Hostname: page-component-5db6c4db9b-bhjbq Total loading time: 0 Render date: 2023-03-24T04:43:20.181Z Has data issue: true Feature Flags: { "useRatesEcommerce": false } hasContentIssue true

Analysing the travel behaviour of older adults: what are the determinants of sustainable mobility?

Published online by Cambridge University Press:  21 November 2022

Basar Ozbilen*
Affiliation:
City and Regional Planning, Knowlton School, The Ohio State University, Columbus, Ohio, USA
Gulsah Akar
Affiliation:
School of City and Regional Planning, College of Design, Georgia Institute of Technology, Atlanta, Georgia, USA
Katie White
Affiliation:
Central Ohio Area Agency on Aging, City of Columbus Recreation and Parks Department, Columbus, Ohio, USA
Holly Dabelko-Schoeny
Affiliation:
College of Social Work, The Ohio State University, Columbus, Ohio, USA
Qiuchang Cao
Affiliation:
Pepper Institute on Aging and Public Policy & Claude Pepper Center, Florida State University, Tallahassee, Florida, USA
*
*Corresponding author. Email: ozbilen.1@osu.edu
Rights & Permissions[Opens in a new window]

Abstract

In recent years, various authorities launched projects that aim to make their cities more age-friendly. Designing age-friendly cities is a complex and context-dependent process that requires clear implementation guidelines for policy makers. As one of the eight domains of age-friendly cities, transportation is a critical component of making our cities more liveable for older adults and their families. This paper contributes to the literature by exploring the travel behaviour of older adults with a focus on the factors that lead to sustainable mobility patterns. Our empirical analysis is based on survey data collected from 1,221 older adults as part of the Age-Friendly Columbus project in Columbus, Ohio in the United States of America. We develop multinomial logistic regression models to investigate the travel mode choices of older adults (auto only, non-auto options only and multimodal (auto and at least one non-auto option)). We include age and built environment characteristics as the key variables, with lifestyle-related factors and socio-demographics as controls in our analysis. We find older respondents were more likely to use autos only compared to younger respondents. Our analysis also reveals significant associations between built environment characteristics and travel mode choices. Interaction effects show that the relationships between built environment characteristics and travel preferences differed by age cohorts among older individuals. The primary contribution of this study is that it provides evidence on what built environmental improvements help to promote sustainable travel among older adults in mid-sized and auto-dependent metropolitan cities. We argue that these improvements contribute to older adults' sustainable mobility, as well as out-of-home activity behaviour, social engagement and individual health. The results of this study may especially benefit non-driver older adults who lack reliable non-auto alternatives for their daily travel.

Type
Article
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - ND
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided that no alterations are made and the original article is properly cited. The written permission of Cambridge University Press must be obtained prior to any commercial use and/or adaptation of the article.
Copyright
Copyright © The Author(s), 2022. Published by Cambridge University Press

Introduction

The world is at the edge of a great demographic shift. The number of persons aged 65 years or older is projected to double by 2050 worldwide (United Nations, 2019). To support governments in making their communities more age-friendly, the World Health Organization (WHO) launched the Global Age-friendly Cities and Communities project. Age-friendliness of a community can be evaluated under eight domains: outdoor spaces and buildings, transportation, housing, social participation, respect and social inclusion, civic participation and employment, communication and information, and community support and health services (WHO, 2007). Impacting all of the domains, transportation is a key element for the wellbeing of older adults (Banister and Bowling, Reference Banister and Bowling2004; Adorno et al., Reference Adorno, Fields, Cronley, Parekh and Magruder2018). The lack of mobility is associated with an increased rate of social isolation and depression among older adults (Klicnik and Dogra, Reference Klicnik and Dogra2019; Ragland et al., Reference Ragland, MacLeod, McMillan, Doggett and Felschundneff2019).

As of 2017, the estimated older adult population (65 years and older) in the United States of America (USA) was over 50 million, which is equal to 15.6 per cent of the population (The Administration for Community Living, 2018). Parallel with the ageing trend in the world, this number is projected to exceed 71 million by 2030 (Colby and Ortman, Reference Colby and Ortman2015). This means over 20 per cent of the US population will be 65 years and older by 2030. Identifying the mobility needs of the ageing population is a fundamental challenge for policy makers in the USA. Therefore, pro-actively analysing and understanding the complex travel needs of this rapidly growing population segment is imperative.

In the USA, the preferred travel mode among older adults is privately owned vehicles (over 91%) (Shen et al., Reference Shen, Koech, Feng, Rice and Zhu2017), which suggests an auto-dependent lifestyle. The share of private vehicle users among adults aged 25–64 years is 90 per cent (Shen et al., Reference Shen, Koech, Feng, Rice and Zhu2017) and it varies between 85 and 93 per cent across different cohorts of adults (Dutzik and Inglis, Reference Dutzik and Inglis2014). This is expected considering the auto-oriented urban development in most of the US cities that makes the private automobile a convenient transportation mode for all (McMillan and Lee, Reference McMillan and Lee2017). Despite its convenience, reliance on privately owned vehicles can cause community-level challenges such as high levels of traffic congestion, increasing rates of crashes, fatalities and injuries, growth of greenhouse gas emissions, oil dependence, etc. (Steg and Gifford, Reference Steg and Gifford2005; Black, Reference Black2010; Gudmundsson et al., Reference Gudmundsson, Hall, Marsden and Zietsman2016). Car dependency can also cause individual-level challenges for older adults such as loss of mobility due to driving cessation/disabilities (Marottoli et al., Reference Marottoli, de Leon, Glass, Williams, Cooney, Berkman and Tinetti1997; Alsnih and Hensher, Reference Alsnih and Hensher2003; Adorno et al., Reference Adorno, Fields, Cronley, Parekh and Magruder2018), driving safety concerns as a result of cognitive/physical impairments, etc. (Rosenbloom, Reference Rosenbloom2001; Dumbaugh, Reference Dumbaugh2016; Centers for Disease Control and Prevention, 2020). Sustainable transportation, which refers to accessible, affordable, high quality, multimodal transportation that minimises the environmental and economic costs, is the only viable solution to all these problems (Litman, Reference Litman1999; Schiller et al., Reference Schiller, Bruun and Kenworthy2010). The promotion of sustainable transportation modes, namely transit, bicycling, and walking as alternatives to auto have the potential to provide positive health outcomes for older adults and help policy makers to create a balanced and equitable transportation network for all (Litman, Reference Litman2002, Reference Litman2019; Simonsick et al., Reference Simonsick, Guralnik, Volpato, Balfour and Fried2005; Kerr et al., Reference Kerr, Rosenberg and Frank2012; Cheng et al., Reference Cheng, Chen, Yang, Cao, De Vos and Witlox2019a). Switching from driving to active travel (e.g. walking and bicycling) can result in numerous positive outcomes on physical and mental health of older adults due to the physical activity included and the reductions in pollutants (Frank et al., Reference Frank, Kerr, Rosenberg and King2010; Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017; Yang et al., Reference Yang, Xu, Rodriguez, Michael and Zhang2018). Beyond walkable/bikeable distances, transit services may provide travel alternatives to older adults who experience driving cessation or to those living in lower-income households and cannot afford to drive. Emerging evidence also shows that public transit use is associated with health benefits for older adults because of the first- and last-mile connections that require walking (Voss et al., Reference Voss, Sims-Gould, Ashe, Mckay, Pugh and Winters2016). Provision of a reliable, accessible and affordable transit service, as well as proper infrastructure for walking/bicycling, are critical components of an equitable transportation system for all (WHO, 2002; Adorno et al., Reference Adorno, Fields, Cronley, Parekh and Magruder2018; Litman, Reference Litman2019; Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Wachs and Pinski2019).

This paper contributes to the literature by analysing the determinants of sustainable mobility among older adults. Specifically, our study responds to the following questions:

  1. (1) What are the built environment characteristics that affect the travel outcomes of older adults?

  2. (2) Are there cohort-specific differences in the associations between built environment characteristics and travel mode choices among older adults?

  3. (3) What can transportation planners do to promote sustainable mobility for older adults?

We employ multinomial logistic regression (MNL) models to analyse the links between travel mode choices, age and built environment while accounting for socio-demographics and lifestyle-related factors. Our focus on age and built environment characteristics is driven by the environmental gerontological perspective, which posits old age as a critical stage in the lifespan that is significantly influenced by the physical environment (Lawton and Nahemow, Reference Lawton, Nahemow, Eisdorfer and Lawton1973; Lawton, Reference Lawton, Lawton, Windley and Byerts1982; Wahl et al., Reference Wahl, Iwarsson and Oswald2012). Previous research on ageing shows that the built environment influences daily mobility, independence, social engagement, physical activity levels and wellbeing among older adults (Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017; Adorno et al., Reference Adorno, Fields, Cronley, Parekh and Magruder2018; Bigonnesse and Chaudhury, Reference Bigonnesse and Chaudhury2019; Cao et al., Reference Cao, Dabelko-Schoeny, White and Choi2019; Li, Reference Li2020; Lyu and Forsyth, Reference Lyu and Forsyth2022). Built environment characteristics such as urban density, diversity of land use, walking infrastructure and prevalence of transit stops are facilitators of walking, bicycling and transit use among older adults (Kemperman and Timmermans, Reference Kemperman and Timmermans2009; Kerr et al., Reference Kerr, Rosenberg and Frank2012; Cerin et al., Reference Cerin, Lee, Barnett, Sit, Cheung, Chan and Johnston2013; Figueroa et al., Reference Figueroa, Nielsen and Siren2014; Chudyk et al., Reference Chudyk, Winters, Moniruzzaman, Ashe, Gould and McKay2015). Due to declines in cognitive and physical capacity as people age, designing pedestrian, bicycling and transit-friendly environments that promote sustainable mobility of older adults is particularly important for relatively older age groups (Marottoli et al., Reference Marottoli, de Leon, Glass, Williams, Cooney, Berkman and Tinetti1997; Mezuk and Rebok, Reference Mezuk and Rebok2008; Schouten et al., Reference Schouten, Blumenberg, Wachs and King2021). The assessment of the links between travel behaviour and built environment characteristics is even more crucial in auto-dependent North American cities because older residents in these cities face higher risk of mobility disadvantage as they age and experience driving cessation (Adorno et al., Reference Adorno, Fields, Cronley, Parekh and Magruder2018; Dabelko-Schoeny et al., Reference Dabelko-Schoeny, Maleku, Cao, White and Ozbilen2021). It is important to control for other well-known determinants of travel behaviour, namely socio-demographics and lifestyle-related factors, to examine the true associations between the built environment and mobility of older adults (Schwanen et al., Reference Schwanen, Dijst and Dieleman2001; Rosenbloom, Reference Rosenbloom2009; Chudyk et al., Reference Chudyk, Winters, Moniruzzaman, Ashe, Gould and McKay2015; Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017; Shrestha et al., Reference Shrestha, Millonig, Hounsell and McDonald2017; Ulfarsson and Kim, Reference Ulfarsson and Kim2019). Consequently, our study controls for factors such as gender, race, employment status, household income, living arrangements, health status and having others available to ask for a ride.

This paper is organised as follows. We first summarise the literature describing the factors influential on travel outcomes of older adults. We then present our study area, data and methods. Next, we demonstrate descriptive statistics and model estimations, and draw from these to inform policy makers about the relevant policies and future research directions. We conclude with potential strategies to enhance sustainable mobility for older adults.

