Integral capacity management & planning in hospitals
Due to the COVID-19 crisis measures the PhD defence Thomas Schneider will take place (partly) online in the presence of an invited audience.
The PhD defence can be followed by a live stream.
Thomas Schneider is a PhD student in the group Mathematics of Operation Research (MOR). His supervisors are prof.dr. R.J. Boucherie from the Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS) and prof.dr.ir. E.W. Hans from the Faculty of Behavioural, Management and Social Sciences (BMS).
Care pathways in hospitals usually encompass multiple resources and healthcare pro- fessionals. This makes managing hospital processes and capacities challenging. To prevent myopic optimization, process improvements should consider multiple steps in care pathways. This dissertation aims to improve complex decision-making that in- tegrally manages capacity for care pathways. Operations research may play a crucial role by analyzing such capacity decisions in a safe environment before actual imple- mentation. However, despite the vast amount of available research and its potential, it appears that the actual implementation of operations research models and results in healthcare practice is rarely described in the literature. This is surprising, as imple- mentation is the ultimate step in realizing improvement. We try to improve this final step by distinguishing two approaches: (1) organizing the timing and alignment of the optimal decisions among related capacities and (2) analyzing (near) optimal capacity decisions considering multiple capacities.
Part I Integral Capacity Management in Hospitals
We start this thesis analyzing the organization of capacity decisions in hospitals in Chapter 2. We observe that current capacity management (CM) in hospitals orga- nizes departments as silos, or even as single cost centers, with their own operations management systems and a top-down deployment of decision-making processes. We aim to realize this potential by breaking through the siloed system, by optimizing flow rather than myopically optimizing utilization. We do this by aligning capacity in care pathways. We propose integral capacity management (ICM) as the successor to CM. This is the first theoretical introduction of ICM. We distinguish three dimensions for organizational integration: hierarchical, patient-centeredness, and domain. We discuss alignments on and between these dimensions to integrally organize capacity decisions. Hierarchical integration concerns top-down and bottom-up decision-making processes, in which higher levels set boundaries, targets and planning objectives (i.e. increasingly disaggregated information) for lower levels and lower levels provide input for improvement of decision-making on higher levels. Patient-centeredness concerns the coordination and alignment of capacity across departments and organizations to optimize care pathways. Domain integration encompasses alignment of managerial domains: clinical, financial and nonrenewable resources. This study is a first step for theoretical development of ICM. We therefore derive multiple directions for future research.
In Chapter 3 we review operations research (OR) literature applied to hospital wards. Based on logistical characteristics and patient flow problems, we distinguish the fol- lowing particular ward types: intensive care, acute medical units, obstetric wards, weekday wards, and general wards. We analyze typical trade-offs of performance in- dicators for each ward type, review OR models commonly applied to it, and discuss typical capacity management and planning decisions. Additionally, we provide three illustrative cases, discuss both theoretical and practical challenges, and provide di- rections for future research. With this review we aim to guide both researchers and healthcare professionals dealing with hospital bed capacity and on which OR models best suits each specific capacity decision and the type of ward.
Part II Integral Capacity Planning in Hospitals
In Chapter 4 we analyze the process of emergency admissions. The increasing number of admissions to hospital emergency departments (EDs) during the past decade has resulted in overcrowded EDs and decreased quality of care. The emergency admission flow that we discuss in this study relates to three types of hospital departments: EDs, acute medical unit (AMUs), and inpatient wards. The study in this chapter has two objectives: (1) to evaluate the impact of allocating beds in inpatient wards to accom- modate emergency admissions and (2) to analyze the impact of pooling the number of beds allocated for emergency admissions in inpatient wards. To analyze the impact of various allocations of emergency beds, we develop a discrete event simulation model. We evaluate the bed allocation scenarios using three performance indicators: (1) the length of stay in the AMU, (2) the fraction of patients refused admission, and (3) the utilization of allocated beds. We develop two heuristics to allocate beds to wards and show that pooling beds improves performance. The partnering hospital has embedded a decision support tool based on our simulation model into its planning and control cycle. The hospital uses it every quarter and updates it with data on a 1-year rolling horizon. This strategy has substantially reduced the number of patients who are re- fused emergency admission.
