Abstract

Objectives: to compare walking speed in the UK older population with the speed required to utilise pedestrian crossings (≥1.2 m/s), and determine health and socio-demographic associations with walking impairment.

Design: cross-sectional study using Health Survey for England 2005 data.

Setting: private households in England.

Participants: random population sample of 3,145 adults (1,444 men) aged ≥65 years.

Main outcome measures: walking speed was assessed by timing a walk of 8 feet at normal pace. Walking impairment was defined as walking speed <1.2 m/s or non-participation in the test due to being unsafe or unable.

Results: the mean walking speed was 0.9 m/s in men and 0.8 m/s in women; 84% of men and 93% of women ≥65 years had walking impairment. Female gender, increasing age, lower socio-economic status, poorer health and lower grip strength were predictors of walking impairment.

Conclusion: most older adults either cannot walk 8 feet safely or cannot walk fast enough to use a pedestrian crossing in the UK. The health impacts on older adults include limited independence and reduced opportunities for physical activity and social interaction. An assumed normal walking speed for pedestrian crossings of 1.2 m/s is inappropriate for older adults and revision of these timings should be considered.

Introduction

The ability to cross the road safely is important for the health of older people. Walking activity among older adults, which has direct health benefits, is greater in pedestrian-friendly neighbourhoods [1, 2]. An inability to cross the road safely may reduce access to goods, health services and social contacts and thus adversely affect health. The divisive effects a road has on local residents, known as ‘community severance’ [3], may have a disproportional impact on the health of older adults because they are more likely to avoid crossing busy roads than younger adults [4].

Older pedestrians are more likely to die [5–7] or be seriously injured [6] in road traffic collisions than younger people due to decreased walking speed, slower decision-making and perceptual difficulties [8].

Having enough time is important for crossing the road safely. UK pedestrian crossing timings assume a minimum walking speed of 1.2 m/s (2.7 miles/h). Normal gait speeds of healthy people range from 0.94 m/s (2.1 miles/h) for women aged 80–99 to 1.43 m/s (3.2 miles/h) in men aged 40–49 [9]. However, these norms are not representative of the population who would like to use pedestrian crossings.

Studies in Ireland [10], the USA [11, 12], South Africa [13] and Spain [14] have shown that older adults have insufficient time at pedestrian crossings. Yet most studies are limited by the ‘healthy’ sample [10], small size [13], non-random sample [10, 13] and/or selection only of individuals actually crossing the road [11, 13].

There is some evidence that walking speed is socially patterned [15], suggesting that the negative health impacts of inappropriate crossing timings may be greatest among more deprived groups.

This cross-sectional study aims to describe mean ‘normal’ walking speeds of older adults in the UK and the proportion of the older population who were able to walk at ≥1.2 m/s, to assess the appropriateness of this speed as the basis for pedestrian crossing timings. We also investigated socio-demographic and health predictors of walking impairment.

Method

The Health Survey for England (HSE) is an annual, cross-sectional survey of a nationally representative sample of adults and children living in private households in England. HSE 2005 included a boost sample of people aged ≥65 [16, 17]. The London Multi-centre Research Ethics Committee approved the study.

Household response rates were 71% in the core and 74% in the boost samples. Data collection took place at an interview and a nurse visit. Totally, 4,269 people (1,897 men) aged ≥65 were interviewed, of whom 74% had a nurse visit.

‘Normal’ walking speed was assessed by timing how long it took the participant to walk 8 feet at their normal pace. The test was not carried out if they were unable to walk the distance, were unsafe, were unwilling, if there was no suitable space, or if their walking aid was unavailable. The walk was carried out twice, and the mean result used. Maximal grip strength was measured with a gripometer on alternate hands. Walking speed and grip strength measurement followed standard protocols [18].

At the interview, data were collected on health (self-reported health, limiting longstanding illness, mobility, falls, functional limitations and BMI), health behaviours (smoking and alcohol consumption) and demographic information (age, sex and ethnicity). Area deprivation was assessed using the Index of multiple deprivation (IMD) 2004 [19]. Further details of the sampling, recruitment and data collection have been reported [16].

