Predictors of Functional Gains

Prognostic factors derived from studies that model recovery may allow clinicians to choose the most appropriate patients to place on an inpatient service, allows comparisons of individual patients with the average rate of gains over time, and offers a means to stratify patients for clinical trials.

The overall likelihood of achieving gains in ADLs by the FIM or BI over the first few months following a stroke runs in parallel to the level of impairment measured by a scale such as the Scandinavian, Canadian, or National Institutes of Health Stroke Scales. Less impairment is associated with less disability, especially in relation to the degree of hemiparesis. The higher the admission score on the BI or FIM, both of which are valid and reliable measure of burden of care, the higher the discharge score and the greater the likelihood that the patient will return to living at home. A BI score over 60 by the time of inpatient discharge is associated with living in the community 6 months later. No single score on the BI or FIM serves as a complete predictor.

Urinary incontinence is about as good a predictor of a poorer outcome as any grouping of impairments.197 Indeed, one classification tree approach to predicting outcome for inpatient rehabilitation found that the level of independence in toilet and bladder management and toilet transfers, along with adequacy of financial resources, best predicted community discharge, survival for more than 3 months after discharge, and no more than minimal physical asistance for ADLs.198 For patients under age 75 years who have rehabilitation admission FIM scores less than 37, 3 FIM items on admission (bladder management, toilet transfers, and memory), and 3 discharge FIM variables (upper body dressing, bed/chair transfers, and comprehension) predict discharge placement with 75% accuracy.74 Another study, which did not include an impairment measure, employed the BI to determine the average pattern of gains over weeks after the stroke.199 The presence of prestroke disability, urinary incontinence, dysarthria, and female sex were associated with lower BI scores throughout the time to reach maximal gains, whereas greater age, dysphasia, and weakness of any limb had varying effects over time. With this model, the predicted BI score at a given time was within 3 points of the observed score on the 20-point variation of the index approximately 70% of the time.

The admission score on the FIM may be used to anticipate the burden of care on the provider and the discharge placement. A classification and regression tree analysis revealed a mean rate of increase of 1.7 points per day across all patients regardless of admission FIM score. Thus, the efficiency or rate of improvement of inpatient gains in ADLs is independent of initial disability.97 The FIM motor score, however, does correlate with the daily FIM point gain (Table 9-12).

A retrospective American study of 464 non-hemorrhagic stroke patients admitted in 1991 to 1 inpatient rehabilitation unit after a mean acute hospitalization of 18 days examined many variables.71 Patients spent an average of 34 days at the facility. Admission FIM scores and age were analyzed in relationship to discharge FIM and placement. Admission FIM positively correlated with discharge FIM and admission FIM negatively correlated with length of stay. Lesser gains on the FIM followed a right, compared to a left, cerebral infarct in patients with the lowest admission FIM scores, less than 40. The greatest FIM change over time occurred in patients with admission FIM scores of 4 to 80. Patients with admission scores over 80 and age less than 55 years returned home. A score of less than 40 and age over 65 produced a nursing home discharge for 62%. For the rest of the FIM-age subgroups, only 13% went to nursing facilities.

A few predictors for functional outcome were found in the Copenhagen Stroke Study for patients with the most severe impairments who survive for at least 3 months.83 Decreasing age, having a spouse, and greater gains in neurologic impairments at 1 week predicted a higher BI score.


A study assessed quality of life (QOL) in 442 patients with a mean age of 73 years using the Sickness Index Profile 6 months after a stroke.200 The investigators found that 60% reported mildly diminished QOL, 33% reported more dysfunction on the physical and psychosocial dimensions, and 7% had marked psychosocial dysfunction. Comorbidities, severity of stroke, and supratentorial location produced the most impaired pattern of QOL. No differences were found for patients with intracerebral hemorrhage and supratentorial infarct. Patients with lacunar infarcts had significantly less dysfunction, with the exception of emotional discomfort, than patients with other subcortical lesions.


Metabolic and functional activation studies may reveal information relevant to functional recovery (see Chapter 3). For example, a higher global and contralateral cerebral metabolic rate for glucose within 1 to 2 weeks of an acute stroke correlated with a better functional outcome in survivors at a mean of 3 and 50 months.201 Low glucose consumption within the unaffected hemisphere in hypertensive patients was associated with poorer ADLs, perhaps because of a subclinical hypertensive ar-teriopathy producing tissue damage that limited compensation. Perilesion metabolic activation and activations associated with greater motor recovery tend to correlate with functional gains, but too few serial studies have been completed to establish meaningful predictors. Activations that change over the course of training, especially for a motor task or a working memory task, may come to serve as physiologic markers of the capacity for improvement in ADLs and the effectiveness of rehabilitative training.

Your Heart and Nutrition

Your Heart and Nutrition

Prevention is better than a cure. Learn how to cherish your heart by taking the necessary means to keep it pumping healthily and steadily through your life.

Get My Free Ebook

Post a comment