There is currently widespread policy concern about the low level of physical activity and the increasing incidence of obesity and chronic disease in the UK. The promotion of walking and cycling is a promising way to increase physical activity across the population by integrating it into daily life. These public health goals are consistent with cross-sectoral interest in the wider benefits of a shift away from car travel to non-motorised transport.

The recent systematic review by Ogilvie and colleagues 1 2 has provided a comprehensive summary of the evidence of effectiveness of interventions intended to promote walking and cycling (active transport) as an alternative to travel by car. They found relatively few interventions rigorously evaluated for their impact with children, young people or parents. The interventions included those in the following categories:

1. Behaviour-change programmes, eg social marketing, where messages about active transport are targeted at individuals.

2. Publicity campaigns, eg mass-media advertising.

3. Infrastructural or engineering measures, eg traffic calming and cycle routes.

Ogilvie and colleagues1 concluded that there is some evidence that behaviour-change interventions can bring about a shift to walking and cycling among motivated groups; the evidence regarding publicity campaigns and engineering measures is inconclusive.

However, looking at effectiveness alone is limited. Other study designs, especially studies of people’s views, are valuable sources of data to inform public health interventions. Such “qualitative” research is often capable of opening up the so-called black box of quantitative trials and epidemiology, to expose why interventions work or do not work.3 In particular, it can help to identify and develop promising interventions.4^–6

This study reports a systematic review of studies of the public’s views of walking and cycling as modes of transport, which was commissioned by the Department of Health for England.

METHODS

Research questions and user involvement

For systematic reviews to be relevant to policy and practice, the involvement of potential users in the review process is of value.7 8 User involvement in this review was sought through an advisory group, including policy specialists, practitioners and researchers, representing a number of organisations.

The Advisory Group’s remit was to advise the research team on the scope of the in-depth review and to refine the research questions, including the relevant populations and aspects of walking and cycling. At its first meeting, the Advisory Group was presented with a summary of the research conducted in the UK since 1995. This research “map” was designed to answer the initial question:

• What research has been undertaken about the public’s views of walking and cycling as modes of transport?

A consensus emerged in the Advisory Group that there was a need to understand what works for children, young people and parents. Thus, the final (in-depth) review question was:

• Do children’s, young people’s and parents’ views illuminate the findings about effectiveness of interventions to promote their walking and cycling?

Group members were also asked to identify potentially relevant research, to comment on draft documents, and to help to produce recommendations for policy and practice.

Searching

Studies were located using searches on seven major databases (CSA, IBSS, PubMed, Geobase, HMIC, HELMIS, and Transport), two databases including grey literature (Dissertation Abstracts, the British Library catalogue), and 31 websites containing databases of relevant material. Searches combined terms relating to views studies, and terms relating to health promotion and transport. A list of all sources searched can be found in the full report at http://eppi.ioe.ac.uk/cms/Default.aspx?tabid = 942; full search strategies are available from the authors on request. In addition, two conference proceedings (European Transport Conference, Velo-City) and two journals (Transport Policy, Transport Reviews) were hand-searched for the years 2000–5. The Advisory Group were asked to help identify relevant research, and contacted experts in the field. Finally, the reference lists of included studies were scanned for relevant references, and citation indexes (SCI/SSCI) were searched for authors of included studies.

Inclusion criteria

Studies were included in the descriptive map presented to the Advisory Group if they:

• were conducted in the UK;

• were published between 1995 and 2005; and

• accessed people’s ideas about, or experiences of: walking and cycling initiatives; what influences whether they walk or cycle; and their ideas about what could be done to promote walking and cycling.

All study designs were eligible for inclusion, as long as they investigated people’s views. For the in-depth review focussed on children and young people, the definition of young people was up to age 21.

Description of studies

Two reviewers independently applied framework analysis to code each included study, and then met to compare their analyses and arrive at a consensus. The methods and quality of each study were also coded by two reviewers independently employing a standard data extraction tool.

Focus of studies

Framework analysis9 was applied to all the studies included in the map. Hypotheses were derived from the background literature about the factors influencing people’s transport choices, and their views regarding these factors. Background literature was also drawn on for information about social determinants of health,10 tackling inequalities11 and theoretical models in health promotion.12 13 Together, these suggest that the most promising interventions will be multifaceted, and target determinants at the following interrelated levels: the individual (eg knowledge, attitudes, self-esteem), the community (eg social support networks, family relationships), and the wider society (eg social class, access to resources and services). The levels of influence and hypotheses together provided an initial framework for coding the studies. In addition, new topics (eg theft) were incorporated as they emerged from the data. The resulting topics included both known advantages and disadvantages of walking and cycling (eg health, environmental benefits, costs, convenience) and variables which may influence the relevance of these factors (eg journey type, age, gender). The number of studies in which people identified each factor as potentially or actually influencing their behaviour was recorded.