Statistical analysis

The median, mean and standard deviation of walking speed and the proportion of participants with a walking speed <1.2 m/s were calculated. Walking impairment was defined as the participant being unable or unsafe to take the walking speed test or having a walking speed <1.2 m/s.

Logistic regression modelling was used to determine the associations with walking impairment. Possible explanatory variables were tested; significant variables were included in the final model. Statistics were adjusted for clustered stratified sampling and weighted to reduce non-response bias, except when describing participant characteristics. Statistical analysis was conducted in Stata Version 11.0.

Results

A total of 3,145 older adults (46% men) received a nurse home visit. Supplementary data are available in Age and Ageing online; Table w1 shows participants' characteristics.

90% of men and 87% of women took the walking speed test. 5.7% of men and 7.2% of women did not participate because they were unable to walk short distances or felt unsafe and 133 (4.3%) participants were not tested because of unwillingness (1.8%) or technical problems (2.5%).

The mean ‘normal’ walking speed was 0.9 m/s in men and 0.8 m/s in women, with a decrease in speed as age increased (See supplementary data available in Age and Ageing online, Figures w1 and w2); 76% men and 85% women had a walking speed <1.2 m/s; 93% of woman and 84% of men had walking impairment (Table 1).

Table 1.

Walking speed test performance by age and sex (= 3,145)

  Age
 
65–69 70–74 75–79 80–84 85+ All 65+ 
Men 
 Unable to do testa (%) 13 
 Walking speed <1.2 m/s (%) 73 76 80 85 85 76 
 Total walking impaireda (%) 77 82 87 91 98 84 
Walking speed (m/s)b 
 Mean (standard deviation) 1.0 (0.3) 0.9 (0.3) 0.9 (0.3) 0.8 (0.3) 0.7 (0.2) 0.9 (0.3) 
 Standard error 0.01 0.02 0.02 0.02 0.03 0.01 
 Median (inter-quartile range) 1.0 (0.4) 0.9 (0.3) 0.9 (0.4) 0.8 (0.4) 0.7 (0.4) 0.9 (0.4) 
Women 
 Unable to do testa (%) 14 17 
 Walking speed <1.2 m/s (%) 82 84 89 84 83 85 
 Total walking impaireda (%) 87 89 96 98 100 93 
Walking speed (m/s)b 
 Mean (standard deviation) 0.9 (0.3) 0.9 (0.3) 0.8 (0.3) 0.7 (0.2) 0.5 (0.2) 0.8 (0.3) 
 Standard error 0.02 0.01 0.02 0.02 0.02 0.01 
 Median (inter-quartile range) 0.9 (0.4) 0.9 (0.4) 0.8 (0.4) 0.7 (0.5) 0.5 (0.3) 0.8 (0.4) 
  Age
 
65–69 70–74 75–79 80–84 85+ All 65+ 
Men 
 Unable to do testa (%) 13 
 Walking speed <1.2 m/s (%) 73 76 80 85 85 76 
 Total walking impaireda (%) 77 82 87 91 98 84 
Walking speed (m/s)b 
 Mean (standard deviation) 1.0 (0.3) 0.9 (0.3) 0.9 (0.3) 0.8 (0.3) 0.7 (0.2) 0.9 (0.3) 
 Standard error 0.01 0.02 0.02 0.02 0.03 0.01 
 Median (inter-quartile range) 1.0 (0.4) 0.9 (0.3) 0.9 (0.4) 0.8 (0.4) 0.7 (0.4) 0.9 (0.4) 
Women 
 Unable to do testa (%) 14 17 
 Walking speed <1.2 m/s (%) 82 84 89 84 83 85 
 Total walking impaireda (%) 87 89 96 98 100 93 
Walking speed (m/s)b 
 Mean (standard deviation) 0.9 (0.3) 0.9 (0.3) 0.8 (0.3) 0.7 (0.2) 0.5 (0.2) 0.8 (0.3) 
 Standard error 0.02 0.01 0.02 0.02 0.02 0.01 
 Median (inter-quartile range) 0.9 (0.4) 0.9 (0.4) 0.8 (0.4) 0.7 (0.5) 0.5 (0.3) 0.8 (0.4) 

aIncluding those who felt unable or unsafe to perform the test.

bAmong those doing the walking speed test.

