We examined predictors of occupational injury and assault among a population of mostly Hispanic, Black, and Asian food delivery gig workers in NYC using unique data from the NYC-DCWP survey. Two key findings emerged from our analysis. First, among the 1,650 respondents in our sample, most (67%) were fully dependent on platform-based work and over 21% reported being injured or assaulted while on the job. Second, full dependence on platform work was significantly associated with greater probability of injury and assault, by 61% and 36%, respectively, relative to partial dependence, even after adjusting for work experience, delivery mode, and weekly hours. Together, these findings point to the clear dangers associated with platform-based delivery work, especially for two-wheeled drivers, and reinforce the importance of dependence as a key moderator of risk. They also challenge company narratives that most people engage in gig delivery work as a supplemental, “flexible” form of paid work.
The diversity in levels of economic dependence found in our study population is a distinct characteristic of platform work associated with platform companies’ willingness to accept workers irrespective of their other work commitments.11 The high proportion of fully dependent workers in our and in other study samples,10,26 along with the fact that 40% of fully dependent respondents worked 40 + hours per week, raises questions regarding whether workers are able to benefit from the flexibility that food delivery companies advertise.
Importantly, dependence on platform-based work was an indicator of high social vulnerability in our sample. Although most respondents were of a racialized minority population, fully dependent respondents were less likely to use cars and more likely to have limited English language proficiency than partially dependent respondents. In this context, the limited (or non-existent) access to worker protections, such as workers’ compensation insurance, among high-risk dependent respondents can exacerbate existing health disparities; foreign-born workers of low socio-economic position may be less able to cope with the acute and downstream consequences of a work-related injury (e.g., lost wages, medical expenses, difficulty paying important bills, and threats to housing).
The positive association between dependence and work injury among food delivery gig workers echoes recent research on this emerging topic. Using qualitative data, Schor et al (2023) explored the relationship between dependence level and work risks and concluded that those with greater economic insecurity and who were more dependent on the income from the platform job were less likely to feel in control of key aspects of their work, including flexibility to choose which times, places, specific tasks and platforms maximize compensation.28 Similar to our findings, Jing et al (2023) found that dependence on platform-based work was associated with 23% greater odds of work injury compared to non-dependence in a study among platform-based food delivery workers in China.10 They also found that workload was a mediator on the pathway between dependence and injury.10 Further research is needed to replicate these findings and identify other mechanisms for how higher levels of dependence place workers at greater risk for injuries and assaults.
Research on the health effects of algorithmic management suggests other avenues in which dependence on platform-based work may precipitate occupational injury and assault. Specific features of algorithmic management, used by platform-based food delivery companies for performance monitoring, scheduling, compensation, and hiring and termination, have the potential to contribute to income insecurity, schedule instability, isolation, and limited decision authority, which are linked with feelings of stress, fatigue, anxiety, and anger.3 In this state of heightened stress, combined with a high workload, workers may be more easily distracted and likely to engage in risk-taking behaviors, such as speeding or not waiting at traffic lights, which are behaviors reported among food delivery workers9,10 and risk factors of traffic collisions and accidents.17 Moreover, platform companies incentivize workers with bonuses to make deliveries during inclement weather when delivery demand is high and road conditions are unsafe.10 Other risk factors of injury among two-wheeled delivery drivers include insufficient infrastructure for safe cycling, the use of damaged equipment and the lack of adequate personal protective equipment (PPE) (e.g., helmets, lights, protective clothing). In all of these scenarios, platform companies are not legally obligated to help workers recoup expenses related to medical treatment, lost wages, or stolen and damaged equipment given their ability to (mis)classify workers as independent contractors rather than employees.1,29
The mechanisms leading to assault among food delivery workers may be similar to those leading to injury but are not as well studied. There is evidence that bike theft is a common catalyst of assault in addition to reports of mistreatment and harassment from employees and customers at pick-up and drop-off delivery locations.10,26 Our finding that limited English-language ability was positively associated with assault, but not injury, suggests that discrimination among foreign-born workers should be considered as a potential risk factor of assault in future studies.
Strengths and limitations
This study used existing data from the NYC-DCWP survey. The collection of data from a population of workers who are otherwise invisible to most occupational safety and health data surveillance systems is a strength of the parent study on which our analyses are based. Also, while our sample comes from one segment of the gig economy in NYC, our findings likely apply to cities with similar dense urban cores and more broadly to workers under similar “gig” working conditions.
Despite the strengths of the NYC-DCWP survey, its design posed several important limitations to our study. Our analysis was based on a convenience sample and may not be representative of the underlying population of food delivery gig workers in NYC, which has yet to be fully characterized. Given the cross-sectional design, former workers were not asked questions related to job dependence and, therefore, were excluded from our sample, which could lead to an under- or over-estimate of the association between dependence and injury and assault. Since the NYC-DCWP survey only included platform workers, we are unable to distinguish occupational risks that are inherent in food delivery work from those that are associated with the platformization of the food delivery industry. We used the items related to “main job” as a marker of dependence level, though dependence is a more complex construct that deserves further research. The term "main job" was not defined in the survey question and may have been interpreted in different ways. Relatedly, the survey did not include data on other key drivers of dependence (e.g., household composition, earnings and resources, spousal employment, and education). Future research will need to further develop the theory and drivers of dependence and its relationship to health. Finally, our results may not be generalizable to food delivery gig workers in urban settings with less e-bike and moped density than in NYC; however, we did find that, among car riders only, dependence remained a significant risk factor of injury.
Recommendations and future research
Our findings provide evidence in support of labor protection laws and policies designed to address work conditions, urban transportation issues (e.g., physical infrastructure and traffic enforcement), and platform company-specific practices related to the (mis)classification of independent contractors and the design of algorithms. Immediate efforts to improve worker safety can draw from the core elements of safety management programs (e.g., management commitment, worker participation, hazard identification and assessment, hazard prevention and control, education and training, program evaluation) described by the U.S. Department of Labor Occupational Safety and Health Administration30 and food delivery safety guidelines by local authorities in the U.S.31 and in Australia.32 At a minimum, platform-based companies should be required to ensure that workers have adequate PPE and report work-related injuries, illnesses, and fatalities. Improved surveillance systems can be used to both monitor the rates of injury and assault among platform-based workers and assess the efficacy of policies, such as the recent NYC law governing minimum pay, delivery distances, and other aspects of workers’ safety.33 Future research should include longitudinal and alternative study designs to better understand the work-related mechanisms on the pathway between dependence and occupational injury, identify circumstances leading to assault, and describe downstream physical, mental, and economic consequences associated with occupational hazards among platform-based workers.