Journal of Gerontology: MEDICAL SCIENCES Copyright 2008 by The Gerontological Society of America

2008, Vol. 63A, No. 1, 76–82

Square-Stepping Exercise and Fall Risk Factors in

Older Adults: A Single-Blind, Randomized
Controlled Trial
Ryosuke Shigematsu,1 Tomohiro Okura,2 Masaki Nakagaichi,3 Kiyoji Tanaka,2
Tomoaki Sakai,4 Suguru Kitazumi,1 and Taina Rantanen5

Downloaded from https://academic.oup.com/biomedgerontology/article/63/1/76/672319 by guest on 20 January 2024

1
Faculty of Education, Mie University, Japan.
2
Graduate School of Comprehensive Human Sciences, University of Tsukuba, Japan.
3
Research and Development Center for Higher Education, Nagasaki University, Japan.
4
Faculty of Social Studies, Doshisha University, Japan.
5
Department of Health Sciences and Finnish Centre for Interdisciplinary Gerontology, University of Jyväskylä, Finland.

Background. Decreased fitness of the lower extremities is a potentially modifiable fall risk factor. This study aimed to
compare two exercise programs—square-stepping exercise (SSE), which is a low-cost indoor program, and walking—for
improving the fitness of the lower extremities.
Methods. We randomly allocated 68 community-dwelling older adults (age 65–74 years) to either the SSE or walking
group (W group). During the 12-week regimen, the SSE group participated in 70-minute exercise sessions conducted
twice a week at a local health center, and the W group participated in outdoor supervised walking sessions conducted
weekly. The W group was instructed to increase the number of daily steps. Prior to and after the program, we obtained
information on 11 physical performance tests for known fall risk factors and 3 self-reported scales. The fall incidence was
followed-up for 8 months.
Results. At 12 weeks postregimen, significant differences were observed between the two exercise groups with respect
to leg power (1 item), balance (2 items), agility (2 items), reaction time (2 items), and a self-reported scale (1 item); the
SSE group demonstrated a marked improvement in the above-mentioned items with Group 3 Time interactions.
Significant time effects were observed in the tests involving chair stands, functional reach, and standing up from a lying-
down position without Group 3 Time interactions. During the follow-up period, the fall rates per person-year in the SSE
and W groups were 23.4% and 33.3%, respectively ( p ¼ .31).
Conclusion. Although further studies are required, SSE is apparently more effective than walking in reducing fall risk
factors, and it appears that it may be recommended as a health promotion exercise in older adults.

Key Words: Functional fitness—Walking—Fall risk—Health status.

W ALKING is a widely accepted exercise (1) and is used

as a means to develop functional fitness in population-
based fall prevention programs (2). However, older adults
METHODS
The Institutional Review Board of the Kawage Health
Center approved the research protocol. All persons provided
may experience difficulty in walking in unfavorable weather written informed consent prior to enrollment in the study.
conditions such as rain, wind, cold, or heat waves. Further- The study complied with the CONSORT (Consolidated
more, the fear of injury, disease, accident, and crime may Standards of Reporting Trials) checklist for randomized
prevent them from walking outdoors (3,4). controlled trials.
Considering that older adults face these situations in daily
life, we have attempted to develop a square-stepping exer- Participants
cise (SSE) that they can easily perform indoors, composed Persons aged 65–74 years (n ¼ 2164) were recruited from
of movements similar to walking (Figure 1) (5). Walking Kawage, Mie, Japan. A letter containing information
involves only forward-stepping movements, whereas SSE regarding the schedule of the exercise sessions was sent to
involves varied movements in multiple directions and is 700 noninstitutionalized persons (350 women and 350 men)
performed on a thin mat (100 3 250 cm) that is partitioned who were randomly selected community residents from the
into 40 squares (25 cm each). As suggested in previous town of Kawage. After consenting to participate, each
studies, corrective steps in certain directions are necessary person was randomly allocated to either the SSE or walking
for recovering balance after tripping in order to prevent a fall group (W group) by a public health nurse who used a com-
(6–8). Therefore, it appears logical to hypothesize that the puterized random number generation program in which the
functional ability of the lower extremities is improved to numbers 0 and 1 corresponded to the two groups, respec-
a greater extent with SSE than with regular walking; thus, tively. The walking and SSE sessions were conducted on
SSE is more effective in preventing falls. This study aimed different days. The presence of severe neurological or
to compare the effects of SSE and regular walking on the cardiovascular diseases or mobility-limiting orthopedic
fall risk factors in older adults. conditions was considered as an exclusion criterion.

