System of Reporting and Comparing Influence of

Ambulatory Aids on Gait

GARY L. SMIDT, PhD,

and M. A. MOMMENS, MS

The purposes of this study were to 1) present a standardized approach for

describing gait when assistive devices are used, 2) report reference data for
unassisted and assisted gait patterns for normal adults, and 3) discuss clinical
implications for selected variables of gait. Using an automated gait system,
measurements for temporal and distance factors and accelerometry were ob-
tained for 25 normal young adults. In addition to the formulation of a new system
for describing gait patterns when assistive devices are used, the results of the
study were that 1) subjects walked slower with ambulatory aids than without
them, 2) assisted gaits with the same number of counts per cycle tended to
have similar measurements, 3) reciprocal swing times and stance times were
symmetrical for all types of gait studied, 4) double stance times and step times
were asymmetrical for three types of assisted gait, and 5) vertical accelerations
were disproportionately elevated for most assisted gaits.

Key Words: Crutches, Gait, Physical therapy.

Assistive devices such as canes, crutches, and walk-
ers are commonly recommended for problems of
pain, fatigue, equilibrium, joint instability, muscular
weakness, excessive skeletal loading, and cosmesis.
The physical therapist may evaluate the patient for
abnormalities and identify the form of assistive device
to be used and type of assisted gait to be learned. The
clinician may judge the quality of the patient's walk-
ing performance using standards such as precon-
ceived expectations of normality, information pre-
sented in scientific publications, or his clinical expe-
rience in dealing with manifestations of disorders.
Numerous reports on ambulatory aids and gait
may be found in the literature. 1-7 A few articles have
reported the effect of a cane or a crutch on patients'
walking.8"15 Data on the unassisted gait of normal
subjects may be compared with patients who walk
without an assistive device, in order to determine the
Dr. Smidt is Professor and Director of Programs in Physical
Therapy Education, College of Medicine, The University of Iowa,
Iowa City, IA 52242.
Ms. Mommens was a doctoral student in physical therapy, The
University of Iowa, when this study was conducted. She is now self-
employed and can be reached at 707 9th Ave, Coralville, IA 52241.
This paper was presented at the Fifty-second Annual Conference
of the American Physical Therapy Association, New Orleans, June-
July 1976.
This article was submitted October 3, 1978, and accepted November
7, 1979.
extent of the patients' abnormality. No information
is available, however, on normal gait for walking
patterns associated with the variety of commonly
employed types of assisted gait.
Inasmuch as humans walk by reciprocally placing
each of their two feet in front of the other in forward
progression, the sequence and number of contacts
between the body and the walking surface are con-
fined to any combination of two. When one or two
ambulatory aids are added, the complexity of the
sequencing and timing of floor contact is magnified.
For assisted gait, the number of contact points is thus
greater than two, and necessarily the possible order
of contacts on the walking surface is exponentially
increased. Walking techniques with canes and
crutches have been described by Hoberman,14 but
explanations for assisted gait patterns could be im-
proved to be more explicit and to convey a clear
mental image of the walking pattern. One example of
ambiguity is that a two-point gait may refer to the
use of either one or two assistive devices, and these
devices may be canes or crutches. A standardized
method that accurately describes the type of assisted
gait is needed.
The purposes of this paper are to 1) present a
standardized approach for describing gait when assis-
tive devices are used, 2) report reference data for

