The IVF laboratory : from coping and workarounds to a structured and controlled model for lab processes to adhere to physiological time constraints using computer simulation

Kaffel, Aida (2025) The IVF laboratory : from coping and workarounds to a structured and controlled model for lab processes to adhere to physiological time constraints using computer simulation. Doctoral thesis (DClinSci), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (12MB) | Preview

Abstract

IVF laboratory procedures are dynamic and increasingly complex. Most procedures are manual and reliant on the IVF lab staff (embryologists and practitioners). Time spent carrying out procedures and timing in relation with oocyte retrieval are closely linked to performance (ICSI, oocyte cryopreservation, fertilisation). Labs are resourced using estimations to meet timing recommendations but they are often reliant on workarounds that are challenging to plan. Discrete event simulation (DES) is a computer modelling simulation tool used to understand and analyse workflows in systems subject to resource constraints and random variation. It is a tool that can help in identifying problems and testing potential change ideas virtually, and so is sometimes used to support quality improvement work. The aim of this project is to demonstrate the intricate link between staffing in the IVF lab, timing of IVF lab procedures and final IVF outcomes by examining the lab using a discrete event simulation (DES) software Simul8©. The study was carried out at Guy’s Assisted conception Unit ACU GSTT-ACU. The steps in the study were first map and model the IVF laboratory processes: replicating every touchpoint of the patient or their gametes/embryo journey in a Simul8© DES model, then validate the model by comparing retrospective data (2022) to data generated by the simulation. Following validation of the model, identify bottlenecks and deviations from optimal physiological timings, and ultimately test strategies (workload, staffing, equipment and technologies) to take to mitigate against deviations. Key variables generated by the simulation model were Processes metrics (PM) : Number of processes completed (egg collections, transfers, egg freezing, embryos freezing) within a time period. Bottlenecks was a variable represented by both queuing times for procedures and unfinished tasks at the end of staff shifts. The simulation model generated a variable called staff time utilisation (expressed in percentage) as a daily or yearly mean rate. Staff utilisation and job queue were compared with timing of procedures and KPI (IVF and ICSI fertilisation rate) in comparison with targets informed by the clinical literature. Following the validation stage, preliminary results showed a clear dynamic visualisation of processes (inputs and outputs with a timeline). Challenges and limitations with the modelling included representing deviations from set behaviours or unpredictable staff choices. Traditionally, IVF laboratory workload/resources are measured by number of weekly egg collections per whole-time equivalent (WTE) embryologists in post. Preliminary simulation results allowed a dynamic understanding of workload and resources in real time and raised awareness with stakeholders of the IVF lab complexity for allocating resources. The validation step focused on comparing the simulation model outputs to the 2022 real life data (egg collections, egg freezing, embryos transfers completed) using the same number of inputs and was successful delivering similar behaviour in most areas. The second validation step compared the time durations delivered by the simulation model versus real life data considering in the comparison that the model assumption allowed tasks to be carried out when the resources were available and travel time between tasks was not accounted for. Once validated, the aim of this project was to identify bottlenecks and test “what if scenarios” to improve patient outcomes, optimise workflow and balance workload with resources in the IVF laboratory. Scenarios included removing the andrology service, adding equipment, adding staff on busy days and adding 1 hour overtime for staff. The improvements observed with the different scenarios allowed identification of staffing bottlenecks that could be resolved. Challenges of modelling the IVF laboratory include difficulty incorporating embryologists’ and practitioners’ non-clinical tasks due to task difficulty to give a tangible time duration. This project is a first simulation modelling exploring a novel method to display the complexity of the lab workflows but also to analyse the complex dynamics and give answers to make improvements. The idea is to analyse the IVF laboratory differently to what has been done so far and bring solutions to a recurrent problem : carry out the lab procedures on time. Taking this idea forward in the IVF community would benefit 3 stakeholders: patients, staff and organisations.