor vehicles. 7. Describe the roles of stakeholders who can influence the use of CRS. 8. Develop targeted behavioral change strategies and messages to promote child passenger safety, including at-risk and hard-to-reach populations. 9. Propose strategies to evaluate the guidance. Additional evidence of the prevalence of children in rideshare use was found in a 2019 survey conducted by members of the research team (Ehsani et al., 2021). This study, based on a nation- ally representative sample, reported that more than two thirds (68%) of parents of children below 10 years of age had used rideshare and that 82% of these parents had used rideshare with their children. Rideshare use with children, as measured by this survey, was significantly higher among individuals with college education than among those with high school education or less (57% compared to 47%) and in households with incomes of $50,000 or higher compared to those with incomes below $50,000 (60% compared to 43%). Adults between 18 and 34 years of C H A P T E R   5

18 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles age and those 35 to 44 years old reported the highest percentage of ridesharing use (71% and 65%, respectively). The Centers for Disease Control and Prevention reports that unintentional injuries are the leading cause of death for children and youth above age 1, and the third leading cause of death for those under age 1 (West et al., 2021). NHTSA further indicates that motor vehicle crashes are the second leading cause of unintentional death for children through age 3 and the leading cause for ages 4 through 20 years (Webb, 2020). NHTSA estimates that CRS are 71% effective in preventing car crash death for infants (under 1 year old) and 54% effective for toddlers (1 to 4 years old) in pas- senger cars (Hertz, 1996). From 1975 to 2017, more than 11,000 children 5 years old and younger are estimated to have had their lives saved by restraints (CRS or adult seat belts). While restraint use for children age 4 and under is greater than 90%, a 2021 NHTSA report estimated that an additional 46 lives in this age group could be saved annually if CRS use was 100% (Enriquez, 2021). Evidence indicates that increasing the CRS usage rate to this level will require special attention to child seat use in rideshare vehicles and taxis.

19   Research Approach This chapter describes the methods used to complete each task: (1) document review, (2) surveys, and (3) expert opinions from an advisory group. Document Review Task 1: Annotate Current State and Local Policies for Ensuring Child Passenger Safety in For-Hire Motor Vehicles State policies governing the use of CRS in for-hire motor vehicles, including for rideshare and taxi services, were collected for all 50 states and the 50 largest U.S. cities (determined through 2019 U.S. Census estimates; note: Washington, DC, was included among the cities). To review state legislation, the research team used the Thomson Reuters Westlaw database, a resource for online legal research. The team examined each state’s statutes and regulations using an advanced search that allowed for multiple keywords at once. The keywords included child safety seat, car seat, booster seat, seat belt, safety belt, lap belts, safety restraints, child passenger, ridesharing, taxi, for-hire, livery, child restraint, ride source, omnibus, jitney, and transportation network. Alternate versions of some of these keywords were included in the advanced search to avoid missing any relevant laws. For example, the team searched for both “seat belt” and “seatbelt” to cover two common spellings of the same term. Once the search results were retrieved, research assistants reviewed each of the first 100 results to determine their pertinence to the research question. The research team abstracted the collected policies to determine whether the policy applies to specific vehicles (i.e., specific standards for ridesharing or taxi vehicles), whether the policy requires specific safety restraints in those vehicles, the specific requirements for use of those safety restraints, how compliance was enhanced, how noncompliance was penalized, and whether there were any exceptions. A similar method for the city-level review was used, but because Westlaw does not include many relevant local policies, the research team used city-specific databases. For each city, the team located the official city charter or code. Many large cities use similar database services to host their municipal code, namely, Municode and American Legal Publishing Corporation. These databases allow for city-specific code searches. When these databases were available, the team used the same search terms to identify relevant code sections. When city code databases were not hosted by one of these services, the team searched the city code by using search tools provided by the city’s website. In rare cases when the city code was not searchable, the relevant subsections pertaining to transportation and public safety were identified and searched manu- ally. As a check on these searches, the team also conducted basic Internet searches using city names and select search terms. Once the city search results were retrieved, research assistants reviewed each of the results to determine their relevance to the study. The team abstracted the city policies using the same method used for abstraction of the state policies. C H A P T E R   6

20 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles These search and coding strategies were tested on five states and five cities to determine their feasibility and adequacy. Once the strategies were finalized, one team member performed all searches and all coding for the states and cities to ensure consistency. To validate these results, a second team member independently reviewed a random 20% sample of the states and the cities (10 states and 10 cities). For this random sample, both sets of results were compared. Any discrepancies were reviewed and resolved through discussion with the entire study team. When possible, the findings were compared to collections of state CRS, seat belt, and rideshare policies (IIHS/HLDI, 2022; VTTI/TTI, 2022). Task 2: Describe Current Company Practices, Both National and International, for Providing CRS in For-Hire Motor Vehicles The research team began by defining the inclusion and exclusion criteria for rideshare companies relevant for this review. Transportation companies that do not provide drivers were excluded, such as services that connect riders with traditional taxi companies (e.g., FREE NOW, Arro, Easy Taxi), private chauffeur companies (e.g., Blacklane), or companies that provide both services (e.g., Gett, Talixo). A ridesharing company (also known as a transportation network company, ride-hailing service, app-taxi, e-taxi, or mobility service provider) was defined as a company that, via websites and mobile apps, matches passengers with drivers of vehicles for hire that, unlike taxicabs, cannot legally be hailed from the street. These companies typically do not employ the drivers but contract services to them on a trip-by-trip basis, using software to facilitate the transaction. To identify major rideshare companies located in the 37 Organization for Economic Cooperation and Development (OECD) countries, the team searched for the leading ridesharing companies by market share in each country. The team used Statista, a market research database, to search for multiple keywords such as rideshare market, ride-hailing market, for-hire vehicle market, and leading rideshare companies. The search of each OECD country was followed by a search of all keywords on both Google and Statista (e.g., “Belgium leading rideshare companies”). As the rideshare industry is relatively new, market share reports for only three countries were retrieved from the search results. Statista included reports for the U.S. and Australia rideshare markets, and a report by Advanced Symbolics, a market research company, included the rideshare market breakdown for Canada. To determine the rideshare companies in the remaining OECD countries for which market summaries were not available, a professional library service referred the team to Rideguru’s Ride- shares Worldwide, a resource that lists every rideshare company and its areas of operation. If a rideshare service was available in an OECD country, it was included in a spreadsheet that tracks every OECD country and the rideshare companies operating there. The team’s findings on the United States, Australia, and Canada using Google and Statista were consistent with the findings on the same three countries using Rideshares Worldwide. A total of nine companies were identified. This list was broadly cross-checked by a search for the leading rideshare companies worldwide, not specifically the OECD countries, using the following keywords on Google and Statista: “international rideshare market” and “leading international rideshare companies.” The team found two reports by market research companies through Google and one report by Statista, and the team’s list covered every company included by these three resources except companies that did not operate in OECD countries. Only one company on the list was not included in any of the reports. After identifying the major rideshare companies located in OECD countries, the research team reviewed the companies’ policies and practices requiring or encouraging CRS or seat belt

Research Approach 21 use for children by searching for the company name followed by multiple keywords on Google (i.e., car seat policy, child safety policy, and seat belt policy). The search results were used to review the relevant section of each company’s website to determine whether the company provides information pertinent to the research question. For companies that had separate pages or web- sites based on region, the team reviewed the relevant practices in the OECD countries where they operated. If a practice was deemed relevant, the information was copied into a spreadsheet that tracked answers to specific parts of the research question. The identified practices were used to determine whether the company requires drivers and riders to comply with applicable passenger safety laws, whether the company provides appropriate child restraint devices if the parent or caregiver does not have such a device available, the specifications for providing these devices, and any exceptions that the company defines for these practices. Along with identifying rideshare company practices, the team also sought company practices for both domestic and international taxi companies. To start, the research librarian at the Johns Hopkins Welch Medical Library was engaged to obtain a list of taxi companies both in the United States and in OECD countries. Once the list of taxi companies was compiled, Google was searched using the company names and the keywords described above to locate taxi company policies related to CRS use. This search did not reveal any relevant policies. In addition to performing Google searches for company policies, the team searched for contact information for rideshare and taxi company representatives using Google and tools available to the Johns Hopkins research librarian. After obtaining this contact information, the team created two short surveys and distributed them to these rideshare and taxi company representatives to gather more information about their company practices. Each survey was hosted on Qualtrics XM experience management software (https://www.qualtrics.com), a widely used and secure web application for building and managing online surveys and databases. The rideshare company practices survey consisted of three questions. Respondents were first asked to describe their role within the company and then asked about the steps their company has taken to ensure children are protected by appropriate CRS and adult passengers are protected against crash injury by use of seat belts. This survey was distributed to 14 different email addresses for rideshare company representatives. The taxi company practices survey was identical to the rideshare company practices survey. This survey was distributed to nine taxi company representatives via email and sent to an additional 11 companies via a “contact us” form located on their websites. Surveys Task 4: Determine the CRS Experiences and Attitudes of Parents, Other Caregivers, and For-Hire Drivers Task 5: Document the Barriers and Facilitators to CRS Use (e.g., Culture, Geography, Income, Education) with a Focus on Companies, Drivers, and Caregivers Design, Conduct, and Analyze Surveys The research team designed two separate surveys to understand safety behaviors and usage patterns for CRS among (1) parents and caregivers of children under the age of 13 who use for-hire services, including ridesharing and taxis, and (2) drivers of for-hire vehicles. The purpose of these surveys was to understand the current behaviors, experiences, and attitudes of parents/