After mutual adjustment, functional disabilities (excluding walking disabilities), alcohol consumption and falls were not associated with walking impairment, and so were excluded from the final model. Female gender, current smoking, living in a deprived area, fair or poor self-reported health, low grip strength and limiting longstanding illness were associated with walking impairment in the unadjusted and fully adjusted analyses (Table 2).

Table 2.

Prevalence of walking impairment and univariable and multivariable associations

Variable Walking impairment (%) Univariable associations P-value Multivariable associationsa P-value 
OR (95% CI) OR (95% CI) 
Sex 
 Male 1,105 (84) <0.001 <0.001 
 Female 1,566 (93) 2.40 (1.89–3.05) 2.64 (2.02–3.34) 
Age (years)      
 65–69 730 (83) <0.001 <0.001 
 70–74 653 (86) 1.26 (0.96–1.66) 1.09 (0.81–1.45) 
 75–79 563 (92) 2.33 (1.65–3.28) 1.54 (1.07–2.22) 
 ≥80 726 (97) 6.63 (3.89–11.28) 3.65 (2.12–6.27) 
Index of multiple deprivation (IMD, 2004) 
 0.59 ≤8.35 (least deprived) 574 (83) <0.001 <0.001 
 8.35 ≤21.16 (middle tertile) 1,141 (87) 1.41 (1.06–1.86) 1.40 (1.03–1.89) 
 21.16–86.36 (most deprived) 956 (94) 3.40 (2.37–4.86) 2.50 (1.70–3.68) 
Highest educational qualificationb 
 University degree 185 (74) <0.001 — — 
 Other qualification 917 (85) 1.91 (1.39–2.61) — 
 None 1,566 (94) 5.20 (3.74–7.23) — 
Smoking status 
 Never smoker 1,238 (87) 0.010 0.012 
 Ex-smoker 1,109 (89) 1.20 (0.94–1.53) 1.39 (1.06–1.82) 
 Current smoker 322 (93) 1.99 (1.24–3.19) 1.84 (1.11–3.05) 
Alcohol consumption (estimated units consumed on heaviest drinking day in last week)      
 None 702 (92) <0.001 — — 
 Less than or equal to recommended limitc 1,272 (86) 0.54 (0.40–0.72) — 
 Over recommended limit 267 (84) 0.47 (0.33–0.66) — 
General health 
 Good or very good 1,402 (83) <0.001 <0.001 
 Fair 892 (95) 4.41 (3.14–6.20) 2.87 (1.99–4.14) 
 Poor or very poor 377 (99) 36.32 (8.98–146.99) 15.99 (3.96–64.51) 
Longstanding illness 
 No longstanding illness 704 (83) <0.001 0.012 
 Non-limiting longstanding illness 681 (84) 1.10 (0.82–1.46) 0.98 (0.72–1.34) 
 Limiting longstanding illness 1,286 (95) 4.04 (2.95–5.54) 1.54 (1.10–2.15) 
Functional disabilitiesd 
 0 2,189 (88) 0.005 — — 
 ≥1 481 (92) 1.66 (1.16–2.38) — 
Fall in previous 12 months 
 No 1,928 (88) 0.001 — — 
 Yes 742 (92) 1.63 (1.21–2.19) — 
Grip strengthe 
 Median or above 1,161 (82) <0.001 <0.001 
 Below median 1,407 (95) 4.44 (3.36–5.85) 2.49 (1.84–3.37) 
Variable Walking impairment (%) Univariable associations P-value Multivariable associationsa P-value 
OR (95% CI) OR (95% CI) 
Sex 
 Male 1,105 (84) <0.001 <0.001 
 Female 1,566 (93) 2.40 (1.89–3.05) 2.64 (2.02–3.34) 
Age (years)      
 65–69 730 (83) <0.001 <0.001 
 70–74 653 (86) 1.26 (0.96–1.66) 1.09 (0.81–1.45) 
 75–79 563 (92) 2.33 (1.65–3.28) 1.54 (1.07–2.22) 
 ≥80 726 (97) 6.63 (3.89–11.28) 3.65 (2.12–6.27) 
Index of multiple deprivation (IMD, 2004) 
 0.59 ≤8.35 (least deprived) 574 (83) <0.001 <0.001 
 8.35 ≤21.16 (middle tertile) 1,141 (87) 1.41 (1.06–1.86) 1.40 (1.03–1.89) 
 21.16–86.36 (most deprived) 956 (94) 3.40 (2.37–4.86) 2.50 (1.70–3.68) 
Highest educational qualificationb 
 University degree 185 (74) <0.001 — — 
 Other qualification 917 (85) 1.91 (1.39–2.61) — 
 None 1,566 (94) 5.20 (3.74–7.23) — 
Smoking status 
 Never smoker 1,238 (87) 0.010 0.012 
 Ex-smoker 1,109 (89) 1.20 (0.94–1.53) 1.39 (1.06–1.82) 
 Current smoker 322 (93) 1.99 (1.24–3.19) 1.84 (1.11–3.05) 
Alcohol consumption (estimated units consumed on heaviest drinking day in last week)      
 None 702 (92) <0.001 — — 
 Less than or equal to recommended limitc 1,272 (86) 0.54 (0.40–0.72) — 
 Over recommended limit 267 (84) 0.47 (0.33–0.66) — 
General health 
 Good or very good 1,402 (83) <0.001 <0.001 
 Fair 892 (95) 4.41 (3.14–6.20) 2.87 (1.99–4.14) 
 Poor or very poor 377 (99) 36.32 (8.98–146.99) 15.99 (3.96–64.51) 
Longstanding illness 
 No longstanding illness 704 (83) <0.001 0.012 
 Non-limiting longstanding illness 681 (84) 1.10 (0.82–1.46) 0.98 (0.72–1.34) 
 Limiting longstanding illness 1,286 (95) 4.04 (2.95–5.54) 1.54 (1.10–2.15) 
Functional disabilitiesd 
 0 2,189 (88) 0.005 — — 
 ≥1 481 (92) 1.66 (1.16–2.38) — 
Fall in previous 12 months 
 No 1,928 (88) 0.001 — — 
 Yes 742 (92) 1.63 (1.21–2.19) — 
Grip strengthe 
 Median or above 1,161 (82) <0.001 <0.001 
 Below median 1,407 (95) 4.44 (3.36–5.85) 2.49 (1.84–3.37) 

