Published online before print November 14, 2002, doi: 10.1161/01.STR.0000040407.44712.C7
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2002;33:2888.)
© 2002 American Heart Association, Inc.
Original Contributions |
Brief Comprehensive Quality of Life Assessment After Stroke
The Assessment of Quality of Life Instrument in the North East Melbourne Stroke Incidence Study (NEMESIS)
From the National Stroke Research Institute and Department of Neurology, Austin and Repatriation Medical Centre (J.W.S., H.M.D., G.A.D., R.A.L.M., A.G.T.); Centre for Health Program Evaluation, School of Population Health (R.H.O.); Department of Medicine, University of Melbourne (R.H.O., H.M.D., G.A.D., R.A.L.M., A.G.T.); and Department of Epidemiology and Preventive Medicine, Monash Medical School, Alfred Hospital (A.G.T.), Melbourne, Australia.
Correspondence to Dr Jonathan Sturm, National Stroke Research Institute, Repatriation Campus, Austin and Repatriation Medical Centre, Banksia St, West Heidelberg Vic 3081, Australia. E-mail jsturm{at}austin.unimelb.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAppendixReferences |
|---|
Background and Purpose— Generic utility health-related quality of life instruments are useful in assessing stroke outcome because they facilitate a broader description of the disease and outcomes, allow comparisons between diseases, and can be used in cost-benefit analysis. The aim of this study was to validate the Assessment of Quality of Life (AQoL) instrument in a stroke population.
Methods— Ninety-three patients recruited from the community-based North East Melbourne Stroke Incidence Study between July 13, 1996, and April 30, 1997, were interviewed 3 months after stroke. Validity of the AQoL was assessed by examining associations between the AQoL and comparator instruments: the Medical Outcomes Short-Form Health Survey (SF-36); London Handicap Scale; Barthel Index; National Institutes of Health Stroke Scale; and Irritability, Depression, Anxiety scale. Sensitivity of the AQoL was assessed by comparing AQoL scores from groups of patients categorized by severity of impairment and disability and with total anterior circulation syndrome (TACS) versus non-TACS. Predictive validity was assessed by examining the association between 3-month AQoL scores and outcomes of death or institutionalization 12 months after stroke.
Results— Overall AQoL utility scores and individual dimension scores were most highly correlated with relevant scales on the comparator instruments. AQoL scores clearly differentiated between patients in categories of severity of impairment and disability and between patients with TACS and non-TACS. AQoL scores at 3 months after stroke predicted death and institutionalization at 12 months.
Conclusions— The AQoL demonstrated strong psychometric properties and appears to be a valid and sensitive measure of health-related QoL after stroke.
Key Words: cerebrovascular disorders • health status • outcome assessment • quality of life
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendixReferences |
|---|
Health-related quality of life (HRQoL) is an important index of outcome after stroke and is likely to be more relevant to the patient than impairment or disability.1–3 Measurement of HRQoL may be useful in natural history studies, in assessing the effectiveness of acute interventions, and in the evaluation of rehabilitation and community support programs. From a societal perspective, it is also important to measure the efficiency of interventions, including health gains per monetary value expended. Although there is debate about the relative merits of generic versus stroke-specific HRQoL measures,2,3 generic utility HRQoL instruments can directly measure health status, allow comparison across diseases, and facilitate economic evaluation.