Determinants of older adults' travel behaviour

A significant body of work demonstrates that ageing-related limitations (e.g. functional and cognitive impairments, etc.) and environmental barriers (e.g. spatial segregation, lack of infrastructure, etc.) may limit older adults' mobility (Collia et al., Reference Collia, Sharp and Giesbrecht2003; Dumbaugh, Reference Dumbaugh2016; Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017; Forsyth et al., Reference Forsyth, Molinsky and Kan2019). Previous research also shows that the lack of mobility is associated with negative health outcomes and, consequently, decreased quality of life for older adults (Marottoli et al., Reference Marottoli, de Leon, Glass, Williams, Cooney, Berkman and Tinetti1997; Fonda et al., Reference Fonda, Wallace and Herzog2001; Alsnih and Hensher, Reference Alsnih and Hensher2003; Kerr et al., Reference Kerr, Rosenberg and Frank2012; Adorno et al., Reference Adorno, Fields, Cronley, Parekh and Magruder2018).

Research shows older adults travel less than younger ones (Collia et al., Reference Collia, Sharp and Giesbrecht2003; Szeto et al., Reference Szeto, Yang, Wong, Li and Wong2017; Rahman et al., Reference Rahman, Deb, Strawderman, Smith and Burch2019). However, they are physically more active and drive more than previous generations (Rosenbloom, Reference Rosenbloom2001; Nordbakke and Schwanen, Reference Nordbakke and Schwanen2015; Wood and Horner, Reference Wood and Horner2019). Given mobility is among the top factors that influence the wellbeing of older adults (Banister and Bowling, Reference Banister and Bowling2004; Hjorthol, Reference Hjorthol2013), the provision of diverse travel options for older adults is crucial. In many Western countries – particularly in the USA – older adults possess auto-oriented travel behaviour and lifestyles (Buehler and Pucher, Reference Buehler and Pucher2012; Shen et al., Reference Shen, Koech, Feng, Rice and Zhu2017). Therefore, driving cessation may significantly increase social isolation and depression levels among older adults (Marottoli et al., Reference Marottoli, de Leon, Glass, Williams, Cooney, Berkman and Tinetti1997; Davey, Reference Davey2007; Mezuk and Rebok, Reference Mezuk and Rebok2008). Additionally, changing travel behaviours can be challenging for older adults due to their established habits and perceptions (Rosenbloom and Waldorf, Reference Rosenbloom and Waldorf2001; Dill et al., Reference Dill, Mohr and Ma2014). Recent studies demonstrate that habitual behaviour may be the true determinant of older adults' driving preference and, thus, driving cessation may limit their daily mobility (Mifsud et al., Reference Mifsud, Attard and Ison2017; Caragata, Reference Caragata2021). These issues call for further research that focuses on the promotion of non-auto transportation alternatives for older adults in meeting their transportation needs and challenges.

Previous research shows that the travel outcomes of older adults depend on socio-demographics, lifestyle-related factors and built environment characteristics (Rosenbloom and Waldorf, Reference Rosenbloom and Waldorf2001; Schmöcker et al., Reference Schmöcker, Quddus, Noland and Bell2008; Cao et al., Reference Cao, Mokhtarian and Handy2010; Van den Berg et al., Reference Van den Berg, Arentze and Timmermans2011; Figueroa et al., Reference Figueroa, Nielsen and Siren2014; Chudyk et al., Reference Chudyk, Winters, Moniruzzaman, Ashe, Gould and McKay2015; Hahn et al., Reference Hahn, Kim, Kim and Ulfarsson2016; Böcker et al., Reference Böcker, van Amen and Helbich2017; Forsyth et al., Reference Forsyth, Molinsky and Kan2019; Klicnik and Dogra, Reference Klicnik and Dogra2019; Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Wachs and Pinski2019; Cheng et al., Reference Cheng, De Vos, Shi, Yang, Chen and Witlox2019b; Kan et al., Reference Kan, Forsyth and Molinsky2020). In this section, we present the existing research findings on these three categories.

Socio-demographics

Socio-demographics are important factors when it comes to older adults and their travel behaviour. Prior studies indicate that age is a significant factor that reduces the travel demand of older adults (Collia et al., Reference Collia, Sharp and Giesbrecht2003; Figueroa et al., Reference Figueroa, Nielsen and Siren2014). As the risk for cognitive and physical decline increases as people age (Marottoli et al., Reference Marottoli, de Leon, Glass, Williams, Cooney, Berkman and Tinetti1997; Mezuk and Rebok, Reference Mezuk and Rebok2008; Schouten et al., Reference Schouten, Blumenberg, Wachs and King2021), the decrease in daily trips is expected. Previous research focusing on Europe and Asia shows that ageing older adults drive less and use non-auto alternatives more (Rosenbloom, Reference Rosenbloom2001; Schwanen et al., Reference Schwanen, Dijst and Dieleman2001; Alsnih and Hensher, Reference Alsnih and Hensher2003; Szeto et al., Reference Szeto, Yang, Wong, Li and Wong2017; Cheng et al., Reference Cheng, Chen, Yang, Cao, De Vos and Witlox2019a). However, studies focusing on auto-dependent geographies such as the USA and Canada show that ageing older adults use autos more and non-auto options such as transit, walking and bicycling less (Hess, Reference Hess2009; Rosenbloom, Reference Rosenbloom2009; Buehler, Reference Buehler2011; Shen et al., Reference Shen, Koech, Feng, Rice and Zhu2017).

Older adults differ in terms of travel needs and preferences depending on their gender and race (Collia et al., Reference Collia, Sharp and Giesbrecht2003; Hess, Reference Hess2009; Cao et al., Reference Cao, Mokhtarian and Handy2010; Böcker et al., Reference Böcker, van Amen and Helbich2017). Based on previous research, older women drive less and ride transit more than older men (Nobis and Lenz, Reference Nobis and Lenz2005; Ulfarsson and Kim, Reference Ulfarsson and Kim2019). This is because women are less likely to have a driver's licence, and thus rely on alternative modes of transportation more (Metz, Reference Metz2003; Rosenbloom, Reference Rosenbloom2009; Wacker and Roberto, Reference Wacker and Roberto2014). Older adults with different racial and ethnic backgrounds show various travel patterns and mode choices. Research shows non-White older adults, i.e. Asians, Blacks and Hispanics, are less likely to drive and more likely to use alternative modes such as transit and walking than their White counterparts (Rosenbloom and Waldorf, Reference Rosenbloom and Waldorf2001; Beckman and Goulias, Reference Beckman and Goulias2008; Ding et al., Reference Ding, Sallis, Norman, Frank, Saelens, Kerr, Conway, Cain, Hovell, Hofstetter and King2014). Most non-White older adults, especially those who are immigrants, on a low income, etc., are more likely to live in lower-income inner-city neighbourhoods (Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Wachs and Pinski2019), which suggests that they rely on transit and non-motorised transportation options.

Employment status is another significant determinant that affects travel needs and mode choices of older adults (Schwanen et al., Reference Schwanen, Dijst and Dieleman2001; Hahn et al., Reference Hahn, Kim, Kim and Ulfarsson2016). Employed older adults are more likely to be auto users due to their relatively higher income levels (Hjorthol et al., Reference Hjorthol, Levin and Sirén2010; Shrestha et al., Reference Shrestha, Millonig, Hounsell and McDonald2017). For many older adults, ‘retiring from work’ means ‘retiring from driving’ (Alsnih and Hensher, Reference Alsnih and Hensher2003; Ang et al., Reference Ang, Lee, Chen, Yap, Song and Oxley2020). Being employed decreases non-work auto travel among older adults due to limited time availability (Figueroa et al., Reference Figueroa, Nielsen and Siren2014). Berg et al. (Reference Berg, Levin, Abramsson and Hagberg2015) argue that even if newly retired people might have an auto-oriented lifestyle, they have very positive attitudes towards walking and bicycling, especially for discretionary travel. Contrary to these findings, Buehler (Reference Buehler2011) shows that retirement increases car use among older adults in the USA. Older adults with disabilities are more likely to prefer walking (Schmöcker et al., Reference Schmöcker, Quddus, Noland and Bell2008). Additionally, those who are unable to work are more likely to use transit over autos (Bardaka and Hersey, Reference Bardaka and Hersey2019). Linked with the employment status, household income is found to have significant effects on the travel behaviour of older adults. Previous research shows that retired older adults' household incomes are lower than their younger counterparts (Shrestha et al., Reference Shrestha, Millonig, Hounsell and McDonald2017), therefore, they are more likely to have limited transportation options due to financial constraints (Novek and Menec, Reference Novek and Menec2014; Lehning et al., Reference Lehning, Kim, Smith and Choi2018; Cheng et al., Reference Cheng, De Vos, Shi, Yang, Chen and Witlox2019b). Older adults with relatively lower disposable incomes are more likely to prefer walking, bicycling and transit as compared to auto (Kim and Ulfarsson, Reference Kim and Ulfarsson2004; Böcker et al., Reference Böcker, van Amen and Helbich2017; Rahman et al., Reference Rahman, Deb, Strawderman, Smith and Burch2019). According to Schmöcker et al. (Reference Schmöcker, Quddus, Noland and Bell2008), there is a strong association between mode choice and household income because income level is a significant factor that determines the auto ownership of an individual. All above socio-demographic factors are influential on the travel behaviour of older adults, thus transportation studies need to account for these factors wherever possible.

Lifestyle-related factors

Lifestyle-related factors such as health status and living arrangements (living alone versus living with others) are also influential on older adults' travel behaviour (Hess, Reference Hess2009; Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017). Better health status is positively associated with out-of-home activity participation, and thus promotes overall travel among older adults (Nordbakke and Schwanen, Reference Nordbakke and Schwanen2015; Ragland et al., Reference Ragland, MacLeod, McMillan, Doggett and Felschundneff2019). Mezuk and Rebok (Reference Mezuk and Rebok2008) and Chihuri et al. (Reference Chihuri, Mielenz, DiMaggio, Betz, DiGuiseppi, Jones and Li2016) show that worsening health conditions can cause driving cessation and limit older adults' mobility. In other words, those with better health conditions may drive more in the long run. Contrary to this argument, some studies argue better health status can lead to more walking and transit use rather than driving (Naumann et al., Reference Naumann, Dellinger, Anderson, Bonomi, Rivara and Thompson2009; Böcker et al., Reference Böcker, van Amen and Helbich2017; Klicnik and Dogra, Reference Klicnik and Dogra2019). It is also important to note that active modes of transportation, such as walking and bicycling, promote positive health outcomes among older adults (Simonsick et al., Reference Simonsick, Guralnik, Volpato, Balfour and Fried2005; Kerr et al., Reference Kerr, Rosenberg and Frank2012; Cheng et al., Reference Cheng, Chen, Yang, Cao, De Vos and Witlox2019a).

