This review assessed evidence from randomised controlled trials (RCTs) and controlled before-after studies (CBAs), on whether 'mobility management' can be used to prevent, reduce, or delay car driving in teenagers aged 15 to 19 years.
Despite safety-based educational programmes and programmes designed to restrict teenage driving (called graduated drivers licensing, or GDL), rates of injury and death caused by teenage car drivers remain high. Car use also leads to health risks caused by being less active physically, and environmental risks from burning fossil fuels.
'Mobility management' is the term applied to strategies that encourage people of different ages to drive less. Examples include encouraging walking or cycling, use of public transport, or car sharing. People are more likely to make changes to how they live their lives in the period between the ages of 15 and 19 years. This may be the ideal time to introduce mobility management to prevent, reduce, or delay driving. This may influence long-term car use, and reduce risks associated with driving.
We searched for RCTs (clinical studies where people are randomly put into one of two or more treatment groups), and CBAs (CBAs are not randomised, but measure outcomes before and after an intervention is applied), published up to 16 August 2019.
We found one small RCT. This study gave 178 university students (average age 18 years) who did not hold a driving licence, different types of information to educate them about the negative aspects of car use. Students received information about cost, risk, or stress related to car use, or a combination of all three, while a fifth group of students received no information.
We also found one CBA with 860 participants. In this study, 17- to 18-year-old school students taking a driving theory course received an additional lesson about active transport, such as walking or cycling. Some were also invited to join a relevant Facebook group. Students in a third group received no additional information. Students were asked questions about their intention to use active transport after getting a driving licence, but many did not complete the study.
The RCT reported that fewer students obtained a driving licence 18 months after they were given negative car-related information than students given no information.
The CBA reported a decreased intention to use active transport immediately after the lesson in students invited to the Facebook group only, but eight weeks later this was the only group where intention to use active transport had increased.
We judged the quality of the evidence in both studies as very low. The studies did not report frequency of driving, driving distance, driving hours, use of alternative modes of transport, or car crashes.
Quality of the evidence
We found only two small studies investigating different types of mobility management, so could not compare their findings with other similar experiments to see if they produced the same results. Similarly, we could not be certain whether these programmes would be effective in real-world settings. Some of the methods in the RCT were not well reported, and we expected that students might share information they were given with other study participants, which might have affected the study findings.
CBA study designs have higher risks of bias because participants are not randomised to groups, and, in this CBA, the characteristics of students varied between groups. There was also a very high number of participants who did not complete the questionnaires at each time point.
Consequently, we judged the evidence from these studies to be of very-low quality.
We found only two small studies that investigated ways to present mobility management information to teenagers. We could not be sure of their effect on obtaining a driving licence, or intention to use active transport (walking or cycling) after obtaining a driving licence, because the quality of the evidence was very low. Risks from driving are a public health concern, and, because we did not find enough evidence, large studies with a long follow-up (monitoring) period are needed to find out whether teenagers can be encouraged to reduce or delay driving.
We found only two small studies, and could not determine whether mobility management interventions were effective to prevent, reduce, or delay car driving in teenagers. The lack of evidence in this review raises two points. First, more foundational research is needed to discover how and why young people make decisions surrounding their personal transport, in order to find out what might encourage them to delay licensing and driving. Second, we need longitudinal studies with a robust study design – such as RCTs – and with large sample sizes that incorporate different socioeconomic groups in order to evaluate the feasibility and effectiveness of relevant interventions. Ideally, evaluations will include an assessment of how attitudes and beliefs evolve in teenagers during these transition years, and the potential effect of these on the design of a mobility management intervention for this age group.
