Evidence-Based Ophthalmology and Ethics
This section covers evidence-based ophthalmology and ethics: study design, diagnostic test interpretation, bias, consent, and resource allocation. The exam tests numerical reasoning and professional judgement.
Core rules and concepts
- Randomised controlled trials best assess treatment effect; cohort studies assess prognosis, and case-control studies suit rare diseases.
- Sensitivity is the proportion of diseased patients who test positive; specificity is the proportion of non-diseased who test negative.
- Positive predictive value depends on prevalence: it falls when disease is rare, even with high sensitivity and specificity.
- Number needed to treat is the reciprocal of absolute risk reduction; confidence intervals show precision.
- Ethical practice requires informed consent, confidentiality, avoidance of conflicts of interest, and honest reporting of outcomes.
Worked example
A screening test has 90% sensitivity and 95% specificity. Prevalence is 1%. In 10,000 people, 100 have disease; 90 test positive and 10 test negative. Of 9,900 without disease, 495 test falsely positive. PPV is 90/(90+495)=15.4%. Most positives are false positives. This explains why screening rare diseases needs confirmatory testing and why prevalence must be stated when interpreting a test.
Traps that cost marks
Confusing sensitivity with PPV is the commonest calculation error. A highly sensitive test is useful for ruling out disease, not for confirming it. Do not ignore prevalence when comparing tests. Ethical marks are lost by breaching confidentiality or failing to declare a conflict of interest.