Volume 60 / Number 5, May 1980 551

Downloaded from https://academic.oup.com/ptj/article-abstract/60/5/551/2727029

by guest
on 06 March 2018
 unassisted and assisted gait patterns, and 3) discuss
clinical implications for selected variables of gait.
DEFINITIONS OF GAIT VARIABLES
For purposes of this study, the following definitions
were used.
Walking Velocity: The rate of linear forward mo-
tion of the body; Product of distance and time; Mea-
sured in centimeters per second or meters per second.
Stance Time: The elapsed time that one foot is in
contact with walking surface; Measured in seconds.
Swing Time: The elapsed time that one foot is not
in contact with walking surface; Measured in seconds.
Gait Cycle: Usually considered the interval be-
tween successive ipsilateral foot contacts on the walk-
ing surface, but may be the interval between any
recurring event during walking.
Cycle Time: The elapsed time during one gait cycle;
Measured in seconds.
Double-Stance Time: The elapsed time during the
gait cycle when both feet are simultaneously in con-
tact with the walking surface; Measured in seconds.
Stride Length: The distance between two consecu-
tive ipsilateral foot contacts on the walking surface;
Measured in centimeters or meters.
Step Length: The distance in the direction of walk-
ing between the left and right feet; Measured in
centimeters or meters.
Step Time: The elapsed time between consecutive
foot-floor contacts; Measured in seconds.
Functional Lower Extremity Length: The distance
between the center of the femoral head and a point
on the floor located slightly anterior and medial to
the medial malleolus on the ipsilateral side; Measured
in centimeters or meters.
Ratios Reflecting Symmetry: Mathematically de-
rived quotients obtained by dividing a measurement
associated with one side of the body by a measure-
ment for the same variable associated with the other
side of the body.
a. Stride Length/Lower Extremity Length Ratio:
Stride length divided by lower extremity length;
Measurement unit is dimensionless.
b. Swing/Time Ratio: Right swing time divided by
left swing time or, in a disabled individual, most
involved side divided by least involved side; Mea-
surement unit is dimensionless.
c. Swing/Stance Ratio: Swing time divided by stance
time; Measurement unit is dimensionless.
Other ratios for which measurements were obtained
during the course of this study were stance time,
double-stance time, step time, step length, heel strike-
foot flat time, foot flat-heel off time, and heel off-toe
off time.
Body Acceleration: The rate of change of velocity
of a point posterior to the sacrum; Measured in meters
per second per second. Maximum values for acceler-
ation in vertical, medial-lateral, and fore-aft direc-
tions were obtained in the study.
Harmonic Ratio: An index of smoothness of gait.
A Fourier series analysis applied to the acceleration
curves yields even- and odd-numbered coefficients.
The harmonic ratio is the quotient derived from the
sum of the coefficients for the even-numbered har-
monics divided by the sum of the coefficients for the
odd-numbered harmonics.
Cadence (stride or step frequency): The number of
strides or steps per unit of time; Number of steps
divided by time; Measured in strides or steps per
minute or steps per second.
METHOD
A recently developed automated gait analysis sys-
tem was used to obtain measurements for a large
number of variables of gait.15 The system included
triaxial accelerometers, placed posterior to the sac-
rum, which were sensitive to changes in velocity in
the anterior-posterior, medial-lateral, and vertical di-
rections. The system also included pressure-sensitive
foot switches attached to the heel and forefoot of each
shoe, a signal-amplification unit, and a laboratory
computer. Graduated strips of tape on the walkway
permitted acquisition of step-distance measurements.
After each walking sequence in the laboratory, mea-
surement of temporal and distance factors (eg, walk-
ing velocity, step length, step time) and body accel-
eration could be viewed on the teletype. A detailed
Fig. 1. Scope of study.

552 PHYSICAL THERAPY

Downloaded from https://academic.oup.com/ptj/article-abstract/60/5/551/2727029

by guest
on 06 March 2018
 Approach for Labeling and Describing
Assisted Gait Patterns

Specific terms are needed for describing assisted

gait using ambulatory aids. The term point refers to
the number of floor contacts on a line perpendicular
to the direction of walking that simultaneously occurs
during any part of stance phase for the lead foot.
When the lead foot is clearly slower than the assistive
device in making floor contact, the term delayed
should be used. Laterality describes the associated
placement on the walking surface of the side of the
upper extremity holding the assistive device and the
side of placement for the lead foot in the cycle. For
example, laterality in the case of "ipsilateral left"
indicates that both the assistive device held on the left
and the left foot are concurrently in contact with the
walking surface during a portion of the cycle. Another
example is "contralateral, left hand-right foot" when
the assistive device held on the left side and the right
foot are concurrently in contact with the walking
surface during a portion of the cycle. For complete-
ness, the specific type of assistive device is identified.
The recommended sequence for presenting descrip-
tors for types of assisted gait is 1) delayed (when
appropriate), 2) number of points, 3) laterality (when
appropriate), and 4) type of assistive device. Common

