Article Preview
TopIntroduction
Delivering equitable healthcare at sustainable costs is one of the most pressing economic challenges currently facing the US system. Consequently, leveraging big data with advanced computerized decision support and technology solutions provides a unique opportunity to assist medical professionals in delivering both intelligent and efficient patient-level decision-making policies. In this paper, we attempt to consider one such opportunity: US heart attack patients entering the hospital system via the emergency room. The failure to provide consistent treatment for different patient groups has long been a serious concern for those studying the US healthcare system. Accordingly, some of the primary motivations of the 2010 Affordable Care Act (ACA) legislation were to expand patient care and standardize healthcare delivery practices (US Department of Health & Human Services, 2018). One crucial area where procedural differences may be witnessed is in the emergency department (ED) discharge setting related to the inter-facility transfers (IFT) of heart attack patients. Acute myocardial infarction (AMI) – or a “heart attack” – is one of many high-transfer-rate medical conditions (Kindermann et al. 2015) where decisions at ED discharge can have important implications for health outcomes. During the patient encounter, providers must quickly gather and process information to determine if the ED heart attack patient requires admission to the hospital or if the patient (or hospital) would be better served by an inter-facility transfer (IFT) between hospitals, dedicated nursing facilities, and/or primary or secondary support centers for different types of downstream cardiac care (Joseph et al. 2020). Because heart-attack cases often involve multiple care decisions at both the time of admission and after ED discharge, the procedures involved in coordinating heart attack care have been considered extensively by both healthcare (e.g., Currie et al. 2016; Ward et al. 2016; Kindermann et al. 2015) and operations management scholars (e.g., Youn et al. 2022, Lu and Lu, 2017; Theokary and Ren, 2011). Yet, empirical analysis of the discharge decision-making process and the impact of recent government and industry reform on ongoing coordination of care in the ED remains unclear (Dobrzykowski 2019).
Given the complexity of diagnosis and multi-step treatment required in EDs for heart attack cases, it is expected that modeling the decision-making processes in this area would rarely conform to a fixed array of treatment guidelines and procedures and would be highly prone to process workarounds depending on the surrounding environment (Tucker et al. 2014). This fact has also been clear from both the data and resulting analyses. For instance, some non-clinical factors such as the time of patient arrival at a hospital (Anderson et al. 2014), patient payer status with insurance (Ward et al. 2016; Kindermann et al. 2015; Spencer et al. 2013), hospital ownership status (Ding, 2014), government regulations (Ho et al. 2017), local population density (Jarman et al. 2016), and patient income (Hisam et al. 2016) have been found to affect healthcare clinical processes, outcomes, and costs in different ways for different groups of patients. Recent healthcare research studying operations has advocated for standardizing process routines to avoid workarounds, emphasizing the need for metrics to manage the multiple dimensions of conformance and experiential quality (e.g., Smith et al. 2022; Senot et al. 2016). Yet, wide differences observed in the discharge practices for patients facing similar medical conditions, particularly within similar ED settings, would appear to undermine these efforts.