Living arrangement is another important determinant of travel mode choice among older adults. Those living alone are more likely to walk and use transit as compared to those living with others (Hess, Reference Hess2009; Chudyk et al., Reference Chudyk, Winters, Moniruzzaman, Ashe, Gould and McKay2015). We want to highlight that the share of older women living alone is higher than older men because they tend to outlive men (Ortman et al., Reference Ortman, Velkoff and Hogan2014; Reher and Requena, Reference Reher and Requena2018). Hjorthol et al. (Reference Hjorthol, Levin and Sirén2010) underline that if a woman is dependent on her husband for travel, the loss of the husband means the loss of the driver and auto travel opportunity, especially among the oldest cohorts. Linked partially with living arrangement, having others to ask for rides is another important factor that affects travel behaviour. Rides given by others are the second most preferred travel mode after driving alone among older adults (Rahman et al., Reference Rahman, Strawderman, Adams-Price and Turner2016; Ragland et al., Reference Ragland, MacLeod, McMillan, Doggett and Felschundneff2019). For those with mobility constraints, rides given by others are extremely important (Dumbaugh, Reference Dumbaugh2016). Those who provide rides to older adults may be family members such as spouses, significant others, children and friends (Burkhardt, Reference Burkhardt1999; Choi et al., Reference Choi, Adams and Kahana2012). Those living alone without others to ask for rides are prone to crucial mobility limitations (Hess, Reference Hess2009; Tsai et al., Reference Tsai, Rantakokko, Portegijs, Viljanen, Saajanaho, Eronen and Rantanen2013). Therefore, travel behaviour studies need to consider the availability of this option for older adults, especially non-drivers.

Built environment characteristics

A substantial body of literature demonstrates that built environment characteristics are influential on travel behaviour, particularly on non-auto travel such as transit use, bicycling and walking (Handy et al., Reference Handy, Boarnet, Ewing and Killingsworth2002; Ewing and Cervero, Reference Ewing and Cervero2010; Shrestha et al., Reference Shrestha, Millonig, Hounsell and McDonald2017; Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Wachs and Pinski2019). Considering the possible decline of physical/cognitive capacity as people age, the association between built environment and travel behaviour is also important for the mobility and wellbeing of older adults (Nahemow and Lawton, Reference Nahemow, Lawton and Preiser1973; Lawton, Reference Lawton1989). Therefore, controlling for both objective and perceived built environment measures wherever possible is crucial. It is important to point out that objective measures may not reflect the perspectives of older adults perfectly (Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Wachs and Pinski2019). In this manner, previous studies using both perceived and objective measures show that inclusion of both would be helpful to capture the true associations between built environment measures and travel behaviour (Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017).

Previous research shows that built environment characteristics such as residential and commercial density, land-use mix, proximity to open and green spaces, visually appealing and aesthetically pleasing scenery, and quality of pedestrian and bicycling infrastructure are particularly influential on older adults' travel behaviour (Kemperman and Timmermans, Reference Kemperman and Timmermans2009; Kerr et al., Reference Kerr, Rosenberg and Frank2012; Cerin et al., Reference Cerin, Lee, Barnett, Sit, Cheung, Chan and Johnston2013, Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017; Figueroa et al., Reference Figueroa, Nielsen and Siren2014; Chudyk et al., Reference Chudyk, Winters, Moniruzzaman, Ashe, Gould and McKay2015; Forsyth et al., Reference Forsyth, Molinsky and Kan2019; Cheng et al., Reference Cheng, Chen, Yang, Cao, De Vos and Witlox2019a; Kan et al., Reference Kan, Forsyth and Molinsky2020). Due to health and safety concerns, well-lit sidewalks, safe urban environments and parks, pedestrian countdown timers at crosswalks, and non-slippery pavements for walking are also important facilitators of active transportation amongst older adults (Metz, Reference Metz2003; Dumbaugh, Reference Dumbaugh2016; Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Levy-Storms, Chen and Brozen2016; Böcker et al., Reference Böcker, van Amen and Helbich2017; Ragland et al., Reference Ragland, MacLeod, McMillan, Doggett and Felschundneff2019). Adkins et al. (Reference Adkins, Makarewicz, Scanze, Ingram and Luhr2017) and Loukaitou-Sideris et al. (Reference Loukaitou-Sideris, Wachs and Pinski2019) point out that in neighbourhoods in which safety is a major concern due to high crime rates or disorders, the effects of built environment on travel behaviour can be suppressed. Thus, wherever possible, controlling for the environmental safety measures is important.

It is important to keep in mind that the association between travel behaviour and built environment can be spurious. Individuals with positive attitudes towards specific transportation modes such as walking or bicycling may choose to live in neighbourhoods that meet their travel needs and preferences. Therefore, residential preferences might be the true determinants of travel behaviour rather than neighbourhood built environment for these individuals (Cao et al., Reference Cao, Mokhtarian and Handy2010). In previous literature, this is defined as residential self-selection (Cao et al., Reference Cao, Mokhtarian and Handy2009). Controlling for the residential self-selection would provide more accurate estimations on travel behaviour (Adkins et al., Reference Adkins, Makarewicz, Scanze, Ingram and Luhr2017).

Transportation service characteristics are influential on older adults' travel preferences. Accessible, affordable, convenient, frequent and reliable transit service provision is important, especially for those who do not own autos and/or cannot drive (Alsnih and Hensher, Reference Alsnih and Hensher2003; WHO, 2007; Novek and Menec, Reference Novek and Menec2014; Dumbaugh, Reference Dumbaugh2016; Szeto et al., Reference Szeto, Yang, Wong, Li and Wong2017; Ragland et al., Reference Ragland, MacLeod, McMillan, Doggett and Felschundneff2019). The quality of transit infrastructure, such as transit stops with benches and shelters, and the service reliability are particularly important for transit preferences during extreme weather conditions (Shrestha et al., Reference Shrestha, Millonig, Hounsell and McDonald2017; Klicnik and Dogra, Reference Klicnik and Dogra2019). Considering the importance of the first- and last-mile access issues for transit, high-quality pedestrian infrastructure and prevalence of transit stops are also important for older adults' travel behaviour (Kerr et al., Reference Kerr, Rosenberg and Frank2012; Cerin et al., Reference Cerin, Lee, Barnett, Sit, Cheung, Chan and Johnston2013). Dill et al. (Reference Dill, Mohr and Ma2014) indicate individuals' perceptions about transportation service and infrastructure can influence their mode choices. They argue that perceived quality of service and infrastructure can have mediating effects on the relationship between built environment characteristics and travel behaviour. Given that older adults are more likely to experience mobility limitations that may influence their perceptions negatively, it is important to include variables regarding perceived transportation service quality in travel behaviour analyses.

Promoting sustainable mobility among older adults in the USA

Considering the ageing population of the USA, the above-mentioned built environment factors related to infrastructure, design and service characteristics need to be incorporated into the neighbourhood design processes. For that reason, researchers in the field of ageing conducted various empirical studies in recent years to test the potential effects of the built environment on older adults' mobility in US cities. These studies elaborated the impacts of various urban planning concepts such as transit-oriented developmentFootnote 1 and smart growth,Footnote 2 and different built environment variables such as building density, land-use mix, transit service quality, street connectivity, quality of sidewalks and proximity to nearest parks on sustainable travel mode choices of older adults (Boschmann and Brady, Reference Boschmann and Brady2013; Yang et al., Reference Yang, Xu, Rodriguez, Michael and Zhang2018; Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Wachs and Pinski2019; Bai et al., Reference Bai, Steiner and Zhai2021; Schouten et al., Reference Schouten, Blumenberg, Wachs and King2021). These studies demonstrated that the built environment interventions can significantly influence older adults' travel behaviour, particularly in reducing auto trips and promoting non-auto modes of transportation. Providing non-auto alternatives is crucial in solving the accessibility issues of transportation-disadvantaged communities with limited access to health care, goods, services and social networks, particularly in auto-oriented US cities (Adorno et al., Reference Adorno, Fields, Cronley, Parekh and Magruder2018; Lehning et al., Reference Lehning, Kim, Smith and Choi2018; Dabelko-Schoeny et al., Reference Dabelko-Schoeny, Maleku, Cao, White and Ozbilen2021).

Based on this understanding, this study focuses on the impacts of various built environment factors and age cohorts of older adults on their travel preferences while controlling for well-known and widely used control variables such as socio-demographics and lifestyle-related factors (for a summary of the factors that influence older adults' travel behaviour, see Table 1). Since environmental gerontology literature points out the need for the assessment of individual-level factors together with social and physical environmental components for ageing studies (Lawton and Nahemow, Reference Lawton, Nahemow, Eisdorfer and Lawton1973; Wahl et al., Reference Wahl, Iwarsson and Oswald2012), we used a diverse set of variables in our analyses. These variables provide us with the opportunity to examine the true associations between the built environment and sustainable mobility of older adults while controlling for well-known determinants of travel behaviour in later life. Our study presents on-site implementation guidelines for urban planners and decision-makers that are transferable to other cities with similar urban development patterns.

Table 1. Summary table for determinants of older adults' travel behaviour

Data and methodology

This study focuses on the travel behaviour of older adults in Columbus, Ohio, in the USA. Central Ohio residents aged 50 years and older make up 31 per cent of the total population (American Community Survey, 2020), and the number of older adults in the region is expected to double in the next 35 years (Mid-Ohio Regional Planning Commission (MORPC), 2017). In order to respond to this demographic change, the authorities launched the Age-Friendly Columbus (AFC) initiative in 2016. Since then, AFC has been working on research projects and on-site applications that are intended to make the city more liveable for individuals of all ages.

Columbus is the largest metropolitan area in Central Ohio and the 32nd most populous metropolitan area in the nation (Statista, 2021). It has a diverse representation of age groups, cultures and ethnicities. Columbus metropolitan area has a sprawled urban form, a well-connected highway system and a relatively sparse transit network (Wang and Chen, Reference Wang and Chen2015; Vyas et al., Reference Vyas, Famili, Vovsha, Fay, Kulshrestha, Giaimo and Anderson2019). As a major Midwestern city that is more auto-dependent than coastal cities such as New York, San Francisco and Boston, the analysis of the travel behaviour of older adults can offer valuable insights to numerous US cities with a similar urban form and transportation network.

Data

The primary data for this study were collected by the Center for Community Solutions through engagement with the MORPC as a part of the AFC and Franklin County initiative. Through two surveys conducted in Columbus, Ohio from September to November in 2016, the respondents were asked questions about eight age-friendly domains (MORPC, 2017). The dataset provides 1,221 responses from registered voters aged 50 and older residing in Columbus (stratified by ZIP codes and age groups according to the voter registration list). Age 50 years was selected as the cut-off point because racial and ethnic minorities and those living in poverty experience age-related changes at younger ages than White and/or wealthier counterparts. The same age cut-off point is used by several earlier studies that focus on older adults' travel behaviour (Hess and Russell, Reference Hess and Russell2012; Fordham et al., Reference Fordham, Grisé and El-Geneidy2017; Li and Tilahun, Reference Li and Tilahun2017). Both surveys provide detailed information on respondents' individual and household characteristics, transportation mode choices and impressions regarding amenities, services, infrastructure and barriers within their neighbourhood.