Chapter 5 analyzes optimal surgery schedules considering multiple resources. Surgery groups are clustered surgery procedure types that share comparable characteristics (e.g. expected duration). Scheduling operating theater (OT) blocks leaves many op- tions for operational surgery scheduling and this increases the variation in usage of both the OT and downstream beds. Therefore, we schedule surgery groups to reduce the options for operational scheduling, ultimately bridging the gap between tactical and operational scheduling. We propose a single step mixed integer linear programming (MILP) approach that approximates the bed and OR usage along with a simulated annealing approach. Both approaches are compared on a real-life data set and results show that the MILP performs best in terms of solution quality and computation time. Furthermore, the results show that our model may improve the OR utilization from 71% to 85% and decrease the bed usage variation from 53 beds to 11 beds compared to historical data. To show the potential and robustness of our model, we discuss several variants of the model requiring minor modifications. The use of surgery groups makes it easier to implement our model in practice as, for operational planners, it is instantly clear where to schedule different types of surgery.
Chapter 6 presents an innovative methodology to overcome Markov decision process (MDP) intractability for online multi-appointment scheduling problems. As a result of increasing treatment options and far-reaching specialization, the number of appoint- ments for patients has increased. This makes appointment schedules fragile as depen- dencies between schedules increase. Decomposing the decisions (e.g. accept/reject and allocation decisions) allows us to analytically solve practical online multi-appointment scheduling problems. We use an MDP to derive optimal decisions for accepting or re- jecting new arrivals based on capacity availability and future arrivals. Once accepted, we developed an ILP to allocate patients to their next appointment. Based on a case study at the Leiden University Medical Center cardiology outpatient clinic, we then present the implementation of our model for a real-life instance. We compare the per- formance of our approach with a heuristic and show that our approach outperforms that of the heuristic. Furthermore, we show a full implementation and analyze the impact in practice.
Future developments for Hospital Capacity Management & Planning
We see multiple future directions for ICM and planning in hospitals. In Western coun- tries, most hospitals have emerged from the digitization era and are now discovering the value of the newly available information. This will explode the number of research opportunities for all types of analytics (i.e. descriptive, predictive and prescriptive). To increase their impact in practice, researchers should embrace data-driven optimiza- tion. For example, both descriptive and predictive analytics may be used to improve the input data for prescriptive analytics and therefore improve the results of prescrip- tive analytics. Prescriptive analytics quickly become intractable when the number of decisions increases. Therefore, the number of decisions resulting from descriptive and predictive analytics should be balanced to ensure tractable prescriptive analyt- ics. This may result in decreasing quality of both descriptive and predictive analytics. Analyzing trade-offs between quality measures of different types of analytics could be an interesting research topic. Capacity planning and management software systems that in real-time automate all steps of capacity planning and management may be developed using data-driven optimization. Ultimately, this may further reduce the waste caused by failures in care coordination.
Currently, regulation of data is organized well. As the aforementioned scalability of innovations is finally taking place, the exchange of data and information has proven to be difficult as there are still compatibility and data ownership problems. ”Data is the new oil” . First and foremost, health data will increase the quality of care as information availability improves. For operations research this will raise opportunities as more data about the total care pathways of patients becomes, to some extent, available through, for example, wearables.
In the coming years, clinical practice will become more technical as a result of the digitization era, technical medicine and the introduction of health analytics. The first small steps are currently being taken. For example, predictive medicine already sup- ports medical decision-making on antibiotic dosage in sepsis for intensive care patients . This also has an impact on capacity planning and management. Therefore, both analytical and medical scholars should further embrace each other’s field of expertise. Together, they can integrally shape the hospital of the future.