Computerized adaptive testing with the I-HaND scale for monitoring patients with upper limb nerve pathology.

Abstract

The Impact of Hand Nerve Disorders scale is a patient-reported outcome measure for upper limb nerve pathology. We aimed to assess its structural validity using item response theory and to develop computerized adaptive testing algorithms. We conducted a series of psychometric studies to assess constructs measured, applied an item response theory model to the data, then developed computerized adaptive testing algorithms. The results showed two distinct factors: 'Motor Function' and 'Pain'. We developed two separate computerized adaptive tests which reduced the number of questions from six ('Pain') and 16 ('Motor Function') to a median of three questions each at a high level of precision (standard error of measurement <0.3). For optimal measurement precision, the Impact of the Hand Nerve Disorders scale should be divided into two subscales and administered through computerized adaptive testing. In some circumstances (such as screening for specific issues) it may be preferable to administer all 32 items.