Assessment of risk for venous thromboembolism in hospital emergency departments: an analysis of concordance between scales

Background and objective: Little information is available on differences between scales

used to assess risk for venous thromboembolism (VTE) in patients with medical

conditions admitted from hospital emergency departments. We aimed to evaluate the

agreement between 2 risk prediction tools: a scale based on the PRETEMED guideline

and the Padua Prediction Score (PPS).

Methods: Prospective multicenter cohort study in 6 Spanish hospital emergency

departments. Patients with medical conditions requiring hospital admission were

included between December 2011 and July 2012. The concordance correlation between

the 2 scales was calculated. The development of VTE was registered for 90 days.

Results: Of 610 patients recruited, data for 580 patients (mean [SD] age, 70.1 [16.9]

years; 45.3% women) were valid for assessment. Patients classified as having moderate

or high risk numbered 256 (44.1%) with the PRETEMED scale and 368 (63.4%) with the

PPS. The κ index of concordance between the 2 scales was 0.39 (95% CI, 0.32–0.46)

(discordance, 31.3%). Among patients who did not receive thromboprophylaxis, the

incidence of VTE was 2.3% among the 129 patients the PRETEMED scale classified as

being at moderate or high risk and 3.5% among patients at high risk according to the

PPS.

Conclusions: There is disagreement between PRETEMED and PPS scale assessments of

risk for VTE in patients with medical conditions. The PPS classified more patients as being

at high risk, and the predictions were more accurate based on their later development of

VTE.