Relationships between initial motor assessment scale scores and length of stay, mobility at discharge and discharge destination after stroke.
Stroke is the second biggest cause of death among Australians annually and accounts for one quarter of chronic adult disability in Australia, making it one of the nation's highest causes of morbidity (Australian Institute of Health and Welfare 2008, National Stroke Foundation 2008). Every year approximately 60,000 strokes occur, with around 70% of these being first time events (Australian Institute of Health and Welfare 2008). Strokes have been estimated to cost the health system $2.14 billion annually (National Stroke Foundation 2008). Of the people suffering a stroke, one in five will die within 1 month, and one in three will die within one year (National Stroke Foundation 2008). Approximately 88% of stroke survivors live at home and most have a permanent disability. An increase in the overall number, as well as the proportion, of people over 65 years of age; in addition to advances in medical technology have resulted in an increase in the number of older people surviving stroke and requiring rehabilitation and ongoing care (Australian Institute of Health and Welfare 2008).
The average length of stay (LOS) for someone undergoing rehabilitation within an Australian public health service following a stroke is 71.3 days (McKenna et al 2002). As many as 17% of stroke survivors are unable to return home following discharge from rehabilitation (Tooth et al 2005). Inpatient rehabilitation following stroke and discharge to institutionalised care, such as nursing home accommodation, are associated with high social and economic costs (Australian Institute of Health and Welfare 2008). In 2000-2001, the breakdown of stroke care costs within the health system costs were as follows: inpatient rehabilitation ($150 million), nursing home care ($63 million) and provision of allied health services ($4.8 million) (Senes 2006). These costs place a considerable burden on health resources and are steadily increasing (Australian Institute of Health and Welfare 2008, Senes 2006).
The clinician's ability to measure the effectiveness of rehabilitation treatment underpins good practice. This is achieved using context appropriate, widely used, reliable, valid and responsive outcome measures (Barak and Duncan 2006, Salter et al 2005). If such tools are administered correctly, they can provide a guide to treatment planning and prioritisation, document progress made by the individual, indicate the need for modifying treatment regimes and assist in prognostication. In stroke rehabilitation settings, outcomes following rehabilitation have been measured in numerous ways (Salter et al 2005). Length of stay and discharge destination are commonly used measures of outcome, though both strongly depend on many factors making accurate prediction complex. To date no clear relationship has been established between location and severity of the stroke, and LOS or discharge destination. Indeed many studies examining predictive models of LOS have reported consistently low correlations between these factors (McKenna et al 2002, Tooth et al 2005, Wee et al 2003). Small sample sizes (McKenna et al 2002, Tooth et al 2005), retrospective study designs (Lutz 2004, McKenna et al 2002) and the multifactorial nature of contributors to LOS (McKenna et al 2002, Shah et al 1989, Wee et al 2003) have been cited as possible reasons for these findings.
Several studies analysing correlations between LOS and stroke category using the classification developed by Bamford et al (Table 1) have also reported conflicting results (Bamford et al 1991, Hakim and Bakheit 1998, Tooth et al 2005). Tooth et al (2005) found that those suffering a LACI had a significantly reduced LOS compared to someone suffering a POCI, whereas Hakim & Bakheit (1998) found there was no real difference between PACI, LACI and POCI classifications and that all these classifications had a shorter LOS than those surviving a TACI. These findings agree with clinical expectations as a TACI classified stroke generally results in the most severe disability; however it is clear that the relationships between LOS and stroke category are complex.
Length of hospital inpatient stay following stroke has been shown to have a moderate association with measures of function at admission such as the Barthel Index (Bohannon et al 2002) and the Berg Balance Scale (Wee et al 2003). Many factors such as motor function (Bohannon et al 2002, Hakim and Bakheit 1998, Tooth et al 2005), pre-morbid living status (Tooth et al 2005), LOS in the acute setting (Tooth et al 2005, Wee et al 2003) and the presence and extent of co-morbidities (Galski et al 1993, Lew et al 2002, Tooth et al 2005) have been reported to predict LOS. A negative correlation between initial motor function and discharge destination has been demonstrated previously (Bohannon et al 2002, Tooth et al 2005), and age has been shown to be a significant predictor of discharge destination, with younger individuals more likely to be discharged home (Bohannon et al 2002).
