Muscle impairments and behavioral factors mediate functional limitations and disability following stroke.

Stroke presents a major public health concern in the United States, with more than 700,000 new or recurrent cases occurring each year. (1) Despite a noteworthy reduction in mortality in the last century, (2) stroke remains the third leading cause of death. Moreover, morbidity in the approximately 4.8 million people who have survived a stroke is substantial, making stroke the foremost cause of serious, long-term disability in the United States. (3)

The impairments (abnormalities occurring in a specific organ or organ system (4)) resulting from stroke encompass motor, sensory, visual, affect, cognitive, and language systems. Of people who have survived a stroke in the long term, 50% demonstrate hemiparesis, 19% demonstrate aphasia, and 35% demonstrate clinical depression. Stroke-related deficits are further manifested in functional limitations (limitations in performing functional tasks at the whole-body level (4)). Approximately 22% of people who have survived a stroke are unable to walk without assistance, and 26% are dependent in activities of daily living. (5) The residual impairments and functional limitations in people who have survived a stroke in the long term represent a major cause of disability (limitations in performing a socially defined role in a physical or social environment (4)) in the population. Therefore, gaining a more thorough understanding of the relationships among impairments, functional limitations, and disability in people who have survived a stroke will provide a framework allowing rehabilitation professionals to identify strategies to better assist this population.

In previous work, various research groups investigated the relationships among age- and disease-associated motor impairments and limitations in physical function (ie, ability to walk, climb stairs, and rise from a chair) and disability. Specifically, muscle strength (maximum force-generating capacity) was demonstrated to have a positive association with measures of habitual walking speed, (6) stair climbing, (7) and chair rising (8) in older adults. More recently, however, researchers observed that impairments in peak skeletal muscle power (the product of the force multiplied by the velocity of shortening) explain more of the variability in function and disability than does strength in older people. (9-11) Although previous studies (12-15) demonstrated that lower-extremity strength is correlated with gait quality and other measures of function following stroke, impairments in muscle power and their association with function and disability have not been well described.

In chronic and complex diseases such as stroke, disability and quality of life are related not only to physical impairments but also to behavioral, emotional, and psychological processes. Strategies to enable patients to improve their outlook and self-manage their chronic diseases so as to optimize health are fundamental. (16) Self-efficacy is a psychological construct representing confidence in one's ability to perform a task or specific behavior or to change a specific state, regardless of circumstances or contexts. (17) Moreover, self-efficacy denotes the importance of an individual's perception of his or her ability and capability to execute and achieve important and valued outcomes. Self-care self-efficacy has been shown to be highly correlated with quality-of-life measures at both 1 and 6 months following stroke. (18) Therefore, high self-efficacy for one's physical abilities may relate to improved function, reduced disability, and improved quality of life in people who have survived a stroke in the long term.

The purpose of this investigation was to quantify the relationships among impairments in lower-extremity strength and power, measures of lower-extremity function, and global disability following stroke. Specifically, we examined whether quantitative measures of muscle strength and power in the involved lower extremity following stroke predict functional limitations and evaluated the contributions of behavioral factors, such as self-efficacy and depression, to mediating disability and quality of life.

Method

Subjects

Subjects were recruited through local newspaper advertisements, volunteer databases, and local stroke support networks. Inclusion criteria included age of 50 years or more; 6 to 24 months following a single, unilateral, ischemic, mild-to-moderate stroke (as classified with the Orpington Prognostic Scale (19)); community dwelling; independent ambulation with or without an assistive device; and willingness to attend the laboratory for 2 testing sessions. Stroke history was confirmed by medical records review. Exclusion criteria included myocardial infarction or fracture within the past 6 months, acute or terminal illness, symptomatic coronary artery disease or congestive heart failure, uncontrolled hypertension (>150/90 rum Hg), and a score 20 or less on the Mini-Mental State Examination (MMSE). (20) All subjects provided written informed consent. All outcome measures were obtained by 2 physical therapists. All study procedures were in accordance with institutional (Boston University) guidelines.

Thirty-one community-dwelling subjects who had experienced an ischemic stroke in the past 6 to 24 months and who met the inclusion criteria volunteered to participate in the study. The sample consisted of 23 men and 8 women (74.2% white, 22.6% black, and 3.2% American Indian). Sample and descriptive characteristics are shown in Table 1.