We linked the AFC survey with the US Environmental Protection Agency's Smart Location Database (SLD) to obtain information about built environment characteristics. SLD is a data product that summarises more than 90 different indicators associated with built environment and location efficiency (for details, see US Environmental Protection Agency, 2014). Most of the SLD attributes are available at the census block group (CBG) level; however, since our primary data source is stratified at the ZIP code level, we recalculated all SLD variables at the ZIP code level. In the condition of lack of available built environment data at a finer scale such as CBG or a buffered zone around the participants' home, the use of ZIP codes for the analyses is a common practice in travel behaviour studies (Freeman et al., Reference Freeman, Neckerman, Schwartz-Soicher, Quinn, Richards, Bader, Lovasi, Jack, Weiss, Konty, Arno, Viola, Kerker and Rundle2012; Yang et al., Reference Yang, Xu, Rodriguez, Michael and Zhang2018; Macleod et al., Reference Macleod, Thorhauge, Villalobos, Van Meijgaard, Karriker-Ja, Kelley-Baker and Ragland2020). The recalculation using ArcGIS software (version 10.6) can be summarised within four steps: (a) using the intersect tool, we calculated intersecting features of each CBG by ZIP codes; (b) using field calculator, we estimated the percentage of area covered by each intersecting feature within the corresponding ZIP code area; (c) using the percentages calculated in the second step, we calculated the normalised values for all SLD variables for each feature within the ZIP code; and (4) using the spatial join operation, we aggregated intersecting features by the corresponding ZIP codes.

The AFC survey respondents were asked about their travel mode choices for running errands, getting medical appointments or attending events. These items were considered the most frequent activities of daily interests. The respondents were allowed to choose multiple options. Based on the respondents' answers to this question, we created three categories to represent individual travel preferences:

  1. (1) Auto user (individual uses the auto option only as a driver and/or as a passenger).

  2. (2) Non-auto user (individual uses only non-auto alternatives such as walking, bicycling, transit, etc.).

  3. (3) Multimodal traveller (individual uses both auto – as a driver and/or as a passenger – and at least one non-auto alternative).

As indicated previously, non-auto options such as walking, bicycling and transit promote positive health outcomes among older adults, minimise environmental and economic costs, and help eliminate isolation issues created by the driving cessation in older ages. Due to these reasons, the multimodal travel preference (indicating that individual uses non-auto transportation modes for some of the trips) and non-auto travel preference (indicating that individual uses solely non-auto transportation modes for travel) can be referred to as more sustainable options as compared to auto-only travel preference. We included three groups of independent variables in our analysis: (a) socio-demographics, (b) lifestyle-related factors, and (c) built environment characteristics. The socio-demographics cover variables such as age, gender, race, household income and employment status. Lifestyle-related factors include health status, living alone and having others available to ask for a ride. Lastly, we included two groups of built environment variables. The first group consists of objective built environment measures, namely retail density, land-use mix and frequency of transit service that come from SLD. The second group includes perceived built environment measures, i.e. access to well-maintained and safe parks that are within walking distance of home, access to crosswalks with pedestrian countdown timers that allow enough time to cross, that come from the AFC survey. Also, we tested for additional objective built environment characteristics such as residential density, population density, regional diversity index, etc., and perceived built environment measures such as perceived transit service quality, perceived sidewalk quality, etc., in different stages of the study; however, excluded most of them because they either caused autocorrelation problems or did not provide any interpretive results.

Methodology

We examined the associations between our dependent variable, travel preference, key variables, i.e. age and built environment characteristics, and control variables, namely socio-demographics and lifestyle-related factors, using a MNL model (for details, see McFadden, Reference McFadden1974; Hausman and McFadden, Reference Hausman and McFadden1984). MNL is a method employed when you have discrete choices such as transportation mode choice. Therefore, we employ MNL for the multivariate analysis similarly to previous studies focusing on mode choice of older adults (Schwanen et al., Reference Schwanen, Dijst and Dieleman2001; Kim and Ulfarsson, Reference Kim and Ulfarsson2004; Buehler, Reference Buehler2011; Böcker et al., Reference Böcker, van Amen and Helbich2017).

The MNL employed in this study takes the following functional form:

(1)$$P_{ij} = \displaystyle{{\exp ( {S_{ij}, \;\;L_{ij}, \;\;B_{ij}} ) } \over {\mathop \sum \nolimits_{\,j = 0}^2 \exp ( {S_{ij}, \;\;L_{ij}, \;\;B_{ij}} ) }}\quad for\quad j = 0, \;\;1, \;\;2$$

where Pij = probability of person i belonging to discrete travel preference category j (0, 1 and 2 refer to auto user, non-auto user and multimodal traveller, respectively); Sij = a vector of individual socio-demographic characteristics such as age, gender, etc. of person i; Lij = a vector of lifestyle-related factors such as health status, living arrangement, etc. for person i; Bij = a vector of built environment characteristics such as access to well-maintained and safe parks that are within walking distance, land-use mix, etc. within neighbourhood of person i.

We set auto user as the reference category and interpret the coefficients for non-auto user and multimodal traveller as compared to the reference category. We also tested interaction terms between our key variables, built environment characteristics and age, to see whether there are differences across different cohorts of older adults in terms of travel behaviour. A significant number of respondents did not provide complete responses to all questions. We removed those with missing information and examined the sample using univariate and multivariate statistics to identify coding errors and outliers (Tabachnick and Fidell, Reference Tabachnick and Fidell2012). The final sample included 689 valid responses. It is important to acknowledge the limitations of using MNL for the analysis. The MNL model assumes that the ratio of probabilities of choosing any two alternatives is independent of the existence of another, the irrelevant alternatives (IIA) assumption (McFadden et al., Reference McFadden, Tye and Train1976; Greene, Reference Greene2018). The IIA assumption is the most serious limitation of MNL models because it may be unrealistic in a number of decision situations (Heinrich and Wenger, Reference Heinrich and Wenger2002; Seo, Reference Seo2016). MNL models also assume homogeneity in tastes, which implies the effects of an attribute are fixed across a population (Willis, Reference Willis2014). Lastly, in panel data settings, MNL models assume there is no serial correlation in the error term, which may not hold true in various cases (Morikawa, Reference Morikawa1994; Seo, Reference Seo2016).

Results and discussion

Table 2 presents the descriptive statistics for the dependent variable, key variables (age and built environment characteristics) and control variables. As expected, consistent with the auto-dependency in the USA (Buehler and Pucher, Reference Buehler and Pucher2012), auto users made up 84 per cent of the sample. The multimodal traveller category which includes individuals who use non-auto modes (bicycles, walking, transit, etc.) as well as autos had the second highest share (10%). Lastly, the non-auto users (those who solely use non-auto modes) accounted for 6 per cent of the sample.

Table 2. Older adults' characteristics and built environment measures, descriptive statistics

Notes:Number of observations = 689. 1. There are other categories in this question that are not included in the final model because either they lack theoretical support or they did not provide any interpretive results. These are ‘I can't afford a car or car maintenance’, ‘I do not drive’, ‘I can't afford public transportation’, ‘There is no bus to take me where I need to go’, ‘Buses are difficult to use and/or unreliable’ and ‘I don't feel safe walking’. 2. There are other categories in this question that are not included in the final analysis because either they lack theoretical support or they did not provide any interpretive results. These are ‘Streets that are visually appealing (trees, flowers, benches and public art make the street a nice place to walk or ride a bike)’, ‘Well-lit public streets and walkways’ and ‘Sidewalks that are in good condition’. SD: standard deviation.

Most of the respondents (80%) were within the 50–69 age range, with those aged 60–69 making up the largest share (44%). Twenty per cent of the respondents were aged 70 and above. The survey sample included more women than men (71% versus 29%). This is not unexpected. The dominance of women in survey samples is reported in some earlier studies using primary data collected from older adults (Naumann et al., Reference Naumann, Dellinger, Anderson, Bonomi, Rivara and Thompson2009; Ragland et al., Reference Ragland, MacLeod, McMillan, Doggett and Felschundneff2019). Most of the respondents were White/Caucasian (80%) and almost half of the respondents had more than US $60,000 household income. The median household income in Columbus is US $54,902 (US Census Bureau, 2021), and those who have more than US $60,000 household income have a relatively higher income level. Almost half of the respondents were employed (48%). More than 87 per cent of the participants indicated that their health status is good, very good or excellent. This suggests that for most of the respondents, health status may not be a limitation for walking and bicycling. About one-third of survey respondents lived alone, while 5 per cent did not have others to ask for a ride. This suggests some respondents may not have the option to travel as a passenger, which may reduce their auto use. When it comes to perceived built environment characteristics, 59 per cent of the respondents stated they have access to well-maintained and safe parks, and 55 per cent stated they have access to crosswalks with pedestrian countdown timers that allow enough time to cross in their neighbourhood. We included three objective built environment characteristics in the final model, namely retail density, land-use mixFootnote 3 and frequency of transit during evening peak.Footnote 4 We examined the correlations between all independent variables included in the analysis to rule out multicollinearity. Since all results are modest (all Pearson correlations ≤ 0.51), we proceeded with multivariate analysis (Table 3).

Table 3. Multinomial logistic regression model results

Notes: The base category is ‘auto user’. RRR: relative risk ratio (i.e. the odds ratio equivalent in multinomial logit model). CIs: confidence intervals (upper and lower bounds). 1. All variables related to the built environment characteristics are tested for interactions with age. This is the only significant interaction between the key variables.

Significance level: Bold values are significant at the 10 per cent level.

Factors affecting sustainable mobility preferences of older adults

We further examined the effects of independent variables on travel preferences of older adults using MNL models. We also tested for moderation effects through interactions between the key variables to assess differences across different age cohorts. We tested our models for multicollinearity using variance inflation factors (VIFs). Results show the mean VIF value is 1.49 and all individual VIF values are less than 2.86 (lower than the widely used cut-off value of 5.0; Craney and Surles, Reference Craney and Surles2002), which suggests that multicollinearity is not a concern for the model. We also tested all our models for IIA using Hausman–McFadden and Small–Hsiao tests (Hausman and McFadden, Reference Hausman and McFadden1984; Small and Hsiao, Reference Small and Hsiao1985; McFadden, Reference McFadden1987), that are widely used in the literature. The models meet the pre-defined criteria for both tests, which suggest that IIA assumption holds (Long and Freese, Reference Long and Freese2005). We acknowledge that these tests are sensitive to model parameterisation (Cheng and Long, Reference Cheng and Long2007; Hamre and Buehler, Reference Hamre and Buehler2014). Based on discrete choice theory and subjective judgement, we decided that all three categories are distinct choice sets for older adults.

We report relative risk ratios (RRR), p-values, and lower and upper bounds of the confidence intervals. RRR refers to the probability of choosing the corresponding outcome category over the probability of choosing the base category for a unit change in the predictor variable. In our models, being an auto user is set as the base category. If RRR is greater than 1 for an explanatory variable, the probability of being in the corresponding outcome category relative to the base outcome category increases as the value of the variable increases. As expected, if the RRR value is less than 1, it refers to otherwise. The results of these models are shown in Table 3.

Controlling for other variables (hereafter this applies to all interpretations), those aged 60–69 were less likely to prefer using non-auto modes as compared to those aged 50–59. This shows the increasing auto-only travel preferences of ageing older adults. This is consistent with the previous literature demonstrating that ageing older adults drive more and use non-auto options less, particularly in auto-dependent geographies like the USA and Canada (Hess, Reference Hess2009; Rosenbloom, Reference Rosenbloom2009; Shen et al., Reference Shen, Koech, Feng, Rice and Zhu2017). We do not find any significant relationship between age and the multimodal traveller category. It is important to note that the cross-sectional structure of our dataset may mask the cohort effects and, thus, the assessment of these findings in a longitudinal design may provide more accurate estimates about the associations between older adults' age and transportation mode choices.

