Cost-effectiveness of a shared decision-making intervention for patients receiving treatments for cardiovascular disease and diabetes in primary care: a cluster randomised controlled trial based on real-world data (IMA-cRCT study)

Background: Cardiovascular risk (CVR) and medication non-adherence are major public health issues. The Initial Medication Adherence (IMA) intervention promotes Shared Decision-Making (SDM) in primary care (PC) to enhance medication adherence and reduce CVR. This study evaluated the cost-effectiveness of the IMA intervention compared to usual care (UC). Methods: This was a cluster randomized controlled trial conducted in primary and community care in Spain. Primary care centers (PCC) were randomly selected from all Catalan PCC. Using a 1:1 paired randomization based on center characteristics, PCC (and their catchment areas) were allocated to control (12 PCC, 1553 patients) and intervention (12 PCC, 2189 patients) groups. Professionals in the IMA group were trained in SDM and decision aids. Physicians, nurses and pharmacists applied the intervention to patients receiving a first prescription for antihypertensive, lipid-lowering, antiplatelet and oral/injectable antidiabetic drugs. Professionals in the control group received no training, and patients received UC. Clinicians were not blinded to the intervention, but outcome variables were obtained from electronic medical records. Direct medical costs included the number of visits, hospital stays, medical tests, medications, and the IMA intervention. Indirect costs from absenteeism were also considered. Effects included CVR, and medication initiation and implementation (proportion of days covered>80%). Using an intention-to-treat (ITT) approach, multilevel regression models estimated differences in costs and effects between groups. Incremental cost-effectiveness ratios were calculated. Uncertainty was assessed through sensitivity analysis and bootstrapping. Results : 89/100 physicians; 75/78 nurses and 61/63 pharmacists completed the training; all were analyzed (ITT approach). There were no differences between intervention and control patients or professionals. More training PCC were present in the intervention group (67% vs. 25%). At 12 months, the IMA intervention was cost-effective for CVR, dominant in most scenarios from both societal and healthcare perspectives, with >98% probability of being cost-effective at relevant thresholds. For treatment initiation and implementation outcomes, results showed greater uncertainty, indicating lower efficiency and variable cost-effectiveness depending on the scenario and analysis assumptions. Conclusions: The SDM-based IMA intervention was cost-effective compared with usual care for PC patients needing cardiovascular disease and/or diabetes treatment. Trial registration: ClinicalTrials.gov, NCT05026775. Registered 30th August 2021. Funding: ERC Starting Grant (Grant Agreement 948973).