212
P
REDICTING
THE
G
RADE
OF
D
ISABILITY
1 Y
EAR
A
FTER
S
TROKE
F
OLLOWING
R
EHABILITATION
Jau-Hong Lin, Athena Yi-Jung Tsai,1 Sing-Kai Lo,2 Jyh-Jong Chang,1 and Mao-Hsiung Huang3
Faculties of Physical Therapy and 1Occupational Therapy, Kaohsiung Medical
University, 3Department of Rehabilitation, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan, and 2The George Institute for International Health,
University of Sydney, Australia.
The purpose of this study was to identify predictors of grades of disability at least 1 year after stroke rehabilitation therapy. We recruited stroke patients from the inpatient rehabilitation department of a university hospital. The degree of disability was graded using the disability evaluation at least 1 year after stroke onset. Functional ability was evaluated using the Functional Independence Measure instrument on admission, on discharge from the inpatient rehabilitation program, and at the 6-month follow-up visit after discharge. Major sociodemographic, medical, and rehabilitative factors were also collected during the hospitalization period. Of the 109 patients surveyed, 64 (58.7%) had severe or very severe grades of disability. The correlates of severe or very severe disability in logistic regression analyses were bilaterally affected (odds ratio, OR, 10.8), impaired orientation (OR, 3.6), and poorer functional ability at discharge (OR, 7.6). Based on the significant predictors identified, the logistic regression model correctly classified severe or very severe disability in 68.0% of subjects. The higher frequency of severe or very severe disability in this study may have been due to the relatively more severely affected stroke patient population in the inpatient rehabilitation service and the use of unique disability evaluation criteria. These results may provide information useful in planning continuous rehabilitation care and setting relevant socio-welfare policies for stroke victims.
Key Words: disability, rehabilitation, stroke
(Kaohsiung J Med Sci 2005;21:212–9)
Received: August 23, 2004 Accepted: February 24, 2005 Address correspondence and reprint requests to: Dr. Jau-Hong Lin, Faculty of Physical Therapy, Kaohsiung Medical University, 100 Shih-Chuan 1st Road, Kaohsiung 807, Taiwan.
E-mail: [email protected]
Although most stroke patients who receive rehabilitation may improve in function, the rate and quality of im-provement vary [1,2]. Some patients may experience per-manent disability and remain dependent in many ways [3, 4]. Their need for continuing care and management has significant impacts on family members and society, and is a major concern for health care policy makers. Early identification of the factors affecting the grade of disability after recovery will be helpful in establishing reasonable rehabilitation programs and relevant socio-welfare policies.
Disability evaluation (DE) is an official measurement of disablement especially in type and grade/severity. Disable-ment is defined as “a collective term referring to any experi-ence that is a consequexperi-ence of disease and which may be identified as impairment, disability or handicap” [5]. The final determination of the disability grade represents a legal identification of the disabled person and specifies and classifies the level of need for assistance and compensa-tion [6]. The purposes of measuring disablement include: planning for future services, monitoring patient care, eval-uating intervention effects, epidemiology, and assessing eligibility for benefits [7]. The last purpose is different from the common medical outcome measurements such as basic or instrumental function of activities of daily living (ADL). Without official approval from the DE, a patient cannot be eligible for formal compensation, assistance, or legal
exemption despite ADL limitations.
In Taiwan, stroke patients with motor, cognition and/ or language limitations usually have their DE by medical specialists 1 year after onset. Qualified stroke victims will receive their Physically and Mentally Disabled Manual as an identification document to apply for welfare, health, education, environmental accommodation, communi-cation, vocational and financial assistance or exemptions [8]. The grade from the DE becomes a crucial function-al outcome index because it represents the level of dis-ability or handicap recognized by society. The purpose of this prospective follow-up study was to identify predictors of DE grades at least 1 year after stroke for victims in Taiwan.
