From cars to bikes – the feasibility and effect of using e-bikes, longtail bikes and traditional bikes for transportation among parents of children attending kindergarten: design of a randomized cross-over trial

Physical activity levels and active transportation

It is well documented that regular physical activity (PA) implies reduced risk for overweight and chronic diseases [1]. Still, one third of adults and four fifths of adolescents globally do not reach PA guidelines [2], and recent European data showed that only 28% of the adult population comply with the PA recommendations [3]. Nonetheless, only ten additional minutes of PA daily would make two thirds of inactive persons adhere to current PA guidelines [4].

For our ancestors, food procurement was inextricably linked to PA and energy expenditure [5], as they needed to hunt and forage in order to get food. Today, this link is broken- we can drive our car to the grocery shop and buy the foods we need with minor energy expenditure. In turn, these changes imply that being physically active requires conscious choices to a larger degree. Physical inactivity is estimated to cause approximately 6–10% of the non-communicable diseases of coronary heart disease, type II diabetes, breast- and colon cancer, and 9% of premature deaths worldwide, i.e. similar health effects as the established risk factors of obesity and smoking [6]. Car use and other forms of motorized transportation favour neither health nor environmental sustainability, as it entails sedentariness and emissions of greenhouse gases (GHGs).

Active transportation, like walking or cycling for transportation purposes, may be a feasible and time efficient way to incorporate PA into daily routines, potentially increasing PA levels [7, 8]. Thus, active transportation could promote health [7, 9–15], prevent obesity [16, 17], and decrease future healthcare costs [18]. It has been calculated that increased PA would translate into significant health gains, entailing major socioeconomic gains for the society [19]. For bicycling specifically, a tripling in cycling in five urban areas in Norway between 2006 and 2009 has been estimated to imply health benefits of 250 million NOK annually [20].

Next to direct effects on fitness and health, an additional advantage of active transportation is the potential to decrease GHG emissions [21, 22], as well as noise and pollution [7]. Currently, motorized transportation is responsible for about 23% of global climate gas discharges [7]. The ambitious goal of the Paris Agreement, entailing carbon neutrality before the end of the century [23], demands initiatives to be generated within all areas of society, not the least within the transportation sector. It is likely that an increased share of travels in Norway could be conducted as active transportation, considering that 25% of daily travels done by car are shorter than 2.5 km [20], and average distance of bicycle trips is 5.1 km [24]. Accordingly, 20% of all travels in the UK are shorter than one mile [25].

Parents as facilitators of physical activity

Lifestyle behaviors in childhood constitute the foundation for health throughout the lifespan, and research suggests that lifestyle habits, such as PA, track from childhood into adulthood [26, 27]. Likewise, overweight and obese children are more likely to become overweight and obese adults, than normal weight children [28, 29]. Parents are important facilitators and role models of PA for their kids, and among the significant correlates are parental PA, as well as parent participation in child PA [30]. Being transported to kindergarten by bicycle instead of by car could teach children that alternative modes of transport exist, hence representing early adaptation to healthy and sustainable transportation and PA habits. Since parental PA behaviors are crucial for their own and their children’s current and future health, parents of toddlers is a target group of utmost importance. In terms of parental own PA habits, lack of time [31] and stress [32] are repeatedly documented to be negatively associated with PA in adults. In this regard, active transportation could potentially decrease the impact of time scarcity as a barrier, and may also reduce overall perceived stress through incorporating PA into daily transportation, implying less need for additional time-consuming exercise.

E-bikes and longtails

Electric assisted bicycles (e-bikes) represent an unexploited potential in terms of increased bicycle use. If replacing other motorized modes, and not replacing other PAs, e-bikes could favor both public health and the environment through increased levels of PA and decreased emissions of climate gases [22, 33]. Empirical evidence indicates that mode substitution depends on local context, culture and available transport alternatives, entailing that a larger proportion of car trips is replaced in cardominated countries such as Australia, US and Canada, than in countries with a bicycle culture, such as Denmark and the Netherlands [34]. For illustration, recent Dutch data showed that e-bike ownership reduced car and public transport use, although usage of the traditional bike decreased even more [35]. Nonetheless, e-bike owners reduced their car and public transport use more than traditional bicycle owners [35]. Moreover, in a convenience sample of Norwegian car owners, it was found that those who cycled the least were most interested in buying an e-bike, which in turn could result in mode shifts from cars to bikes [36].

