Factors Influencing Stroke Survivors’ Quality of Life
During Subacute Recovery
Deborah S. Nichols-Larsen, PhD, PT; P.C. Clark, PhD, RN, FAHA; Angelique Zeringue, MS, MA;
Arlene Greenspan; Sarah Blanton, DPT, NCS
Background and Purpose—Health-related quality of life (HRQOL) is an important index of outcome after stroke and may
facilitate a broader description of stroke recovery. This study examined the relationship of individual and clinical
characteristics to HRQOL in stroke survivors with mild to moderate stroke during subacute recovery.
Methods—Two hundred twenty-nine participants 3 to 9 months poststroke were enrolled in a national multisite clinical trial
(Extremity Constraint-Induced Therapy Evaluation). HRQOL was assessed using the Stroke Impact Scale (SIS),
Version 3.0. The Wolf Motor Function Test documented functional recovery of the hemiplegic upper extremity.
Multiple analysis of variance and regression models examined the influence of demographic and clinical variables across
Results—Age, gender, education level, stroke type, concordance (paretic arm?dominant hand), upper extremity motor
function (Wolf Motor Function Test), and comorbidities were associated across SIS domains. Poorer HRQOL in the
physical domain was associated with age, nonwhite race, more comorbidities, and reduced upper-extremity function.
Stroke survivors with more comorbidities reported poorer HRQOL in the area of memory and thinking, and those with
an ischemic stroke and concordance reported poorer communication.
Conclusions—Although results may not generalize to lower functioning stroke survivors, individual characteristics of
persons with mild to moderate stroke may be important to consider in developing comprehensive, targeted interventions
designed to maximize recovery and improve HRQOL. (Stroke. 2005;36:1480-1484.)
Key Words: quality of life ? rehabilitation ? stroke
are often limited to the resulting neurological impairment and
functional disability, neglecting to evaluate the total influence
of the event on a patient’s well-being. Health-related quality
of life (HRQOL) “is a self-reported measure consisting of
multiple dimensions that includes but is not limited to the
concepts of physical, social, and emotional health.”1
HRQOL measurements are potentially more relevant to
patients than measures of impairment or disability and are an
important index of outcome after stroke that can facilitate a
broader description of the disease and outcomes.2–4The most
commonly used measurements of stroke outcomes, the
Rankin Scale,5Barthel Index,6and National Institutes of
Health Stroke Scale,7have not demonstrated sensitivity to
change in mild strokes8and do not address HRQOL dimen-
sions, such as emotion, communication, and role function. In
1999, the Stroke Impact Scale (SIS) emerged as a tool to
measure these important multidimensional consequences of
stroke. This diagnosis-specific measure considers the per-
troke is often a catastrophic event affecting all aspects of
an individual’s life. Current stroke outcome assessments
spective of the patient in multiple domains, ranging from
hand function and activities of daily living to memory and
Multiple factors, including age,10–12gender,10,13–15depen-
dency in activities of daily living (ADL)/disability,10–12social
tes14,17have been associated with poorer HRQOL in stroke
survivors. Results from these studies are inconclusive and
conflicting in part because of the variability in HRQOL
measurements (Short Form 36,10,13–15Sickness Impact Pro-
file,13,16Stroke Impact Scale,17and many others11,16,18) and
the inherent heterogeneity of stroke severity and symptoms.
Using the SIS, the present study is a comprehensive analysis
investigating the impact of individual characteristics (eg, age
and gender) and clinical correlates (eg, functional level,
concordance, stroke type, comorbidities) on HRQOL in
patients with mild to moderate stroke. This information will
be helpful in developing more comprehensive interventions
in conjunction with those specifically for improvement in
Received February 10, 2005; final revision received April 5, 2005; accepted April 15, 2005.
From the Physical Therapy Program (D.S.-L.), School of Allied Medical Professions, The Ohio State University, Columbus, Ohio; the School of
Nursing (P.C.C.) and the Department of Rehabilitation Medicine (A.G., S.B.), Emory University, Atlanta, Ga; and the Division of Biostatistics (A.Z.),
Washington University, St Louis, Mo. Current address for A.G. is Centers for Disease Control and Prevention, Atlanta, Ga.
