Table 2

Sensitivity and specificity of cut-off points in CV risk scores

True cases (n)Positive tests (n)True positive (n)False positive (n)False negative (n)True negative (n)TotalSensitivity (%)Specificity (%)PPV (%)NPV (%)
SCORE
 >10%1044107933125615105076651996
 >20%1042476318441762105061802695
FRS
 >10%14948713035719544105087602797
 >20%1491448816161740106059823592
RRS
 >10%872845925528694100668762196
 >20%871443311154808100640882394
QRisk II
 >10%12651610641020514105084552196
 >20%1262988221644708105065772894
  • True cases represent all CV events that occurred, using the criteria for each individual risk score. The number of positive tests are the number of patients who are classified as being at intermediate-high or high risk (a risk score of >10% or >20%, respectively) for CVD, out of the total number of patients with RA. As the RRS is not applicable to patients with diabetes, these patients (n=44) are excluded, leaving a total of 1006 patients for analysis. Sensitivity is calculated as the proportion of correctly classified patients given a positive result on the outcome variable, that is, proportion of patients with a predicted risk >10% or >20% out of all the patients who experienced an event (true positive). Specificity is calculated as the proportion of correctly classified patients given a negative result on the outcome variable, that is, proportion of patients with a predicted risk <10% or <20% out of all the patients who did not have an event (true negative).

  • CV, cardiovascular; CVD, cardiovascular disease; FRS, Framingham risk score; NPV, negative predicting value; PPV, positive predicting value; RA, rheumatoid arthritis; RRS, Reynolds risk score; SCORE, Systematic Coronary Risk Evaluation.