Persisting Consequences of Stroke Measured by the Stroke Impact Scale
Background and Purpose— The purpose of this study was to compare disability and quality of life as measured by the Stroke Impact Scale (SIS) of stroke patients deemed recovered (Barthel Index ≥95) with 2 stroke-free populations of community-dwelling elderly.
Methods— Eighty-one stroke patients who participated in the Kansas City Stroke Registry and achieved a Barthel Index of ≥95 at 3 months after stroke and 246 stroke-free subjects enrolled in the Community Elders Study were enrolled in this study. The Community Elders Study group was further divided into 2 groups, those recruited from the Department of Veterans Affairs Health System (VA) and a those from a local health maintenance organization (HMO). Stroke patients were administered the SIS ≈90 days after stroke, and the stroke-free community dwellers were administered a version of the SIS adapted for nonstroke subjects, the Health Impact Scale (HIS). A general linear model was used to examine differences in health outcomes measured by the SIS or HIS between the KCSR stroke patients and VA and HMO community-dwelling elders after controlling for medical comorbidities and demographics.
Results— Kansas City Stroke Registry participants were significantly older than the community study groups (P=0.0052). Selected medical conditions were similar among the 3 study groups. Old age and a history of diabetes mellitus were more likely to be associated with more deficits and poor quality of life. In stroke patients deemed recovered, stroke still affected hand function, activities and independent activities of daily living, participation, and overall physical function compared with the stroke-free community dwellers in the HMO health system even after adjustment for age and diabetes status. Stroke-free community dwellers in the VA health system also had worse social participation than the stroke-free community dwellers in the HMO health system.
Conclusions— Research and clinicians have consistently underestimated the impact of stroke with the Barthel Index. This has major implications for the design of therapeutic trial designs and adequate assessments of social and economic sequelae of stroke.
Stroke is often associated with distressing and disabling sequelae. However, among those who are deemed recovered by traditional standards, the long-term disability resulting from stroke remains misunderstood and underestimated. Previous studies found that 40% to 50% of patients achieved 100% on the Barthel Index (BI) by 6 months.1–3⇓⇓ A BI of ≥95 has traditionally been used to define minimal or no disability. In fact, a BI of ≥95 has been chosen to define favorable outcome in a number of acute stroke trials.4,5⇓ Yet little or no information has been available to characterize the well-being of these individuals who were perceived to be functionally independent. Investigation in this regard is particularly important because the traditionally used BI instrument has suffered from ceiling effects and thus not captured higher levels of physical function or diminished quality of life or has measured only the physical aspect of stroke.6 Stroke affects not only physical function but also emotion, memory and thinking, communication, and quality of life.
The Stroke Impact Scale (SIS) has been developed to be a more comprehensive measure of health outcomes for stroke populations.7–9⇓⇓ The SIS incorporates meaningful dimensions of function and health-related quality of life into 1 self-report questionnaire. The SIS version 3 includes 59 items and assesses 8 domains [strength, hand function, activities and independent activities of daily living (ADL and IADL), mobility, communication, emotion, memory and thinking, and participation/role function]. Duncan et al7,9⇓ have shown the SIS to have less ceiling and floor effects and to be valid, reliable, and sensitive to change compared with other commonly used measures such as the BI and Short Form-36 in stroke populations.
The aim of the present study was to compare the disability and quality of life as measured by the SIS of stroke patients deemed recovered by the traditional method of BI score ≥95 to 2 stroke-free populations of community-dwelling elderly.
Study subjects for the present investigation included stroke patients participating in the Kansas City Stroke Registry (KCSR), a core program in the Kansas Claude D. Pepper Older Americans Independence Center, and subjects participating in the Community Elders Study to evaluate the benefit of physical performance measures in predicting older adults’ future healthcare use. As expected, the inclusion and exclusion criteria of these 2 studies were different and therefore are discussed in detail below. Both studies were approved by the university’s Human Subjects Committee. For KCSR patients to be included in the present investigation, subjects must have completed a 90-day poststroke assessment and achieved a BI score of ≥95. Case ascertainment for the KCSR started in January 1998. Recruitment and the follow-up effort are still ongoing. The eligible study participants were recruited from any of 17 participating hospitals in the Greater Kansas City area. Eligible stroke patients were identified by (1) review of daily admission records; (2) referrals from physicians, clinical nurse specialists, or therapists on medical, neurology, and rehabilitation units; and (3) review of discharge codes.
