Relationship Between Disability and Health-Related Quality of Life and Caregiver Burden in Patients With Upper Limb Poststroke Spasticity
Article Outline
Objective
To evaluate the relationship between disability and both health-related quality of life (HRQoL) and caregiver burden in patients with upper limb poststroke spasticity.
Design
Multicenter open-label study.
Setting
Thirty-five sites in North America.
Participants
Patients (N = 279) with upper limb poststroke spasticity.
Methods
Post hoc analyses of data from an open-label study were performed to estimate HRQoL and caregiver burden at study baseline across levels of disability in 4 problem domains: hygiene, dressing, limb posture, and pain. Disability severity in these areas was determined by using the 4-point Disability Assessment Scale rated by the physicians.
Main Outcome Measurements
HRQoL measured by the patient-reported EuroQol 5 Dimensions questionnaire and the Stroke-Adapted Sickness Impact Profile and caregiver burden.
Results
At study baseline, increasing disability in the hygiene, dressing, and pain domains of the Disability Assessment Scale was associated with diminishing HRQoL scores (P < .002) measured by the EuroQol 5 Dimensions. By using the Stroke-Adapted Sickness Impact Profile, greater disability scores in all problem domains were significantly associated with higher overall dysfunction scores (P ≤ .05). Within the physical dimension of the Stroke-Adapted Sickness Impact Profile, significant associations also were observed in all domains. At baseline, caregiver burden was significantly related to increasing levels of hygiene and dressing domain severity (P ≤ .05). Caregiver assistance requirement increased from approximately 9.0-28.2 hours per week in the hygiene domain and 3.3-32.1 hours per week in the dressing domain as disability increased from “none” to “severe.”
Conclusions
In patients with upper limb poststroke spasticity, increasing disability in the hygiene, dressing, and pain domains of the Disability Assessment Scale were associated with diminishing HRQoL. Furthermore, these patients required caregiver assistance proportionally related to the severity of their disability in the hygiene and dressing domains.
Introduction
Each year, approximately 795,000 persons experience a new or recurrent stroke [1]. Among persons 45-64 years of age, approximately 8%-12% of patients with ischemic stroke and 37%-38% of patients with hemorrhagic stroke die within 30 days [1]. Among survivors, 30.7% have received outpatient rehabilitation for various conditions, including spasticity [1]. It is estimated that residual spasticity occurs in 19%-38% of stroke patients [2, 3]. Spasticity is characterized by increased muscle tone, exaggerated tendon jerks, and poor control of voluntary movement that occurs as part of an upper motor neuron syndrome. Motor dysfunctions can arise in the upper (shoulder, elbow, and wrist) or lower (knee and ankle) limb, or both. The principal goals of managing spasticity are to improve active and passive functions, to avoid progression of impairment, and to provide relief of symptoms [4].
CME Question In this study, caregiver assistance/burden was found to be proportionally related to severity of disability in which of the following domains?
Answer online at me.aapmr.org
Several published studies have reported either the impact of upper limb poststroke spasticity on health-related quality of life (HRQoL) or how treatments for upper limb poststroke spasticity have affected HRQoL [5, 6, 7, 8, 9]. Both generic (Medical Outcomes Study 36-Item Short-Form Health Survey [9], EuroQol questionnaire [EQ-5D] [10], Assessment of Quality of Life [7]) and stroke-specific (Stroke-Adapted Sickness Impact Questionnaire [SA-SIP30] [11], Niemi QoL scale [12], the Ferrans and Powers QoL Index—Stroke Version [13]) instruments have been used to measure HRQoL [14]. Nevertheless, little is known about the relationship between disability level and outcomes such as HRQoL and caregiver burden in poststroke patients with spasticity. The Food and Drug Administration has established that HRQoL data can be used to characterize health outcomes for patients [15].