Impairment Measures

The muscle strength and power of the involved and uninvolved lower extremities were quantified with previously described methods (21,22) that were demonstrated to have good-to-excellent reliability in people following stroke. (23) Briefly, measurements of 1-repetition maximum (1RM) and peak power were obtained for the knee extensors (KEs) with computer-interfaced pneumatic resistance machines. * The 1RM is defined as the maximum load that can be moved one time only through the full range of motion (ROM) while maintaining proper form. An ultrasonic system measuring position and therefore relative motion aided the examiner in establishing a subject's full ROM during performance of the measurement with minimal resistance. The examiner progressively increased the resistance for each successful repetition until the subject could no longer move the lever arm one time through the full ROM. The maximum isometric strength of ankle plantar flexion and dorsiflexion was captured with an isokinetic dynamometer. ([dagger]) The 1RM and maximum isometric strength measurements were obtained twice, with the second evaluation occurring 3 to 7 days after the initial evaluation. The better of the 2 measurements was recorded as the 1RM and maximum isometric strength. In this sample, the intraclass correlation coefficients (ICC [3,1]) of repeated 1RM measures of involved and uninvolved KEs were both .88. For repeated maximum isometric strength measurements of the involved and uninvolved ankles in plantar flexion and dorsiflexion, the ICCs ranged from .69 to .84.

Peak skeletal muscle power is the product of the force and velocity of muscle shortening. Briefly, the power of the KEs was evaluated at 6 relative intensities (40%, 50%, 60%, 70%, 80%, and 90% the 1RM). Beginning with 40%, subjects performed 5 lifts at each established percentage of their 1RM (separated by 30 seconds) as quickly as possible through the full ROM. The software engineered for the testing equipment calculated power (in watts) between 5% and 95% of the concentric phase. Peak power for ankle plantar flexion and dorsiflexion was measured with the Cybex isokinetic dynamometer. Subjects performed 5 repetitions at angular velocities of 60[degrees], 120[degrees], and 180[degrees]/s. For KEs and ankle plantar flexion and dorsiflexion, the highest power achieved during the 2 testing sessions was recorded as the peak muscle power. In this sample, the ICCs of repeated peak power measurements of the involved and uninvolved KEs were .86 and .87, respectively. For peak power of the involved and uninvolved ankles in plantar flexion, the ICCs were .83 and .87, respectively, and for peak power of the involved and uninvolved ankles in dorsiflexion, the ICCs were .79 and .84, respectively.

Measures of Function

Habitual gait speed, stair climbing, and chair rising were measured as previously described. (24) Briefly, habitual gait speed was assessed over a 10-m distance with an ultrasonic gait speed monitor ([double dagger]) and recorded as the average of 2 trials (ICC=.98). The time to climb a single flight of stairs (10 steps, 17.7 cm per step) was determined with a handheld timer, and the better of 2 measurements was used for analyses (ICC=.98). The time to perform 5 sit-to-stand sequences from a standard chair was measured once. Repeated measures were not performed due to fatigue.

Disability Measure

The limitation dimension of the Late-Life Function and Disability Instrument (LLFDI) was used to assess disability (inability to perform major life tasks and social roles). The limitation dimension of the LLFDI evaluates self-reported limitations (capabilities) in taking part in 16 major life tasks. The limitation dimension comprises 2 domains: the instrumental role and the management role. The instrumental role domain reflects limitations in the ability to perform activities in the home and in the community. The management role domain reflects limitations in the organization and management of socially defined tasks that involve minimal mobility or physical activity. The raw scores from each item response are transformed into linear scaled scores (0-100) and subsequently summed to represent component and domain values. (25) The test-retest reliability of data for the LLFDI domains has not been established in people following stroke; however, test-retest reliability of data for the LLFDI domains was previously determined in ethnically and racially diverse adults aged 60 years and older and was found to be moderate to high (ICC=.69-.82). (25)

Quality-of-Life Measure

The shortened version of the Sickness Impact Profile (SIP) (26) was administered to evaluate 6 domains of health-related behavior (somatic autonomy, mobility control, psychological autonomy and communication, social behavior, emotional stability, and mobility range), referred to as quality of life. In people who have had a stroke, the SIP exhibits reliability, validity, and responsiveness. (27)

Behavioral Measures

The Geriatric Depression Scale was used to identify physical and nonphysical symptoms that are related to depression and that may have been present over the preceding week. (28) The Geriatric Depression Scale is a reliable and valid self-rating depression screening scale for older people and people who have had a stroke. (29) Cognitive impairment was assessed with the MMSE.

