Factors Associated with Outdoor Winter Walking in Older Adults: A Scoping Review

Abstract The purpose of this study was to identify internal and external factors associated with outdoor winter walking in older adults. In this scoping review, 12 databases were searched. Inclusion criteria included English language, focus on adults 65 years of age or older, and evaluation of factors associated with outdoor winter walking. Two authors screened titles/abstracts and full text. Conflicts were resolved by consensus. Data were extracted, organized into tables, and summarized as pertaining to barriers/facilitators and internal/external factors associated with outdoor winter walking. A total of 6,843 articles were identified, 1,898 duplicates were removed, 4,789 were excluded during title/abstract screening, and 148 were excluded during full-text review. Eight studies were included. Four categories of factors affecting outdoor winter walking in older adults were identified: adverse weather conditions, physical environment, physical function, and perceptions relating to winter walking conditions. Rehabilitation and exercise professionals can use the results to educate their clients and implement the facilitators of and alternatives and solutions to barriers to outdoor winter walking.


Importance of Physical Activity in Aging
The importance of regular physical activity as a person ages is well known; it is associated with improved health, function, and quality of life (Anokye, Trueman, Green, Pavey, & Taylor, 2012; Older adults may choose to walk for many reasons. Krogstad, Hjorthol, & Tennoy (2015) identified three main categorical reasons why older adults may choose to walk: to accomplish necessary activities, because they have the ability to, and because they find it to be an enjoyable leisure activity. As a result, walking plays a role in accomplishing daily living activities, expressing and maintaining independence, and improving physical and mental functioning for older adults (Krogstad et al., 2015). Warm summer months evidently provide good conditions for walking, but seasonal changes can bring many environmental challenges to winter walking for older adults. In the winter months social isolation can occur if walking outdoors becomes limited, as older adults use walking as a means of transportation (Krogstad et al., 2015;Choi & DiNitto, 2016). An older adult may normally walk to a local drop-in centre once a week, but this may be reduced or limited as a result of factors affecting their ability to walk outdoors in the winter. This may decrease their engagement and participation in their social community.
Seasonal changes impact gait speed (Montufar, Arango, Porter., & Nakagawa, 2007). Assessed in real life conditions, walking speeds of both younger and older adults were reduced in winter compared with summer (Montufar et al., 2007). This is an important factor, as increased gait speed in older adults is related to survival (Studenski et al., 2011). A reduction in gait speed is a risk factor for a variety of outcomes such as falls, cognitive impairment, institutionalisation, and death (Abellan van Kan et al., 2009). Understanding more clearly the factors facilitating or preventing outdoor winter walking among older adults could help develop better strategies to deliver the benefits that walking, as a type of physical activity, can provide during the winter months. Factors may be described as internal to the person, such as physical function, perceptions, or beliefs; or external, which refers to factors outside of the person such as neighbourhood or environment.

Winter and Reduced Activity
The geographical lens of this scoping review is the climate conditions experienced by residents of Winnipeg, Manitoba in Canada. Canadian winters are some of the most severe in the world (Government of Canada, 2013). For example, in Winnipeg, Manitoba, the average ambient temperature between December and February is -14.2 degrees Celsius and the extreme minimum temperature is -45.0 degrees Celsius (Government of Canada, 2019a). Wind chill greatly amplifies the cold weather that Canadians experience. In Winnipeg, the average number of days per year with a wind chill below -20 degrees Celsius is 85.1, with 43.9 of these days being below -30 degrees Celsius and 12.1 of these days being below -40 degrees Celsius (Government of Canada, 2019a). The snow, ice, wind, decreased daylight hours, and cold weather experienced during Canadian winters can make outdoor winter walking challenging for older adults.
Environmental factors have been identified as both facilitators and barriers in relation to activity and participation (Schneidert, Hurst, Miller, Ustun, 2003). For example, snow-covered sidewalks could be a barrier to walking outdoors, whereas snow clearing and sanding could be a facilitator of outdoor walking. The model developed by Patla and Shumway-Cook (1999) identifies ambient conditions and terrain characteristics as two of eight factors that may limit mobility in older adults. They discuss two aspects in regard to ambient conditions: weather conditions and light level. In the winter months, Canada has decreased daylight hours and snowfall that can reduce visibility, wind that can affect balance, and cold weather, which requires warmer, bulkier clothing (Government of Canada, 2013;2019a, 2019b. Terrain characteristics associated with winter, such as ice and snow, add to the complexity of winter walking for older adults in addition to the usual obstacles such as curbs and inclines (Patla & Shumway-Cook, 1999). Environmental factors can pose challenges to mobility for older adults, and if they have difficulty overcoming these challenges, inactivity can result (Patla & Shumway-Cook, 1999).
In order to address the barriers to walking outdoors in winter and the resultant influence on physical activity, mobility, and safety, we need to have a comprehensive understanding of the associated factors that appear in the literature. As a result of the review, we hope to develop a basis for further research and strategies that will identify barriers as well as the factors that could facilitate participation in safe winter walking.