The purpose of this study was to examine the relationship among levels of disability in patients with upper limb poststroke spasticity and both HRQoL and caregiver burden. The hypothesis evaluated by this study was that greater disability is associated with worse HRQoL and a greater need for caregiver assistance. Although this may be an apparent relationship, we specifically aimed to measure the effect size of the association among varying levels of disability and HRQoL and caregiver burden in patients with upper limb poststroke spasticity. To the best of our knowledge, this information is currently not available in the medical literature. Disability in the hygiene, dressing, limb posture, and pain domains, which are common among patients with upper limb poststroke spasticity, was the focus of this study.
Methods
Study Design
Post hoc analyses were conducted on data from a large, prospective, multicenter, open-label study of poststroke spasticity in the upper extremity, with a 12-month follow-up [8]. A brief description of the original study design is provided here, but complete details can be found elsewhere [8]. The study was conducted at 35 clinical sites across North America between April 2003 and December 2004, and was originally designed to assess the safety and efficacy of repeated treatments with onabotulinumtoxinA on functional disability, quality of life, and muscle tonicity. The direct effect of study treatment on outcomes is not of interest in the current analysis. The study was approved by the institutional review board at each participating center. All subjects provided written informed consent before any study-related procedure.
Subjects
All subjects recruited had a history of stroke that resulted in spasticity limited to one upper limb [8]. Criteria for selection into this study included age ≥21 years, weight ≥40 kg, and the need for onabotulinumtoxinA injection of at least 200 units in the wrist and finger flexors. Exclusion criteria were as follows: a stroke within 6 months before enrollment; any condition that might place the patient at greater risk with botulinum treatment; any disorder that would interfere with neuromuscular function; infection or dermatologic condition at the injection sites; fixed contracture of the study limb; inflammation in the study limb that limited joint movement; past or planned treatment for spasticity; casting of the study limb during the study or within 3 months before treatment; current treatment with intrathecal baclofen; profound atrophy of the arm muscles; and hypersensitivity to the study medication [8].
Outcome Measures
Investigators (physicians) rated patients' disability at study baseline by using the Disability Assessment Scale (DAS) [16]. Problem domains of hygiene, dressing, limb posture, and pain were assessed on a 4-point scale: 0, none; 1, mild; 2, moderate; and 3, severe. The key outcomes measures were HRQoL and caregiver burden. HRQoL was assessed by using the EQ-5D questionnaire and the SA-SIP30.
The EQ-5D questionnaire is a patient-reported description of 5 health status dimensions, including mobility, self-care, usual activities, pain and/or discomfort, and anxiety and/or depression [10, 17]. Each dimension of the EQ-5D is rated on a 3-point scale: 1, no problems; 2, some or moderate problems; and 3, extreme problems. EQ-5D health states across all dimensions were converted into a single summary index that ranged from 0 (equivalent to death) to 1 (perfect health) by using the U.S. value sets [18]. Mean EQ-5D index scores across DAS levels for each domain at study baseline were calculated.
The patient-reported validated 30-item shorten version of the full 136-item Sickness Impact Profile also was used to assess HRQoL after stroke [11]. It consisted of questions in 8 domains: body care and movement, mobility, household management, ambulation, social interaction, communication, emotional behavior, and alertness behavior [11, 14]. From these domains, an overall score and 2 dimension (physical and psychosocial) functioning scores were generated, which ranged from 0 (no dysfunction) to 100 (maximum dysfunction). Mean scores across disability levels for each domain at study baseline were estimated.
For caregiver burden, the subjects were asked to report whether they received assistance in their homes from a friend, family member, or health care professional over the past 4 weeks. If yes, then the subjects reported the average hours of assistance received per week. If assistance was provided by a friend or a family member, then the subjects reported any time taken off from work to provide assistance. Pairwise comparisons were performed for mean hours per week of caregiver assistance across DAS scores for each domain at study baseline.
Statistical Analysis
Study data were managed and analyzed by using the SAS software (version 9.1; SAS Institute Inc, Cary, NC). Means and standard deviations were used to summarize continuous variables and proportions used for categorical variables. For EQ-5D, SA-SIP30, caregiver burden, and age, outcomes were compared across DAS levels by using one-way analysis of variance. F tests were used for statistical comparisons with a significance level of .05. For gender distribution at baseline, the χ2 test was used for the test of significance (P < .05).