Methods of studies

Studies were described in terms of their aim and rationale, design, sampling strategy, recruitment and consent procedures, description of the sample, data-collection methods, data-analysis methods, and methods of the study, and their appropriateness for eliciting the views of children, young people and parents.

Quality assessment of studies

All studies reviewed in depth were assessed using ten criteria addressing four main quality issues.14 Two relate to the quality and reporting of the study’s sampling methods (identification of the sampling frame, and of the methods used to obtain the sample). Four relate to the quality of the description of the sample in the study (the total number of participants, their ages, sex, and socioeconomic status). Two relate to the reliability and validity of the tools used to collect data. Finally, one criterion relates to whether studies used appropriate methods for helping people to express their views, and another the relevance of the study to the review question. Each study scored 0, 0.5 or 1 point on each criterion; only studies with a total of 7 points or more were included in the synthesis of findings.

Synthesis of views evidence

The framework analysis was taken further to review in depth the findings of studies addressing the views of children, young people and parents. Tables grouped factors according to whether they might influence behaviour individually, or through families, the community, wider society or the environment. Factors of influence within these levels were noted, as were demographic data; who expressed the view (child, young person or parent); children’s age and sex; the socioeconomic status of the area; urban, suburban or rural location; and whether the view expressed was about walking or cycling. Findings relating to these factors and circumstances were summarised and collated from all the contributing studies to provide as complete a picture as possible. The summaries of findings were drawn on to describe the range and nature of views, and find associations and discrepancies between these views with a view to providing explanations for the findings. Explanations were presented as overarching themes, from which implications for interventions were developed.

Comparison with effectiveness evidence

The findings of the present views synthesis and those of the synthesis of evaluations of interventions in the review by Ogilvie and colleagues1 were compared. The implications for interventions which had been developed from the views synthesis were tabulated against the characteristics of evaluated interventions.6 This enabled identification of which influences on walking and cycling from the views studies have been addressed by evaluated interventions. It also enabled investigation of whether more effective interventions have addressed factors identified by the views studies, and which recommendations for intervention development derived from the views studies have yet to be addressed by interventions evaluated in the outcome studies.

RESULTS

Included studies

The present searches identified a total of 6484 unique records. Following screening of these records, 106 reports of 97 UK studies of people’s views were systematically identified, retrieved and described. Of these, 34 studied the views of children, young people and parents, and were assessed for their quality (see Quality assessment of studies above). The 16 studies judged to be higher quality were reviewed in depth.15^–30

These studies covered a range of topic areas and included children (up to age 10, n  =  8), young people (aged 11–21, n  =  10) and parents (n  =  5). A diverse range of methods was used to elicit participants’ views, including questionnaires (n  =  12), discussion or focus groups (n  =  9), and individual interviews (n  =  5). Six focussed on walking, and one on cycling; the remaining nine studies elicited views on transport in general. Three studies were process evaluations of UK-based interventions designed to increase walking and/or cycling. In total, studies included at least 8776 respondents. A summary of the results is presented here; full analysis can be found in the technical report (http://eppi.ioe.ac.uk/cms/Default.aspx?tabid = 942).

Themes: factors of influence

A list of themes found in the views studies, categorised by type and level of influence, is presented in tables 1–4. For reasons of space, tables 1–4 present only themes found in more than one study.

Data were also extracted on the principal findings for each theme. Tables 5–6 provide summary examples of the findings relating to barriers and facilitators encountered at one of the levels (wider society and environment).

Synthesis of views

Five overarching explanations of transport choices were derived from the themes identified through framework analysis.

1. Culture of car use

This heading subsumes the perception of cars as more convenient than walking and cycling, the status of car ownership and the perception of cars as “cool”, and the view of car ownership and use as integral to a normal adult lifestyle, which militate against the use of active transport. Whilst the impact of this culture is strongest on parents and older young people, it is already manifest in the views of young children.

2. Fear and dislike of local environments

Fear of the local environment includes barriers present in the social environment such as those relating to safety, traffic, the need for appropriate facilities and concerns about theft and other crime, as well as negative features of the natural environment. This fear was expressed by both children and parents and can take many forms, from immediate practical worries to a diffuse perception of walking and cycling as involving exposure to hostile surroundings.