Objective
The purpose of this study was to identify the internal and external factors associated with outdoor winter walking in older adults. The protocol for this scoping review was published in Open Science Framework (https://osf.io/xebgd/).

Approach
Scoping reviews provide an avenue to comprehensively map areas of interest and do so in a clear and transparent way so that it may be replicated by others (Arksey & O'Malley, 2005). By consolidating many study designs, the scoping review has the advantage of retrieving all literature that is pertinent to the research (Arksey & O'Malley, 2005). Through this method of knowledge synthesis, scoping reviews can summarize research, identify current gaps, inform future directions, and be a source of information for programs and policies (Arksey & O'Malley, 2005). Our review followed the foundational steps of the methodological framework produced by Arksey and O'Malley (2005) in conducting a scoping review as well as following the recommendations proposed by Levac, Colquhoun, and O'Brien (2010) to enhance each stage. Covidence was used to manage search results, develop a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart, and manage article screening and data extraction (Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia, available at www.covidence.org). The PRISMA extension for scoping reviews checklist was used in planning the review (www.equator-network.org/reporting-guidelines/prisma-scr/).
The research topic was created by the 11 members of the "Winter-Walk" team. This team included: seven researchers encompassing professional backgrounds of physical therapy, occupational therapy, kinesiology, geography, and library sciences. The team also included two older adult members of the community from the Manitoba Association of Seniors Centres and Active Aging in Manitoba, and two graduate student trainees. These members had diverse and unique experiences in areas of older adults, physical activity, and winter walking. The team had two, 5-hour meetings in which there were presentations from all the team members, roundtable discussions, and two rounds of voting to determine the team research priorities related to older adults and outdoor winter walking. The team identified and voted on 14 topics. The topic with the highest priority was "identifying internal and external factors associated with outdoor winter walking for older adults". This review will add to the literature of outdoor winter walking barriers and facilitators for older adults and assist as a basis for future winter walking studies.

Study Eligibility
The scoping review included all study designs and all years published within the utilized databases. To be included, studies had to evaluate one or more factors associated with outdoor winter walking and be written in the English language, and the study population had to be 65 years of age or older or contain a sub-group analysis of adults 65 years of age or older. Studies that contain mixed ages were included if 80 per cent or more of the study sample was over 65 years old.
Articles were excluded if participants did not walk outdoors (e.g., used wheeled mobility) or if the study population examined was homogenous of a specific health condition.
An eligibility checklist was developed based on these criteria within Covidence and applied to the articles during the screening process to exclude those that did not meet the criteria for inclusion.

Search Strategy
The search was conducted using the subject headings: Geriatrics, Aging, Aged, Retirement, Walking, and Cold Temperature; along with appropriate keywords and free-text words. Databases The reference list of included articles was examined to identify other relevant literature. Deduplication and references were managed in EndNote. Further deduplication was additionally completed in Covidence. The initial search strategy was conducted in Ovid MEDLINE and then applied within the other databases. See Table 1 for the full search strategy conducted on Ovid MEDLINE.