A commonly used tool to measure motor recovery and functional ability following stroke is the Motor Assessment Scale (MAS) (Carr et al 1985). The MAS is a criterion-based scale assessing 8 domains of functional motor activity with each item scored on a 7-point ordinal scale (ranging from 0= no motor function, to 6= optimal task performance or performance completed within the set time frame) (Carr and Shepherd 1998). The MAS has been shown to be reliable when used by qualified as well as undergraduate therapists (Carr et al 1985, Poole and Whitney 1988) and validity has been confirmed in 2 separate studies comparing total MAS scores with the Fugl-Meyer Assessment, producing correlations of r=0.96 and r=0.88 (Malouin et al 1994, Poole and Whitney 1988). One limitation is that the MAS does not discriminate between early stages of motor recovery (Hill et al 1997), as there are no criteria for flaccidity and the movement synergies typical in early stroke onset. Several studies have also confirmed a limitation in the dimensionality and scalability of the measure as it can be seen that top and/or bottom levels are overrepresented and other items show clustered middle levels (Aamodt et al 2006, Brock et al 2002, Dean and Mackey 1992, English et al 2006).
The ability to accurately predict discharge destination and functional outcomes following rehabilitation could result in improved planning, more efficient service utilisation and reduction in overall inpatient rehabilitation costs. In one recently published report, discharge destination was predicted with 87% accuracy, based on pre-stroke residential status, age and MAS scores for walking (item 5) and supine to side lying (item 1) (Brauer et al 2008). An earlier study suggested that recovery of motor function at discharge (total MAS discharge score) could be predicted based on age, balanced sitting (item 3) and combined arm scores (items 6-8) on admission (Loewen and Anderson 1990).
The present study aimed to examine, within an Australian aged care context, the extent to which MAS admission scores could be used to predict mobility outcomes at discharge, as well as LOS and discharge destination after stroke. It was hoped that this information would assist the therapist with goal setting and overall management of clinical loads, as well as determination of treatment priorities. In addition to informing therapeutic planning, this insight into an individual's potential for recovery could assist the patient, their family and the healthcare service from both financial and social perspectives.
This was a retrospective audit of patients admitted to the Stroke Rehabilitation Unit (SRU) at Osborne Park Hospital, Perth, Western Australia during the period June 2001 and January 2007. The SRU provides comprehensive, multidisciplinary rehabilitation after stroke for adults aged 65 years or older. Individuals entered the SRU by referral from a range of acute medical and neurosurgical services. Stroke survivors were not considered for admission if they were not deemed to have rehabilitation potential, for instance, if they had a shortened life expectancy due to co-morbidity (e.g., cancer), significant co-morbidity likely to impair ability to participate in rehabilitation (e.g., severe cardiac disease), poor functional level prior to admission or a cognitive impairment severe enough to impair rehabilitation potential.
We interrogated the SRU database to extract the required data on all patients admitted during the study period of June 2001 and January 2007. Patients were identified only by Unit Medical Record Number (UMRN) and variables recorded included: gender, age, date of birth, affected side, stroke classification (see Table 1) and stroke type (infarct or haemorrhage) (Bamford et al 1991), length of inpatient stay, admission and discharge MAS scores (total and individual component scores) and discharge destination.
Motor ability was assessed using the MAS within the first 1-2 days of admission to the SRU by one of four senior physiotherapists. To investigate the confounding influence of multiple assessors, evaluation of the inter-rater reliability was undertaken utilising a video of five stroke patients, with varying degrees of disability, performing the tasks within each subset of the MAS. Each physiotherapist scored the patients' performance using the MAS and scores were compared for percentage exact agreement.
Descriptive analyses were undertaken to describe the sample participants and inspect the total and individual subsections of the MAS on admission and discharge, as well as discharge destination and LOS. The total MAS scores on admission were recoded into four categories (<20, 20-29, 30-39, 40+). The term 'mobility' included the first 5 items on the MAS; while the ability to 'walk' at discharge was based on only item 5 (walking), which was dichotomised further into "walking with assistance" (scoring 1-3) or "walking independently" (scoring 4-6). The database recorded discharge destination as either being independent at home, home with carer or other support or discharge to an institution. For the prediction of discharge destination in this study, the variables were dichotomised to being discharged home (with or without support) or discharge to an institution.