The Assessment of Quality of Life (AQoL) instrument4,5 was designed (1) to ensure coverage of the full universe of HRQoL, (2) to meet standard psychometric requirements for reliable and valid measurement, (3) to be sensitive to a wide range of health states from very good to extremely poor health, and (4) to deliver psychometrically sound scores on the different dimensions of HRQoL. The AQoL has been successfully used in nonstroke populations.6–10 Other generic instruments have been used to collect HRQoL data in stroke survivors. These include the Medical Outcomes Short-Form Health Survey (SF-36),11 and the Euroqol.12 Both have been criticized for a lack of sensitivity.2 Similar to the Euroqol, the AQoL yields both a descriptive profile and a single value that quantifies overall HRQoL. The Euroqol is simple and slightly quicker to complete than the AQoL but appears to have poorer coverage of stroke-relevant HRQoL. The SF-36 takes longer to complete than the AQoL. Scores are provided from 8 subscales (or 2 summary subscales). The use of multiple subscales may be viewed as advantageous because of the diversity in the recovery process from stroke; however, it may also make the interpretation of overall outcome difficult because gains in 1 domain can be offset by losses in other domains. The AQoL appears to have advantages over existing generic measures of HRQoL and health status in that it offers brief assessment, better coverage of stroke-specific domains (eg, communication and vision), and 1 overall score facilitating analysis and can be used in economic analysis. However, the validity of the AQoL as a measure of HRQoL in stroke has not been established. The aim of this study was to validate the AQoL as a measure of HRQoL in a population of stroke survivors.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendixReferences |
|---|
The North East Melbourne Stroke Incidence Study
Stroke survivors were recruited from the North East Melbourne Stroke Incidence Study (NEMESIS), a community-based study of the incidence, risk factors, and outcomes of stroke in the northeastern suburbs of Melbourne, Australia. The methods have previously been described in detail.13 Briefly, NEMESIS was conducted in an 8–postal code region containing a population of
133 816 in northeast Melbourne between May 1, 1996, and April 30, 1997. The methodology used to obtain stroke cases was based on the recommendations for the conduct of "ideal" stroke incidence studies.14 All potential cases of stroke were reviewed by a panel of at least 2 neurologists and 1 epidemiologist before inclusion in the study. Cases meeting the World Health Organization (WHO) definition of stroke were registered.15 Pathological subtype of stroke (cerebral infarction [CI], intracerebral hemorrhage [ICH], and subarachnoid hemorrhage [SAH]) was determined on the basis of neuroimaging, lumbar puncture, or autopsy findings as described previously.16 Patients who had neither neuroimaging nor autopsy were classified as undetermined. Patients were further categorized by an expert panel blinded to brain imaging findings that used the Oxfordshire Community Stroke Project stroke classification17 using symptoms and signs at the time of maximal deficit. Patients with SAH were included in incidence and mortality counts but were not formally followed up in NEMESIS.
Follow-Up
Patients were administered the AQoL and comparator instruments by trained research nurses via face-to-face interview 3 months after stroke. If a patient was unable to complete the interview because of dementia or aphasia, a face-to-face proxy interview was sought from the best available informant, usually the next of kin or, if necessary, a professional caregiver.
Subjects
For the purposes of the present study, all patients who had a first-ever-in-a-lifetime or recurrent stroke (CI, ICH, or undetermined but not SAH) between July 13, 1996, and April 30, 1997, were eligible for inclusion. Two hundred sixty-one patients had a stroke (CI, ICH, or undetermined) in this period. At 3 months after stroke, 72 (28%) were deceased. Thirty-three patients (13%) were notified to the NEMESIS study >3 months after stroke, and 63 (24%) refused to take part. Therefore, 93 patients (60% of those approached, 49% of those alive at 3 months) were enrolled in the study. The mean age was 72 years (range, 28 to 89 years), 45% were male, and 70% were born in Australia; for 83%, this was a first-ever-in-a-lifetime stroke. Of the total patients, 10% had total anterior circulation infarction, 37% had partial anterior circulation infarction, 19% had posterior circulation infarction, 26% had lacunar infarction, and 6% had ICH; in 2 patients, stroke type was undetermined. Patients included in this study did not differ significantly from those who refused to take part or were notified late to the NEMESIS study in terms of age, sex, stroke subtype, or prestroke disability. Those not included were significantly more likely to have been born outside Australia (P=0.05) and to have been admitted to a private hospital (P=0.02).
