Functional Status and Patient-Reported Outcome 10 Years After Stroke The Lund Stroke Register Ann-Cathrin Jönsson, RN, PhD; Hossein Delavaran, MD; Susanne Iwarsson, OT, PhD; Agneta Ståhl, PhD, Eng; Bo Norrving, MD, PhD; Arne Lindgren, MD, PhD Background and Purpose—Long-term studies of outcome after stroke are scarce. Our aim was to study functional status and patient-reported outcome 10 years after a first-ever stroke. Methods—Ten-year follow-up was conducted among the survivors from a population-based group of 416 patients included in the Lund Stroke Register, Sweden, between March 1, 2001, and February 28, 2002. The Barthel index was used to assess the functional status and the modified Rankin Scale to assess the degree of disability. The EQ-5D scale was used for survivors’ self-reports about health outcome and the specific Short-Form 36 (SF-36) question for rating their overall health. The patients also reported their frequency of physical activity. Results—Among 145 survivors 10 years after stroke (median age, 78 years), 59% were men, 90% lived in their ordinary housing, 73% were assessed as independent, and 71% had no or slight disability. The need of assistance with mobility and self-care was reported by 14% and with usual activities by 22%. Moderate pain was reported by 39%, and 4% had a high degree of pain. Moderate anxiety/depression was reported by 28% and high degree only by 1%. Overall health status was reported in positive terms by more than two thirds of the survivors. Almost half the cohort reported the same frequency of physical activity (≥4× weekly) as before stroke onset. Conclusions—This study indicates that 10-year stroke survivors in Sweden are mostly independent in daily activities and report good overall health and frequent physical activity, although half of them are ≥78 years. (Stroke. 2014;45:1784-1790.) Key Words: long-term survivors ◼ patient outcome assessment ◼ physical activity ◼ recovery of function ◼ self report ◼ stroke
S
troke is one of the most dominant causes of morbidity and mortality worldwide and poses a major global healthcare challenge.1 Stroke survivors are often affected by neurological impairments causing functional disability and need of assistance, which may lead to institutionalization in nursing or residential care setting.2 Quality of life has been found to decline after stroke in relation to age, mood, stroke severity, urinary incontinence, functional status, and cognition.3 Reliable up-todate knowledge on long-term outcome after stroke is of importance for the optimization of healthcare planning and to obtain more knowledge on how to further reduce the burden of stroke. However, there is scarcity of population-based long-term studies of stroke outcome ≥10 years after stroke onset. Most studies available were published several years ago and are limited by focusing on a few measures such as survival,4 survival and health-related quality of life,5 risk factors related to stroke and survival,6–8 recurrent stroke and disability,9 and effects of stroke unit treatment.10 One recent report on longterm outcome after stroke was limited by a substantial loss to
follow-up.11 Another recent study excluded patients >49 years of age in a 12-year follow-up after stroke covering mortality, functional status, psychosocial factors, and treatment goals.12 This is, to the best of our knowledge, the first study describing outcome 10 years after stroke onset in a population-based cohort of patients with first-ever stroke, reporting on objectively assessed physical function and comparing these findings with patient-reported outcome measures and also including age and sex comparisons.