Correspondence to Deborah Nichols-Larsen, PhD, PT, 516 Atwell Hall, The Ohio State University, 1583 Perry St, Columbus, OH 43210. E-mail
© 2005 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org DOI: 10.1161/01.STR.0000170706.13595.4f
by guest on November 21, 2015http://stroke.ahajournals.org/Downloaded from
As part of a national multisite clinical trial of stroke recovery, 229
subjects were recruited at 6 clinical sites (Extremity Constraint-
Induced Therapy Evaluation [EXCITE]). Inclusion /exclusion crite-
ria for all participants included: (1) enrollment between 3 to 9
months of their first clinical stroke; (2) at least 10 degrees of wrist
extension and metacarpophalangeal (MP) and interphalangeal (IP)
extension of 2 digits and the thumb; (3) ability to transfer to and from
the toilet independently and maintain standing for two minutes; and
4) no major cognitive deficit (Mini-Mental Status Examination,
?24). (Refer to Winstein et al19for more details on clinical trial
methods.) Institutional review boards at each site approved the study
methods, and all participants provided written informed consent.
The SIS (Version 3.0; Rehabilitation Outcomes Center; Gainesville,
Fla), a 59-item self-report assessment of stroke outcome was used to
assess HRQOL in 8 domains: (1) strength, (2) hand function, (3)
mobility, (4) physical (ADL) and instrumental (IADL) activities of
daily living, (5) memory and thinking, (6) communication, (7)
emotion, and (8) social participation. Potential scores for each
domain range from 0 to 100; higher scores indicated higher HRQOL.
Reliability and validity have been established in the stroke popula-
tion, and the SIS has been found to be sensitive to change in function
across domains.9Four of the scales (strength, hand function, ADL/
instrumental ADL, mobility) can be combined into an overall
physical component score.17
The Wolf Motor Function Test (WMFT),20,21an impairment-
based assessment, was used to document functional level of the
upper extremity. The WMFT consists of 15 timed performance items
(maximum time 120 seconds), progressing from simple joint move-
ment to complex movements, and 2 strength items. The reliability
and validity of the WMFT has previously been established for
evaluating upper-extremity functional impairment in stroke.20
Comorbidities were measured by summing the major health
problems (eg, diabetes, cardiovascular disease) reported by stroke
survivors as a proxy for severity of illness. This total score was used
in the analysis.
Descriptive statistics were computed for each domain. Data were
examined for potential covariates. Multiple analysis of variance
(MANOVA), which allows comparison of group means across
several dependent variables without the multiple testing problems
associated with repeated ANOVA, was used to examine variables
that exerted influence across multiple SIS domains (SAS Version
8.2; SAS Institute). All SIS domains were used in the MANOVA to
prevent diminishing the influence of the 4 components of the
physical domain. The MANOVA appeared to be robust to a lack of
normality in some domains resulting from ceiling and floor effects.
Mean WMFT times were skewed but easily corrected to normal with
a natural-log transformation. One covariate (education) was missing
for one participant; thus, the overall mean was imputed. Regression
models were used to examine the influence of demographic and
clinical variables for the combined physical component and each
additional domain of the SIS. Both because of its reported robust-
ness22and a marked lack of normality in some other domains, the
combined physical domain was used for the regression. Potential
interactions were tested and only significant ones retained in the final
Of the 229 subjects randomized for EXCITE, 13 were
excluded from analysis because of nursing home residence,3
incomplete SIS,1or withdrawal before the baseline visit.9
Participants were primarily white, male, living at home, had
at least a high school education, and had a mean age of 62
(Table 1). On average, they had 2.79?1.42 comorbidities,
were primarily right-hand dominant with ischemic stroke, and
presented with right hemiparesis.
