Skip to main content

Table 1 Diagnostic accuracy definitions

From: Accuracy of risk scales for predicting repeat self-harm and suicide: a multicentre, population-level cohort study using routine clinical data

Sensitivity The proportion of episodes that were followed by a repeat self-harm episode and were correctly identified by the scale as high risk
Specificity The proportion of episodes that were not followed by a repeat self-harm episode and were correctly identified by the scale as low risk
Positive predictive value The probability that the episode identified as high risk by the scale was followed by repeat self-harm
Negative predictive value The probability that the episode identified as low risk by the scale was not followed by repeat self-harm
Positive likelihood ratio The increased likelihood of a high-risk scale result in an episode that is followed by repeat self-harm versus one that is not
Negative likelihood ratio The decreased likelihood of a low-risk scale result in an episode that is followed by repeat self-harm versus one that is not
Diagnostic odds ratio The odds of a high-risk result in an episode that is followed by repeat self-harm versus one that is not (interpreted the same as an odds ratio)
Receiver operating characteristic (ROC) curve Graphically shows the overall discrimination ability of a scale to identify episodes that were followed by a repeat self-harm episode compared with those that did not at various cut-off points (plotted as sensitivity versus 1-specifcity). The performance of the scale is indicated by the calculation of the area under the curve (AUC). Higher AUC indicate greater discriminatory power.