An analysis of the sensitivity and stability of patients’ preferences can lead to more appropriate medical decisions

Sava, M. Gabriela; Vargas, Luis G.; May, Jerrold H.; Dolan, James G.

doi:10.1007/s10479-018-3109-3

Cited by 8 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A new sensitivity and stability analysis of preferences elicited using ANP models was proposed by Sava et al (2020a). Their work analyzes, in the perturbation space, not just the impact of simultaneous changes applied to the criteria weights on the final ranking of the alternatives, but which changes are necessary in order to modify the decision of selecting one alternative over another.…”

Section: Background and Research Methodology Motivationmentioning

confidence: 99%

“…The two measures were originally introduced by Sava et al (2020aSava et al ( , 2020b to enable the practical implementation of the full-dimensional stability analysis of preferences. As the decision-maker obtains additional information about the criteria, the information results in perturbations applied to the criteria weights via a continuous function.…”

Section: Full-dimensional Stability Analysis For Anp Modelsmentioning

confidence: 99%

“…The full‐dimensional stability analysis of preferences provides insight into the potential changes or shifts that might occur. Previous work related to the effect of additional changes in the criteria weights considered one, two and three simultaneous changes (May et al, 2013; Sava et al, 2020a). When analyzng the possible changes of the decision maker's preferences, however, it is not enough to perturb three or fewer criteria weights, because the majority of ANP models developed correspond to situations characterised by many more than three criteria.…”

Section: Stability Analysis Of the Preferences For Anp Modelsmentioning

confidence: 99%

“…Thus, the initially most preferred alternative

A_{i}

contains the origin of the perturbation space, the point with coordinates

((),,, 0, \dots, 0)

. Sava et al (2020a) showed that the perturbation space is compact, that is, it does not contain holes, and as mentioned above it is simply connected, that is, one can move from one point in the space to any other point in the space by means of a continuous transformation. Hence, the region in which an alternative is most preferred must also be simply connected, and it cannot consist of two disjointed regions.…”

Section: Stability Analysis Of the Preferences For Anp Modelsmentioning

confidence: 99%

“…We model the impact of the provider's responses to those questions using a full-dimensional stability analysis of the patient's preferences (Sava et al, 2020b). Previous work related to understanding the patient's preferences considered only two-and three-dimensional results (Sava et al, 2018(Sava et al, , 2020a.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Implications of the stability analysis of preferences for personalised colorectal cancer screening

Sava

Stanciu

Dolan

et al. 2021

Multi Criteria Decision Anal

Self Cite

View full text Add to dashboard Cite

Patients are increasingly interested in becoming involved in the medical decisionmaking process. As a result, healthcare providers and researchers are concerned with finding new ways to integrate patients' preferences, by understanding their commitment to and the stability of those preferences. Preventive medicine, such as colorectal cancer screening, presents an opportunity for personalising the decision-making trajectory based on patients' preferences. In this paper, we propose a framework for a joint decision-making process, capable of integrating patients' changing preferences, as described by a stability analysis of those preferences and design scenarios for implementing the process in clinical practice. In this study, a secondary data analysis, we present scenarios that demonstrate how the stability analysis of an Analytic Network Process (ANP) model supports personalising the process of agreeing on an appropriate colorectal cancer screening option. We illustrate the framework using two patients whose preferences have different stabilities and for whom the healthcare provider has different rankings for the recommended alternatives. The results show the differences in additional medical information the healthcare provider might need to provide as part of the joint decision-making process in order to reach an agreement between the patient and the provider. A stability analysis of the patient's preferences provides the healthcare provider with a mapping of the preferred options. It also shows how the patient's most preferred alternative might change as the patient obtains additional relevant medical information. Knowing how the patient's priorities might change supports a personalisation of the medical decision-making process. We conclude that the healthcare provider can utilise the stability analysis of a patient's preferences to identify possible dialogue paths that would enable reaching a consensus about an appropriate screening option.

show abstract

Section: Background and Research Methodology Motivationmentioning

confidence: 99%

Section: Full-dimensional Stability Analysis For Anp Modelsmentioning

confidence: 99%