Descriptive data for the SIS subscales can be found in
Table 2 along with the internal consistency coefficient ? for
TABLE 2.Descriptors of SIS Domains
SubscaleMean (SD)Actual Range
Strength: subscale I
Hand function: subscale 7
Mobility: subscale 6
ADL: subscale 5
Emotion: subscale 3
Memory: subscale 2
Communication: subscale 4
Participation: subscale 8
(combined domains 1, 5, 6, 7)
*Internal consistency coefficient ?.
TABLE 1.Characteristics of Stroke Survivors (n?216 )
Years of Education
Nichols-Larsen et alQuality of Life of Stroke Survivors
by guest on November 21, 2015http://stroke.ahajournals.org/Downloaded from
each domain. On average, poorer quality of life was found in
the areas of hand function, strength, and social participation
with the highest areas in memory and communication.
Age, gender, education level, stroke type, concordance
(paretic arm?dominant hand), upper extremity motor func-
tion (log mean WMFT-Time), and comorbidities were sig-
nificantly associated across all SIS domains (Table 3). Be-
cause the overall model was significant, individual domains
Table 4 has the results of the separate regression models. It
should be noted that the R2values, although significant, were
relatively low. Importantly, these results may indicate that
factors other than those measured may have a relatively large
influence on the variability of the HRQOL scores. Older
stroke survivors, nonwhites, and those with more comorbidi-
ties and lower upper-extremity function reported poorer
HRQOL in the physical domain. Stroke survivors with more
comorbidities reported poorer HRQOL in the area of memory
and thinking; those with an ischemic stroke and concordance
of paretic arm with dominant arm reported poorer communi-
cation. The 3-way interaction with age, gender, and race was
significant for both the emotion and social participation
subscales. (See the Figure, which illustrates the interaction
for social participation.) For social participation, this interac-
tion indicated improved social participation for nonwhite men
with increasing age, decreased participation for white men
and nonwhite women with increasing age, and low scores
across ages for white women. For emotion, this interaction
delineated higher emotional QOL for younger white males
than the other 3 groups, with nonwhite males having the
lowest emotional QOL at younger ages. Women, both white
and nonwhite, tended to have relatively stable emotional
HRQOL across ages. All groups responded similarly at older
TABLE 3. MANOVA Results Across SIS Domains
VariableWilks ? ValueFP
Log mean WFMT-Time*†
Note: A proc generalized linear model was used in SAS to do the MANOVA.
*Denotes a continuous variable. Binary variables were coded as follows: male
?1, female?0; concordant?1, discordant?0; ischemic?1, hemorrhagic?0;
white?1, nonwhite?0; lives with spouse/relative?1, lives alone/with house-
keeper?0. †Mean of the Wolf times was log transformed to correct to normal.
TABLE 4.QOL Regression Models With Stroke Survivor Characteristics
Physical Domain Subscales 1, 5, 6, 7Memory Subscale 2Emotion Subscale 3Communication Subscale 4Participation Subscale 8
Mean Wolf time
Note: *Denotes a continuous variable. Binary variables were coded as follows: male?1, female?0; concordant?1, discordant?0; ischemic?1, hemorrhagic?0;
white?1, non-white?0; live with spouse/relative?1, lives alone/with housekeeper?0).
†If none of the interaction terms were significant, they were removed from the model and a main effects model was fit.
Social participation SIS domain vs age. Lines were drawn using
a smooth of the data.
by guest on November 21, 2015http://stroke.ahajournals.org/Downloaded from
The association of age and gender with all of the SIS domains
indicates the powerful influence that these demographic
variables have overall on the quality of life of stroke
survivors. Other factors influencing HRQOL across domains
were stroke type, concordance, and upper-extremity motor
function. Previous studies have also reported that age,10–12
gender,10,13–15disability,10–12and diabetes (as a common
comorbidity)14,17negatively influence HRQOL. However, in
examining the individual domains of the SIS, it is apparent
that demographic and clinical variables have disparate im-
pacts, which has not previously been delineated in the
subacute mild-to-moderate stroke population.