Instruments
The AQoL instrument is a generic HRQoL utility instrument that includes 5 dimensions of HRQoL: independent living, social relationships, physical senses, psychological well-being, and illness (reproduced in full in the Appendix). Scores from the first 4 dimensions only are used to calculate an overall utility score. Each individual scale is weighted to extend between 0.0 (death) and 1.0 (full health). However, the overall utility score range is from -0.04 (worst possible HRQoL state) to 0.00 (death-equivalent HRQoL state) to 1.00 (full HRQoL). The utilities were derived using time tradeoff methodology in a sample of the general population.4 The AQoL was designed to be administered by self-completion, through face-to-face interview, or via proxy. It takes 5 minutes to complete and has been well validated in the general population.4,8
The SF-3611 is a widely used and validated generic HRQoL instrument.18 It comprises 8 subscales: physical functioning, role physical, bodily pain, general health, vitality, social function, role emotional, and mental health. Each of the subscales is scored separately using item weighting and additive scaling. Summed data are transformed onto a 0-to-100-point scale. The 8 subscales can be combined into 2 summary health status measures, physical function and mental health. The acute (1-week time frame) version of the SF-36 was used.
Impairment was measured with the National Institutes of Health Stroke Scale (NIHSS)19 (range, 0 to 42). Patients were categorized into prespecified groups by severity of impairment with 5-point intervals. Categorization into 5-point intervals has been used in previous studies and represents a clinically relevant score threshold.20,21 Because of the small number of patients with very high impairment scores in our sample, we collapsed the patients scoring >10 into 1 category, ie, 0 to 5, 6 to 10, and >10.
Disability was assessed with the Barthel Index.22 Patients were scored according to current actual performance of activities of daily living. Severity of disability was determined by stratifying patients into prespecified categories: very severely (Barthel 0 to 4), severely (5 to 9), moderately (10 to 14), or mildly (15 to 19) disabled or nondisabled (20 of 20).23
Handicap was assessed with the London Handicap Scale (LHS).24,25 This provides a descriptive profile of disadvantages experienced in the domains of mobility, physical independence, occupation, orientation, social functioning, and economic self-sufficiency. A weighted total handicap score is produced, with scores ranging from 0 (maximum disadvantage) to 100 (no disadvantage).
The Irritability, Depression, Anxiety Scale (IDA)26 was used to assess mood. This validated 18-question instrument is used to assess inward and outward irritability, depression, and anxiety. For the purposes of this study, only the two 5-question subscales pertaining to depression and anxiety were used. For each of these subscales, a score ranging from 0 to 15 is produced, with higher scores reflecting greater mood impairment.
These questionnaires were chosen to enable comparison between AQoL scores and scores from established instruments that measure different levels of outcome according to the WHO International Classification of Impairments, Disabilities, and Handicaps (ICIDH).27
Statistical Analysis
The construct validity of the AQoL was assessed by examining the relationships between the AQoL (individual dimensions and total utility score) and the NIHSS, Barthel Index, LHS, SF-36 subscales, and the depression and anxiety subscales of the IDA at 3 months after stroke using Spearman’s rank correlation coefficient (R). For a sample size of 93 and the null hypothesis R=0, the critical values are P=0.05, R±0.204; P=0.01, R±0.267; and P=0.001, R±0.338.28 We hypothesized that the magnitude of correlation would be highest between scales measuring similar constructs (providing evidence of convergent validity) and the weakest correlations would be between scales measuring conceptually unrelated constructs (discriminant validity). For example, we expected that the independent living dimension of the AQoL would be more highly correlated with the Barthel Index than with measures of mental health and, conversely, the psychological well-being dimension to be more highly correlated with the IDA subscales than with the Barthel Index.
Criterion validity is assessed by comparing a measure with a gold standard. Because there is no gold standard for HRQoL, we compared the AQoL with instruments that measure different levels of outcome according to the WHO ICIDH.27 The AQoL was constructed with a primary emphasis on handicap but with supplementation from the concepts of impairment and disability.4 Given this framework for the construction of the AQoL, we hypothesized that the overall AQoL utility scores would be more strongly correlated with the LHS (handicap) than with the Barthel Index (disability) or NIHSS (impairment).
