Enter the reverend: introduction to and application of Bayes' theorem in clinical ophthalmology

Abstract

Knowledge of Bayes' theorem helps in interpreting test results and supports the clinical teaching that testing for conditions with a low prevalence has a poor predictive value. In some clinical situations Bayes' theorem can be used to calculate vital data required for patient management.

Review of null hypothesis significance testing in the ophthalmic literature: are most ‘significant’ P values false positives?

“…While much debate exists regarding the value of Bayesian reasoning, arguably it is more consistent with the epistemology of science, as knowledge cannot arise through experimentation alone but depends also on prior data as well as subjective knowledge, experience and judgement. The application of Bayesian methods in ophthalmic research has yet to become established but inroads are being made in an attempt to popularize the techniques Finally, it has been suggested to curtail usage of the term ‘significant’ due to its arbitrary and misleading implications .…”

Looking deeper than (just) below the surface: response

“…This should occur with support from journal editorial staff and the efforts of knowledgeable participants such as Thomas et al who are prepared to highlight alternate methodology in clear, practical terms. 4,5 Perhaps, a prudent observation of the difficulty in introducing statistical methodology appears in the Lisboa article; Bayes theorem represented only 0.5% of all statistical methods used within the duration of the review.…”

Visual symptoms following iridotomy

“…In contrast to conventional frequentist analysis and P values, Bayesian analysis permits incorporation of prior data with current data to update evidence, and provides direct probability statements about any differences that might be of clinical interest …”