M
ETHODSSubjects
Participants were recruited from stroke patients con-secutively admitted to the rehabilitation department of Kaohsiung Medical University Hospital. This study was approved by the Ethics Committee of the University. In-formed consent was obtained from all participants prior to participation. Stroke was defined as a rapidly developing clinical manifestation of a focal loss of cerebral function lasting more than 24 hours [9]. Diagnosis was made by consultant physicians and neurologists based on clinical symptoms and confirmed by findings of neuroimaging studies. During the study period, 172 stroke patients were admitted consecutively to the department with a diagnosis of cerebrovascular disease (International Classification of Diseases, ICD-9-CM, codes 430–434, 436–438) [10]. Of these, nine (5%) were excluded because there was no writ-ten consent or the data were incomplete. To minimize the influence of potential confounding effects, 12 patients who had a diagnosis of dementia at discharge or coma in a persistent vegetative state because of recurrent stroke or deteriorating condition during hospitalization were excluded. The follow-up DE was conducted at a mean of 13.8 ± 2.6 months after discharge. Of the 151 partici-pants, 42 dropped out before the completion of the DE due to death (20), loss to follow-up (4), or incomplete DE (18). A total of 109 patients completed the DE in the rehabilitative or neurological clinics during the follow-up period from January 1, 1998, to July 31, 1999.
Instruments and procedure
Subjects underwent functional status assessment on
ad-mission, on discharge from the inpatient rehabilitation program, and at the 6-month follow-up visit after discharge. Functional status was assessed based on the ability to perform items of the Functional Independence Measure (FIM) [11]. All subjects were evaluated by the same senior physical therapist trained in using the FIM. The FIM is an 18-item, 7-level scale for assessing the patient’s need for assistance (or devices) in six areas of daily activities: self care (eating, bathing, grooming, dressing upper body, dressing lower body, toileting); sphincter control (bowel, bladder); transfers (bed/chair transfer, toilet transfer, tub transfer); locomotion (walk/wheelchair, stairs); commu-nication (comprehension, expression); and social cogni-tion (social interaccogni-tion, problem-solving, memory). Each FIM item is scored on seven ordinal levels, with 1 being the most disabled condition and 7 being the least. This makes a total possible FIM score of 18–126. In addition, orientation status at discharge was identified using the orientation scale of the Scandinavian Stroke Scale (SSS), which asks the patient to identify time (month), place (hospital name), and date of birth [12]. A patient answering correctly in all three items was regarded as having no disorientation.
Sociodemographic information, including gender, age at onset, years of education, occupation, living and marital status, and main caregiver, was collected using a question-naire during hospitalization. Major medical data, including lesion area, side of paralysis, whether the attack was recur-rent, stroke etiology, risk factors, medical complications, interval from stroke onset to rehabilitation commence-ment, rehabilitation stay, and whether the patient received continuous outpatient rehabilitation therapy, was collected from medical records within 1 month after discharge.
The Physically and Mentally Disabled Citizen Protection Law defines 14 official categories of disabilities with 3–4 grades (mild, moderate, severe, and very severe) under each category, and specifies a comprehensive evaluation procedure [8]. Disabled stroke victims might fit into one of three categories: limbs disabled, with irreversible limb dysfunction; voice\speech mechanism disabled, with dif-ficulties in language comprehension or expression; and multi-disabled, with more than one category of disability. The evaluation of disability was determined by either a senior physiatrist or neurologist based on the criteria of the Physically and Mentally Disabled Citizen Protection Law and related regulations [8].
Statistical analysis
In order to strengthen the power of comparison, the grade of disability was grouped into mild/moderate and severe/
very severe. Repeated measures analysis of variance (ANOVA) was used to compare the functional status at admission, discharge, and 6-month follow-up after stroke between these two groups. Univariate analysis (Chi-squared test or Fisher’s exact test, as appropriate) was used to examine the associations between categorical clinical and demographic factors among stroke patients and between the two disabled groups. For this reason, the discharge FIM score was pooled into either good (64–126) or poor-to-middle (18–63) categories [13]. Factors significant in the univariate analysis were then considered potential predic-tors of the grade of disability, and fitted into a stepwise logistic regression to construct a predictive model. All analyses were performed using the SAS statistical package version 6.12 for Windows (SAS Institute Inc, Cary, NC, USA). The level of significance was set at a p of 0.05 or less.