The predictive value of initial MAS scores to determine LOS was investigated using a multiple linear regression procedure, while prediction of discharge destination and mobility on discharge utilised multiple logistic regression. The variables entered into each regression equation were age, gender, side, type and classification of stroke and MAS score on admission. Length of stay had a skewed distribution so these data were logarithm transformed to create a normal distribution. All analyses were performed using SPSS v 15.0. A probability of p<0.05 was considered to represent meaningful differences for all statistical tests.
Baseline Descriptive Data
Data were collected on a total of 253 people admitted to the SRU between June 2001 and January 2007. Data from fourteen individuals were excluded from analysis: seven died during their hospital admission period, two were transferred to another hospital and discharge clinical outcomes were not measured, two had other diagnoses, two had a second admission due to a stroke during the time period of this study, and in one case clinical data was incomplete (see figure 1). Consequently data from 239 individuals were included in the analysis. Demographic data for the study sample are presented in Table 2.
Percentage exact agreement between raters' scores of the eight MAS items for the five patient videos ranged from 55% for the balanced sitting (item 3) to 100% for the task of supine to sitting over side of the bed (item 2). The mean percentage agreement for the four raters over the five patients was 71%.
Relationships between MAS admission scores, patient demographics, length of stay, mobility on discharge and discharge destination.
1. Length of stay--There were moderate correlations between LOS and total MAS admission scores (r=-0.706) and admission mobility scores (items 1-5) (r=-0.716); with lower scores on admission indicating a greater likelihood for a longer LOS. Age, gender, type and side of stroke showed no association with LOS.
2. Mobility on discharge--Total MAS scores on admission were strongly associated with walking function (item 5) on discharge (#2= 119.4, 45DF, p=0.001). The lower the total MAS admission score, the less likely the patient was to be 'walking independently' (scoring between 4-6 for this item) on discharge.
3. Discharge destination--Correlations are summarised in Table 3. A significant association was observed between discharge destination and total MAS admission scores and mobility on admission (MAS items 1-5). Of the 169 patients discharged home, 68.3% had a total admission MAS score >30, whereas 85.7% of the 70 patients discharged to an institution had an MAS score <30. When the mobility component scores are considered, the score for supine to sitting over the side of bed (item 2) on admission showed greatest association with eventual discharge destination, followed by balanced sitting (item 3).
[FIGURE 1 OMITTED]
Predicting Discharge Destination
The MAS admission scores as well as age were included in the final model ([R.sup.2] = 0.249, P<0.001), (Table 4). The regression analysis illustrated that the lower the total MAS admission score, the more likely an individual was to be discharged to an institution (p<0.001). Similar trends were observed for age, with older individuals more likely to be discharged to an institution (p<0.001). This model correctly predicted a greater percentage of those discharged home, (with or without support) compared to discharge to an institution (Table 5).
Predicting Walking Function at Discharge
Of the variables examined (age, gender, side, type and classification of stroke and MAS admission scores) only age and total MAS admission scores were significantly related to the likelihood of walking independently on discharge (Table 6). The likelihood of walking also was shown to decrease with age (see table 6).
This study aimed to explore the utility of admission MAS scores to predict LOS, mobility at discharge and discharge destination, as well as to explore associations between other variables and these outcomes. Of the patient demographics explored for possible relationships with LOS, discharge destination and walking function, affected side and type of stroke, as well as gender showed no association with any of the discharge outcomes. Age was only significant in relation to predicting discharge destination and walking function on discharge.
Moderate associations were identified between eventual LOS admission and total MAS as well as MAS mobility items (r=-0.706 and r=0.716 respectively). However accurate prediction of LOS was limited by the retrospective nature of the study. While Tooth et al (2005) found the variables predictive of a longer LOS were poor admission FIM scores, living alone, longer acute LOS, co-morbidity and stroke classification, our model revealed total MAS scores on admission to be the only significant variable predicting LOS. Both models accounted for around half of the variance associated with LOS. This underlines the fact that there are many potential confounding factors that need to be taken into consideration when predicting LOS for an individual undergoing rehabilitation post stroke, some of which reflect social and economic perspectives. Overall LOS is affected by time spent in the acute setting as well as time spent waiting for placement. For instance, Wee et al (2003) found that one factor responsible for increased LOS in the stroke unit was the lack of bed availability in the intended discharge destination.
