Abstract
Keywords
Introduction
Perceived difficulty in walking increases with increasing age (Sainio et al., 2006; Simonsick et al., 2008) and may threaten participation in terms of running errands or engaging in social, cultural, or physical activities (Hovbrandt, Stahl, Iwarsson, Horstmann, & Carlsson, 2007; Rantanen, 2013). Typically, mobility gradually declines with age, following a pattern where performance of the most strenuous tasks, such as walking longer distances, deteriorates first (Rantanen, 2013). These kinds of early mobility limitations increase the risk of more advanced mobility limitations (Manty et al., 2007). Eventually, when no longer able to walk a few hundred meters, the individual’s independent community mobility becomes threatened (Andrews et al., 2010).
The neighborhood environment in which older people most often move may affect their walking ability. Applying the rationale of the person–environment (P-E) fit model (Lawton & Nahemow, 1973), an imbalance between the person and the environment can be assumed to underlie perceived difficulties in walking. Earlier follow-up studies have shown that environmental deficits, such as long distance to services and lack of resting places (Rantakokko, Iwarsson, Manty, Leinonen, & Rantanen, 2012), or reporting two or more such deficits (poor lighting, heavy traffic, poor access to public transportation, excessive noise; Balfour & Kaplan, 2002) may explain incident difficulties in mobility functions. Individual factors, such as fear of falling, may also have negative effect on mobility in old age (Donoghue, Cronin, Savva, O’Regan, & Kenny, 2013; Viljanen et al., 2012). Conversely, outdoor recreational facilities that are easy to access and located within walking distance from home may delay mobility decline among older community-dwelling people (Eronen, von Bonsdorff, Rantakokko, & Rantanen, 2014).
In the above-mentioned studies, however, the participants were studied as one group in spite of the wide variation in functional capacity and mobility generally found in older people. In addition, most studies that have investigated the relationship between environmental factors and mobility have relied on self-assessments of mobility (Balfour & Kaplan, 2002; Clarke & Nieuwenhuijsen, 2009). Self-reports of mobility are typically based on the individual’s experiences when doing the task in his or her actual living environment, and thus have face validity in relation to everyday life. In turn, performance-based tests of mobility describe the individual’s motor capacity to function optimally in standardized circumstances (Mannerkorpi, Svantesson, & Broberg, 2006; Sainio et al., 2006). It is necessary to evaluate whether the information obtained with performance-based tests of motor functioning, allowing classification of the participants to mobility performance subgroups, would add anything important to the relationship between environment and self-assessed mobility.
The aim of this study was to analyze the associations between perceived environmental and individual characteristics and perceived walking difficulties in older people. We studied the correlates for early and advanced walking limitations so as to distinguish between minor and more severe perceived difficulties. In addition, we wanted to find out to what extent the associations between neighborhood environment and perceived walking limitations are linked to objectively measured mobility performance level. Our hypothesis was that the associations would differ for people with intact or poorer lower extremity performance.
Method
Study Design and Participants
This study is part of the Life-Space Mobility in Old Age (LISPE) project, a 2-year follow-up study examining how the characteristics of home and neighborhood influence older people’s health, functioning, quality of life and life-space mobility (Rantanen et al., 2012). In short, 2,550 persons aged 75 to 89 years and resident in the municipalities of Jyväskylä or Muurame in central Finland were drawn from the population register. This was done in three samples of 850 persons from each 5-year age group (persons aged 75-79, 80-84, and 85-89). Inclusion criteria were community living, ability to communicate, living in the research district, and willingness to participate. The present study was a cross-sectional analysis of the baseline data for all those who met the inclusion criteria, in total 848 participants. The data used for the present analyses were collected through interviews and performance tests in the participants’ homes. Due to internal missing values on the mobility performance test (Short Physical Performance Battery [SPPB]), nine participants were excluded. In addition, four persons had missing values in the question concerning the ability to walk 2 km and were excluded, as well as one person due to inconsistency in mobility questions. Thus, the final sample was 834 participants from three age groups (75- to 79-year-olds, n = 352; 80- to 84-year-olds, n = 279; 85- to 89-year-olds, n = 203). Data quality was good, with less than 2% of the data missing for all the variables in the present database. The study was approved by the University of Jyväskylä Ethical Committee, and all participants gave their informed consent in writing.
