Implementation and Evaluation of a Complex Intervention to Reduce Avoidable Transitions Between Nursing Home and Hospital

The researchers will conduct a pilot cluster randomized controlled trial (RCT) of the RATHER-NH intervention with an embedded process evaluation. This study has two specific objectives: 1) to evaluate the implementation of the RATHER-NH intervention, the perceived impact and underlying mechanisms of impact, and the contextual factors influencing implementation and impact; and 2) To assess the feasibility of the trial methods and procedures, and the potential primary and secondary outcome measures for a future full-scale RCT.

The study will be conducted in twelve nursing homes in Flanders, of which six will be randomly assigned to the intervention group and six to the control group.

The researchers will use the MRC guidance on complex intervention development and evaluation combined with the integrated Promoting Action on Research Implementation in Health Services (i-PARIHS) framework to guide the implementation and evaluation of RATHER-NH.

Based on the results of the qualitative study with nursing home residents, family carers and professional stakeholders, and on existing scientific literature, a first version of a logic model was developed, which illustrates the different determinants and mechanisms that influence the anticipated outcomes and impact of the RATHER-NH intervention. The logic model was used to select relevant intervention tools from the INTERACT Quality Improvement Program and our own self-developed tool concerning a GP communication tool:

1. Stop and Watch Early Warning Tool This is a checklist that marks the most relevant changes in the physical and behavioural condition of a nursing home resident that eventually may lead to a potentially avoidable hospitalisation.
2. Decision trees Decision trees have been developed for several Ambulatory Care Sensitive Conditions (ACSCs), which are conditions that can lead to unplanned hospitalisation which may be avoided if the condition is promptly detected and managed correctly through access to medical care in the nursing home.
3. General Practitioner communication tool This tool does tries to standardize the communication between the general practitioner and nursing home staff members.

The tools from the RATHER-NH intervention toolkit have fixed and flexible elements. To adapt elements of the tools to the local context, each participating nursing home will engage in a facilitated group discussion to assess the context and their specific needs, facilitated by the researcher and guided by the i-PARIHS framework. While the fixed elements of each tool need to be maintained to preserve the intervention fidelity, the flexible and hence adaptable elements can be tailored to fit the specific needs, routines, and structures of each nursing home. This ensures that the intervention is both evidence-informed and context-sensitive, increasing the likelihood of successful implementation.

Based on evidence from our qualitative study with key stakeholders and findings from other relevant studies, several implementation barriers are anticipated when implementing complex nursing home interventions such as the RATHER-NH intervention. These barriers were linked to the implementation strategies as described in the Expert Recommendations for Implementing Change (ERIC) Matching Tool. These strategies combined will form the implementation strategy for RATHER-NH (e.g. conduct educational meetings, ongoing training and identify and prepare champions). To ensure the implementation strategy is responsive to the specific needs of each nursing home, this will also be discussed during the facilitated group discussion to tailor and operationalise the implementation strategies by adapting the flexible elements of the implementation strategies to the local context. Next to the researcher, an external trainer will be recruited to support the implementation of the intervention within the nursing homes.