76
 SSE AND FALL RISK FACTORS 77

Downloaded from https://academic.oup.com/biomedgerontology/article/63/1/76/672319 by guest on 20 January 2024

Figure 1. Top: Examples of the square-stepping exercise patterns in Elementary 1 and 2, Intermediate 1, and Advanced 3 categories. Bottom: Square-stepping exercise.

Outcome Measures frequency (4-point Likert scale: 1 ¼ not at all, 2 ¼ once or

At baseline, the persons completed a questionnaire on twice a month, 3 ¼ once a week, and 4 ¼ two or more times
vision (5-point Likert scale: 1 ¼ poor and 5 ¼ excellent, a week); and occurrence of falls in the previous year (yes or
higher scores indicate better vision) (9,10); common medi- no). In addition, body weight and height were measured.
cal conditions (from among 21 possible common medical Body fat was estimated by bioimpedancemetry (HBF-354;
conditions such as cerebrovascular disease, hypertension, Omron Healthcare Co., Ltd., Kyoto, Japan).
and heart disease); medication use (yes or no); exercise The physical performance tests for the fall risk factors
78 SHIGEMATSU ET AL.

were adopted from previous studies and included the fol- detailed description of the SSE method has been provided in
lowing items: number of chair stands in 30 seconds (11), leg another study (5). In brief, SSE was performed on a thin felt
extension power (12), single-leg balance with eyes closed mat (100 3 250 cm) that was partitioned into 40 squares (25
(5), functional reach (13), forward/backward tandem walk- cm each). The persons were instructed to walk (step) from
ing over a 20-foot distance (14), standing up from a lying- one end of the mat to the other according to the step pattern
down position (5), stepping with both feet in 10 seconds provided (Figure 1). When the persons reached the end of
[persons stepped as quickly as possible for 10 seconds by the mat, they were instructed to return to their start positions
using a 60 3 55 cm stepping sheet (TKK 5301; Takei by walking normally off the mat and then stand in line for
Scientific Instruments Co. Ltd., Niigata, Japan)], walking the next stepping. The SSE included forward, backward,
around two cones (5), vertical jump reaction time after a lateral, and oblique step patterns. After the persons became

Downloaded from https://academic.oup.com/biomedgerontology/article/63/1/76/672319 by guest on 20 January 2024

