Use this URL to cite or link to this record in EThOS:
Title: Improving the measurement of visual acuity in clinical practice and clinical research
Author: Rosser, Daniel Ashley
ISNI:       0000 0001 3538 3134
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2005
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This thesis considers the relationship between the design of a visual acuity test and various aspects of its performance. Using contemporary test design theory, novel tests are developed and evaluated in an attempt to better meet the requirements of a visual acuity test most pertinent to clinical practice, clinical research, and population based surveys. The acuity test of choice in clinical practice is the Snellen chart, a test whose usefulness is limited by several design flaws. Clinical researchers favour the ETDRS logMAR chart which employs robust design principles, but is time consuming to use. A chart featuring an abbreviated ETDRS design was developed and its performance compared with that of the ETDRS and Snellen charts. The prototype chart allows acuities to be measured in half the time of the ETDRS chart with greater precision than the Snellen chart. A tumbling-E version of this chart has been successfully employed in population based surveys in Thailand, Bangladesh and Mongolia. It was noted during the study that the precision of even ETDRS acuities was relatively poor. A computerised version of the ETDRS test was developed and used to investigate the repeating and averaging of acuities as a means to improve precision. Whilst prolonging test time, the computerised test allowed acuities to be measured with improved precision. Optical defocus was investigated as a potential source of reduced precision in visual acuity testing. It was shown that even small degrees of defocus may significantly reduce test precision. An approach which considers test performance in terms of sensitivity and specificity was developed. A mathematical model was used to show that current methods of using estimates of precision to identify clinically important change, are overly optimistic. Predictions derived using the model were shown to agree well with empirical findings.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available