Study Selection
The health sciences librarian (H.L.) aided in the development of the search and offered search strategy suggestions. After conducting the search, titles and abstracts were screened and then selected using the inclusion and exclusion criteria checklist, which is available in Covidence. This involved each title and abstract being assessed by two authors for eligibility (K.B., M.J., K.S., E.S., S.M.). Lack of agreement on titles and abstracts at this stage resulted in carrying the study forward to the next stage. Two authors then applied the same inclusion and exclusion criteria to each full text article that passed the first stage. Lack of agreement between reviewers resulted in further analysing the article until a consensus was reached. Once this was completed, the reference lists of included studies were analysed to determine if any additional studies met the inclusion and exclusion criteria that were not covered by the databases. Covidence was used for citation management and the screening process.

Data Extraction and Data Charting
Data was extracted in Covidence independently by the first four authors, ensuring that at least two authors addressed each source and compared results. After extracting data from two articles, minor revisions were made to the extraction chart format based on team member experience and discussion. In cases for which there was a lack of agreement, the two individuals who had extracted the data of the study met and formed a consensus on the final data extracted. Data were exported from Covidence to Microsoft Excel. Tables were created from the following information that was extracted: author, title, publication year, country, study design, study purpose, age, sex, health conditions or physical function, number of participants in the study, statistical methods used, outcome measures used, how the study measured winter walking, the factors described in the study, and the factors that are associated with winter outdoor walking or are predictors of winter outdoor walking.

Synthesis of Results
The first four authors and R.B. of the "Winter-Walk" team analysed the results of the final studies included. For each study, the factors associated with outdoor winter walking in older adults were classified as either facilitators or barriers and then categorized as either internal or external in nature. Furthermore, the factors identified were classified into body functions, activity, or participation, according to the International Classification of Functioning Disability and Health (ICF) (World Health Organization, 2002). Once the factors affecting outdoor winter walking from all the studies were organized together in tables, factors were grouped into categories.
A stage of consultation occurred to offer additional knowledge and perspectives, as well as relevance and applicability to the scoping review (Levac et al., 2010). A consultation occurred with the two older adult community members of the "Winter-Walk" team, who represented two community organizations which support senior centres and active aging. Scoping review results were shared with these team members, and their impressions of the results and next steps are included in the discussion.

Results
The database searches resulted in 6,843 studies, with 1,893 duplicates removed via EndNote. The remaining 4,950 studies were imported to Covidence for screening. Further deduplication occurred in Covidence where five more studies were removed. After reviewing the titles and abstracts of 4945 studies, 4,789 were excluded for reasons including being unrelated to the review topic or because the study focused on a specific health condition. This resulted in 156 full-text studies that were assessed for eligibility. We excluded 148 of these, with the main reason for exclusion being that the study did not evaluate factors associated with outdoor winter walking. Other reasons for exclusion can be seen in Figure 1. Following full-text screening, eight studies were included in the scoping review.

Study Characteristics
As seen in Table 2 Kyle Bergen et al. ("65 and up" or "65 and older" or "65 or older" or "65þ" or "65 plus" or "65plus" or "sixty-five plus" or "sixty five and up" or "sixty five and older" or "sixty five or older").mp. 8 ((old or older) adj5 (person* or people* or female* or woman or man or men or males*) Outdoor Winter Walking in Older Adults 319 conducted in Finland , Iceland (Bjornsdottir, Arnadottir, & Halldorsdottir, 2012), Sweden (Wennberg, Stahl, & Hyden 2009), the United States (Finlay, 2018) and England (Wu, Luben, & Jones, 2017). Average temperature in the month of January ranged from -15 to þ5 degrees Celsius at the included study sites. In total, these studies collectively included 5,484 older adults with ages ranging from 49 to 100. On average, 70 per cent of the participants were female. As is common in sport medicine literature (Parsons, Coen, & Bekker, 2021), the terms "sex" and "gender" were used interchangeably with male/female and man/woman in a number of the included studies. In most of the eight included studies, the description appeared to be related to the biological description of males and females, as opposed to gender identities. How the determination of male/female or man/woman was made was not clear in all studies. Therefore, we chose to use the term "sex" along with male/female in the description of study participants, as described elsewhere (Government of Canada, Canadian Institutes of Health Research, 2015).
Although our inclusion criterion was age 65 or older, two studies Wu et al., 2017) included participants outside the age range of this review, but either completed subgroup analysis or had more than 80 per cent of participants within the age range criterion for inclusion. Health conditions and physical function of the participants were described very differently across studies. For example: by walking aid use , physical function assessment Sakari et al., 2017), number of chronic conditions Sakari et al., 2017 or self-rated health . The study designs encompassed qualitative approaches including surveys , interviews (Finlay, 2018;Schmidt et al., 2016;Bjornsdottir et al., 2012) and focus groups . Other study types included cross-sectional studies Sakari et al., 2017), exploratory mixed methods ) and a prospective cohort study .
The included studies utilized multiple methods to identify outdoor winter walking including: surveys/questionnaires Li et al., 2013;Wu et al., 2017), accelerometry , interviews Finlay, 2018;Sakari et al., 2017;Schmidt et al., 2016;Wennberg et al., 2009), focus groups ) and participant observation (Finlay, 2018;Wennberg et al., 2009). Weather conditions and weather data were also collected in some studies Wu et al., 2017). Multiple studies used a combination of methods; for example, the Neighbourhood Environmental Walkability Survey was utilized by Clarke et al. (2017), the results of which were connected to archived weather information. Wennberg