light signal (simple reaction time) (15), and weight transfer familiar with each of these step patterns, they were
time recorded while stepping in the forward/backward/right/ instructed to walk with their heels lifted, that is, on their
left direction after a light signal (choice reaction time) (16). toes, without treading on the frames of the squares. Each
These tests were conducted by individuals who were un- step pattern was repeated 4–10 times to ensure that the
aware of the study group assignment (such as public health persons could complete the pattern, and was followed by the
nurses other than those involved in the randomization, introduction of a more complex step pattern. In total, 196
exercise instructors other than those who served in the step patterns were developed and categorized (based on
regimens, and university students who had specialized in progressively increasing levels of complexity) into 8
exercise gerontology). Each test was measured by the same categories (Elementary, 1–2; Intermediate, 1–3; and Ad-
staff preregimen and postregimen. vanced, 1–3). The persons were encouraged to concentrate
Self-reported scales consisted of the fear of falling (17), in order to successfully perform each progressively more
perceived health status (18), and pleasure during exercise complicated step pattern. Step cadence was not determined;
(using a line scale: left end ¼ not pleasant ‘‘0’’ and right therefore, the persons performed the pattern at their
end ¼ very pleasant ‘‘100’’; higher scores indicate consider- preferred pace. Although they required 15–20 seconds to
able pleasure). complete each step pattern initially, they eventually
The occurrence of falls and trips was also measured completed each pattern in , 15 seconds.
during the 8-month follow-up period at the end of the The persons in the W group were instructed to attend an
program. A fall was defined as a sudden unintentional outdoor supervised walking session at the Kawage Health
change in position that caused an individual to land at Center once a week for 12 weeks. These sessions were
a lower level, that is, on an object, the floor, or the ground, structured in a manner similar to that of the SSE sessions
due to reasons other than sudden-onset paralysis, epileptic except that SSE was substituted with a long-distance 40-
seizures, or overwhelming external forces (19). A trip was minute outdoor walking session. Furthermore, the W group
defined as the act of stumbling over an object without was also instructed to increase the number of daily steps,
landing on any part of the body. Trips may cause false- particularly during long-distance walking.
positive results because some individuals may report a trip The SSE (n ¼ 32) and W (n ¼ 36) groups were further
as a fall (20); therefore, the persons were explained the divided into 2 subgroups (n ¼ 16 and 18 for the SSE and W
difference between a fall and a trip and were instructed to subgroups, respectively), and the respective sessions were
record the occurrence of falls and trips separately on a daily conducted for each subgroup from December 2004 through
basis. All the persons received a prepaid postcard at the February 2005 (winter season). These sessions were always
beginning of each month, which they returned at the supervised by the same instructors who were certified in first
beginning of the subsequent month. A telephonic or face-to- aid and were encouraged to report any negative signs or
face interview was conducted to ascertain the reported symptoms that they observed in the persons during the
occurrence of falls and trips. sessions due to the exercises.
Pedometers (Walking Style HJ-710IT; Omron Healthcare
Co.) were provided to the persons of each group one week
prior to the study. During the first week, as a pre-regimen, Statistical Analysis
the persons were instructed to continue their routine daily An outcome analysis was performed using the intention-
activities and were advised against performing any new to-treat principle, and only two-tailed tests of significance
exercises. During the period between preregimen and were used. All baseline characteristics were compared
postregimen, they were instructed to wear the pedometers between the groups by using the Student t test except for
at all times when awake except when bathing. The recorded sex, vision, medications (proportion of medicated persons),
number of steps also included those completed during all the exercise frequency, and falls in the last year because these
exercise sessions. characteristics were assessed using the chi-square test.
Analysis of covariance (ANCOVA) was used to determine
Exercise Regimen the effect of the exercise program on each of the outcome
The SSE group participated in the supervised group measures by using the baseline characteristics as covariates.
sessions twice a week over the 12-week period at the For both groups, the proportional hazards models were used
Kawage Health Center; each session comprised 15 minutes to determine the relative hazards associated with the first
of warm-up activities such as stretching and calisthenics, 40 fall, and these relative hazards were calculated using the
minutes of SSE, and 15 minutes of cool-down activities. A Cox model. SPSS 11.5 software (SPSS Inc., Chicago, IL)
 SSE AND FALL RISK FACTORS 79

Table 2. Daily Steps from One Week Before

Regimen and the End of Regimen
Measurement SSE W
Pre-regimen 7548 6 453 5060 6 468
Weeks 1–4 7404 6 493 6972 6 509
Weeks 5–8 7124 6 486 6773 6 503
Weeks 9–12 6084 6 505 7732 6 522
Notes: Values express mean 6 standard error. Time effect and Group 3
Time interaction were both significant (p , .001).
SSE ¼ square-stepping exercise; W ¼ walking.

Downloaded from https://academic.oup.com/biomedgerontology/article/63/1/76/672319 by guest on 20 January 2024

Adherence and Adverse Effects of the Intervention
The persons in the SSE and W groups attended 21.8 6
2.9 of 24 sessions (90.9% 6 12.1%) and 9.3 6 2.6 of 11
sessions (84.2% 6 23.7%), respectively (p ¼ .15). None of
the persons in the SSE group dropped out of the study,
Figure 2. Stagewise progress of the square-stepping exercise and walking whereas five persons of the W group did (Figure 2). Of
groups, including the flow of persons and withdrawals. these, one of the male persons had developed knee pain
due to twisting of the knee during a daily activity (not due
was used for all statistical analyses. A value of p , .05 was to the prescribed walking regimen). The SSE persons con-
accepted as significant. scientiously performed SSE for 40 minutes throughout the
regimen. No adverse events such as falls or episodes of fear
were experienced by the persons during the sessions. All the
RESULTS persons completed the 8-month follow-up.