Results
Demographic and Baseline Characteristics
The patients recruited into the study were 21-88 years of age (mean age, 58.2 years). Approximately, 53.8% (150 of 279) were men and 82.1% (229/279) were white. Thrombotic stroke (45.5%) was the most frequent cause of stroke. The mean time since the onset of stroke was 5.0 years (range, 0.2-31.5 years). The mean time since the onset of upper limb spasticity was 4.4 years (range, 0.1-31.5 years). The mean age was similar across DAS disability levels in each problem domain at study baseline (Table 1). Gender distribution also was similar across DAS disability levels in each problem domain (Table 1).
Table 1. Patient demographics at baseline and week 48 by disability levels across problem domains
| Domain (DAS score) | N | Age at Baseline, Mean ±SD, y⁎ | % Men at Baseline⁎ |
|---|---|---|---|
| Hygiene | |||
| None (0) | 55 | 55.3±15.3 | 62 |
| Mild (1) | 88 | 57.0±11.0 | 44 |
| Moderate (2) | 89 | 59.9±13.2 | 55 |
| Severe (3) | 47 | 60.6±14.0 | 60 |
| Dressing | |||
| None (0) | 29 | 56.1±12.4 | 62 |
| Mild (1) | 66 | 57.0±12.8 | 47 |
| Moderate (2) | 116 | 58.0±13.7 | 54 |
| Severe (3) | 68 | 60.6±13.2 | 56 |
| Limb posture | |||
| None (0) | 20 | 60.3±14.4 | 50 |
| Mild (1) | 50 | 58.6±13.7 | 56 |
| Moderate (2) | 115 | 58.9±13.6 | 47 |
| Severe (3) | 94 | 56.8±12.3 | 62 |
| Pain | |||
| None (0) | 145 | 59.3±14.0 | 59 |
| Mild (1) | 61 | 57.2±11.6 | 54 |
| Moderate (2) | 55 | 58.6±13.0 | 44 |
| Severe (3) | 18 | 51.1±11.9 | 44 |
⁎ Mean age (one-way analysis of variance, F test) and percentage of men (χ2 test) across DAS disability levels were not statistically significant. |
Relationship Between Disability and HRQoL
At study baseline, increasing disability in the hygiene, dressing, and pain domains was associated with diminishing EQ-5D index scores (P < .002) (Table 2). For the limb posture domain, increasing disability was not associated with statistically significant decrease in the EQ-5D score. The mean EQ-5D scores in severe disability (ie, DAS score of 3) ranged from 0.47 to 0.64 across any of the 4 problem domains. The patients with no disability (ie, DAS score of 0) reported mean EQ-5D scores between 0.70 and 0.74 across any of the 4 problem domains. In general, the greatest change in EQ-5D score appears to be between “moderate” and “severe” disability. The difference in mean EQ-5D scores between “severe” and “no” disability ranged from 0.12 to 0.25.
