May 2006
Volume 47, Issue 13
Free
ARVO Annual Meeting Abstract  |   May 2006
Validity and Repeatability of Computerized and Paper–Based Subjective Analogue Scales
Author Affiliations & Notes
  • C.A. Woods
    CCLR, School of Optometry, University of Waterloo, Waterloo, ON, Canada
  • B.B. Cumming
    CCLR, School of Optometry, University of Waterloo, Waterloo, ON, Canada
  • T. Simpson
    CCLR, School of Optometry, University of Waterloo, Waterloo, ON, Canada
  • Footnotes
    Commercial Relationships  C.A. Woods, None; B.B. Cumming, None; T. Simpson, None.
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science May 2006, Vol.47, 102. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      C.A. Woods, B.B. Cumming, T. Simpson; Validity and Repeatability of Computerized and Paper–Based Subjective Analogue Scales . Invest. Ophthalmol. Vis. Sci. 2006;47(13):102.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: : Using subjective analogue scales to determine a rating for subjective symptoms is of value during clinical trials. These scales are usually comprised of a line on which the subject places a mark to indicate their opinion of the variable in question, e.g. degree of comfort or degree of ocular redness. These scales could be delivered using a computer and thus via the internet to facilitate home–based assessments; however, it would not be possible to control screen resolution and thus the length of the line in these situations. Would variability in line length for the scales influence accuracy and reproducibility?

Methods: : 20 subjects were asked to mark predetermined values on a non demarcated horizontal line, repeating the exercise with two different line lengths (6cm and 10 cm) using a paper–based scale as well as a computer–based scale accessed with their own desktop computer. The lines were converted to a 1 to 100 scale.

Results: : The mean difference between the two line lengths was 0.33 units. The coefficient of repeatability (2SD of the differences) was 5.80 units, and the 95% confidence interval (2SE of the difference) was 0.42 units. These results demonstrate high repeatability (5.80) as this lies within what we considered to be a clinically significant change of 10 units and no bias (0.33 ± 0.42). The regression equation between the values specified (x) and recorded (y) for the paper scales was y=0.71+0.97x (r=0.99) and for the computer by y=–2.52+1.02x (r=0.99). For both functions the slope did not differ statistically from 1.0 (p>0.05).

Conclusions: : Discrepancy analysis yielded no significant difference and no bias between the different line lengths, suggesting a high degree of repeatability. Regression analysis demonstrated that varying the line length is valid regardless of the medium used. These results suggest that the length of the line does not influence the result, indicating that computers can be used for subjective assessment regardless of screen resolution.

Keywords: clinical research methodology • contact lens • conjunctiva 
×
×

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.

×