22 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles caregivers and drivers toward child restraints and to determine barriers and facilitators for CRS or seat belt use in these two groups. Parent/Caregiver Survey In a national pilot study conducted in 2019, the research team found that 82% of rideshare users with children below the age of 10 reported traveling with their children on rideshare trips (Ehsani et al., 2021). On the basis of this finding, the team targeted a sample size of 1,000 U.S. parents and/or caregivers (18+ years) who reported using ridesharing services and/or taxis with their children below the age of 13 in the previous 12 months. Survey respondents were a nation- ally representative sample recruited from the National Opinion Research Center at the University of Chicago (NORC) AmeriSpeak Panel. The survey was conducted in English and Spanish. Recruitment Strategy for the Parent/Caregiver Survey Respondents were NORC AmeriSpeak Panel members invited to participate in the survey through email invitations. The survey included a short screener to confirm qualification criteria for this study. Upon successfully completing the survey, qualified respondents received panel points, equivalent to approximately $5 USD. To understand the CRS experiences and attitudes of parents and caregivers, the research team asked questions about frequency of use of for-hire vehicles with children, trip purpose, and self-reported CRS use (never, sometimes, mostly, always). Parents were also asked about their knowledge of state laws regarding CRS use for children and liability of the parent/caregiver if ticketed. To understand barriers and facilitators to CRS use in for-hire vehicles, the research team com- pared the characteristics of parents/caregivers who do and those who do not always use CRS in for-hire vehicles. The team also compared the characteristics of parents/caregivers whose CRS use patterns are different in their personal vehicles from those when for-hire vehicles are used and explored reasons for those differences (e.g., convenience, portability of appropriate seats, influence of laws/policies, and vehicle characteristics). Survey development was an iterative process that went through multiple rounds of pilot testing, feedback, and revision. The survey questions were tested for face validity through cognitive inter- viewing with pilot test participants. Cognitive interviewing is a method used to identify potential problems with survey questions. Critical decisions were made after weighing the risks and benefits with the research team and receiving feedback from pilot testing and the panel. One important decision was to restructure the parent/caregiver survey to only ask about the parent’s or caregiver’s youngest child. This survey focused on the youngest child because of all the children in the parent’s or caregiver’s family, the youngest is most likely to be using a child passenger restraint. This decision also had implications for survey length; if the survey asked about every child, it could become too long and burdensome. The research team also contemplated the impacts of the COVID-19 pandemic on usage of for-hire vehicles, and how to restructure the parent/caregiver survey accordingly. Initially, the research team considered asking people about their use of ridesharing before the pan- demic. As time progressed and the protracted nature of the pandemic became apparent, it seemed that people’s current behaviors may be most aligned with future behavior. The research team eliminated questions about pre-pandemic behaviors and focused on current behaviors instead. The final parent/caregiver survey included 30 items and took most respondents approxi- mately 15 minutes to complete. This survey instrument can be found in Appendix I.

Research Approach 23 Driver Survey Given that driver retention for ridesharing services and taxis is low and there is a high attrition rate (Cook et al., 2018), the research team restricted the sample to those who have driven for a rideshare or taxi service within the previous 12 months. This helped ensure the survey captured current drivers and minimized recall bias regarding the behaviors of interest. Further, the type of ridesharing service the driver provides needed to be relevant to transporting families with children. Therefore, drivers who exclusively provide delivery services such as Uber Eats were excluded. The driver survey was conducted in English. Inclusion Criteria for the Driver Survey The sample included 198 U.S. adults who reported driving for a rideshare company in the pre- vious year and 93 U.S. adults who reported driving for a taxi company in the previous year. Due to ridesharing company policies requiring drivers to be above a minimum age (Lyft Help, 2020) and/or have a minimum of 3 years of driving experience if below the age of 23 (www.uber.com/ us/en/drive/requirements), the sample was restricted to drivers aged 21 or older. Recruitment Strategy for the Driver Survey Ideally, rideshare and taxi companies would have helped the research team circulate a survey to their drivers. However, perhaps because of the tendency of ridesharing companies to engage drivers as contractors (Burns, 2019), and consistent with them declining to participate in a survey on company practices (as described above), partnering with these companies to distribute the survey was not considered a viable recruitment strategy. As an alternative, the team engaged Harry Campbell, founder of the website TheRideshareGuy.com. This website includes a blog and podcast that provide information to drivers of for-hire vehicles, including rideshare and taxi drivers. A one-time payment was issued to TheRideshareGuy.com for an email advertisement to its sub- scribers to encourage them to complete the survey. The research team programmed the for-hire driver survey using Qualtrics. The survey included a short screening section consisting of four questions that confirmed qualification criteria for the survey and sorted participants as rideshare or taxi drivers. Upon successfully completing the survey, qualified respondents were directed to a separate form where they could leave their email address to receive a $5 Amazon.com gift card. Each week, a research team member reviewed the emails obtained from this form and sent a gift card to each eligible email. Drivers were asked to report for which company or companies they are currently driving, the number of hours they work each week, and the state where they spend most of their time working. To understand the CRS experiences and attitudes of drivers of for-hire vehicles, the research team asked questions about current practices for passengers who are less than 8 years old. Questions also tested drivers’ knowledge of company policies and state laws regarding CRS use for children younger than 8 years. To understand barriers and facilitators to CRS use by drivers of for-hire vehicles, the research team compared the characteristics of drivers who do and those who do not always require pas- sengers to use CRS or seat belts, and those who do and do not provide CRS devices for children who are passengers in their for-hire vehicles. The for-hire driver survey instrument was pilot tested on a convenience sample of 12 individuals residing in Maryland or the District of Columbia that included rideshare drivers. In-depth cogni- tive interviewing was done with one rideshare driver via Zoom, eliciting responses to the survey and identifying questions needing refinement or clarification. Seven completed surveys were received. Pilot survey responses were collected from June 29 to July 2, 2021; drivers took an average of 12 minutes to complete the survey.

24 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles The rideshare section of the final for-hire driver survey included 30 items, whereas the taxi section included 28 items. Drivers took an average of 14 minutes to complete the survey. The rideshare and taxi driver survey instruments can be found in Appendices J and K, respectively. Measures For the parent/caregiver survey, the primary outcome measure is self-reported CRS or seat belt use for children under the age of 13 years by parents or caregivers. For the rideshare and taxi driver survey, the primary outcome measure is self-reported CRS or seat belt use for children under the age of 8 years. Analyses For both the parent/caregiver and for-hire driver surveys, analyses described current behaviors, experiences, and attitudes of parents/caregivers and drivers toward child restraints. Some of this information was provided to the Advisory Group (described in more detail below) for their use in further clarifying barriers or facilitators to CRS use among these two groups based on the survey findings and their individual expertise. Analyses also examined CRS use by gender, race/ ethnicity, education, and income level. Univariate associations for covariates were tested, followed by multivariate analyses using groups of related variables (e.g., individual characteristics). The surveys assessed these variables, and the analyses identified barriers and facilitators to CRS or seat belt use for children by regression analyses that treat predisposing, reinforcing, enabling, and environmental factors as the independent variables or correlates of CRS use outcomes. In both surveys, the sample size allowed the exploration of how CRS use in for-hire vehicles may be affected by key independent variables, including individual characteristics such as gender, age, education, and attitudes toward CRS use and passenger safety. Expert Opinion from an Advisory Group • Task 3: Identify noteworthy international policies and practices for CRS use applicable to the United States • Task 6: Identify objective techniques to measure the use of CRS in for-hire motor vehicles • Task 7: Describe the roles of stakeholders who can influence the use of CRS • Task 8: Develop targeted behavioral change strategies and messages to promote child passenger safety, including at-risk and hard-to-reach populations • Task 9: Propose strategies to evaluate the guidance Establish and Convene an Advisory Group The research team established an advisory group (AG) of domestic and international opinion leaders and experts in child passenger safety. The role of the AG was to provide recommendations for each of the tasks listed above that reflect the scope of perspectives necessary for broad rele- vance, effectiveness, and acceptability. Appropriate recommendations were technically feasible, research-informed, and consistent with the needs, constraints, and expectations of families, safety and technical experts, business, and academia. Members of the AG were selected to represent this range of perspectives. Table 1 lists expertise areas, candidates, and qualifications for each member of the AG. Candidates for the AG were asked to commit 6–8 hours to this research project, including two 3-hour meetings, which were held virtually via Zoom on May 17 and May 31, 2022. An hono- rarium of $500 was offered to compensate members for their involvement. Before the meetings, the research team contacted AG members to inform them of expecta- tions and the process for AG deliberations. The research team developed briefing materials and

Research Approach 25 Expertise Area AG Member and Qualifications Child biomechanics Kristy Arbogast Children’s Hospital of Philadelphia Expertise: Child seat testing and correlation of engineering design with crash injury outcomes. Compatibility and safe fitment of child safety seats in vehicles Tammy Franks Chairman, National Child Passenger Safety Board Expertise: Synthesis of research into practice and development of training curricula for 40,000 National Child Passenger Safety Technicians. Pediatrics Alisa Baer Principal, thecarseatlady.com Expertise: Over 25 years of experience in consulting with parents and caregivers on optimal child passenger protection. Child seat manufacturing Joe Colella Director of Child Passenger Safety, Juvenile Products Manufacturers Association Expertise: Over 25 years of experience in translating safety research into guidance for child safety seat design. Vehicle manufacturing Stephen Gehring Hyundai Motor Company Expertise: Over 30 years of experience in translating child safety research and practice into vehicle design and marketing. Child passenger safety advocacy Lorrie Walker Child Passenger Safety Technical Advisor Expertise: Developing, implementing, and evaluating national and international campaigns to improve child passenger protection practices. Child passenger safety analysis and reporting Denise Donaldson Editor, Safe Ride News Expertise: Editor of the most comprehensive independent source of child passenger safety information for health and safety professionals. Author of widely endorsed technical manual on the LATCH System. Public health law and policy Jon Vernick Johns Hopkins Center for Injury Research and Policy Expertise: Legal review, synthesis, analysis across a range of safety issues. Public health and injury prevention policy Joyce Pressley Columbia University Mailman School of Public Health Expertise: Developing policies and interventions to prevent injury and increase life expectancy; chair of the TRB Occupant Protection Committee. Road safety— international Sjaan Koppel Monash University Accident Research Centre Expertise: One of the world’s most prominent authors of literature related to child passengers and other vulnerable road users. Table 1. Expertise areas, AG members, and qualifications. (continued on next page)

26 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles a presentation for the AG members. These materials contained data generated by the document review (Tasks 1 and 2) and the surveys (Tasks 4 and 5). AG members were asked to provide evidence-informed comments and recommendations during each of the two meetings. Following the AG meetings, the research team drafted final recommendations for each task and circulated this draft among AG members for comment. Comments were considered by the research team during development of final recommendations. Expertise Area AG Member and Qualifications Child seat manufacturing Mark LaPlante Graco Children's Products Inc. Expertise: Engineering and regulatory compliance. Child seat manufacturing Alison Schmidt Britax Child Safety Expertise: Product research and development, testing and regulatory compliance. Table 1. (Continued).

