Value-based healthcare – impact on quality and costs of healthcare

In the last three to four years several initiatives have started in The Netherlands to develop and use outcome measures for quality improvement and transparency of care. Outcome measures play a central role in value-based healthcare, organizing healthcare delivery around patient value, defined as health outcomes that matter to patients relative to the costs of achieving those outcomes. However, it is not shown yet whether and how outcome measures can be used to establish high value healthcare. Two specific research questions are investigated in this study: 1. Does steering on outcome measures lead to improved patient value, i.e. better health outcomes relative to costs? 2. What are the key aspects for successful implementation strategies in establishing high value healthcare? Study design The type of research is quality of care research by implementing quality improvement strategies for establishing high value healthcare. It can therefore be classified as a “developmental evaluation design.” The research is set up as a cohort study in a controlled before-and-after design. Study population/dataset Two key areas of St. Antonius Hospital are investigated: Aortic Valve Disease (AVD; Heart Center) and Interstitial Lung Disease (ILD; Lung Center). We will establish a cohort of patients with AVD, using an existing patient registry of Measurably Better (Dutch: Meetbaar Beter). Measurably Better is a national initiative of 12 collaborating Heart Centers in The Netherlands. Outcomes for AVD in St. Antonius Hospital will be compared with outcomes in 11 other Heart Centers in The Netherlands. For ILD we will establish a cohort of patients in an international collaborative of four hospitals. Outcomes for ILD in St. Antonius Hospital will be compared with outcomes in three other Lung Centers. Intervention For both ILD and AVD the implementation strategy of outcome reporting within St. Antonius Hospital is the main intervention. The implementation strategies will be developed based on four information sources: (a) a systematic review of the literature to assess the effectiveness of the implementation of interventions aimed at outcomes measurement; (b) analysis of baseline data to identify key quality gaps in health outcomes; (c) the identification of best practices within a group of hospitals; (d) analysis of barriers and facilitators for implementing outcomes measurement Core component of the implementation strategies will be the tailored presentation, and systematic and cyclic feedback of selected outcomes to the participating healthcare professionals and management of St. Antonius Hospital. Strategies to support St. Antonius Hospital for interpreting and acting upon the presented outcomes will be multifaceted and tailored to the specific circumstances. Outcome measures For both medical conditions a set of outcome measures is evaluated. For AVD we will use existing outcome measures as established in Measurably Better. For ILD part of the research will address the selection and definition of outcome measures for this chronic and rare disease. Costs are the other side of the focus on patient value. We will define a model in which both direct costs and relevant indirect costs will be held against activities that belong to the multidisciplinary care chain of ILD and AVD. Sample size For AVD, all patients of St. Antonius Hospital will be analyzed for the period 2009-2017 with ~180 patients treated by AVR and 100 patients treated with TAVI annually, leading to the inclusion of ~1,600 patients treated by AVR and 900 patients treated with TAVI. For ILD, the patient population is ~2,000 patients with specific focus on pulmonary sarcoidosis. We will establish a cohort of patients treated from 2009-2013 for baseline measurements, and a cohort of patients treated from 2015-2017 for evaluating the implementation strategies. Data analysis Patient value, including both health outcomes and costs will be compared before and after the implementation of outcome measures reporting. Outcomes will be case-mix adjusted for patient initial conditions. Costs will be analyzed according to the activity-based costing principle. We will use a mixed (multilevel) regression model for analyzing the data. The difference in outcomes between intervention and control hospitals will be modeled with an interaction term for the before-after implementation time periods. Data analyses will be controlled for baseline differences between intervention and control groups. Process evaluation The implementation strategies for the two conditions will be evaluated and compared in a process evaluation with participating professionals and stakeholders via semi-structured interviews in order to gain understanding in why the implementation strategies were successful or not.