Baseline Characteristics and Daily Physical Activity Outcomes

Of the 70 older adults who consented to participate in the
The preregimen and postregimen group statistics and
study, 68 were assessed as eligible for the study (Figure 2).
Group 3 Time interactions are presented in Tables 3 and 4.
Both the SSE and W groups were comparable and well
After the 12-week regimen, significant differences were
matched with regard to the baseline characteristics (Table 1).
observed between the two exercise groups with respect to leg
The time effect and Group 3 Time interaction of the daily
extension power, forward/backward tandem walking, step-
steps were both significant. The means 6 standard errors
ping with both feet, walking around two cones, simple/choice
of the daily average were 7548 6 453 and 5060 6 468 at
reaction time, and perceived health status with significant
pre-regimen, 7404 6 493 and 6972 6 509 at weeks 1–4,
Group 3 Time interactions; the SSE group persons demon-
7124 6 486 and 6773 6 503 at weeks 5–8, and 6084 6
strated a marked improvement in the above-mentioned test
505 and 7732 6 522 at weeks 9–12 for the SSE and W
items. Significant time effects without Group 3 Time
groups, respectively (Table 2).
interactions were observed for three items, that is, chair
Table 1. Baseline Characteristics of Study Participants by
stands, functional reach, and standing up from a lying-down
Randomized Groups position; persons of both the groups demonstrated a marked
improvement in these tests. Although the number of steps at
Characteristic SSE W p pre-regimen was statistically higher for the SSE group than
Female, n (%) 18 (56) 25 (69) .32 for the W group, when this number was included in the
Age, y 68.6 6 2.4 69.5 6 2.9 .18 analyses as a covariate, the results remained unchanged.
Body weight, kg 59.3 6 11.2 55.6 6 7.6 .13 During the 8-month follow-up period, five falls in four
Height, cm 157.3 6 10.0 154.4 6 6.6 .17
persons of the SSE group (fall rate per person-year, 23.4%)
Body fat, % 30.1 6 5.7 31.5 6 5.5 .30
Vision* 2.77 6 0.80 3.09 6 0.61 .08 and eight falls in seven persons of the W group (fall rate per
Common medical conditions, ny 0.97 6 0.97 0.89 6 1.11 .75 person-year, 33.3%; p ¼ .31) were reported. During the
Medications, n (%) 20 (63) 16 (47) .23 same period, 46 and 60 trips were recorded in the SSE and
Exercise frequencyz 2.67 6 1.37 2.76 6 1.42 .78 W groups, respectively, indicating that the rate of falls per
Falls in the last year, n (%) 8 (26) 5 (15) .36 trip reported [fall/(fall þ trip)] in the SSE group (9.8%) was
Notes: Values express mean 6 standard deviation. ‘‘Female,’’ ‘‘Medications,’’ not significantly lower than that in the W group (11.8%; p ¼
and ‘‘Falls’’ refer to the number (percentage) of participants who were female, .50). The hazard ratio of the W group to the SSE group with
used one or more medications, or had one or more falls in the last year. respect to the first fall was 2.32 (95% confidence interval
*Measured on a 5-point Likert scale: 1 ¼ poor and 5 ¼ excellent; higher
[CI], 0.59–9.04; p ¼ .23).
scores indicate better vision.
y
Measured of 21 possible common medical conditions (e.g., cerebrovascular
disease, hypertension, and heart disease).
z
Measured on a 4-point Likert scale: 1 ¼ not at all, 2 ¼ 1 or 2 d/mo, 3 ¼ 1 d/ DISCUSSION
wk, and 4 ¼ 2 d/wk. This single-blind randomized controlled trial was de-
SSE ¼ square-stepping exercise; W ¼ walking. signed to examine whether SSE, which is a novel exercise
80 SHIGEMATSU ET AL.

Table 3. Functional Fitness Items by Group at Preregimen and Postregimen

Crude Effect Adjusted Effect Group 3 Time Time Effect
Item Preregimen Postregimen (95% CI) (95% CI)* Interaction p Value* p Value*
Leg strength and power
Chair stands, n  30 s1
SSE 14.6 6 0.5 15.8 6 0.5 1.2 (0.4 to 2.0) 1.2 (0.5 to 2.0) .42 , .001
W 14.7 6 0.5 16.3 6 0.5 1.6 (1.0 to 2.3) 1.7 (0.9 to 2.4)
Total 14.7 6 0.3 16.1 6 0.3 1.4 (0.9 to 1.9) 1.4 (0.9 to 1.9)
Leg extension power, W
SSE 318.2 6 21.8 343.0 6 19.5 24.8 (1.3 to 48.4) 27.4 (6.9 to 47.9) .03 .14

