Psychometric evaluation of a new proxy‐instrument to assess participation in children aged 3–6 years with visual impairment: PAI‐CY 3‐6. (29th August 2019)

• Record Type: Journal Article
• Authors: Elsman, Ellen B M; van Nispen, Ruth M A; van Rens, Gerardus H M B
• Is Part Of: Ophthalmic and physiological optics. Volume 39:Number 5(2019:Sep.)
• Page Start: 378
• Page End: 391
• Publication Date: 2019-08-29
• Subjects: children -- item response theory -- low vision -- participation -- questionnaire -- validation; Ophthalmology -- Periodicals; Physiological optics -- Periodicals; Optometry -- Periodicals; Optics -- Periodicals; Vision -- Periodicals; 617.75
• DOI: 10.1111/opo.12642
• Languages: English
• ISSNs: 0275-5408

Abstract

Abstract: Purpose: The Participation and Activity Inventory for Children and Youth 3–6 years (PAI‐CY 3‐6) was recently developed to assess the participation needs of children with visual impairment (VI) by means of parent‐proxy report. This study reports on its psychometric properties. Methods: Parents of children aged 3–6 years registered at two low vision rehabilitation centers in the Netherlands were invited to participate and completed the 52‐item PAI‐CY. Satisfaction with the PAI‐CY 3‐6 was determined using an evaluation form. Basic item analyses was conducted, after which an item response theory (IRT) model (i.e. the graded response model, GRM) was fitted. Deletion of items was informed by results of item analyses, fulfillment of IRT assumptions, differential item functioning, fit to the GRM and item information content. Face and content validity were considered; professionals from low vision rehabilitation centers were asked for their opinion in the item deletion process. After obtaining a satisfactory set of items, known‐group validity, concurrent validity and test‐retest reliability were also investigated. Results: Data of 237 parents were included in the analyses. Various aspects of the PAI‐CY 3‐6 were perceived as neutral to positive by over 85% of the respondents. After removing 17 items, the remaining 35 items reflected satisfactory fit to the GRM. Known‐group validity was supported, since participants with more severe VI and comorbidity scored significantly worse than those with less severe VI and without comorbidity after correcting for potential confounders. Test‐retest reliability was adequate, and the PAI‐CY showed to have good concurrent validity. Feedback from professionals motivated the maintenance of 3 of the 17 deleted items, although not included in the scoring. Furthermore, two new items were added, resulting in a 40‐item instrument. Conclusions: The PAI‐CY 3‐6 has sound psychometric properties and can now be used to assess the participation needs of children aged 3–6 years with VI by means of proxy. Implementation in routine low vision rehabilitation care enables further optimization and investigation of its acceptability and feasibility.