aAll results are mutually adjusted.

bEducation was not included in the multivariable model due to collinearity with area level deprivation (IMD, 2004) [19].

cFour units for men, three units for women.

dExcluding walking disabilities.

eMedian grip strength 36 kg in men, 21 kg in women.

Discussion

The mean walking speed in both men and women was below the speed required to use a pedestrian crossing in the UK and many other countries [10, 12, 13]; 93% of women and 84% of men aged ≥65 years either could not walk 8 feet safely or their normal walking speed was too slow to cross the road in time.

The mean walking speeds were lower than established norms [9], possibly because our study did not exclude unhealthy participants. Our findings are consistent with other studies in showing that many older pedestrians have insufficient time to use pedestrian crossings [10, 12, 14]. The walking speeds determined in this study were generally similar to, or lower than, those found in other studies.

It is well established that walking speed is lower in women and decreases with age [9, 20]. In addition to women, and the ‘oldest old’, those in deprived areas, smokers, those with poor grip strength [indicative of sarcopenia (loss of muscle mass)] and those whose general health is only fair or worse or who had longstanding illness were most likely to have walking impairment. The walking impaired participants may therefore be characterised as ‘frail’ [21], though definitions of this term vary. Grip strength is known to be an independent predictor of walking impairment [10]. However, poor mobility and a history of falls have also been found to be risk factors [10], but falls were not significant in our multivariable model.

Similar patterns of social inequality in walking speed have also been found in early old age [15]. Residual confounding by health problems not captured in our study [16, 22] may explain some of the association between low socio-economic status and walking impairment. However, only one-third of observed social inequality in walking speed can be explained by health conditions or demographic, psychosocial, biological and behavioural factors [15].