Our results show that most of the built environment characteristics were associated with older adults' mode choices. Having access to well-maintained and safe parks that are within walking distance of residential locations increased the likelihood of preferring multimodal travel over the auto-only option. This finding is consistent with the previous literature which demonstrates open and green spaces in the neighbourhood are important facilitators of active travel among older adults (Cerin et al., Reference Cerin, Lee, Barnett, Sit, Cheung, Chan and Johnston2013; Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Levy-Storms, Chen and Brozen2016). Those who have access to crosswalks with countdown timers in their neighbourhood were more likely to be in more sustainable travel categories. This is in parallel with the previous studies that show pedestrian countdown timers are crucial factors that affect older adults' active travel (Metz, Reference Metz2003; Kerr et al., Reference Kerr, Rosenberg and Frank2012). Those living in neighbourhoods with higher levels of land-use mix were more likely to be multimodal travellers, which suggests land-use mix is a facilitator of sustainable mobility amongst older adults. The transit service frequency variable had slightly positive associations with both non-auto user and multimodal traveller categories. This indicates those who live in neighbourhoods with higher transit service frequencies were more likely to be in more sustainable travel categories. This is consistent with previous studies linking transit service quality and sustainable mobility among older adults (Alsnih and Hensher, Reference Alsnih and Hensher2003; Szeto et al., Reference Szeto, Yang, Wong, Li and Wong2017; Klicnik and Dogra, Reference Klicnik and Dogra2019).

The findings regarding the control variables are mostly consistent with studies conducted elsewhere. Women were more likely to be non-auto users, as expected (Nobis and Lenz, Reference Nobis and Lenz2005; Ulfarsson and Kim, Reference Ulfarsson and Kim2019). Those with higher household incomes were less likely to prefer multimodal travel and non-auto travel over the auto-only option. This is consistent with the previous literature as presented by Böcker et al. (Reference Böcker, van Amen and Helbich2017), Kim and Ulfarsson (Reference Kim and Ulfarsson2004) and Schmöcker et al. (Reference Schmöcker, Quddus, Noland and Bell2008). Those who are not disabled and unable to work were less likely to be multimodal travellers as compared to being auto users. This is consistent with the literature that shows those who are disabled and unable to work are more likely to use non-auto modes of transportation (Schmöcker et al., Reference Schmöcker, Quddus, Noland and Bell2008; Bardaka and Hersey, Reference Bardaka and Hersey2019). Older adults with better health status were more likely to be auto users as compared to being non-auto users. This finding contradicts the previous literature that shows better physical capacity is associated with more non-auto trips (Naumann et al., Reference Naumann, Dellinger, Anderson, Bonomi, Rivara and Thompson2009; Böcker et al., Reference Böcker, van Amen and Helbich2017). This might be due to the auto-dependent lifestyle and sprawled urban form in Columbus, Ohio. Some of the previous studies focusing on similar North American cities show similar findings about health status and driving behaviour (Mezuk and Rebok, Reference Mezuk and Rebok2008; Chihuri et al., Reference Chihuri, Mielenz, DiMaggio, Betz, DiGuiseppi, Jones and Li2016). Older adults living alone were considerably more likely to prefer multimodal travel and travel with non-auto modes only, as expected. This is consistent with the previous studies that demonstrate living alone is positively associated with non-auto travel (Hess, Reference Hess2009; Chudyk et al., Reference Chudyk, Winters, Moniruzzaman, Ashe, Gould and McKay2015). Those who do not have others to ask for rides were more likely to be non-auto users. Previous research shows that rides given by others are mostly preferred by older adults who cannot drive (Rahman et al., Reference Rahman, Deb, Strawderman, Smith and Burch2019). Given the auto user category includes auto use as a driver and/or as a passenger, not having anyone to ask for a ride is expected to reduce the probability of being in this category.

Age moderates the associations between built environment and travel preferences

We tested all possible interactions between age and built environment characteristics. The only significant interaction was between age and access to crosswalks with pedestrian countdown timers. In this section, we discuss the effect of age and access to crosswalks interaction only since the effects of other variables are consistent with the model without interaction except for the employment variable (see Table 3). In the model with interaction, ‘unemployed/retired and seeking work’ and ‘retired and not looking for work’ categories of the employment variable were not significantly associated with the multimodal traveller category.

The results of the model with interaction show that among those who have access to crosswalks with pedestrian countdown timers, older adults aged 70 and more were significantly more likely to choose more sustainable travel preferences as compared to those between 50 and 59. This shows that pedestrian countdown timers that allow enough time to cross are particularly important for older adults who are relatively older and more likely to experience physical limitations regarding active travel (Collia et al., Reference Collia, Sharp and Giesbrecht2003; Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017). This finding shows that a minor improvement such as the adjustment of crosswalk timers for older adults can significantly increase the active travel preferences of older cohorts. These findings are consistent with the previous studies (Metz, Reference Metz2003; Kerr et al., Reference Kerr, Rosenberg and Frank2012).

Conclusion and limitations

Our study demonstrates that age and built environment characteristics were associated with sustainable mobility preferences of older adults, controlling for socio-demographics and lifestyle-related factors. We find ageing may not cause a transition from auto to other alternatives, which is consistent with the previous research (e.g. Rosenbloom, Reference Rosenbloom2009). Findings regarding age show that ageing older adults were more likely to drive (or ask for rides) as compared to their younger counterparts. The research findings also show that transportation planners and policy makers can promote sustainable mobility by four built environment interventions: by improving park access of older-adult neighbourhoods, by adjusting the timing of pedestrian countdown timers on crosswalks to allow enough time to cross or implementing countdown timers to new intersections, by designing urban environments with higher levels of land-use mix that will increase the diversity of activities and opportunities, and by providing higher-frequency transit services. All of these built environment interventions are imperative for a well-designed and inclusive urban environment for all. They are particularly relevant to cities that are auto-dependent, sprawled and have relatively sparse transit networks because all four interventions contribute to the improvement of built environment and transportation service quality and, consequently, advance non-auto alternatives such as walking, bicycling and transit.

Among these environmental factors, designing urban environments with higher levels of land-use mix, improving access of older adults to parks, and improving transit service frequency require long-term strategies and land-use plans. Taking these policies into account in future planning efforts may help promote sustainable mobility among older adults. Given the upcoming demographic change that will reshape our society, we believe the re-design of our neighbourhoods to meet the needs of the older adults will gain more attention in the future. Therefore, these long-term strategies need to be assessed carefully by planning scholars and practitioners. Our findings regarding the pedestrian countdown timers are considerably important in the short term. These easy-to-implement interventions to the intersections might be more effective than transportation planners might think. The interaction of age with the pedestrian countdown timers shows that this intervention is considerably important for people 70 or older with possibly limited physical capabilities. The adjustment of the existing countdown timers to allow more time and/or implementation of countdown timers to new intersections might be a starting point in making the environment more age-friendly given that the other findings may have larger financial implications and require long-term planning.

Our paper builds upon previous studies conducted by environmental gerontological researchers, who underline the need to extend examinations beyond individual level to the social and built environments (Lawton and Nahemow, Reference Lawton, Nahemow, Eisdorfer and Lawton1973; Lawton, Reference Lawton, Lawton, Windley and Byerts1982; Wahl et al., Reference Wahl, Iwarsson and Oswald2012). We argue that built environment improvements contribute not only to older adults' sustainable mobility but also their out-of-home activity behaviour, social engagement, and physical and mental health (Wahl et al., Reference Wahl, Iwarsson and Oswald2012; Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017; Cao et al., Reference Cao, Dabelko-Schoeny, White and Choi2019; Lyu and Forsyth, Reference Lyu and Forsyth2022). Age-friendly neighbourhoods are an integral part of independence in later life for older people, and thus they are crucial elements of ageing-in-place policies and practices (Bigonnesse and Chaudhury, Reference Bigonnesse and Chaudhury2019). Considering the upcoming demographic change worldwide (United Nations, 2019), we encourage authorities to have a more proactive role in analysing the needs of older adults and preparing environments for their needs.

This research supports other recent studies which demonstrate that creating age-friendly cities requires an in-depth understanding of the perspectives of older adults (Dabelko-Schoeny et al., Reference Dabelko-Schoeny, Fields, White, Sheldon, Ravi, Robinson, Murphy and Jennings2020). A recent comprehensive review of older adults, mobility and living environment literature shows the importance of collaborations between urban planning and other disciplines such as social work and public health, which are traditionally more experienced in working with older adults, to design age-friendly urban environments (Li, Reference Li2020). As a multi-disciplinary team consisting of urban planners and social workers, we suggest future studies to expand these connections with other disciplines to understand age-friendliness in more comprehensive ways.

The unique contribution of this study is that it provides specific guidelines on what built environmental factors help to promote sustainable travel among older adults in mid-sized metropolitan cities. Additionally, our study demonstrates that the heterogeneity in the older population calls for specific policies that will address the varying needs of different age cohorts. The findings can assist policy makers in prioritising certain built environment-related improvements to support the mobility of older adults (e.g. the adjustment of existing countdown timers). This prioritisation can be particularly helpful for policy makers in solving the mobility issues of relatively older cohorts. The paper also provides insights into the factors that promote the sustainable mobility of older adults. The promotion of sustainable mobility options such as walking, bicycling and transit contributes to positive health outcomes among older adults and help policy makers to develop an equitable transportation system (Kerr et al., Reference Kerr, Rosenberg and Frank2012; Adorno et al., Reference Adorno, Fields, Cronley, Parekh and Magruder2018; Litman, Reference Litman2019; Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Wachs and Pinski2019; Cheng et al., Reference Cheng, Chen, Yang, Cao, De Vos and Witlox2019a). Finally, the findings of this paper can complement existing sustainable transportation policies and help authorities to develop more accessible, affordable and high-quality transportation service provisions for older adults.

This is one of the few older adult travel behaviour studies using quantitative data collected from a mid-sized metropolitan area in the USA. The policy recommendations that we draw from the data analyses are crucial, especially for non-driver older adults who suffer from the lack of reliable non-auto travel options in these cities. To be prepared for the upcoming demographic change in our society, local policy makers should take a proactive role and prepare the built environment and transportation services for the ageing population's needs. Our results can contribute to these efforts that aim to improve the quality of life for all in our communities. It is important to acknowledge that our recommendations may be more relevant for North American cities that are sprawled and auto-dependent. Similar studies conducted in other cities that have a more compact urban form, well-connected transit network and lower levels of auto-dependency show different results than our study. For example, Buehler (Reference Buehler2011) shows that while retired older adults in the USA drive more, their counterparts in Germany drive less compared to those who are not retired. Older adults living in compact cities, such as those in Denmark, Norway, the United Kingdom, the Netherlands and China, with affordable and well-developed transit networks, are less likely to drive and more likely to walk, bicycle and ride transit (Schwanen et al., Reference Schwanen, Dijst and Dieleman2001; Schmöcker et al., Reference Schmöcker, Quddus, Noland and Bell2008; Cerin et al., Reference Cerin, Lee, Barnett, Sit, Cheung, Chan and Johnston2013; Szeto et al., Reference Szeto, Yang, Wong, Li and Wong2017; Cheng et al., Reference Cheng, Chen, Yang, Cao, De Vos and Witlox2019a). In brief, we can argue that what we can conclude from this study will be more relevant for North American cities with similar urban development patterns, transportation network characteristics and auto-dependency levels.