Downloaded from https://academic.oup.com/biomedgerontology/article/63/1/76/672319 by guest on 20 January 2024

W 256.0 6 21.4 253.1 6 19.1 2.9 (20.5 to 14.7) 5.5 (26.0 to 15.0)
Total 286.6 6 15.7 297.3 6 14.7 10.7 (3.9 to 25.4) 10.9 (3.8 to 25.6)
Balance
Single-leg balance with eyes closed, s
SSE 9.1 6 1.2 9.9 6 1.7 0.8 (2.2 to 3.8) 0.8 (1.8 to 3.4) .99 .39
W 7.9 6 1.2 8.7 6 1.7 0.7 (1.6 to 3.0) 0.8 (1.9 to 3.5)
Total 8.5 6 0.9 9.3 6 1.2 0.8 (1.1 to 2.6) 0.8 (1.0 to 2.6)
Functional reach, cm
SSE 27.8 6 0.9 31.4 6 0.7 3.6 (1.6 to 5.6) 3.5 (1.9 to 5.2) .06 , .001
W 29.5 6 0.9 30.6 6 0.7 1.1 (0.1 to 2.3) 1.2 (0.5 to 2.9)
Total 28.7 6 0.6 31.0 6 0.5 2.4 (1.2 to 3.5) 2.4 (1.2 to 3.6)
Forward tandem walking, s
SSE 21.1 6 0.8 16.8 6 0.8 4.3 (2.8 to 5.9) 4.3 (2.6 to 5.9) .01 , .001
W 19.0 6 0.8 18.1 6 0.8 0.8 (0.9 to 2.5) 1.0 (0.7 to 2.6)
Total 20.0 6 0.6 17.5 6 0.6 2.6 (1.4 to 3.8) 2.6 (1.4 to 3.8)
Backward tandem walking, s
SSE 26.3 6 1.2 21.2 6 1.5 5.2 (3.1 to 7.3) 5.1 (2.2 to 8.0) .03 .01
W 24.3 6 1.2 23.9 6 1.5 0.4 (3.1 to 3.8) 0.4 (2.6 to 3.5)
Total 25.3 6 0.8 22.5 6 1.1 2.8 (0.8 to 4.9) 2.9 (0.7 to 5.0)
Agility
Standing up from a lying-down position, s
SSE 3.48 6 0.27 3.19 6 0.28 0.30 (0.02 to 0.60) 0.26 (0.02 to 0.51) .86 .01
W 3.58 6 0.27 3.33 6 0.28 0.26 (0.07 to 0.44) 0.30 (0.04 to 0.55)
Total 3.53 6 0.19 3.26 6 0.19 0.27 (0.10 to 0.45) 0.28 (0.11 to 0.45)
Stepping with both feet, n  10 s1
SSE 50.9 6 2.1 60.7 6 1.9 9.8 (6.9 to 12.7) 10.1 (7.6 to 12.7) .04 , .001
W 50.6 6 2.2 57.1 6 2.0 6.6 (3.9 to 9.2) 6.2 (3.6 to 8.9)
Total 50.7 6 1.5 59.0 6 1.4 8.2 (6.3 to 10.2) 8.2 (6.4 to 10.1)
Walking around two cones, s
SSE 24.0 6 0.7 21.3 6 0.6 2.7 (1.4 to 3.9) 2.7 (1.7 to 3.7) .03 , .001
W 21.9 6 0.7 20.9 6 0.6 1.0 (0.4 to 1.7) 1.0 (0 to 2.0)
Total 22.9 6 0.5 21.1 6 0.4 1.8 (1.1 to 2.5) 1.9 (1.1 to 2.6)
Reaction
Simple reaction time, 1000 ms1
SSE 461 6 14 426 6 12 35 (11 to 60) 34 (11 to 57) , .001 .94
W 419 6 14 453 6 12 34 (54 to 13) 32 (55 to 9)
Total 440 6 10 439 6 9 1 (17 to 19) 1 (17 to 18)
Choice reaction time, 1000 ms1
SSE 982 6 15 920 6 14 62 (35 to 89) 60 (37 to 84) , .001 .01
W 938 6 15 954 6 14 16 (36 to 4) 14 (39 to 11)
Total 961 6 11 936 6 10 25 (5 to 44) 25 (6 to 44)
Notes: Values in preregimen and postregimen indicate mean 6 standard error.
*Adjusted for baseline characteristics as shown in Table 1.
SSE ¼ square-stepping exercise; W ¼ walking; CI ¼ confidence interval.