Self-Efficacy Measure

The Ewart Self-Efficacy Scale measures self-perceived ability, or confidence, to perform a number of physical tasks (eg, walking and jogging various distances, climbing stairs, lifting objects of different weights). Scores are 0 to 100, with higher scores indicating higher self-efficacy. (30) The Ewart Self-Efficacy Scale has been used extensively in studies of people who have coronary artery disease, but it has not been validated or reliability tested in people who have survived a stroke.

Data Analysis

Descriptive statistics were calculated for all subjects. Paired sample t tests were calculated for all KE and ankle plantar-flexion and dorsiflexion muscle strength and power measurements to determine differences between the involved and uninvolved limbs. A Bonferroni test-wise correction adjusted the P value to <.008 (.05/6). Pearson correlations were calculated to examine the relationships between potential adjustment variables and dependent variables. Prior to regression modeling, the normality of the dependent variables was determined with the Shapiro-Wilk test (sample size under 50). If the Shapiro-Wilk test was significant (P<.05), then the data were considered nonnormal and the dependent variable was log transformed. For each regression model, linearity was checked by adding a quadratic term to each model. If the quadratic term was significant, then the independent variable was log transformed to achieve linearity and then was used in the regression model.

The relationships of impairment with function, disability, and quality of life for both the involved and uninvolved limbs were examined by stepwise regression modeling. The relationships of our strength and power impairment measures (KEs and ankle plantar flexion and dorsiflexion) with function and disability were very similar. In this report, we have presented KE strength and power analyses, given the fundamental role of the involved musculature in the performance of lower-extremity physical functions and the previously reported excellent reliability of strength and power measurements. (23) Thus, we fit 2 regression models for each of the 3 measures of function (habitual gait speed, chair rising rime, and stair-climbing time), the disability index (limitation dimension of the LLFDI), and the quality-of-life measure (SIP) by using the KE power of the involved and uninvolved limbs. The covariates of cognition, depression, and self-efficacy were chosen because of their significant associations with dependent variables (P<.05) and their potential effect on the relationships of impairment with function and disability. The statistical significance for all multivariate regression models was accepted at P<.05. All data were analyzed with SPSS statistical software. ([section])

Results

Impairments in Muscle Strength and Power

Paired sample t tests indicated significant differences between the involved and uninvolved limbs for all KE and ankle plantar-flexion and dorsiflexion strength and power measurements (P<.004 for all measurements). Measurements and comparisons of muscle strength and power for the involved side versus the uninvolved side are shown in the Figure.

[FIGURE OMITTED]

Relationships Between Impairments and Function

Stair-climbing times (Shapiro-Wilk test: P<.006) and chair rising times (Shapiro-Wilk test: P<.001) were nonnormal, and log transformations were performed. As shown in Table 2, regression model 1 (strength) and model 2 (power) were significantly associated with habitual gait speed, stair-climbing time, and chair rising time (P<.001 for all measurements). The strength of the relationships between the 2 models and the measures of function ranged from [R.sup.2]=.43 to [R.sup.2]=.78. The KE strength and power in the involved limb were significantly associated with habitual gait speed (P<.05) and explained similar degrees of variability ([R.sup.2]=.13 and [R.sup.2]=.12, respectively). In addition, self-efficacy and sex were associated with habitual gait speed in both regression model 1 and model 2 (P<.05) but explained less of the variability than either KE strength or power in the 2 models. Analysis of stair-climbing time revealed that KE power in the involved limb explained more of the variability in performance than KE strength ([R.sup.2]=.24 versus [R.sup.2]=.11). The KE power in the uninvolved limb also was significantly associated with stair-climbing time. The relationship between self-efficacy and stair-climbing time also was significant and proved stronger in the strength model ([R.sup.2]=.11) than in the power model ([R.sup.2]=.07). In contrast to the other measures of function, neither KE strength nor power of the involved or uninvolved limb was associated with chair rising time (not retained in the stepwise model; P>.05). Self-efficacy, however, demonstrated a strong relationship with this measure in both models ([R.sup.2]=.43). As shown in Table 2, the KE strength in the uninvolved limb, cognition, and depression all failed to explain a significant portion of the variability in the stepwise regression models for measures of function (P>.05 for all measurements).

