eCite Digital Repository

Development of a Score to Predict Hospitalisation due to Adverse Drug Reactions in Older Patients

Citation

Nair, N and Chalmers, L and Bereznicki, BJ and Castelino, RL and Peterson, GM and Curtain, CM and Connolly, MJ and Bereznicki, LRE, Development of a Score to Predict Hospitalisation due to Adverse Drug Reactions in Older Patients, 2015 Joint APSA-ASCEPT Annual Conference, 29 November - 2 December, 2015, Hobart, Tasmania (2015) [Conference Extract]

Abstract

Introduction: Adverse drug reactions (ADRs) are a significant cause of hospitalisation in elderly patients.

Aim: To develop a risk score to predict ADR-related hospitalisation in people aged ≥65 years.

Methods: We conducted a prospective cross-sectional study at the Royal Hobart Hospital, Tasmania over a 12-month period to identify the proportion of patients whose admission was caused by an ADR. Identification of ADRs was based on reviewing medical records and patient interview. A wide range of potential predictors of ADR were evaluated in 70 patients admitted due to ADRs that were deemed preventable and 698 controls. Variables included the number of regular medications, number of comorbid conditions, dementia, heart failure, liver failure, renal failure, recent hospital admission and use of potentially inappropriate medications defined by Beer’s criteria. The variables associated with ADRs (p<0.05) in the bivariate analyses were entered into a multivariate logistic regression model. Variables retained in the final model were used to compute the ADR risk score. A score of 1 was assigned to variables with an odds ratio (OR) between 1.00 and 1.99; and a score of 2, to those with an OR between 2.00 and 2.99. The ADR risk score was computed based on the sum of scores of individual variables. Further analyses were performed to determine the cut-off score, sensitivity, specificity and model performance.

Results: The variables that predicted admission due to an ADR included number of regular medications (≥8), number of comorbid conditions (>6), presence of dementia and hospital admission within the past month. An ADR score cut off of 3 had a sensitivity of 70% and a specificity of 59%. The predictive ability of the risk score was assessed from a calculation of the area under the receiver operator characteristic curve and found to be 0.69 (95% CI, 0.63-0.75), suggesting that the ability of the model to predict ADRs is better than chance alone.

Discussion: The ADR risk score developed, after further validation, may be useful in clinical practice as a tool to identify patients at risk of hospitalisation due to preventable ADRs, and to target a subgroup of elderly that could benefit from interventions aimed to reduce this risk.

Item Details

Item Type:Conference Extract
Research Division:Medical and Health Sciences
Research Group:Pharmacology and Pharmaceutical Sciences
Research Field:Clinical Pharmacy and Pharmacy Practice
Objective Division:Expanding Knowledge
Objective Group:Expanding Knowledge
Objective Field:Expanding Knowledge in the Medical and Health Sciences
Author:Nair, N (Mr Nibu Parameswaran Nair)
Author:Chalmers, L (Dr Leanne Chalmers)
Author:Bereznicki, BJ (Dr Bonnie Bereznicki)
Author:Castelino, RL (Dr Ronald Castelino)
Author:Peterson, GM (Professor Gregory Peterson)
Author:Curtain, CM (Mr Colin Curtain)
Author:Connolly, MJ (Mr Michael Connolly)
Author:Bereznicki, LRE (Professor Luke Bereznicki)
ID Code:105035
Year Published:2015
Deposited By:Pharmacy
Deposited On:2015-12-03
Last Modified:2015-12-03
Downloads:0

Repository Staff Only: item control page