Functional Reach
Functional Reach
Functional Reach
Type of test:
Time to administer: This test takes a few minutes and is very reliable
Clinical Comments: There are some recent discussions whether this test examines limits of stability. This
test may predict falling in some community dwelling populations better than patient populations.
Purpose/population for which tool was developed: Developed as a clinically feasible measure of the margin of
stability (in balance assessment) in adults. The forward reach was chosen as the test task because it is a common
functional movement and because it is similar to the leaning movements used to measure the excursion of the center
1
of pressure on a force platform (an accepted dynamic balance measure). A recent article challenges that FR and
2
limits of stability should not be used interchangeably.
When appropriate to use: 1) to document change over time in patients with balance problems, 2) to assess
likelihood that patient will fall , 3) to complete a balance assessment.
Scaling: Results in the literature have been reported in inches and centimeters. The functional reach score equals
the difference (in inches or centimeters) between the ‘end’ and the ‘start’ hand positions. (2.54 cm = 1 inch)
Equipment needed:
Yardstick and/or large paper, tape. Mackenzie (1999) suggests a modified form of the measuring device
using a self-recording tape measure connected to a handle. 3
Directions: Subject must be able to stand 1 minute without support in order to have this test administered
Set-up/Instructions:
Tape a level yardstick to wall at patient’s acromion height. Patient stands perpendicular to yardstick, with arm flexed
to 90 degrees and hand in a fist. Record position of 3 rd metacarpal head on the yardstick. Instruct pt. to reach as far
forward as possible without losing his/her balance,, lifting his heels, or taking a step. Record position of 3 rd
metacarpal head on the yardstick. [Note: pt. needs to keep hand at level of yardstick when reaching forward but
cannot be allowed to touch the wall. Beyond these restrictions, DO NOT control the method of reach]. A large
piece of paper could be taped to the wall for marking the start & end positions. Allow 2 practice trials then average
the next 3 trials to obtain the score for the session. A paper by Billek-Sawhney (2005) found the reliability between
2 trials to be r=.975 meaning one can use 2 trials. 4
Arnadottir and Mercer (2000) 5 found 35 women age 65 to 93 performed better on FR when they were barefoot or
wore walking shoes than when subjects wore dress shoes regardless of whether they performed the test on carpet or
linoleum. There was no difference between barefoot walking shoe conditions on either floor surface.
The foot placement is the typical stance of the client. No studies were found that compared foot placement for FR.
An article by Mcllroy and Maki (1996) 6 suggests the wide range of preferred foot placements highlights the need
for standardization during foot placement. Functional reach increases significantly with additional sensory
information from the fifth metacarpal surface of the dominant hand 7 and if a target is given. 8
Contraindications: Blurred vision has less of an effect on FR than the Tinetti or TUG. 9
Learning Effect: Clients who have a target reach further than those who do not. 8
Reliability:
Reference N= Sample description Reliability statistic
Intrarater reliability: same rater within one session (or one day)
Mecagni, 200010 8 2 trials ICC =.96
Franchignoni, 199811 45 healthy women 55-71 ICC(2,1) = .97
Rockwood, 2000 12 1161 3 trials: persons with cognitive impairment ICC=.92
Interrater Reliability
Duncan, 1990 1 17 normal subjects, age 20-87 ICC = .98
Franchignoni, 1998 11 45 healthy women 55-71 ICC(2,1) = .86
Light, 1995 13 30 5 trials each for 2 subjects, in community- r = .98
dwelling elderly
Wolf, 1999 14 56 For 4 raters observing the same test, ICC = .99
Kileff, 2005 15 8 (2 raters) people with MS Friedman Test Mean difference;
.5 on FR left arm and .25 FR
