Objectives To develop, implement, and evaluate a context-sensitive approach to reduce low value imaging by answering specific research questions (Q1-5):

1. What are the relevant low-value imaging examinations to target with an intervention?.

2. What are the relevant interventions to target low-value examinations?.

3. Which examinations should be targeted by an intervention in this specific context (Norway)?.

4. Which intervention is best suited to target the selected examinations, and how can this be implemented?.

5. What are the outcomes of the intervention to reduce low-value imaging examinations in this context (in terms of number of examinations and costs)?.

Method For question Q1 and Q2 we searched Medline-Ovid, Embase-Ovid, Scopus, and Cochrane Library for studies published from 2010 to 2020, and included articles in English, German, Dutch, Swedish, Danish, or Norwegian for review. To address Q1 we performed a scoping review and for Q2 a systematic review.

For Q3 and Q4 we applied an expert panel with representatives from the Norwegian Medical Association, Norwegian Radiological Association, the Norwegian Patient Organization, and two private providers of radiological services that deliberated on identifying the most suitable examinations and intervention.

For Q5 we analysed data from the Control and Reimbursement of Healthcare Claims (CRHC) registry, from Radiological Information Systems (RIS) in Norway from 2013 til 2023.

Results The scoping review encompassing 39,986 records and including 370 studies identified 84 low-value imaging examinations. The systematic review counting 15,659 records and including 95 studies showed that multi-component interventions were more effective than single-component interventions and that musculoskeletal, neurological, and vascular imaging were the most targeted types of imaging.

The expert panel together with the research group shortlisted 13 imaging examinations and recommended targeting three specific examinations: MRI of the lower back, MRI of the head, MRI of the knee. Correspondingly, an intervention based on a combination of a rejection letter, an information campaign towards referrers and patients in primary care waiting rooms was rolled out.

Preliminary results show a substantial use of the return letters at the start of the intervention period, and initial results in use of low-value imaging examinations will be presented for discussion in Copenhagen.

Conclusion Reducing low-value imaging examinations requires careful deliberation on which examinations to target, which interventions to apply, and how to implement them. Sensitivity to context is key. This Norwegian approach and its results can be of interest and inspirations to others. More specific examples of reducing low-value services are needed.