An aging population, an increasing number of chronically ill people and the rapidly growing number of diagnostic and treatment options are putting our healthcare system and its (financial) sustainability under increasing pressure. Policymakers are therefore looking for ways to increase the value for money of care and to achieve the best possible outcomes for patients at the lowest possible cost. In doing so, the aim is to reduce over- and undertreatment of patients, undesirable variation in outcomes and fragmentation of care. Two specific areas of focus that are critical to developing and optimizing appropriate policies in this context are (1) gaining insight into patient outcomes and using this information to improve the quality of care, and (2) developing and evaluating reform of the financial incentives embedded in the methods used to pay health care providers for the care they provide. Currently, these incentives are primarily centred on volume rather than "value," and are unrelated to quality of care and patient outcomes.
In practice, "best possible outcomes" means maximum scores on a balanced set of clinical and patient-reported outcome measures, developed in consultation with patients. These outcome measures can then be used internally by a healthcare provider to assess and improve care for individual patients. In aggregated form, healthcare providers can use this data to track outcomes over time and identify areas for improvement within their own patient population. In addition, aggregated outcome information can also be used externally for comparison between healthcare providers ("benchmarking") and can be incorporated into alternative payment models of healthcare providers.
Despite many efforts in research and policy in recent years, insight into outcomes and their relationship with the quality and cost of care for many diseases remains very limited. There is still a lack of knowledge about suitable methods for quantifying and reporting outcomes and variation therein between healthcare providers. Moreover, the "feasibility" of using aggregated outcome measures to improve quality of care is questioned. In addition, little is known about how outcome information can best be integrated into alternative cost models, and what the implementation of these models can deliver. These knowledge gaps impede the transition to better care with better outcomes at lower costs.
Objectives and Relevance
This Action Line aims to contribute to the scientific underpinnings of the pursuit of outcome-based care. Two specific objectives are:
- To develop methods for quantifying outcomes and differences therein among healthcare providers, as well as to assess the reliability (e.g., the influence of chance variation) and validity (e.g., the influence of case-mix differences) of outcome indicators.
- Developing and testing methods to manage aggregated outcomes (both internal and external) and costs - including benchmarking and alternative payment models (such as bundled payments and outcome costs) - and evaluating the effects of these new methods in terms of improving care processes, (re)designing/improving workflows, multidisciplinary collaboration, and quality and cost of care.
To achieve these objectives, the following studies will be designed and conducted within the Action Line:
- Quantitative analysis of large databases using 'comparative effectiveness research' methods. The use of outcome information to guide policy in clinical practice assumes that outcome indicators are valid and reliable indicators of quality of care. It is important that outcome indicators are sensitive to determine improvement in the quality of care. We are therefore investigating the relationship between variation in quality of care (measured by process indicators based on treatment guidelines from clinical practice) on the one hand and clinical and patient-reported outcomes on the other. In doing so, using innovative methods, we take into account a number of methodological challenges in comparing outcomes between healthcare providers, such as chance variation, missing values, and differences in case-mix. In addition, we apply innovative methods (such as the Regression Discontinuity design) that allow us to establish causal relationships between treatment and outcome in observational data. In doing so, we are collaborating with other Action Lines (e.g., Prevention) to make the most of the knowledge about these quasi-experimental methods within Smarter Choices for Better Health.
- For the research on internal analysis and learning of outcome data, we use routinely collected outcome data for different conditions from Erasmus MC. For research into the 'external' use of outcome data (benchmarking), we use data from clinical registries, including data from the Integral Cancer Center Netherlands (INKL) and clinical registries from the Dutch Institute for Clinical Auditing (DICA). If possible, we will also conduct our own data collection in carefully selected (hospital) settings.
- Designing and evaluating methods to manage outcomes. This Action Line will focus on two methods: (a) new approaches to benchmarking that use aggregated outcomes to manage or learn from quality, including determining the conditions under which managing for outcomes might be useful and feasible; and (b) alternative payment models for healthcare providers, using outcome information. Based on the international literature and previous experiences in the Netherlands, we will develop different modalities for (a) and (b) that will be tested for feasibility in different (hospital) settings.
- Conducting (small-scale) experiments with steering based on aggregated outcomes (and possibly costs), using the insights obtained from the previous studies. The main question to be answered is whether the new approach (an intervention based on aggregated outcomes) leads to useful information and actually improves care. These experiments will take place primarily at Erasmus MC, but ideally other healthcare providers in the region will also be involved. In addition, in collaboration with several external partners, we will evaluate ongoing experiments with alternative payment models, both quantitatively (i.e., effects on cost and quality) and qualitatively (e.g., what contextual factors facilitate or hinder the implementation of these types of models in practice?).
We are building on the work and insights gained during SCBH 1.0. In doing so, this Action Line will yield a number of insights that are of great scientific and societal importance. First, a better understanding of the relationship between variation in quality of care on the one hand and (variation in) medical and patient-reported outcomes on the other. Second, a better understanding of the appropriate approach and use of different methods for steering for outcomes and costs, and their effects. We are convinced that these insights will contribute to better approaches to achieve better quality of care at lower costs. From our Action Line, we will actively contribute to the implementation of our findings both at Erasmus MC and at other providers in the Rotterdam region, but also beyond (both nationally and internationally). We will also implement our work and our findings in our education (including relevant courses and subjects of ESHPM and Erasmus MC), and use our (inter)national network to spread our findings as much as possible.
The projects within this Action Line will involve close collaboration with various internal and external partners, including clinical departments at Erasmus MC, Integral Cancer Center Netherlands (IKNL), the Dutch Institute for Clinical Auditing (DICA), health insurer Menzis, the National Institute for Public Health and the Environment (RIVM), the Dutch Healthcare Authority (NZa) and Health Campus The Hague. In addition to strengthening social impact, these collaborations provide access to rich data sources and opportunities to evaluate ongoing interventions/implementations related to steering for outcomes and costs.
The Action Line Outcome Based Healthcare team consists of the following people::
- Erasmus MC: Nikki van Leeuwen (Action Line Leader (ALL)), Margrietha van der Linde (PhD student), Hester Lingsma (supervisor, member Management Team)
- ESHPM: Frank Eijkenaar (ALL), Daniëlle Cattel (assistant professor), Raf van Gestel (assistant professor), Tadjo Gigengack (PhD student), Erik Schut (supervisor)
- Health Campus The Hague: Arthur Hayen (assistant professor)