R
ESULTSOf 109 patients surveyed, there were 57 male (52%) and 52 female subjects. The mean age of stroke onset was 62.8 ± 11.7 years. The average number of years of education was 6.1 ± 4.8, equivalent to primary/elementary education. On average, the rehabilitation intervention started within a month of stroke onset (27.3 ± 21.1 days). All patients had at least limb disability on DE, in which five patients (5%), 59 (54%), 44 (40%), and one (1%) were graded as having very severe, severe, moderate, or mild disability, respectively. FIM scores were significantly different between the two groups at all time points (Table 1).
Various factors were examined to find out whether they were associated with the grade of disability (Tables 2–4). There were no obvious associations between severity and sociodemographic and general medical factors, except for the presence of medical complications (Tables 2 and 3). Among the stroke-related factors (Table 4), patients who had delayed rehabilitation therapy, did not receive continuous outpatient rehabilitation care, were bilaterally
affected or orientation impaired, or had lower FIM score on discharge were more likely to be severely or very severely disabled. Except for the presence of medical complications, the delay of rehabilitation therapy, and no continuous out-patient rehabilitation care, only the following three factors remained significant in the logistic regression (Table 5): lower FIM score on discharge (odds ratio, OR, 7.6; 95% confidence interval, CI, 2.3–24.6), bilaterally affected (OR, 10.8; 95% CI, 1.3–91.9), and orientation impaired (OR, 3.6; 95% CI, 1.0–12.5); these were associated with a more severe grade of disability.
The resulting regression coefficients could be used to estimate the probability of developing severe or very severe disability at 12 months after stroke. This model correctly classified 68% of patients in this study as having severe or very severe disability. For example, for a stroke patient with bilateral hemiplegia due to brain stem infarction, intact orientation, and a 36-point FIM score on discharge, the probability of severe or very severe disability at 12 months after stroke may be estimated as follows:
Probability = 1/[1 + e–(–2.57 + A×1 + B×0 + C×1)] = 1/[1 + e–(–2.57 + 2.38 + 0 + 2.02)] = 0.86
Where e is the exponential function, –2.57 is the regression coefficient constant, A is the regression coefficient of side of hemiplegia, B is the coefficient of orientation, and C is the coefficient of the discharge FIM score.
D
ISCUSSIONThis prospective study is the first to investigate the grade of disability in stroke victims according to the DE in Taiwan. We also established a model to predict severe or very severe disability.
Among the 109 stroke victims surveyed, 64 (58.7%) had severe or very severe disability. This percentage is con-siderably higher than that reported in the Copenhagen study (20%) [2].The main reason for the difference may be
Table 1. Mean Functional Independence Measure (FIM) scores in the two disability groups at different time points
Grade of disability Admission FIM* Discharge FIM* Follow-up FIM* p
Mild/moderate (n = 45) 65.4 ± 21.2 87.4 ± 17.4† 100.1 ± 19.6‡ int = 0.016 Severe/very severe (n = 64) 46.7 ± 22.7 65.3 ± 27.4† 69.7 ± 33.9 grp = 0.000 *p < 0.0001 between disabled groups; †p < 0.0001 from admission after Sharpened Bonferroni adjustment; ‡p < 0.0001 from discharge after Sharpened
Table 3. Associations between general medical status and severe or very severe disability
General medical status N Severe or very severe disability, % χ2 p Prior hypertension No 6 46.2 Yes 58 60.4 0.96 0.33 Prior diabetes No 42 62.7 Yes 22 52.4 1.13 0.29
Prior heart disease
No 42 56.8
Yes 22 62.9 0.37 0.55
Other medical complications
No 36 50.0
Yes 28 75.7 6.65 0.01
Table 2. Associations between demographic factors and severe or very severe disability
Demographic factor N Severe or very severe disability, % χ2 p Gender Male 36 63.2 Female 28 53.9 0.97 0.32 Age (yr) < 65 32 56.1 ≥ 65 32 61.5 0.33 0.57 Education (yr) < 7 43 58.1 ≥ 7 21 60.0 0.04 0.85 Literate No 13 48.2 Yes 51 62.2 1.65 0.20 Occupation Labor 25 59.5 Others 39 58.2 0.02 0.89
Living with family members
No 3 25.0
Yes 61 62.9 — 0.01*
Marital status
Married 44 59.5
Single/divorced/widowed 20 57.1 0.05 0.82
Main caregiver during hospitalization
Family members 37 53.6
Professional 27 67.5 2.01 0.16
*Fisher’s exact test.