Our findings provide valuable insights to transportation professionals and decision makers in developing policies that will help to promote sustainable mobility for older adults. We recommend authorities to provide better access to well-maintained and safe parks, adjust pedestrian countdown timers on crosswalks to allow enough time to cross the street, promote higher levels of land-use mix that will provide a more diverse set of services and amenities, and improve the transit service frequency for better service quality. All these interventions can result in many positive outcomes for older adults such as more active travel that will improve physical and mental health, better access to essential services and resources, and more individual independence. These interventions are age-friendly and climate smart, as the promotion of sustainable transportation options such as transit, bicycling and walking helps to overcome numerous community-level challenges such as air pollution, crashes, congestion, etc. Lastly, it is important to indicate that having proper infrastructure and environmental elements does not ensure older adults would go out and/or use alternative modes of transportation. Therefore, in addition to the improvements in built environment, local governments need to introduce programmes that will promote older adults' out-of-home physical activities and non-auto travel (Dill et al., Reference Dill, Mohr and Ma2014; Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Levy-Storms, Chen and Brozen2016).

We want to highlight that the results should be viewed in the light of the sampling limitations of our study. Our final sample is predominantly White/Caucasian, female and auto users. Additionally, those who are relatively older (70 and more) make up only 20 per cent of the final sample. Previous studies show that being White and having access to a private automobile increases car use among older adults (Beckman and Goulias, Reference Beckman and Goulias2008; Schmöcker et al., Reference Schmöcker, Quddus, Noland and Bell2008; Ding et al., Reference Ding, Sallis, Norman, Frank, Saelens, Kerr, Conway, Cain, Hovell, Hofstetter and King2014; Shrestha et al., Reference Shrestha, Millonig, Hounsell and McDonald2017). On the other hand, older women drive less than older men (Shen et al., Reference Shen, Koech, Feng, Rice and Zhu2017; Ulfarsson and Kim, Reference Ulfarsson and Kim2019). Lastly, studies conducted in North American cities demonstrate that ageing older adults drive more than their younger counterparts (Rosenbloom, Reference Rosenbloom2009; Buehler, Reference Buehler2011). Considering the impacts of these factors on older adults' travel behaviour and the composition of our study sample, our findings should be interpreted with caution. While we acknowledge the sampling limitations of our study, we also note that our findings regarding older adults' travel preferences are consistent with other studies using different age cut-off points (e.g. 60 or 65) with samples that are more evenly distributed across gender, race, etc. (Schmöcker et al., Reference Schmöcker, Quddus, Noland and Bell2008; Hess, Reference Hess2009; Buehler, Reference Buehler2011; Shen et al., Reference Shen, Koech, Feng, Rice and Zhu2017; Ulfarsson and Kim, Reference Ulfarsson and Kim2019).

We acknowledge that there are several limitations to our study. This study does not account for travel attitudes, habits and residential self-selection that are found to be mediating the relationship between the built environment and travel behaviour (Cao et al., Reference Cao, Mokhtarian and Handy2009, Reference Cao, Mokhtarian and Handy2010; Dill et al., Reference Dill, Mohr and Ma2014; Adkins et al., Reference Adkins, Makarewicz, Scanze, Ingram and Luhr2017; Cerin et al., Reference Cerin, Nathan, Cauwenberg, Van, Barnett and Barnett2017) due to data limitations. As identified by previous studies, modal choice by older adults may be strongly influenced by habitual practice (Rosenbloom and Waldorf, Reference Rosenbloom and Waldorf2001; Dill et al., Reference Dill, Mohr and Ma2014; Mifsud et al., Reference Mifsud, Attard and Ison2017; Caragata, Reference Caragata2021). Additionally, older adults' habits and travel-related attitudes may cause them to reside in certain neighbourhoods (Cao et al., Reference Cao, Mokhtarian and Handy2010), and the residential location might be one of the primary determinants of modal choice. Considering their mediating effects on older adults' travel behaviour, we encourage future research to include these variables in the analysis. Second, we used a cross-sectional dataset that can mask cohort effects and behavioural variations based on socio-economic factors affecting specific groups (Blumenberg and Smart, Reference Blumenberg and Smart2010). The assessment of these long-term effects calls for longitudinal approaches for more accurate and robust estimations (Cao et al., Reference Cao, Mokhtarian and Handy2009; Figueroa et al., Reference Figueroa, Nielsen and Siren2014). Third, we used ZIP code-level built environment variables due to the data limitations. We acknowledge that built environment data at a finer scale such as CBG can capture the variability in the relationship between spatial characteristics and travel behaviour more accurately. Therefore, we encourage future empirical studies to test our findings with finer-scale built environment data. Fourth, since the majority of survey respondents were registered voters, the results may not be representative of the travel behaviour of non-registered voters (e.g. immigrants, refugees and non-citizens), who make up over 10 per cent of the Columbus metropolitan area population (VERA Institute of Justice, 2017). Additionally, we acknowledge that the effects of the built environment on travel mode choice may vary between individuals based on their demographics (income level, race, etc.). While we did not find any significant associations between these two, we encourage future research to explore the variation of built environment impacts across individuals with different demographic characteristics. Lastly, our dataset is limited in terms of the environmental safety perceptions of older adults. Considering previous studies claim that there is a significant association between perceived environmental safety and non-motorised travel for older adults (Loukaitou-Sideris et al., Reference Loukaitou-Sideris, Levy-Storms, Chen and Brozen2016, Reference Loukaitou-Sideris, Wachs and Pinski2019; Adkins et al., Reference Adkins, Makarewicz, Scanze, Ingram and Luhr2017), we suggest further studies to take safety factors into account.

Conflict of interest

The authors declare no conflicts of interest.

Ethical standards

Ethical approval was not required.

Footnotes

1 Transit-oriented development is an urban development approach that includes a mix of commercial, residential, office and entertainment centred around transit stations (Federal Transit Administration, 2019).

2 According to the definition of the American Planning Association (2012), smart growth can be defined as an urban development approach ‘which supports choice and opportunity by promoting efficient and sustainable land development, incorporates redevelopment patterns that optimise prior infrastructure investments, and consumes less land that is otherwise available for agriculture, open space, natural systems, and rural lifestyles’.

3 Land-use mix refers to the employment mix (entropy). It uses eight employment categories to calculate the employment mix (entropy). These categories are retail, office, service, industrial, entertainment, education, health care and public administration.

4 Transit service frequency during evening peak refers to the aggregate frequency of transit service within 0.25 miles of block group boundary per hour during evening peak period.