Methods Sample Patients with a first-ever stroke were consecutively included in the Lund Stroke Register between March 1, 2001, and February 28, 2002. The Lund Stroke Register is a population-based stroke register that covers the catchment area of Skåne University Hospital in Lund, Sweden, including 8 municipalities with a total of 234 505 inhabitants (as of December 31, 2001). Several overlapping sources were used to detect patients with stroke. Methods for the detection and registration
Received February 25, 2014; final revision received March 31, 2014; accepted April 7, 2014. From the Department of Health Sciences (A.-C.J., S.I.), Department of Clinical Sciences, Lund, Neurology (H.D., B.N., A.L.), and Department of Technology and Society (A.S.), Lund University, Lund, Sweden; and Department of Neurology, Skåne University Hospital, Lund, Sweden (H.D., B.N., A.L.). Guest Editor for this article was Markku Kaste, MD, PhD. Correspondence to Ann-Cathrin Jönsson, RN, PhD, Department of Health Sciences, Lund University, Box 157, 221 00 Lund, Sweden. E-mail [email protected] © 2014 American Heart Association, Inc. Stroke is available at http://stroke.ahajournals.org
DOI: 10.1161/STROKEAHA.114.005164
1784
Jönsson et al Outcome 10 Years After Stroke 1785 of all patients with first-ever stroke during the defined period have been previously described.13,14 Stroke was defined according to the World Health Organization criteria.15
Follow-Up Procedures The registered patients have been prospectively followed up at 4 and 16 months after stroke onset. Different aspects of stroke outcome obtained from these initial follow-ups have been reported, including self-reported health-related quality of life,13 pain,16 shoulder pain,17 and weight loss.18 For the present study, a follow-up of survivors was organized 10 years after stroke onset in a similar way as in the previous follow-ups. The sample flow from baseline to 10-year follow-up is illustrated in the Figure. All survivors from the previous follow-up at 16 months after stroke participated at the follow-up after 10 years. Seventy-three percent of patients were followed up at the outpatient clinic of the Department of Neurology; 8% in skilled nursing facilities, in some cases with assistance of a nurse who knew the patient well; 6% while staying in hospital (ie, if the patient was hospitalized at the time of follow-up); and 12% were assessed by home visits. Three patients had moved ≈250 km from the study area and were followed up by telephone and in cooperation with physicians or nurses in primary health care in the districts where they lived at the follow-up.
Outcome Assessments For the present study, the following assessments were performed: 1. Objective evaluation of outcome: (1) The Barthel index (score 0–100) to assess functional status,19 divided into 3 grades: independence (score 95–100), minor dependence (score 60–90), and major dependence (score 0–55)20; (2) modified Rankin Scale (mRS; score 0–5) to assess the degree of disability21; (3) housing situation (ordinary housing, ordinary housing with home care, skilled nursing facilities).
2. Patient-reported outcome measures: (1) EQ-5D, a standardized instrument to measure health outcome, was used to obtain selfreports on mobility, self-care, usual activities (eg, work, study, housework, family or leisure activities), pain/discomfort, and anxiety/depression22; (2) the first question of Short-Form 36 (SF-36) to assess general health status expressed as follows: In general, would you say your health is: excellent, very good, good, fairly good, or poor?23; (3) questions on physical activity at baseline and follow-up were asked as follows: How often do you walk, bike, run, or practice physical activities in other ways? The respondents could use the following response alternatives: never, less than once weekly, once weekly, 2 to 3× weekly, or ≥4× weekly. Individuals who were physically active were also asked how long they were physically active each day with the following response alternatives: never, 1 hour.
Ethical Considerations This study was approved by the Regional Ethical Review Board in Lund, Sweden, Registration No. 2011/278. At the follow-ups, patients with newly detected health problems considered to need medical intervention were referred to their respective general practitioners. If a more urgent intervention was needed, a neurologist at the university hospital was consulted.
Statistics Data analysis was performed using the SPSS statistical software package, version 21. Sex and age differences were analyzed using the Mann–Whitney U test, and for comparisons of the same individuals between baseline and 10 years later, the Wilcoxon signed-rank test was used. Spearman ρ correlation coefficient was used to analyze possible relationships between outcome variables and age. P values ≤0.05 were considered statistically significant.
Results 416 registered 6 no consent 12 dropouts
68 deceased
330 Follow-up 4 months after stroke 21 deceased
309 Follow-up 16 months after stroke
Ten years after stroke, 145 (35%) of the initially registered 416 patients with a first-ever stroke had survived and were thus included (Figure). As presented in Table 1, 59% of survivors were men and 41% were women (P=0.03), compared with 56% versus 44% among the 416 patients at stroke onset (P=0.01). The median age at 10-year follow-up was 78.1 (range, 28–97) years. Nearly one quarter (23%) of the survivors were ≥85 years of age (30.5% of women and 17% of men in the total group of 145 patients). Among the 145 survivors who were followed up, 126 (86.9%) had cerebral infarction at baseline, 10 (6.9%) had intracerebral hemorrhage, 8 (5.5%) Table 1. Baseline Characteristics for All 416 Patients Included in the Lund Stroke Register and for the 145 Who Had Survived at 10 Years After Stroke Onset Total number of patients, n Men/women,%
164 deceased
All Patients Registered
10-y Survivors
416
145 (35% of 416)
56%/44%
59%/41%
Age at time of index stroke,* 74.3/76.5 (17.5–102.3) 66.2/67.9 (17.5–87.1) y, mean/median (range) Subtype of index stroke, n (%)
145 survivors ten years after stroke Figure. Patient flow chart describing the 10-year follow-up of the first year cohort in the Lund Stroke Register.