Another component of this study was to investigate the potential of admission MAS scores to predict discharge destination. Like many previous investigations (Bohannon et al 2002, Brauer et al 2008, Loewen and Anderson 1990, McKenna et al 2002, Tooth et al 2005, Wee et al 2003), our data show a strong association between lower MAS admission scores and the likelihood of being discharged to an institution; a relationship which is not unexpected. Using multiple regressions to predict discharge destination, Brauer et al (2008) found that previous living situation and age, as well as MAS scores for walking (item 5) and supine to side lying (item 1) were able to predict discharge destination with 86% accuracy ([r.sup.2] = 0.373, p < 0.001). Our logistic regression analyses correctly predicted a higher percentage of those who were discharged home, with or without support (86.4%) compared with only 51.4% of those discharged to an institution. Our data differed from Brauer et al's (2008) in the variables entered into the final regression equation. These researchers reported that pre-stroke residence was the most significant as this greatly influences patient's subsequent discharge destination; however this variable was not available in our database. Many factors such as cognition, age and social support determine requirements for high levels of care following a stroke, which partly explains the lower prediction accuracy for institutionalised care compared to predicting discharge home. The strongly predictive nature of our model (86.4% correct) indicates that age and total MAS admission scores can be considered to be key considerations when determining those who are likely to be discharged home following stroke.
Different correlations between MAS sub-sections and discharge destination have been reported previously. In Brauer's study (2008) the greatest association was with the MAS item 5 (walking) followed by MAS item 1: supine to side lying and then MAS item 2: supine to sitting over side of bed. In our study, MAS item 2 was most highly correlated with discharge destination, followed by item 4: sit to stand and item 3: balanced sitting. Although different sub-sections of the MAS have been reported to predict discharge destination in different studies, it is clear that it is the mobility items (MAS items 1-5), rather than arm and hand function, that appear to be the most influential. It has been hypothesised that the complexity of these tasks, which integrate motor control, perceptual ability, upper limb strength and trunk control, is the reason for their high predictive value of eventual discharge outcome. This premise is supported by previous findings that the presence of perceptual problems (Kalra et al 1997) and poor trunk control (Massucci et al 2006) are outcomes highly predictive of discharge to an institution.
One aspect of this investigation was to explore the value of the MAS to predict walking ability on discharge, an outcome of considerable importance to patients and their families post-stroke. The multiple logistic regression model indicated a limited ability to predict discharge walking function based solely on MAS total admission scores. The variables entered in the multiple logistic regression were age and MAS admission scores, although these were associated with significant variance. A major factor limiting the ability to predict gait function on discharge from admission scores was that most individuals scored 0 on the MAS item 5 (walking) on admission. This has been previously described as a downfall of the MAS due to the floor effect of this walking measure (Aamodt et al 2006, Brock et al 2002, English et al 2006). Loewen and Anderson (1990) have reported that balanced sitting score on admission (item 3), and MAS admission walking score (item 5) and bowel control scores (from the Barthel Index) were significant predictors of discharge walking function. These authors also found that integration of a second outcome measure, the Barthel Index, accounted for a greater proportion of the variability as it included impairments that are not measured by the MAS. It is apparent that prediction of walking function is difficult using a scale that measures only motor recovery, as the majority of patients score 0-1 on admission.
