Zombie plot calculators

Michael L. Richardson, M.D.

Why would you want a zombie plot? Because it provides you with a rapid, visual way to tell whether a diagnostic test is worthwhile. It does this by dividing the space on an ROC plot into “zones of mostly bad imaging efficacy”, or ZOMBIE for short. It then plots the sensitivity, specificity and 95% CIs for your test in that space.

Figure 1. The zones of mostly bad imaging efficacy.
Figure 1. The zones of mostly bad imaging efficacy.

If life is good, the 95% confidence region of your test will be completely contained within the optimal zone in the upper left-hand corner of the plot. In this zone, you have reasonable assurance that your test will be useful for ruling in or ruling out disease.

If your test ends up contained in one of the green zones, that means it should be useful for either ruling in or ruling out disease, but not both.

If your test lies in the mediocre or perverse zones, then it is not a very credible test, and you should use something else.

Zombie Plot Calculators

Here are two, count ’em, two calculators for your delectation. Depending on what information you have available for your test, I’m pretty sure that you will be able to figure out which calculator to use:

  1. Use calculator 1 if you already know sensitivity and specificity and their 95% CIs.
  2. Use calculator 2 when you have a 2 x 2 table of the categorical data.

References

  1. Richardson ML. The zombie plot: a simple graphic method for visualizing the efficacy of a diagnostic test. AJR 2016 Aug 9:W1-W10. [Epub ahead of print]