Ambulances & Vehicle Ops, Operations

Survey Measures EMS Personnel’s Perceptions of Self-Driving Ambulances

Issue 4 and Volume 43.

Survey measures provider perceptions of self-driving ambulances

You arrive on-scene to find a gunshot victim who’s bleeding profusely. You and your partner share a knowing look—you’ll need a second set of hands to work on this patient. Luckily you brought the automated ambulance to the scene.

You and your partner jump in back with the patient as the ambulance drives itself to the closest trauma center, allowing the two of you to work as a team on a rapidly declining patient when normally one of you would have to drive.

An artist’s concept of the exterior of a self-driving ambulance. Image courtesy Joseph Keebler

Background: As self-driving cars become a reality, manufacturers and legislators alike are exploring ways to maximize the benefits and reduce the risks, of self-driving technology.1,2 Continued evolution of the technology may eliminate the need for a driver altogether, which may also eliminate the personal and public health hazard of distracted, impaired and fatigued drivers.

Emergency vehicles aren’t immune to the risks of imperfect drivers. Every year in the United States, one-third of the 4,500 non-fatal traffic accidents involving ambulances cause injuries. More than two dozen of these accidents are fatal.3 EMS providers are more likely to sustain a fatal injury than any other profession, and the most common fatal injury occurs in a motor vehicle collision.4 Adequate driver training, in addition to proper passenger restraints, is a focused recommendation.5 A 2015 publication from the Department of Homeland Security details dozens of recommendations on best driving practices for emergency vehicle operations.6 Self-driving ambulances may offer a solution to these risks.

Current ambulances are most commonly staffed with two providers; when transporting a patient to the hospital, one drives while the other renders aid. In cases of severely ill or injured patients, additional personnel from other emergency vehicles will assist in the back of the ambulance with the patient.

Research has demonstrated that having two individuals delivering patient care will generally lead to better outcomes due to a team being able to coordinate, support and check each other’s actions.7,8

There are practical constraints to this scenario. Staffing response vehicles with three providers is expensive, and therefore, not often the norm in EMS agencies. Self-driving ambulances could potentially ensure that two medical providers are always present in the back of the ambulance to deliver care while traveling to a hospital.

Methods: Although this technology doesn’t currently exist, it’s important to understand how professionals in the field would react to self-driving ambulances.

This study surveyed 168 paramedics and EMS personnel on their reaction to potentially working in a self-driving ambulance.

This study was conducted using a correlational design, with data collected via a survey delivered in Google Forms. The survey included demographics questions concerning the EMS provider’s organization, background, experience and other individual difference factors.

Providers were randomly assigned to one of two conditions. Each participant was either entered into a “normal” condition, where they were told they were in an ambulance with another medical provider who was driving, or in a “self-driving” condition, where they were with another provider who was administering care in the back while the ambulance drove itself.

After reading a brief vignette describing either the normal or self-driving ambulance, providers rated their willingness to ride in the ambulance using a Willingness to Ride (WTR) scale.9 They were also asked to rate their emotions with regards to the vignette using a universal emotional rating scale that evaluates each of the six main emotions: happiness, disgust, anger, sadness, surprise and fear.10

Respondents were solicited from JEMS readers, through email listservs and social media posts (e.g., Facebook). The data was analyzed through Preacher and Hayes regression mediation modeling with the two conditions used as a two-level factor (self-driving or standard) and the six emotions utilized as mediators.

Results: In total, 168 providers responded from across the U.S. completed the survey. Of those who responded, 86% stated their normal configuration is one provider in the back while another provider drives; 13% stated there’s one driver with two in the back, and < 1% stated a different configuration.

Results indicated a significant model
(R2 = 0.64, F (7, 157) = 40.44, p < 0.0001), with a difference in WTR for those who were in the self-driving ambulance condition, such that those in the self-driving condition were less willing to ride in the ambulance than those in the standard configuration condition.

There was a direct effect on WTR coming from the manipulation, and indirect effects for the emotions of fear and happiness. In other words, overall, providers were less willing to ride in the self-driving ambulance, and this seems to be explained through high amounts of fear and low amounts of happiness when compared to the normal condition.

Discussion: This research has important implications for the future of self-driving emergency vehicles.

First, it seems that the current community of providers isn’t very willing to utilize this technology at this time. It’s possible that because this concept seems abstract, providers are fearsome of the technology, unhappy about it, and therefore are unwilling to ride. This could potentially be solved in the intervening years through education and increased familiarity of this new technology.

Just as Uber and Google are working to familiarize riders with driverless vehicles, the EMS community may benefit from an informative campaign that demonstrates benefits of driverless vehicles. By the time automated ambulances become a reality, there might be a cohort effect such that new providers are much more accepting of automated ambulances.

The second important finding is that providers were unwilling to ride in the automated condition, even though most of them (86%) stated that their ambulances have providers who work alone in the back. This is in contrast to the idea that excellent teamwork has been demonstrated to increase patient outcomes, including mortality.11 This finding suggests that despite being limited in resources and staffing, EMS providers don’t yet feel that the technology of driverless vehicles is ready to provide a worthy solution, even though it arguably would enhance teamwork in this setting.

Conclusion: It’s no longer a question of if, but of when technology will allow for automated ambulances. This survey reveals a significant distrust and unwillingness to use this technology, despite the potential benefits of having larger teams available to deliver patient care. This may have important implications for the introduction of this technology, and suggests a need for educational outreach to introduce the positive potential of automated ambulances into a population that apparently doesn’t want to ride in them.


1. Greene JD. Our driverless dilemma: When should your car be willing to kill you? Science Magazine. 2016;352(6293):1514–1516.

2. Arbogast C. (May 9, 2017.) Experiments show that a few self-driving cars can dramatically improve traffic flow. Illinois College of Engineering. Retrieved March 6, 2018, from

3. Smith N. A national perspective on ambulance crashes and safety. EMS World. 2015;44(9):91–2, 94.

4. Jaynes C. The price of safety: Comparing the return on investment in safe driving systems. JEMS. 2016;41(10):44–48.

5. Centers for Disease Control and Prevention. Ambulance crash-related injuries among Emergency Medical Services workers—United States, 1991–2002. MMWR Morb Mortal Wkly Rep. 2003;52(8):154–156.

6. U.S. Department of Homeland Security. (2015.) A research study of ambulance operations and best practice considerations for emergency medical services personnel. Retrieved March 6, 2018 from

7. Keebler JR, Lazzara EH, Misasi PM, editors: The human factors and ergonomics of pre-hospital emergenc care. CRC Press: Boca Raton, Fla., 2017.

8. Salas E, Rosen MA, Burke CS, et al: The wisdom of collectives in organizations: An update of the teamwork competencies. In Team effectiveness in complex organizations: Cross-disciplinary perspectives and approaches. Routledge: New York, pp. 39–79, 2009.

9. Winter SR, Rice S, Keebler JR, et al. (2017, March). Patient perceptions on the use of an auto-piloted emergency medical transport: An affective perspective [conference presentation]. International Symposium on Human Factors and Ergonomics in Health Care: New Orleans, 2017.

10. Rice S, Winter SR. Which emotions mediate the relationship between type of pilot configuration and willingness to fly? Journal of Aviation Psychology and Applied Human Factors. 2015;5(2):83–92.

11. Hughes AM, Gregory ME, Joseph DL, et al. Saving lives: A meta-analysis of team training in healthcare. J Appl Psychol. 2016:101(9);1266–1304.