HRQOL in the area of overall physical functioning (sub-
scales 1, 5, 6, and 7) was significantly influenced by age,
race, comorbidities, and upper-extremity function. The last is
not surprising, as upper-extremity function is an underlying
component of the physical domain scales; additionally, the
level of disability has repeatedly been found to correlate with
diminished HRQOL in the area of physical functioning. The
impact of multiple comorbidities, most commonly high blood
pressure (n?142), diabetes (n?49), and arthritis (n?48),
might be expected to impact physical function more than
other HRQOL areas, as all 3 disorders affect general health
and physical function.16,17,23,24Finally, the effect of race
(white/nonwhite) on HRQOL needs further evaluation, as no
other study has reported a differential effect of race on
HRQOL outcomes after stroke. The existing knowledge
regarding health disparities and the fact that low-income
blacks have significantly lower functional recovery during
the first year after stroke compared with whites25may be a
partial explanation for these differences. Blacks with osteo-
arthritis, which affects physical function, have also reported
worse HRQOL.26There was a lack of association and little
variance explained among many of the clinical characteristics
and remaining domains of the SIS with only comorbidities
(memory and thinking), stroke type and affected side (com-
munication), age, race, and gender (social participation,
emotion) affecting HRQOL in these areas. This lack of
association may be attributable to the relatively high function
of this sample and the rigorous inclusion criteria, including
screening for aphasia and cognition. An alternative explana-
tion may be that the perceptions of stroke survivors about
changes in their thinking and emotions may be less accurate
than their perceptions about physical limitations. This expla-
nation is consistent with proxy ratings where others have
rated stroke survivors HRQOL lower than survivors did.9
Stroke type and upper-extremity concordance were differ-
entially associated with communication. The influence of
concordance is consistent with the unilateral representation of
speech in the left hemisphere such that those with right
hemiparesis and right-handedness would be most likely to
experience communication difficulties secondary to damage
of the left hemisphere. The effect of stroke type may be
indicative of the disparity in the number of ischemic versus
hemorrhagic strokes (189 versus 27) in this cohort.
The presence of diabetes mellitus has previously been
associated with poorer HRQOL scores.14,17The present study
supports this finding and suggests that additional comorbidi-
ties may be associated specifically with changes not only in
physical function but in memory and thinking as well.
Mackenzie and Chang16also report a decrease in psycholog-
ical HRQOL with heart disease. Diabetes has previously been
reported to affect only physical HRQOL and social partici-
pation.17Nonetheless, comorbidities of stroke survivors are
important contributors to HRQOL and should be included in
data collection to gain a greater understanding of potential
stroke research outcomes.
The 3-way interactions of race, gender, and age for social
participation and emotion both compliment and contradict
other studies, reflecting the complexity of evaluating the
impact of these variables and emphasizing the need to include
them in any analysis. In our study, white women reported
poorer HRQOL in the area of social participation across ages,
which is consistent with findings in other studies10,13–15;
conversely, nonwhite women, primarily black, reported
higher initial social participation that decreased with age.
This finding was similar to the pattern for white men. Both
white males and nonwhite females mirror previous findings
for the effects of aging.10,15Nonwhite men reported relatively
low participation initially with improved participation with
increased age. Others have reported poorer HRQOL for black
men with arthritis26; however socioeconomic status (SES)
was a confounding variable. Lower SES may be related to
employment type (white versus blue collar); blue collar
workers are less likely to return to work after stroke as are
those with less education.27The potential confound of return-
ing to work may also explain the better emotional QOL
reported by white men and poorer QOL for nonwhite men at
younger ages in the present study. These confounding vari-
ables of SES and employment may help explain the findings
for race but were not evaluated in this study. Consistent with
prior reports of decreased HRQOL for women,10,13–15women,
regardless of race, had a stable emotional HRQOL across
ages that was poorer than that for white men until after age 70
years. However, it is important that the HRQOL of minorities
and women after stroke be further explored.
In summary, it should be noted that the participants in this
study were relatively high functioning for the subacute stroke
population and had, on average, a high school education;
thus, these findings may not generalize to more severely
involved stroke survivors nor lower socioeconomic groups.