Table 2. Relationship between disability (DAS) and EQ-5D index score across problem domains
| Domain (DAS score) | N | EQ-5D Index Score at Baseline, Mean±SD |
|---|---|---|
| Hygiene | ||
| None (0) | 54 | 0.70±0.17⁎ |
| Mild (1) | 87 | 0.68±0.18 |
| Moderate (2) | 89 | 0.67±0.17 |
| Severe (3) | 47 | 0.58±0.20 |
| Dressing | ||
| None (0) | 28 | 0.74±0.17⁎ |
| Mild (1) | 65 | 0.72±0.14 |
| Moderate (2) | 116 | 0.66±0.17 |
| Severe (3) | 68 | 0.59±0.21 |
| Limb posture | ||
| None (0) | 20 | 0.70±0.14 |
| Mild (1) | 48 | 0.70±0.18 |
| Moderate (2) | 115 | 0.67±0.19 |
| Severe (3) | 94 | 0.64±0.18 |
| Pain | ||
| None (0) | 144 | 0.72±0.15⁎ |
| Mild (1) | 60 | 0.67±0.14 |
| Moderate (2) | 55 | 0.56±0.20 |
| Severe (3) | 18 | 0.47±0.20 |
⁎ P < .002, 1-way analysis of variance; comparison across DAS disability levels. |
At study baseline, greater disability scores in the hygiene, dressing, limb posture, and pain domains were significantly (P < .05) associated with higher overall SA-SIP30 dysfunction scores (Table 3). Greater DAS scores were significantly (P < .05) associated with higher SA-SIP30 physical dimension scores in all domains. The relationship between greater disability and higher psychosocial dimension scores was generally observed; however, statistical significance was achieved only for the domains of dressing, pain, and hygiene. The greatest change in SA-SIP30 score appears to be between “moderate” and “severe” disability. The difference in scores between the “none” and “mild” group was not discernible. The difference in overall mean SA-SIP30 scores between “severe” and “no” disability ranged from 8.2 to 21.2.
Table 3. Relationship between disability (DAS) and SA-SIP30 across problem domains
| Domain (DAS score) | N | SA-SIP30, Mean±SD | ||
|---|---|---|---|---|
| Overall | Physical Dimension | Psychosocial Dimension | ||
| Hygiene | ||||
| None (0) | 55 | 33.5±18.9⁎ | 41.8±24.0⁎ | 23.8±19.8⁎ |
| Mild (1) | 88 | 37.7±19.9 | 43.5±25.1 | 30.9±21.8 |
| Moderate (2) | 87 | 43.2±18.8 | 50.4±22.8 | 34.8±22.1 |
| Severe (3) | 47 | 54.7±16.7 | 66.4±16.4 | 41.0±26.3 |
| Dressing | ||||
| None (0) | 28 | 34.6±18.9⁎ | 38.3±21.1⁎ | 30.4±23.0† |
| Mild (1) | 66 | 31.3±16.7 | 37.3±21.6 | 24.2±19.3 |
| Moderate (2) | 115 | 43.0±20.8 | 49.4±23.7 | 35.5±24.7 |
| Severe (3) | 68 | 51.5±16.2 | 64.8±20.2 | 36.0±21.2 |
| Limb Posture | ||||
| None (0) | 20 | 39.7±18.3† | 51.5±26.5⁎ | 25.9±14.4 |
| Mild (1) | 50 | 39.3±22.8 | 44.8±24.8 | 32.9±26.5 |
| Moderate (2) | 115 | 37.5±20.2 | 44.1±24.5 | 29.9±22.3 |
| Severe (3) | 92 | 47.9±16.8 | 57.5±21.1 | 36.7±22.6 |
| Pain | ||||
| None (0) | 145 | 38.0±19.5⁎ | 46.3±24.5† | 28.2±21.1⁎ |
| Mild (1) | 61 | 38.2±19.9 | 44.9±24.0 | 30.4±23.3 |
| Moderate (2) | 53 | 51.0±17.1 | 59.2±21.6 | 41.5±23.1 |
| Severe (3) | 18 | 52.6±20.5 | 57.7±22.3 | 46.6±23.7 |
⁎ P < .001, one-way ANOVA; comparison across DAS disability levels. |
† P < .05, one-way ANOVA; comparison across DAS disability levels. |
Relationship Between Disability and Caregiver Burden
At study baseline, patients exhibited levels of caregiver burden that were significantly (P < .05) and directly related to increasing disability in hygiene and dressing domains (Table 4). As disability in the hygiene domain increased from “none” to “severe,” the hours of caregiver assistance required per patient per week increased from approximately 9.0 hours per week to approximately 28.2 hours per week. Similarly, increasing disability in the dressing domain from “none” to “severe” was associated with a 10-fold increase in caregiver hours (from approximately 3.3 to 32.1 hours per week). No statistically significant differences were observed in caregiver time for increasing disabilities in the limb posture and pain domains (Table 4).
