The relationship between physical function in the elderly and judgment error in walking speed.

[Purpose] "Judgment error," defined as a difference between the actual and the imagined performance, is often observed in elderly persons. The aims of this study were to assess subjective judgment errors in elderly persons, and to evaluate the relationship between physical function and judgment error in walking speed.
[Subjects and Methods] A total of 106 community-dwelling elderly individuals participated. Subjects observed video footage of a model walking an obstacle course, and were asked to subjectively compare the model's gait speed with their own gait speed. When the subjective comparison differed from the actual difference, it was considered as a judgment error. Physical function was compared between those with and without judgment error.
[Results] Significant interaction effects between the actual performance and subjective perception were found for the walking time on the obstacle course and the Activities-specific Balance Confidence Scale score (utilized as an index of self-confidence in own balance ability and a fear of falling).
[Conclusion] The results demonstrate that some elderly persons tend to overestimate their balance and ambulation function compared to another person, even though they had low physical function and low self-confidence in terms of balance. These elderly individuals might have a high risk of accidents.

INTRODUCTION

Physical functioning declines with age; however, individual differences in the rate of decline exist. When an individual fails to appreciate the changes in physical ability caused by age, he/she is unable to choose an appropriate balance strategy, which may increase the risk of accidents. Moreover, properly imagining the time required for an action is an important factor in safely and smoothly navigating everyday life. For example, when judging whether a road can be crossed safely before the green light on a traffic signal changes, properly imagining one’s own movement and its result influences the risk of a traffic accident. Thus, there are numerous daily situations where imaging the results of actions, as well as the required time, are important, and we are constantly predicting the required time for actions in our daily living.

Previous studies on self-awareness of physical function in elderly individuals have utilized a method that measures the time or distance of the imagined performance of a movement without actual physical execution of the movement1,2,3,4,5,6,7,8). The method used for such a temporally predictive task is called the ‘mental chronometry’ method. Previous studies have reported that the duration of an imagined movement corresponds well to the duration of the actual movement in healthy young individuals9). However, many studies have reported that there are temporal differences between the actual physical execution of a movement and the imagined performance of the same movement in elderly individuals2, 6, 8, 10,11,12). That is, “judgment error” tends to be observed in elderly individuals. We have previously studied judgment error with regard to balance function among community-dwelling elderly individuals using spatially and temporally predictive tasks13). This previous study demonstrated that elderly individuals underestimate their abilities in spatially predictive tasks, such as the Functional Reach Test (FR)14) and upward reaching task, and overestimate their abilities in temporally predictive tasks, such as the Timed Up and Go test (TUG)15) and the Standardized Walking Obstacle Course (SWOC)16). The reason for this difference remains unknown; however, our research has also indicated that temporally predictive tasks might better evaluate judgment errors in the elderly compared to spatially predictive tasks.

In the situation of a pedestrian crossing a road before the green light changes as mentioned above, the mental chronometry method is considered to be performed naturally in a static environment. However, in our daily living, many situations exist that must be judged by imagining the results of our action in a dynamic environment (e.g., walking in a crowd with many people coming in, attempting to cross in front of others walking in various directions, crossing a road with cars and bicycles, etc.). In these examples, it is conceivable that individuals choose an appropriate action strategy by considering the speed of the strangers or vehicles while imaging the results of one’s own action. In present study, we investigated the subjective perceptions of elderly individuals regarding their balance function in comparison with an observed model walking an obstacle course. The subjects then walked the same obstacle course. Disagreement between the subjective perception and actual results was considered a judgment error. The aim of the present study was to examine relationships between subjective judgment errors and assessments of physical function, including a fear of falling.

SUBJECTS AND METHODS

The subjects were community-dwelling elderly individuals who had participated in physical fitness assessments for health promotion and guidance in a rural area of the Ibaraki Prefecture in Japan between 2012 and 2015. Individuals who felt unwell or were medically unstable were excluded from participation. For individuals who participated multiple times, the data from the initial participation was used in the analyses. A total of 106 subjects participated, aged 63 to 89 years (mean age ± SD: 74.5 ± 5.8), with 34 men and 72 women (mean age ± SD: 74.6 ± 6.6 and 74.4 ± 5.4, respectively). Eligible subjects were informed of the research purpose and procedures, and voluntarily provided written informed consent before the measurements were collected.

The following parameters of physical function for the elderly were assessed: 1) grip strength, 2) duration of the one-leg standing test (OLS) with open/closed eyes, 3) FR score, 4) time on the TUG, 5) comfortable and maximum gait speeds, 6) time on the SWOC, and 7) the Berg Balance Scale (BBS) score. The grip strength, duration of the OLS with open/closed eyes, FR score14), time on the TUG15), and BBS score17) were measured using standardized methods.