In e-bikes categorized as pedelecs, propulsion is caused partly from the pedal- power of the rider, and partly from an electric engine supplementing power up to 25 km/h, or a maximum power output of 250 watt [34]. It has been claimed that energy efficiency of an e-bike is greater than that of any other mode of transport, except from traditional bicycles [37]. Compared with regular bicycles, e-bikes enable maintenance of speed with less effort, which in turn helps overcoming some of the most common barriers to traditional pedal cycling, such as lack of fitness needed to cycle, hilly terrain, longer distances, lack of time, and lack of end of trip facilities (e.g. change rooms and shower) [34]. Current knowledge suggests that e-bike users cycle more often, and to more distant locations than those using regular bicycles [34, 37, 38], hence possessing greater capacity for exchanging car use than regular bicycles.

A major limitation with traditional bicycles, and also standard e-bikes, is the carrying capacity [39]. At present, there are several different cargo bikes on the market, both human powered and with electric assistance, as well as various bike trailers for carrying goods and/or children. However, carrying stuff on a trailer may be less convenient than directly on a bike. In this regard, so-called longtail bikes possess a great potential, being constructed for carrying an adult, two children and additional goods. Such longtails could possibly promote health through increased cycling, and reduce anthropogenic CO₂ emissions related to motorized transportation, while simultaneously meeting a practical need not sufficiently accomplished by a traditional bike or e-bike. Considering that a great share of car travels are done within a limited range [20, 25], longtail bikes might represent a feasible mode of transportation, yet current scientific evidence is scarce. Still yet, our research group has tested human powered longtail bikes in different families for periods lasting up to five years, elucidating the potential of longtails for various trip purposes, and for all seasons and weather conditions. In line with this, American data support possible mode substitution and a decline in car travel among cargo bike owners [40]. Giving specific attention to females with children, representing a minority group in the bicycling community, Schwartz and Riggs [41] newly suggested that under certain circumstances and subject to a certain culture, the cargo bike has emerged as an option for women to choose.

Health effects of e-bicycling and use of longtails/transport bikes

It is repeatedly found that both active [42] and inactive [43, 44] subjects reach PA levels sufficiently high to meet the moderate-to-vigorous-intensity standard (3–9 metabolic equivalents (METs)) of the physical activity guidelines [1, 45], when e-biking. Although e-bikes seem to entail lower intensity than traditional bikes [46], e-bikes could still boost overall levels of PA [34] and thereby promote health, if combined with more frequent and longer trips as proposed [37, 38]. Current evidence regarding health effects of commuting with an e-bike is scarce. De Geus and colleagues [47] conducted a quasi-experimental study in twenty untrained men and women, who were provided with an e-bike for six weeks. No change in maximum oxygen uptake (VO₂ max) was found, yet a significant gain in maximal power output was achieved after six weeks of e-biking [47]. A recent Norwegian pilot study equipped 25 inactive adults with an e-bike for eight months, measuring participants′ VO₂ max at baseline and at intervention determination [48]. Results showed an average 8% improvement in VO₂ max, and cycling distance was positively associated with the increase, yet no control group was included. Focus group interviews were conducted and analyzed as well, revealing that e-biking contributed to highly positive experiences regarding active commuting (unpublished results). Moreover, a recent American study in twenty sedentary commuters reported four weeks of pedelec commuting to result in significant improvements in 2-h post-plasma glucose levels, VO₂ max and maximal power output [49]. In addition, no compensatory changes in overall PA levels were observed, hence pedelec commuting helped the participants to meet PA recommendations [49]. In terms of potential health effects of using human powered longtail bikes or other cargo bikes, no previous studies have addressed these associations. It seems reasonable to assume though [46], that the intensity will be higher than for traditional bikes due to the weight of the cargo, entailing additional health effects if used as frequently as traditional bikes.