Relationships Between Impairments and Disability and Between Impairments and Quality of Life

Regression model 1 and model 2 were significantly associated with the limitation dimension, the instrumental role domain, and the management role domain (P<.001 for all measurements) of the LLFDI. The strength of the relationships between the 2 models and the 3 dimensions of disability ranged from [R.sup.2]=.43 to [R.sup.2]=.70. For the limitation dimension, self-efficacy demonstrated a strong association ([R.sup.2]=.55, P<.001), and depression also was significant (P<.05) but explained a much smaller degree of variability ([R.sup.2]=.09) in both model 1 and model 2. Self-efficacy was the only variable associated with the instrumental role domain (P<.001) and explained 63% of the variability in this measure in both models. Both self-efficacy (P<.01) and depression (P<.01) were significantly associated with the management role domain and explained similar degrees of variability in this measure ([R.sup.2]=.23 and [R.sup.2]=.24, respectively). Neither KE strength (model 1) nor power (model 2) in the involved or uninvolved limb, cognition, or sex was significantly associated with disability in either model (P>.05). These relationships are shown in Table 3.

Similar to the results for disability, regression model 1 and model 2 were significantly associated with health-related quality of life, as measured by the SIP (P<.001). The strength of the relationships of both models with quality of life was [R.sup.2]=.69. Depression and self-efficacy in model 1 and model 2 were significantly associated with quality of life (P<.001). In contrast, KE strength and KE power of both the involved and uninvolved limbs, cognition, and sex were not associated with quality of life (P>.05 for all measurements). These relationships are shown in Table 4.

Discussion and Conclusions

In the United States, both stroke prevalence and survivorship continue to heighten (3) and have made evident the need to attenuate stroke-related disability and optimize quality of life. In this report, we examined the disablement process in community-dwelling people 6 to 24 months following ischemic stroke. We observed a strong association between residual impairments in skeletal muscle strength and power on the involved side and performance on measures of gait speed and stair-climbing rime. Moreover, we demonstrated that behavioral factors, including depression and self-efficacy, more than physical impairments, are significantly related to disability in socially defined life tasks and quality of life in people who have survived a stroke. These findings have important implications for the design of both clinical interventions and future research initiatives aiming to optimize rehabilitation following stroke.

Hemiparesis is a hallmark of acute stroke and a persistent burden in people who have survived a stroke in the long term. (5,31) In this study, we observed significant deficits in lower-extremity muscle strength (>30%) and lower-extremity muscle power (>40%) for the involved side compared with the uninvolved side 6 to 24 months following the onset of stroke. Residual impairments in skeletal muscle strength in people who have survived a stroke previously were correlated with gait capacity and other measures of function. (12-15) In this study, we confirmed these findings and demonstrated, for the first time, the significant contribution of stroke-related deficits in muscle power to measures of function, namely, habitual gait speed and ability to climb a standard flight of stairs. Muscle power was the strongest predictor of stair climbing time and explained nearly twice the variability as muscle strength. These observations suggest that measures of function that require a lower percentage of maximum strength to perform (eg, gait speed on level surfaces and stairs) may be more sensitive to the velocity of movement. Thus, efforts to optimize the power of the involved musculature may confer improvements on the performance of lower-extremity physical functions following stroke. In contrast to gait speed and stair-climbing time, KE muscle strength and power failed to demonstrate an association with chair rising time. This finding may be attributed to the task relying more heavily on muscle groups (eg, core musculature), coordinated movement patterns, balance, endurance, or motor planning not assessed in this study.

Despite the associations between motor impairments and measures of function, we failed to demonstrate an association between muscle power and disability in statistical models that also included sex, depression, and self-efficacy. However, when these 3 variables were removed from our regression model, both KE strength and KE power of the involved limb were significantly associated with the limitation dimension of the LLFDI (P<.04 for both measures; data not shown). Thus, although various investigators have demonstrated the efficacy of progressive resistance training in improving skeletal muscle strength (23,32,33) and, more recently, power (23) in the involved lower extremity of people who have survived a stroke in the long terra, the effectiveness of this strategy in improving function and reducing disability independently remains ambiguous. Therefore, future studies are warranted.