Fig. 2. Diagrammatic view of assisted gaits.

description of the system can be found in a previous

publication.15
Twenty-five physical therapy students (12 men, 13
women) served as subjects for this study. The mean
age was 22 years, mean height was 168 cm (5 ft, 6 in),
and mean weight was 64.6 kg (142 lbs). After appli-
cation of the equipment, the subjects were permitted
to walk until they felt comfortable in the laboratory
environment. During the first of two sessions, mea-
surements of gait were randomly obtained for two
walking sequences for each of the four categories of
unassisted gait: a self-selected velocity, moderate ve-
locity (71-110 cm/sec), slow velocity (31-70 cm/sec),
and very slow velocity (30 cm/sec or less). At the end
of the first session each subject was provided with
properly adjusted ambulatory aids and received in-
struction to proficiency in nine different types of
assisted gaits (Fig. 1). During the second testing ses-
sion on another day, the foot switches and accelero-
metry apparatus were applied to the subject, and
measurements of gait for two walking sequences of
each assisted gait pattern were obtained. Before the
test session, subjects were again permitted to walk
with the measurement equipment attached and be-
come acclimated to the laboratory environment. Fig. 3. Diagrammatic view of assisted gaits.

Volume 60 / Number 5, May 1980 553

Downloaded from https://academic.oup.com/ptj/article-abstract/60/5/551/2727029

by guest
on 06 March 2018

Fig. 4. Diagrammatic view of assisted gaits.
forms of assisted gait are illustrated in Figures 2 to 5.
The figures show three of the descriptors (assistive
device not included) and the numerical order of
walking surface contact. The type of assistive device
might be a crutch or a cane and, in the example of
five-point gait (Fig. 5), a walker.
Fig. 5. Diagrammatic view of assisted gaits.
554
Downloaded from https://academic.oup.com/ptj/article-abstract/60/5/551/2727029
by guest
on 06 March 2018
A final term designated as a count identifies the
number of separate floor-contact events that occurs
in one walking cycle. Unassisted gait has two counts.
Several assisted-gait patterns also have two counts:
two-point, three-point, five-point, swing-to, and
swing-through. Three counts are involved in the de-
layed two-point, three-point, and five-point gait pat-
terns. The four-point contralateral assisted gait with
two devices requires four counts.
To further illustrate the use of this system, the two-
point contralateral (left hand-right foot) assisted-gait
pattern shown in Figure 2 will be described. As
depicted in the diagram, only one assistive device is
being used, in this example, a crutch. The first event
in the gait cycle is concurrent forward movement and
floor placement of the crutch and the right foot, at
which time the cycle is complete. The proper labeling
of this type of assisted gait is "two-point contralateral
crutch gait, left hand-right foot," a description that
should be one of the first pieces of information in-
cluded in an evaluation report of a gait abnormality.
Fig. 6. Walking velocity.
RESULTS
Walking Velocity
Figure 6 shows, for walking velocity, the means
and (95%) confidence intervals for the four unassisted
gaits and nine assisted gaits studied. When using
ambulatory aids, the subjects walked considerably
slower than their customary self-selected velocity.
Assisted gaits with two counts (two-point contralat-
eral cane-right hand/left foot, two-point contralat-
eral-two canes, two-point contralateral crutch-right
hand/left foot, two-point ipsilateral cane-left, and
PHYSICAL THERAPY
 Fig. 7. Cycle time.
Fig. 8. Swing ratio L/R.

three-point crutch-left foot) were similar to the un-
assisted moderate velocity. The delayed three-point
crutch gait and delayed two-point contralateral cane
are similar to unassisted slow velocity; delayed five-
point walker gait is similar to unassisted very slow
velocity. The velocity for the four-point crutch is
located between slow and very slow velocities.
Two clinical implications may be derived from
these results. First, either the introduction of an assis-
tive device to a patient or changes in the types of
assisted-gait patterns will tend to alter the walking
velocity. Secondly, types of assisted gaits with three
and four counts may contain the upper limit for
walking velocity. Patients with severe locomotor dis-
orders may require a three- or four-count gait to
permit walking; nevertheless, there appears to be a
rather low upper limit for walking velocity.
Cycle Time
The cycle time, or time elapsed during one stride,
is related to velocity and cadence (Fig. 7). The cycle

Fig. 9. Stance ratio L/R.

