Bundled payments (BPs) are increasingly being adopted to enable the delivery of high-value care. For BPs to reach their goals, accounting for differences in patient risk profiles (PRPs) predictive of spending is crucial. However, insight is lacking into how this is done in practice. This study aims to fill this gap.
We conducted a systematic review of literature published until February 2024, focusing on BP initiatives in the Organization for Economic Cooperation and Development countries. We collected data on initiatives’ general characteristics, details on the (stated reasons for) approaches used to account for PRP, and suggested improvements. Patterns within and across initiatives were analyzed using extraction tables and thematic analysis.
We included 95 documents about 17 initiatives covering various conditions and procedures. Across these initiatives, patient exclusion (n = 14) and risk adjustment (n = 12) of bundle prices were the most applied methods, whereas risk stratification was less common (n = 3). Most authors stated mitigating perverse incentives as the primary reason for PRP accounting. Commonly used risk factors included comorbidities and sociodemographic and condition/procedure-specific characteristics. Our findings show that, despite increasingly sophisticated approaches over time, key areas for improvement included better alignment with value and equity goals, and enhanced data availability for more comprehensive corrections for relevant risk factors.
BP initiatives use various approaches to account for PRP differences. Despite a trend toward more sophisticated approaches, most remain basic with room for improvement. To enable cross-initiative comparisons and learning, it is important that stakeholders involved in BPs be transparent about the (reasons for) design choices made.
This systematic literature review examines how bundled payment initiatives address variations in patient risk profiles to ensure high-value healthcare delivery. Bundled payments are designed to streamline costs by providing a single fixed payment for all services related to a specific medical condition over a defined period. For these programs to be effective, understanding and adjusting for the differences in patient risk is essential.
The review analyzed 95 documents related to 17 bundled payment initiatives across countries in the Organization for Economic Cooperation and Development (OECD). The findings reveal that the most common methods for accounting for differences in patient risk profiles are patient exclusion and risk adjustment. Patient exclusion involves removing certain patients from the bundle based on risk factors, while risk adjustment modifies the payment amounts based on the complexity of the patients' conditions, often focusing on comorbidities and demographic factors.
While bundled payment initiatives are evolving toward more sophisticated methods, many still rely on basic approaches, indicating a need for improvement. Commonly, initiatives aim to limit unintended consequences, such as providers avoiding higher-risk patients, and to ensure fairness among providers treating varied patient populations. There is a growing emphasis on using a wider range of risk factors to enhance the accuracy of risk adjustments, particularly incorporating clinical data alongside historical claims data.
Healthcare decision makers, patients, and researchers should note the importance of transparent reporting of the methodologies used in these initiatives. This transparency is crucial for comparing different programs and learning from their successes and failures. The study suggests that improvements in data availability and the incorporation of more diverse risk factors are necessary to align bundled payment initiatives with broader goals of healthcare value and equity.
In conclusion, while bundled payment initiatives have made strides in addressing patient risk profiles, there remains significant room for enhancement in methodology and data utilization. By refining these approaches, stakeholders can better support high-value care and ensure that payment models fairly reflect the complexities of patient needs.