The Nagi model of physical disability outlines the progression of active pathology to impairment, impairment to functional limitation, and functional limitation to disability. (24) Modifications of this scheme include the addition of internal and external factors that may augment or attenuate the disablement process. (4) Relevant to stroke, we have examined the influence of cognitive impairments and depression. At 3 months and at 1, 2, and 3 years following stroke, the prevalences of cognitive impairments (MMSE score of <24) have been reported to be 39%, 35%, 30%, and 32%, respectively, (35) and associated with institutionalization 4 years following stroke. (36) The fact that we did not detect an association between cognitive impairments and disability may be attributable to the relatively low level of these impairments in our study volunteers. In contrast, we did observe a relatively high prevalence of depression in our study participants. This finding is in agreement with a recent prospective epidemiological study that reported a high occurrence of depression in people who have survived a stroke in the long term (odds ratio=3.5, 95% confidence interval= 1.4-8.3). (37) In our statistical model, depression was not associated with measures of function; however, it was strongly correlated with disability and quality of life. The contribution of depression to disability was particularly evident in tasks that involved more socially defined roles, such as organization and management of social activities that have little reliance on an individual's physical capacities. These findings corroborate those of previous investigations (38,39) and underscore the importance of early detection and treatment of stroke-related depression in attenuating the disablement process and improving the quality of life.

Self-efficacy, the perception of one's ability, has been described as an intraindividual factor modifying the disablement process (4) and recently was proposed to be a component of the disability pathway that directly influences functional limitations. (40) In this study of people who have survived a stroke, self-efficacy, akin to muscle power, emerged as a strong predictor of measured functions. Moreover, self-efficacy was the only independent variable associated with all dimensions of self-reported disability and, in accord with a previous report, (18) was related to quality of life. These findings strongly suggest that the perception of one's ability may be as important as objective physical impairments in mediating the disablement process. Given the social-environmental context of disability and the compensatory mechanisms used by people who have survived a stroke to cope with new challenges, strategies to improve self-efficacy may have a direct and beneficial influence on multiple components of the disability pathway and quality of life.

Potential limitations of the present study also must be considered. First, we focused on selected impairments on the basis of our expertise and experience. Admittedly, the population studied has a multitude of impairments not evaluated here (eg, in sensory and language systems) that undoubtedly further contribute to the disablement process. Moreover, in a comparable sample of people that had survived a stroke (3-9 months), Nichols-Larsen et al (41) determined that individual characteristics (ie, age, race, and comorbidities) also may have a significant influence on the physical domain of health-related quality of life and are worthy of further investigation. Second, the sample size was relatively small, and the sample consisted of people 6 to 24 months following mild-to-moderate stroke, a time frame that we selected on the basis of our objective to examine residual impairments that persist despite acute rehabilitation efforts. Although the sample that we studied was representative of the population that has had a stroke on the basis of age and health status (eg, stroke severity, age, comorbidities, medication use), a larger and more functionally diverse sample would have allowed further evaluation of the impairment-function-disability relationships across various magnitudes of the chosen measures. Collectively, these data reflect the complexity of the disablement process initiated by stroke. On the basis of the work of our group and others, future studies are warranted and necessary to determine whether interventions that address residual impairments in muscle strength and power improve function and whether strategies to address depressive symptoms and low self-efficacy help attenuate functional limitations and disability and optimize quality of life in people who have survived a stroke.

This article was received May 17 2005, and was accepted May 18, 2006.

References

(1) Heart Disease and Stroke Statistics: 2004 Update. Dallas, Tex: American Heart Association; 2003:13-16.

(2) Whisnant JP. The decline of stroke. Stroke. 1984;15:160-168.

(3) Muntner P, Garrett E, Klag MJ, Coresh J. Trends in stroke prevalence between 1973 and 1991 in the US population 25 to 74 years of age. Stroke. 2002;33:1209-1213.

(4) Verbrugge LM, Jette AM. The disablement process. Soc Sci Med. 1994;38:1-14.

(5) Helgason CM, Wolf PA. American Heart Association Prevention Conference IV: prevention and rehabilitation of stroke--executive summary. Circulation. 1997;96:701-707.

(6) Rantanen T, Guralnik JM, Ferrucci L, et al. Coimpairments: strength and balance as predictors of severe walking disability. J Gerontol A Biol Sci Med Sci. 1999;54:M172-M176.

(7) Saleta GJ, Wang MY, Young JT, et al. Knee strength and lower- and higher-intensity functional performance in older adults. Med Sci Sports Exerc. 2000;32:1679-1684.