right arm
Giorgetti, 1998 16 21 Mean age = 73, without disability ICC = .73
21 (2 examiners) Mean age = 75, with ICC = .79
disability
Holbein-Jenny, 2005 17 26 Community-dwelling ICC (1,1)
Forward = .98; Backward = .96
Right = .94; Left = .91
Schenkman, 1997 18 15 patients with early to middle stages of PD. ICC = .90
Frzovic, 2000 19 28 (N=14) people with MS ; (N=14) Control ICC=.89
Test-retest reliability
Time Btw. Mean SD Test-retest
Reference Population MDC (cm)
Testing (cm) (cm) Reliability
1
Duncan, 1990 Community- 1 week Forward ICC= Unable to calculate-
dwelling elderly 0.92 no X or SD given
(n=128)
Franchignoni, (n=45) Females, 24 hrs. Forward ICC= Unable to
1998 11 ages 55-75 0.87 calculate—no X or
SD given
Hageman, 1995 20 Community- 1 week Forward ICC= Unable to
dwelling healthy 0.92 calculate—no X or
adults (n=12) SD given
Holbein-Jenny, Elderly (n=21), 1-2 weeks 14.22, 6.54, Forward ICC= Forward= 10.54
2005 17 ages 74-92 7.37, 5.59, 0.75 Backward= 8.33
8.38, 6.35, Backward ICC= Right= 10.26
9.40 7.87 0.71 Left= 8.99
Right ICC= 0.66
Left ICC= 0.83
Lim, 2005 21 Idiopathic 1 week Forward ICC= SDD= 11.5
Parkinson’s 0.74
Disease (n=26)
Marsh, 2005 22 Community- 2 weeks Lateral ICC= unable to calculate-
dwelling elderly 0.86 X and SD not given
(n=44) for subset
Schenkman, 1998 Parkinson’s 1 day 32.3 Forward ICC= Unable to calculate-
23
Disease, (n=14) 0.84 no SD given for
74.5 yrs (mean initial measurement
age)
Sherrington, 2005 Hopital 1 day 14 9.6 Forward ICC= 8.83
24
inpatients and 0.89
community
dwelling elderly,
fallers and
previous fallers
(n=30)
Fallers 217 Duncan, 1992 39 found FR to have predictive validity in identifying recurrent
fallers (i.e., 2 or more falls during the 6-month follow up period); n= 217
community-dwelling male veterans (age 70-104). Logistic regression shows that:
If FR = 0 inches: 8 times more likely to have 2 falls in 6 mos than
person with FR=10”
If FR < or equal to 6 inches: 4 times more likely to have 2 falls
If FR > 6 inches but < 10 inches: 2 times more likely to have 2 falls
16 Cho & Kamen (1998) 40 found no group differences on FR for 8 healthy older
subjects compared to 8 age-matched idiopathic fallers.
705 Having a long functional reach ( 35 cm) and being able to perform a full tandem
stand with eyes closed for at least 10 seconds were associated with decreased rates
of falls. 41
67 Any improvement in FR during PT Rx in a geriatric day hospital can predict
subsequent decrease in falling 42
Older adults 436 FR did not predict disability in a large cohort study of women. 43
705 FR was positively associated with quadriceps and grip strength; 44 as well as BMI
in studies of 705 elderly Japanese women in Hawaii. 41
Community- 402 FR was not associated with falls 45 which averaged 24 cm
dwelling
Fallers 15 Mean age = 73 46 No difference on FR between 2 groups
Non-Fallers 10 Mean age = 75 46
Community- 99 Duncan (1990 ) found that only 3/99 male veterans who could ascend/descend
dwelling elderly stairs foot over foot had FR of 6 inches or less. 1
45 No subject with FR less than 7 inches: was able to complete more than 6/11 items
on the mobility skills protocol; could balance for greater than 1 second during
SLS; was able to tandem walk; or was able to leave his/her neighborhood without
help. 26
Women community- 99 Mean age = 71 47 No significant differences
dwelling Non-fallers (N=65): FFR= 30(1) Right FR = between groups
20(1)
Fallers (N=35): FFR= 29(1) Right FR= 20(1)
Frequent Fallers (N=16): FFR= 29(2) Right FR=
19(1)
Recurrent Fallers (N=19): FFR= 29(2) Right FR=
20(1)
Community- 15 Steady patients No statistical differences
dwelling 23 Unsteady patients 48 between 2 groups
Sensitivity/specificity:
Population N= Cutoff Score and Description Results
Fallers 54 Cutoff of 25 cm: (identifying multiple fallers vs Sensitivity of 63%
nonmultiple fallers Specificity of 59%
(N=54; outpatients over the age of 65 attending
community rehab) 49
Dx/o Parkinson’s 58 Cutoff of 25.4 cm: (identifying fallers) Sensitivity of 30%
50
Disease Specificity of 92%.