However, our results indicate the need for further examina-
tion into the issues that may be related to the HRQOL of
stroke survivors. Other areas may need exploration to better
explain factors affecting stroke survivor HRQOL. General
family functioning and family conflict surrounding stroke
recovery have been associated with negative psychological
outcomes in caregivers of stroke survivors28and may affect
stroke survivors as well. Socioeconomic status, return to
work, and other support systems not explored in the current
study may also offer greater understanding of the effect of
stroke on HRQOL. Yet, the findings of this study emphasize
the impact of and the need to take demographic and personal
characteristics, including age, race, gender, concordance, and
comorbidities, into account in the planning of poststroke
rehabilitation programs and discharge preparation. Use of the
SIS throughout the subacute stage of recovery may facilitate
Nichols-Larsen et al Quality of Life of Stroke Survivors
by guest on November 21, 2015 http://stroke.ahajournals.org/Downloaded from
a better understanding of individual needs and, thereby,
planning for programming during recovery.
This study was supported by EXCITE grant RO1 HD 37606 (from
the National Institute of Child Health and Human Development and
the National Institute of Neurological Disorders and Stroke; to
Steven L. Wolf, principal investigator) and Family Function, Stroke
Recovery, and Caregiver Outcomes grant RO1 NR07612-01 (from
the National Institute of Nursing Research; to P.C.C.). Participating
EXCITE sites include Emory University, University of Southern
California, University of North Carolina at Chapel Hill, Wake Forest
University, The Ohio State University, University of Florida–
Gainesville, University of Alabama at Birmingham, and Washington
University School of Medicine, St Louis, Mo.
1. WHOQOL Group. Development of the World Health Organization
WHOQOL-BREF quality of life assessment. Psychol Med. 1998;28:
2. Leplege A, Hunt S. The problem of quality of life medicine. JAMA.
3. Moon Y, Kim S, Kim H, Won M, Kim D. Correlates of quality of life
after stroke. J Neurol Sci. 2004;224:37–41.
4. Sturm JW, Osborne RH, Dewey HM, Donnan GA, Macdonell RAL,
Thrift AG. Brief comprehensive quality of life assessment after stroke:
the assessment of quality of life instrument in the North East Melbourne
Stroke Incidence Study (NEMESIS). Stroke. 2002;33:2888–2894.
5. Rankin J. Cerebral vascular accidents in patients over the age of 60. Scott
Med J. 1957;2:200–215.
6. Mahoney F, Barthel D. Functional evaluation: the Barthel Index. Md State
Med J. 1965;14:61–65.
J, Eberle R, Hertzberg V, Walker M. Measurements of acute cerebral
infarction: a clinical examination scale. Stroke. 1989;20:864–870.
8. Wade DT. Measurement in neurological rehabilitation. In Harrison, MA,
ed. Physiotherapy in Stroke Management. New York: Churchill Liv-
ingstone; 1995;197 (Abstract).
9. Duncan PW, Wallace D, Lai SM, Johnson D, Embretson S, Laster LJ.
The Stroke Impact Scale version 2.0: evaluation of reliability, validity,
and sensitivity to change. Stroke. 1999;30:2131–2140.
10. Hackett ML, Duncan JR, Anderson CS, Broad JB, Bonita R. Health-
related quality of life among long-term survivors of stroke: results from
the Auckland Stroke Study, 1991–1992. Stroke. 2000;31:440–447.
11. Rochette A, Desrosiers J, Noreau L. Association between personal and
environmental factors and the occurrence of handicap situations fol-
lowing a stroke. Disabil Rehabil. 2001;23:559–569.
12. Sturm JW, Donnan GA, Dewey HM, Macdonell RA, Gilligan AK, Thrift
AG. Determinants of handicap after stroke: the North East Melbourne
Stroke Incidence Study (NEMESIS). Stroke. 2004;35:715–720.
13. Carod-Artal J, Egido JA, Gonzalez JL, de Seijas EV. Quality of life
among stroke survivors evaluated 1 year after stroke: experience of a
stroke unit. Stroke. 2000;31:2995–3000.
14. Suenkeler IH, Nowak M, Misselwitz B, Kugler C, Schreiber W, Oertel
WH, Back T. Timecourse of health-related quality of life as determined
3, 6 and 12 months after stroke. Relationship to neurological deficit,
disability and depression. J Neurol. 2002;249:1160–1167.