Cerebral infarction
354 (85.1%)
126 (86.9%)
Intracerebral hemorrhage
40 (9.6%)
10 (6.9%)
Subarachnoid hemorrhage
13 (3.1%)
8 (5.5%)
9 (2.2%)
1 (0.7%)
Undefined
*Index stroke indicates the event of first-ever stroke.
1786 Stroke June 2014 Table 2. Demographics, Living Conditions, Functional Status, Disability, and Health Status at 10 Years After a First-Ever Stroke Participants Age, y, mean/median (range)
All
Men
145
86 (59%)
76.4/78.1(28–97)
75.2/77.7 (28–97)
Women 59 (41%) 78.1/79 (55–96)
S
troke is one of the most dominant causes of morbidity and mortality worldwide and poses a major global healthcare challenge.1 Stroke survivors are often affected by neurological impairments causing functional disability and need of assistance, which may lead to institutionalization in nursing or residential care setting.2 Quality of life has been found to decline after stroke in relation to age, mood, stroke severity, urinary incontinence, functional status, and cognition.3 Reliable up-todate knowledge on long-term outcome after stroke is of importance for the optimization of healthcare planning and to obtain more knowledge on how to further reduce the burden of stroke. However, there is scarcity of population-based long-term studies of stroke outcome ≥10 years after stroke onset. Most studies available were published several years ago and are limited by focusing on a few measures such as survival,4 survival and health-related quality of life,5 risk factors related to stroke and survival,6–8 recurrent stroke and disability,9 and effects of stroke unit treatment.10 One recent report on longterm outcome after stroke was limited by a substantial loss to
follow-up.11 Another recent study excluded patients >49 years of age in a 12-year follow-up after stroke covering mortality, functional status, psychosocial factors, and treatment goals.12 This is, to the best of our knowledge, the first study describing outcome 10 years after stroke onset in a population-based cohort of patients with first-ever stroke, reporting on objectively assessed physical function and comparing these findings with patient-reported outcome measures and also including age and sex comparisons.
Methods Sample Patients with a first-ever stroke were consecutively included in the Lund Stroke Register between March 1, 2001, and February 28, 2002. The Lund Stroke Register is a population-based stroke register that covers the catchment area of Skåne University Hospital in Lund, Sweden, including 8 municipalities with a total of 234 505 inhabitants (as of December 31, 2001). Several overlapping sources were used to detect patients with stroke. Methods for the detection and registration
Received February 25, 2014; final revision received March 31, 2014; accepted April 7, 2014. From the Department of Health Sciences (A.-C.J., S.I.), Department of Clinical Sciences, Lund, Neurology (H.D., B.N., A.L.), and Department of Technology and Society (A.S.), Lund University, Lund, Sweden; and Department of Neurology, Skåne University Hospital, Lund, Sweden (H.D., B.N., A.L.). Guest Editor for this article was Markku Kaste, MD, PhD. Correspondence to Ann-Cathrin Jönsson, RN, PhD, Department of Health Sciences, Lund University, Box 157, 221 00 Lund, Sweden. E-mail [email protected] © 2014 American Heart Association, Inc. Stroke is available at http://stroke.ahajournals.org
DOI: 10.1161/STROKEAHA.114.005164
1784
Jönsson et al Outcome 10 Years After Stroke 1785 of all patients with first-ever stroke during the defined period have been previously described.13,14 Stroke was defined according to the World Health Organization criteria.15
Follow-Up Procedures The registered patients have been prospectively followed up at 4 and 16 months after stroke onset. Different aspects of stroke outcome obtained from these initial follow-ups have been reported, including self-reported health-related quality of life,13 pain,16 shoulder pain,17 and weight loss.18 For the present study, a follow-up of survivors was organized 10 years after stroke onset in a similar way as in the previous follow-ups. The sample flow from baseline to 10-year follow-up is illustrated in the Figure. All survivors from the previous follow-up at 16 months after stroke participated at the follow-up after 10 years. Seventy-three percent of patients were followed up at the outpatient clinic of the Department of Neurology; 8% in skilled nursing facilities, in some cases with assistance of a nurse who knew the patient well; 6% while staying in hospital (ie, if the patient was hospitalized at the time of follow-up); and 12% were assessed by home visits. Three patients had moved ≈250 km from the study area and were followed up by telephone and in cooperation with physicians or nurses in primary health care in the districts where they lived at the follow-up.