To be accepted into the KCSR, the subject had to have a confirmed eligible stroke as defined by World Health Organization (WHO) criteria as “rapid onset and of vascular origin reflecting a focal disturbance of cerebral function, excluding isolated impairments of higher function and persisting longer than 24 hours.”10 Stroke was confirmed by clinical assessment and/or a CT or MRI scan. Trained nurses and physical therapists reviewed medical records and interviewed both patients and physicians to determine whether the patient was eligible and consented to enrollment. Subjects were excluded if they (1) were <50 years of age; (2) had stroke onset >28 days previously; (3) experienced stroke as a result of subarachnoid hemorrhage; (4) had New York Heart Association class IV heart failure (ie, patients with cardiac disease resulting in marked limitation to carry on any physical activity without discomfort); (5) were expected to live <1 year; (6) were unable to follow a 2-step command; (7) had a major limb amputated; (8) lived in a nursing home before stroke; (9) were unable to take care of their own affairs before stroke; (10) were lethargic, obtunded, or comatose; and (11) lived >50 miles from the participating hospital. For the present investigation, only subjects ≥65 years of age were included in the analysis.
The patients were evaluated with a variety of standardized assessments at enrollment and followed up at 3 months after stroke onset by a research staff member at home or at a chronic care facility. Baseline assessments included demographics, medical history, stroke characteristics, stroke severity by use of the Orpington Prognostic Scale11 and the National Institute of Health Stroke Scale,12 and BI.13 Approximately 90 days after stroke, all patients were assessed again, including the SIS and BI. Administration of the SIS is discussed in detail below.
Study participants in the Community Elders Study are community-dwelling elders who participated in the study to evaluate predictive ability of physical performance measures. These participants were recruited from a Veteran’s Affairs (VA) network site and a large primary care clinic sponsored by a Medicare health management organization (HMO). Eligibility for this study included age of ≥65 years, community dwelling within 20 miles, and enrollment in the healthcare system for at least 1 year. Subjects were excluded if scores on the Mini-Mental Status Examination14 were <16 or if the persons were not able to walk at least 4 m. Two hundred forty-six subjects who were enrolled in this study, had no history of stroke, and had completed a Health Impact Scale (HIS) questionnaire 3 years after initial enrollment in the study were included in the present analysis. The HIS questionnaire was designed to parallel the SIS and is discussed below. The Community Elders Study subjects were further divided into 2 groups: VA (n=66) and HMO (n=180). The latter group will serve as the reference group.
For the KCSR group, the SIS was administered to stroke patients by a research assistant at home or at a chronic facility. The same SIS questionnaire that was administered to KCSR patients was administered to subjects in the Community Elders Study after removing the phase “since your stroke” from the questionnaire. This questionnaire was renamed the HIS and administered to the Community Elders Study group by a research assistant as well. The SIS and HIS are basically the same instrument and have 8 domains (strength, hand function, ADL/IADL, mobility, communication, emotion, memory and thinking, and participation/role function). The 4 physical function domains (strength, hand function, ADL/IADL, and mobility) were further collapsed and reduced to be a physical function subscale of the SIS or HIS. This composite domain of physical function included 16 items and was named the SIS 16. All domain scores ranged from 0 to 100, with a 100 being the best.
Methods of assessing comorbid conditions for the KCSR and Community Elders Study were slightly different. For the KCSR group, medical records were reviewed for comorbid conditions as documented and not documented. For the Community Elders Study group, self-report of comorbid conditions was obtained by interviewing study subjects. The subjects were asked, for example, “Has your doctor ever told you that you have or have had a heart attack?”
Statistical analyses were carried out with SAS software. A few patients had missing data on some of the variables, so some analyses were performed with a slightly lower number of patients. Descriptive statistics were carried out to characterize the sample. The effect of selected medical conditions on SIS or HIS assessment was also evaluated. Contingency tables and t tests were applied when appropriate. General linear model with the PROC GLM procedure was used to examine differences in health outcomes measured by the SIS or HIS between the KCSR stroke patients, VA community-dwelling elders, and HMO community-dwelling elders after controlling for demographics and medical conditions that were found to affect SIS or HIS assessment.
The sample included 327 individuals (81 from KCSR, 66 community dwellers from the VA health system, and 180 community dwellers from the HMO health system). The mean ages were 76.0±6.56, 73.8±5.02, and 73.6±5.18 years, respectively (P=0.0052; Table 1). As expected, community elders from the VA system were all male, and the KCSR study had more men than did the community-dwelling HMO group. There were more whites in both the KCSR and Community Elders Study HMO groups than in the VA group (P<0.0001). Diabetes mellitus (DM) was more commonly reported in the KCSR stroke group than in the 2 stroke-free community-dwelling groups, although the difference was not statistically significant (Table 1). Other medical conditions were similar among the 3 groups (Table 1). Of the 81 KCSR patients with stroke, 24 were right hemispheric, 34 were left hemispheric, 1 was bilateral, 5 were cerebellar, and 17 were brainstem.