27   Findings and Applications Task 1: Annotate Current State and Local Policies for Ensuring Child Passenger Safety in For-Hire Motor Vehicles All 50 states have policies that require the use of child restraint systems (CRS) for children under a certain age, weight, or height. Seven states (14%) exempt rideshare vehicles from the CRS requirement (Hawaii, Idaho, Indiana, Maryland, New Hampshire, Texas, Washington) and have policies that contain a broad exemption for “commercial” or “for-hire” vehicles, defining those terms to include both rideshare and taxis. In the remaining 43 states, child passengers in rideshare vehicles are required to be secured in an appropriate CRS. Twenty-eight states (56%) exempt taxis from the CRS requirement. Among the 43 states requiring child passengers in ride- share vehicles to be secured in an appropriate CRS, 21 states (49%) exempt taxis and 22 states (51%) do not exempt taxis. Every state policy describes penalties for failure to comply with CRS requirements. These penalties generally include monetary fines and/or driver’s license points. The fines vary, depend- ing on the state, type of CRS, and other circumstances, with a minimum of $10 and a maximum of $500. In personal vehicles, the driver is generally responsible for traffic and safety violations, but when children are riding in rideshare or taxi vehicles, the responsibility may shift. Of the 43 states that require CRS in rideshare vehicles and/or taxis, 36 (84%) place responsibility on the driver. Four states (9%) (California, Colorado, Iowa, Florida) hold the parent or guardian respon- sible, though California and Colorado note that this is only true if the parent or guardian is in the vehicle with the child. In three states (7%) (Montana, North Dakota, Oregon), the legal respon- sibility for CRS violations in this context is somewhat ambiguous or unclear. In those cases, it is likely that the driver is responsible. Of the 50 cities included in this study, only 12 had relevant laws or ordinances. The fact that no relevant policies appeared during a search in the municipal code of the remaining 38 cities does not mean that CRS use is not required or that taxis or rideshare vehicles are unregulated. Rather, in those cases, localities have either chosen not to legislate on top of state requirements or are prevented from doing so. Of the 12 cities with relevant policies, eight (67%) require CRS in rideshare vehicles, but not taxis; two cities (17%) (Raleigh, NC, and Seattle, WA) require CRS use in both rideshare vehicles and taxis. In one city (Louisville, KY), there is an explicit require- ment for taxis, but no specific language for rideshare vehicles. Finally, in one city (Chicago, IL), the law does not indicate any specific requirements or exceptions for rideshare vehicles or taxis. Importantly, these laws generally reiterate state law; only one city (Seattle) has an ordinance that differs from that of the state (Washington) law. While Seattle and Washington state both have a CRS requirement that exempts “for-hire” vehicles, Seattle has a provision that defines “for-hire” in a way that excludes “taxicabs . . . [and] transportation network company endorsed vehicles,” which includes rideshare vehicles (Seattle Ord. §§ 6.310.110, 11.14.235, 11.58.195). C H A P T E R   7

28 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles Further details of this analysis are available in a peer-reviewed scientific paper by the research team included in Appendix H. Task 2: Describe Current Company Practices, Both National and International, for Providing CRS in For-Hire Motor Vehicles Detailed findings of this search are included in Appendix E. A summary of the main results is provided below. Eight of the nine major rideshare companies in the Organization for Economic Coopera- tion and Development (OECD) countries require drivers and riders to comply with child safety laws in their for-hire vehicles. However, only one company provides appropriate child restraint devices if the parent or caregiver does not have such a device available; this program is available exclusively in Chile and Mexico. Four companies state that riders are responsible for providing their own CRS, including the two companies (Uber and Lyft) that make up most of the U.S. ride- share market. The remaining four companies do not specify policies for CRS use in their vehicles. Only Uber specifies that drivers can deny rides if they are not comfortable with the rider installing a car seat in their car or feel that the child cannot be safely transported in the seat. Although it is ambiguous whether drivers for Lyft, Bolt, and Carmel may cancel rides even if the rider provides the proper car seat for their child, they may deny riders who provide improper CRS or do not provide a car seat at all. Uber and Lyft specify that it is the rider’s responsibility to provide a car seat, but a car seat option is available exclusively in New York City for an additional fee of $10. The car seat option for both companies includes one forward-facing car seat suitable for a child who is at least 2 years old, 22–48 pounds, and 31–52 inches. Uber’s and Lyft’s car seat services in New York City may suggest that companies will adapt practices to address a large need for car seats among rideshare users in regions where for-hire vehicles are frequently used. In Chile and Mexico, Cabify has four different car seats available depending on the child’s size and weight (groups 0, 1, 2, and 3), accounting for all children under the height of 53 inches. A maximum of three seats may be requested per ride; the first car seat is free, but there is an extra fee for each additional seat. According to Cabify’s website, riders who order their journey several hours in advance are guaranteed the appropriate car seat for their child. Compared to Uber’s and Lyft’s car seat option in New York City, Cabify’s car seat selection addresses children of all sizes and weights who require CRS by law, including one car seat that is rear-facing (group 0). The research team did not receive any responses to the rideshare company practices survey. A member of the research team corresponded with a Lyft representative who made internal inquiries but ultimately responded as follows: “Regrettably, Lyft is declining participation in the survey. As independent contractors, our drivers are obligated to comply with state laws on the subject and we do not collect data from them on the issue.” Similarly, no responses were received for the taxi company practices survey. Task 3: Identify Noteworthy International Policies and Practices for CRS Use Applicable to the United States As described in the preceding section, a review of international rideshare and taxi prac- tices pertaining to child seat use identified at least one example of a corporate practice that facilitates the availability of appropriate CRS for customers with young children (see Cabify example above). This precedent may indicate that the logistic necessities of such a practice are

Findings and Applications 29 surmountable within a rideshare business model. However, further research would be necessary to determine how such a practice might translate to the U.S. environment. Beyond this example, the research indicates that the situation regarding CRS policy and practice in rideshare vehicles is relatively consistent among OECD countries. When asked for insights regarding the potential relevance of international practices in the United States, the AG pointed out differences in product availability between the United States and Europe that may offer advantages in child seat portability and consequently influence child passenger protection in rideshare. One specific example provided was the multifunctional Trunki BoostApak. The group believed that increased availability of such devices in the United States could be an effective means for increasing child restraint use in rideshare vehicles. The group suggested that an analysis be conducted of the extent to which Federal Motor Vehicle Safety Standards (FMVSS) prevent marketing of portable child restraint products in the United States and that some consideration be given to accommodating and promoting their use if this is feasible without compromising safety. Task 4: Determine the CRS Experiences and Attitudes of Parents, Other Caregivers, and For-Hire Drivers Description of Survey Sample Inclusion Criteria Parents and caregivers of a child below the age of 13 who used ridesharing or taxis with their child were invited to participate in the parent/caregiver survey through email invitations. The sample was selected from the AmeriSpeak Panel using sampling strata based on age, race, Hispanic ethnicity, education, and gender. The survey was fielded between May 4 and June 10, 2022, in both English and Spanish. Most respondents (95%) completed the survey in English, and 5% of respondents completed the survey in Spanish. A total of 2,836 parents/caregivers had children below the age of 13. Of these individuals, 32 had missing data on ridesharing use, leaving 2,804 parents/caregivers with children under the age of 13 who had complete data about ridesharing or taxi use. In total, 718 respondents had used ridesharing with their children and 393 had used taxis with their children. The weighted proportion of rideshare use with children in the sample was 24.3% [95% Confidence Interval (CI): 22.4% to 26.3%] and taxi use with children was 14.7% [95% CI: 13.1% to 16.6%]. Parents and caregivers who reported using rideshare with their children (N = 718) were more likely to be Black (22% among users versus 9% among non-users) or Hispanic (24% among users versus 16% among non-users), never married (31% among users versus 19% among non-users), have $100K+ income (31% among users versus 25% among non-users), live in metropolitan areas (93% among users versus 82% among non-users), rent rather than own (44% among users versus 33% among non-users), and live in a building instead of a house (33% among users versus 18% among non-users). Parents and caregivers who used taxis with their children (N= 393) were more likely to be Black (21% among users versus 11% among non-users) or Hispanic (27% among users versus 17% among non-users), have high school or less education (29% among users versus 19% among non- users), be never married (34% among users versus 20% among non-users), currently not have a job (33% among users versus 26% among non-users), have less than $60K income (57% among users versus 47% among non-users), rent rather than own (51% among users versus 34% among non-users), and have no Internet in the home (16% among users versus 10% among non-users). Using the total sample of respondents, the overall estimated prevalence of child restraint use in personal vehicles among rideshare users (N =718) was 92.2% [95% CI: 88.9% to 94.5%] while

30 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles the prevalence of child restraint use in rideshare vehicles was 87.2% [95% CI: 83.6% to 90.1%]. The overall estimated prevalence of child restraint use in taxis among taxi users (N = 393) was 79.7% [95% CI: 74.3% to 84.2%]. Description of Analytic Sample Of the 718 respondents who reported using ridesharing with their children, 45 had missing data on child restraint use in either rideshare or personal vehicles. A further 46 respondents said that their child rides either not in a restraint or on a lap in personal vehicles. As the outcome of interest was differences in restraint use between personal vehicles and rideshare, these individuals were excluded from the analysis. The final analytic sample for rideshare users was 627. Of the 393 respondents who reported using a taxi with their children, 31 had missing data on child restraint use in either taxis or personal vehicles. A further 38 respondents said that their child rides either not in a restraint or on a lap in personal vehicles. As the outcome of interest was differences in restraint use between personal vehicles and taxis, these individuals were excluded from the analysis. The final analytic sample for taxi users was 324. The projected survey sample size of 1,000 parents/caregivers with children under the age of 13 who used ridesharing was not reached despite the large initial sample and multiple reminder invitations that were sent to respondents. This was due to a lower-than-expected incidence rate of parents/caregivers with children below the age of 13 who used ridesharing. The demographic characteristics of the sample are described in Table 2. Data are presented as mean and standard deviation (SD) or median and interquartile range (IQR) for continuous measures, and N (%) for categorical measures. The characteristics of participants’ rideshare use are described in Table 3. The majority of participants (87.4%) reported using rideshare services a few times a year or monthly. Higher frequency rideshare users represented the remaining 12.6% of the sample. Most respondents reported using rideshare during vacation, although everyday activities such as going to and from school, shopping, and attending social activities and medical appointments were also common reasons for using rideshare. Many parents reported using rideshare because they did not have access to a vehicle, and it was more convenient than other mobility options. Data are presented as N (%) for categorical measures. Approximately half (48.5%) of the respondents reported that the COVID-19 pandemic did not impact their ridesharing use. Over one-third (38.4%) stated that they had decreased rideshare use during the COVID-19 pandemic. The characteristics of parents and caregivers who used taxis with children are described in Table 4. Among those who rode taxis, 82.4% used them a few times a year or monthly. Most taxi trips occurred during vacation and for medical appointments. Many parents reported using taxis because they did not have access to a vehicle, and it was more convenient than other mobility options. Over one-third (40.7%) of the sample stated that they had decreased taxi use during the COVID-19 pandemic. Data are presented as N (%) for categorical measures. Child restraint use in personal vehicles and rideshare vehicles is described in Table 5. All respondents reported that their child usually rode restrained when personal vehicles were used. This dropped to 90.4% for rideshare vehicles. Restraint use is defined as being restrained by a car seat, booster seat, or seat belt. While appro- priate restraint use according to child height and weight could not be determined from these self-reported data, it appears that age-appropriate restraint use diminished when children went from being passengers in personal vehicles to being passengers in rideshare vehicles (see shaded cells). For example, a small percentage of individuals appeared to be using age-inappropriate restraints (i.e., booster or seat belts for infants or children between ages 1 and 4).

