A prediction model for activities of daily living for stroke patients in a convalescent rehabilitation ward

Rather than useless crap research like this our stroke leaders should be focusing on SOLVING  the problems in stroke. Not just predicting the problems.
https://www.jstage.jst.go.jp/article/jalliedhealthsci/7/1/7_1/_article

Atsushi Sato¹⁾, Takaaki Fujita^{2) 3)}, Yuji Ohashi⁴⁾, Yuichi Yamamoto⁴⁾, Kyouichi Suzuki⁵⁾, Koji Otsuki⁴⁾

1) Department of Physical Therapy, Yachiyo Rehabilitation College 2) Department of Rehabilitation, Tohoku Fukushi University 3) Department of Rehabilitation Sciences, Gunma University Graduate School of Health Sciences 4) Department of Rehabilitation, Northern Fukushima Medical Center 5) Department of Neurosurgery, Fukushima Red Cross Hospital

Released 20160401  

Keywords: activities of daily living, convalescent rehabilitation ward, prediction model

• Full Text PDF [118K]

• Abstracts
• References(18)

PURPOSE: To create a model to predict independence in the activities of daily living at discharge in stroke patients in the convalescence stage. The study also examined whether the predictability of functional independence at discharge would be improved by creating a specific prediction model for each rehabilitation facility. METHODS: To create the prediction model, data of 65 first stroke patients were analyzed using stepwise multiple regression analysis. Age, time post-stroke, Functional Independence Measure motor subscale score, Functional Independence Measure cognitive subscale score, Stroke Impairment Assessment Set, Berg Balance Scale, and Vitality Index at admission were selected as predictor variables of Functional Independence Measure motor subscale score at discharge. The accuracy of this model was compared with an existing prognosis model using data from 98 first-stroke patients, comparing the difference between actual and predicted Functional Independence Measure motor subscale score at discharge for each model. RESULTS: The prediction formula created included admission Functional Independence Measure motor subscale score, Vitality Index, age, and Stroke Impairment Assessment Set score. The adjusted R square value was 0.60. The prediction errors of the new and previous models were −2.5 ± 10.8 and −18.3 ± 18.7, respectively, which were significantly different. CONCLUSION: Our results suggest that prediction accuracy may be improved by creating prediction formulas specifically for each institution.