Dementia is a global health issue, and no cure currently exists. The majority of people diagnosed with the disease experience behavioral and psychological symptoms of dementia (BPSD). These symptoms can be a burden for both the patient and their relatives and are a reason for relocation to a nursing home as well as being prescribed psychotropic medication. The first line of treatment for BPSD is non-pharmacological, but antidepressants are often used in older nursing home residents with dementia, sometimes in an attempt to lower BPSD although the treatment is off-label. Deprescribing in this population is often complicated and requires collaboration between general practitioners (GP), nursing home staff, patients, and relatives.

We hypothesized that improving GP and nursing home staff's knowledge about BPSD, antidepressants, and deprescribing in combination with using a more structured approach to strengthen communication would increase their collaboration and willingness to attempt deprescribing, thus improving the treatment of BPSD by reducing unnecessary antidepressants. The aim of this thesis was to develop a new model situated in general practice for the management of BPSD in nursing home residents with dementia. Moreover we sought to investigate if the model could effectively and safely reduce the use of psychotropic medication.
As an overall framework, we used the Medical Research Council’s updated guidelines from 2021 to develop and evaluate the intervention. We conducted a qualitative systematic review and further developed an intervention through an iterative process. The effectiveness of the intervention was evaluated in a cluster randomized controlled trial.

Based on the findings from a qualitative systematic review and interviews with stakeholders during the tailoring process, we changed the focus from antipsychotic medication to antidepressants and made it optional to include relatives during the intervention. The final complex intervention included a case-based training course, a pre-visit reflection tool and a dialog tool used during a structured consultation in nursing homes. Results from the evaluation of the cluster randomized controlled trial showed a non-significant increase in the reduction of antidepressants in the intervention group compared to the control group. Reducing antidepressants was not associated with an increased mortality but may have caused worsening of BPSD symptoms. Thus the project suggests that testing the intervention in a larger scale to gain more power and certainty regarding outcomes of reducing antidepressants would be relevant.