- Email: [email protected]
Hand-grip dynamometry provides a valid indication of upper extremity strength impairment in home care patients
Hand-grip Dynamometry Provides a Valid Indication of Upper Extremity Strength Impairment in Home Care Patients Richard W. Bohannon, EdD, PT, NCS Department of Physical Therapy School of Allied Health University of Connecticut Storrs, Connecticut
h erap ists working with patients in a home
T care setting require measurements that are capable of providing a valid indication of the patients' impairments. Impairment is defined by the World Health Organization as any loss or abnormality of psychological, physiological, or anatomical structure or function.' Impairments in upper extremity strength can be quantified by several means, the primary being manual muscle testing and dynamometry. Manual muscle testing has the advantage of simplicity. It requires no equipment and can be performed quickly. Hand-grip dynamometry incorporates the benefits of objectivity and sensitivity. Although the validity of hand-grip dynamometry is well established.i" a review of the literature did not reveal any descriptions of the construct validity of hand-grip dynamometry as a measure of upper extremity strength in home care patients. The purpose of this brief report, therefore, was to describe the discriminant and convergent construct validity of hand-grip dynamometry measurements obtained from home care patients referred for rehabilitation.
METHODS This was a retrospective study based on data retrieved from initial physical therapy notes of 37 home care patients (15 men, 22 women). The mean age of the patients was 77.7 years, SD = 10.9 years. Correspondence and reprint requests to Richard W. Bohannon, EdD, PT, NCS, Department of Physical Therapy, School of Allied Health U-lOl, 358 Mansfield Road, University of Connecticut, Storrs, CT 06269.
258
JOURNAL OF HAND THERAPY
ABSTRACT: This retrospective study investigated the validity of hand-grip dynamometry with respect to its use in predicting generalized upper extremity strength. The records of 37 patients (mean average age, 77.7 years) receiving home care were used. Discriminant construct validity was examined by comparing their dynamometry measurements with measurements of age- and gender-matched healthy individuals reported in the literature. Convergent construct validity was described by the correlations of their dynamometry measurements and manual muscle test scores of the upper extremities. The patients' dynamometer-measured grip forces were significantly less than reported normative values. The patients' dynamometer measured grip forces were correlated significantly with their manual muscle test scores. These findings support the construct validity of hand-grip dynamometry for characterizing upper extremity strength impairment among adults treated in a home care setting. J HAND THER 11:258-260, 1998.
All the patients whose notes were examined were treated by the author over a two-year period. Every patient had some degree of physical disability, but none of the patients was referred for an isolated problem of either upper extremity. Fourteen patients had a neurologic diagnosis (e.g., stroke), six had orthopedic problems (e.g., lower extremity fracture), six had cardiopulmonary disorders (e.g., pneumonia), seven were postsurgical (e.g., bowel resection), and four had general debility. A calibrated Jamar hand-grip dynamometer was used to measure grip strength bilaterally. Testing was performed in accordance with the recommendations of the American Society of Hand Therapists/ but in cases where a patient's strength was insufficient to hold the dynamometer in the recommended test position, it was supported by the therapist. The 0-5 grading scheme was used bilaterally to quantify the manually tested muscle strength of the hand-grip, elbow flexion, and shoulder abduction actions. For the purpose of statistical analysis the manual muscle test grades were converted as follows: 0 = 0, 1 = I, 1+ = 2, 2- = 3, 2 = 4,2+ = 5, 3- = 6, 3 = 7, 3+ = 8,4- = 9, 4 = 10,4+ = II, 5 = 12. The method of muscle testing was essentially that described by Hislop and Montgomery." An exception was a measure of hand-grip strength, for which there is no description provided by Hislop and Montgomery. That test merely involved having the patient make a fist. If able to do so (unconverted manual muscle test grade 3), the patient was asked to maximally grasp the author's second, third, and fourth fingers. If the subject was able to perform such a grasp, he or she was then told to squeeze as hard as possible while the tester
TABLE 1. Descriptive and Inferential Statistics Comparing Actual and Predicted Hand-grip Forces (in Newtons) for 37 Home Care Patients Actual Grip Forces Mean
SD
Left side
158
75
Right side
177
70
Predicted Grip Forces Range
Mean
SD
Range
t
P
0-325
216
66
167-484
- 3.748
0.001
22-325
249
72
189-488
-6.646
0.0001
attempted to withdraw his fingers. A grade was given, based on the subject's forcefulness of grip (e.g., tester hand not withdrawn with moderate effort = unconverted manual muscle test score 4; tester hand not withdrawn with maximal effort = unconverted manual muscle test score 5). Discriminant construct validity was tested by comparing dynamometer measurements obtained from the patients with gender- and age-groupmatched reference values published by Mathiowetz et al." Separate t-tests were used for comparisons on the left and right sides. Convergent construct validity was established by computing the correlations of hand-grip dynamometer forces with the manual muscle test scores of each upper extremity. Spearman correlations (r 5 ) were used to accommodate the ordinal manual muscle test grades. As scatterplots showed the relationships to be somewhat curvilinear, Pearson correlations (r) and multiple correlations (R) were also performed. The Systat statistical program was used to conduct all statistical analyses."