To determine the comfortable and maximum gait speeds, subjects were asked to walk 11 m (consisting of 3 m of acceleration, 5 m of the actual test distance, and 3 m of deceleration) at two different gait speeds: a comfortable speed and their maximum speed. The gait speeds were calculated using the test distance and walking time measured in the actual test phase.

Balance and ambulation were evaluated in a manner that approximated the activities of daily living (ADL). The SWOC consisted of a 12.2 × 0.92 m path with three directional changes and obstacles16). The subjects were asked to rise from a chair, walk along the course while negotiating the obstacles, and sit down on a chair placed on the other side of the course. The ambulation time through the SWOC was measured in seconds.

After the SWOC procedure was explained to the subject, the subject was shown video footage of a woman (age, 57 years; height, 145 cm) walking the SWOC. Her SWOC score was 12.3 seconds, which is approximately the average score for the elderly individuals we have assessed during the past few years (average SWOC score, 12.12 ± 3.11 seconds, based on data from 254 individuals [95 men and 159 women] with an average age of 76.2 ± 6.4 years). After watching the video, subjects were asked to compare their usual gait speed (that used at home) to the video model’s gait speed on the SWOC, using a Likert scale.

The one-year retrospectively recalled incidence of falls and the presence of a fear of falling were obtained from an interview with each subject, performed before the execution of the measurements. In addition, the Activities-specific Balance Confidence (ABC) Scale score18) was assessed. In the ABC Scale, subjects are asked to rate (0 to 100%) his/her confidence in performing specific activities without losing his/her balance18,19,20). The ABC Scale score is the average score of the 16-item questionnaire; a lower score indicates less confidence in one’s balance ability and is interpreted as the presence of a fear of falling18, 20). The ABC Scale is reported to be a more appropriate questionnaire, relative to the Fall Efficacy Scale, for assessing the fear of falling in community-dwelling elderly individuals with relatively a high physical function19, 20).

All participants were supervised and protected from falls or other hazards during testing by physical therapists or trained physical therapy students. This study was approved by the Ethics Committee of Ibaraki Prefectural University of Health Sciences (Approval No. 459).

Subjects were divided into two groups: fast walkers (those with a smaller SWOC score compared to that of the video model), and slow walkers (those with a larger SWOC score compared to that of the video model). Subjects were further divided based on their subjective perceptions: subjects who considered themselves to be faster than the model were classified as ‘assured’, while those who did not consider themselves to be faster than the model were classified as ‘modest’. Thus, subjects were categorized into four groups: fast-assured, fast-modest, slow-assured, and slow-modest.

The associations between age and the measured variables of interest were examined using the Pearson’s product-moment correlation coefficient. Two-way analysis of variance (ANOVA) was performed to estimate the effects of actual SWOC performance and subjective perceptions on the measured physical function variables. For variables strongly correlated with age, a two-way analysis of covariance (ANCOVA) with the age as a covariate was performed instead of an ANOVA. Post hoc analyses were performed with Bonferroni correction for multiple comparisons.

All statistical analyses were conducted using IBM SPSS Statistics for Windows version 22.0 (IBM Corp., Armonk, NY, USA). Test statistics were considered significantly different or related at an alpha level of 0.05.

RESULTS

Descriptive statistics are shown in Table 1. Based on our interviews and the survey on falls, 21.7% of subjects (23/106) had experienced falls during the past year, and 24.5% of subjects (26/106) had a fear of falling. Age was correlated with all measured variables, except the ABC scale score.

Table 1.

Descriptive statistics for the measured variables

	N(n=106)
Age (years)	74.5 ± 5.8
Height (cm)	153.2 ± 8.5
Body weight (kg)	55.5 ± 9.9
Grip strength (kg)	27.3 ± 8.3
One-leg standing time with open eyes (s)	32.02 ± 36.39
One-leg standing time with closed eyes (s)	4.48 ± 4.18
Functional Reach test (cm)	28.4 ± 6.1
Timed Up and Go test (s)	8.11 ± 1.71
SWOC score (s)	11.19 ± 2.45
Comfortable gait speed (m/s)	1.37 ± 0.24
Maximum gait speed (m/s)	1.85 ± 0.31
Activities-specific balance Confidence (ABC) Scale score (%)	77.5 ± 20.2
Berg Balance Scale score (points)	54.2 ± 2.0

Mean ± standard deviation

In terms of actual SWOC performance, 78 subjects were fast walkers and 28 were slow walkers. In terms of subjective perceptions, 24 subjects were “assured” and 82 were “modest”. The number of subjects in the four groups based on the combined factors of actual SWOC performance and subject perception were as follows: fast-assured, n=16; fast-modest, n=62; slow-assured, n=8; and slow-modest, n=20. Although the number of subjects did not appear to evenly distributed among the groups, a chi-square test revealed that the differences were not statistically significant (x=0.764, df=1, p=0.382).

