Tailoring interventions to implement recommendations for the treatment of elderly patients with depression: a qualitative study

To improve adherence to evidence-based recommendations, it is logical to identify determinants of practice and tailor interventions to address these. We have previously prioritised six recommendations to improve treatment of elderly patients with depression, and identified determinants of adherence to these recommendations. The aim of this article is to describe how we tailored interventions to address the determinants for the implementation of the recommendations.

We drafted an intervention plan, based on the determinants we had identified in a previous study. We conducted six group interviews with representatives of health professionals (GPs and nurses), implementation researchers, quality improvement officers, professional and voluntary organisations and relatives of elderly patients with depression. We informed about the gap between evidence and practice for elderly patients with depression and presented the prioritised determinants that applied to each recommendation. Participants brainstormed individually and then in groups, suggesting interventions to address the determinants. We then presented evidence on the effectiveness of strategies for implementing depression guidelines. We asked the groups to prioritise the suggested interventions considering the perceived impact of determinants and of interventions, the research evidence underlying the interventions, feasibility and cost. We audiotaped and transcribed the interviews and applied a five step framework for our analysis. We created a logic model with links between the determinants, the interventions, and the targeted improvements in adherence.

Six groups with 29 individuals provided 379 suggestions for interventions. Most suggestions could be fit within the drafted plan, but the groups provided important amendments or additions. We sorted the interventions into six categories: resources for municipalities to develop a collaborative care plan, resources for health professionals, resources for patients and their relatives, outreach visits, educational and web-based tools. Some interventions addressed one determinant, while other interventions addressed several determinants.

It was feasible and helpful to use group interviews and combine open and structured approaches to identify interventions that addressed prioritised determinants to adherence to the recommendations. This approach generated a large number of suggested interventions. We had to prioritise to tailor the interventions strategies.