(8) Skelton DA, Greig CA, Davies JM, Young A. Strength, power and related functional ability of healthy people aged 65-89 years. Age Ageing. 1994;23:371-377.

(9) Bean JF, Kiely DK, Herman S, et al. The relationship between leg power and physical performance in mobility-limited older people. J Am Geriatr Soc. 2002;50:461-467.

(10) Foldvari M, Clark M, Laviolette LC, et al. Association of muscle power with functional status in community-dwelling elderly women. J Gerontol A Biol Sci Med Sci. 2000;55:M192-M199.

(11) Cuoco A, Callahan DM, Sayers S, et al. Impact of muscle power and force on gait speed in disabled older men and women. J Gerontol A Biol Sci Med Sci. 2004;59:1200-1200.

(12) Nadeau S, Arsenault AB, Gravel D, Bourbonnais D. Analysis of the clinical factors determining natural and maximal gait speeds in adults with a stroke. Am J Phys Med Rehabil. 1999;78:123-130.

(13) Bohannon RW, Walsh S. Nature, reliability, and predictive value of muscle performance measures in patients with hemiparesis following stroke. Arch Phys Med Rehabil. 1992;73:721-725.

(14) Olney SJ, Griffin MP, Monga TN, McBride ID. Work and power in gait of stroke patients. Arch Phys Med Rehabil. 1991;72:309-314.

(15) Patel AT, Duncan PW, Lai SM, Studenski S. The relation between impairments and functional outcomes poststroke. Arch Phys Med Rehabil. 2000;81:1357-1363.

(16) Marks R, Allegrante JP, Lorig K. A review and synthesis of research evidence for self-efficacy-enhancing interventions for reducing chronic disability: implications for health education practice (part i). Health Promot Pract. 2005;6:37-43.

(17) Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychol Rev. 1977;84:191-215.

(18) Robinson-Smith G, Johnston MV, Allen J. Self-care self-efficacy, quality of life, and depression after stroke. Arch Phys Med Rehabil. 2000;81:460-464.

(19) Kalra L, Crome P. The role of prognostic scores in targeting stroke rehabilitation in elderly patients. J Am Geriatr Soc. 1993;41:396-400.

(20) Folstein MF, Folstein SE, McHugh PR. "Mini-mental state": a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.

(21) Fielding RA, LeBrasseur NK, Cuoco A, et al. High-velocity resistance training increases skeletal muscle peak power in older women. J Am Geriatr Soc. 2002;50:655-662.

(22) Suzuki T, Bean JF, Fielding RA. Muscle power of the ankle flexors predicts functional performance in community-dwelling older women. J Am Geriatr Soc. 2001 ;49:1161-1167.

(23) Ouellette MM, LeBrasseur NK, Bean JF, et al. High-intensity resistance training improves muscle strength, self-reported function, and disability in long-term stroke survivors. Stroke. 2004;35:1404-1409.

(24) Sayers SP, Bean J, Cuoco A, et al. Changes in function and disability after resistance training: does velocity matter? A pilot study. Am J Phys Med Rehabil. 2003;82:605-613.

(25) Jette AM, Haley SM, Coster WJ, et al. Late life function and disability instrument, I: development and evaluation of the disability component. J Gerontol A Biol Sci Med Sci. 2002;57:M209-M210.

(26) de Bruin AF, Buys M, de Witte LP, Diederiks JP. The Sickness Impact Profile: SIP68, a short generic version: first evaluation of the reliability and reproducibility. J Clin Epidemiol. 1994;47:863-871.

(27) Buck D, Jacoby A, Massey A, Ford G. Evaluation of measures used to assess quality of life after stroke. Stroke. 2000;31:2004-2010.

(28) Yesavage JA, Brink TL, Rose TL, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res. 1982; 17:37-49.

(29) Johnson G, Burvill PW, Anderson CS, et al. Screening instruments for depression and anxiety following stroke: experience in the Perth community stroke study. Acta Psychiatr Scand. 1995;91:252-257.

(30) Ewart CK, Stewart KJ, Gillilan RE, Kelemen MH. Self-efficacy mediates strength gains during circuit weight training in men with coronary artery disease. Med Sci Sports Exerc. 1986;18:531-540.

(31) Sommerfeld DK, Eek EU, Svensson AK, et al. Spasticity after stroke: its occurrence and association with motor impairments and activity limitations. Stroke. 2004;35:134-139.

(32) Weiss A, Suzuki T, Bean J, Fielding RA. High intensity strength training improves strength and functional performance after stroke. Am J Phys Med Rehabil. 2000;79:369-376.

(33) Teixeira-Salmela LF, Olney SJ, Nadeau S, Brouwer B. Muscle strengthening and physical conditioning to reduce impairment and disability in chronic stroke survivors. Arch Phys Med Rehabil. 1999;80: 1211-1218.

(34) Nagi SZ. Some conceptual issues in disability and rehabilitation. In: Susman MB, ed. Sociology and Rehabilitation. Washington, DC: American Sociological Association; 1965:100-113.

(35) Patel MD, Coshall C, Rudd AG, Wolfe CD. Natural history of cognitive impairment after stroke and factors associated with its recovery. Clin Rehabil. 2003;17:158-166.

(36) Patel MD, Coshall C, Rudd AG, Wolfe CD. Cognitive impairment after stroke: clinical determinants and its associations with long-term stroke outcomes. J Am Geriatr Soc. 2002;50:700-706.

(37) Whyte EM, Mulsant BH, Vanderbilt J, et al. Depression after stroke: a prospective epidemiological study. J Am Geriatr Soc. 2004;52:774-778.

(38) Parikh RM, Robinson RG, Lipsey JR, et al. The impact of poststroke depression on recovery in activities of daily living over a 2-year follow-up. Arch Neurol. 1990;47:785-789.

(39) King RB. Quality of life after stroke. Stroke. 1996;27:1467-1472.

(40) Rejeski WJ, Focht BC. Aging and physical disability: on integrating group and individual counseling with the promotion of physical activity. Exerc Sport Sci Rev. 2002;30:166-170.

(41) Nichols-Larsen DS, Clark PC, Zeringue A, et al. Factors influencing stroke survivors' quality of life during subacute recovery. Stroke. 2005; 30:1480-1484.

* Keiser Sports Health Equipment Inc, 2470 S Cherry Ave, Fresno, CA 93706.

([dagger]) Cybex International, 10 Trotter Dr, Medway, MA 02053.

([double dagger]) OCPB Electronics, G11 6NT, Glasgow, United Kingdom.

([section]) SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.

NK LeBrasseur, PT, PhD, was a doctoral candidate, Human Physiology Laboratory, Department of Health Sciences, Sargent College of Health and Rehabilitation Sciences, Boston University, Boston, Mass, at the time of the study and is currently Assistant Professor of Medicine, Boston University School of Medicine, 670 Albany St, Rm 218, Boston, MA 02118 (USA). Address all correspondence to Dr LeBrasseur at: [e-mail].

SP Sayers, PhD, is Assistant Professor, Department of Physical Therapy, School of Health Professions, University of Missouri-Columbia, Columbia, Mo.

MM Ouellette, PT, MSPT, is Research Associate, Human Physiology Laboratory, Department of Health Sciences, Sargent College of Health and Rehabilitation Sciences, Boston University.

RA Fielding, PhD, was Associate Professor and Director, Human Physiology Laboratory, Department of Health Sciences, Sargent College of Health and Rehabilitation Sciences, Boston University, at the time of the study and is currently Director and Scientist I, Nutrition, Exercise Physiology and Sarcopenia Laboratory, Jean Mayer USDA Human Nutrition Research Center on Aging, Tufts University, Boston, Mass.

Dr LeBrasseur and Dr Fielding provided concept/idea/research design. Dr LeBrasseur, Dr Sayers, and Dr Fielding provided writing. Dr LeBrasseur and Ms Ouellette provided data collection, and Dr LeBrassenr and Dr Sayers provided data analysis.

This study was approved by the Boston University Institutional Review Board.

This work was supported by grants to Dr LeBrasseur from the Boston University Roybal Center for the Enhancement of Late-Life Function (NIH/NIA AG11669) and to Dr Fielding from the Jacob and Valeria Langeloth Foundation. The Claude D Pepper Older Americans Independence Center (AG08112) provided assistance in subject recruitment.