Assessments
Mobility
Early and advanced mobility limitations were assessed as perceived difficulty in walking 2 and 0.5 km. The respondents were asked “Are you able to walk about 2 km (0.5 km)?” The response alternatives were 0 = yes, without difficulties; 1 = yes, but with some difficulty; 2 = yes, but with a great deal of difficulty; 3 = not without help of another person; and 4 = unable. The variables were dichotomized into those without difficulties (Category 0) and those with difficulties (Categories 1-4). We created a three-category variable as follows: no walking limitation (no difficulty in walking 2 and 0.5 km), early walking limitations (difficulty in walking 2 km), and advanced walking limitations (difficulty in walking 2 and 0.5 km). The categorization was valid with the exception of one participant who reported difficulties in walking 0.5 km but not 2 km, and who was thus excluded due to miscoding.
Intact and poorer lower extremity performance was determined with the SPPB (Guralnik et al., 1994), which includes timed measurements of self-paced walking for 2.44 m, five chair stands and standing in three different positions, each with a narrower base of support. The different tests were scored from 0 to 4 and added to form a summary score ranging from 0 to 12, with higher scores indicating better performance. The sample was divided into those with intact lower extremity performance (SPPB = 10-12) and those with poorer lower extremity performance (SPPB = 9 or below), using a previously validated cutoff (Bandinelli et al., 2006; Vazzana et al., 2010).
Environmental Barriers to and Facilitators of Mobility
The participants were asked what environmental barriers to and facilitators of mobility they perceived in their neighborhood. The self-rated barriers were recorded using the “Checklist for Perceived Environmental Barriers to Outdoor Mobility” (PENBOM; Rantakokko et al., 2014). The checklist comprises 15 items and one open-ended question about the environmental barriers that people perceive as hindering their possibilities for outdoor mobility, such as street conditions, high curbs, long distance to services, lack of resting places, snow and ice in winter, and so on. Another checklist of self-rated facilitators of mobility, “Perceived Environmental Facilitators for Outdoor Mobility”(PENFOM; Rantakokko, Iwarsson, Portegijs, Viljanen, & Rantanen, 2015) included 16 items, such as the existence of a park nearby, good walking routes, beautiful landscape, familiar surroundings, good lighting, other walkers whose example inspires one to be active, and so on. Each item was marked as present or absent.
Individual Reasons That Hinder Outdoor Mobility
The participants were also asked to state what are the reasons that hinder or prevent them from outdoor physical activity, such as walking for fitness or walking to a store. This was done using the “Barriers to Outdoor Physical Activity Questionnaire” (BOPA; Eronen, Bonsdorff, Tormakangas et al., 2014). Fifteen items were listed under the themes of health, fear and negative experiences, ambient conditions, and lack of knowledge and negative attitude. From the list, the participants picked out all the items that they felt described their situation. Two items were omitted from the original list of 17 items, as they overlapped with the questions on walking difficulties and self-rated health.
The Covariates
The covariates used in the multivariate analyses were factors that may affect mobility and included age group, sex, education level (higher than primary school vs. primary school or less), self-rated health (good and very good vs. satisfactory, poor and very poor), self-reported chronic conditions (number of physician-diagnosed diseases according to a list of 22 chronic conditions and an additional open-ended question), body mass index (BMI, kg/m2, based on self-reported height and weight), cognitive function according to the Mini-Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975), and depressive symptoms assessed with the Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977). The number of years lived at the current address was obtained from the population register data. This information was missing for 29 participants (of these, 19 had intact and 10 had poorer performance of lower extremities).
Statistical Analyses
Polytomous logistic regression, which allows use of an outcome variable with more than two categories, was used with the three-category walking difficulties variable. The outcome variable was used as a nominal scale variable, with the group without walking difficulties as the reference group. The independent variables were perceived environmental barriers to and facilitators of outdoor mobility and the individual reasons that hinder outdoor mobility. From each questionnaire, the items that had significant or near significant bivariate associations to the outcome variable were chosen (p ≤ .100 in chi-square test, see Table 1). Possible confounding variables included in the models were sex, age, self-rated health, number of chronic conditions, BMI, cognitive function (MMSE), and depressive symptoms (CES-D). Two models were created, one for those with intact and another for those with poorer lower extremity performance (SPPB scores = 10-12, n = 526, and SPPB scores = 0-9, n = 308, respectively). Due to the sampling technique, the oldest age group was overrepresented and the youngest age group underrepresented. Thus, in the analyses, where all three age groups were included simultaneously, the data were weighted by the coefficients 1.30 in the age group 75 to 79, 1.04 in the age group 80 to 84, and by 0.66 in the age group 85 to 89.