The main strength of this study is that it provides an accurate picture of the proportion of people aged ≥65 years in the general population who are likely to be unable to use pedestrian crossings safely. The large sample size, random sample and the fact that participants were not excluded on the basis of disability mean that the data are representative of the older population who may wish to use a pedestrian crossing.

A further strength is that those people who were unable or for whom it was unsafe to participate in the walking speed test were included in the analysis (classed as walking impaired alongside those with gait speed <1.2 m/s). The advantage of using a general population sample rather than surveying people using a pedestrian crossing [11, 13, 23] is that those people who have difficulty using pedestrian crossings, and are therefore not utilising them, are captured.

A limitation of this study is the non-response bias that would result from differential participation in the survey. It is likely that older people with worse health were less likely to respond (including those temporarily in hospital). This study could have underestimated the prevalence of walking impairment, although the data were weighted to adjust for non-response.

Insufficient crossing time among older adults may not increase the risk of pedestrian fatalities, which are uncommon at pedestrian crossings, but it is likely to deter this group from even trying to cross the road. For older people, maintenance of mobility outside the home not only has direct health benefits but is also an important way to maintain independence and social networks [3, 24, 25]. Physical activity in older people may depend on the ability to negotiate their local environment, including crossing the road safely. The groups we have identified as being most likely to having walking impairment are also those least likely to have access to other, more expensive, forms of transport. Puffin crossings (with timings regulated by sensors) may enable older adults to cross in time, but more are needed and their profile must be raised for benefits to be realised.

The assumed ‘normal’ walking speed of 1.2 m/s is utilised internationally as the basis for pedestrian crossing timings. Our results show that pedestrian crossings requiring a walking speed of 0.8 m/s may be more appropriate, as this would allow the ‘average’ man or woman ≥65 years sufficient time to cross. The current assumed walking speed excludes most of the older population in England from using pedestrian crossings and therefore should be revised.

Key points

  • The vast majority of people over 65 years old in England are unable to walk fast enough to use a pedestrian crossing.

  • Those affected are more likely to be from deprived areas.

  • It is important for older adults to be able to cross the road safely to keep physically active and maintain social contacts.

  • Current pedestrian crossing timings should therefore be reviewed.

Conflicts of interest

None declared.

Funding

The Health Survey for England 2005 was funded by the English Department of Health. The funder played no part in the decisions to undertake these secondary analyses or to submit this paper for publication. The views expressed are those of the authors, not the funder. Three of the authors are currently, or have previously been, funded to work on the Health Survey for England series, but this study received no funding.

Supplementary data

Supplementary data mentioned in the text is available to subscribers in Age and Ageing online.