The descriptive statistics for the measured variables stratified by group are shown in Table 2. Significant main effects of the actual SWOC performance, indicated by (*) in Table 2, were found for age, TUG, SWOC score, comfortable gait speed, maximum gait speed, ABC Scale score, and BBS score (F(1,102)=12.671, F(1, 101)=37.923, F(1, 101)=123.298, F(1, 101)=17.977, F(1, 101)=14.939, F(1, 102)=11.415, and F(1, 101)=9.991, respectively). Significant main effects of the subjective perception, indicated by (‡) in Table 2, were found for the OLS with open/closed eyes and SWOC score (F(1, 101)=6.958, F(1, 101)=4.142, and F(1, 101)=6.479, respectively). In addition, significant interaction effects between actual SWOC performance and subjective perceptions, indicated by (†) in Table 2, were found for the SWOC score and the ABC Scale score (F(1, 101)=8.282 and F(1, 102)=5.378, respectively). Subjects in the slow-assured group were significantly slower on the SWOC compared to subjects in the other three groups, and subjects in the slow-modest group were significantly slower on the SWOC compared to subjects in the fast-assured and fast-modest groups. However, there were no significant differences in SWOC scores between the fast-assured and fast-modest groups. In addition, the ABC Scale score was significantly lower in the slow-assured group compared to that in fast-assured and fast-modest groups. That is, the slow-assured group was significantly less self-confident in their balance ability compared to fast-assured and fast-modest groups.

Table 2.

Descriptive statistics stratified by group

	Fast walker		Slow walker
	Assuredn=16	Modestn=62	Assuredn=8	Modestn=20
Age (years)*	73.4 ± 4.5	73.2 ± 6.0	79.1 ± 3.9	77.4 ± 5.1
Grip strength (kg)	28.9 ± 8.0	27.7 ± 8.9	23.7 ± 5.9	26.3 ± 7.7
One-leg standing time with open eyes (s)‡	58.26 ± 46.02	31.94 ± 37.11	26.4 ± 24.13	13.53 ± 7.66
One-leg standing time with closed eyes (s)‡	6.93 ± 5.38	4.3 ± 4.25	4.33 ± 3.34	3.13 ± 2.18
Functional Reach test (cm)	28.1 ± 5.3	29.6 ± 5.7	26.2 ± 6.0	25.4 ± 7.1
Timed Up and Go test (s)*	7.52 ± 1.02	7.51 ± 1.41	10.66 ± 1.83	9.4 ± 1.33
SWOC score (s)†‡ *	9.98 ± 1.21	10.09 ± 1.45	15.83 ± 2.97	13.69 ± 1.09
Comfortable gait speed (m/s)*	1.46 ± 0.23	1.42 ± 0.22	1.06 ± 0.21	1.25 ± 0.16
Maximum gait speed (m/s)*	1.98 ± 0.31	1.92 ± 0.27	1.48 ± 0.33	1.68 ± 0.24
Activities-specific balance Confidence (ABC) Scale score (%)† *	83.6 ± 14.1	79.7 ± 18.8	55.9 ± 23.2	74.5 ± 23.0
Berg Balance Scale score (points)*	55.4 ± 0.6	54.4 ± 1.8	53.0 ± 2.5	53.1 ± 2.4

Mean ± standard deviation. *ANOVA or ANCOVA indicated a significant main effect of the actual SWOC performance (p<0.05). ‡ANOVA indicated a significant main effect of subjective perception (p<0.05). †ANOVA or ANCOVA indicated a significant interaction between actual SWOC score and subjective perception (p<0.05). SWOC score was significantly lower in the slow-assured groups compared to that in the other three groups, and was lower in the slow-modest group compared to the fast-assured/modest groups. The ABC Scale score was significantly lower in the slow-assured group compared to that in the fast-assured/modest groups.

DISCUSSION

Individuals must adopt appropriate strategies and behaviors compatible with their physical abilities for safe daily living. Failure to judge one’s own physical ability appropriately may result in inappropriate strategies or behavior, increasing the risk for accidents, including falls. Several studies have examined the relationships among aging, judgment errors regarding physical ability, and the incidence of falls3, 7). Judgment error is considered as one of the risk factors for falls.

