This study was designed to develop the Korean Activities of Daily Living (ADL) evaluation tools for persons with stroke applying the Rasch model based on item response theory. The total subjects for this study were 477 stroke patients, 308 inpatients ...
This study was designed to develop the Korean Activities of Daily Living (ADL) evaluation tools for persons with stroke applying the Rasch model based on item response theory. The total subjects for this study were 477 stroke patients, 308 inpatients and 169 outpatients, at 18 different rehabilitation facilities in Korea. Questionnaires containing ADL evaluation items selected in advance with the Q-sorting method from the preexisting ADL related literatures were distributed to the subjects by mail or visit. Rasch analysis was then done by means of the Winsteps program to determine the validity and reliability of the Korean ADL evaluation tools for persons with stroke. The results were as follows :$$a$$a1. In case of inpatients, three items were found to be misfits and the order of item difficulty of the remaining 20 items was arranged. Stroke inpatients'' ADL ability is indicated by -6.77~6.73 logit, and the transformation formula is “score=(logit score+6.77)/(6.77+6.73)×100.”$$a$$a2. In case of outpatients, seven items were found to be misfits and the order of item difficulty of the remaining 16 items was arranged. Stroke outpatients'' ADL ability is indicated by -6.32~6.23 logit, and the transformation formula is “score=(logit score+6.32)/(6.32+6.23)×100.”$$a$$a3. In the order of difficulty of evaluation items, inpatients and outpatients responded equally. The most difficult item was “going up and down stairs,” which was one from the functional mobility domain. And the easiest item was “drinking,” which was one from the feeding domain.$$a$$aIn conclusion, the Korean ADL evaluation tools for persons with stroke have been proven reliable and valid. Therefore they are expected to be used as effective tools for treatment planning and intervention for persons with stroke.$$a$$a