One limitation on the present study was that the MAS data utilised were collected by four different therapists over the time period under evaluation. The circumstances under which the data were collected in this investigation represent real life situations where frequently patients are assessed and treated by more than one therapist during their inpatient stay. An inter-rater reliability study was undertaken to investigate the impact that this may have had on the data utilised. The variations in scoring the five cases by four senior physiotherapists, while surprising, in part reflects the use of videoed cases for this procedure. In contrast to the study performed by Carr & Shepherd (1985), therapists did not receive any specific training and did not review the guidelines for scoring the MAS prior to the evaluation of video cases. Percentage agreements are considered to be a 'stringent measure of consistency' (Carr et al 1985). However it would have been unlikely that raters would have 100% percent exact agreements on scores made from video recordings of patients performing the movement tasks, compared to actual assessments. In a previous exploration of inter-rater reliability, Carr and Shepherd (1985) showed that balanced sitting (item 3) had the highest degree of consistency between raters; however the opposite was true for our data. Raters in the present investigation indicated that this was due to the difficulty of assessing equal weight distribution in sitting, based on the video camera angle. It was apparent from our data that some variation exists in the manner that the test is routinely administered, even by experienced staff. From these findings it might be recommended that formal training be undertaken by new staff using this scale and regular review of scoring consistency be undertaken if the MAS is to be considered to be a reliable tool for patient assessment across multiple therapists or rehabilitation settings. Wherever possible, investigations of this nature should use the same therapist to record all evaluations for a single patient or establish clearly inter-rater reliability between different evaluators prior to commencement of the data collection.
Attention to inter-rater reliability issues in clinical as well as in research settings is critical to obtain data that is useful for outcome prediction.
Another limitation of the present study was the use of retrospective data. It is recommended that further investigation of the value of specific stroke outcome measures for discharge prediction should use a prospective study design. Further research is needed into the effect that stroke complications and other comorbidities have on LOS and discharge destination, as these factors have been shown to be significant confounders in some previous investigations.
This study has provided further support of the utility of the Motor Assessment Scale to measure and predict outcomes after stroke using data from a large, representative cohort of patients in an Australian aged care stroke rehabilitation setting. Total MAS admission score, combined with age, could predict discharge destination with an overall accuracy of 76.2%. Total MAS and MAS mobility scores (items 1-5) on admission were only modestly correlated with eventual LOS, however total MAS scores and age significantly predicted the ability to walk independently on discharge. These data have potential to assist the rehabilitation physiotherapist with treatment planning as well as to inform patients and carers regarding likely outcomes post rehabilitation.
Thank you to Karen Joesbury for assistance with data analysis and to Leanne Cormack, Tracy Beckwith, and Carol Unkovich for their involvement in the set up of the database and data collection,
Aamodt G, Kjendahl A and Jahnsen R (2006): Dimensionality and scalability of the Motor Assessment Scale (MAS). Disability and Rehabilitation 28: 1007-1013.
Australian Institute of Health and Welfare (2008): Australia's Health 2008. The Eleventh Biennial Health Report of the Australian Institute of Health and Welfare. Canberra: AIHW.
Bamford J, Sandercock P, Dennis M, Burn J and Warlow C (1991): Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet 337: 1521-1526.
Barak S and Duncan P (2006): Issues in selecting outcome measures to assess functional recovery after stroke. NeuroRx 3: 505-524.
Bohannon R, Lee N and Maljanian R (2002): Postadmission function best predicts acute hospital outcomes after stroke. American Journal of Physical Medicine and Rehabilitation 81: 726-730.
Brauer S, Bew P, Kuys S, Lynch M and Morrison G (2008): Prediction of discharge destination after stroke using the motor assessment scale on admission: a prospective, multisite study. Archives of Physical Medicine and Rehabilitation 89: 1061-1065.
Brock K, Goldie P and Greenwood K (2002): Evaluating the effectiveness of stroke rehabilitation: choosing a discriminative measure. Archives of Physical Medicine and Rehabilitation 83: 92-99.
Carr J and Shepherd R (1998): Neurological Rehabilitation: Optimising Motor Performance. (2nd ed.) Edinburgh: Butterworth Heinemann.
Carr J, Shepherd R, Nordholm L and Lynne D (1985): Investigation of a new motor assessment scale for stroke patients. Physical Therapy 65: 175-180.
Dean C and Mackey F (1992): Motor assessment scale score as a measure of rehabilitation outcome following stroke. Australian Journal of Physiotherapy 38: 31-35.
English K, Hillier S, Stiller K and Warden-Flood A (2006): The sensitivity of three commonly used outcome measures to detect change amongst patients receiving inpatient rehabilitation following stroke Clinical Rehabilitation 20: 52-55.