Accessibility, social support and intrinsic motivation

When developing programs aiming to promote certain behaviors, such as bicycling with e-bikes and longtail bikes, relevant determinants for the behaviors of interest should be known. Existing literature and behavioral theories provide relevant determinants at several levels, that is, the personal (e.g. motivation, intention, knowledge, attitude), interpersonal (social support, modelling) and/or environmental (availability, accessibility, policy) level. The socio-ecological framework suggested by Sallis et al. [50] describes most of these determinants, especially the environmental determinants, while theories such as the Social cognitive theory [51] and the Self-determination theory [52] describe the more interpersonal and cognitive processes to increase goals and motivation for various behaviors, like PA. The Self-determination theory focuses on diverse types of motivation, and argues that intrinsic motivation should be strived for, in order to result in sustained behavior [52]. According to the socio-ecological framework [50], accessibility is one important environmental determinant for PA, including active transportation. Supporting this, a recent British study indicated that when made accessible, e-bikes could facilitate active travel and have substantial effects on travel behavior, also in subjects traditionally undertaking less PA or feeling unable to use a conventional bike [53]. In total 80 employees were loaned an e-bike from two major employers in Brighton, during a six- to eight-week trial period. Across all participants 75% chose to use the e-bike at least once a week, car mileage was reduced by 20%, and 59% of the employees reported that their overall PA increased. At the end of the trial, 73% of the employees said they would cycle to work at least once a week if they had an e-bike available [53]. Bike share programs may be considered another aspect of accessibility, and a number of cities have now introduced e-bike shares, potentially encouraging new users to bike share [54]. Underpinning this, a pilot study trialing a university based e-bike share in North-America reported that new users were attracted to cycling [55]. Multi-city analyses of regular bike shares’ impact on car use and PA suggest that car use decreases, yet of limited magnitude [56]. Nonetheless, PA levels could increase [57] due to mode shifts, likely resulting in overall positive health effects [58]. Another important behavioral interpersonal determinant described in the socio-ecological framework is social support, and the workplace environment, entailing both the social and the physical environment, could facilitate active transportation [50]. A previous study by Wen and colleagues [59] assessing the role of workplaces in promoting active commuting, reported a significant inverse association between employees’ perception of workplace encouragement for active travel and driving to work. Also, physical support at work, such as available bike parking and presence of showers, as well as cultural and social support for active transportation, has shown to be relevant for female employees’ transport choices [60]. Accordingly, Yang et al. [61] found that worksite support and policies tended to associate with active commuting and the use of public transit. Grounded in this, together with previously reported higher income and education among e-bike owners [62], it is reasonable to assume that initiatives providing bike accessibility could increase cycling. In line with Self-determination theory, we hypothesize that increased accessibility and social support could facilitate intrinsic motivation for biking [52] through meeting the basic psychological needs, i.e. feelings of autonomy, competence and relatedness [63], which in turn could result in higher levels of bicycling.

Objectives

We aim to assess the effect of an intervention where participants will be provided access to an e-bike (including a trailer), a longtail bike and a traditional bike (including a trailer), each bike for three months, on the following aspects:

Study design

The present study will have a cross-over design, entailing that participants in the intervention group (n = 18) will complete the following intervention arms in random order: (i) three months access to an e-bike with trailer (n = 6), (ii) three months access to a longtail bike (n = 6), and (iii) three months access to a traditional bike with trailer (n = 6), in total nine months. Also, a control group (n = 18) maintaining usual transportation and PA habits will be included. Randomization of participants into intervention or control group will be stratified according to sex and cardiorespiratory fitness. As incentive for participants randomized into the control group, those who fulfill the study will be in the draw of one traditional bicycle, including a trailer. To reduce the risk of accidents and injuries, bike helmets (parent and child), a reflex vest, lights and winter tyres with spikes will be handed out. The present study was approved by The Norwegian Social Science Data Services (NSD), and all participants will provide informed consent prior to study start. The trial is registered at clinicaltrials.gov, with number NCT03131518.

Study sample

Measurements

Quantitative and qualitative data analyses

The statistical analyses for the quantitative data will be performed using the statistical software package IBM SPSS Statistics version 24.0 (IBM Corp., Somers, New York, USA). A two-sided p-value of <0.05 will be considered statistically significant. Descriptive analyses will be conducted and continuous variables will be presented as means and standard deviations (SD), categorical variables as proportions. Moreover, regression analyses will be performed according to Twisk and Proper [73], using data from the post-measures adjusting for baseline measures, to assess differences between (1) the control group and the intervention group (between design) for all outcome variables, and (2) within the intervention group when comparing the three different arms regarding total amount of bicycling and PA, transportation habits, and self-reported health and HRQoL.

A qualitative content analysis will be conducted from a phenomenologic-hermeneutic perspective [74]. Recordings of the qualitative interviews will be transcribed verbatim and read repeatedly to grasp the meaning of the data as a whole. Data will then be imported into the software analysis program NVivo 11 for further analysis. The next step will be to separate the text into meaning units; i.e. the words, sentences or paragraphs containing aspects related to each other through their content and context. Following this, the meaning units will be classified into subcategories, by pooling data with similar characteristics together in a category defined by its content. Finally, the subcategories will be combined and categorized into main categories.