Findings and Applications 31 Characteristic Total N = 627 Parent gender Male 207 (33.0%) Female 420 (67.0%) Parent age (mean, SD) 38.4 (8.0) Parent age (categorial) 18–29 80 (12.8%) 30–44 418 (66.7%) 45–59 121 (19.3%) 60+ 8 (1.3%) Child age (mean, SD) 6.7 (3.7) Child age (categorical) 31 (4.9%) 177 (28.2%) 198 (31.6%) 206 (32.9%) 15 (2.4%) Parent race/ethnicity 281 (44.8%) 132 (21.1%) 146 (23.3%) 36 (5.7%) 23 (3.7%) 9 (1.4%) Parent education 35 (5.6%) 83 (13.2%) 196 (31.3%) 163 (26.0%) 150 (23.9%) Parent marital status 368 (58.7%) 4 (0.6%) 48 (7.7%) 20 (3.2%) 187 (29.8%) Parent employment 390 (62.2%) 75 (12.0%) 37 (5.9%) 42 (6.7%) 5 (0.8%) 16 (2.6%) 62 (9.9%) Income 162 (25.8%) 133 (21.2%) 122 (19.5%) 210 (33.5%) Residence 43 (6.9%) Infant, <1 Toddler, 1–4 Young child, 5–8 Older child, 9+ Missing White Black Hispanic Asian Multiple Other Less than high school High school Vocational training Bachelor Postgraduate Married Widowed Divorced Separated Never married Working/employee Working/self-employed Temporary not working Looking for work Retired Disabled Other not working <$30K $30K–60K $60K–<100K $100K+ Non-metro area Metro area 584 (93.1%) Number of vehicles (median, IQR) 2 (1–2) Table 2. Demographic characteristics of parents/ caregivers who used ridesharing with children.

32 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles Taxi Use Characteristics Total N = 324 Frequency of taxi use Daily 23 (7.1%) A few times a week 34 (10.5%) A few times a month 63 (19.4%) A few times a year 204 (63.0%) Taxi use (select all that apply)* Vacation 149 (46.0%) School/daycare 48 (14.8%) Shopping 65 (20.1%) Social activities 45 (13.9%) Medical appointments 81 (25.0%) Other 20 (6.2%) Reason for taxi use (select all that apply)* Didn't have access to a vehicle 126 (38.9%) Cheaper than other options 62 (19.1%) Faster than other options 67 (20.7%) Safer than other options 34 (10.5%) More convenient than other options 110 (34%) Someone else made the decision 38 (11.7%) Other 12 (3.7%) Pandemic impact on taxi use No change 135 (41.7%) Increased use 49 (15.1%) Decreased use 132 (40.7%) Missing 8 (2.5%) *For questions for which respondents could select multiple responses, the total percentages will not sum to 100%. Table 4. Parent/caregiver taxi use with children. Frequency and Trip Characteristics Total N= 627 Frequency of rideshare use Daily 33 (5.3%) A few times a week 46 (7.3%) A few times a month 97 (15.5%) A few times a year 451 (71.9%) Rideshare use (select all that apply)* Vacation 325 (51.8%) School/daycare 91 (14.5%) Shopping 135 (21.5%) Social activities 110 (17.5%) Medical appointments 147 (23.4%) Other 72 (11.5%) Reason for rideshare use (select all that apply)* Didn't have access to a vehicle 260 (41.5%) Cheaper than other options 138 (22.0%) Faster than other options 126 (20.1%) Safer than other options 62 (9.9%) More convenient than other options 251 (40.0%) Someone else made the decision 61 (9.7%) Other 19 (3.0%) Pandemic impact on rideshare use No change 304 (48.5%) Increased use 79 (12.6%) Decreased use 241 (38.4%) Missing 3 (0.5%) Table 3. Frequency and trip characteristics of rideshare use with children.

Findings and Applications 33 Child restraint use in personal vehicles and taxi vehicles is described in Table 6. All respon- dents reported that their child usually rode restrained when personal vehicles were used. This dropped to 81.0% for taxi vehicles. Restraint use is defined as being restrained by a car seat, booster seat, or seat belt. While appro- priate restraint use according to child height and weight could not be determined from these self- reported data, it appears that age-appropriate restraint use diminished when children went from being passengers in personal vehicles to being passengers in taxi vehicles (see shaded cells). As was observed for ridesharing vehicles, a small percentage of individuals appeared to be using age- inappropriate restraints (i.e., booster or seat belts for infants or children between ages 1 and 4). Participants’ knowledge of laws and penalties associated with restraint use in personal vehicles and rideshare vehicles is described in Table 7. While most respondents stated they knew there was a law requiring the use of child restraints in personal vehicles “where they live,” less than half stated there was a law for using CRS in rideshare vehicles. The same was true when respon- dents were asked about fines for not using CRS in rideshare vehicles. Notably, about half of the respondents stated that parents and caregivers should be responsible for providing restraints during rideshare trips. This response is notable because most parents and caregivers perceive that they are responsible for the safe transportation of their children, regardless of whether they are driving the vehicle. Data are presented as N (%) for categorical measures. Participants’ knowledge of laws and penalties associated with restraint use in personal vehicles and taxis is described in Table 8. While two thirds of respondents stated they knew there was a law requiring the use of child restraints in personal vehicles “where they live,” less than half stated there was a law for using CRS in taxis. The same was true when respondents were asked about fines for not using CRS in taxis. When asked who should be providing CRS in taxis, most Restraint Use Infant, <1 Toddler, 1–4 Young Child, 5–8 Older Child, 9–12 Total N = 31 N = 177 N = 198 N = 206 N = 612* Child restraint use in personal vehicle** Yes 31 (100.0%) 177 (100.0%) 198 (100.0%) 206 (100.0%) 612 (100.0%) No 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) Restraint type in personal vehicle Car seat 25 (80.6%) 115 (65.0%) 47 (23.7%) 16 (7.8%) 203 (33.2%) Booster seat 3 (9.7%) 44 (24.9%) 106 (53.5%) 27 (13.1%) 180 (29.4%) Seat belt 3 (9.7%) 18 (10.2%) 45 (22.7%) 163 (79.1%) 229 (37.4%) On a lap 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) None 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) Child restraint use in rideshare vehicle** Yes 29 (93.5%) 150 (84.7%) 181 (91.4%) 193 (93.7%) 553 (90.4%) No 2 (6.5%) 27 (15.3%) 17 (8.6%) 13 (6.3%) 59 (9.6%) Restraint type in rideshare vehicle Car seat 20 (64.5%) 76 (42.9%) 21 (10.6%) 14 (6.8%) 131 (21.4%) Booster seat 3 (9.7%) 32 (18.1%) 63 (31.8%) 17 (8.3%) 115 (18.8%) Seat belt 6 (19.4%) 42 (23.7%) 97 (49.0%) 162 (78.6%) 307 (50.2%) On a lap 2 (6.5%) 15 (8.5%) 3 (1.5%) 3 (1.5%) 23 (3.8%) None 0 (0.0%) 12 (6.8%) 14 (7.1%) 10 (4.9%) 36 (5.9%) *Fifteen respondents did not report child age. **Restraint use is defined as being restrained by a car seat, booster, or seat belt. Appropriate restraint use according to child height and weight could not be determined from these data. Table 5. Child restraint use by personal vehicle and rideshare vehicle by age group.

34 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles Restraint Use Infant, <1 Toddler, 1–4 Young Child, 5–8 Older Child, 9–12 Total N = 17 N = 95 N = 92 N = 107 N = 311* Child restraint use in personal vehicle** Yes 17 (100.0%) 95 (100.0%) 92 (100.0%) 107 (100.0%) 311 (100.0%) No 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) Restraint type in personal vehicle Car seat 14 (82.4%) 58 (61.1%) 22 (23.9%) 14 (13.1%) 108 (34.7%) Booster seat 2 (11.8%) 26 (27.4%) 49 (53.3%) 16 (15.0%) 93 (29.9%) Seat belt 1 (5.9%) 11 (11.6%) 21 (22.8%) 77 (72.0%) 110 (35.4%) On a lap 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) None 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) Child restraint use in taxi** Yes 13 (76.5%) 69 (72.6%) 75 (81.5%) 95 (88.8%) 252 (81.0%) No 4 (23.5%) 26 (27.4%) 17 (18.5%) 12 (11.2%) 59 (19.0%) Restraint type in taxi Car seat 11 (64.7%) 23 (24.2%) 11 (12.0%) 15 (14.0%) 60 (19.3%) Booster seat 0 (0.0%) 21 (22.1%) 20 (21.7%) 9 (8.4%) 50 (16.1%) Seat belt 2 (11.8%) 25 (26.3%) 44 (47.8%) 71 (66.4%) 142 (45.7%) On a lap 2 (11.8%) 14 (14.7%) 7 (7.6%) 5 (4.7%) 28 (9.0%) None 2 (11.8%) 12 (12.6%) 10 (10.9%) 7 (6.5%) 31 (10.0%) *Twelve respondents did not report child age. **Restraint use is defined as being restrained by a car seat, booster, or seat belt. Appropriate restraint use according to child height and weight could not be determined from these data. Table 6. Child restraint use by personal vehicle and taxis by age group. Table 7. Parent/caregiver knowledge of laws and penalties related to CRS in personal vehicles and rideshare vehicles. Missing 9 (1.4%) Knowledge of CRS Law and Penalty Total N = 627 Knowledge of CRS law in personal vehicle Yes 456 (72.7%) No 97 (15.5%) Don't know 67 (10.7%) Missing 7 (1.1%) Knowledge of CRS laws in rideshare Yes 288 (45.9%) No 104 (16.6%) Don't know 223 (35.6%) Missing 12 (1.9%) Knowledge of fines if no CRS in rideshare Yes, for parent 157 (25.0%) Yes, for driver 67 (10.7%) Yes, for both 58 (9.3%) No 37 (5.9%) Don't know 303 (48.3%) Missing 5 (0.8%) Who should provide CRS in rideshare Rideshare company 174 (27.8%) Parent 316 (50.4%) No opinion 116 (18.5%) Other 12 (1.9%)

