Using Normalisation Process Theory to design and test a complex intervention to implement opioid stewardship in acute hospitals

Author: Champika Pattullo. 

The challenges of creating meaningful and sustainable change in healthcare have been well documented in the literature, highlighting issues such as a limited understanding of the complexity and the lack of knowledge and skills in implementation. This can lead to the assumption that improvement depends solely on the characteristics of an intervention, while overlooking the unpredictability of complex systems which may respond differently to the same interventions. Implementation science is an approach that can be used to navigate these challenges. 

Opioid stewardship (OS) is a broad term that encompasses a range of interventions to improve the safety of prescription opioids. Normalisation Process Theory (NPT) is an implementation science theory, which uses four domains (coherence, cognitive participation, collective action and reflexive monitoring) to guide implementation. We describe below how we used NPT to guide the implementation of OS programs by examining the relationship between the intervention and its recipients (e.g. clinicians, policymakers and patients). 

Coherence (sense-making): Opioid-associated safety concerns are well accepted by clinicians. However, they play a central role in management of moderate to severe acute pain in hospital settings. Therefore, a balance is required in providing optimal acute pain management while minimising the potential harm of opioids if prescribed in excess. This is complicated, as pain is subjective and there are risks associated with under-treatment of pain, especially after surgery. In addition, OS is a complex intervention that is built up from a number of components, which may act both independently and inter-dependently. Therefore, a structured, holistic, patient-centred approach is particularly important. 

Cognitive participation (engagement): The OS Framework was developed to support the implementation of this 'complex' intervention and to ensure that the focus remained on patient and clinician priorities. We established an expert panel, with a majority (23/29) regularly involved in providing direct patient care and two patient representatives. The panel created the OS Framework by consensus using the Delphi method. 

Collective action (enactment): Subsequently, we tested the OS Framework on a wider group of clinicians who were tasked with implementing OS programs (across 16 hospitals). The testing provided insights into how the tool was used in real-world settings and provided participants the opportunity for feedback. 

Reflexive monitoring (appraisal): This process is used throughout by proactively seeking and incorporating end-user feedback for the OS Framework, initially by using the Delphi technique and subsequently by testing. 

NPT allowed us to use a structured approach to understand the problem, by taking a 'work as done' approach, instead of a 'work as imagined' approach. In addition, by using NPT, we were able to develop and test an intervention that can accommodate contextual nuances and one that focuses primarily on patient safety rather than one that limits opioid prescribing. This is particularly important in the absence of validated and safe opioid prescribing rates. The OS Framework promotes a holistic, patient-centred approach by identifying approaches to improve pain management and opioid prescribing. 

Champika Pattullo 
Assistant Director for Safety and Quality (Innovation & Implementation) 
The Royal Brisbane and Women’s Hospital 
Queensland Health, Queensland, Australia 

Acknowledgements 
Peter Donovan 
Benita Suckling 
Lisa Hall