Principal Investigator - Dr Sally Wheelwright
Researcher - Dr Jo Middleton
Funder: Motor Neurone Disease Association
Sponsor: University of Sheffield
Introduction
Motor neurone disease (MND) is a life-shortening neurological condition that affects the nerves controlling movement. It causes progressive muscle weakness, which can affect how a person walks, talks, eats, drinks and breathes. Symptoms vary from person to person and usually get worse over time, but support, treatment and equipment can help people live as well as possible. As people living with MND (plwMND) experience progressive decline, they are faced with several complex decisions around medical and care interventions. One reason decision-making is challenging is that it is difficult to predict when interventions are going to be needed. Healthcare professionals face uncertainty about the best time to start having conversations about interventions, as they try to balance providing enough time for quality decision-making whilst not overwhelming people with information about several forthcoming decisions simultaneously.
A project led by Professor Chris McDermott at the University of Sheffield is using machine learning to try and improve prognostic predictions. This involves applying mathematical algorithms to large sets of data to examine relationships between multiple predictors (e.g., patient demographic and clinical characteristics) and the probability of future clinical outcomes, including when plwMND would require a wheelchair and when they would benefit from a gastrostomy feeding tube. The current project will investigate how to use the outcomes from Professor McDermott's project in clinical practice.
This project is a collaboration between the Department of Neuroscience at University of Sheffield and SHORE-C at University of Sussex.
Aim
The aim of this project is to promote personalised patient-centred care, support HCP practice, and enhance shared decision-making in MND. This will be achieved by gaining an in-depth understanding of i) the complex needs, attitudes, and preferences of plwMND, and their informal carers, towards personalised prognostic information about their care; ii) HCP practices and preferences for applying a prediction model in their clinical practice; and iii) adapting an existing patient decision aid supporting people make an informed decision around gastrostomy insertion.
Method
The project includes three work-packages: (WP1) Stakeholder values, attitudes and preferences regarding prognostic information and modelling; (WP2) Co-design development of a web-based platform integrating a prediction model, its outputs and information; (WP3) Preliminary evaluation of the implementation of the integrated time-to-event prognostic model in clinical practice.
Impact
This study will support a new approach to communicating individual prognostic information to people living with MND in the context of the UK healthcare services. Knowing the likely timing a decision about gastrostomy insertion would need to be made will help people consider their options and plan without pressure. Knowing the likely timing of loss of ambulation for individuals would allow service planners to organise the provision of wheelchairs in a proactive manner.
For related work see:
Wheelwright S, Maunsell R, Taylor S, et al. (2023) Development of 'gastrostomy tube - is it for me?', a web-based patient decision aid for people living with motor neurone disease considering having a gastrostomy tube placed. Amyotroph Lateral Scler Frontotemporal Degener 24:605-613.
Current Stage:
Recruiting