Abstract
Purpose :
To test the assumption that visual performance with ultra-low vision (ULV) is similar regardless of vision modality (i.e native, retinal implant, sensory substitution) and can be measured using the ULV activities of daily living (ULV-ADL) assessment toolkit.
Methods :
Thirty-five participants were recruited for this study: 25 with native ULV – i.e., from natural causes, 4 with an Argus II retinal implant and 6 with a Brainport (sensory substitution device using tongue stimulation). The 17-item ULV-ADL toolkit was administered in their home environment. Tasks in the ULV-ADL are administered at 3 difficulty levels and include everyday activities such as detection of a towel on a rack, a moving cursor, a lit candle on a table, etc. Item and person measures were calculated from the ULV-ADL scores using Winsteps®.
Results :
The item measure distribution of the ULV-ADL items in native ULV participants ranged from -3.94 (less difficult to complete) to +2.24(more difficult to complete) logits. A narrowed difficulty range of -2.02 to 2.32 logits was obtained when scores from all participants were included. Person measures in our sample ranged from -2.02 to 5.35 logits. The Argus II and Brainport users were within the same range as the native ULV subjects. The Argus II users in our sample had lower ability compared to others while subjects with higher ability had larger standard errors (i.e. cruder estimates) since their ability exceeded the difficulty of the ULV-ADL items. A principal components analysis indicated that 56% of the variance was explained by the model, and showed no evidence of additional dimensions (largest unexplained component, 5.1%).
Conclusions :
Despite the small sample size, our data set confirms the assumption that native, prosthetic and substitute ULV all exhibit similar behavior and can be measured using the ULV-ADL toolkit. This demonstrates the instrument's potential across the ULV rehabilitation field, irrespective of the cause of the ultra low vision. We intend to continue examining this in a larger sample and compare person measures from our self-report questionnaire (ULV-VFQ) and the ULV-ADL.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.