Bivariate Associations Between Perceived Environmental and Individual Barriers to and Environmental Facilitators for Outdoor Mobility and Walking Limitations.
Note. Seven barriers are not shown and were excluded from further analyses due to low reporting rate (>20% of cells in cross-tabulation had expected count less than 5). These were vehicles on walkways, hearing problems, fear of getting hit by a car, environment around home not suitable for walking, health care personnel or relatives have told not to go walking outdoors, too old for outdoor walking, and not used to do outdoor activities.
Results
Those with intact lower extremity performance (SPPB score = 10-12) reported fewer chronic conditions and better self-rated health, had better cognitive function (higher MMSE score) and less depressive symptoms (lower CES-D score), and perceived fewer environmental barriers to mobility and individual barriers to outdoor physical activity as well as less walking difficulties than those with poorer lower extremity performance (SPPB score = 0-9; p < .001 for all comparisons; Table 2).
Participant Characteristics According to Lower Extremity Performance Assessed With the SPPB (N = 834).
Note. SPPB = Short Physical Performance Battery; MMSE = Mini-Mental State Examination; BMI = body mass index; CES-D = Center for Epidemiologic Studies Depression Scale.
p value for chi-square and t tests.
The factors associated with early and advanced walking limitations among those with intact and poorer lower extremity performance, respectively, are shown in Table 3. In general, perceiving environmental barriers to outdoor mobility and reporting personal reasons that hinder outdoor mobility increased the odds for early and advanced walking limitations. In contrast, perceiving environmental facilitators for mobility decreased the odds of walking limitations among participants with intact as well as poorer lower extremity performance.
Polytomous Logistic Regression Models for Early and Advanced Walking Limitations in Participants With Intact (Model 1) and Poorer (Model 2) SPPB.
Note. The two models include different variables; the ORs and CIs of all the variables that were included in each model are shown. The bold figures indicate statistically significant ORs. Early mobility limitation: Difficulty walking 2 km; advanced mobility limitation: Difficulty walking 2 and 0.5 km. SPPB = Short Physical Performance Battery; OR = odds ratio, CI = confidence interval; BMI = body mass index; MMSE = Mini-Mental State Examination; CES-D = Center for Epidemiologic Studies Depression Scale.
Among those with intact lower extremity performance, perceiving environmental barriers related to distances (long distances to the community amenities, lack of resting places) as well as lack of interest and company for outdoor mobility increased the odds for walking difficulties. In contrast, reporting good lighting on walking routes, shops and services nearby and other walkers’ inspiring example lowered the odds.
Among those with poorer lower extremity performance, perceiving snow and ice as barriers and reporting fear of falling and feelings of insecurity when walking outdoors increased the odds for walking limitations, while perceiving familiar surroundings and reporting having a park or a green area nearby lowered the odds.
Discussion
This study showed that overall, perceived neighborhood environmental characteristics were associated with walking limitations among older community-living people. Among those with poorer lower extremity performance the likelihood for advanced walking limitations was related, in particular, to safety issues, and among those with intact lower extremity performance to social factors and long distances.
The P-E fit model (Lawton & Nahemow, 1973) suggests that individuals with low competence are more sensitive to environmental characteristics than those with high competence. This has been confirmed in several longitudinal studies (Balfour & Kaplan, 2002; Rantakokko et al., 2012) and is also evident from the results of the present cross-sectional study. Moreover, earlier research has reported that aspects of the physical environment affect mobility more among those with existing mobility impairments than those without (Shumway-Cook et al., 2003; Yang & Sanford, 2012); however, this finding was not supported in the present study. Instead, the environmental characteristics that were associated with walking limitations were different among those with poorer mobility performance than those with intact mobility performance. This was a new finding, and it is supported by some earlier studies in which different functional limitations have been associated with different environmental barriers (Slaug, Schilling, Iwarsson, & Carlsson, 2011).
The finding that perceiving snow and ice in winter as barriers to outdoor mobility increased the likelihood of walking limitations in participants with poorer lower extremity performance indicates that winter conditions, in particular, hamper outdoor activities among those whose prerequisites for safe locomotion are compromised. During winter, existing problems with walking seem to intensify due to snow and ice, often resulting in reduced outdoor mobility (Hjorthol, 2013). Older people themselves also report that snow and ice hamper their walking in wintertime (Van Cauwenberg et al., 2012) and regard the prevention of ice and slipperiness as highly important (Wennberg, Stahl, & Hyden, 2009). In the Nordic countries, falls in wintertime occur most often when walking surfaces consist of ice covered with snow (Gao, Holmer, & Abeysekera, 2008) or when the weather conditions are favorable for the formation of ice (Morency, Voyer, Burrows, & Goudreau, 2012). From the viewpoint of the mobility of older people, this means that more should be done to keep the environment safe for walking in wintertime.