Previous studies utilizing the mental chronometry method have reported that the duration or distance of an imagined movement corresponds poorly to the duration or distance of the actual movement in elderly individuals3, 7, 8). In the present study, we utilized a different method to evaluate judgment error in elderly individuals. Elderly subjects were asked to subjectively compare their gait speed to that of another person videotaped while walking the SWOC. Subjects then walked the same SWOC. Given the results of the “Physical Fitness and Athletic Ability Survey” conducted by a Japanese administrative agency in 201421), the subjects in the present study appear to have average Japanese elderly’s physical function (Table 1). Judgment errors (disagreement between subjective perceptions and actual SWOC performance) occurred in 70 subjects, with 36 subjects appropriately judging their gait speed relative to the video model.

Among fast walkers (those who actually walked the SWOC faster than the model), there were no differences in physical function between “assured” and “modest” individuals, except for the OLS with open/closed eyes and the SWOC score. Fast walkers with judgment errors (the fast-modest group) underestimated themselves. It is conceivable that such individuals seldom have urgent issues, as the nature of their judgment error would tend to play a safeguarding role. That is, individuals in this group would be favorable from a viewpoint of risk avoidance. However, in the long-term, such individuals may limit their own activities more than necessary because of their underestimation of themselves; this limitation in activity might further decrease their physical function and increase the risk of falls.

More difficulty might arise in slow walkers with judgment errors (the slow-assured group). Slow walkers had significantly poorer physical function compared with fast walkers in all measured variables, except grip strength, OLS with open/closed eyes, and FR (Table 2). Among slow walkers, “assured” individuals had a tendency for poorer physical function and a decreased self-confidence for balance compared to that for “modest” individuals. Thus, some of the elderly individuals overestimated their abilities even though their physical function was decreased. It is presumable these elderly individuals would be at high risk for accidents.

Interestingly, slow-assured elderly individuals indicated that the walking speed of the model was slower than their own, even though their actual SWOC performance and ABC Scale scores were the poorest among the four groups. A low score on the ABC Scales indicates a lack of self-confidence in one’s balance ability and is interpreted as a fear of falling18, 19). In the present study, even though slow-assured elderly individuals lacked self-confidence in their balance ability, they misjudged their walking speed as faster than another person. Why someone with less confidence in balance ability overestimates him/herself when compared to other person remains an open question. It is presumable that slow-assured individuals might restrict their activities because of their decreased self-confidence in their balance ability as indicated by the low ABC Scale scores. Low self-confidence in balance ability might be useful in reducing the total risk. However, it is also conceivable that a loss in self-confidence may result in elderly individuals staying at home all day, which may not only decrease physical strength and function, but also reduce the opportunity for numerous experiences, contributing to misjudgments when comparing their ability to that of others. Since the number of subjects in the slow-assured group was small, further research is needed. In addition, further research evaluating the relationship between the amount of activity and judgment errors is needed, as well as follow-up and prospective studies on fast-modest individuals.

It is possible that some individuals overestimate themselves during comparisons with other individuals because of competition and vanity. Such individuals might have a high risk of accidents in crowds since it would be difficult for them to judge the situation calmly. As mentioned in the introduction, there are many situations in daily life that require appropriate judgments involving the action of ourselves in relation to the action of others. In these situations, selection of the appropriate strategy must consider the speed and/or distance of strangers and vehicles while imaging the result of one’s action. Slow-assured individuals might utilize an inappropriate strategy when they are in such situations due to a difficulty in judging their own balance function relative to that of others. Therefore, slow-assured individuals might have a greater risk for accidents. Further research is needed regarding the relationship between accidents and judgment errors occurring in a comparison with another person.

One of the limitations of the present study concerns the lopsided distribution of subjects among the four groups. Although there were no statistically significant differences in the subject distribution, the number of subjects in the high-risk, slow-assured group was low. Although this result is desirable from a risk perspective, it is necessary to increase the number of subjects for further detailed study. In addition, subjective judgments might change when subjects are in a dynamic situation, such as during the SWOC. Further research regarding the relationship between overestimation/underestimation of one’s abilities and accidents is needed.

The results of the present study demonstrate that judgment errors involving a comparison between one’s own ability and that of another are common among elderly individuals, with some overestimating their abilities and some underestimating their abilities. Interestingly, those who overestimated their ability did so despite a lack of self-confidence in balance function.