was demonstrated by the difference between the dynamometer-measured forces of patients and healthy subjects, the former being significantly lower. The convergent construct validity was demonstrated by the significant correlations of handgrip force with upper extremity manual muscle test scores. This validity does not justify using dynamometrically measured grip force as a surrogate for other upper extremity strength measures. If a therapist needs to know the strength of a specific action-for example, the elbow extensors-that action should be tested. On the other hand, if a sensitive quantitative measure is sought to characterize the upper extremity strength of a home care pa-
TABLE 2. Descriptive Statistics for Manual Muscle Test Measurements of Upper Extremity Strength Measurement Left side
RESULTS Forces measured with the hand-grip dynamometer as well as average forces predicted from data presented by Mathiowetz et al.' are listed in Table 1. The patients' measured forces were significantly lower than their predicted forces on both the left and right sides (p ::5 0.001). Table 2 provides descriptive statistics for manual muscle test measurements of upper extremity strength. Table 3 lists correlations of hand-grip force measured by dynamometry with manual muscle test grades. All correlations tr; 0.421-0.848; r, 0.537-0.799; R, 0.589-0.934) were significant, with the highest correlations being between dynamometer and manual measures of grip strength. The curvilinear nature of these relationships is illustrated in Figure 1. The next highest correlations were between the dynamometer measures of grip force and the sum of upper extremity manual scores. These relationships were also curvilinear (Figure 2).
DISCUSSION The results of this study demonstrate that in a home care setting measurements of grip strength obtained with a hand-grip dynamometer possess construct validity. Discriminant construct validity
Hand grip (HG) Elbow flexion (EF) Shoulder abduction (SA) HG + EF + SA
Right side Hand grip Elbow flexion Shoulder abduction HG + EF + SA
Median Minimum Maximum 11.0
1.0
12.0
11.0
2.0
12.0
9.0 31.0
2.0 5.0
12.0 36.0
11.0 11.0
7.0 7.0
12.0 12.0
9.0 31.0
2.0 19.0
12.0 36.0
TABLE 3. Correlations of Hand-grip Dynamometer Forces with Manual Muscle Test Measurements of Upper Extremity Strength
Measurement Left side
Right side
Hand grip (HG) Elbow flexion (EF) Shoulder abduction (SA) HG + EF + SA Hand grip Elbow flexion Shoulder abduction HG + EF + SA
Spearman Correlation (rJ
Pearson Multiple Correia- Correiation tion (r) (R)
0.818
0.799
0.934
0.421*
0.537
0.654
0.672 0.752
0.724 0.773
0.729 0.852
0.848 0.450*
0.778 0.577
0.840 0.694
0.543 0.701
0.574 0.731
0.589 0.750
"Correlations significant at P < 0.01; all other values significant at P < 0.001.
October-December 1998
259
15,-----,------,----,------,
o(/)
10
0-
il:
o
...J
:§ z
~
r
w
a::
o
~ 100
200
300
I
400
5
o
o
DYNAMOMETER GRIP FORCE (N)
li!
I
401,---,---,-----r-~ --,
o(/)
li!
o
tt. j
~
w
~ 20
a::
w
00-
:::J 10 ...J
~ o'--_-----L_ _----L o 100 200
----L--~ 300
400
DYNAMOMETER GRIP FORCE (N)
200
300
400
40,..----,------,---,---,
o >-
a:: ......
100
DYNAMOMETER GRIP FORCE (N)
(/)
~ 30
. ·.
· .... . .... ·. ... .