Galski T, Bruno R, Zorowitz R and Walker J (1993): Predicting length of stay, functional outcome, and aftercare in the rehabilitation of stroke patients. The dominant role of higher-order cognition. Stroke 24: 1794-1800.
Hakim E and Bakheit A (1998 ): A study of the factors which influence the length of ho spital stay of stroke patients. Clinical Rehabilitation 12: 151-156.
Hill K, Ellis P, Bernhardt J, Maggs P and Hull S (1997): Balance and mobility outcomes for stroke patients: a comprehensive audit. Australian Journal of Physiotherapy 43: 173-180.
Kalra L, Perez I, Gupta S and Wittink M (1997): The influence of visual neglect on stroke rehabilitation. Stroke 28: 1386-1391.
Lew H, Lee E, Date E and Zeiner H (2002): Influence of medical comorbidities and complications on FIM change and length of stay during inpatient rehabilitation. American Journal of Physical Medicine and Rehabilitation 81: 830-837.
Loewen S and Anderson B (1990): Predictors of stroke outcome using objective measurement scales. Stroke 21: 78-81.
Lutz B (2004): Determinants of discharge destination for stroke patients. Rehabilitation Nursing 29: 154-163.
Malouin F, Pichard L, Bonneau C, Durand A and Corriveau D (1994): Evaluating motor recovery early after stroke: comparison of the Fugl-Meyer Assessment and the Motor Assessment Scale. Archives of Physical Medicine and Rehabilitation 75: 1206-1212.
Massucci M, Perdon L, Agosti M, Celani M, Righetti E, Recupero E, Todeschini E, Franceschini M and Italian Cooperative Research (ICR2) (2006): Prognostic factors of activity limitation and discharge destination after stroke rehabilitation. American Journal of Physical Medicine and Rehabilitation 85: 963-970.
McKenna K, Tooth L, Strong J, Ottenbacher K, Connell J and Cleary M (2002): Predicting discharge outcomes for stroke patients in Australia. American Journal of Physical Medicine and Rehabilitation 81: 47-56.
National Stroke Foundation (2008) http://www.strokefoundation. com.au/facts-figures-and-stats [accessed February 17, 2010]
Poole J and Whitney S (1988): Motor assessment scale for stroke patients: concurrent validity and interrater reliability. Archives of Physical Medicine and Rehabilitation 69: 195-197.
Salter K, Jutai J, Teasell R, Foley N and Bitensky J (2005): Issues for selection of outcome measures in stroke rehabilitation: ICF Body Functions. Disability and Rehabilitation 27: 191207.
Senes S (2006). How we manage stroke in Australia. AIHW cat. no. CVD 31. Canberra, Australian Institute of Health & Welfare.
Shah S, Vanclay F and Cooper B (1989): Predicting discharge status at commencement of stroke rehabilitation. Stroke 20: 766-769.
Tooth L, McKenna K, Goh K and Varghese P (2005): Length of stay, discharge destination, and functional improvement: Utility of the Australian National Subacute and Nonacute Patient Casemix Classification. Stroke 36: 1519-1525.
Wee J, Wong H and Palepu A (2003): Validation of the Berg Balance Scale as a predictor of length of stay and discharge destination in stroke rehabilitation. Archives of Physical Medicine and Rehabilitation 84: 731-735.