program, was more effective than regular walking in functional fitness of the lower extremities—was improved to
improving the functional fitness of the lower extremities in a greater extent in the SSE group than in the W group.
older adults. After the 12-week regimen, we observed that Furthermore, the perceived health status was significantly
one of the most common risk factors for falls (9)—the improved in the SSE group. Our study provides new
 SSE AND FALL RISK FACTORS 81

Table 4. Self-Reported Items by Group at Preregimen and Postregimen

Crude Effect Adjusted Effect Group 3 Time Time Effect
Item Preregimen Postregimen (95% CI) (95% CI)* Interaction p Value* p Value*
Fear of falling*
SSE 2.00 6 0.11 2.22 6 0.12 0.22 (0.03 to 0.47) 0.21 (0.03 to 0.46) .91 .35
W 2.06 6 0.11 2.28 6 0.12 0.22 (0.01 to 0.44) 0.23 (0.01 to 0.48)
Pleasure during exercisey
SSE 72.3 6 5.0 90.6 6 3.1 18.3 (9.9 to 26.7) 18.9 (10.2 to 27.7) .20 .43
W 78.4 6 4.9 89.3 6 3.1 10.9 (2.5 to 19.3) 10.6 (1.7 to 19.5)
Perceived health statusz

Downloaded from https://academic.oup.com/biomedgerontology/article/63/1/76/672319 by guest on 20 January 2024

SSE 2.75 6 0.13 3.28 6 0.14 0.53 (0.11 to 0.95) 0.58 (0.24 to 0.91) .002 .01
W 2.81 6 0.13 2.69 6 0.14 0.13 (0.34 to 0.09) 0.21 (0.55 to 0.13)
Notes: Values in preregimen and postregimen indicate mean 6 standard error.
*Measured on a 3-point Likert scale: 1 ¼ very fearful, 2 ¼ fearful, and 3 ¼ not fearful.
y
Measured using a line scale: left end ¼ not pleasant; ‘‘0’’ and right end ¼ very pleasant; ‘‘100’’; higher scores indicate considerable pleasure.
z
Measured on a 5-point Likert scale: 1 ¼ poor, 2 ¼ fair, 3 ¼ good, 4 ¼ very good, and 5 ¼ excellent.
CI ¼ confidence interval; SSE ¼ square-stepping exercise; W ¼ walking.