FIGURE 2. Scatterplots illustrating the curvilinear nature of the relationship between dynamometer grip forces and the sum of manual scores of the left upper extremity (left) and the right upper extremity (right).
LI----'-----'----'---------' o
100
200
300
5.
REFERENCES
6.
1. World Health Organization. International Classification of
Impairments, Disabilities, and Handicaps. Geneva, Switzerland: WHO, 1980. 2. Sunderland A, Tinson D, Bradley, L, Langton-Hewer R. Arm
JOURNAL OF HAND THERAPY
400
DYNAMOMETER GRIP FORCE (N)
tient, hand-grip dynamometry is appropriate. The internal consistency of limb muscle strength measurements supports this judgment." Hand-grip dynamometers are not particularly expensive. They are easily transported to patients' homes, where they can be used to obtain bilateral measures of grip strength in less than a minute. These facts combined with their validity lend backing to the use of hand-grip dynamometry in a home care setting.
260
FIGURE 1. Scatterplots illustrating the curvilinear nature of the relationship between dynamometer grip forces and manual grip scores of the left hand (left) and the right hand (right).
...J
o o
o
1
J ~~-
5
~·:-1
15
3.
4.
7. 8.
function after stroke: an evaluation of grip strength as a measure of recovery and a prognostic indicator. J Neurol Neurosurg Psychiatry. 1989;52:1267-72. Pincus T, Callahan LF. Rheumatoid function tests: grip strength, walking time, button test and questionnaires document and predict long-term morbidity and mortality in rheumatoid arthritis. J Rheumatol. 1992;19:1051-7. Webb AR, Newman LA, Taylor M, Keogh JB. Hand grip dynamometry as a predictor of postoperative complications: reappraisal using age standardized grip strengths. J Parenter Enteral Nutr. 1989;13:30-3. Mathiowetz V, Kashman N, Volland G, Weber K, Dowe M, Rogers S. Grip and pinch strength: normative data for adults. Arch Phys Med Rehabil. 1985;66:69-72. Hislop HG, Montgomery J. Daniels and Worthingham's Muscle Testing. 6th Ed. Philadelphia, PA: WB Saunders, 1995: 58-166. Systat 6.0 for Windows. Chicago, IL: SPSS, 1996. Bohannon RW. Internal consistency of manual muscle testing scores. Percept Motor Skills. 1997;85:736-8.
h erap ists working with patients in a home
T care setting require measurements that are capable of providing a valid indication of the patients' impairments. Impairment is defined by the World Health Organization as any loss or abnormality of psychological, physiological, or anatomical structure or function.' Impairments in upper extremity strength can be quantified by several means, the primary being manual muscle testing and dynamometry. Manual muscle testing has the advantage of simplicity. It requires no equipment and can be performed quickly. Hand-grip dynamometry incorporates the benefits of objectivity and sensitivity. Although the validity of hand-grip dynamometry is well established.i" a review of the literature did not reveal any descriptions of the construct validity of hand-grip dynamometry as a measure of upper extremity strength in home care patients. The purpose of this brief report, therefore, was to describe the discriminant and convergent construct validity of hand-grip dynamometry measurements obtained from home care patients referred for rehabilitation.
METHODS This was a retrospective study based on data retrieved from initial physical therapy notes of 37 home care patients (15 men, 22 women). The mean age of the patients was 77.7 years, SD = 10.9 years. Correspondence and reprint requests to Richard W. Bohannon, EdD, PT, NCS, Department of Physical Therapy, School of Allied Health U-lOl, 358 Mansfield Road, University of Connecticut, Storrs, CT 06269.
258
JOURNAL OF HAND THERAPY
ABSTRACT: This retrospective study investigated the validity of hand-grip dynamometry with respect to its use in predicting generalized upper extremity strength. The records of 37 patients (mean average age, 77.7 years) receiving home care were used. Discriminant construct validity was examined by comparing their dynamometry measurements with measurements of age- and gender-matched healthy individuals reported in the literature. Convergent construct validity was described by the correlations of their dynamometry measurements and manual muscle test scores of the upper extremities. The patients' dynamometer-measured grip forces were significantly less than reported normative values. The patients' dynamometer measured grip forces were correlated significantly with their manual muscle test scores. These findings support the construct validity of hand-grip dynamometry for characterizing upper extremity strength impairment among adults treated in a home care setting. J HAND THER 11:258-260, 1998.