Findings and Applications 35 respondents (43.2%) stated that it was their own responsibility and 30.6% said it was the respon- sibility of the taxi company. This response is notable because most parents and caregivers perceive that they are responsible for the safe transportation of their children, regardless of whether they are driving the vehicle. Data are presented as N (%) for categorical measures. Respondents’ perceptions toward social expectations and attitudes toward CRS use are pre- sented in Table 9. Respondents were instructed to select a value between 1 and 10, where “1” Table 8. Parent/caregiver knowledge of laws and penalties related to CRS in personal vehicles and taxi vehicles. Other 4 (1.2%) Missing 11 (3.4%) Knowledge of CRS Law and Penalty Total N = 324 Knowledge of CRS law in personal vehicle Yes 215 (66.4%) No 63 (19.4%) Don't know 43 (13.3%) Skipped on web 3 (0.9%) Knowledge of CRS laws in taxi Yes 130 (40.1%) No 75 (23.1%) Don't know 112 (34.6%) Skipped on web 7 (2.2%) Knowledge of fines if no CRS in taxi Yes, for parent 64 (19.8%) Yes, for driver 34 (10.5%) Yes, for both 42 (13.0%) No 24 (7.4%) Don't know 151 (46.6%) Missing 9 (2.8%) Who should provide CRS in taxi Taxi company 99 (30.6%) Parent 140 (43.2%) No opinion 70 (21.6%) Installing/uninstalling a car seat or booster seat for an individual rideshare or taxi trip is a hassle. 6.7 (2.8) 7 (5–9) Subjective and Descriptive Norms N = 627 Mean (SD) Median (IQR) People expect me to use a car seat or booster seat. 7.3 (3.0) 8 (5–10) Important people in my life expect me to use a car seat or booster seat. 7.5 (3.0) 9 (5–10) Most people would understand if I didn’t use a car seat or booster seat. 4.6 (3.2) 5 (1–7) I don’t want people to see me riding with my child without using a car seat or booster seat. 6.7 (3.2) 7 (5–10) I would use a car seat or booster seat for my child because I am someone who follows laws. 7.9 (2.6) 9 (6–10) The risk of being in a crash during an individual rideshare/taxi trip is low. 4.8 (2.7) 5 (3–6.5) Carrying a car seat or booster seat for an individual rideshare or taxi trip is inconvenient. 6.8 (2.9) 7 (5–10) Parents/caregivers who use a car seat or booster seat for their child for an individual rideshare or taxi trip are being responsible. 7.9 (2.5) 9 (6–10) Parents/caregivers who do not use a car seat or booster seat for their child during an individual rideshare or taxi trip are being practical. 4.7 (2.8) 5 (2–7) Table 9. Parent/caregiver perceptions related to CRS.

36 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles represents strong disagreement and “10” represents strong agreement. The mean and standard deviation (SD) and the median and interquartile range (IQR) of the responses are listed in the table. The mean and median show the differences in the distribution of responses between items. While responses to some items were normally distributed, others were skewed. Overall, parents and caregivers expressed moderately strong levels of agreement with state- ments related to social expectations for CRS use with their children. Consistent with this pattern, respondents disagreed with statements that people would understand if they did not use CRS. Respondents reported the highest level of agreement with a statement related to following laws, highlighting the importance of legislation in establishing social expectations. Respondents reported neutral agreement with the statement that the risk of being in a crash during an individual trip is low, but the range of responses was skewed toward disagreement. There was a moderate level of agreement with factors related to the inconvenience of carrying and installing a CRS for a rideshare or taxi trip, but at the same time most parents and caregivers disagreed with the statement that not using a CRS was being practical. The central research question concerned the level of child restraint use in personal vehicles compared to that in for-hire vehicles. Table 10 compares the individual-level characteristics of Individual-Level Characteristics CRS Use in Personal Vehicle AND Ridesharing Vehicle CRS Use in Personal Vehicle but NOT in Ridesharing Vehicle N = 567 N = 60 Child age in categories* Infant, <1 29 (5.1%) 2 (3.3%) Toddler, 1–4 150 (26.5%) 27 (45.0%) Younger child, 5–8 181 (31.9%) 17 (28.3%) Older child, 9–12 193 (34.0%) 13 (21.7%) Missing 14 (2.5%) 1 (1.7%) Parent gender Male 182 (32.1%) 25 (41.7%) Female 385 (67.9%) 35 (58.3%) Parent age 18–29 74 (13.1%) 6 (10.0%) 30–44 371 (65.4%) 47 (78.3%) 45–59 115 (20.3%) 6 (10.0%) 60+ 7 (1.2%) 1 (1.7%) Parent race/ethnicity White 255 (45.0%) 26 (43.3%) Black 124 (21.9%) 8 (13.3%) Hispanic 129 (22.8%) 17 (28.3%) Asian 22 (3.9%) 1 (1.7%) Multiple 31 (5.5%) 5 (8.3%) Other 6 (1.1%) 3 (5.0%) Parent education Less than high school 31 (5.5%) 4 (6.7%) High school 73 (12.9%) 10 (16.7%) Vocational training 182 (32.1%) 14 (23.3%) Bachelor 151 (26.6%) 12 (20.0%) Postgraduate 130 (22.9%) 20 (33.3%) Table 10. Individual-level factors associated with CRS use in personal vehicles and rideshare vehicles.

Findings and Applications 37 Number of vehicles 2 (1–2) 2 (1–2) Knowledge of CRS law for personal vehicle Yes 419 (73.9%) 37 (61.7%) No 85 (15.0%) 12 (20.0%) Don't know 57 (10.1%) 10 (16.7%) Knowledge of CRS law for rideshare vehicle Yes 268 (47.3%) 20 (33.3%) No 94 (16.6%) 10 (16.7%) Don't know 193 (34.0%) 30 (50.0%) Knowledge of fines if no CRS in rideshare vehicle Yes, for parent 139 (24.5%) 18 (30.0%) Yes, for driver 65 (11.5%) 2 (3.3%) Yes, for both 53 (9.3%) 5 (8.3%) No 35 (6.2%) 2 (3.3%) Don't know 272 (48.0%) 31 (51.7%) * p <.05. Individual-Level Characteristics CRS Use in Personal Vehicle AND Ridesharing Vehicle CRS Use in Personal Vehicle but NOT in Ridesharing Vehicle N = 567 N = 60 Parent marital status Married 327 (57.7%) 41 (68.3%) Widowed 4 (0.7%) 0 (0.0%) Divorced 43 (7.6%) 5 (8.3%) Separated 19 (3.4%) 1 (1.7%) Never married 174 (30.7%) 13 (21.7%) Parent employment status Working/employee 351 (61.9%) 39 (65.0%) Working/self-employed 69 (12.2%) 6 (10.0%) Temporary not working 29 (5.1%) 8 (13.3%) Looking for work 39 (6.9%) 3 (5.0%) Retired 5 (0.9%) 0 (0.0%) Disabled 15 (2.6%) 1 (1.7%) Other not working 59 (10.4%) 3 (5.0%) Household income <$30K 145 (25.6%) 17 (28.3%) $30K–60K 119 (21.0%) 14 (23.3%) $60K–<100K 114 (20.1%) 8 (13.3%) $100K+ 189 (33.3%) 21 (35.0%) Residence Non-metro area 39 (6.9%) 4 (6.7%) Metro area 528 (93.1%) 56 (93.3%) Table 10. (Continued).

38 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles those who used restraints in both personal vehicles and rideshare vehicles to those who used restraints exclusively in personal vehicles. Univariate analyses were conducted to test the asso- ciation between individual-level characteristics and child restraint use in personal and ride- share vehicles. The only individual-level or behavioral factors that were significantly different between the two groups was child age. The differences between restraint use in personal vehicles and ride- share vehicles were largest for toddlers. This age group represented almost half the children who went from being restrained in personal vehicles to being unrestrained in rideshare vehicles (i.e., 27 out of 60). Other individual-level factors were not significantly different between groups. Knowledge of the laws or associated penalties was also not associated with restraint use in ride- share vehicles. Table 11 compares the individual-level characteristics of those who used restraints in both personal vehicles and taxis to those who used restraints exclusively in personal vehicles. Univari- ate analyses were conducted to test the association between individual-level characteristics and child restraint use in personal vehicles and taxis. The only individual-level or behavioral factors that were significantly different between the two groups was child age. The differences between restraint use in personal vehicles and taxis were largest for toddlers. This age group represented almost half the children who went from being restrained in personal vehicles to being unrestrained in taxis (i.e., 26 out of 62). Other individual-level factors were not significantly different between groups. Knowledge of the laws or associated penalties was also not associated with restraint use in taxis. Toddler, 1–4 69 (26.3%) 26 (41.9%) Younger child, 5–8 75 (28.6%) 17 (27.4%) Older child, 9–12 95 (36.3%) 12 (19.4%) Missing 10 (3.8%) 3 (4.8%) Parent gender Male 86 (32.8%) 23 (37.1%) Female 176 (67.2%) 39 (62.9%) Parent age 18–29 37 (14.1%) 10 (16.1%) 30–44 163 (62.2%) 44 (71.0%) 45–59 60 (22.9%) 7 (11.3%) 60+ 2 (0.8%) 1 (1.6%) Parent race/ethnicity White 103 (39.3%) 24 (38.7%) Black 45 (17.2%) 15 (24.2%) Hispanic 76 (29.0%) 12 (19.4%) Asian 22 (8.4%) 5 (8.1%) Multiple 12 (4.6%) 4 (6.5%) Other 4 (1.5%) 2 (3.2%) Individual-Level Characteristics CRS Use in Personal Vehicle AND Taxi Vehicle CRS Use in Personal Vehicle but NOT in Taxi Vehicle N = 262 N = 62 Child age in categories* Infant, <1 13 (5.0%) 4 (6.5%) Table 11. Individual-level factors associated with CRS use in personal vehicles and taxis.

Findings and Applications 39 Knowledge of CRS law for personal vehicle Yes 177 (67.6%) 38 (61.3%) No 48 (18.3%) 15 (24.2%) Don't know 36 (13.7%) 7 (11.3%) Knowledge of CRS law for taxi vehicle Yes 109 (41.6%) 21 (33.9%) No 60 (22.9%) 15 (24.2%) Don't know 87 (33.2%) 25 (40.3%) Knowledge of fines if no CRS in taxi vehicle Yes, for parent 54 (20.6%) 10 (16.1%) Yes, for driver 28 (10.7%) 6 (9.7%) Yes, for both 32 (12.2%) 10 (16.1%) No 20 (7.6%) 4 (6.5%) Don’t know 123 (46.9%) 28 (45.2%) * p <.05. Household income <$30K 84 (32.1%) 27 (43.5%) $30K–60K 54 (20.6%) 9 (14.5%) $60K–<100K 46 (17.6%) 6 (9.7%) $100K+ 78 (29.8%) 20 (32.3%) Residence Non-metro area 27 (10.3%) 6 (9.7%) Metro area 235 (89.7%) 56 (90.3%) Number of vehicles 2 (1–2) 2 (1–2) Parent education Less than high school 24 (9.2%) 5 (8.1%) High school 48 (18.3%) 8 (12.9%) Vocational training 67 (25.6%) 21 (33.9%) Bachelor 67 (25.6%) 9 (14.5%) Postgraduate 56 (21.4%) 19 (30.6%) Parent marital status Married 151 (57.6%) 35 (56.5%) Divorced 18 (6.9%) 5 (8.1%) Separated 11 (4.2%) 1 (1.6%) Never married 82 (31.3%) 21 (33.9%) Parent employment status Working/employee 147 (56.1%) 34 (54.8%) Working/self-employed 34 (13.0%) 5 (8.1%) Temporary not working 22 (8.4%) 6 (9.7%) Looking for work 23 (8.8%) 6 (9.7%) Retired 2 (0.8%) 1 (1.6%) Disabled 8 (3.1%) 3 (4.8%) Other not working 26 (9.9%) 7 (11.3%) Individual-Level Characteristics CRS Use in Personal Vehicle AND Taxi Vehicle CRS Use in Personal Vehicle but NOT in Taxi Vehicle N = 262 N = 62 Table 11. (Continued).

