Article Text

Download PDFPDF

4CPS-359 Development and validation of a 30 day revisit risk prediction model in patients admitted to the emergency department due to drug related problems
  1. J Ruiz Ramos,
  2. L Gras Martín,
  3. A Juanes Borrego,
  4. N Mas Malagarriga,
  5. D Medina Catalán,
  6. N Jorba Beltran,
  7. MA Mangues Bafalluy
  1. Hospital Santa Creu I Sant Pau, Pharmacy, Barcelona, Spain


Background and importance Drug related problems (DRPs) are an important cause of admission to the emergency department (ED), and one of the most frequently implicated drugs are those used for cardiovascular diseases. However, information regarding the risk factors associated with ED revisits is this group of patients is scarce.

Aim and objectives The aim of this study was to develop a predictive model of 30 day revisits to the ED in patients with a first visit for an episode of DRP.

Material and methods A retrospective cohort study was carried out including patients who attended an ED in 2019 due to DRPs caused by drugs classified in the ATC classification system as A, B and C. A 30 day prediction model was created in a derivation cohort using backward logistic regression. Those variables significant at p<0.100 in a multivariate analysis were assigned an integer score proportional to the regression coefficient. The model was then internally validated by k-fold cross validation and in the validation cohort.

Results 580 patients were included (mean age 80.0 (12.6) years) and 133 (22.9%) patients revisited the ED at day 30. Five independent risk factors (moderate to severe chronic kidney disease (5 points), previous ED visit within 3 months (6), high anticholinergic burden (8), DRPs related to heparin use (12) and safety DRPs (8)) were identified in the derivation cohort and were combined into an overall score. The model achieved an area under the curve–receiver operating curve of 0.71 (95% CI 0.66 to 0.75) in the derivation cohort and 0.70 (95% CI 0.65 to 0.74) in the validation cohort (p=0.273). Patients were classified into three risk categories (high, medium and low) and had the following rates of risk: 11.1% (0–6 points), 20.0% (8–13 points) and 39.5% (>13 points). Findings were similar in the validation cohort. The optimal cut-off point in the model was 9, having a sensitive of 67.09%, a specificity of 69.06%, a positive predictive value of 36.78%, and a negative predictive value of 87.61%.

Conclusion and relevance This score could be used by clinicians from the ED to identify those patients at high risk of 30 day revisits, and could be useful to design specific interventions at discharge in this group of patients.

Conflict of interest No conflict of interest

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.