320
Kyle Bergen et al.  Wu et al., 2017), one study included both urban and rural settings , whereas the remainder of studies focused on the urban setting. Within these eight studies, factors that were associated with outdoor winter walking were described as barriers or facilitators ( Table 3). The barriers and facilitators were then further divided into internal or external factors and categorized using the ICF (Table 4). Internal factors were considered to be person-related, such as one's physical capacities, perceptions, beliefs, and fears, whereas external factors referred to those factors apart from the person such as weather, ice, snow, neighbourhood, and environment. Factors were aligned with the ICF classifications of body function, activity, participation, and the environment. Common categories emerged in the results that allowed an understanding of the categories of factors that were relevant to the study population, both internal and external, that impacted walking outdoors in winter. These categories were physical function, perceptions and fears relating to winter walking conditions (internal factors), and adverse weather conditions and physical environmental conditions (external factors). See Table 4.

Discussion
The objective of this scoping review was to identify internal and external factors associated with outdoor winter walking in older adults. The eight included studies revealed many different factors that were described as external or internal barriers or facilitators. These factors were further summarized into four categories: adverse weather conditions (external), physical environmental conditions (external), physical function (internal), and perceptions and fears relating to winter walking conditions (internal).
The first category that emerged was the impact that weather conditions had on outdoor winter walking in older adults. All of the included studies described at least one weather-related factor that was described as a barrier to or facilitator of outdoor winter walking. These included conditions producing ice, slush, and strong winds Schmidt et al., 2016), cold temperatures , and higher levels of snow  or precipitation , which were all negatively associated with outdoor walking. These results were perhaps unsurprising, as it seems sensible that harsh weather would tend to decrease older adults' participation in outdoor walking, and this was confirmed by the results. These results appear to confirm that it is realistic to think that, as a general principle, categorically harsher winter days will result in less outdoor winter walking whereas more pleasant winter days will result in more outdoor winter walking. This appears to be consistent with literature that describes decreased gait speed (Montufar et al., 2007) and steps per day in winter (Kimura, Kobayashi, Nakayama, & Kakihana, 2015). Beyond examining weather variables in isolation, Clarke et al. (2017) examined the interplay among neighbourhood walkability, vehicle dependency, and weather conditions. Walkable neighbourhoods encompassed those with shorter block lengths, a greater mix of land use, and more street connectivity . Snow was found to be negatively associated with outdoor walking in general but much more strongly so in car-dependent neighbourhoods. This implies that properly designed, built environments can attenuate some of the negative associations between winter weather and outdoor walking. These results highlight the important overlap between winter-related factors and the built, physical environment, pointing to the relevancy of considering how the physical environment an older adult lives in can affect their outdoor winter walking. This is exemplified by the second category.
The second category that emerged were factors related to the physical environment. These factors included conditions of the walking terrain, such as ice-covered surfaces Finlay, 2018;Li et al., 2013;Schmidt et al., 2016;Wennberg et al., 2009) and lack of snow removal (Finlay, 2018). Li et al. (2013) examined what specific areas of the physical environment were of most concern to outdoor walkers. The most problematic were sidewalks, followed by street crossings and curb ramps. In these three areas, commonly reported barriers were ice, snow, and slush accumulation that made walking difficult. These physical environmental barriers led to winter coping strategies that were explored further by Li et al. (2013) who found that 16 (59%) of the older adult participants coped with these winter challenges by going out less, and consequently reducing outdoor winter walking. Therefore, we see how the physical environment, as affected by winter weather, can be a barrier to outdoor walking.