To assess the sensitivity of the AQoL (and to provide further evidence of construct validity), we performed a series of known-group validations. We compared mean AQoL utility scores (using 1-way analysis of variance [ANOVA]) between groups of patients categorized by severity of disability and severity of impairment, hypothesising that AQoL scores would be lower (worse HRQoL) in those patients with greater disability and impairment. The sensitivity of the AQoL was compared with the NIHSS, Barthel Index, LHS, SF-36, and depression and anxiety subscales of the IDA by examining scores obtained from patients with total anterior circulation syndrome (TACS) stroke and patients with non-TACS stroke. Relative efficiencies (RE), the ratio of the squares of the t statistic, were calculated for each instrument and compared with the RE of the AQoL. The RE is a commonly used statistic that allows a simple comparison of the relative abilities of different instruments to detect differences between groups.29,30 The RE was applied to a known-groups comparison in which patients categorized as TACS would be expected to have lower health status than patients with other subtypes of stroke at 3 months after stroke because they have significantly greater disability, handicap, and levels of dependency.17,31,32 Patients with definite ICH were excluded from the TACS/non-TACS analysis because these patients may be drowsy initially (and classified as TACS) but make a good recovery by 3 months.
The predictive validity of the AQoL was assessed by testing the hypothesis that the AQoL utility score at 3 months after stroke would predict the outcomes of death, institutionalization, or both 12 months after stroke. Logistic regression was used to produce models describing the relationship between 3-month AQoL scores and these outcomes. Discrimination of the predictive models was assessed by calculating the area under a receiver-operating characteristic curve (AUC) of sensitivity versus 1 minus specificity. An AUC of 1 implies perfect prediction, whereas an AUC of 0.5 implies the that predictions of that model are no better than chance. Multiple logistic regression was performed to determine whether AQoL score at 3 months remained an independent predictor of death and institutionalization at 12 months when adjusted for age, sex, and impairment score.
Analyses were performed with the SPSS package (version 10.0 for Windows).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendixReferences |
|---|
All 93 patients completed the AQoL. Fifteen subjects (16%) were assessed by means of a proxy response. The mean AQoL utility score was 0.404, the median was 0.365, and the SD was 0.333. The data were not significantly skewed, and scores were distributed evenly throughout the available range. Two patients (2%) scored -0.04 (lowest possible score), and only 1 patient (1%) scored 1.0 (highest possible score), demonstrating no substantial floor or ceiling effects. Ninety-two patients were assessed with the Barthel Index, LHS, and SF-36; 88 were assessed with the NIHSS; and 79 were assessed with the IDA.
The association between the AQoL and its individual dimensions and the other instruments is shown in Table 1. The magnitude of the correlations was greatest between scales with the most similar construct. The independent living dimension of the AQoL correlated most strongly with the Barthel Index (R=0.85), followed by the LHS (R=0.82), NIHSS (R=-0.74), and SF-36 physical functioning subscale (R=0.77). The social relationships dimension was most highly correlated with the LHS (R=0.56), anxiety (R=-0.49), and SF-36 social functioning (R=0.48) and mental health (R=0.50) subscales. The physical senses dimension correlated most strongly with the LHS (R=0.52), Barthel Index (R=0.45), and NIHSS (R=-0.40). The highest correlations for the psychological well-being dimension were with the depression (R=-0.53) and anxiety (R=-0.47) subscales and the SF-36 bodily pain (R=0.61) and mental health (R=0.41) subscales. The total AQoL score was most highly associated with the LHS (R=0.83) and Barthel Index (R=0.77).
|
The relationships between mean AQoL score and categories of disability (Figure 1) and impairment were in the expected direction. Highest HRQoL was observed when disability and impairment were least. The differences in total AQoL scores across categories of disability and impairment were significant (P<0.001, ANOVA).
|
The total AQoL and independent living dimension, SF-36 physical function subscale, LHS, Barthel Index, and NIHSS discriminated significantly between the health status of TACS and non-TACS patients in both parametric (t test) and nonparametric (permutations test33) analyses (Table 2).
|
The AQoL utility score (RE referenced to 1.0) discriminated between TACS and non-TACS patients more efficiently than the SF-36 (RE, 0.01 to 0.54), LHS (RE, 0.45), Barthel Index (RE, 0.50), IDA (RE, 0.02 to 0.10), and NIHSS (RE, 0.47). Only the independent living subscale of the AQoL discriminated between these groups of patients with more efficiency than the total AQoL score (RE, 1.84).
