1) Taking a benzodiazepine, phenothiazine, or antidepressant drug (positive likelihood ratio of 27)

2) Dementia (positive likelihood ratio of 17)

In other words, patients on the drugs listed above are 27 times more likely to fall in the next year than patients not on those drugs. Patients with reported dementia are 17 times more likely to fall in the next year than patients without dementia.

None of the diagnostic tests in the table show a negative LR that would significantly change the pre to post-test probability of a patient falling in the next year.

The reason that none of the diagnostic tests have a strong negative LR can be explained by looking at the negative LR equation once again.

In order for a negative LR to be significant the number must be small (<0.1), therefore the numerator (the false negative rate) must be small and the denominator (the true negative rate) must be large. For most of these diagnostic tests or markers, a large number of patients will not have the finding in question but will have a fall (False negative) therefore the false negative rate will be high and the negative LR will be a larger number.