Claire Tucak BSc (Physio) (Hons),
Senior Physiotherapist, Osborne Park Hospital, Perth, Australia
Jennifer Scott, B Physio (Hons),
Physiotherapy, School of Health Sciences, University of Notre Dame Australia
Alison Kirkman, BSc (Physio),
Lecturer, Physiotherapy, School of Health Sciences, University of Notre Dame Australia
Barbara Singer, Dip PT, MSc, PhD
Associate Professor, Physiotherapy, School of Health Sciences, University of Notre Dame Australia
ADDRESS FOR CORRESPONDENCE
Alison Kirkman, Division of Physiotherapy, School of Health Sciences, The University of Notre Dame, Fremantle, WA 6959, Australia, Email: firstname.lastname@example.org; Phone: +61 8 94330902; Fax: +61 8 94330210
Table 1: Classification of Stroke in the study * Stroke Classification Features Total anterior cerebral Deficits of higher cerebral function, infarct (TACI), homonymous hemianopia and ipsilateral hemiparesis with or without sensory loss. Partial anterior cerebral Only two features of TACI present infarct (PACI), Lacunar anterior cerebral Pure or combined sensorimotor infarct (LACI) deficit, ataxic hemiparesis or acute movement disorder without disturbances of higher cerebral function or consciousness. Posterior circulation Presence of signs of brainstem, infarct (POCI), cerebellum, vertebrobasilar or occipital lobe dysfunction, including ipsilateral cranial n erve palsy, bilateral motor and/or sensory deficit, disorder of conjugate eye movement, cerebellar dysfunction without ipsilateral long tract deficit or isolated hemianopia or cortical blindness. * after Bamford et al Table 2: Demographics of study sample n= 239 Characteristic Number Mean+SD Age 78.1 [+ or -] 7.1 Males 125 76.2 [+ or -] 6.9 Females 114 80.1 [+ or -] 6.8 Side of stroke Left 115 Right 118 Bilateral 6 Type of stroke Ischaemic 193 TACI 25 PACI 105 POCI 33 LACI 30 Haemorrhagic 46 Length of Stay (LOS) LOS Total 50.4 days [+ or -] 26.8 days LOS Acute 17.9 days [+ or -] 12.4 days LOS SRU 32.6 days [+ or -] 20.74 days MAS admission score <20 79 20-29 48 30-39 75 40+ 37 Discharge Destination Home Independent 104 Home with carer 65 Institution 70 Other hospital 13 Slowstream rehab 22 Hostel 14 Nursing home 21 Characteristic Percentage Age Males 52.3% Females 47.7% Side of stroke Left 48.1% Right 49.6% Bilateral 2.5% Type of stroke Ischaemic TACI 13.0% PACI 54.4% POCI 17.1% LACI 15.5% Haemorrhagic Length of Stay (LOS) LOS Total LOS Acute LOS SRU MAS admission score <20 33.1% 20-29 20.1% 30-39 31.1% 40+ 15.5% Discharge Destination Home Independent 43.5% Home with carer 27.2% Institution 29.3% Other hospital 5.4% Slowstream rehab 9.2% Hostel 5.9% Nursing home 8.8% Table 3: Associations between total MAS admission scores and discharge destination, and individual MAS items and discharge destination [chi MAS Item on Admission square] df P 1--Supine to side lying 35.43 2 0.001 2--Supine to sitting over side of bed 64.04 2 0.001 3--Balanced Sitting 50.22 2 0.001 4--Sit to Stand 33.48 2 0.001 5--Walking 18.02 2 0.001 Mobility--Items 1-5 56.09 4 0.001 Total MAS scores 62.22 6 0.001 Table 4: Prediction of discharge destination using total MAS admission scores (a,b,c) Variable B Standard P Coefficient Error MAS admission score -0,109 0,016 0,001 Age 0,064 0,001 (a) Discharge destination dichotomised as either home (with or without support) or discharge to an institution (b) adjusted [R.sup.2]= 0,249 (c) non-significant variables: gender, type, side and stroke classification, Table 5: Actual versus predicted discharge destinations Home with/ without PREDICTED ACTUAL care Institution % Correct Home with/ without care 146 23 86,4 Institution 34 36 51,4 Overall % 76,2 Table 6: Prediction of the likelihood of walking on discharge (a) using total MAS admission scores (a,b,c) Variable B Standard P Coefficient Error MAS admission score 0,146 0,026 0,001 Age -0,094 0,026 0,001 (a) Walking defined as MAS item 5 score with assistance, criterions 1-3 or without assistance, criterions 4-6 (b), adjusted R2= 0,3851 (c), non-significant variables: gender, type, side and stroke classification,
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Research Report|
|Author:||Tucak, Claire; Scott, Jennifer; Kirkman, Alison; Singer, Barbara|
|Publication:||New Zealand Journal of Physiotherapy|
|Date:||Mar 1, 2010|
|Previous Article:||The future of the New Zealand Journal of Physiotherapy.|
|Next Article:||Effect of continuous passive motion following total knee arthroplasty on knee range of motion and function: a systematic review.|