Univariately, KCSR patients had lower mean scores of mobility, hand function, ADL/IADL, SIS 16, and participation compared with the VA and HMO participants (all P<0.05). No statistically significant differences were observed among the 3 groups in strength, emotion, memory and thinking, and communication (Table 2). An inverse relationship between age and each of the 8 SIS/HIS domains was observed (all P<0.05). Health status was also altered by DM. Subjects with DM had significantly lower mean scores of ADL/IADL, mobility, hand function, and participation than those without DM (all P<0.05; Table 3). The SIS 16 score, which includes items from domains of ADL/IADL, mobility, hand function, and strength, was lower in subjects with DM than in subjects without DM. Other selected medical conditions were not significantly associated with higher or lower SIS/HIS domains scores.
To evaluate persisting consequences of stroke, all 8 SIS domains were analyzed individually for a comparison among the 3 groups after adjustment for age, sex, and DM. In highly recovered stroke patients (BI score ≥95), stroke status affected hand function, ADL/IADL, physical function, and participation. However, stroke status did not affect strength, mobility, emotion, memory and thinking, and communication. After adjustment for differences in age, sex, and status of DM, significantly lower mean scores of hand function, ADL/IADL, and SIS 16 were observed for the well-recovered KCSR stroke patients compared with those observed in community-dwelling HMO elders (Table 4). For example, mean hand function score was 9.0 points lower in the KCSR stroke group and 2.5 points lower in the community-dwelling VA elders compared with community-dwelling HMO elders. Mean score for participation was 12.8 points lower in the KCSR stroke group compared with the stroke-free community-dwelling HMO elders (Table 4). Mean scores of strength, emotion, memory and thinking, and communication were similar among the 3 study groups. The multifactorial effects of age, sex, DM status, and study groups on selected SIS domains are demonstrated separately as Tables 5 through 7⇓⇓ for hand function, ADL/IADL, and social participation, respectively. As expected, both age and DM status had a negative impact on health status.
Highly recovered stroke patients continued to have significant residual disability in hand function, ADL/IADL, and the combined physical function (SIS 16) compared with stroke-free community-dwelling elders even after controlling for age and comorbid conditions like DM (Tables 4 through 6⇑⇑). As shown in Tables 4 and 5⇑, mean hand function score was 9.0 points lower in the KCSR stroke group compared with community-dwelling HMO elders. Previous studies have shown that up to 85% of stroke patients have an initial deficit in the arm.15–19⇓⇓⇓⇓ Functional recovery of the arm includes grasping, holding, and manipulating objects. This recovery process requires the recruitment and complex integration of muscle activity from shoulder to fingers. Subsequently, these problems remain in 55% to 75% of patients even 3 to 6 months later. The hemiplegic upper limb is affected in many stroke patients and has not been properly evaluated with traditional measurements. The SIS captures persisting difficulties in hand function, ADL/IADL, and overall physical function (SIS 16) in stroke patients who have traditionally been considered functionally independent. The patients’ ability to look after themselves and live independently is one of many areas that clinicians should comprehensively assess after stroke. Traditional measures such as the BI have consistently underestimated limitations in these areas.
Many other important aspects, including social functioning, from the patients’ point of view are their ultimate level of everyday performance and functioning.20–23⇓⇓⇓ Unfortunately, the BI and other outcome measures such as the modified Rankin Scale give little or no information. In addition to persisting difficulties in physical functioning, stroke patients who were perceived functionally independent at 3 months after stroke still experienced social isolation (Table 3). Difficulties in resuming patients’ normal level of physical functioning resulted in poor quality of life. Less active social participation for stroke patients compared with community control subjects was also reported by investigators using the Framingham study cohort.24 The SIS is a newly developed, comprehensive measure of health impact after stroke and is able to capture the impact of stroke across multiple dimensions from impairment to disability and participation (handicap). This instrument has undergone evaluation in stroke populations and has been shown to have good validity, reliability, and responsiveness.7,8⇓
Traditionally, in therapeutic trials, individuals with BI scores ≥95 were considered recovered.4,5,13,25⇓⇓⇓ Yet as shown in our study, stroke patients who have achieved a BI score ≥95 continued to have residual disability and impaired quality of life. Because of the ceiling effect inherent in the BI, disabilities and diminishing quality of life were not captured by the BI. The SIS, on the other hand, is able to show differences in fine motor function (hand function) and social participation between stroke-free community-dwelling elders and the stroke cohort that has been regarded as functionally independent. An instrument such as the SIS that is able to capture ADL, IADL, mental health, dysphasia, mobility and walking, and social functioning is recommended for assessment of the recovery process after stroke. With these measures, areas that require long-term disability rehabilitation can be identified and patients’ quality of life can be maximized.
