Process mining in healthcare management

In this article, we discuss process mining applicability in healthcare management.

Process mining in healthcare management

The provision of quality hospital services depends on the suitable and efficient execution of processes, understood as a series of activities aimed to diagnose, treat and prevent diseases and improve a patient’s health. Changes in processes – such as the adherence to the Enhanced Recovery After Surgery protocol [1] or the prompt administration of appropriate antibiotics in the management of patients with sepsis [2] – have proven its worth to improve outcomes.

These processes are supported by clinical and non-clinical activities, executed by different types of professionals, and can vary from one organization to another, according to the resources and infrastructures available. However, reality often deviates from the clinical practice guidelines. That’s why, in this article, we are going to explore process mining - a method for Business Process Management (BPM) and analysis, focused on extracting knowledge from data generated and stored in corporate information systems in order to analyze executed processes – and its applicability in healthcare.

First things first: what does process mining mean and what does this concept brings to healthcare?

Process mining is gaining increasing attention in healthcare. Adopted in the domain of BPM, it is useful to extract knowledge and non-trivial information from the process executions, to discover, monitor, and improve processes. It provides a comprehensive set of tools to return fact-based insights and, thus, to support process improvement.

In short, process mining applied to healthcare involves 3 variables: processes - recommended set of activities to be performed by clinical and non-clinical teams -; performers - people that execute processes -; and data (technically named event log) - all the insights recorded before, during, and after interventions, that is to say the real execution of processes, including deviations from recommended actions that happen in practice.

A literature review performed by Rojas et al. identified and classified the most relevant common aspects in which the application of process mining is useful in healthcare. Today, we would like to highlight the following:

1. Process mining methods to identify different types of hospital processes

  • The Medical Treatment Processes are the clinical processes responsible for managing patients, including tasks ranging from diagnosis to the execution of actions for alleviating each patient.
  • The Organizational Processes focus on the organizational understanding of processes, capturing collaborative information from professionals and their organizational units.

2. Process mining methods to provide specialists in medical processes the ability to respond to frequently posed questions about these processes

  • What happened: identifying the need to discover the process executed and its activities.
  • Why did it happen: understanding the activities and circumstances characterizing the situation/action.
  • What will happen: identifying the circumstances of when or how a specific activity will take place.
  • What is the best that can happen: identifying possible steps towards specific improvements.

That being so, process mining ensures that not only the healthcare processes are well-understood, but also allows to achieve a high level of process efficiency. Healthcare organizations can discover how processes are executed in reality, check whether certain practices were really followed, and gain insights into resource utilization and performance-related aspects. At this point, we’ve arrived at the three main current types of process mining: [3]

1. Discovery is related to inferring process models that are able to reproduce the observed behavior.

2. Conformance checking allows comparing an a priori model with the observed behavior, to analyze if the reality conforms with the process model.

3. Extension/Enhancement corresponds to the projection of the information extracted from the observed behavior onto the model, in order to enrich it.

Clinical pathways on the spotlight

The first two dimensions mentioned – discovery and conformance checking – can be applied to implement and evaluate clinical pathways compliance. In a first moment to design or redesign processes within a healthcare unit, and after for performance analysis and conformance checking.

This perspective is particularly interesting when it comes to assessing the compliance between a clinical pathway - the term assigned to the evidence-based medical treatment processes of operational nature that healthcare professionals should act in accordance with - and the real medical behaviors.

Clinical pathways provide detailed guidance for each stage in the management of a patient (treatments and interventions), but multiple variances may still unavoidably occur due to individual complexities and subjectivity. [4] Due to the gap observed between what is stated by updated scientific evidence and effective care settings, clinical pathway compliance checking is of utmost importance in the healthcare environment.

For this purpose, when applying process mining techniques and technologies, healthcare organizations are able to compare a given process model with all process executions in reality. The objective is to output a set of compliance metrics to measure the observed deviations. Gathering insights on the actual clinical practice is receiving increasing attention, due to the potential of the information extracted. In the mentioned scenario, clinical pathways compliance analysis is the process of discovering knowledge about how clinical activities impact the patients careflow and using the discovered knowledge for: [6]

  1. Care pathway redesign
  2. Care pathway optimization
  3. Clinical decision support
  4. Medical deviation detection
  5. Business management

Main takeaways:

Despite the difficulties due to the richer diversity and complexity of medical behaviors over other business processes, discovering process models and analyzing their performance provides relevant opportunities for extracting knowledge out of the abundant information recorded in a hospital’s EHR.

  • Process mining’s wide applicability in healthcare can help to construct or redesign the processes, to analyze and improve the performance and collaboration between physicians, to identify which activities are bottlenecks, and to add alternative or supplementary data to the activities of the process.
  • Compliance monitoring is a crucial step to be considered when a healthcare organization intends to accomplish a higher degree of process maturity.
  • Optimized processes are the guarantee for a sustained and long-lasting outcomes improvement.

References:

[1] Ripollés-Melchor J, Ramírez-Rodríguez JM, Casans-Francés R, et al. Association Between Use of Enhanced Recovery After Surgery Protocol and Postoperative Complications in Colorectal Surgery: The Postoperative Outcomes Within Enhanced Recovery After Surgery Protocol (POWER) Study. JAMA Surg. 2019;154(8):725–736. doi:10.1001/jamasurg.2019.0995

[2] Seymour, Christopher & Gesten, Foster & Prescott, Hallie & Friedrich, Marcus & Iwashyna, Theodore & Phillips, Gary & Lemeshow, Stanley & Osborn, Tiffany & Terry, Kathleen & Levy, Mitchell. (2017). Time to Treatment and Mortality during Mandated Emergency Care for Sepsis. The New England journal of medicine. 376. 10.1056/NEJMoa1703058.

[3] Rojas, E., Munoz-Gama, J., Sep´ulveda, M., and Capurro, D. (2016). Process mining in healthcare: A literature review. Journal of Biomedical Informatics, 61:224–236.

[4] Yan, H., Gorp, P. V., Kaymak, U., Lu, X., Ji, L., Chiau, C. C., Korsten, H. H. M., and Duan, H. (2018). Aligning Event Logs to Task-Time Matrix Clinical Pathways in BPMN for Variance Analysis. IEEE Journal of Biomedical and Health Informatics, 22(2):311–317.

[5] Braithwaite Jeffrey. Changing how we think about healthcare improvement BMJ 2018; 361:k2014

[6] Huang, Z., Lu, X., and Duan, H. (2012). On mining clinical pathway patterns from medical behaviors. Artificial Intelligence in Medicine, 56(1):35–50.

[7] Barros, Sebastião. Conformance Checking for Care Pathway Compliance Assessment [Tese]. [Lisboa]: Instituto Superior Técnico; 2018.