Table 4. Relationship between disability (DAS) and caregiver burden (h/wk) across problem domains
| Domain (DAS score) | n | Caregiver Burden (h/wk), Mean±SD |
|---|---|---|
| Hygiene | ||
| None (0) | 53 | 9.0±25.3⁎ |
| Mild (1) | 85 | 13.6±27.5 |
| Moderate (2) | 84 | 21.5±39.3 |
| Severe (3) | 42 | 28.2±38.6 |
| Dressing | ||
| None (0) | 29 | 3.3±7.8† |
| Mild (1) | 61 | 5.2±11.2 |
| Moderate (2) | 111 | 19.6±35.8 |
| Severe (3) | 63 | 32.1±43.7 |
| Limb posture | ||
| None (0) | 16 | 11.8±25.8 |
| Mild (1) | 50 | 17.7±36.7 |
| Moderate (2) | 109 | 15.9±32.0 |
| Severe (3) | 89 | 20.3±35.2 |
| Pain | ||
| None (0) | 138 | 17.6±33.4 |
| Mild (1) | 59 | 14.0±32.8 |
| Moderate (2) | 49 | 17.1±31.3 |
| Severe (3) | 18 | 29.0±43.1 |
⁎ P < .05, one-way ANOVA; comparison across DAS disability levels. |
† P < .00, one-way ANOVA; comparison across DAS disability levels. |
Discussion
Published studies have reported either the impact of upper limb poststroke spasticity on HRQoL or how treatments for upper limb poststroke spasticity have affected HRQoL [5, 6, 7, 8]. However, the relationship among levels of disability and outcomes such as HRQoL and caregiver burden has not been fully explored in adults with upper limb poststroke spasticity in the manner that was conducted in this study. This study confirms that increasing levels of disability are associated with an important reduction in HRQoL, as captured by 2 independent validated instruments, and an increase in caregiver burden. Evidence for such an association is lacking in the medical literature. These findings are potentially useful in future research in 2 ways: (1) to characterize the disease burden of patients with upper limb poststroke spasticity in terms of patient-reported outcomes and societal impact, and (2) to distinguish the disease burden as a function of the degree of disability that patients with upper limb poststroke spasticity experience. Through better understanding of the impact of varying amount of disabilities on HRQoL and caregiver burden, one can indirectly predict the improvement in health and the reduction in costs if therapies become available to decrease disability. Therefore, this information can be important to guide medical and/or policy decision making. Moreover, this information can be useful in formal cost-effectiveness analysis to evaluate the value of potential new therapies, because the cost of the therapies can be offset through reduction in caregiver burden, and outcomes such as life expectancy can be adjusted by the quality of life impact. In many countries (such as the United Kingdom, Canada, and Australia), formal cost-effectiveness analyses are used to inform reimbursement decisions.
In all problem domains evaluated, with the exception of limb posture, reduction in disability translated into improvement in HRQoL as measured by the EQ-5D. In this study, substantial differences in the EQ-5D scores between severe disability (DAS = 3) and no disability (DAS = 0) were observed for hygiene (0.12), dressing (0.15), and pain (0.25) domains. By comparison, minimally important differences in the EQ-5D reported in other studies for various medical conditions are much smaller, such as 0.04 for coronary heart disease, 0.09-0.10 for cancers, and 0.09-0.20 for psoriasis [19, 20, 21]. Although the minimally important difference from an HRQoL instrument should be uniquely established for each medical condition, the magnitude of these differences noted in this study would suggest that these changes are likely to be clinically meaningful in addition to being statistically significant. A similar analysis was repeated by using data at week 48 (results not shown) and revealed a consistent relationship between disability and HRQoL as observed at study baseline. Other factors, such as age and gender, could potentially cause shifts in patient-reported HRQoL; however, we did not observe significant differences in the age and gender distributions between disability levels at study baseline, which suggests that these factors are unlikely to bias our findings.