The third category that emerged was participants' physical function acting as a barrier or facilitator to outdoor winter walking. Older adults with a higher physical function were more likely to walk outdoors in the winter than those with poorer physical function , as measured using the Short Physical Performance Battery (SPPB). Sakari et al. (2017) also measured physical function through assessment of the lower extremities using the SPPB, which examines functional strength through walking, sit-to-stand, and standing activities. Sakari et al. (2017) discussed that for older adults with poor lower extremity performance, fear stemming from safety issues was a major barrier. These barriers for those with intact lower extremity performance focused on aspects of outdoor walking such as the distance required to walk, and less so on fears or safety concerns . This finding indicated that older adults with poorer physical function may walk less, simply because of their perceptions of winter as a dangerous time to walk. Other functional limitations, including mobility, vision, and hearing impairments, were also found to be barriers to outdoor winter walking for older adults . Many of the aforementioned barriers to outdoor winter walking have also been identified as barriers to outdoor walking in general in a conceptual framework for an outdoor walking intervention, including various aspects of physical function as well as fear and self-efficacy (Salbach et al., 2019).
The fourth category that emerged was how participants perceived that environmental conditions influenced their decision to walk outdoors. Specifically, fear of falling was a consistent subject across many of the included studies, which led to avoidance of outdoor winter walking among older adults Finlay, 2018;Sakari et al., 2017;Schmidt et al., 2016;). This consistent factor indicated that older adults may perceive the risk of walking outdoors in winter to be greater than the benefit and choose to avoid doing so. Individuals with poor lower extremity performance had higher fear of falls .
The results of this scoping review have important implications for our study population, as well as other key stakeholders. As active aging, functional independence, and social participation are all impacted by maintaining one's mobility (Krogstad et al., 2015), being able to walk year-round in a safe and self-directed manner are important goals. By identifying the known facilitators of and Kyle Bergen et al.  Li et al., 2013 Questionnaire regarding difficulty with outdoor winter walking and at 9 specific outdoor locations, such as sidewalks and ramps The presence of snow or ice Those with functional limitations were more likely to describe difficulties in getting about outdoors in winter than those with no functional limitations. Winter coping strategies including remaining home, reducing frequency of going out, avoiding areas/locations with snow/ice/ slush, utilizing vehicles rather than walking, attaining exercise through a gym instead of walking. The following physical environment problems were identified: icy, snowy, or slushy surfaces, snow banks, puddles, needing to traverse over metal manhole covers, tall snow piles preventing appropriate vision between pedestrians and drivers, snow obscuring the pedestrian's ability to tell the difference between curb ramp and the sidewalk.
Winter coping strategies including utilizing ice grip devices and enlisting other individuals for assistance. Outdoor Winter Walking in Older Adults barriers to outdoor winter walking in older adults, there can be knowledge regarding why older adults may walk less in the winter, what factors surrounding winter pose difficulties for them, and how policy, programs, or other strategies may assist in facilitating walking in winter. Understanding that ice, snow, or slush-covered walkways will negatively impact walking for older adults identifies that prioritizing walkway clearance, sanding iced pathways, and removing snow from benches, as examples, are important tasks in the built environment. Specifically, Li et al. (2013) and Finlay (2018) identified snow at locations such as curb ramps, bus stops, crosswalks, and sidewalks as barriers to outdoor walking. To support this as a priority, Wennberg et al. (2009) found that a facilitator to walking was prompt snow removal so that walking surfaces were visible. Departments in cities or towns can benefit from this information by utilizing it to focus resources on tasks that can reduce barriers to outdoor walking for older adults. The finding that poorer physical function will negatively affect outdoor winter walking  through fear identifies that interventions to improve physical function in older adults could potentially facilitate walking. This is supported by , who found that higher physical function was positively associated with the number of destinations walked to in winter among their older adult participants.
A number of studies were excluded, as they evaluated outdoor physical activity, but did not specifically identify walking. All eight studies used interviews and questionnaires to identify outdoor walking in this scoping review. One study also used accelerometry , while two used observation (Finlay, 2018;Wennberg et al., 2009). It should be noted that although an older adult may be less physically active and walk outdoors in winter less, indoor walking alternatives may be used as a compensatory strategy to allow older adults to remain physically active .

