Beth Shinkins, now part of the CEBM staff, recently passed her DPhil (PhD) – a great excuse to get baking and celebrate! Beth’s area of expertise is diagnostics and so the “two by two table” Battenberg cake fitted the occasion perfectly.
Two by two tables are used to present data from diagnostic accuracy studies. The columns represent whether the patient has the condition or not, whilst the rows report their test result.
Patients with and without the condition who are correctly classified are known as true positives (TP) and true negatives (TN) respectively. Patients who score a negative test despite actually having the condition are false negatives (FN). And patients without the condition that have a positive test, are – well, you can work it out! From the two by two table, all sorts of “interesting” stats can be calculated…
Beth’s research looked into the limitations of presenting diagnostic accuracy in this binary way which glosses over many real life scenarios, for example when a test result is inconclusive or uninterpretable. And so, in Beth’s honour, the 3×3 Battenberg was created complete with grey bits to represent diagnostic uncertainty.
If all this has wetted your appetite for diagnostics and you want to learn more, CEBM runs a dedicated Evidence-Based Diagnosis and Screeningmodule in January. And there are a few places left for the new biomarkers workshop at the end of the month.