Day Hospital 30 Using cut off of 18.5 to predict fall; Mean score Sensitivity of 75%
fallers (N=18) 15.5(6.5); non-fallers (N=12) Specificity of 67%
19.4(4.2); Mean age = 80-81 51 OR 5.28, p < .08
Community 203 Using a cutoff of 30 for able vs. not able 52 Sensitivity 86%
dwelling elderly Specificity 38%
Using a cutoff of 24 for decreased disability vs. Sensitivity 81%
disabled 52 Specificity 52%
NOTE: Clinicians need to choose a cut-off score based on the specific purpose for which the test is used
Responsiveness/sensitivity to change
Population N Responsive
Reference and Intervention Data Supporting Responsiveness
Descriptor = Yes/No
53
Community- 42 Okumiya, 1996 Yes Exercisers improved significantly
dwelling Healthy Japanese elderly; mean age greater than controls
elderly = 79
Experimental group:
Exercisers
Control group:
Non-exercising
Length / frequency of intervention
6 months; 54
1 hour, 2x/week
Community- 12 Rogers, 2001 Yes Significant Improvement from 33
dwelling Balance intervention program; mean cm initial to 40 cm
elderly age=70
Length / frequency of intervention
10 weeks
20 Barrett, 2002 55 Yes Progressive
Healthy elderly persons Initial: 34 (5) cm
Progressive resistive exercise Final: 38 (3) cm; p < .003
program Flexibility Initial 33(5) to 33(6); NS
Flexibility training; 2x per wk; 10 Significant change between groups
weeks
Community- 14 Shigematsu, 2001 56 Yes Initial: 23 (5) to 27(3); p<.05
dwelling Exercise program Control 26(8) to 25(7); NS
elderly Length / frequency of intervention
(Continued) 60 min, 3x/week for 3 months
19 Dennis, 1999 57 Yes Initial: 7 (3) inches
Health ambulatory women over 65 Final: 8(2) inches; p <.025
Intervention: Alexander Technique Control: FR decreased by .74
Instruction inches; p<.005
Length / frequency of intervention
1 hr, 2x/week, 4 weeks
134 Morey, 1999 58 No Both with baseline measure of 13”;
Group 1: spinal flexibility plus No significant gains in either group
aerobic exercise
Group 2: aerobic only exercise
Community- 52 Simmons, 1996 59 Yes Significant improvement in water
dwelling Subjects mean age=80, with a fear exercisers (p<.001), land exercisers
elderly of falling (p<.03)
(Continued) 4 groups: water exercisers, land No change in other 2 groups
exercisers, water sitters, land sitters
94 Hakim, 2004 60 Yes Group 1: Better FR (p<.01)
Healthy older adults Group 2: Better at Forward (p<.01),
Control group: no exercise Backward (p<.001) and Left FR
Group 1: structured exercise (p<.001)
Group 2: Tai Chi intervention
Reference data: All studies that reported inches were converted to centimeters (cm)
Resource N= Subjects FR Scores
Weiner, 1992 26 45 Community-dwelling elderly; male & mean (SD) = 27.68cm (7.87)
female; mean age = 78 (8.4)
Newton, 1997 251 Seniors; average age 74 mean (SD)= 22.60 (8.38) cm
89
Schenkman, 251 Community dwelling adults (mean age subjects with PD:
2000 91 = 71) This study assessed spinal mean (SD) = 31.50cm (7.62)
flexibility and balance.
n=56 with PD subjects without PD:
n=195 without PD mean(SD) =34.29cm (5.84)
Aoyagi, 2000 92 447 Community-dwelling Japanese Exercisers:
persons (mean age =66). This was a Women = 29.6 (.5) cm
study of bone mineral density (BMD). Men= 29.5 (1.0) cm
Non-exercisers:
Women = 28.6 (.4) cm
Men= 29.2 (.8) cm
Purser, 1999 93 185 Older women with osteoporosis and Women with osteoporosis and vertebral fractures
vertebral fractures. 28.96cm (5.84)
Lehmann, 2006 50 Persons with late effects of polio, Men: 21.3(9.5)cm (N=21)
94
mean age = 60 Women: 25.2(8.9)cm (N=29)
All: 23.5(9.3)cm (N=50)
Frzovic, 2000 14 Subjects with MS and 14 controls in AM: 39.19 (5.88) cm
19
in PM: 39.92 (6.66) cm
Davis, 1999 41 705 Japanese women in Hawaii (mean age Mean (SD) =30.9 (6.1) cm
= 74)
Stack, 2005 95 51 (N=33) Grade III, PD 18cm (13-23)
(N=18) Grade IV, PD 15cm (7-21)
Marsh, 2005 22 140 Community dwelling; Mean age = 75 30.5(6.6)cm
Cim biz, 2005 30 Diabetic neuropathic; Mean age = 58 34(13)cm
96
30 Control; Mean age = 67 44(14)cm
Chow, 2004 32 16 Females with osteoporosis/Osteopenia; 30(9)cm
Mean age=67
Hageman, 1995 24 (N=12) Younger adults; Mean age = 43(4)cm
20
25 37(6)cm
(N=12) Older adults; Mean age = 65
Smith, 2004 97 75 Stroke 23(9)cm
Stankovic, 30 PD Mean age = 68 Without Falls: 30(6)cm
2004 98 Mean age = 72 With Falls: 21(6)cm
20 Control Mean age = 70 32(6)cm
Teri, 1998 99 30 Alzheimer’s 25(15)cm
Wolf, 2003 100 145 Tai Chi participants; Mean age = 81 30(8)cm
141 Wellness Class; Mean age = 81 27(8)cm
Goldberg, 2005 8 Young; Mean age = 24 34.80cm (2.29)
