Neuropsychiatric symptoms (NPS) are common in affected individuals and can be challenging for (in)formal caregivers. Therefore, they are also referred to as challenging behaviors (CBs). Sensor technology measuring context and behavior can be assistive to effectively manage CBs in an objective fashion. Sensors can help support healthcare professionals, such as nurses, by enabling remote monitoring and alarming on early-stage behavioral changes associated with CBs. This might/ will improve the quality of life (QoL) for both caregivers and clients living in a nursing homes (NH).

In the project “MOnitoring Onbegrepen Gedrag bij Dementie met sensortechnologie” (MOOD-Sense), we aim to develop such a monitoring system. Our research focuses on two questions 1) How to develop and implement a monitoring system within the context of nursing homes with parameters on environment, physiology, and behavior, identify and process relevant precursors of challenging behavior with this monitoring system and 2) gain insight in which behaviors are challenging according to nurses and how they are described. This will be represented in an ontology such that sensor data can be translated into the same conceptual information.

The first research question will be examined with a set of experiments in the field (in NH) with an iterative approach. Insights from previous experiments on usability and added value of sensors will be used to improve successive experiments. During each experiment, multiple participants (clients with dementia and CBs) are monitored with both ambient and wearable sensors. For the second research question a qualitative approach is employed, using focus groups (FG) and consensus methods. These FGs will be held amongst nursing staff who are involved in daily care tasks for people with dementia. Subsequently, consensus methods are used to align behavioral descriptors/labels.

early findings will be presented at the symposium

Within this project we expect to find precursors of challenging behavior in a personalized fashion based on nurse’s expert knowledge and sensor data. In order to develop a monitoring system that can be embedded within NH’s, real-time alarming, in-situ behavior recognition and trustworthiness are part of our technological requirements. Just-in-time interventions may then be deployed to prevent behavior escalation or the persistence of undesirable situations.