3. Children as responsible transport users

Children expressed clear and definite views, often diverging from the views of their parents, relating to their transport preferences, the environmental impact of different transport modes, and the benefits of walking and cycling in terms of safe accompanied travel, sociability, and health and fitness.

4. Parental responsibility and behaviour

Both children and parents mentioned the pressure on parents to fulfil expectations about being a “good parent” by ensuring the safety of their children. Children viewed their parents’ car use as part of a pattern of behaviour whereby the independence of the children is limited in the name of safety.

5. Contextual factors

The views of children, young people and parents differ depending on, among other factors, age, sex and location (urban, suburban or rural). Any interventions designed to increase walking and cycling may have to be tailored according to these factors.

Implications for policy, research and development

The final stage of synthesis was to draw out implications for interventions from the findings of the views studies, utilising the broad topics identified in the thematic synthesis. These implications were then compared with the findings of Ogilvie and colleagues,1 both the interventions identified as soundly evaluated and effective, and the interventions which may be promising but have not been soundly evaluated. The goal was to identify effective interventions that are appropriate for UK populations, appropriate interventions needing more rigorous evaluation, and to suggest other interventions based on people’s views.

1. Culture of car use

The views studies highlighted the culture of car use as an important influence, implying that educational or marketing-based interventions may gain from more explicitly addressing the perception of car travel as normal. The present findings suggest that such interventions should be targeted towards younger children, as these preferences are already well established by early adolescence.18–20 26 30

No interventions sought directly to change children’s or parents’ preference for car use, or perceptions of the high status of car ownership and use. However, some behaviour-change interventions31^–33 addressed perceptions of car travel as more convenient than walking or cycling; these interventions were successful with motivated population subgroups.

Parents’ and older children’s emphasis on the convenience of car travel may relate to cultural or preference-related influences as much as to structural factors. In particular, the perception of walking and cycling as less convenient may indirectly reflect an association of these modes with low social status. Nonetheless, behaviour-change programmes are likely to be more successful if they are undertaken in conjunction with structural and environmental interventions to make walking and cycling more convenient.

2. Fear and dislike of local environments

As well as barriers arising from the natural environment such as weather and hills, which were mentioned in several studies,15 20 26 negative perceptions of the social environment were widely reported. Parents cited risks from traffic and “stranger danger” as a reason to drive children from place to place, particularly to school.16 18 23 Implications for policy suggest a need to improve the environment, and to encourage accurate perceptions of risk.

Interventions trying to change accident rates by engineering work (ie, changes to the physical environment) produced ambiguous results. Two studies conducted in England34 35 reported reductions, and a third 36 reported an increase in accidents, although these changes were not shown to be statistically significant. The need for appropriate facilities for walking and cycling was addressed in a controlled evaluation in Delft, in the Netherlands, in which an improved and extended cycle network was installed.37 A positive shift toward cycling was observed, but was not shown to be statistically significant.

Only one intervention study directly measured perceptions of risk. The study of school travel plans in Camden and Islington38 assessed parental fears of traffic danger, abduction danger and bullying after one year. These fears were not significantly reduced in the intervention group compared with the control group. None of these interventions resulted in a significant increase in walking or cycling.

Addressing perceptions of safety in designing new interventions is likely to be challenging for two reasons. First, as noted above, environmental interventions to improve the safety of walking and cycling are often ambiguous in effect. No interventions which work by improving safety have been shown to be effective in promoting walking and cycling. Second, perceptions of safety do not appear to be highly correlated with measured rates of accidents or crime. For example, Department for Transport statistics on child deaths and serious injuries in Great Britain indicate that rates for child pedestrians and cyclists fell by 49 and 54%, respectively, between 1994–98 and 2005;39 whilst cycle mileage by children decreased by 18% over the same period, pedestrian mileage increased by 5%.40 Although these figures indicate that both modes have become substantially safer in recent years, the present data suggest that parents are still concerned about the safety of walking and cycling.

These reservations aside, this analysis provides several prompts for future intervention research. Researchers should continue to evaluate the safety effects of infrastructural interventions, and their effects on attitudes to the local environment. Behaviour-change and publicity campaigns should emphasise that walking and cycling are not intrinsically dangerous activities, and promote perceptions of local environments as safe where this is appropriate. Concerns about safety from accidents particularly need to be addressed amongst younger children, and those about safety from personal attack amongst older girls and in urban locations. Interventions could also attempt to address perceptions of natural factors, such as weather, as barriers to walking and cycling.