Outcome Assessments For the present study, the following assessments were performed: 1. Objective evaluation of outcome: (1) The Barthel index (score 0–100) to assess functional status,19 divided into 3 grades: independence (score 95–100), minor dependence (score 60–90), and major dependence (score 0–55)20; (2) modified Rankin Scale (mRS; score 0–5) to assess the degree of disability21; (3) housing situation (ordinary housing, ordinary housing with home care, skilled nursing facilities).
2. Patient-reported outcome measures: (1) EQ-5D, a standardized instrument to measure health outcome, was used to obtain selfreports on mobility, self-care, usual activities (eg, work, study, housework, family or leisure activities), pain/discomfort, and anxiety/depression22; (2) the first question of Short-Form 36 (SF-36) to assess general health status expressed as follows: In general, would you say your health is: excellent, very good, good, fairly good, or poor?23; (3) questions on physical activity at baseline and follow-up were asked as follows: How often do you walk, bike, run, or practice physical activities in other ways? The respondents could use the following response alternatives: never, less than once weekly, once weekly, 2 to 3× weekly, or ≥4× weekly. Individuals who were physically active were also asked how long they were physically active each day with the following response alternatives: never, 1 hour.
Ethical Considerations This study was approved by the Regional Ethical Review Board in Lund, Sweden, Registration No. 2011/278. At the follow-ups, patients with newly detected health problems considered to need medical intervention were referred to their respective general practitioners. If a more urgent intervention was needed, a neurologist at the university hospital was consulted.
Statistics Data analysis was performed using the SPSS statistical software package, version 21. Sex and age differences were analyzed using the Mann–Whitney U test, and for comparisons of the same individuals between baseline and 10 years later, the Wilcoxon signed-rank test was used. Spearman ρ correlation coefficient was used to analyze possible relationships between outcome variables and age. P values ≤0.05 were considered statistically significant.
Results 416 registered 6 no consent 12 dropouts
68 deceased
330 Follow-up 4 months after stroke 21 deceased
309 Follow-up 16 months after stroke
Ten years after stroke, 145 (35%) of the initially registered 416 patients with a first-ever stroke had survived and were thus included (Figure). As presented in Table 1, 59% of survivors were men and 41% were women (P=0.03), compared with 56% versus 44% among the 416 patients at stroke onset (P=0.01). The median age at 10-year follow-up was 78.1 (range, 28–97) years. Nearly one quarter (23%) of the survivors were ≥85 years of age (30.5% of women and 17% of men in the total group of 145 patients). Among the 145 survivors who were followed up, 126 (86.9%) had cerebral infarction at baseline, 10 (6.9%) had intracerebral hemorrhage, 8 (5.5%) Table 1. Baseline Characteristics for All 416 Patients Included in the Lund Stroke Register and for the 145 Who Had Survived at 10 Years After Stroke Onset Total number of patients, n Men/women,%
164 deceased
All Patients Registered
10-y Survivors
416
145 (35% of 416)
56%/44%
59%/41%
Age at time of index stroke,* 74.3/76.5 (17.5–102.3) 66.2/67.9 (17.5–87.1) y, mean/median (range) Subtype of index stroke, n (%)
145 survivors ten years after stroke Figure. Patient flow chart describing the 10-year follow-up of the first year cohort in the Lund Stroke Register.
Cerebral infarction
354 (85.1%)
126 (86.9%)
Intracerebral hemorrhage
40 (9.6%)
10 (6.9%)
Subarachnoid hemorrhage
13 (3.1%)
8 (5.5%)
9 (2.2%)
1 (0.7%)
Undefined
*Index stroke indicates the event of first-ever stroke.
1786 Stroke June 2014 Table 2. Demographics, Living Conditions, Functional Status, Disability, and Health Status at 10 Years After a First-Ever Stroke Participants Age, y, mean/median (range)
All
Men
145
86 (59%)
76.4/78.1(28–97)
75.2/77.7 (28–97)
Women 59 (41%) 78.1/79 (55–96)