Volume 60 / Number 5, May 1980 555

Downloaded from https://academic.oup.com/ptj/article-abstract/60/5/551/2727029

by guest
on 06 March 2018

Fig. 10. Step-distance ratio RL/LR.
time for this study ranged from 1.13 sec for self-
selected speed to slightly greater than 3 sec for the
four-point crutch gait and the very slow unassisted
gait, a three-fold difference. The results demonstrate
that clinical isolation of movement abnormalities by
observational gait analysis requires rapid scrutiny of
body segments, particularly when walking is per-
formed at a reasonable rate.
Swing/Time Ratio
The swing/time ratio was obtained by dividing the
right swing time into the left swing time so that
equality of contralateral swing times for the left and
right sides would yield a ratio of 1.00 (Fig. 8). Results
Fig. 11. Step-time ratio LR/RL.
556
Downloaded from https://academic.oup.com/ptj/article-abstract/60/5/551/2727029
by guest
on 06 March 2018
for the swing ratio illustrate remarkable symmetry of
left and right swing times for all 13 gait patterns.
Stance/Time Ratio
Left and right stance times were symmetrical for
all 13 gait patterns (Fig. 9). These results indicate that
left and right stance and swing times may be equal
despite the type of assisted gait. In patients, other
factors may preclude a symmetrical gait performance.
Step Length
Step length was not found to be symmetrical for all
types of walking (Fig. 10). Variability among subjects
and asymmetry of step length as reflected by the step-
length ratios tends to be greatest for the three-count
assisted gaits (delayed two-point contralateral cane,
delayed three-point crutch, and delayed five-point
walker). The length of the lead step was largest.
Therefore, a clinical objective of symmetrical step
lengths for walking may be realistic, but will probably
be most difficult to accomplish for the types of as-
sisted gait that require three counts per walking cycle.
Fig. 12. Double-stance ratio RL/LR.
Step Time
Like step length, step times for all patterns were
symmetrical except for the three-count gaits (Fig. 11),
in which the step times associated with the lead lower
limbs were faster than the trail limb by approximately
60 to 100 percent. Pursuit of symmetrical step times
for the three-count assisted gaits is probably unreal-
istic, but can most closely be approximated by swing-
PHYSICAL THERAPY
 Fig. 13. Symmetrical-stance and swing; asymmetrical-step and double stance.

ing the assistive devices forward in a more rapid
fashion. Step times for the remaining types of gaits
were symmetrical and variability was small.
Double-Stance Time
The double-stance times tend to be symmetrical for
all types of gait studied except the three-count assisted
gaits, in which the right-left double stance was dra-
matically greater than the counterpart double-stance
times associated with the left-right foot placement
(Fig. 12). The results for three-count assisted gait
patterns showed that stance and swing times were
essentially symmetrical but that step times and double
stance times were asymmetrical for the same types of
assisted gait. This problem can be explained by the
occurrence of a phase shift in the timing of the foot
placement for either the right or left lower limb (Fig.
13).
Vertical Acceleration
The vertical acceleration near the center of gravity
during walking tended to be similar for the assisted

Fig. 14. Vertical acceleration—peak to peak.