In summary, research and clinicians have underestimated the impact of stroke with traditional measurements. This has major implications for the design of therapeutic trials and adequate assessments of social and economic sequelae of stroke.
This study was funded by the American Heart Association, Merck Research Laboratories, and the Kansas Claude D. Pepper Older Americans Independence Center (NIA-P60 AG 14635-03). Participating facilities in the greater Kansas City area include Baptist Medical Center, Bethany Hospital, VA Medical Center at Kansas City, Independence Regional Health Center, John Knox Village Care Center, Liberty Hospital, Life Care Center of Grandview, Menorah Medical Center, Mid-American Rehabilitation Hospital, Overland Park Regional, Rehabilitation Institute, Research Medical Center, Shawnee Mission Medical Center, St Luke’s Medical Center, St Joseph Health Center, Trinity Lutheran Hospital, and University of Kansas Medical Center.
- Received October 1, 1001.
- Revision received March 14, 2002.
- Accepted March 15, 2002.
- ↵Langton Hewer R. Rehabilitation after stroke. Q J Med. 1990; 76: 659–674.
- ↵Wade DT, Langton Hewer RL. Functional abilities after stroke: measurement, natural history and prognosis. J Neurol Neurosurg Psychiatry. 1987; 50: 177–182.
- ↵Hacke W, Kaste M, Fiechi C, von Kummer R, Davalos A, Meier D, Larrue V, Bluhmki E, Davis S, Donnan G, Schneider D, Diez-Tejedor E, Trouillas P, for the Second European-Australasian Acute Stroke Study Investigators. Randomised double-blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II). Lancet. 1998; 352: 1245–1251.
- ↵Duncan PW, Samsa GP, Weinberger M, Goldstein LB, Bonito A, Witter DM, Enarson C, Matchar D. Health status of individuals with mild stroke. Stroke. 1997; 28: 740–745.
- ↵Duncan PW, Lai SM, Bode RK, Perera S, for the GAIN Americas Investigators. Rasch analysis of a new stroke specific outcome scale: the Stroke Impact Scale. Arch Phys Med Rehabil. In press, 2002.
- ↵Duncan PW, Lai SM, Bode RK, Perera S, DeRosa JT, for the GAIN Americas Investigators. Development of the SIS-16 and comparison with the Barthel Index. Neurology. 2002.
- ↵Proposal for the Multinational Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA Project). Geneva, Switzerland: WHO; 1983.WHO/MNC/82.1. Revision 1.
- ↵Lai SM, Duncan PW, Keighley J. Prediction of functional outcome after stroke: comparison of the Orpington Prognostic Scale and the NIH Stroke Scale. Stroke. 1998; 29: 1838–1842.
- ↵Brott T, Adams H, Olinger C. Measurements of acute cerebral infarction: a clinical examination scale. Stroke. 1989; 20: 864–870.
- ↵Feys HM, De Weerdt WJ, Selz BE, Cox Steck GA, Spichiger R, Vereeck LE, Putman KD, Van Hoydonck GA. Effect of a therapeutic intervention for the hemiplegic upper limb in the acute phase after stroke: a single-blind, randomized, controlled multicenter trial. Stroke. 1998; 29: 785–792.
- ↵Wade DT, Langton-Hewer R, Wood VA, Skilbeck CE, Ismail HM. The hemiplegic arm after stroke: measurement and recovery. J Neurol Neurosurg Psychiatry. 1983; 46: 521–524.
- ↵Olsen TS. Arm and leg paresis as outcome predictors in stroke rehabilitation. Stroke. 1990; 21: 247–251.
- ↵Greveson GC, Gray CS, French JM, James OFW. Long-term outcome for patients and carers following hospital admission for stroke. Age Ageing. 1991; 20: 337–344.
- ↵Silliman RA, Wagner EH, Fletcher RH. The social and functional consequences of stroke for elderly patients. Stroke. 1987; 18: 200–203.
- ↵Brocklehurst JC, Morris P, Andrews K, Richard B, Laycock P. Social effects of stroke. Soc Sci Med. 1981; 15A: 35–39.
- ↵Gresham GE, Kelly-Hayes M, Wolf PA, Beiser AS, Kase CS, D’Agostino RB. Survival and functional status 20 or more years after first stroke: the Framingham study. Stroke. 1998; 29: 793–797.
- ↵Sulter G, Steen C, De Keyser J. Use of the Barthel Index and modified Rankin Scale in acute stroke trials. Stroke. 1999; 30: 1538–1541.