All the patients whose notes were examined were treated by the author over a two-year period. Every patient had some degree of physical disability, but none of the patients was referred for an isolated problem of either upper extremity. Fourteen patients had a neurologic diagnosis (e.g., stroke), six had orthopedic problems (e.g., lower extremity fracture), six had cardiopulmonary disorders (e.g., pneumonia), seven were postsurgical (e.g., bowel resection), and four had general debility. A calibrated Jamar hand-grip dynamometer was used to measure grip strength bilaterally. Testing was performed in accordance with the recommendations of the American Society of Hand Therapists/ but in cases where a patient's strength was insufficient to hold the dynamometer in the recommended test position, it was supported by the therapist. The 0-5 grading scheme was used bilaterally to quantify the manually tested muscle strength of the hand-grip, elbow flexion, and shoulder abduction actions. For the purpose of statistical analysis the manual muscle test grades were converted as follows: 0 = 0, 1 = I, 1+ = 2, 2- = 3, 2 = 4,2+ = 5, 3- = 6, 3 = 7, 3+ = 8,4- = 9, 4 = 10,4+ = II, 5 = 12. The method of muscle testing was essentially that described by Hislop and Montgomery." An exception was a measure of hand-grip strength, for which there is no description provided by Hislop and Montgomery. That test merely involved having the patient make a fist. If able to do so (unconverted manual muscle test grade 3), the patient was asked to maximally grasp the author's second, third, and fourth fingers. If the subject was able to perform such a grasp, he or she was then told to squeeze as hard as possible while the tester
TABLE 1. Descriptive and Inferential Statistics Comparing Actual and Predicted Hand-grip Forces (in Newtons) for 37 Home Care Patients Actual Grip Forces Mean
SD
Left side
158
75
Right side
177
70
Predicted Grip Forces Range
Mean
SD
Range
t
P
0-325
216
66
167-484
- 3.748
0.001
22-325
249
72
189-488
-6.646
0.0001
attempted to withdraw his fingers. A grade was given, based on the subject's forcefulness of grip (e.g., tester hand not withdrawn with moderate effort = unconverted manual muscle test score 4; tester hand not withdrawn with maximal effort = unconverted manual muscle test score 5). Discriminant construct validity was tested by comparing dynamometer measurements obtained from the patients with gender- and age-groupmatched reference values published by Mathiowetz et al." Separate t-tests were used for comparisons on the left and right sides. Convergent construct validity was established by computing the correlations of hand-grip dynamometer forces with the manual muscle test scores of each upper extremity. Spearman correlations (r 5 ) were used to accommodate the ordinal manual muscle test grades. As scatterplots showed the relationships to be somewhat curvilinear, Pearson correlations (r) and multiple correlations (R) were also performed. The Systat statistical program was used to conduct all statistical analyses."
was demonstrated by the difference between the dynamometer-measured forces of patients and healthy subjects, the former being significantly lower. The convergent construct validity was demonstrated by the significant correlations of handgrip force with upper extremity manual muscle test scores. This validity does not justify using dynamometrically measured grip force as a surrogate for other upper extremity strength measures. If a therapist needs to know the strength of a specific action-for example, the elbow extensors-that action should be tested. On the other hand, if a sensitive quantitative measure is sought to characterize the upper extremity strength of a home care pa-
TABLE 2. Descriptive Statistics for Manual Muscle Test Measurements of Upper Extremity Strength Measurement Left side
RESULTS Forces measured with the hand-grip dynamometer as well as average forces predicted from data presented by Mathiowetz et al.' are listed in Table 1. The patients' measured forces were significantly lower than their predicted forces on both the left and right sides (p ::5 0.001). Table 2 provides descriptive statistics for manual muscle test measurements of upper extremity strength. Table 3 lists correlations of hand-grip force measured by dynamometry with manual muscle test grades. All correlations tr; 0.421-0.848; r, 0.537-0.799; R, 0.589-0.934) were significant, with the highest correlations being between dynamometer and manual measures of grip strength. The curvilinear nature of these relationships is illustrated in Figure 1. The next highest correlations were between the dynamometer measures of grip force and the sum of upper extremity manual scores. These relationships were also curvilinear (Figure 2).
