Risk assessment of medication safety in pharmacotherapeutic practice

Abstract

Approximately 4% of hospitalizations are associated with preventable adverse drug events, estimated to cost over €2 million annually for our national healthcare service.

The aim of the research is to identify key areas for improving medication safety and evidence-based strategies for reducing the margin of error in pharmacotherapeutic practice.

A method for classification of medication error was developed using a quality risk management approach, based on the clinical stage of pharmacotherapy and the cognitive psychology of task-oriented behaviour. Reports of adverse drug events recorded in the EudraVigilance System between June 2012 and September 2015 were assessed for the type of medication error and predetermined case variables, which included healthcare setting, patient age, gender, number of drugs, route of administration and pharmacological group. The severity of the medication errors identified from the case reviews was graded according to the WHO Severity index. Hypothesis testing was carried out using the Pearson Chi-squared test to determine the correlation between the severity and explanatory variables, and the Kruskal Wallis one way analysis of variance to compare the mean severity rating scores. The relative significance of the predictors was established using a Multinomial logistic model for ordinal data. From a total of 2294 adverse drug events reported to the European Medicines Agency over a period of three years, 1926 (84%) were confirmed cases of medication errors, of which 1300 (57%) were considered preventable. The majority of medication errors were encountered in the prescribing (28%) and the administration (27%) of medicine, compared to the dispensing (14%) stage. The type of human errors were classified as case based in the prescribing stage (60%), skill based in the dispensing stage (66%) and rule based during administration of drug treatment (75%). A positive correlation was observed between the severity of patient harm and pharmacological group, number of drugs and route of administration (p < 0.001), as well as type of error and patient age (p < 0.003).

The research identifies nine different types of pharmacotherapeutic failure in clinical practice and quantifies the relative severity of patient harm associated with each. The risk analysis provides evidence for the potential of significant reductions in medication error through the use of quality risk management systems by licensed healthcare institutions, and the regulation of medication safety standards as a complementary strategy to existing pharmacovigilance measures required at law.

The results of the risk assessment support the need for robust decision support systems in prescribing of drugs, standard quality control checks during administration of treatment, better monitoring and improvement of patient counselling to reduce the margin of clinical error, and screening of specific pharmacological groups as priority areas for reducing severe adverse drug events, particularly in elderly patients and cases where patients are being treated with ten drugs or more.