The lack of resting places in winter increased the likelihood of walking limitations among those with intact lower extremity performance, a factor which also is related to weather conditions in the Nordic countries. In wintertime, benches may not be accessible due to snow and ice, or they have been removed to make way for (snow) plowing. Earlier studies have shown that older people find benches and resting places very important for the usability of their outdoor environment (Hovbrandt, Fridlund, & Carlsson, 2007; Wennberg et al., 2009), including wintertime (Wennberg et al., 2009). Long distances and lack of resting places may even increase the probability for developing difficulties in walking (Rantakokko et al., 2012).
Among the participants with poorer lower extremity performance, familiarity with their surroundings decreased the risk for advanced walking limitations, while fear of falling and feelings of insecurity increased it. In combination with the weather conditions mentioned above, these findings show that safety aspects are particularly significant for mobility, if performance in basic mobility tasks requiring balance control, coordination, and muscle force is impaired. Balance and other sensorimotor functions are important determinants of mobility performance (Sakari et al., 2010), and controlling these functions during locomotion is probably more difficult if extra attention has to be paid to the environment. Associations between balance problems, alone or in combination with other sensory difficulties, and fear of falling have been shown earlier (Austin, Devine, Dick, Prince, & Bruce, 2007; Viljanen et al., 2013).
The finding that having services or shops near increased the risk for early walking difficulties in those with poorer lower extremity performance was unexpected and difficult to explain. It might be possible that these persons had already moved near the services due to their worsening health condition.
Among the participants with intact lower extremity performance, walking difficulties were explained by lack of company and lack of interest. Lack of company has often been reported to be one reason for not being sufficiently physically active in old age (Moschny, Platen, Klaassen-Mielke, Trampisch, & Hinrichs, 2011; Reichert, Barros, Domingues, & Hallal, 2007) and to hinder physical activity, particularly among those with mobility difficulties (Rasinaho, Hirvensalo, Leinonen, Lintunen, & Rantanen, 2007). Lack of interest has been shown earlier to be an important barrier to physical activity (Moschny et al., 2011). Here, the importance of social factors for mobility was also evident in the association found between other walkers’ inspiring example and lower risk for advanced walking limitations in those with intact lower extremity performance. It is likely that the relationship of social factors and lack of interest with walking limitations is explained indirectly by the effects of these factors on physical activity.
Strengths and Limitations of the Study
The strength of this study is the large population sample with a very small amount of missing data. Overall, the participants had lived at the same address for a long time, which indicates that the associations found were not due to recent changes in their living environment. Measuring mobility performance with the SPPB test was useful as it enabled splitting the sample according to the motor capacity of the participants. The analyses were, however, carried out in small subgroups, which may have somewhat underestimated the true associations.
The data on environmental factors and walking difficulties are based on self-reports. Self-reports are often thought to be unreliable and some studies have shown considerable discrepancy between subjective and objective measurements of the environment (Kirtland et al., 2003) and mobility of older people (Sainio et al., 2006). However, from the point of view of prevention, the subjective view is important. For example, if an older person does not perceive the distance to shops to constitute a barrier to mobility, he or she may walk there, which promotes walking ability. On the other hand, if the actual distance is short and the older person experiences it as long and harmful, some rehabilitative actions are definitely needed.
Conclusion
Clearly, older people with poorer mobility performance—which indicates less than optimal physical prerequisites for walking, and possibly also balance problems—are more vulnerable to insecurity aspects in their neighborhood environment than older people with intact mobility performance. In addition to rehabilitative actions, increasing the safety of the walking environment and arranging for official or unofficial company on walking trips would probably be beneficial. On the contrary, older people with good mobility performance might benefit from better support in the environment for walking recreation. This could be achieved, for example, by increasing possibilities to socialize and rest in connection with walking trips. The possible barriers to, and facilitators of, mobility in older people’s neighborhood environments should be systematically evaluated in connection with mobility assessments. The environmental aspects that are important for mobility should be taken into account in urban planning and street maintenance, particularly in city areas with a high density of older residents.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Academy of Finland (the Future of Living and Housing Program; Grants 255403, 263729), Finnish Ministry of Education and Culture, and the Ribbingska Foundation in Lund, Sweden.