Our analysis showed that shifts in disability are strongly correlated with changes in HRQoL. The implication of this finding is that treatment that demonstrates disability reduction can potentially show a considerable improvement in HRQoL. Moreover, a simple 4-point scale such as DAS, rated by the physician, demonstrated a remarkable ability to discern important differences in patient-reported outcomes. It should be noted that these differences were observed in a population of stroke patients with significant spasticity. For hygiene, the greatest improvement in HRQoL coincided with a disability reduction from a severe to a moderate level. However, for the pain domain, marked improvement in HRQoL was observed for each disability level from a DAS of 3 to a DAS of 0.
There was a strong correlation between the patient-rated EQ-5D questionnaire and the investigator-rated DAS when used to classify problems relevant to upper limb poststroke patients (ie, hygiene, dressing, limb posture, pain). The primary reason for this was the domains included in the EQ-5D questionnaire. The pain and/or discomfort, usual activities, and self-care domains of the EQ-5D corresponded particularly well with the pain, dressing, and hygiene problem domains, respectively. It is important to note that, although some domains in the EQ-5D overlap with problem domains measured by DAS, these scales are rated by separate individuals (patients versus physicians). Therefore, a strong correlation between the improvement in disability that the physicians observed and the benefit that patients perceived provides confirmation that changes in disability are meaningful and consistent for physicians and patients. None of the EQ-5D domains corresponded well with the limb posture domain, which could explain why EQ-5D index scores were not markedly different across disability levels in the limb posture domain. An alternative explanation is that limb posture has limited impact on quality of life in general.
Post et al [22] summarized the health utility of health states after stroke based on a systematic review of the literature. Health utility is often used in economic evaluation to conduct formal cost-utility analysis. A utility is a net expression of several dimensions of HRQoL summarized into a single score between 0 (dead) and 1 (perfect health) that can be used as weights. The EQ-5D index score used in this study is an example of a health utility. Post et al [22] also reported that stroke survivors assign a utility of 0.41 for a major stroke and 0.72 for a minor stroke. When using the EQ-5D questionnaire, stroke survivors reported utilities of 0.32 and 0.71 for major and minor stroke, respectively [22]. This range is similar to the results based on the EQ-5D from this study. Post et al [22] noted considerable variability in reported health utility values due to the participants of the survey and the instrument used to elicit HRQoL.
To put the impact of poststroke spasticity on HRQoL into perspective, Sullivan et al [23] reported that the mean EQ-5D among noninstitutionalized adults ≥18 years of age in the U.S. population was 0.867. This was based on a nationally representative sample (N = 38,678) collected in the Medical Expenditure Panel Survey (MEPS). However, the mean age in MEPS was 46, which is younger than patients in this study with poststroke spasticity. By comparison, in the MEPS, the patients with angina, myocardial infarction, and coronary heart disease, who were, on average, in their 60s, reported mean EQ-5D scores of 0.709, 0.725, and 0.725, respectively [23]. Patients with stroke in MEPS reported mean EQ-5D of 0.694 [23]. Patients with upper limb poststroke spasticity with more severe disabilities had similar HRQoL to that of patients with nonhypertensive congestive heart failure (0.636) and rheumatoid arthritis (0.661) [23].
In general, the relationship between DAS and HRQoL as measured by the SA-SIP30 substantiated the findings of the analysis on disability and EQ-5D. Although not every association between DAS and SA-SIP30 reported in Table 3 was determined to be statistically significant, the majority were, and all exhibited trends in the same direction (ie, more disability was associated with worse HRQoL and vice versa). The results also showed that the greatest change in HRQoL is between “moderate” and “severe” disability. When taken together, the 2 instruments used in this study corroborated the findings of each other and confirmed that differences in disability levels have meaningful differences in HRQoL. Across all problem domains, increasing disability was significantly associated with higher overall SA-SIP30 scores at study baseline. An analysis conducted on data at week 48 (results not shown) revealed the same relationship. The physical dimension of the SA-SIP30 appeared to have greater emphasis compared with the psychosocial dimension due to the stronger effect observed, which can be partly explained by the alignment of the 5 questions covered by the “body care and movement” domain contained in the SA-SIP30 with the hygiene and dressing problem domains.