In logistic regression models, the AQoL 3-month score was a significant predictor of the outcomes of death (13 cases), institutionalization (7 cases), or both 12 months after stroke. The following models describe how lower (worse) AQoL scores are associated with increased odds of an adverse outcome: ln(odds of death)=-1.059-2.404(AQoL score), P=0.034; ln(odds of institutionalization)=-1.193-7.149(AQoL score), P=0.037; and ln(odds of either)=-0.173-3.808(AQoL score), P=0.001.
The AUC for the model predicting death or institutionalization was 0.773 (SE, 0.056; P=<0.001 for the null hypothesis that AUC=0.5). For institutionalization alone, AUC=0.829, SE=0.05, and P=0.004. For death alone, AUC=0.692, SE=0.74, and P=0.027.
When adjusted for age, sex, and 3-month impairment score, AQoL score at 3 months remained a significant predictor of death or institutionalization (P=0.006). Actual and predicted rates of death or institutionalization at 12 months after stroke according to AQoL utility score at 3 months are shown in Figure 2.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendixReferences |
|---|
We have provided evidence that the AQoL is a valid instrument for assessing HRQoL after stroke. The strongest associations of AQoL and its dimensions with other measures were observed between scales measuring similar constructs, and the weakest relationships were between conceptually unrelated scales, indicating convergent and discriminant validity. The independent living dimension, as expected, correlated best with measures of physical impairment, disability, handicap, and physical functioning. The social relationships dimension correlated most highly with measures of disability, social functioning, and mood impairment, suggesting that both physical and psychological functioning is important for satisfying relationships. The physical senses dimension correlated best with disability and handicap measures. Low correlations between this dimension and SF-36 scales are expected because the items covered (seeing, hearing, communication) are not specifically addressed by the SF-36. The psychological well-being dimension was correlated most highly with other measures of psychological function. The psychological well-being incorporates pain assessment, which explains the high correlation with the SF-36 bodily pain scale.
Total AQoL utility scores were more strongly correlated with scores from the LHS (handicap) than with the Barthel Index (disability) and NIHSS (impairment), and this, together with the moderate correlation between the AQoL and most SF-36 subscales, suggests strong criterion validity. The highest correlation was with the LHS, which was expected because the design of the AQoL included a substantial emphasis on handicap and because the LHS is also a utility instrument that incorporates community preferences of the patients’ health status. There are major differences between the AQoL and the SF-36, so high correlations between these instruments should not be expected. The AQoL is a multiattribute utility instrument designed to measure HRQoL. Although a description of the patient’s difficulties relating to health is provided, these levels of health are then substituted during scoring by population preference weights established through the time tradeoff procedure whereby respondents "trade off" life length for health level. The AQoL utility value is presented, therefore, on a life-death scale in which the lower boundary is -0.04 (HRQoL states valued worse than death), 0.00 (HRQoL states valued equivalently to death), to 1.0 (HRQoL states valued as perfect HRQoL). The appropriateness of a possible "worse than death" utility for major stroke has been previously discussed.34 The SF-36 is used to describe the patient’s health status largely in terms of the impact of health on activities, although perception of general health status and mood are also assessed. Dimension scores for the SF-36 are computed with a non–preference-weighted additive model in which the scores representing health status are quantified on a 0-to-100-point rating scale. Given these differences between instruments, the correlations are of the magnitude expected.
