To explore whether large item sets, as used in item banking
, enable important latent traits, such as driving
, to form stand-alone measures.
The 88-item activity limitation (AL) domain of the glaucoma
module of the Eye-tem Bank
was interviewer-administered to patients with glaucoma
. Rasch analysis was used to calibrate all items in AL domain on the same interval-level scale and test its psychometric properties. Based on Rasch dimensionality metrics, the AL scale was separated into subscales. These subscales underwent separate Rasch analyses to test whether they could form stand-alone measures. Independence of these measures was tested with Bland and Altman (B&A) Limit of Agreement (LOA).
The AL scale was completed by 293 patients (median age, 71 years). It demonstrated excellent precision (3.12). However, Rasch analysis dimensionality metrics indicated that the domain arguably had other dimensions which were driving
, luminance, and reading. Once separated, the remaining AL items, driving
and luminance subscales, were unidimensional and had excellent precision of 4.25, 2.94, and 2.22, respectively. The reading subscale showed poor precision (1.66), so it was not examined further. The luminance subscale demonstrated excellent agreement (mean bias, 0.2 logit; 95% LOA, −2.2 to 3.3 logit); however, the driving
subscale demonstrated poor agreement (mean bias, 1.1 logit; 95% LOA, −4.8 to 7.0 logit) with the AL scale.
These findings indicate that driving
items in the AL domain of the glaucoma
module were perceived and responded to differently from the other AL items, but the reading and luminance items were not. Therefore, item banking
enables stand-alone measurement of driving
ability in glaucoma