Psychological factors influencing unsafe behaviour during medication administration

Psychological factors influencing unsafe behaviour during medication administration

The health system in Australia delivers safe and effective health care to millions of patients each year. However, health care is not as safe as it could be with research indicating that errors involving medications are a leading cause of unintended harm to patients both in Australia and internationally. Historically, hospital authorities have attempted to reduce incidents by focusing on the actions of individuals. However, the health system is now taking advantage of research carried out in other complex industries which indicates that error is inevitable and that identifying individuals as the ultimate cause of adverse incidents is of limited value unless the context in which the incident occurred is well understood. This series of studies used Reason's (1990) model of accident causation as the basis for the search into possible contributing factors to unsafe behaviour by nurses during medication administration. Structural equation modelling was used to operationalise Reason's theory by developing a model linking organisational and individual factors to unsafe behaviour in the hospital system. Study 1 in this series was a preliminary investigation of the role of organisational factors in contributing to violations by nurses in rural and remote areas in Queensland, Australia. Data were collected using a self-report questionnaire with this instrument being used to develop a structural model wherein organisational variables predicted 23% of the variance in self-reported violations. Study 2 extended the number of organisational factors measured by using a validated instrument that is widely used in public sector hospitals in Queensland. This instrument measures organisational climate and also a number of individual factors. In addition to the outcome variable, violation behaviour, a measure of errors was included. Data were collected from nurses working in two rural health service districts. A structural model was developed from this instrument wherein organisational variables predicted 7% of the variance in selfreported violations and 24% of the variance in errors. The hypothesised relationships between the individual factors and errors were not supported in this study. Study 3 investigated the impact of individual factors and a specific type of organisational climate, that is, safety climate on unsafe behaviour. The violation behaviour and error scales were extended and improved in this study, for example, the error scale was expanded to include near misses. In addition, a new scale measuring reporting behaviour was developed and included. Data were collected from nurses working in a large rural centre. The structural model developed from the instrument indicated that safety climate predicted 27% of the variance in violation behaviour, 61% of the variance in errors and near misses, and 20% of the variance in willingness to report. This series of studies identified underlying contributing factors to unsafe behaviour during medication administration, indicated the strength of the relationships among the various elements, and illustrated how the various parts of the system link together to influence safety outcomes. By identifying which elements are important by the use of structural equation modelling, this research provides the basis for predicting unsafe organisational conditions and leads to suggestions for suitably targeted interventions to reduce unsafe behaviour and adverse incidents.