The AQoL was sensitive in discriminating differences in health status between patients with varying severity of stroke, as defined by disability and impairment scores. The utilities observed for severe and mild stroke are similar to those reported elsewhere.34 Through the use of the RE, it was observed that the AQoL was more efficient than the SF-36, LHS, Barthel Index, NIHSS, and the anxiety and depression subscales of the IDA in discriminating between the health status of TACS and non-TACS patients. It is possible that some severe cases of stroke may have been included in the non-TACS group (eg, some posterior circulation strokes may be severe); however, this would tend to minimize differences, and the AQoL was able to clearly differentiate between these groups. No substantial ceiling or floor effects were observed for the overall utility score. Finally, AQoL scores at 3 months after stroke predicted future death or institutionalization, demonstrating the predictive validity of the instrument.
The lack of consistency between studies in the choice of outcome measure makes interpretation and comparison between studies difficult.35 The use of a single measure in place of multiple scales would be less burdensome on patients and would simplify the interpretation of results. The concordance between data obtained from the AQoL and those obtained from measures of physical and mood impairment, disability, and handicap, coupled with the demonstrated sensitivity in discriminating between different health states, suggests that further study is warranted to assess whether the AQoL could potentially be used as a sole outcome measure in trials or natural history studies.
The illness domain was included in the original development of the AQoL scale, but this scale is not used to derive an overall utility score. For those researchers who wish to have the most parsimonious questionnaire, the 3 questions related to the Illness scale could be omitted, resulting in a briefer 12- item instrument. Although not designed specifically for stroke patients and having no specific questions about upper-extremity function and cognition, as included in stroke-specific HRQoL instruments,36,37 the AQoL performs well. Despite its brevity, it captures much of the variance of the handicap, disability, and impairment instruments and returns information on broader aspects of HRQoL. The performance of the AQoL is most likely due to the relevance of its domains to stroke survivors (self-care, household tasks, mobility, relationships, social isolation, family role, seeing, hearing, communication, sleep, anxiety and depression, and pain). The questionnaire is easy to use, taking 5 minutes to administer to most patients. It can be used in a variety of settings because it was designed for self-completion or use at interview, and it appears suitable for proxy assessments.
Recently, a stroke-specific QOL scale (12 domains, 78 items)36 and a stroke impact scale (8 domains, 64 items) have been developed.37 These have the advantage of returning substantial detail on stroke-specific outcomes, but their length, lack of a single outcome score, and inability to be used in nonstroke comparison groups may make them unsuitable for use in some studies. Further validation of the AQoL using stroke-specific HRQoL scales would be useful.
The AQoL can be recommended for use in cross-sectional studies of outcome after stroke. Further evaluation is required to replicate these findings in other stroke populations, to confirm its usefulness in longitudinal studies, and to formally examine the reliability of proxy assessment. Despite these caveats, we have provided initial evidence that the AQOL is a valid, sensitive measure of HRQoL in stroke patients.
|
Acknowledgments |
|---|
This work was supported by grants from the Public Health Research and Development Committee of the National Health & Medical Research Council, the Victorian Health Promotion Foundation, Foundation for High Blood Pressure Research, and the National Stroke Foundation. Dr Sturm was supported by a postgraduate medical scholarship provided by the National Health & Medical Research Council. Lichun Quang provided assistance with database management and analysis. Donna Bradford and Lucie O’Malley provided administrative support. The contribution of the following research nurses is also gratefully acknowledged: Stephen Cross, Barbara Dowell, Elspeth Freeman, Jodi Haartsen, Meg Hooton, Amanda Loth, Sally Roberts, Catherine Sharples, Mary Staios, Cathy Taranto and Dennis Young. We thank Dr John Ludbrook (Biomedical Statistical Consulting Service) for assistance with the statistical analyses.
|
Appendix |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendixReferences |
|---|
|
|
Received November 1, 2001; revision received July 15, 2002; accepted July 17, 2002.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendixReferences |
|---|
1. Seale C, Davies P. Outcome measurement in stroke rehabilitation research. Int Disabil Stud. 1987; 9: 155–160.[Medline] [Order article via Infotrieve]
2. Williams LS. Health-related quality of life outcomes in stroke. Neuroepidemiology. 1998; 17: 116–120.[CrossRef][Medline] [Order article via Infotrieve]
3. Buck D, Jacoby A, Massey A, Ford G. Evaluation of measures used to assess quality of life after stroke. Stroke. 2000; 31: 2004–2010.