3. Children as responsible transport users

The findings indicate that children and young people may be particularly receptive to messages emphasising the environmental benefits of walking and cycling, their role in promoting health and fitness, and the opportunities they offer for social contact with friends and family.15 19–21 26 27 26 27As children’s (particularly younger children’s) perceptions of walking and cycling are substantially more positive than those of adults, adapting existing interventions to incorporate children’s own perspectives is a promising direction.

None of the evaluated interventions was specifically tailored to children’s and young people’s perceptions; only one uncontrolled study41 evaluated an intervention which emphasised environmental benefits and was implemented in schools as well as community settings.

4. Parental responsibility and behaviour

The present findings indicate that parents perceive a trade-off between ensuring children’s safety and fostering their independent mobility.15 18 30 Intervention design needs to take this perception into account, and, where appropriate, to address social norms that discourage parents from allowing their children to walk and cycle. It may also be of value to encourage children to engage critically with these norms.

As noted above, the relation between perceived safety and objectively measured risk is unclear. Parents who express concerns about the safety of active transport may believe that car travel is less dangerous than walking or cycling, or may wish to accompany their children to minimise the risk of traffic accidents, abduction or attack. If the latter, these concerns might be addressed by accompanying their children on foot or by bicycle – or having another trusted adult accompany them – and not implying that the car must be used. This example suggests that parents’ concerns about the safety of active transport may reflect social pressures to be a “good parent” by adopting cultural norms defining safety, as much as assessments of the risks associated with different transport modes. Children and young people themselves frequently query the restrictions that are placed on their mobility in the name of safety.19 23 Many parental concerns are well founded, and ensuring that walking and cycling are safe should remain a priority. However, addressing these concerns also requires attention to expectations and norms about parenting.

No intervention studies directly addressed such expectations and norms. The only intervention to encourage children to think critically about car use in their own families was the programme in Adelaide,42 which encouraged children participating in the school-based part of the intervention to relay the lessons learnt to other members of the household.

DISCUSSION

The synthesis of views from qualitative research reveals a culture of car use, fed by a fear and dislike of local environments and parental responses that emphasises children’s safety at the expense of developing their independence, despite children expressing responsible attitudes towards transport choices.

Strengths and limitations of the review

Previously, syntheses of the public’s views and effectiveness evidence prepared simultaneously have been integrated in a cross-study-type synthesis.6 This is the first time that a views synthesis has been integrated with an effectiveness synthesis prepared and published by another research team. As such, it provides a novel way of extending the existing evidence base.

However, the scope of Ogilvie and colleagues’ review of interventions1 was not identical to that of the present review. First, while the review of interventions benefited from including international literature, the present review of views studies included UK-based studies only. Second, the review of interventions included only population-wide strategies reporting a shift away from car transport towards walking and cycling. Third, our cross-study-type synthesis included only interventions evaluated in controlled trials, whilst Ogilvie and colleagues’ review imposed no such limitation.

The use of summary scores for critical appraisal in research syntheses is not uncontroversial.43 The present quality assessment process resulted in the exclusion of 18 lower-quality studies from the review. The choice of a different cut-off point for defining quality would have resulted in a different mix of studies. In particular, most of the excluded studies focussed on cycling.

CONCLUSIONS

In this review, a range of views was examined concerned with factors influencing walking and cycling among children, young people and parents. There were four themes that stood out clearly: a strong culture of car use, fear and dislike of local environments, children as responsible transport users, and parental responsibility for children.

These findings indicate strong public support for structural measures such as cycle lanes and traffic calming, although further intervention research is needed to establish the effectiveness of such measures. Communicative interventions such as behaviour-change programmes and publicity campaigns, for which there is some evidence of effectiveness, may benefit from taking the public’s views into account. In particular, perceptions of walking and cycling as dangerous activities are an important barrier to the promotion of active transport. Whilst research and policy should seek to promote the safety of pedestrians and cyclists wherever possible, the present findings suggest that over-emphasising safety may discourage walking and cycling by focussing attention on the risks involved, at the expense of the potential benefits of an increase in physical activity. Parents’ concerns about safety, and their perception of a conflict between promoting their children’s activity and independent mobility and protecting them from harm, are well founded. Nonetheless, policy-makers should be careful not to exacerbate this conflict, and seek to help parents to achieve a better understanding of the risks involved in not allowing their children to walk or cycle to school.