DISCUSSION The results of this study demonstrate that in a home care setting measurements of grip strength obtained with a hand-grip dynamometer possess construct validity. Discriminant construct validity
Hand grip (HG) Elbow flexion (EF) Shoulder abduction (SA) HG + EF + SA
Right side Hand grip Elbow flexion Shoulder abduction HG + EF + SA
Median Minimum Maximum 11.0
1.0
12.0
11.0
2.0
12.0
9.0 31.0
2.0 5.0
12.0 36.0
11.0 11.0
7.0 7.0
12.0 12.0
9.0 31.0
2.0 19.0
12.0 36.0
TABLE 3. Correlations of Hand-grip Dynamometer Forces with Manual Muscle Test Measurements of Upper Extremity Strength
Measurement Left side
Right side
Hand grip (HG) Elbow flexion (EF) Shoulder abduction (SA) HG + EF + SA Hand grip Elbow flexion Shoulder abduction HG + EF + SA
Spearman Correlation (rJ
Pearson Multiple Correia- Correiation tion (r) (R)
0.818
0.799
0.934
0.421*
0.537
0.654
0.672 0.752
0.724 0.773
0.729 0.852
0.848 0.450*
0.778 0.577
0.840 0.694
0.543 0.701
0.574 0.731
0.589 0.750
"Correlations significant at P < 0.01; all other values significant at P < 0.001.
October-December 1998
259
15,-----,------,----,------,
o(/)
10
0-
il:
o
...J
:§ z
~
r
w
a::
o
~ 100
200
300
I
400
5
o
o
DYNAMOMETER GRIP FORCE (N)
li!
I
401,---,---,-----r-~ --,
o(/)
li!
o
tt. j
~
w
~ 20
a::
w
00-
:::J 10 ...J
~ o'--_-----L_ _----L o 100 200
----L--~ 300
400
DYNAMOMETER GRIP FORCE (N)
200
300
400
40,..----,------,---,---,
o >-
a:: ......
100
DYNAMOMETER GRIP FORCE (N)
(/)
~ 30
. ·.
· .... . .... ·. ... .
FIGURE 2. Scatterplots illustrating the curvilinear nature of the relationship between dynamometer grip forces and the sum of manual scores of the left upper extremity (left) and the right upper extremity (right).
LI----'-----'----'---------' o
100
200
300
5.
REFERENCES
6.
1. World Health Organization. International Classification of
Impairments, Disabilities, and Handicaps. Geneva, Switzerland: WHO, 1980. 2. Sunderland A, Tinson D, Bradley, L, Langton-Hewer R. Arm
JOURNAL OF HAND THERAPY
400
DYNAMOMETER GRIP FORCE (N)
tient, hand-grip dynamometry is appropriate. The internal consistency of limb muscle strength measurements supports this judgment." Hand-grip dynamometers are not particularly expensive. They are easily transported to patients' homes, where they can be used to obtain bilateral measures of grip strength in less than a minute. These facts combined with their validity lend backing to the use of hand-grip dynamometry in a home care setting.
260
FIGURE 1. Scatterplots illustrating the curvilinear nature of the relationship between dynamometer grip forces and manual grip scores of the left hand (left) and the right hand (right).
...J
o o
o
1
J ~~-
5
~·:-1
15
3.
4.
7. 8.
function after stroke: an evaluation of grip strength as a measure of recovery and a prognostic indicator. J Neurol Neurosurg Psychiatry. 1989;52:1267-72. Pincus T, Callahan LF. Rheumatoid function tests: grip strength, walking time, button test and questionnaires document and predict long-term morbidity and mortality in rheumatoid arthritis. J Rheumatol. 1992;19:1051-7. Webb AR, Newman LA, Taylor M, Keogh JB. Hand grip dynamometry as a predictor of postoperative complications: reappraisal using age standardized grip strengths. J Parenter Enteral Nutr. 1989;13:30-3. Mathiowetz V, Kashman N, Volland G, Weber K, Dowe M, Rogers S. Grip and pinch strength: normative data for adults. Arch Phys Med Rehabil. 1985;66:69-72. Hislop HG, Montgomery J. Daniels and Worthingham's Muscle Testing. 6th Ed. Philadelphia, PA: WB Saunders, 1995: 58-166. Systat 6.0 for Windows. Chicago, IL: SPSS, 1996. Bohannon RW. Internal consistency of manual muscle testing scores. Percept Motor Skills. 1997;85:736-8.