A scoping review of fair machine learning techniques when using real-world data.

OBJECTIVE: The integration of artificial intelligence (AI) and machine learning (ML) in health care to aid clinical decisions is widespread. However, as AI and ML take important roles in health care, there are concerns about AI and ML associated fairness and bias. That is, an AI tool may have a disparate impact, with its benefits and drawbacks unevenly distributed across societal strata and subpopulations, potentially exacerbating existing health inequities. Thus, the objectives of this scoping review were to summarize existing literature and identify gaps in the topic of tackling algorithmic bias and optimizing fairness in AI/ML models using real-world data (RWD) in health care domains. METHODS: We conducted a thorough review of techniques for assessing and optimizing AI/ML model fairness in health care when using RWD in health care domains. The focus lies on appraising different quantification metrics for accessing fairness, publicly accessible datasets for ML fairness research, and bias mitigation approaches. RESULTS: We identified 11 papers that are focused on optimizing model fairness in health care applications. The current research on mitigating bias issues in RWD is limited, both in terms of disease variety and health care applications, as well as the accessibility of public datasets for ML fairness research. Existing studies often indicate positive outcomes when using pre-processing techniques to address algorithmic bias. There remain unresolved questions within the field that require further research, which includes pinpointing the root causes of bias in ML models, broadening fairness research in AI/ML with the use of RWD and exploring its implications in healthcare settings, and evaluating and addressing bias in multi-modal data. CONCLUSION: This paper provides useful reference material and insights to researchers regarding AI/ML fairness in real-world health care data and reveals the gaps in the field. Fair AI/ML in health care is a burgeoning field that requires a heightened research focus to cover diverse applications and different types of RWD.

Step-down units are cost-effective alternatives to coronary care units with non-inferior outcomes in the management of ST-elevation myocardial infarction patients after successful primary percutaneous coronary intervention.

Percutaneous coronary interventions (PCIs) within a door-to-balloon timing of 90 min have greatly decreased mortality and morbidity of ST-elevation myocardial infarction (STEMI) patients. Post-PCI, they are routinely transferred into the coronary care unit (CCU) regardless of the severity of their condition, resulting in frequent CCU overcrowding. This study assesses the feasibility of step-down units (SDUs) as an alternative to CCUs in the management of STEMI patients after successful PCI, to alleviate CCU overcrowding. Criteria of assessment include in-hospital complications, length of stay, cost-effectiveness, and patient outcomes up to a year after discharge from hospital. A retrospective case-control study was done using data of 294 adult STEMI patients admitted to the emergency departments of two training and research hospitals and successfully underwent primary PCI from 1 January 2014 to 31 December 2015. Patients were followed up for a year post-discharge. Student t test and χ2 test were done as univariate analysis to check for statistical significance of p < 0.05. Further regression analysis was done with respect to primary outcomes to adjust for major confounders. Patients managed in the SDU incurred significantly lower inpatient costs (p = 0.0003). No significant differences were found between the CCU and SDU patients in terms of patient characteristics, PCI characteristics, in-hospital complications, length of stay, and patient outcomes up to a year after discharge. The SDU is a viable cost-effective option for managing STEMI patients after successful primary PCI to avoid CCU overcrowding, with non-inferior patient outcomes as compared to the CCU.