evidence that SSE is a more useful exercise program than exercises is important for health promotion. Third, SSE
regular walking for older adults; thus, it may serve as a new requires minimum investment because it involves the use
form of exercise to prevent falls. of low-tech equipment. Fourth, because of the significantly
A study by Orr and colleagues (21) revealed that leg- small reaction time, which is a cognitive function, SSE may
strengthening exercises at light loads (20% of maximal improve information-processing speed and psychomotor
strength) improve balance because they ensure that the processes (23). Based on the results of the current study,
muscles remain active throughout the concentric phase of we suggest that the variety of step patterns and the level
the movement and maintain the level of force output. The of muscle coordination involved in SSE make it more
exercise intensity and movement in the above-mentioned beneficial than regular walking in reducing fall risk factors.
study were rather similar to those of our step exercises, in- This observation supports the well-known principle re-
cluding the slight extension of the knees and ankles. A leg garding specificity of training as the skills targeted by the
exercise such as this is assumed to enhance neural function exercise program were improved. However, walking is
by reducing response latency, effectively recruiting postural known to have beneficial effects on balance and gait speed
muscles, and improving the interpretation of sensory infor- (24) as well as on cardiorespiratory fitness, blood pressure,
mation (21). In addition, the multidirectional steps in the and cholesterol levels (25), which were not assessed in this
forward, backward, lateral, and oblique directions during study. Furthermore, from the pedometer readings, we ob-
SSE lead to better activation of the synergist and agonist leg served that walking could increase the amount of physical
muscles. Therefore, it is possible that the SSE regimen activity even during the winter season, whereas SSE might
consequently improves many aspects of the functional decrease it. Therefore, walking could still be recommended
fitness of the lower extremities, which is a fall risk factor. as a health-enhancing form of exercise in older adults.
After the SSE persons were familiar with the step This study has notable limitations. First, although fall risk
patterns, they were instructed to walk with their heels lifted. factors were lowered in the SSE group, fall rates were not
This movement, which involved small hopping steps, also different in the two intervention groups. Furthermore, the
improved their leg strength (22). A study by Pijnappels and occurrence of falls among the elderly adults in each group
colleagues (8) revealed that during a trip, when the balance was not very high. The possible reasons for this observa-
of one leg is lost, the other leg is immediately lifted off the tion may be the short follow-up period and the inclusion
floor, in a manner similar to hopping, in order to prevent of persons with a low fall risk. Second, the statistical anal-
a fall. This mechanism can explain the reason for the lower ysis of each of the 15 outcome measures, including physical
number of falls observed in the SSE group than in the W performance tests, self-reported scales, and fall occur-
group, as the former appears to have adequate functional rence, was performed separately; therefore, there was an
ability to prevent falls. increased risk of false-positive findings (type 1 error). The
The results of this study imply that SSE could be used as self-reported scales would not be adequately sensitive to
a means of rehabilitation and public health promotion a change because each of these scales contained a single
because it has a number of advantages. First, it is possible item. Third, the pedometer readings revealed that the
for fewer staff members (including physicians, public health number of steps in the SSE group was smaller than that in
nurses, and exercise instructors) to simultaneously supervise the W group, although our finding is that SSE as an exercise
several older adults with high risk of falling because SSE form has a favorable effect on fall risk factors. However, the
can be performed within a small indoor space. Second, intensity of the walking regimen, which was not recorded in
outdoor walkers can substitute walking with SSE when it this study, might not be sufficient to reduce the fall risk. In
rains. In this context, our study proposed a new form of addition, we did not attempt to standardize the amount of
exercise for older adults. Increasing the number of feasible daily activity in the two groups, that is, the W group persons
82 SHIGEMATSU ET AL.

were advised to walk by themselves without supervision; 8. Pijnappels M, Bobbert MF, van Dieen JH. Push-off reactions in
thus, the effects may reflect the persons’ interpretation of recovery after tripping discriminate young subjects, older non-fallers
and older fallers. Gait Posture. 2005;21:388–394.
and compliance with the prescribed exercise regimen. In the 9. American Geriatric Society, British Geriatric Society, and American
same 12-week period, the supervised SSE sessions were Academy of Orthopaedic Surgeons Panel on Falls Prevention.
conducted twice a week, whereas the supervised walking Guideline for the prevention of falls in older adults. J Am Geriatr
sessions were conducted only once a week; therefore, Soc. 2001;49:664–672.
10. Lord SR. Visual risk factors for falls in older people. Age Ageing.
unfortunately, we cannot rule out that better participation in 2006;35–S2:ii42–ii45.
the supervised SSE sessions may have improved the 11. Rikli RE, Jones CJ. Development and validation of a functional fitness
outcome in the SSE group. test for community-residing older adults. J Aging Phys Act. 1999;
7:129–161.
12. Bassey EJ, Short AH. A new method for measuring power output in

Downloaded from https://academic.oup.com/biomedgerontology/article/63/1/76/672319 by guest on 20 January 2024

