April 2010
Volume 51, Issue 13
ARVO Annual Meeting Abstract  |   April 2010
Effectiveness of Population-Based Vision Screening at Different Ages
Author Affiliations & Notes
  • S. Schutte
    BioMechanical Engineering, Delft University of Technology, Delft, The Netherlands
  • F.-W. Goudsmit
    Public Health, Erasmus MC: University Medical Centre, Rotterdam, The Netherlands
  • H. de Koning
    Public Health, Erasmus MC: University Medical Centre, Rotterdam, The Netherlands
  • H. J. Simonsz
    Ophthalmology, EMC, Rotterdam, The Netherlands
  • Footnotes
    Commercial Relationships  S. Schutte, None; F.-W. Goudsmit, None; H. de Koning, None; H.J. Simonsz, None.
  • Footnotes
    Support  None.
Investigative Ophthalmology & Visual Science April 2010, Vol.51, 4356. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      S. Schutte, F.-W. Goudsmit, H. de Koning, H. J. Simonsz; Effectiveness of Population-Based Vision Screening at Different Ages. Invest. Ophthalmol. Vis. Sci. 2010;51(13):4356.

      Download citation file:

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

  • Supplements

In preparation of a disinvestment study of population-based screening of amblyopia we developed a model to evaluate the effectiveness of the current screening program in the Netherlands, which consists of 7 screens, using data of the 7-year (N=4624) birth-cohort study RAMSES, which considered 6 of the 7 screens. We determined the effectiveness of the screens at different ages.


The input of the model was the number of true positives per screen and the numbers of existing amblyopes (n=100 of 2964) found at age 7 in RAMSES, differentiated for amblyopia type. As in RAMSES only the age of detection of amblyopia was determined and not the age of onset, this had to be estimated. We assumed that amblyopia development started after 3 months and made optimistic and pessimistic predictions about the development rate of amblyopia, resulting in upper and lower boundaries for prevalence curves. With the prevalence-curves and the number of detected amblyopia cases the sensitivity per screen could be calculated. The effect of occlusion treatment was assumed to be linearly declining with age, from 1 (=100%) at time of development to 0 at the time treatment has no effect. Together with the prevalence and sensitivity we calculated the effect per screen. The simulation was repeated under exclusion of the least effective screens.


The sensitivity of the screens at 14 and 24 months was lower than for screens at other ages. The screen at age 3 was most effective (Fig). This is caused by the ability to measure visual acuity from age 3. The overall effect of the screening program remained similar after excluding 1 or 2 screens around age two.


Based on the results of this simulation model, the screen at age 2 will be omitted in the disinvestment study. The cost savings will be reallocated to a closed-loop referral procedure after positive screening, as in RAMSES, it was found that 23% of the positively screened children had been referred unsuccessfully.  

Keywords: screening for ambylopia and strabismus • amblyopia 

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.