Breast cancer-related fatigue risk and intervention recommendations: What can and cannot be personalised?
Lian Beenhakker is a PhD student in the Department of Biomedical Signals and Systems. (Co)Promotors are prof.dr. M.M.R. Hutten, prof.dr. S. Siesling and dr. A. Witteveen from the Faculty of Electrical Engineering, Mathematics and Computer Science.
Cancer-related fatigue (CRF) is one of the most common and underdiagnosed long-term effects after breast cancer. Many factors influence the development of CRF, however, on individual level it is unknown who is going to develop CRF. There are many interventions to reduce CRF, but unfortunately, not a gold-standard intervention that works best for all patients. These both aspects need personalisation, and so the goal of this thesis was to determine what can and cannot be personalised in risks for fatigue and intervention recommendations for breast cancer-related fatigue.
In Chapter 2, we studied the personalisation of the risk of developing CRF. It was not possible to accurately predict CRF, as CRF is a complex construct. So, in Chapter 3, focus groups with patients and interviews with healthcare professionals showed the complexity of CRF and the factors that are important to CRF.
In Chapter 4, an overview of existing interventions was created, to show the large variation and possibility to give a personalised intervention recommendation. In Chapter 5, breast cancer patients indicated their preferences for interventions and decision rules were developed to create a simple personalised intervention recommendation. In Chapter 6, we tried to predict intervention effectiveness on individual level, again to further personalise the intervention recommendation. As in Chapter 2, it showed to be difficult to predict CRF.
In the general discussion of Chapter 7, two themes emerged: the personalisation in predictions in fatigue, and the personalisation of intervention recommendations. For the first theme, improvements of the work of this thesis lies in either the data used, or the modelling approaches. For the second theme, we dived into the extension of the decision rules, and how patients can still receive a personalised intervention recommendation. Future research should focus on the standardisation of data to have a common method to measure CRF, if this is possible at all, and the implementation of the results of this thesis into clinical practice.
It can be concluded that based on current available data, personalisation in predictions in fatigue is not accurately possible, while in the personalisation of intervention recommendations, first important steps were made.