References

1
Li
F
Fisher
KJ
Brownson
RC
Bosworth
M
Multilevel modelling of built environment characteristics related to neighbourhood walking activity in older adults
J Epidemiol Community Health
 , 
2005
, vol. 
59
 (pg. 
558
-
64
)
2
Clarke
P
Ailshire
JA
Bader
M
Morenoff
JD
House
JS
Mobility disability and the urban built environment
Am J Epidemiol
 , 
2008
, vol. 
168
 (pg. 
506
-
13
)
3
Mindell
JS
Karlsen
S
Community severance and health: what do we actually know?
J Urban Health
 , 
2012
, vol. 
89
 (pg. 
232
-
46
)
4
Shumway-Cook
A
Patla
A
Stewart
A
Ferrucci
L
Ciol
MA
Guralnik
JM
Environmental components of mobility disability in community-living older persons
J Am Geriatr Soc
 , 
2003
, vol. 
51
 (pg. 
393
-
8
)
5
Kim
JK
Ulfarsson
GF
Shankar
VN
Kim
S
Age and pedestrian injury severity in motor-vehicle crashes: a heteroskedastic logit analysis
Accid Anal Prev
 , 
2008
, vol. 
40
 (pg. 
1695
-
702
)
6
Martin
AJ
Hand
EB
Trace
F
O'Neill
D
Pedestrian fatalities and injuries involving Irish older people
Gerontology
 , 
2010
, vol. 
56
 (pg. 
266
-
71
)
7
Nicaj
L
Wilt
S
Henning
K
Motor vehicle crash pedestrian deaths in New York City: the plight of the older pedestrian
Inj Prev
 , 
2006
, vol. 
12
 (pg. 
414
-
6
)
8
Dommes
A
Cavallo
V
Vienne
F
Aillerie
I
Age-related differences in street-crossing safety before and after training of older pedestrians
Accid Anal Prev
 , 
2012
, vol. 
44
 (pg. 
42
-
7
)
9
Bohannon
RW
Williams Andrews
A
Normal walking speed: a descriptive meta-analysis
Physiotherapy
 , 
2011
, vol. 
97
 (pg. 
182
-
9
)
10
Romero-Ortuno
R
Cogan
L
Cunningham
CU
Kenny
RA
Do older pedestrians have enough time to cross roads in Dublin? A critique of the Traffic Management Guidelines based on clinical research findings
Age Ageing
 , 
2010
, vol. 
39
 (pg. 
80
-
6
)
11
Hoxie
RE
Rubenstein
LZ
Are older pedestrians allowed enough time to cross intersections safely?
J Am Geriatr Soc
 , 
1994
, vol. 
42
 (pg. 
241
-
4
)
12
Langlois
JA
Keyl
PM
Guralnik
JM
Foley
DJ
Marottoli
RA
Wallace
RB
Characteristics of older pedestrians who have difficulty crossing the street
Am J Public Health
 , 
1997
, vol. 
87
 (pg. 
393
-
7
)
13
Amosun
SL
Burgess
T
Groeneveldt
L
Hodgson
T
Are elderly pedestrians allowed enough time at pedestrian crossings in Cape Town, South Africa?
Physiother Theory Pract
 , 
2007
, vol. 
23
 (pg. 
325
-
32
)
14
Romero Ortuno
R
The regulation of pedestrian traffic lights in Spain: do older people have enough time to cross the road?
Rev Esp Geriatr Gerontol
 , 
2010
, vol. 
45
 (pg. 
199
-
202
)
15
Brunner
E
Shipley
M
Spencer
V
, et al.  . 
Social inequality in walking speed in early old age in the Whitehall II study
J Gerontol A Biol Sci Med Sci
 , 
2009
, vol. 
64
 (pg. 
1082
-
9
)
16
Craig
R
Mindell
J
The Health Survey for England 2005: The Health of Older Adults
 , 
2007
London
The Information Centre
17
Mindell
J
Biddulph
JP
Hirani
V
, et al.  . 
Cohort profile: the health survey for England
Int J Epidemiol
 , 
2012
18
Guralnik
JM
Ferrucci
L
Pieper
CF
, et al.  . 
Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery
J Gerontol A Biol Sci Med Sci
 , 
2000
, vol. 
55
 (pg. 
M221
-
31
)
19
Statistics Authority UK
2004
 
20
Bohannon
RW
Comfortable and maximum walking speed of adults aged 20–79 years: reference values and determinants
Age Ageing
 , 
1997
, vol. 
26
 (pg. 
15
-
9
)
21
Sternberg
SA
Wershof Schwartz
A
Karunananthan
S
Bergman
H
Mark Clarfield
A
The identification of frailty: a systematic literature review
J Am Geriatr Soc
 , 
2011
, vol. 
59
 (pg. 
2129
-
38
)
22
Marmot
M
Shipley
M
Brunner
E
Hemingway
H
Relative contribution of early life and adult socioeconomic factors to adult morbidity in the Whitehall II study
J Epidemiol Community Health
 , 
2001
, vol. 
55
 (pg. 
301
-
7
)
23
Sterling
TK
Sharrat
C
Walter
L
Narine
S
The Effect of Re-timed Invitation to Cross Periods on Road User Behaviour at Signalised Junctions in London
 , 
2009
Wokingham
Transport Research Laboratory
24
Appleyard
DGM
Lintell
M
Liveable Streets
 , 
1981
Berkeley
University of California Press
25
Mindell
JS
Mindell
JS
Watkins
SJ
Cohen
JM
Stress, social support and community severence
Health on the Move 2
 , 
2011
Stockport
Transport and Health Study Group

Supplementary data

Comments

0 Comments