40 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles Restraint Use Total N = 627* Restraint use in front seat of personal vehicle Always 518 (82.6%) Nearly always 34 (5.4%) Sometimes 53 (8.5%) Seldom 19 (3.0%) Never 2 (0.3%) Restraint use in rear seat of personal vehicle Always 418 (66.7%) Nearly always 70 (11.2%) Sometimes 83 (13.2%) Seldom 34 (5.4%) Never 10 (1.6%) Restraint use in front seat in rideshare vehicle Always 493 (78.6%) Nearly always 35 (5.6%) Sometimes 60 (9.6%) Seldom 22 (3.5%) Never 16 (2.6%) Restraint use in rear seat in rideshare vehicle Always 405 (64.6%) Nearly always 77 (12.3%) Sometimes 93 (14.8%) Seldom 25 (4.0%) Never 16 (2.6%) Crashes while using rideshare 0 537 (85.6%) 1 25 (4.0%) 2 26 (4.1%) 3 8 (1.3%) 4 or more 3 (0.5%) *Missing data account for the differences between the column totals within each category. Table 12. Parent/caregiver restraint use in personal and rideshare vehicles and crashes while using ridesharing. Parent/caregiver restraint use in personal and rideshare vehicles in both front and rear seating positions is presented in Table 12. In total, 88% of adults reported “always” or “nearly always” using a seat belt when seated in the front seat of a personal vehicle in contrast to 84.2% of adults in rideshare vehicles. Restraint use in the rear seat was lower in both personal vehicles and rideshare vehicles. In total, 77.9% of adults reported “always” or “nearly always” using a seat belt in the rear seat of a per- sonal vehicle compared to 76.9% of adults in rideshare vehicles. While most respon- dents (85.6%) reported not being involved in any crashes while using rideshare, 9.9% reported being in at least one crash while using rideshare. Parent/caregiver restraint use in personal vehicles and taxis in both front and rear seating positions is presented in Table 13. In total, 81.8% of adults reported “always” or “nearly always” using a seat belt when seated in the front seat of a personal vehicle in contrast to 76.9% of adults in taxis. Restraint use in the rear seat was lower in both personal vehicles and taxis. In total, 68% of adults reported “always” or “nearly always” using a seat belt in the rear seat of a personal vehicle compared to 66% of adults in taxis. While most respondents (83%) reported not being involved in any crashes while using taxis, 11.3% reported being in at least one crash while using a taxi.

Findings and Applications 41 Table 13. Parent/caregiver restraint use in personal and taxi vehicles and crashes/near-crashes while using taxis. Restraint Use Total N = 324* Restraint use in front seat of personal vehicle 241 (74.4%) 24 (7.4%) 39 (12.0%) 17 (5.2%) Always Nearly always Sometimes Seldom Never 2 (0.6%) Restraint use in rear seat of personal vehicle 191 (59.0%) 29 (9.0%) 62 (19.1%) 26 (8.0%) Always Nearly always Sometimes Seldom Never 6 (1.9%) Restraint use in front seat in taxi 217 (67.0%) 32 (9.9%) 45 (13.9%) 19 (5.9%) Always Nearly always Sometimes Seldom Never 8 (2.5%) Restraint use in rear seat in taxi 163 (50.3%) 52 (16.0%) 68 (21.0%) 25 (7.7%) Always Nearly always Sometimes Seldom Never 7 (2.2%) Number of crashes/near-crashes while using taxi 269 (83.0%) 17 (5.2%) 11 (3.4%) 8 (2.5%) 0 1 2 3 4 or more 1 (0.3%) *Missing data account for the differences between the column totals within each category. Task 5: Document the Barriers and Facilitators to CRS Use with a Focus on Companies, Drivers, and Caregivers The target sample size for the rideshare and taxi driver survey was 500 respondents (250 ride- share drivers and 250 taxi drivers). The rideshare and taxi survey received 291 total responses; 198 of the survey respondents reported only driving for rideshare companies, while the remaining 93 respondents indicated they drive for a taxi company. Of these 93 respondents, 19 reported only driving for a taxi company and 74 reported driving for both rideshare and taxi companies. Most survey respondents (80.8%) were male, which reflects the overall demographics of ride- share and taxi drivers (Hall and Krueger, 2016). Additional demographic data are described in Table 14. Survey respondents were first asked how many trips they complete in a typical day and how long they have worked as a rideshare or taxi driver. Respondents were then asked to describe their seat belt use along with the typical seat belt usage of their front and back seat passengers. The results from these questions are displayed in Table 15.

42 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles Driver Type Rideshare Taxi Total p-Value* Characteristics N = 198 N = 93 N = 291 Race .002 Black 7 (3.5%) 6 (6.5%) 13 (4.5%) Asian 3 (1.5%) 7 (7.5%) 10 (3.4%) Multiple/Native 16 (8.1%) 15 (16.1%) 31 (10.7%) White 170 (85.9%) 61 (65.6%) 231 (79.4%) Missing 2 (1.0%) 4 (4.3%) 6 (2.1%) Hispanic ethnicity .010 Yes 51 (25.8%) 20 (21.5%) 71 (24.4%) No 145 (73.2%) 68 (73.1%) 213 (73.2%) Missing 2 (1.0%) 5 (5.4%) 7 (2.4%) Gender .24 Male 166 (83.8%) 69 (74.2%) 235 (80.8%) Female 31 (15.7%) 19 (20.4%) 50 (17.2%) Missing 1 (0.5%) 5 (5.4%) 6 (2.1%) Has children .66 Yes 51 (25.8%) 20 (21.5%) 71 (24.4%) No 145 (73.2%) 68 (73.1%) 213 (73.2%) Missing 2 (1.0%) 5 (5.4%) 7 (2.4%) *Fisher’s exact test was used to compare group differences. Table 14. Rideshare and taxi driver demographics. The following tables show that a significant amount of data is missing from taxi drivers. There are likely multiple reasons for this, but the survey length may have exacerbated this issue. It is possible that these drivers skipped the taxi section due to respondent fatigue. Child Passengers, Restraint Use, Awareness of Local Laws, and Company Training All survey respondents (N = 291) were asked how many of their daily trips include a child younger than 8 years old with a parent/guardian. Seventy-three respondents noted no rides by children less than 8 years of age, and 19 did not respond to this question. The remaining 199 respon- dents were asked how these children typically ride in their vehicle, whether they carry a CRS in their vehicle, and whether they encourage parents/guardians to bring a CRS with them for their next trip. Table 16 shows the responses to those questions. All survey respondents (N = 291) were asked additional questions about CRS that covered CRS guidance from their rideshare or taxi company, knowledge of state CRS laws, and information/ training about local CRS laws. One question, which asked about company guidance, allowed multiple responses, so the question choices are presented separately in Table 17. Overall, 46 respondents did not respond, 225 gave one choice, 18 gave two choices, and two respondents gave three choices. Of the drivers who selected one response, 37% (N = 83) said that they received no guidance from their ride- share or taxi company, 27% (N = 60) said that their company told them passengers are respon- sible for bringing their own CRS, 23% (N = 51) said their company told them to recommend that passengers bring a CRS, and only 3% (N = 7) said that their company told them they must carry a child seat/booster seat in their vehicle. The remaining 10% (N = 24) indicated that they had received some other kind of guidance from their company. Some of those drivers explained that their company is vague on guidance, others indicated that they will not drive a child with- out a child seat, and one driver mentioned how it is not practical to carry a CRS because it takes up valuable space in the trunk, which is frequently used by passengers to carry items like luggage.

Findings and Applications 43 Table 15. Rideshare and taxi driver trip characteristics and safety behaviors. Driver Type Characteristics and Behavior Rideshare N = 198 Taxi N = 93 Rideshare type Multiple companies 137 (69.2%) Lyft or Uber 61 (30.8%) Daily number of trips (median, IQR) 10 (7–15) 5 (2–10) Driving duration <3 months 1 (0.5%) 27 (29.0%) 3–6 months 7 (3.5%) 3 (3.2%) 7–12 months 12 (6.1%) 5 (5.4%) >12 months 178 (89.9%) 17 (18.3%) Missing 0 (0.0%) 41 (44.1%) Driver seat belt use Always 194 (98.0%) 45 (48.4%) Almost always 0 (0.0%) 2 (2.2%) Sometimes 1 (0.5%) 1 (1.1%) Almost never 1 (0.5%) 1 (1.1%) Never 2 (1.0%) 0 (0.0%) Missing 0 (0.0%) 44 (47.3%) Passenger seat belt use in front seat Always 126 (63.6%) 32 (34.4%) Almost always 41 (20.7%) 7 (7.5%) Sometimes 25 (12.6%) 6 (6.5%) Almost never 3 (1.5%) 2 (2.2%) Never 1 (0.5%) 2 (2.2%) Missing 2 (1.0%) 44 (47.3%) Drives passenger who refuses to use seat belt in front seat Always 10 (5.1%) 7 (7.5%) Almost always 7 (3.5%) 0 (0.0%) Sometimes 11 (5.6%) 11 (11.8%) Almost never 46 (23.2%) 7 (7.5%) Never 123 (62.1%) 23 (24.7%) Missing 1 (0.5%) 45 (48.4%) Passenger seat belt use in back seat Always 46 (23.2%) 12 (12.9%) Almost always 63 (31.8%) 13 (14.0%) Sometimes 63 (31.8%) 11 (11.8%) Almost never 17 (8.6%) 7 (7.5%) Never 0 (0.0%) 2 (2.2%) Missing 9 (4.5%) 48 (51.6%) Drives passenger who refuses to use seat belt in back seat Always 34 (17.2%) 6 (6.5%) Almost always 30 (15.2%) 2 (2.2%) Sometimes 40 (20.2%) 19 (20.4%) Almost never 47 (23.7%) 8 (8.6%) Never 46 (23.2%) 13 (14.0%) Missing 1 (0.5%) 45 (48.4%)

