Can we do better? Sustainability and efficiency in intervention development and implementation

Developing interventions has become much more robust and rigorous, guided by the Medical Research Council’s Framework on developing complex interventions [1]. There is an emphasis on reference to existing evidence, theory, feasibility and pilot testing, and process evaluation. However, this is time-consuming, prolongs the generation of evidence, and if practice and policy evolve over the course of the research, the intervention may no longer be relevant. A research programme that we led (PolyPrime), focussing on the prescribing of appropriate polypharmacy by general practitioners, has taken 9 years to achieve completion of a pilot study, as we worked our way through the generation of a systematic review, the intervention development process, a small feasibility study and a randomized pilot trial that was affected by the COVID-19 pandemic [2]. Over this time period, the practice and policy landscape changed, which has implications for the future implementation of this intervention, particularly with the advent of general practice pharmacists (GPPs) assuming many of the responsibilities for medicines optimization in primary care [3]. It has been recognized that there are insufficient resources to develop and evaluate interventions for every health care issue separately [4]. This is particularly pertinent when we consider interventions focussed on prescribing and medicines optimization in which we need to consider different medication categories, different health care settings or contexts, and different populations [patients and health care professionals]. Therefore, how can we develop and test interventions targeting prescribing and medicines optimization, more efficiently and sustainably, while retaining methodological robustness? Identification of common intervention components may help accelerate intervention development. Increasingly, ‘behaviour change techniques’ (BCTs) are being used as the ‘active ingredients’ or components that bring about change in prescribing and other healthcare behaviours [5]. The BCTs used in PolyPrime were ‘Action planning’, ‘Prompts and cues’, ‘Modelling or demonstrating the behaviour’, and ‘Salience of consequences’ [6]. Tang et al. [7] reporting on an intervention focussing on a range of ‘drug therapy risks’ involving multiple medicines to be delivered by pharmacists in general practices, noted that there was a broad similarity between the BCTs in their intervention compared to PolyPrime. A systematic review of studies targeting deprescribing noted that effective interventions often contained the BCTs ‘Instructions on how to perform the behaviour’, ‘Credible source’, ‘Social support’ (unspecified), ‘Action planning’, and ‘Feedback on behaviour’ [8]. Several studies have focussed on antimicrobial stewardship interventions, and across this research, there has been commonality in intervention content, notably, the use of the BCTs ‘Feedback on behaviour’ and ‘Environmental restructuring’ [9]. However, although there may be overlap in intervention content, could these interventions be implemented in different settings or contexts? Bohlen et al. [10] stated that delivering the same interventions containing the same BCTs and being operationalized in the same way may not be appropriate due to different contexts and populations. Nair et al. [11] highlighted that most interventions targeting antibiotic prescribing had come from ‘developed’ countries and involved complex multi-faceted strategies such as electronic decision support, electronic health record prompts and automated peer comparison interventions. It was unlikely that such interventions would be as effective or even applicable in low- and middle-income countries, which have a high burden of communicable diseases and may not be able to support expensive or high-tech interventions involving electronic health records. There is an increasing interest in intervention adaptation that has been defined as ‘an intentional modification(s) of an evidence-informed intervention, in order to achieve a better fit with a new context’ [12]. This includes planned adaptions (changes made prior to introducing a new intervention) and responsive adaptations (changes made intentionally but in response to emerging contextual issues occurring during implementation) [12]. Context has been defined as ‘any feature of the circumstances in which an intervention is implemented that may interact with the intervention to produce variation in outcomes, including geographical, organisational, cultural and economic circumstances’ [12]. Recently published guidance (known as ADAPT) outlines four steps, which should be overseen by an adaptation team. Step 1: Assess the rationale for the intervention and consider intervention-context fit. This step requires defining the problem to be targeted and identifying candidate interventions that may be suitable for adaptation. Following the review of candidate interventions, the guidance recommends selecting one intervention, obtaining detailed information on this intervention (including robustness of effectiveness claims), and mapping the similarities and differences between the original and new contexts [12]. Step 2: Plan for and undertake adaptations. This step requires the adaptation team to consider the adaptations that are required for the selected intervention in the new context. For example, if the selected intervention contains a training component, does this need to be updated or changed? This may also lead to a discussion on the resources needed to adapt and implement the intervention, and if there are any possible unintended consequences of introducing and implementing the intervention in the new context [12]. Step 3: Plan for and undertake piloting and evaluation. The extent of evaluating the adapted intervention will be dependent on the existing evidence for the selected intervention, how applicable this evidence is to the new context, and the extent of intervention adaptation that may be required. The ADAPT guidance indicates that some re-evaluations may be quite cursory, while others may require a full RCT [12]. Step 4: Implement and maintain the adapted intervention at scale. This final step is dependent on the outcome of Step 3. If the adapted intervention is shown to be effective, it may be possible to implement at scale with a full roll-out. Alternatively, if evaluation in Step 3 is inconclusive, this may require consideration of further adaptation, or a decision not to proceed further [12]. In the case of PolyPrime [2], the context that has changed is organizational (i.e. the role of GPPs in primary care), and this would probably be the main consideration for adaptation. We would also need to consider the type and extent of evaluation needed for an adapted PolyPrime intervention as the pilot study did not assess effectiveness [2]. Adapting interventions is in its relative infancy but the ADAPT guidance does provide researchers with a systematic approach that may lead to faster implementation of effective interventions across different contexts. Both authors contributed equally to this editorial. None declared. The PolyPrime study referred to in this editorial was funded by the HSC R&D Division Cross-border Healthcare Intervention Trials in Ireland Network (CHITIN) programme, funded by the European Union’s INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB) project reference CHI/5431/2018. The views and opinions expressed in this editorial do not necessarily reflect those of the European Commission or the Special EU Programmes Body (SEUPB). The funding body and study sponsor were not involved in the writing of this editorial.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.