Results of this study showed that, with increasing disability in the hygiene and dressing domains, greater hours of caregiver assistance were required. The increase in hours of required caregiver assistance from no disability to severe disability on the DAS ranged from approximately 3- to 10-fold for the hygiene and dressing domains, respectively. The results suggest that patients with severe disability may require approximately 30 hours on average per week of caregiver assistance. If such help is provided as formal care, then this equates to nearly three-fourths of a full-time equivalent based on U.S. standard. Even patients with moderate disability may require about 20 hours of help on average, which is equivalent to one-half of a full-time equivalent. These findings suggest that the caregiver costs of upper limb poststroke spasticity could be very high among individuals with moderate-to-severe disability but may be reduced substantially if there are effective treatments.
Disabilities in limb posture and pain domains were found not to be positively correlated with caregiver burden. The lack of association in the pain domain validates the notion that caregiver time would not be expected to vary across disability levels in the pain domain because caregivers are unlikely to directly help with pain relief. However, it was unclear why a significant association between DAS and caregiver burden for limb posture was not observed. One would expect that reducing limb posture disability should lead to fewer hours of caregiver help. In the limb posture domain, patients with “none,” “mild,” and “moderate” disability required similar hours of caregiver assistance, and only those with “severe” disability required many more hours (Table 4). This observation suggests that there may be a disability threshold in limb posture beyond which additional caregiver assistance is needed.
The study results need to be considered in light of several limitations that exist. First, the study was conducted on a sample of 279 subjects. It was a subdivision of stroke patients who required intervention for their spasticity. To generalize this information for the stroke population overall, it would be necessary to repeat this study in a group without significant tone. When the study population was stratified into DAS disability subgroups, the sample sizes used for comparison were further reduced. Second, when using the EQ-5D questionnaire, the patients reported their overall HRQoL across 5 dimensions included in the instrument and not their quality of life solely affected by an individual problem domain. Therefore, the EQ-5D index scores reported in hygiene, dressing, limb posture, and pain domains cannot be viewed as completely independent, and some degree of overlap does exist across these domains. Third, patients reported hours of caregiver help based on their total need across all problem domains. Therefore, hours of caregiver help reported for each specific domain were not necessarily due to only that problem. Fourth, patients in this study may have other deficits (eg, cognitive and linguistic deficits) in addition to the 4 poststroke-related disability domains that were examined in this study. These deficits, as well as other comorbid medical conditions, may contribute to a reduction in HRQoL and increase in caregiver burden. Therefore, it is not possible to conclude definitively that the size of the association is exclusively due to upper limb poststroke spasticity. However, it is assumed that the other stroke-related deficits and comorbid conditions are distributed equally across the 4 disability levels (ie, “severe,” “moderate,” “mild,” and “none”) of each disability domain. If this is true, then the size of the association observed in this study is most likely due to disability in the hygiene, dressing, and pain domains. Fifth, one-way analysis of variance was used to compare outcomes across different levels of disabilities; however, this test does not account for the ordinal nature of DAS, which could result in a loss of statistical power. In spite of this, many of the associations were found to be statistically significant.
Conclusion
This study provides quantitative evidence to conclude that increasing disability in the hygiene, dressing, and pain domains is associated with diminishing HRQoL in patients with upper limb poststroke spasticity. In addition, these patients required caregiver assistance that is proportionally related to the severity of their disability in the hygiene and dressing domains.
Acknowledgments
We thank Ying Jiang and Paul M. Conforti of Allergan Inc for verifying all statistical analyses conducted for this project.