15. Hopman WM, Verner J. Quality of life during and after inpatient stroke
rehabilitation. Stroke. 2003;34:801–805.
16. Mackenzie AE, Chang AM. Predictors of quality of life following stroke.
Disabil Rehabil. 2002;24:259–265.
17. Lai S, Studenski S, Duncan PW, Perera S. Persisting consequences of
stroke measured by the Stroke Impact Scale. Stroke. 2002;33:1840–1844.
18. Mercier L, Audet T, Hebert R, Rochette A, Dubois MF. Impact of motor,
cognitive, and perceptual disorders on ability to perform activities of daily
living after stroke. Stroke. 2001;32:2602–2608.
19. Winstein C, Miller J, Blanton S, Morris D, Uswatte G, Taub E, Nichols
D, Wolf S. Methods for a multi-site randomized trial to investigate the
effect of constraint-induced movement therapy in improving upper
extremity function among adults recovering from cerebrovascular stroke.
Neurorehabil Neural Repair. 2003;17:137–152.
20. Wolf SL, Catlin PA, Ellis M, Archer AL, Morgan B, Piacentino A.
Assessing Wolf Motor Function Test as outcome measure for research in
patients after stroke. Stroke. 2001;32:1635–1639.
21. Morris DM, Uswatte G, Crago JE, Cook ER, Taub E. The reliability of
the Wolf Motor Function Test for assessing upper extremity function after
stroke. Arch Phys Med Rehabil. 2001;82:750–755.
22. Duncan PW, Bode RK, Lai SM, Perera S. Rasch analysis of a new
stroke-specific outcome scale: the Stroke Impact Scale. Arch Phys Med
23. Lalonde L, O’Connor A, Joseph L, Grover SA, Group CCCA. Health-
related quality of life in cardiac patients with dyslipidemia and hyper-
tension. Qual Life Res. 2004;13:793–804.
24. Briggs A, Scott E, Steele K. Impact of osteoarthritis and analgesic
treatment on quality of life of an elderly population. Ann Pharmacother.
25. Horner RD, Swanson JW, Bosworth HB, Matchar DB; VA Acute Stroke
(VAST) Study Team. Effects of race and poverty on the process and
outcome of inpatient rehabilitation services among stroke patients. Stroke.
26. Ibrahim SA, Burant CJ, Siminoff LA, Stoller EP, Kwoh CK. Self-
assessed global quality of life: a comparison between African-American
and white older patients with arthritis. J Clin Epidemiol. 2002;55:
27. Vestling M, Tufvesson B, Iwarsson S. Indicators for return to work after
stroke and the importance of work for subjective well-being and life
satisfaction. J Rehabil Med. 2003;35:127–131.
28. Clark PC, Dunbar SB, Shields CG, Viswanathan B, Aycock DM, Wolf
SL. Influence of stroke survivor characteristics and family conflict sur-
rounding recovery on caregivers’ mental and physical health. Nurs Res.
by guest on November 21, 2015http://stroke.ahajournals.org/ Downloaded from
Blanton Download full-text
Deborah S. Nichols-Larsen, P.C. Clark, Angelique Zeringue, Arlene Greenspan and Sarah
Factors Influencing Stroke Survivors' Quality of Life During Subacute Recovery
Print ISSN: 0039-2499. Online ISSN: 1524-4628
Copyright © 2005 American Heart Association, Inc. All rights reserved.
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
2005;36:1480-1484; originally published online June 9, 2005;
World Wide Web at:
The online version of this article, along with updated information and services, is located on the
is online at:
Information about subscribing to
Information about reprints can be found online at:
document. Permissions and Rights Question and Answer process is available in the
Request Permissions in the middle column of the Web page under Services. Further information about this
Once the online version of the published article for which permission is being requested is located, click
can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Requests for permissions to reproduce figures, tables, or portions of articles originally published
by guest on November 21, 2015 http://stroke.ahajournals.org/Downloaded from