References

Adkins, A, Makarewicz, C, Scanze, M, Ingram, M and Luhr, G (2017) Contextualizing walkability: do relationships between built environments and walking vary by socioeconomic context? Journal of the American Planning Association 83, 296314.CrossRefGoogle ScholarPubMed
Adorno, G, Fields, N, Cronley, C, Parekh, R and Magruder, K (2018) Ageing in a low-density urban city: transportation mobility as a social equity issue. Ageing & Society 38, 296320.CrossRefGoogle Scholar
Alsnih, R and Hensher, DA (2003) The mobility and accessibility expectations of seniors in an aging population. Transportation Research Part A: Policy and Practice 37, 903916.Google Scholar
American Community Survey (2020) S0101 | Age and Sex 2020: ACS 5-Year Estimates Subject Tables. Available at https://data.census.gov/cedsci/table?q=age&g=310XX00US18140&tid=ACSST5Y2020.S0101&moe=false.Google Scholar
American Planning Association (2012) APA Policy Guide on Smart Growth. Available at https://www.planning.org/policy/guides/adopted/smartgrowth.htm.Google Scholar
Ang, BH, Lee, SWH, Chen, WS, Yap, KK, Song, KP and Oxley, JA (2020) Retiring from driving and riding: be safe and not a risk. Journal of Transport & Health 18, 100874.CrossRefGoogle Scholar
Bai, X, Steiner, RL and Zhai, W (2021) Beyond neighborhood design: exploring the effects of smart growth on older adults’ travel behavior over time. Journal of Planning Education and Research. https://doi.org/10.1177/0739456X211020352.CrossRefGoogle Scholar
Banister, D and Bowling, A (2004) Quality of life for the elderly: the transport dimension. Transport Policy 11, 105115.CrossRefGoogle Scholar
Bardaka, E and Hersey, J (2019) Comparing the travel behavior of affordable and market-rate housing residents in the transit-rich neighborhoods of Denver, CO. Travel Behaviour and Society 15, 7487.CrossRefGoogle Scholar
Beckman, JD and Goulias, KG (2008) Immigration, residential location, car ownership, and commuting behavior: a multivariate latent class analysis from California. Transportation 35, 655671.CrossRefGoogle Scholar
Berg, J, Levin, L, Abramsson, M and Hagberg, JE (2015) ‘I want complete freedom’: car use and everyday mobility among the newly retired. European Transport Research Review 7, 31.CrossRefGoogle Scholar
Bigonnesse, C and Chaudhury, H (2019) The landscape of ‘aging in place’ in gerontology literature: emergence, theoretical perspectives, and influencing factors. Journal of Aging and Environment 34, 233251.CrossRefGoogle Scholar
Black, WR (2010) Sustainable Transportation: Problems and Solutions. New York, NY: The Guilford Press.Google Scholar
Blumenberg, E and Smart, M (2010) Getting by with a little help from my friends … and family: immigrants and carpooling. Transportation 37, 429446.CrossRefGoogle Scholar
Böcker, L, van Amen, P and Helbich, M (2017) Elderly travel frequencies and transport mode choices in Greater Rotterdam, the Netherlands. Transportation 44, 831852.CrossRefGoogle Scholar
Boschmann, EE and Brady, SA (2013) Travel behaviors, sustainable mobility, and transit-oriented developments: a travel counts analysis of older adults in the Denver, Colorado metropolitan area. Journal of Transport Geography 33, 111.CrossRefGoogle Scholar
Buehler, R (2011) Determinants of transport mode choice: a comparison of Germany and the USA. Journal of Transport Geography 19, 644657.CrossRefGoogle Scholar
Buehler, R and Pucher, J (2012) Walking and cycling in Western Europe and the United States. TR News, pp. 3442.Google Scholar
Burkhardt, JE (1999) Mobility changes: their nature, effects, and meaning for elders who reduce or cease driving. Transportation Research Record 1671, 1118.CrossRefGoogle Scholar
Cao, X, Mokhtarian, PL and Handy, SL (2009) Examining the impacts of residential self-selection on travel behaviour: a focus on empirical findings. Transport Reviews 29, 359395.CrossRefGoogle Scholar
Cao, X, Mokhtarian, PL and Handy, SL (2010) Neighborhood design and the accessibility of the elderly: an empirical analysis in northern California. International Journal of Sustainable Transportation 4, 347371.CrossRefGoogle Scholar
Cao, Q, Dabelko-Schoeny, HI, White, KM and Choi, MS (2019) Age-friendly communities and perceived disconnectedness: the role of built environment and social engagement. Journal of Aging and Health 32, 937948.CrossRefGoogle ScholarPubMed
Caragata, G (2021) Should I drive today? Development of a daily driving decisions model for older adults. Transportation Research Part F: Psychology and Behaviour 81, 158172.CrossRefGoogle Scholar
Centers for Disease Control and Prevention (2020) Older Adult Drivers. Available at https://www.cdc.gov/motorvehiclesafety/older_adult_drivers/index.html.Google Scholar
Cerin, E, Lee, K, Barnett, A, Sit, CHP, Cheung, M, Chan, W and Johnston, JM (2013) Walking for transportation in Hong Kong Chinese urban elders: a cross-sectional study on what destinations matter and when. International Journal of Behavioral Nutrition and Physical Activity 10, 78.CrossRefGoogle Scholar
Cerin, E, Nathan, A, Cauwenberg, J, Van, Barnett, DW and Barnett, A (2017) The neighbourhood physical environment and active travel in older adults: a systematic review and meta-analysis. International Journal of Behavioral Nutrition and Physical Activity 14, 15.CrossRefGoogle ScholarPubMed
Cheng, S and Long, JS (2007) Testing for IIA in the multinomial logit model. Sociological Methods & Research 35, 583600.CrossRefGoogle Scholar
Cheng, L, Chen, X, Yang, S, Cao, Z, De Vos, J and Witlox, F (2019 a) Active travel for active ageing in China: the role of built environment. Journal of Transport Geography 76, 142152.CrossRefGoogle Scholar
Cheng, L, De Vos, J, Shi, K, Yang, M, Chen, X and Witlox, F (2019 b) Do residential location effects on travel behavior differ between the elderly and younger adults? Transportation Research Part D: Transport and Environment 73, 367380.CrossRefGoogle Scholar
Chihuri, S, Mielenz, TJ, DiMaggio, CJ, Betz, ME, DiGuiseppi, C, Jones, VC and Li, G (2016) Driving cessation and health outcomes in older adults. Journal of the American Geriatrics Society 64, 332341.CrossRefGoogle ScholarPubMed
Choi, M, Adams, KB and Kahana, E (2012) The impact of transportation support on driving cessation among community-dwelling older adults. Journals of Gerontology: Series B 67, 392400.CrossRefGoogle ScholarPubMed
Chudyk, AM, Winters, M, Moniruzzaman, M, Ashe, MC, Gould, JS and McKay, H (2015) Destinations matter: the association between where older adults live and their travel behavior. Journal of Transport & Health 2, 5057.CrossRefGoogle ScholarPubMed
Colby, SL and Ortman, JM (2015) Projections of the Size and Composition of the U.S. Population: 2014 to 2060. US Census Bureau. Available at https://www.census.gov/content/dam/Census/library/publications/2015/demo/p25-1143.pdf.Google Scholar
Collia, DV, Sharp, J and Giesbrecht, L (2003) The 2001 National Household Travel Survey: a look into the travel patterns of older Americans. Journal of Safety Research 34, 461470.CrossRefGoogle ScholarPubMed
Craney, TA and Surles, JG (2002) Model-dependent variance inflation factor cutoff values. Quality Engineering 14, 391403.CrossRefGoogle Scholar
Dabelko-Schoeny, H, Fields, NL, White, K, Sheldon, M, Ravi, K, Robinson, SR, Murphy, IE and Jennings, C (2020) Using community-based participatory research strategies in age-friendly communities to solve mobility challenges. Journal of Gerontological Social Work 63, 117.CrossRefGoogle ScholarPubMed
Dabelko-Schoeny, H, Maleku, A, Cao, Q, White, K and Ozbilen, B (2021) ‘We want to go, but there are no options’: exploring barriers and facilitators of transportation among diverse older adults. Journal of Transport and Health 20, 100994.CrossRefGoogle Scholar
Davey, JA (2007) Older people and transport: coping without a car. Ageing & Society 27, 4965.CrossRefGoogle Scholar
Dill, J, Mohr, C and Ma, L (2014) How can psychological theory help cities increase walking and bicycling? Journal of the American Planning Association 80, 3651.CrossRefGoogle Scholar
Ding, D, Sallis, JF, Norman, GJ, Frank, LD, Saelens, BE, Kerr, J, Conway, TL, Cain, K, Hovell, MF, Hofstetter, CR and King, AC (2014) Neighborhood environment and physical activity among older adults: do the relationships differ by driving status? Journal of Aging and Physical Activity 22, 421431.CrossRefGoogle ScholarPubMed
Dumbaugh, E (2016) Designing communities to enhance the safety and mobility of older adults: a universal approach. Journal of Planning Literature 23, 1736.CrossRefGoogle Scholar
Dutzik, T and Inglis, J (2014) Millennials in Motion: Changing Travel Habits of Young Americans and the Implications for Public Policy. US PIRG Education Fund. Available at https://uspirg.org/sites/pirg/files/reports/Millennials in Motion USPIRG.pdf.Google Scholar
Ewing, R and Cervero, R (2010) Travel and the built environment. Journal of the American Planning Association 76, 265294.CrossRefGoogle Scholar
Federal Transit Administration (2019) Transit-oriented Development. US Department of Transportation. Available at https://www.transit.dot.gov/TOD.Google Scholar
Figueroa, MJ, Nielsen, TAS and Siren, A (2014) Comparing urban form correlations of the travel patterns of older and younger adults. Transport Policy 35, 1020.CrossRefGoogle Scholar
Fonda, SJ, Wallace, RB and Herzog, AR (2001) Changes in driving patterns and worsening depressive symptoms among older adults. Journals of Gerontology: Series B 56, 343351.CrossRefGoogle ScholarPubMed
Fordham, L, Grisé, E and El-Geneidy, A (2017) When I'm 64: assessing generational differences in public transit use of seniors in Montreal, Quebec, Canada. Transportation Research Record 2651, 3141.CrossRefGoogle Scholar
Forsyth, A, Molinsky, J and Kan, HY (2019) Improving housing and neighborhoods for the vulnerable: older people, small households, urban design, and planning. Urban Design International 24, 171186.CrossRefGoogle Scholar
Frank, L, Kerr, J, Rosenberg, D and King, A (2010) Healthy aging and where you live: community design relationships with physical activity and body weight in older Americans. Journal of Physical Activity and Health 7, 8290.CrossRefGoogle ScholarPubMed
Freeman, L, Neckerman, K, Schwartz-Soicher, O, Quinn, J, Richards, C, Bader, MDM, Lovasi, G, Jack, D, Weiss, C, Konty, K, Arno, P, Viola, D, Kerker, B and Rundle, AG (2012) Neighborhood walkability and active travel (walking and cycling) in New York City. Journal of Urban Health: Bulletin of the New York Academy of Medicine 90, 575585.CrossRefGoogle Scholar
Greene, WH (2018) Econometric Analysis, 8th Edn. New York, NY: Pearson.Google Scholar
Gudmundsson, H, Hall, RP, Marsden, G and Zietsman, J (2016) Sustainable Transportation: Indicators, Frameworks, and Performance Management. Heidelberg, Germany: Springer-Verlag.CrossRefGoogle Scholar
Hahn, J, Kim, H, Kim, J and Ulfarsson, GF (2016) Trip making of older adults in Seoul: differences in effects of personal and household characteristics by age group and trip purpose. Journal of Transport Geography 57, 5562.CrossRefGoogle Scholar
Hamre, A and Buehler, R (2014) Commuter mode choice and free car parking, public transportation benefits, showers/lockers, and bike parking at work: evidence from the Washington, DC region. Journal of Public Transportation 17, 6791.CrossRefGoogle Scholar
Handy, SL, Boarnet, MG, Ewing, R and Killingsworth, RE (2002) How the built environment affects physical activity: views from urban planning. American Journal of Preventive Medicine 23, 6473.CrossRefGoogle ScholarPubMed
Hausman, J and McFadden, D (1984) Specification tests for the multinomial logit model. Econometrica 52, 12191240.CrossRefGoogle Scholar
Heinrich, CJ and Wenger, JB (2002) The economic contributions of James J. Heckman and Daniel L. McFadden. Review of Political Economy 14, 6989.CrossRefGoogle Scholar
Hess, DB (2009) Access to public transit and its influence on ridership for older adults in two U.S. cities. Journal of Transport and Land Use 2, 327.CrossRefGoogle Scholar
Hess, DB and Russell, JK (2012) Influence of built environment and transportation access on body mass index of older adults: survey results from Erie County, New York. Transport Policy 20, 128137.CrossRefGoogle Scholar
Hjorthol, R (2013) Transport resources, mobility and unmet transport needs in old age. Ageing & Society 33, 11901211.CrossRefGoogle Scholar
Hjorthol, RJ, Levin, L and Sirén, A (2010) Mobility in different generations of older persons. The development of daily travel in different cohorts in Denmark, Norway and Sweden. Journal of Transport Geography 18, 624633.CrossRefGoogle Scholar
Kan, HY, Forsyth, A and Molinsky, J (2020) Measuring the built environment for aging in place: a review of neighborhood audit tools. Journal of Planning Literature 35, 180194.CrossRefGoogle Scholar
Kemperman, A and Timmermans, H (2009) Influences of built environment on walking and cycling by latent segments of aging population. Transportation Research Record 2134, 19.CrossRefGoogle Scholar
Kerr, J, Rosenberg, D and Frank, L (2012) The role of the built environment in healthy aging: community design, physical activity, and health among older adults. Journal of Planning Literature 27, 4360.CrossRefGoogle Scholar