44 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles Who is responsible for CRS Driver 2 (1.0%) 1 (1.1%) Rideshare/taxi company 9 (4.5%) 6 (6.5%) Parent/guardian 180 (90.9%) 33 (35.5%) Don't know/no opinion 7 (3.5%) 12 (12.9%) Missing 0 (0.0%) 41 (44.1%) Company guidance: required to carry CRS in vehicle Yes 6 (3.0%) 4 (4.3%) No 191 (96.5%) 44 (47.3%) Missing 1 (0.5%) 45 (48.4%) Company guidance: passengers are responsible for providing CRS Yes 69 (34.8%) 8 (8.6%) No 128 (64.6%) 40 (43.0%) Missing 1 (0.5%) 45 (48.4%) Company guidance: recommend passengers bring CRS Yes 57 (28.8%) 12 (12.9%) No 140 (70.7%) 36 (38.7%) Missing 1 (0.5%) 45 (48.4%) Company guidance: no guidance received Yes 64 (32.3%) 19 (20.4%) No 133 (67.2%) 29 (31.2%) Missing 1 (0.5%) 45 (48.4%) Does driver’s state have any CRS law for rideshare or taxi vehicles? Yes 159 (80.3%) 34 (36.6%) No 3 (1.5%) 49 (52.7%) Don't know 36 (18.2%) 10 (10.8%) Driver Type Practice and Law Awareness Rideshare N = 198 Taxi N = 93 Table 17. Awareness of laws and company practices among rideshare and taxi drivers. CRS Use Practice Driver Type Rideshare N = 142 Taxi N = 57 How children ride in vehicle In a seat belt 54 (38.0%) 2 (3.5%) In an infant, child, or booster seat 80 (56.3%) 6 (10.5%) Not in a seat belt, e.g., on a parent's lap 6 (4.2%) 2 (3.5%) Other 2 (1.4%) 0 (0.0%) Missing 0 (0.0%) 47 (82.5%) Does driver carry CRS in vehicle? Yes 11 (7.7%) 2 (3.5%) No 131 (92.3%) 8 (14.0%) Missing 0 (0.0%) 47 (82.5%) Does driver recommend passengers bring CRS for future trips? Yes 116 (81.7%) 8 (14.0%) No 25 (17.6%) 2 (3.5%) Missing 1 (0.7%) 47 (82.5%) Table 16. CRS use practices of rideshare and taxi drivers.

Findings and Applications 45 Rideshare/taxi company provided information on local CRS laws Yes 60 (30.3%) 19 (20.4%) No 137 (69.2%) 28 (30.1%) Missing 1 (0.5%) 46 (49.5%) Rideshare/taxi company provided education on local CRS laws Yes 70 (35.4%) 19 (20.4%) No 125 (63.1%) 28 (30.1%) Missing 3 (1.5%) 46 (49.5%) Rideshare/taxi company provided training on local CRS laws Yes 8 (4.0%) 5 (5.4%) No 190 (96.0%) 42 (45.2%) Missing 0 (0.0%) 46 (49.5%) Does driver’s state have any CRS law for children younger than 8 who are riding in a rideshare or taxi? Yes 130 (65.7%) 26 (28.0%) No 49 (24.7%) 61 (65.6%) Don’t know 19 (9.6%) 6 (6.5%) Driver Type Practice and Law Awareness Rideshare N = 198 Taxi N = 93 Does driver’s state have a fine if a child younger than 8 is not properly secured in a rideshare or taxi? Yes 101 (51.0%) 24 (25.8%) No 49 (24.7%) 62 (66.7%) Don’t know 48 (24.2%) 7 (7.5%) Table 17. (Continued). Task 6: Identify Objective Techniques to Measure the Use of CRS in For-Hire Motor Vehicles The AG was asked to provide insights based on their individual experience on potential tech- niques for objectively measuring the use of CRS in rideshare vehicles and taxis. The AG interpreted “objective techniques” to mean methods that will minimize measurement bias. The strengths and limitations of self-reported, observational and video-recordings of CRS use in for-hire vehicles were discussed. Self-reported surveys are prone to sampling bias, social desirability bias, and recall bias. How- ever, with these limitations recognized, periodic survey data collected on CRS use in ridesharing and taxi vehicles in a nationally representative sample with appropriate sampling weights could inform understanding of trends and factors related to use and non-use of CRS, particularly if the survey probed barriers and facilitators that underlie behavior choices. Roadside observation of CRS use in ridesharing and taxi vehicles may also be prone to sampling bias if site selection is not carefully considered. For example, observing CRS use only in airport or hotel pickup areas for rideshare and taxi services may be biased toward higher income families. Using the approach developed for the 2019 National Survey of the Use of Booster Seats (Enriquez, 2021) involving data collection at a range of sites, including gas stations, fast-food restaurants, daycare centers, and recreation centers, may be one way to improve the rigor of sampling. Unlike other observational studies of restraint use, if such a method were used for rideshare vehicles, attention would need to focus on identifying ridesharing vehicles by their identifying marks (e.g., logos for Uber or Lyft) on the vehicle. However, these logos are not universally used, so a certain proportion of ridesharing trips are likely to be missed.

46 Identifying and Prioritizing Behavioral Interventions to Improve Child Passenger Safety in For-Hire Vehicles Video data from in-vehicle video recorders could be another approach to collecting objective data on the use of CRS in for-hire vehicles. The use of cameras in rideshare vehicles is not uncom- mon (e.g., to monitor passenger behavior). However, there may be a self-selection bias toward drivers who are willing to participate in a program that collects video data, and this could limit the generalizability of the findings. A naturalistic study of ridesharing and taxi drivers could be one approach to collecting objective data about CRS use in these vehicles. Any approach to col- lecting in-vehicle data in for-hire vehicles would also need to comply with institutional review board requirements and consider issues such as passenger consent and privacy. Task 7: Describe the Roles of Stakeholders Who Can Influence the Use of CRS The AG was asked to consider a range of stakeholders who could influence the use of CRS in for-hire vehicles. Examples identified by the group include the following: 1. Ridesharing corporations: a. There is a need to explore the limits of rideshare corporation collaboration on this safety problem and find ways to improve the safety of children without hurting the economic situation of the drivers. Rideshare corporations should share responsibility for protection of child passengers rather than placing the onus on drivers or parents. b. State highway safety offices may consider advocating for further rideshare corporation collaboration specifically on the protection of child passengers in their vehicles. Strategies need to be identified to engage corporate responsibility on this issue. This could include a pilot whereby a state or city requires rideshare corporations to provide and sanitize appro- priate CRS for all vehicles, alongside appropriate training for drivers. c. Corporations could be required to provide appropriate child restraint devices in a designated portion of their vehicles as a requirement to receive a permit to operate in a city or state. This type of arrangement already exists with micro-mobility companies whereby, to receive permission to operate in more profitable sections of the city, a certain number of electric scooters are required to be distributed in neighborhoods identified as mobility deserts. 2. Rideshare and taxi drivers: With corporate support and financial incentives, drivers could be encouraged to install appropriate CRS in their vehicle. This could enable them to offer special- ized rideshare services for parents with younger children that the parent would request when hailing the vehicle. 3. Pediatricians, other health professionals, and child passenger safety technicians: Health and safety professionals who interact with parents and caregivers could provide guidance and education on best practices for child safety in for-hire vehicles. The AG was not aware of any outreach to these groups on this issue. 4. Safety and consumer advocacy groups: Organizations such as Advocates for Highway and Auto Safety, the Insurance Institute for Highway Safety, and Consumer Reports could be involved in advocacy efforts to encourage ridesharing corporations to share responsibility for child passenger safety and to disseminate information to parents and caregivers. 5. Travel industry: Airlines, travel agents, and hotels routinely provide information to travelers about their destination. This could include information about child restraint laws in the state or city destination and ways to travel safely with children. 6. U.S. Department of Transportation: Funding opportunities or other incentives could be explored to stimulate manufacturers to develop products that may be in the public interest but not otherwise profitable. One example is portable or lightweight child seats, which would be convenient for rideshare users but may be unprofitable for manufacturers. Consumer infor- mation programs, such as the New Car Assessment Program, could be used to recognize innovation and stimulate development and marketing of practical travel versions of child safety seats.

Findings and Applications 47 Task 8: Develop Targeted Behavioral Change Strategies and Messages to Promote Child Passenger Safety, Including At-Risk and Hard-to-Reach Populations The AG suggested a three-pronged approach to promote child safety in for-hire vehicles. These approaches involve education and environmental changes that are likely to influence behavior: 1. Education of parents and caregivers: Parents and caregivers need to be reminded of the injury risks of traveling with children who are not appropriately restrained, even on short trips. In addition, parents and caregivers should be provided with the knowledge and tools to make informed decisions about the appropriate seats in different situations. 2. Child seat integration directly into vehicles: In the current market only a few car models are equipped with integrated child seats, and there seems to be little innovation in this area. Regula- tions for integrated child seats would face cost/benefit challenges and would require decades to develop and disseminate. Increased consumer interest could generate market demand for this feature, and vehicle manufacturers could be encouraged or incentivized to produce vehicles that have integrated child safety seats through non-regulatory means, as was the case with auto- matic emergency braking. 3. Travel-friendly CRS by car seat manufacturers: Although there are technical challenges in designing a compliant car seat that is also portable, there may be potential for compliant designs for certain weight/size ranges. Such designs may lack the utility of seats that can accommodate a wide range of weights and sizes, but their convenience could increase use in rideshare. Increased consumer demand for such seats could motivate manufacturers to invest in their development. The group discussed whether a harm reduction approach might be appropriate for CRS use in for-hire vehicles. In general, children should be restrained according to the safest possible standard, but could this standard be lowered in the context of for-hire vehicles? For example, a 3-year-old child should not use a backless booster, but considering that the alternative is likely to be no CRS use, this might be appropriate for occasional ridesharing trips. As part of a harm reduction approach, the group considered whether a special category of FMVSS could be developed for CRS for use in for-hire vehicles. While intriguing from a harm reduction perspective, this approach is problematic in other respects and would need to be carefully studied to ensure that the safety benefit added by increased use in rideshare vehicles was not offset by decreased safety when such seats are used in place of conventional child seats in other vehicles. Task 9: Propose Strategies to Evaluate the Guidance Evaluation of these approaches would need to be ongoing, using objective observation of CRS use in for-hire vehicles and market scans to see if new products are being introduced to enhance child safety, either by vehicle manufacturers or by restraint manufacturers. As described in the preceding section, the use of child safety seats in rideshare vehicles and taxis should continue to be assessed in the context of child seat use in all passenger cars. This perspective is necessary to point to gaps in use that might indicate opportunities to improve child passenger safety in rideshare vehicles and taxis and to ensure that any steps taken to increase child safety in rideshare vehicles and taxis do not have unintended effects on child safety in personal vehicles. This is especially critical if a harm reduction approach might be considered to improve child safety in rideshare vehicles and taxis. Since child risk exposure in terms of annual miles traveled is much higher in personal cars, even a small compromise in child passenger pro- tection in personal vehicles—brought about, for example, by portable child seats with reduced safety expectations—might more than offset any benefit in rideshare and taxi safety.