Conclusion
a single leg extension: feasibility, reliability and validity. Eur J Appl
The findings suggest that SSE is safe and acceptable, and Physiol. 1990;60:385–390.
it improves the functional fitness of the lower extremities, 13. Shigematsu R, Tanaka K. Age scale for assessing functional fitness in
which is a fall risk factor, in older adults. The efficacy of older Japanese ambulatory women. Aging Clin Exper Res. 2000;
this exercise in improving perceived health status was 12:256–263.
14. Nelson ME, Fiatarone MA, Morganti CM, Trice I, Greenberg RA,
also substantiated. Therefore, this new activity apparently Evans WJ. Effects of high-intensity strength training on multiple risk
provides an effective, therapeutic, and health-promoting factors for osteoporotic fractures. A randomized controlled trial. JAMA.
exercise alternative. 1994;272:1909–1914.
15. Morozumi K, Yamamoto I, Fujiwara T, et al. Effect of dentures
wearing on motor reaction time and balance function in elderly people.
ACKNOWLEDGMENTS J Physiol Anthropol Appl Human Sci. 2004;23:129–137.
This study was supported by a grant from the Ministry of Education, 16. Lord SR, Fitzpatrick RC. Choice stepping reaction time: a composite
Culture, Sports, Science and Technology (Encouragement of Young measure of falls risk in older people. J Gerontol Med Sci. 2001;
Scientists, B) and from the Hayao Nakayama Foundation for Science & 56A:M627–M632.
Technology and Culture. 17. Myers AM, Powell LE, Maki BE, Holliday PJ, Brawley LR, Sherk W.
Psychological indicators of balance confidence: relationship to actual
and perceived abilities. J Gerontol Med Sci. 1996;51A:M37–M43.
CORRESPONDENCE 18. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health
Address correspondence to Ryosuke Shigematsu, PhD, Faculty of survey (SF-36). I. Conceptual framework and item selection. Med Care.
Education, Mie University, Kurimamachiya 1577, Tsu, Mie, 514-8507 1992;30:473–483.
Japan. E-mail: [email protected] 19. Feder G, Cryer C, Donovan S, Carter Y. Guidelines for the prevention
of falls in people over 65. Br Med J. 2000;321:1007–1011.
20. Lamb SE, Jorstad-Stein EC, Hauer K, et al. Development of a common
REFERENCES outcome data set for fall injury prevention trials: the Prevention of Falls
1. Simonsick EM, Guralnik JM, Fried LP. Who walks? Factors associated Network Europe Consensus. J Am Geriatr Soc. 2005;53:1618–1622.
with walking behavior in disabled older women with and without self- 21. Orr R, de Vos NJ, Singh NA, Ross DA, Stavrinos TM, Fiatarone-Singh
reported walking difficulty. J Am Geriatr Soc. 1999;47:672–680. MA. Power training improves balance in healthy older adults. J
2. McClure R, Turner C, Peel N, Spinks A, Eakin E, Hughes K. Gerontol A Biol Sci Med Sci. 2006;61A:78–85.
Population-based interventions for the prevention of fall-related injuries 22. Menz HB, Morris ME, Lord SR. Foot and ankle characteristics
in older people. Cochrane Database Syst Rev. 2005;(1):CD004441. associated with impaired balance and functional ability in older people.
3. Weinstein A, Feigley P, Pullen P, Mann L, Redman L. Neighborhood J Gerontol A Biol Sci Med Sci. 2005;60A:1546–1552.
safety and the prevalence of physical inactivity—selected states, 1996. 23. Spirduso WW. Health, exercise, and cognitive-function. In: Spirduso
MMWR Morb Mortal Wkly Rep. 1999;48:143–146. WW, ed. Physical Dimensions of Aging. Champaign, IL: Human
4. Tu W, Stump TE, Damush TM, Clark DO. The effects of health and Kinetics; 1995:249–277.
environment on exercise-class participation in older, urban women. 24. Buchner DM, Cress ME, De Lateur BJ, et al. A comparison of the
J Aging Phys Act. 2004;12:480–496. effects of three types of endurance training on balance and other fall
5. Shigematsu R, Okura T. A novel exercise for improving lower- risk factors in older adults. Aging Clin Exp Res. 1997;9:112–119.
extremity functional fitness in the elderly. Aging Clin Exper Res. 25. Mazzeo RS, Cavanagh P, Evans WJ, et al. ACSM position stand:
2006;18:242–248. exercise and physical activity for older adults. Med Sci Sports Exerc.
6. Blake AJ, Morgan K, Bendall MJ, et al. Falls by elderly people 1998;30:992–1008.
at home: prevalence and associated factors. Age Ageing. 1998;17:
365–372.
7. Madigan ML, Lloyd EM. Age and stepping limb performance dif- Received October 16, 2006
ferences during a single-step recovery from a forward fall. J Gerontol A Accepted April 20, 2007
Biol Sci Med Sci. 2005;60A:481–485. Decision Editor: Luigi Ferrucci, MD, PhD