that all subjects recruited in this study were referred to the inpatient rehabilitation service in a medical center. These patients might have more severe neurologic and physical impairments. Moreover, the present study used DE to
de-fine the grade of disability, which is very different from pre-vious studies that used Barthel Index (BI) scores (0–20 and 25–45) to define very severe and severe disabilities [2,14, 15]. Under the limbs category, the grade of disability was
Table 4. Associations between stroke-related factors and severe or very severe disability
Stroke-related factor N Severe or very severe disability, % χ2 p Onset to rehabilitation therapy
< 31 days 42 52.5
≥ 31 days 22 75.9 4.80 0.03
Rehabilitation stay
< 31 days 24 53.3
≥ 31 days 40 62.5 0.92 0.34
Received continuous outpatient rehabilitation care
Yes 20 46.5
No 44 66.7 4.36 0.04
Number of attack
First 44 54.3
Recurrent 20 71.4 2.51 0.11
Brain lesion area
Cortical 33 62.3 Subcortical 31 55.4 0.54 0.46 Type of stroke Infarction 36 58.1 Hemorrhage 28 59.6 0.03 0.87 Side of hemiplegia Unilateral 52 54.2 Bilateral 12 92.3 6.87 0.009 Orientation Normal 42 50.6 Impaired 22 84.6 9.45 0.002
Discharge FIM score
18–63 points 28 87.5
64–126 points 36 46.8 15.5 0.001
FIM = Functional Independence Measure.
judged according to the patient’s “impairments” as loss or abnormality of psychologic, physiologic, or anatomic structure or function at the organ or system level. However, BI scores focus on the levels of daily activity functions, which measure “disability”. Impairment of the limbs does not necessarily lead to self-care dysfunction. The DE under the Social Security system in the USA has been amended to assess stroke victims’ disability rather than impairment [16]. Since one of the aims of DE is to decide the appropriate level of services and benefits, it should reflect the degree of difficulty in activity performance. From this viewpoint, mea-surement of disability in terms of comprehensive ADL score might better serve this purpose than impairment [17]. In this study, functional status was significantly different between the two disabled groups, as assessed by the FIM instrument. The changes in FIM scores indicate that the
severe/very severe disability group had significant im-provement during the rehabilitation stay, but less signifi-cant improvement at later stages of recovery (discharge to 6-month follow-up). However, patients in the mild/mod-erate disability group, who had higher admission FIM scores, not only improved significantly during rehabilita-tion stay but also showed continued improvement at the 6-month follow-up visit. This finding is consistent with previous reports showing that most recovery occurs within the first 3–6 months [2,14]. Therefore, DE should be held at least 6 months after stroke onset [16,18].
