Investigations, Therapeutics and Patient Safety
What the domain tests
Items ask you to choose the next best investigation, interpret a result in context, prescribe safely and recognise systems errors. The recurring theme is that a test is useful only if it changes management.
Core rules
- Sensitivity and specificity are test properties; predictive values are patient properties. A highly sensitive test rules out disease when negative (SnNout); a highly specific test rules in disease when positive (SpPin). Positive predictive value rises with prevalence.
- Therapeutic index and clearance determine safety. Drugs with narrow therapeutic indices, such as warfarin, digoxin, lithium and aminoglycosides, need monitoring; renal clearance matters for many drugs.
- Patient safety is systematic. Medication reconciliation, allergy documentation, weight-based dosing and the surgical safety checklist prevent predictable errors.
- Report and learn. Incident reporting is non-punitive; root-cause analysis addresses systems rather than individuals.
Worked example
A 70-year-old woman on stable warfarin with an INR of 2.5 is prescribed fluconazole for oral candidiasis. Reasoning: fluconazole inhibits CYP2C9 and CYP3A4, reducing warfarin metabolism. The INR will rise and bleeding risk increases. Check the INR within 3-5 days, reduce the warfarin dose if necessary, and counsel about bleeding. This is a predictable pharmacokinetic interaction, not an idiosyncratic reaction.
Traps
- Ordering a test that will not alter management.
- Confusing sensitivity with positive predictive value.
- Prescribing without checking renal function, allergies and interactions.
- Blaming an individual for what is actually a systems error.