Rates of injury and death caused by car crashes with teenage drivers remain high in most high-income countries. In addition to injury and death, car use includes other non-traffic risks; these may be health-related, such as physical inactivity or respiratory disease caused by air pollution, or have global significance, such as the environmental impact of car use. Research demonstrates that reducing the amount of time driving reduces the risk of injury, and it is expected that it would also reduce other risks that are unrelated to traffic. Mobility management interventions aim to increase mobility awareness and encourage a shift from private car use to active (walking, cycling, skateboarding), and public (bus, tram, train), transportation. 'Soft' mobility management interventions include the application of strategies and policies to reduce travel demand and may be instigated locally or more widely, to target a specific or a non-specific population group; 'hard' mobility management interventions include changes to the built environment or transport infrastructure and are not the focus of this review.
Between the ages of 15 to 19 years, young people enter a development stage known as the 'transition teens' in which they are likely to make long-lasting lifestyle changes. It is possible that using this specific time point to introduce mobility management interventions may influence a person's long-term mobility behaviour.
To assess whether 'soft' mobility management interventions prevent, reduce, or delay car driving in teenagers aged 15 to 19 years, and to assess whether these mobility management interventions also reduce crashes caused by teenage drivers.
We searched the Cochrane Injuries Group Specialised Register, CENTRAL, MEDLINE, Embase, Web of Science, and Social Policy and Practice on 16 August 2019. We searched clinical trials registers, relevant conference proceedings, and online media sources of transport organisations, and conducted backward- and forward-citation searching of relevant articles.
We included randomised controlled trials (RCTs) or controlled before-after studies (CBAs) evaluating mobility management interventions in teenagers aged 15 to 19 years. We included informational, educational, or behavioural interventions that aimed to prevent, reduce, or delay car driving in this age group, and we compared these interventions with no intervention or with standard practice. We excluded studies that evaluated graduated drivers licensing (GDL) programmes, separate components of GDL, or interventions that act in conjunction with, or as an extension of, GDL. Such programmes aim to increase driving experience and skills through stages of supervised and unsupervised exposure, but assume that all participants will drive; they do not attempt to encourage people to drive less in the long term or promote alternatives to driving. We also excluded studies which evaluated school-based safe-driving initiatives.
Two review authors independently assessed studies for inclusion, extracted data, and assessed risks of bias. We assessed the certainty of evidence with GRADE.
We included one RCT with 178 participants and one CBA with 860 participants. The RCT allocated university students, with a mean age of 18 years, who had not yet acquired a driving licence, to one of four interventions that provided educational information about negative aspects of car use, or to a fifth group in which no information was given. Types of educational information about car use related to cost, risk, or stress, or all three types of educational information combined.
In the CBA, 860 school students, aged 17 to 18 years taking a driving theory course, had an additional interactive lesson about active transport (walking or cycling), and some were invited to join a relevant Facebook group with posts targeting awareness and habit.
We did not conduct meta-analyses because we had insufficient studies.
We could not be certain whether educational interventions versus no information affected people's decision to obtain a driving licence 18 months after receiving the intervention (risk ratio 0.62, 95% confidence interval 0.45 to 0.85; very low-certainty evidence). We noted that fewer participants who were given information obtained a driving licence (42.6%) compared to those who did not receive information (69%), but we had very little confidence in the effect estimate; the study had high or unclear risks of bias and the evidence was from one small study and was therefore imprecise.
We could not be certain whether interventions about active transport, given during a driving theory course, could influence behavioural predictors of car use. Study authors noted:
- an increased intention to use active transport after obtaining a driving licence between postintervention and an eight-week follow-up in students who were given an active transport lesson and a Facebook invitation compared to those given only the active transport lesson; and
- a decrease in intention between pre- and postintervention in those given an active transport lesson and Facebook invitation compared to those given the active transport lesson only.
There were high risks of bias in this CBA study design, a large amount of missing data (very few participants accepted the Facebook invitation), and data came from a single study only, so we judged the evidence to be of very low certainty.
These studies did not measure our primary outcome (driving frequency), or other secondary outcomes (driving distance, driving hours, use of alternative modes of transport, or car crashes).