On the other hand, bilateral hemiplegia, impaired orientation, and lower discharge FIM scores were found to be connected to the grade of disability. We found that bi-lateral stroke involvement was an important predictor of severe or very severe disability. The neurologic impairment
Table 5. Significant predictors influencing the results of disability evaluation
Factor Regression coefficient (β) Standard error p Odds ratio 95% CI Side of hemiplegia Bilateral 2.38 1.09 0.0299 10.8 1.26–91.9 Left or right — — — 1.00 — Orientation Impaired 1.28 0.64 0.0442 3.60 1.03–12.5 Normal — — — 1.00 —
Discharge FIM score
18–63 points 2.02 0.60 0.0008 7.55 2.32–24.6
64–126 points — — — 1.00 —
Constant –2.57 0.75 0.0006 — —
CI = confidence interval; FIM = Functional Independence Measure.
or physical disability left by a prior stroke will affect motor recovery from a subsequent stroke. Bilateral involvement in motor function of various severities was common in our patients after recurrent stroke. Apart from the impaired physical ability due to the present neurologic insult, the loss of ability to compensate for such loss from the sound side may also increase the dependency of stroke patients. The present study supports the findings of Pedersen et al that impaired orientation during hospitalization exerts a marked, negative influence on basic ADL and social function and subsequently leads to severe or very severe disability [12]. In this study, clinical and demographic data were collected before discharge instead of at admission, as in previous studies [19,20]. The total FIM score at discharge, divided into poor-to-middle (18–63) and high (64–126) score categories, was a useful predictor of the grade of disability. Although there was no universal agreement on score-ranking, and bias might also have been generated by dif-ferences between raters when conducting FIM measure-ments, FIM in general seems to be a useful tool not only to assess function but also to predict the grade of disability.
Although no related literature can be referred to in pre-dicting the outcome of DE, the three predictors from the current study seem to be similar to those that predict func-tional outcome in the Western literature [2,14,15,19]. Many studies have shown that psychologic and social variables (which were not used in the current study) are important correlates of disability in post-stroke patients [21–23]. Fu-ture studies using a standardized psychologic and social support instrument and exploring the impact of related factors on the grade of disability will further improve the accuracy of the prediction in this study. A limitation of this
study was that the duration of the continuous outpatient rehabilitation care after discharge was not investigated, which might be a very important factor affecting the results of this study. Further studies involving a larger sample from multiple district hospitals as well as medical centers are needed.
In summary, using logistic regression, models can be constructed to predict the probability of severe or very severe disability. The results of this study indicated that bilateral involvement, impaired orientation, and poorer functional ability at discharge were independent correlates of the grade of disability. Recognizing these factors may be helpful in selecting potential patients for aggressive rehabilitative treatment and preparing early relevant socio-welfare policies for stroke victims. Prospective studies to evaluate the usefulness of such a model are required.
A
CKNOWLEDGMENTSThis study was supported by a grant from the National Science Council (NSC 89-2320-B-037-014).
R
EFERENCES1. Iwanenko W, Fiedler RC, Granger CV. Uniform data system for medical rehabilitation: report of first admissions to subacute rehabilitation for 1995, 1996 and 1997. Am J Phys Med Rehabil 1999;78:384–8.
2. Jorgensen HS, Nakayama H, Raaschou HO, et al. Outcome and time course of recovery in stroke. Part 1: outcome. The Copenhagen Stroke Study. Arch Phys Med Rehabil 1995;76:
399–405.
3. Lin SH. Home health care and social supports for community-dwelling stroke patients. J Formos Med Assoc 1993;92(Suppl 4): 177–83. [In Chinese]
4. Clarke PJ, Black SE, Badley EM, et al. Handicap in stroke survivors. Disabil Rehabil 1999;21:116–23.
5. Patrick DL. A socio-medical approach to disablement. In: Patrick DL, Peach H, eds. Disablement in the Community. Oxford: Oxford University Press, 1989:1–18.
6. Nagi SZ. Disability and Rehabilitation: Legal, Clinical, and
Self-concepts, and Measurement. Columbus: Ohio State University
Press, 1969.
7. Charlton JH. Approaches to assessing disability. In: Patrick DL, Peach H, eds. Disablement in the Community. Oxford: Oxford University Press, 1989:39–61.