48 C H A P T E R   8 Areas for Further Research and Intervention The research conducted in pursuit of the objectives prescribed by the BTSCRP provides new and valuable information to inform strategies to improve child safety by increasing child restraint use in rideshare and taxi vehicles. While additional research is needed to refine approaches and prioritize interventions, the findings point to several areas of potential. The findings of the legal review indicate that additional state and/or city laws could be con- structive, both through direct effects on compliance by parents, caregivers, and rideshare and taxi corporations, and through their indirect influence in communicating safety expectations to these groups and other stakeholders. Exemptions for rideshare vehicles in child passenger protec- tion laws could be removed in seven states and exemptions for taxis could be removed in 28 states. Considering the strong effect that use laws had on child restraint use in private vehicles, these legislative amendments will be critical for optimizing child safety. In addition, innovative new legislation that places more responsibility on rideshare and taxi com- panies for providing child safety seats for customers with children and/or facilitating their use may be an effective means for increasing use rates. Such legislation could be pilot tested and evaluated to assess effectiveness and encourage replication. Similarly, advocacy from respected safety leaders may be effective in highlighting the potential role of corporations in reducing barriers to child seat use in rideshare vehicles and encouraging these businesses to act even without legislative mandates. Findings from the surveys of parents, caregivers, and taxi and rideshare drivers and how out- reach to rideshare and taxi companies did not yield responses indicate that responsibility for child safety is currently focused on parents and caregivers and reflect the challenges faced by rideshare and taxi passengers in providing the best protection for their children (National Center for Statistics and Analysis, 2022). Child restraint systems (CRS) are typically bulky, heavy, and inconvenient for parents and caregivers to carry along on trips, short or long, that involve use of for-hire services. Reducing these convenience barriers through engineering innovation holds promise for increasing use. Introducing CRS that are easier to carry and/or introducing vehicle seating systems with integrated child passenger protection are likely to lead to increased use. These types of engineering solutions could be stimulated by regulation or incentivized through consumer demand, perhaps with the help of vehicle safety ratings and awareness campaigns. Limitations While the review of state and city laws was conducted using state-of-the-art legal review methods, it is possible that there is a state or locality with relevant CRS law that the research team did not capture. As there is no comprehensive database of city laws and regulations, this may be more Conclusions and Suggested Research

Conclusions and Suggested Research 49 likely to have occurred for these jurisdictions. The team also limited its search to the 50 largest cities, which constrained the number of legal environments examined. Second, the team only identi- fied the laws themselves, which may not reflect the realities of enforcement and implementation. Self-reported surveys are prone to sampling bias, social desirability bias, and recall bias. How- ever, with these limitations recognized, the survey data collected on CRS use in ridesharing and taxi vehicles with a nationally representative sample can be used to inform an understanding of the factors related to use and non-use of CRS. Conclusions The current high rate of CRS use in private motor vehicles is among the greatest public health successes. Achieving similar use rates in rideshare and taxi vehicles would carry that success further. This report identifies several pathways that show promise for realizing these benefits. As always when considering a public health intervention, precautions need to be taken against unintended consequences. In this case, new policies, programs, or engineering approaches should be carefully tested and evaluated to ensure that they do not compromise current levels of child passenger protec- tion in personal vehicles.

50 Burns J. 2019. Will Uber and Lyft ever stop fighting laws, or their workers? Forbes. https://www.forbes.com/sites /janetwburns/2019/11/08/will-platforms-like-uber-and-lyft-ever-stop-fighting-the-law-or-their-workers/. Accessed January 3, 2020. Cook C, Diamond R, Hall J, List JA, Oyer P. 2018. The Gender Earnings Gap in the Gig Economy: Evidence from over a Million Rideshare Drivers. NBER Working Paper 24732. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/papers/w24732. Ehsani JP, Michael JP, Gielen A. 2021. Rideshare use among parents and their children. Injury Epidemiology 8(1):9. https://doi.org/10.1186/s40621-021-00302-4. Enriquez J. 2021. The 2019 National Survey of the Use of Booster Seats. National Highway Traffic Safety Admin- istration, Washington, DC. Hall JV, Krueger AB. 2016. An Analysis of the Labor Market for Uber’s Driver-Partners in the United States. NBER Working Paper 22843. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org /papers/228843. Hertz E. 1996. Revised Estimates of Child Restraint Effectiveness. National Highway Traffic Safety Administration, Washington, DC. IIHS/HLDI. 2022. Seat belt and child seat laws. https://www.iihs.org/topics/seat-belts/seat-belt-law-table. Accessed July 11, 2022. Kahane CJ. 2015. Lives Saved by Vehicle Safety Technologies and Associated Federal Motor Vehicle Safety Standards, 1960 to 2012. National Highway Traffic Safety Administration, Washington, DC. Li HR, Pickrell T. 2018. The 2017 National Survey of the Use of Booster Seats. Report No. DOT HS 812 617. National Highway Traffic Safety Administration, Washington, DC. Lyft Help. 2020. State and city requirements. https://help.lyft.com/hc/en-us/all/sections/115003494688. Accessed January 3, 2020. Lyft, Inc. 2022, February 8. Lyft Announces Solid Q4’21 and Fiscal 2021 Results. https://investor.lyft.com /news-and-events/news/news-details/2022/Lyft-Announces-Solid-Q421-and-Fiscal-2021-Results/default .aspx. Accessed July 11, 2022. National Center for Statistics and Analysis. 2022. Children: 2020 Data. National Highway Traffic Safety Admin- istration, Washington, DC. Owens JM, Womack KN, Barowski L. 2019. Factors Surrounding Child Seat Usage in Rideshare Services. Safety through Disruption National University Transportation Center. https://www.vtti.vt.edu/utc/safe-d/wp -content/uploads/2019/10/01-005_FinalResearchReport_Final.pdf. Polzin SE, Chu X, Godfrey J. 2014. The impact of millennials’ travel behavior on future personal vehicle travel. Energy Strategy Reviews 5:59–65. https://doi.org/10.1016/j.esr.2014.10.003. Prince P, Hines LM, Bauer MJ, Liu C, Luo J, Garrett M, Pressley JC. 2019. Pediatric restraint use and injury in New York City taxis compared with other passenger vehicles. Transportation Research Record: Journal of the Trans- portation Research Board, Vol. 2673, Issue 7. Published online https://doi.org/10.1177/0361198119843091. Uber Technologies, Inc. 2022, February  9. Uber Announces Results for Fourth Quarter and Full Year 2021. https://investor.uber.com/news-events/news/press-release-details/2022/Uber-Announces-Results-for -Fourth-Quarter-and-Full-Year-2021/. Accessed July 11, 2022. VTTI/TTI. 2022. State Laws for Child Safety Seats in Rideshare Vehicles. https://kidsridesafe.org/map.html. Accessed July 11, 2022. Webb CN. 2020. Motor Vehicle Traffic Crashes as a Leading Cause of Death in the United States, 2016 and 2017. National Highway Traffic Safety Administration, Washington, DC. West BA, Rudd RA, Sauber-Schatz EK, Ballesteros MF. 2021. Unintentional injury deaths in children and youth, 2010–2019. Journal of Safety Research 78:322–330. https://doi.org/10.1016/j.jsr.2021.07.001. References

51   Appendices A–J Appendices A–J are not printed herein but are available for download from the National Academies Press website (nap.nationalacademies.org) by searching for BTSCRP Research Report 6. Appendix A: Implementation of Research Findings and Products Appendix B: Strengths and Limitations of the Research and Recommendations for Addi- tional Research Appendix C: Advisory Group Comments and Recommendations Appendix D: Electronic Presentation of the Guidance Appendix E: Summary of Rideshare Company Practices Appendix F: Summary of State Child Restraint Policies Appendix G: Summary of City Child Restraint Policies Appendix H: Parent/Caregiver Survey Instrument Appendix I: Rideshare Driver Survey Instrument Appendix J: Taxi Driver Survey Instrument

Abbreviations and acronyms used without denitions in TRB publications: A4A Airlines for America AAAE American Association of Airport Executives AASHO American Association of State Highway Officials AASHTO American Association of State Highway and Transportation Officials ACI–NA Airports Council International–North America ACRP Airport Cooperative Research Program ADA Americans with Disabilities Act APTA American Public Transportation Association ASCE American Society of Civil Engineers ASME American Society of Mechanical Engineers ASTM American Society for Testing and Materials ATA American Trucking Associations CTAA Community Transportation Association of America CTBSSP Commercial Truck and Bus Safety Synthesis Program DHS Department of Homeland Security DOE Department of Energy EPA Environmental Protection Agency FAA Federal Aviation Administration FAST Fixing America’s Surface Transportation Act (2015) FHWA Federal Highway Administration FMCSA Federal Motor Carrier Safety Administration FRA Federal Railroad Administration FTA Federal Transit Administration GHSA Governors Highway Safety Association HMCRP Hazardous Materials Cooperative Research Program IEEE Institute of Electrical and Electronics Engineers ISTEA Intermodal Surface Transportation Efficiency Act of 1991 ITE Institute of Transportation Engineers MAP-21 Moving Ahead for Progress in the 21st Century Act (2012) NASA National Aeronautics and Space Administration NASAO National Association of State Aviation Officials NCFRP National Cooperative Freight Research Program NCHRP National Cooperative Highway Research Program NHTSA National Highway Traffic Safety Administration NTSB National Transportation Safety Board PHMSA Pipeline and Hazardous Materials Safety Administration RITA Research and Innovative Technology Administration SAE Society of Automotive Engineers SAFETEA-LU Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (2005) TCRP Transit Cooperative Research Program TEA-21 Transportation Equity Act for the 21st Century (1998) TRB Transportation Research Board TSA Transportation Security Administration U.S. DOT United States Department of Transportation

Transportation Research Board 500 Fifth Street, NW Washington, DC 20001 ADDRESS SERVICE REQUESTED ISBN 978-0-309-69860-3 9 7 8 0 3 0 9 6 9 8 6 0 3 9 0 0 0 0