4. Hawthorne G, Richardson J, Osborne R. The Assessment of Quality of Life (AQoL) instrument: a psychometric measure of health-related quality of life. Qual Life Res. 1999; 8: 209–224.[CrossRef][Medline] [Order article via Infotrieve]
5. Hawthorne G, Richardson J, Day N, McNeil H. Using the Assessment of Quality of Life (AQoL) Instrument. Melbourne, Australia: Centre for Health Program Evaluation; 2000.Technical report 12.
6. Hogan A, Hawthorne G, Kethel L, Giles E, White K, Stewart M, Plath B, Code C. Health-related quality of life outcomes from adult cochlear implantation: a cross-sectional study. Cochlear Implants Int. 2001; 2: 115–128.[CrossRef][Medline] [Order article via Infotrieve]
7. South Australian Health Omnibus Survey. Adelaide, Australia: South Australian Health Commission, Epidemiology Department; 1998.
8. Hawthorne G, Richardson J, Day NA. A comparison of the Assessment of Quality of Life (AQoL) with four other generic utility instruments. Ann Med. 2001; 33: 358–370.[Medline] [Order article via Infotrieve]
9. McNeil H, Segal L. Quality of Life and Obesity. Melbourne, Australia: Centre for Health Program Evaluation; 1999.Report 17.
10. Osborne R, Hawthorne G, Papanicolaou M, Wegmueller Y. Measurement of rapid changes in health outcomes in people with influenza symptoms. J Outcomes Res. 2000; 4: 15–30.
11. Ware J, Snow K, Kosinski M, Gandek B. SF-36 Health Survey: Manual and Interpretation Guide. Boston, Mass: Health Institute, New England Medical Centre; 1993.
12. Euroqol Group. Euroqol: a new facility for the measurement of health-related quality of life. Health Policy. 1990; 16: 199–208.[CrossRef][Medline] [Order article via Infotrieve]
13. Thrift AG, Dewey HM, Macdonell RA, McNeil JJ, Donnan GA. Stroke incidence on the east coast of Australia: the North East Melbourne Stroke Incidence Study (NEMESIS). Stroke. 2000; 31: 2087–2092.
14. Sudlow CL, Warlow CP. Comparing stroke incidence worldwide: what makes studies comparable? Stroke. 1996; 27: 550–558.
15. Hatano S. Experience from a multicenter stroke register: a preliminary report. Bull World Health Org. 1976; 54: 541–553.[Medline] [Order article via Infotrieve]
16. Thrift AG, Dewey HM, Macdonell RAL, McNeil JJ, Donnan GA. Incidence of the major stroke subtypes: initial findings from the North East Melbourne Stroke Incidence Study (NEMESIS). Stroke. 2001; 32: 1732–1738.
17. Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet. 1991; 337: 1521–1526.[CrossRef][Medline] [Order article via Infotrieve]
18. Anderson CS, Laubscher S, Burns R. Validation of the Short-Form 36 (SF36) Health Survey questionnaire among stroke patients. Stroke. 1996; 27: 1812–1816.
19. Brott T, Adams HP, Olinger CP, Marler JR, Barson WG, Biller J, Spilker J, Holleran R, Eberle R, Hertzberg V, Rorick M, Moomaw CJ, Walker M. Measurements of acute cerebral infarction: a clinical examination scale. Stroke. 1989; 20: 864–870.
20. Chiu D, Krieger D, Villar-Cordova C, Kasner SE, Morgenstern LB, Bratina PL, Yatsu FM, Grotta JC. Intravenous tissue plasminogen activator for acute ischemic stroke: feasibility, safety, and efficacy in the first year of clinical practice. Stroke. 1998; 29: 18–22.