Kim, S and Ulfarsson, GF (2004) Travel mode choice of the elderly: effects of personal, household, neighborhood, and trip characteristics. Transportation Research Record 1894, 117126.CrossRefGoogle Scholar
Klicnik, I and Dogra, S (2019) Perspectives on active transportation in a mid-sized age-friendly city: ‘you stay home’. International Journal of Environmental Research and Public Health 16, 4916.CrossRefGoogle Scholar
Lawton, MP (1982) Competence, environmental press, and the adaptation of older people. In Lawton, MP, Windley, PG and Byerts, TO (eds), Aging and the Environment. New York, NY: Springer, pp. 3359.Google Scholar
Lawton, MP (1989) Three functions of the residential environment. Journal of Housing for the Elderly 5, 3550.CrossRefGoogle Scholar
Lawton, MP and Nahemow, L (1973) Ecology and the aging process. In Eisdorfer, C and Lawton, MP (eds), The Psychology of Adult Development and Aging. Washington, DC: American Psychological Association, pp. 619674.CrossRefGoogle Scholar
Lehning, A, Kim, K, Smith, R and Choi, M (2018) Does economic vulnerability moderate the association between transportation mode and social activity restrictions in later life? Ageing & Society 38, 20412060.CrossRefGoogle Scholar
Li, S (2020) Living environment, mobility, and wellbeing among seniors in the United States: a new interdisciplinary dialogue. Journal of Planning Literature 35, 298314.CrossRefGoogle Scholar
Li, M and Tilahun, N (2017) Time use, disability, and mobility of older Americans. Transportation Research Record 2650, 5865.CrossRefGoogle Scholar
Litman, T (1999) Exploring the paradigm shifts needed to reconcile transportation and sustainability objectives. Transportation Research Record 1670, 812.CrossRefGoogle Scholar
Litman, T (2002) The Costs of Automobile Dependency and the Benefits of Balanced Transportation. Victoria, BC: Victoria Transport Policy Institute.Google Scholar
Litman, T (2019) Well Measured: Developing Indicators for Sustainable and Livable Transport Planning. Available at https://www.vtpi.org/wellmeas.pdf.Google Scholar
Long, JS and Freese, J (2005) Regression Models for Categorical Dependent Variables Using Stata, 2nd Edn. College Station, TX: Stata Press.Google Scholar
Loukaitou-Sideris, A, Levy-Storms, L, Chen, L and Brozen, M (2016) Parks for an aging population: needs and preferences of low-income seniors in Los Angeles. Journal of the American Planning Association 82, 236251.CrossRefGoogle Scholar
Loukaitou-Sideris, A, Wachs, M and Pinski, M (2019) Toward a richer picture of the mobility needs of older Americans. Journal of the American Planning Association 85, 482500.CrossRefGoogle Scholar
Lyu, Y and Forsyth, A (2022) Planning, aging, and loneliness: reviewing evidence about built environment effects. Journal of Planning Literature 37, 2848.CrossRefGoogle Scholar
Macleod, KE, Thorhauge, M, Villalobos, V, Van Meijgaard, J, Karriker-Ja, KJ, Kelley-Baker, T and Ragland, DR (2020) To drive or not to drive? A study of travel behavior for a recent drinking occasion. Travel Behaviour and Society 20, 7482.CrossRefGoogle ScholarPubMed
Marottoli, RA, de Leon, CFM, Glass, TA, Williams, CS, Cooney, LM, Berkman, LF and Tinetti, ME (1997) Driving cessation and increased depressive symptoms: prospective evidence from the New Haven EPESE. Journal of the American Geriatrics Society 45, 202206.CrossRefGoogle ScholarPubMed
McFadden, D (1974) The measurement of urban travel demand. Journal of Public Economics 3, 303328.CrossRefGoogle Scholar
McFadden, D (1987) Regression-based specification tests for the multinomial logit model. Journal of Econometrics 34, 6382.CrossRefGoogle Scholar
McFadden, D, Tye, W and Train, K (1976) An application of diagnostic tests for the irrelevant alternatives property of the multinomial logit model. Transportation Research Record 637, 3946.Google Scholar
McMillan, A and Lee, S (2017) Smart growth characteristics and the spatial pattern of multifamily housing in US metropolitan areas. Urban Studies 54, 35003523.CrossRefGoogle Scholar
Metz, D (2003) Transport policy for an ageing population. Transport Reviews 23, 375386.CrossRefGoogle Scholar
Mezuk, B and Rebok, GW (2008) Social integration and social support among older adults following driving cessation. Journals of Gerontology: Series B 63, 298303.CrossRefGoogle ScholarPubMed
Mid-Ohio Regional Planning Commission (MORPC) (2017) Age-Friendly Columbus Findings Report. Columbus, OH: MORPC.Google Scholar
Mifsud, D, Attard, M and Ison, S (2017) To drive or to use the bus? An exploratory study of older people in Malta. Journal of Transport Geography 64, 2332.CrossRefGoogle Scholar
Morikawa, T (1994) Correcting state dependence and serial correlation in the RP/SP combined estimation method. Transportation 21, 153165.CrossRefGoogle Scholar
Nahemow, L and Lawton, MP (1973) Toward an ecological theory of adaptation and aging. In Preiser, WFE (ed). Environmental Design Research. Stroudsburg, PA: Dowden Hutchinson & Ross, pp. 2432.Google Scholar
Naumann, RB, Dellinger, AM, Anderson, ML, Bonomi, AE, Rivara, FP and Thompson, RS (2009) Preferred modes of travel among older adults: what factors affect the choice to walk instead of drive? Journal of Safety Research 40, 395398.CrossRefGoogle ScholarPubMed
Nobis, C and Lenz, B (2005) Gender differences in travel patterns: role of employment status and household structure. In Research on Women's Issues in Transportation, Vol. 2, Technical Papers. Washington, DC: The National Academies Press, pp. 114123.Google Scholar
Nordbakke, S and Schwanen, T (2015) Transport, unmet activity needs and wellbeing in later life: exploring the links. Transportation 42, 11291151.CrossRefGoogle Scholar
Novek, S and Menec, VH (2014) Older adults’ perceptions of age-friendly communities in Canada: a photovoice study. Ageing & Society 34, 10521072.CrossRefGoogle Scholar
Ortman, JM, Velkoff, VA and Hogan, H (2014) An Aging Nation: The Older Population in the United States. Washington, DC: U.S. Census Bureau.Google Scholar
Ragland, DR, MacLeod, KE, McMillan, T, Doggett, S and Felschundneff, G (2019) Assessing and Addressing the Mobility Needs of an Aging Population. Berkeley, CA: The University of California Institute of Transportation Studies.Google Scholar
Rahman, MM, Strawderman, L, Adams-Price, C and Turner, JJ (2016) Transportation alternative preferences of the aging population. Travel Behaviour and Society 4, 2228.CrossRefGoogle Scholar
Rahman, MM, Deb, S, Strawderman, L, Smith, B and Burch, R (2019) Evaluation of transportation alternatives for aging population in the era of self-driving vehicles. IATSS Research 44, 3035.CrossRefGoogle Scholar
Reher, D and Requena, M (2018) Living alone in later life: a global perspective. Population and Development Review 44, 427454.CrossRefGoogle Scholar
Rosenbloom, S (2001) Sustainability and automobility among the elderly: an international assessment. Transportation 28, 375408.CrossRefGoogle Scholar
Rosenbloom, S (2009) Meeting transportation needs in an aging-friendly community. Generations: Journal of the American Society on Aging 33, 3343.Google Scholar
Rosenbloom, S and Waldorf, B (2001) Older travelers: does place or race make a difference? Transportation Research E-Circular, E-C026, 103120.Google Scholar
Schiller, PL, Bruun, EC and Kenworthy, JR (2010) An Introduction to Sustainable Transportation: Policy, Planning and Implementation. London: Earthscan.CrossRefGoogle Scholar
Schmöcker, JD, Quddus, MA, Noland, RB and Bell, MGH (2008) Mode choice of older and disabled people: a case study of shopping trips in London. Journal of Transport Geography 16, 257267.CrossRefGoogle Scholar
Schouten, A, Blumenberg, E, Wachs, M and King, H (2021) Keys to the car: driving cessation and residential location among older adults. Journal of the American Planning Association 88, 314.CrossRefGoogle Scholar
Schwanen, T, Dijst, M and Dieleman, FM (2001) Leisure trips of senior citizens: determinants of modal choice. Tijdschrift voor Economische en Sociale Geografie 92, 347360.CrossRefGoogle Scholar
Seo, SN (2016) Microbehavioral Econometric Methods. Seoul: Academic Press.Google Scholar
Shen, S, Koech, W, Feng, J, Rice, TM and Zhu, M (2017) A cross-sectional study of travel patterns of older adults in the USA during 2015: implications for mobility and traffic safety. BMJ Open 7, e015780.CrossRefGoogle ScholarPubMed
Shrestha, BP, Millonig, A, Hounsell, NB and McDonald, M (2017) Review of public transport needs of older people in European context. Journal of Population Ageing 10, 343361.CrossRefGoogle ScholarPubMed
Simonsick, EM, Guralnik, JM, Volpato, S, Balfour, J and Fried, LP (2005) Just get out the door! Importance of walking outside the home for maintaining mobility: findings from the Women's Health and Aging Study. Journal of the American Geriatrics Society 53, 198203.CrossRefGoogle ScholarPubMed
Small, KA and Hsiao, C (1985) Multinomial logit specification tests. International Economic Review 26, 619627.CrossRefGoogle Scholar
Statista (2021) Population of the Largest Metropolitan Areas in the U.S. as of 2019. Available at https://www.statista.com/statistics/183600/population-of-metropolitan-areas-in-the-us/.Google Scholar
Steg, L and Gifford, R (2005) Sustainable transportation and quality of life. Journal of Transport Geography 13, 5969.CrossRefGoogle Scholar
Szeto, WY, Yang, L, Wong, RCP, Li, YC and Wong, SC (2017) Spatio-temporal travel characteristics of the elderly in an ageing society. Travel Behaviour and Society 9, 1020.CrossRefGoogle Scholar
Tabachnick, BG and Fidell, LS (2012) Using Multivariate Statistics, 6th Edn. Boston, MA: Pearson.Google Scholar
The Administration for Community Living (2018) 2018 Profile of Older Americans. Washington, DC: The Administration for Community Living.Google Scholar
Tsai, L, Rantakokko, M, Portegijs, E, Viljanen, A, Saajanaho, M, Eronen, J and Rantanen, T (2013) Environmental mobility barriers and walking for errands among older people who live alone vs with others. BMC Public Health 13, 1054.CrossRefGoogle Scholar
Ulfarsson, GF and Kim, S (2019) Ageing and transportation. In Urban Environments for Healthy Ageing: A Global Perspective. New York, NY: Routledge, pp. 103116.CrossRefGoogle Scholar
United Nations (2019) World Population Ageing 2019: Highlights. New York, NY: United Nations.Google Scholar
US Census Bureau (2021) QuickFacts – Columbus City, Ohio. Available at https://www.census.gov/quickfacts/columbuscityohio.Google Scholar
US Environmental Protection Agency (2014) Smart Location Database Technical Documentation and User Guide. Available at https://www.epa.gov/smartgrowth/smart-location-database-technical-documentation-and-user-guide.Google Scholar
Van den Berg, P, Arentze, T and Timmermans, H (2011) Estimating social travel demand of senior citizens in the Netherlands. Journal of Transport Geography 19, 323331.CrossRefGoogle Scholar
VERA Institute of Justice (2017) Profile of the Foreign-born Population in Columbus, Ohio. Available at https://www.vera.org/downloads/publications/profile-foreign-born-population-columbus.pdf.Google Scholar
Voss, C, Sims-Gould, J, Ashe, MC, Mckay, HA, Pugh, C and Winters, M (2016) Public transit use and physical activity in community-dwelling older adults: combining GPS and accelerometry to assess transportation-related physical activity. Journal of Transport & Health 3, 191199.CrossRefGoogle Scholar
Vyas, G, Famili, P, Vovsha, P, Fay, D, Kulshrestha, A, Giaimo, G and Anderson, R (2019) Incorporating features of autonomous vehicles in activity-based travel demand model for Columbus, OH. Transportation 46, 20812102.CrossRefGoogle Scholar
Wacker, RR and Roberto, KA (2014) Community Resources for Older Adults: Programs and Services in an Era of Change, 4th edn. Thousand Oaks, CA: Sage.Google Scholar
Wahl, HW, Iwarsson, S and Oswald, F (2012) Aging well and the environment: toward an integrative model and research agenda for the future. The Gerontologist 52, 306316.CrossRefGoogle ScholarPubMed
Wang, C and Chen, N (2015) A GIS-based spatial statistical approach to modeling job accessibility by transportation mode: case study of Columbus, Ohio. Journal of Transport Geography 45, 111.CrossRefGoogle Scholar
Willis, KG (2014) The use of stated preference methods to value cultural heritage. In VA Ginsburgh and D Throsby, eds., Handbook of the Economics of Art and Culture. Amsterdam: Elsevier, pp. 145181.Google Scholar
Wood, BS and Horner, MW (2019) Aging in activity spaces: how does individual accessibility compare across age cohorts? The Professional Geographer 71, 114.CrossRefGoogle Scholar
World Health Organization (WHO) (2002) Active Ageing: A Policy Framework. Geneva: WHO.Google Scholar
World Health Organization (WHO) (2007) Global Age-friendly Cities: A Guide. Geneva: WHO.Google Scholar
Yang, Y, Xu, Y, Rodriguez, DA, Michael, Y and Zhang, H (2018) Active travel, public transportation use, and daily transport among older adults: the association of built environment. Journal of Transport and Health 9, 288298.CrossRefGoogle Scholar
Figure 0

Table 1. Summary table for determinants of older adults' travel behaviour

Figure 1

Table 2. Older adults' characteristics and built environment measures, descriptive statistics

Figure 2

Table 3. Multinomial logistic regression model results