March 2012
Volume 53, Issue 14
Free
ARVO Annual Meeting Abstract  |   March 2012
The Impact of Item Filtering On Patient-reported Low Vision Rehabilitation Outcomes
Author Affiliations & Notes
  • Judith E. Goldstein
    Ophthalmology, Johns Hopkins University, Baltimore, Maryland
  • Robert W. Massof
    Ophthalmology, Johns Hopkins University, Baltimore, Maryland
  • Low Vision Research Network
    Ophthalmology, Johns Hopkins University, Baltimore, Maryland
  • Footnotes
    Commercial Relationships  Judith E. Goldstein, None; Robert W. Massof, None
  • Footnotes
    Support  EY012045 and Reader’s Digest Partners for Sight Foundation
Investigative Ophthalmology & Visual Science March 2012, Vol.53, 4419. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Judith E. Goldstein, Robert W. Massof, Low Vision Research Network; The Impact of Item Filtering On Patient-reported Low Vision Rehabilitation Outcomes. Invest. Ophthalmol. Vis. Sci. 2012;53(14):4419.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract
 
Purpose:
 

Evaluate the effects of item filtering on measures of patient-reported low vision rehabilitation outcomes.

 
Methods:
 

The 510-item adaptive Activity Inventory (AI) was administered by telephone to new low vision patients before and after low vision rehabilitation (LVR) at 28 collaborating clinical centers. The AI was administered prior to the initial low vision evaluation and 6 to 9 months after usual LVR. To be eligible, patients had to be new to the practice, over 18 years of age, and able to complete the telephone interview. Rasch analysis was performed on the AI difficulty ratings both at baseline and 6-9 months post rehabilitation with item measures anchored to baseline values. Two methods of calculating change in average functional reserve (AI Δ score) were compared. The unfiltered method included difficulty ratings of all important goal activities and all tasks under important and difficult goals, irrespective of item difficulty. The filtered method censored items from the analysis if they were rated not difficult at baseline. For both methods, AI Δ scores were expressed as effect sizes (Cohen’s d).

 
Results:
 

Both baseline and post-rehabilitation measures were obtained on 446 patients. Outcome measures for different functional domains were estimated from responses to different subsets of items. For all domains except reading, the effects of LVR were significantly greater for AI Δ scores estimated using the filtered method than for the unfiltered method. Reading was the only domain for which a significant effect of LVR was observed using the unfiltered method.

 
Conclusions:
 

LVR aims to increase functional reserve for items that describe important and difficult activities identified by the patient. The comparison of effect sizes with and without item filtering demonstrates that when items not addressed by LVR are included, average change in functional reserve is diluted by the non-targeted items, which results in an underestimation of the effectiveness of intervention.  

 
Keywords: clinical (human) or epidemiologic studies: outcomes/complications • low vision • aging: visual performance 
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×