A Lean-DEA approach for continuous improvement in Brazilian public hospitals.

Purpose: Propose an integrated approach of Lean Healthcare (LH) with Data Envelopment Analysis (DEA), establishing best practices for waste reduction and process improvement along with hospital efficiency.

Design/methodology/approach: The study proposes a two-stage approach integrating LH and DEA to enhance hospital efficiency. The first stage identifies efficient hospitals post-LH implementation and projects efficiency for inefficient ones using ideal projections. In the second stage, a new DEA model applies to established extreme-efficient hospital sets, guiding LH technique selection for continuous improvement. This approach is tested on six Brazilian public hospitals before and after LH application (HA and HD, respectively), offering strategic insights for decision-making. The inclusion of ideal DMUs adds a dynamic dimension, projecting future efficiency and guiding inefficient hospitals toward effective practices.

Findings: The results confirmed the approach's effectiveness, providing performance metrics and strategic guidance for healthcare service enhancement. Utilizing non-oriented SBM-DEA in the first stage determined DMU efficiency, revealing both efficient and inefficient hospitals post-Lean implementation. The approach extended beyond efficiency assessment by proposing ideal DMUs for inefficient hospitals. This refinement allowed a comprehensive analysis, considering real DMUs and ideal projections, emphasizing the approach's holistic evaluative capability.

Research Limitations/implications: The main limitation of this study lies in the restricted set of hospitals analyzed. Since the sample consisted of a limited number of institutions, the selection of inputs and outputs used in the Lean-DEA model was constrained by the availability of data provided by these hospitals. As a result, some potentially relevant variables for evaluating efficiency and Lean practice adoption could not be included in the analysis, which may have affected the comprehensiveness of the results. Additionally, the use of ideal DMUs generated from a mathematical optimization process - without orientation to the specific context of healthcare - may lead to the suggestion of target values that are unrealistic or unattainable for actual hospitals.

Practical Implications: The findings of this research provide a decision-support tool in scenarios marked by budget constraints and high demand for services. The application of the Lean-DEA model makes it possible to identify underperforming units and determine which managerial practices can be incorporated based on internal benchmarks within the healthcare system. As a result, managers gain access to a more accurate, evidence-based approach to guide interventions and strategically allocate resources, enhancing the efficiency and quality of services delivered to the population.

Social Implications: The potential benefits of applying the Lean-DEA approach are directly reflected in improved healthcare services. More efficient hospitals can deliver faster, higher-quality care with reduced waste, which positively impacts the experience of users of the public health system. By contributing to the optimization of hospital processes, this research also supports the strengthening of the healthcare system as a whole, promoting greater equity in access and better public health outcomes.

Originality/value: The study innovatively integrates LH and DEA, introducing ideal DMUs in this context. The target analysis highlights the consistent reference of ideal projections, even compared to potential alternative outcomes, offering a unique contribution not found in existing literature.