Table 1.
Descriptive Characteristics

Characteristic                   [bar.X]   SE

Age (y)                          66.2      1.5
Time after stroke (mo)           17.5      1.2
Stroke severity (a)               2.9      0.2
Body mass index (kg/[m.sup.2])   27.4      0.6
Comorbidities (N)                 4.4      0.3
Prescribed medications (N)        6.1      0.6
Depression score (b)             10.7      1.4
Cognition score (c)              26.6      0.9
Self-efficacy score (d)          27.0      2.5
Gait speed (m/s)                  0.68     0.1
Stair-climbing time (s)          14.8      2.0
Chair rising time (s)            23.6      1.7

(a) Orpington Prognostic Scale (mild = <3.2, moderate = 3.2-5.2).

(b) Geriatric Depression Scale (>9 = depression
of increasing sex city).

(c) Mini-Mental State Examination (0-30).

(d) Ewart Self-Efficacy Scale (0-100).

Table 2.
Associations Among Impairments, Behavioral Factors, and Functions,
as Determined by Stepwise Regression Modeling (a)

                                       Partial
Parameter              [beta]    SE    [R.sup.2]   [R.sup.2]     P

Habitual gait
  Model 1                                             .71      <.001
    KE strength (In)     0.33   0.10      .13                   .002
    Self-efficacy        1.10   0.37      .10                   .007
    Sex                  0.25   0.10      .08                    .14
  Model 2                                             .70      <.001
    KE power (In)        0.21   0.06      .12                   .003
    Self-efficacy        1.16   0.37      .11                   .005
    Sex                  0.21   0.09      .06                   .029

Stair climbing
  Model 1                                             .66      <.001
    Ke strength (In)    -0.23   0.08      .11                   .006
    Self-efficacy       -0.95   0.32      .11                   .006
  Model 2                                             .78      <.001
    KE power (In)       -0.25   0.05      .24                  <.001
    KE power (Un)        0.00   0.00      .04                   .008
    Self-efficacy       -0.75   0.26      .07                   .005

Chair rising
  Model 1                                             .43      <.001
    Self-efficacy       -0.74   0.17      .43                  <.001
  Model 2                                             .43      <.001
    Self-efficacy       -0.74   0.17      .43                  <.001

(a) All stepwise models included sex, cognition, depression,
self-efficacy, and either knee extensor (KE) strength (model 1)
or KF power (model 2) of both the involved (1n) and tic uninvolved
(Un) limbs.

Table 3.

Associations Among Impairments, Behavioral Factors, and Disability,
as Determined by Stepwise Regression Modeling (a)

                                        Partial
Parameter              [beta]    SE     [R.sup.2]   [R.sup.2]     P

Limitation dimension
  Model 1                                              .70      <.001
    Self-efficacy       53.70    8.58      .55                  <.001
    Depression          -0.42    0.17      .09                   .020
  Model 2                                              .70      <.001
    Self-efficacy       53.70    8.58      .55                  <.001
    Depression          -0.42    0.17      .09                   .020

Instrumental domain
  Model 1                                              .63      <.001
    Self-efficacy       72.50   11.85      63                   <.001
  Model 2                                              .63      <.001
    Self-efficacy       72.50   11.85      .63                  <.001

Management domain
  Model 1                                              .53      <.001
    Self-efficacy       36.90   11.56      .23                   .004
    Depression           0.74    0.23      .24                   .004
  Model 2                                              .43      <.001
    Self-efficacy       36.90   11.56      .23                   .004
    Depression           0.74    0.23      .24                   .004

(a) All stepwise models included sex, cognition, depression,
self-efficacy, and either knee extensor (SE) strength (model 1)
or KE, power (model 2) of both involved and uninvolved limbs. The
limitation dimension and instrumental and management role domains
are from the Late-Life Function and Disability Instrument.

Table 4.

Associations Among Impairments, Behavioral Factors, and Quality
of Life, as Determined by Stepwise Regression Modeling (a)

Sickness Impact                      Partial
Profile           [beta]   SE       [R.sup.2]   [R.sup.2]     P

Model 1                                            .69      <.001
  Self-efficacy   -60.70   14.20       .21                  <.001
  Depression        1.65    0.332      .46                  <.001

Model 2                                            .69      <.001
  Self-efficacy   -60.70   14.20       .21                  <.001
  Depression        1.65    0.332      .46                  <.001

(a) All stepwise models included sex, cognition, depression,
self-efficacy, and either knee extensor (KE) strength (model 1)
or KE power (model 2) of both involved and uninvolved limbs.