8. The Regulations of the Physically and Mentally Disabled Evaluation. Taipei: Department of Health, 1999.
9. Stroke–1989. Recommendations on stroke prevention, diagnosis and therapy. Report of the WHO Task Force on Stroke and Other Cerebrovascular Disorders. Stroke 1989;20: 1407–31.
10. International Classification of Diseases: Manual of the International
Statistical Classification of Disease, Injuries, and Causes of Death,
9th revision. Geneva: World Health Organization, 1977.
11. Guide for the Uniform Data Set for Medical Rehabilitation (including
the FIMTM
instrument), Version 5.1. Buffalo: State University of
New York at Buffalo, 1997.
12. Pedersen PM, Jorgensen HS, Nakayama H, et al. Orientation in the acute and chronic stroke patient: impact on ADL and social activities. The Copenhagen Stroke Study. Arch Phys
Med Rehabil 1996;77:336–9.
13. Brosseau L, Potvin L, Philippe P, et al. Post-stroke inpatient rehabilitation: predicting discharge disposition. Am J Phys
Med Rehabil 1996;75:431–6.
14. Wade DT, Hewer RL. Functional abilities after stroke: mea-surement, natural history and prognosis. J Neurol Neurosurg
Psychiatry 1987;50:177–82.
15. Sze K-H, Wong E, Or KH, et al. Factors predicting stroke dis-ability at discharge: a study of 793 Chinese. Arch Phys Med
Rehabil 2000;81:876–80.
16. Social Security Administration. Disability Evaluation Under Social
Security. Department of Health and Human Services, Blue
Book, 2003.
17. Hsueh IP, Wang WC, Sheu CF, et al. Rasch analysis on combining two indices for assessing comprehensive ADL function in stroke patients. Stroke 2004;35:721–6.
18. Patel AT. Disability evaluation following stroke. Phys Med
Rehabil Clin North Am 2001;12:613–9.
19. Ween JE, Alexander MP, D’Esposito M, et al. Factors predictive of stroke outcome in a rehabilitation setting. Neurology 1996; 47:388–92.
20. Inouye M, Hashimoto H, Mio T, et al. Influence of admission functional status on functional change after stroke rehabili-tation. Am J Phys Med Rehabil 2001;80:121–5.
21. Glass TA, Matchar DB, Belyea M, et al. Impact of social sup-port on outcome in first stroke. Stroke 1993;24:64–70. 22. Tatemichi TK, Desmond DW, Stern Y, et al. Cognitive
impair-ment after stroke: frequency, patterns, and relationship to func-tional abilities. J Neurol Neurosurg Psychiatry 1994;57:202–7. 23. Kwakkel G, Wagenaar RC, Kollen BJ, et al. Predicting disability
in stroke – a critical review of the literature. Age Ageing 1996; 25:479–89.
!"#$%&'()*+,-./'0
N = = O= = P= = O= = Q N !"!= !"#==O !"# P !"#= !"#$%&'= =Q !"!#$ %= !"#$%&'()*+,-./0123456789:;"<=>?@ !"#$%&%'($)*+,-.)/012345016789:;2< !"#$%&'()*+,-.#$/0123456789:;<=8>? !"#$%&'()*+,-./01234= EcìåÅíáçå~ä= fåÇÉéÉåÇÉåÅÉ jÉ~ëìêÉ cfjF= !"#$%&'()*+,-./0123456789 !"#$%&'()*+,-./=NMV= !"#$%&'()*=SQ= ERUKTBF= !"#$%&'(%&)*+,-./01%23456789:;!"#$%&=E !=NMKUF= !"#$%&'()*+,-./0&
E !=PKSF= !"#$%&'()=E !=TKSF !"#$%&'( !"#$%&'()*+,-$./=SUKMB= !" #$%&'() !"#$%&#$'(")*+,-./012345637#89:;< !"#$%&'()*+,-./0*123456789:;<(+=>? !"#$%&'()*+,-./01234567 ! !"#$"%&' !"=OMMRXONWONOJV !"VP==U==OP= !"VQ==O==OQ= !"#$%&' !"!#$%&!' UMT ! NMM