References
- Heart disease and stroke statistics—2009 update: A report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee . Circulation . 2009;119:480–486
- . Spasticity after stroke: Its occurrence and association with motor impairments and activity limitations . Stroke . 2004;35:134–139
- . Prevalence of spasticity post stroke . Clin Rehabil . 2002;16:515–522
- . Spasticity in adults: Management using botulinum toxin . http://www rcplondon ac uk/pubs/contents/6988a14a-1179-4071-8f56-dc2a865f0a43 pdf 2009; Accessed August 5, 2010
- . Assessment of post-stroke quality of life in cost-effectiveness studies: The usefulness of the Barthel Index and the EuroQoL-5D . Qual Life Res . 2004;13:427–433
- . Activity, participation, and quality of life 6 months poststroke . Arch Phys Med Rehabil . 2002;83:1035–1042
- Quality of life after stroke: The North East Melbourne Stroke Incidence Study (NEMESIS) . Stroke . 2004;35:2340–2345
- Repeated treatments with botulinum toxin type a produce sustained decreases in the limitations associated with focal upper-limb poststroke spasticity for caregivers and patients . Arch Phys Med Rehabil . 2008;89:799–806
- . Effect of upper limb botulinum toxin injections on impairment, activity, participation, and quality of life among stroke patients . Stroke . 2009;40:2589–2591
- . Modeling valuations for EuroQol health states . Med Care . 1997;35:1095–1108
- . A stroke-adapted 30-item version of the Sickness Impact Profile to assess quality of life (SA-SIP30) . Stroke . 1997;28:2155–2161
- . Quality of life 4 years after stroke . Stroke . 1988;19:1101–1107
- . Quality of life after stroke . Stroke . 1996;27:1467–1472
- . Evaluation of measures used to assess quality of life after stroke . Stroke . 2000;31:2004–2010
- . Guidance for industry, patient-reported outcome measures: Use in medical product development to support labeling claims . http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM193282.pdf Accessed April, 1, 2011
- Inter- and intrarater reliability of the Ashworth Scale and the Disability Assessment Scale in patients with upper-limb poststroke spasticity . Arch Phys Med Rehabil . 2002;83:1349–1354
- . EQ-5D User Guide: basic information on how to use EQ-5D . http://www.euroqol.org/fileadmin/user_upload/Documenten/PDF/User_Guide_v2_March_2009.pdf Accessed October 21, 2011
- . US valuation of the EQ-5D health states: Development and testing of the D1 valuation model . Med Care . 2005;43:203–220
- . Patient-reported health status in coronary heart disease in the United States: Age, sex, racial, and ethnic differences . Circulation . 2008;118:491–497
- . Estimation of minimally important differences in EQ-5D utility and VAS scores in cancer . Health Qual Life Outcomes . 2007;5:70
- . The validity and responsiveness of three quality of life measures in the assessment of psoriasis patients: Results of a phase II study . Health Qual Life Outcomes . 2006;4:71
- . The utility of health states after stroke: A systematic review of the literature . Stroke . 2001;32:1425–1429
- . A national catalog of preference-based scores for chronic conditions in the United States . Med Care . 2005;43:736–749
- ⁎ Disclosure: 2B, Allergan
- † Disclosure: 2B, Allergan, Ipsen, Merz; 3A; 7B, research support from Allergan, Ipsen, Merz, and NINDS with funds paid to the academic institution (IU or Wake)
- ‡ Disclosure: 1, employed by Allergan
- § Disclosure: 1B, employed by Allergan, stock options
- ¶ Disclosure: 1B, employed by Allergan
- ∥ Disclosure: 1B, employed by Allergan, stock options
- # Disclosure: 2A, Allergan, Merz, Ipsen; 7A, grant support from Allergan
This CME activity is designated for 1.0 AMA PRA Category 1 Credit™ and can be completed online at me.aapmr.org. Log on to www.me.aapmr.org, go to Lifelong Learning (CME) and select Journal-based CME from the drop down menu. This activity is FREE to AAPM&R members and $25 for non-members.
Peer reviewers and all others who control content have no relevant financial relationships to disclose.
Disclosure Key can be found on the Table of Contents and at www.pmrjournal.org
PII: S1934-1482(11)01285-8
doi:10.1016/j.pmrj.2011.10.001
© 2012 American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved.