University of Twente Student Theses


Aligning the work processes of the medical instrument sterilization cycle at the OLVG hospital in Amsterdam : a holistic approach

Vries, D.H.M. de (2017) Aligning the work processes of the medical instrument sterilization cycle at the OLVG hospital in Amsterdam : a holistic approach.

[img] PDF
Abstract:This assignment was conducted in the context of completing the master of Health Sciences at the University of Twente. Here we focused on the improvement of the logistical processes which are involved in the continued usage and sterilization of reusable medical instruments at the OLVG hospital in Amsterdam. The motivation to conduct this research was that the management of the OLVG believed that the hospital can achieve significant financial and qualitative gains by equipment and work process standardization. In order to study the instrument sterilization cycle we chose to formulate the following central research question: “What do the work processes of the instrument logistics cycle entail and what opportunities are there to optimize these processes to impact instrument storage, usage and maintenance?” This central research question can be broken down into two distinct parts for a problem solution process. Here the first part of the question is aimed at describing the current state of the instrument supply chain, while the second part tries to establish a future image of the sterilization cycle. To examine the current state of the instrument sterilization cycle we hypothesized that it is crucial to know a couple of things. First off we needed to know which processes are conducted by which department and we were interested to learn how these work processes are monitored. This question was answered by developing multiple flowcharts of the instrument sterilization cycle which were explained with supplementary process descriptions. Second we needed to identify the bottlenecks that cause the issues which are encountered at OLVG in order to offer possible solutions. This second part was answered by developing a gold standard for instrument sterilization management to which the current sterilization cycle was compared. Through this gold standard comparison, we discovered that instrument flow is currently organized in a sub-optimal way due to a lack of process planning. As a first step to remedy this issue, we proposed to develop a demand forecasting model in order to give an estimate of how many instrument trays would be required to cover demand. The forecast model was built as a proof of principle for the urology department, in which the demand for a selection of urologic procedures is coupled to the demand for the corresponding instrument trays. Utilizing this model we hoped to be able to give a correct estimation of the instrument tray supplies required to cover 97.5% of all possible instrument tray demand. In the end we were able to generate a Poisson based forecast for 29 out of the 84 possible instrument tray types, with an average demand varying between 0.1 and 8.9 trays per week. Using the forecasts for these 29 tray types we were able to cover 82.38% of all possible tray demand as recorded in a 10 week training dataset, opposed to the intended 97.5%. Forecast accuracy was determined over a separate 4 week testing dataset by means of a mean absolute scaled error (MASE). In this method a ratio is calculated between the obtained forecasting errors from our Poisson forecast method, relative to a reference forecasting error obtained from a reference forecasting method. Here a MASE smaller than 1 indicates a smaller forecast error relative to the reference error, whereas a MASE larger than 1 indicates the opposite. For our investigation we utilized an easier stochastic forecasting model over the average tray throughput in the 10 week training dataset as a reference method. Based on a calculated MASE of 0.98 +/- 0.26 for the 29 predicted tray types and a MASE of 0.63 +/- 0.55 over all 84 different tray types, we can conclude that our Poisson based forecasting model offers a forecasting accuracy equal to the more facile reference method. Based on the overall forecasting results, we were able to conclude that this forecasting model does not offer a valid representation of instrument tray demand for all instrument tray types. By expanding the selection of the procedures which we have included to generate the forecast, however, the validity of the model can be improved. Another, perhaps more pragmatic option for accurate instrument tray demand forecasting, is to make use of past instrument tray demand to directly forecast future tray demand. Next to the lack of a centralized instrument sterilization planning, the gold standard comparison revealed that the IT environment currently in place, is of limited use for measuring supply chain performance and thus enforcing a materials planning. This is caused by the fact that the IT environment is highly fragmented because a coherent information management strategy is currently lacking. As a secondary target for our problem solution, we therefore chose to formulate an information management strategy for the OLVG. Due to the expressed interest of the OLVG into Track and Trace (T&T), this technology should play a central role in the development of a new strategy. In order to implement an improved version of the forecasting model, the information management strategy should include all information required to create the aforementioned forecasting model. The OLVG should therefore not only document patient centered performance and quality measures, but should also include measures for procedure demand, instrument tray demand and instrument tray throughput to facilitate demand forecasting. To accommodate this, we hypothesized that it is crucial to address the fragmentation of the current IT systems which hampers error free information documentation. If we redesign the IT environment according to this supposition, we opine that it should consist out of two systems, each covering a distinct part of the instrument sterilization cycle. Here an OR T&T system should cover all the tasks, and record all the data related to the management of instrument tray stocks. A separate EPF system, on the other hand, should be used to document all patient related information. For the OR T&T system we also investigated which T&T technologies are suitable to facilitate its implementation. Through a literature research we were able to distinguish two possible candidate modalities being barcoding and RFID. After weighing all the advantages, disadvantages and cost considerations of the two modalities we reached the conclusion that separate instruments should be traced by the use of UDI complaint 2D barcodes whilst instrument trays can be traced by using RFID.
Item Type:Essay (Master)
Faculty:TNW: Science and Technology
Subject:44 medicine, 85 business administration, organizational science, 88 social and public administration
Programme:Health Sciences MSc (66851)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page