Consultation
Consultation was had with the older adult community members of the Winter Walk Team. Both team members agreed that the results of this study were applicable to their association members. It was noted that much of the information applies to people of all ages. The importance of public works and others to be able to remove snow, ice, and slush was supported. Experience was shared that when this occurs, it makes older adults feel safer and more able to have a good walking experience. It was also supported that planning for green space, trails, and benches is very important. It was noted that in the previous year with COVID-19, an increase in dog ownership and experiences with isolation and physical distancing may be motivations for older adults to get outdoors more; the importance of addressing fear of falling and interventions for fall prevention are essential. It was stressed that it will be important to address other knowledge translation strategies of this project; for example, creating an infographic showing the need for cities/municipalities to clear snow from walking paths and sidewalks, which could convince municipal decision makers of the value of clean streets and sidewalks in winter. It is also important to make the connection between walking and one's own health; winter walking is hampered because municipalities do not connect snow with walking and health.

Limitations
There were some limitations to the current scoping review. Our methodology did not evaluate the quality of the studies included, as is common in a scoping review. In a scoping review of 344 scoping reviews, Pham et al. (2014) found that only 22 per cent assessed the methodological quality of their respective included studies, with the majority stating that scoping reviews do not prioritize quality assessment or that quality assessment is not commonly a part of their methodology. Second, including only studies in English excluded potential research conducted in other cultures or languages. Third, the choice to limit the scope of this review to focus on walking was beneficial for assessing a specific facet of physical activity in older adults. The limitation of this approach was that there were a large number of studies in the screening stages that were excluded because the outcomes examined were physical activity levels and not specifically walking. This identified that there was an abundance of research surrounding winter and its effect on physical activity in older adults in general. Some factors may have been missed by this scoping review by way of studies that discussed walking as a type of physical activity among participants in the study, but in measuring physical activity did not distinguish between types of physical activity, specifically walking, in the results. It was unclear in some studies whether the winter walking challenges were reported during the winter season or if winter weather was reflected upon at another time, in interviews or with questionnaires. Another potential limitation was that most studies focused on the urban setting; challenges specific to rural settings were not identified in the analysis. Perhaps this scoping review may spark interest in more in-depth examinations of walking as a specific type of physical activity in the winter months among older adults.

Conclusions
This scoping review identified the factors associated with outdoor winter walking for older adults. The studies in the review included both observed and perceived facilitators and barriers to winter walking that were internal and external in nature. Four main commonalities were found in the results: the impact of adverse weather conditions, the physical environment, physical function, and perceptions and fears relating to winter walking conditions. Overall, the studies cumulatively and consistently found that harsh winter weather conditions producing snow, ice, wind, and cold temperature decreased outdoor winter walking for older adults Clarke et al., 2017;Schmidt et al., 2016;Wu et al., 2017). Fear of falls was consistently a barrier for older adults, especially those with poorer lower extremity performance . Additional questions result from this review. For example, are challenges in outdoor winter walking for pleasure or exercise different from the requirements for and challenges to safe outdoor walking necessary for transport, to get to and from vehicles for transportation to other activities outside the home? This may assist in the development of indoor programs focusing on skills necessary for safe outdoor winter walking, as determined by the purpose that older adults may identify.
Research findings from this study and further research studies could be used in developing public education strategies geared towards older adults. Rehabilitation and exercise professionals can use the information from this article to educate their clients about the facilitators of and barriers to outdoor winter walking. Being aware of the barriers can assist health care professionals in implementing alternatives and solutions to these barriers, such as home and community exercise programs.