21. Lyden P, Brott T, Tilley B, Welch K, Mascha E, Levine S, Haley E, Grotta J, Marler J, for the NINDS TPA Stroke Study Group. Improved reliability of the NIH Stroke Scale using video training. Stroke. 1994; 25: 2220–2226.[Abstract]
22. Mahoney F, Barthel D. Functional evaluation: the Barthel Index. Md State Med J. 1965; 14: 61–65.[Medline] [Order article via Infotrieve]
23. Wade DT, Hewer RL. Functional abilities after stroke: measurement, natural history and prognosis. J Neurol Neurosurg Psychiatry. 1987; 50: 177–182.
24. Harwood RH, Ebrahim S. Manual of the London Handicap Scale. Nottingham, UK: University of Nottingham; 1995.
25. Harwood RH, Gompertz P, Ebrahim S. Handicap one year after a stroke: validity of a new scale. J Neurol Neurosurg Psychiatry. 1994; 57: 825–829.
26. Snaith RP, Constantopolous AA, Jardine MY, McGuffin P. A clinical scale for the self-assessment of irritability. Br J Psychiatry. 1978; 132: 164–171.
27. World Health Organization. International Classification of Impairments, Disabilities and Handicaps. Geneva, Switzerland: World Health Organization; 1980.
28. Zar JH. Significance testing of the Spearman rank correlation coefficient. J Am Stat Assoc. 1972; 67: 578–580.[CrossRef]
29. Liang MH, Larson MG, Cullen KE, Schwartz JA. Comparative measurement efficiency and sensitivity of five health status instruments for arthritis research. Arthritis Rheum. 1985; 28: 542–547.[Medline] [Order article via Infotrieve]
30. McHorney CA, Ware J, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36), II: psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993; 31: 247–263.[Medline] [Order article via Infotrieve]
31. Sturm JW, Dewey HM, Donnan GA, Macdonell RAL, McNeil JJ, Thrift AG. Handicap after stroke: how does it relate to disability, perception of recovery, and stroke subtype? Stroke. 2002; 33: 762–768.
32. Anderson C, Taylor BV, Hankey GJ, Stewart-Wynne EG, Jamrozik KD. Validation of a clinical classification for subtypes of acute cerebral infarction. J Neurol Neurosurg Psychiatry. 1994; 57: 1173–1179.
33. Ludbrook J, Dudley H. Why permutation tests are superior to t and F tests in biomedical research. Am Stat. 1998; 52: 127–132.[CrossRef]
34. Post PN, Stiggelbout AM, Wakker PP. The utility of health states after stroke: a systematic review of the literature. Stroke. 2001; 32: 1425–1429.
35. Duncan PW, Jorgensen HS, Wade DT. Outcome measures in acute stroke trials. Stroke. 2000; 31: 1429–1438.
36. Williams LS, Weinberger M, Harris LE, Clark DO, Biller J. Development of a stroke-specific quality of life scale. Stroke. 1999; 30: 1362–1369.
37. 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.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  S. L. Paul, J. W. Sturm, H. M. Dewey, G. A. Donnan, R. A.L. Macdonell, and A. G. Thrift Long-Term Outcome in the North East Melbourne Stroke Incidence Study: Predictors of Quality of Life at 5 Years After Stroke Stroke, October 1, 2005; 36(10): 2082 - 2086. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Nichols-Larsen, P.C. Clark, A. Zeringue, A. Greenspan, and S. Blanton Factors Influencing Stroke Survivors' Quality of Life During Subacute Recovery Stroke, July 1, 2005; 36(7): 1480 - 1484. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Sturm, G. A. Donnan, H. M. Dewey, R. A. L. Macdonell, A. K. Gilligan, V. Srikanth, and A. G. Thrift Quality of Life After Stroke: The North East Melbourne Stroke Incidence Study (NEMESIS) Stroke, October 1, 2004; 35(10): 2340 - 2345. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||