Volume 60 / Number 5, May 1980 557

Downloaded from https://academic.oup.com/ptj/article-abstract/60/5/551/2727029

by guest
on 06 March 2018
 gait categorized according to number of counts per
walking cycle (Fig. 14). As we would expect, the
magnitudes of acceleration for unassisted walking
were directly related to the walking velocity. That is,
the faster the subjects walked, the greater was the
vertical acceleration of the body.
The vertical accelerations for the assisted gaits are
disproportionately increased, however, when the dif-
ferences in walking velocity are considered. For ex-
ample, based on the walking velocity, we expected
the accelerations for the two-count assisted gaits to
be equal to or below that for the moderate, unassisted
velocity, but accelerations exceed this level. Similarly,
we expected the acceleration for three- and four-count
assisted gait to be in the vicinity of the very slow
velocity or less than the values for the slow velocity.
However, the acceleration values for the three- and
four-count assisted gaits are well above the level for
the slow velocity.
Newton's second law (F = ma) states that acceler-
ation is proportional to the force causing it. Applica-
tion of this law to vertical acceleration results indicate
that use of assistive devices tends to increase the
vertical loading on the structures of the body as a
REFERENCES
1. Bard G, Ralston HJ: Measurement of energy expenditure
during ambulation with special reference to evaluation of
assistive devices. Arch Phys Med Rehabil 40:414-420,
1959
2. Childs TF: An analysis of the swing-through crutch gait. Phys
Ther 44:804-807, 1964
3. Farmer LW: Mobility devices. Bulletin Prosthet Res 14:47-
118, 1978
4. Ganguli S, Bose KS, Dutta SR, et al: Biomechanical approach
to the functional assessment of the use of crutches for
ambulation. Ergonomics 17:365-374, 1974
5. Kauffman IB, Ridenour B: Influence of an infant walker on
onset and quality of walking pattern of locomotion: An elec-
tromyographic investigation. Percept Mot Skills 45:1323-
1329, 1977
6. Klenerman L, Hutton WC: A quantitative investigation of the
forces applied to walking sticks and crutches. Rheumatology
and Physical Medicine 12(3):152-158, 1973
7. Kljajic M, Krajnik J, Stopar M, et al: Equipment for measuring
the axial force in crutch. Annual Progress Report no. 1.
Ljubljana, Yugoslavia, Rehabilitation Engineering Center,
1978
558
Downloaded from https://academic.oup.com/ptj/article-abstract/60/5/551/2727029
by guest
on 06 March 2018
whole even though use of an assistive device can
reduce forces at one lower extremity. The implication
is that we should be cognizant of potential long-term
effects of loading excesses not only for the joint or
extremity in question, but for other body parts as
well. When assistive devices are used, the movement
of all body parts should be monitored.
SUMMARY
This study involving normal subjects may be sum-
marized as follows: 1) a systematic approach for
describing assisted gait patterns was presented, 2)
when ambulatory aids were used, the subjects walked
slower than their unencumbered self-selected veloc-
ity, 3) measurements of gait tended to be similar for
assisted gaits that require the same number of counts
per cycle, 4) reciprocal swing times and stance times
were symmetrical for all types of gait studied, 5)
double-stance times and step times were asymmetrical
for three types of assisted gait (delayed two-point,
three-point, and five-point), and 6) vertical accelera-
tions were disproportionately elevated for most as-
sisted gaits.
8. Seireg AH, Murray MP, Scholz RC: Method of recording the
time, magnitude and orientation of forces applied to walking
sticks. Am J Phys Med 47(6):307-314, 1968
9. Elson RA, Charnley J: The direction of the resultant force in
total prosthetic replacement of the hip joint. Medical and
Biological Engineering 6:19-27, 1968
10. Ely DD, Smidt GL: Effect of cane on variables of gait for
patients with hip disease. Phys Ther 57:507-512, 1977
11. Murray MP, Seirig AH, Scholz RC: A survey of the time,
magnitude, and orientation of forces applied to walking sticks
by disabled men. Am J Phys Med 48:1-13, 1969
12. Smidt GL, Wadsworth JB: Floor reaction forces during gait:
Comparison of patients with hip disease and normal subjects.
Phys Ther 53:1056-1062, 1973
13. Stauffer RN, Smidt GL, Wadsworth JB: Clinical and biome-
chanical analysis of gait following Charnley total hip replace-
ment. Clin Orthop 99:70-74, 1974
14. Hoberman M: Crutch and cane exercises and use. In Licht S
(ed): Therapeutic Exercise. New Haven, CT, Elizabeth Licht,
Publisher, 1958
15. Smidt GL, Deusinger RH, Arora J, et al: An automated
accelerometry system for gait analysis. J Biomech 10:367-
375, 1977
PHYSICAL THERAPY