101
7 Balance unimpaired; Mean age = 74 26.16cm (1.52)
8 Balance impaired; Mean age = 80 26.92cm (2.03)
Huang, 1996 569 Post-menopausal Japanese American Predicted performance on FR; average FR was
102
women; ages 55-93; s/p vertebral 33.1 (6.1) cm
compression fracture
NOTE: FR declines with age in both genders. 1, 20, 92
Interpreting results:
It measures a subject’s forward limit of stability, which is considered one part of postural control (or balance)
assessment. Duncan (1990) 1 concludes that FR is a good clinical measure of the margin of stability and is
“conceptually related” to the excursion of the center of pressure. Others are suggesting that FR is a weak measure
of stability limits (low correlation with FR and displacement of center of pressure, .38). Movement of the trunk
seems to influence the test more than displacement of center of pressure. 103
When the Functional Reach 104 test and platform measures of postural sway were used with clients with hemiparesis,
they appeared to be evaluating comparable standing-balance abilities. In a kinematic study of 34 young subjects
(20-36) and 33 older subjects (60-76 years), spinal motion during forward FR was characterized by forward and
lateral trunk flexion, thoracolumbar and lower body rotation. Young subjects displaced their center of pressure
further forward (45.2 cm) and through a greater percentage of their initial base of support than older subjects (37.1
cm). The younger group had more forward trunk flexion and thoracolumbar rotation. 105 O’Brien, et al (1997) found
a weak correlation between inclination of the upper thoracic spine and functional reach. 106 Wernick-Robinson
(1999) found FR does not measure dynamic balance because people with vestibular hypofunction did as well. 107
Daubney and Culham (1999) 108 found that ankle plantar-flexion force accounted for 13% of the score on the FR.
Correlations were found between FR and hip extensor strength (.45) and hip flexor strength (.47). 30 Eight hundred
thirty three community dwelling elderly 64-79 years old (457 were Mexican American) participated in a home
assessment. For each degree increase in shoulder ROM, the likelihood of having a short reach was reduced by 3%
and for each degree increase in elbow ROM, the likelihood of having a short reach was reduced by 2%. 109
Other:
Reach in Four Directions:
In 1997, the first “reach in four directions” (RFDT) results were published. This study included a large minority
population. (N=204-250) The mean forward reach was 8.9 inches, right 6.8 inches and left 6.6 inches. 89 A small
study (N=7) found a .43 -.65 correlation between BFR and ankle dorsiflexion (df). Improvements in df improved
BFR. 111
Lateral Reach measurements were published in 1999. Validity of lateral reach results showed a significant
correlation with COPE (r=0.33) measurements and laboratory measure of reach (r=.65). Test-retest reliability
(r=0.94) was also found. 112 Lateral reaches to the right and left were not significantly different between the sides.
For their analysis, right side measurements were used. Age was negatively correlated with Lateral Reach results.
Results were not separated into cohorts. This study included 60 females (mean age=72.5). Lateral Reach in 22
community dwelling females (average age 81) was 14.3 (4.5) cm left and 14.9 (4.6) cm right. 113 Lateral reach in
383 Japanese (mean age = 79) 19(12)cm 31 Lateral reach in sitting of 18 elderly persons showed a -.63 correlation
between rising time and lateral reach in sitting. 114
Results of reach in four directions of 87community dwelling adults is reported in Table 5-1.
Forward, right and left functional reach on 53 seniors who sat in a wheelchair were compared for people sitting on a
cushion vs a sling. 117 Forward and lateral reach for 31 healthy and 31 subjects with hemiparesis in a sitting position
are published by Hsu (2005). 118
Table 5-1
Means (X), Standard Deviations (SD) and 95% Confidence Intervals (CI) of the Multi-Directional Reach Test by Age and Gender
Cohorts (in centimeters).
Forward Backward Left Right
(cm) (cm) (cm)
(cm)
Age
(yrs)
Gender N X SD CI X SD CI X SD CI X SD CI
TOTAL
83 29 7 28-29 19 8 17-20 17 5 16-18 17 5 16-18
SAMPLE
Steffen, TM, Mollinger, LA (2005). Age-and gender-related test performance in community-dwelling adults:
multi-directional reach test, berg balance scale, sharpened Romberg tests, activities-specific balance confidence
scale, and physical performance test. Journal of Neurological Physical Therapy 29(4): 181-188.