Introducing artificial intelligence in acute psychiatric inpatient care: qualitative study of its use to conduct nursing observations

Abstract

BackgroundAll patients admitted to an acute inpatient mental health unit must have nursing observations carried out at night either hourly or every 15 minutes, to ascertain that they are safe and breathing. However, while this practice ensures patient safety, it can also disturb patients’ sleep, which in turn can impact negatively on their recovery.ObjectiveThis article describes the process of introducing artificial intelligence (‘digitally assisted nursing observations’) in an acute mental health inpatient ward, to enable staff to carry out the hourly and the 15 minutes observations, minimising disruption of patients’ sleep while maintaining their safety.FindingsThe preliminary data obtained indicate that the digitally assisted nursing observations agreed with the observations without sensors when both were carried out in parallel and that over an estimated 755 patient nights, the new system has not been associated with any untoward incidents. Preliminary qualitative data suggest that the new technology improves patients’ and staff’s experience at night.DiscussionThis project suggests that the digitally assisted nursing observations could maintain patients’ safety while potentially improving patients’ and staff’s experience in an acute psychiatric ward. The limitations of this study, namely, its narrative character and the fact that patients were not randomised to the new technology, suggest taking the reported findings as qualitative and preliminary.Clinical implicationsThese results suggest that the care provided at night in acute inpatient psychiatric units could be substantially improved with this technology. This warrants a more thorough and stringent evaluation.

Applications of artificial intelligence to improve patient flow on mental health inpatient units - Narrative literature review

“…Nurses are required to take regular observations, which range from hourly to close proximity supervision at all times [ 52 ]. However, these have also been shown to disturb the patient, particularly at night, lowering their chances of a faster recovery, which can then lead to longer length of stay [ 53 ]. Barrera et al.…”

“…Barrera et al. [ 53 ] used AI to introduce digitally assisted nursing observations which improved both patient and staff experiences. The tool allowed nurses to take observations remotely via a sensor utilising computer vision, signal processing and AI to observe micromovements that allow for pulse and breathing rate calculation.…”

“…Future work could focus on the discovery of both objective biomarkers for development of targeted treatments and objective ways to measure therapy's effectiveness [ 53 ]. Studies on AI in therapy aim to enhance decisions and personalise interventions to improve recovery and allocate resources efficiently.…”

A Concept Analysis on the Use of Artificial Intelligence in Nursing

“…A more complex definition of AI is “multiple technologies that can augment human activities in the form of machine learning, to process and learn with raw data and deep learning, [sic] to stimulate decision-making using complex artificial neural networks” [ 25 ]. Definitions are currently scattered with only three articles formally defining AI [ 24 , 26 , 27 ], six not offering any formal definition [ 6 , 7 , 25 , 28 - 30 ], three defining only machine learning [ 4 , 14 , 31 ], three defining only neural networks [ 5 , 16 , 32 ], and one defining only decision trees [ 15 ]. Therefore, the definitions of AI used in the literature include machine learning, neural networks, and decision trees.…”

“…Continuous development is necessary, as the real world can never be truly quantified as a finite number of variables interacting with each other. Thus, development never ends, as long as the AI is in use within a nursing context and all require careful evaluation and implementation [ 28 ]. In addition, there is a need to continuously quantitatively test the algorithm, to ensure that there are no errors, and to compare between competing AIs [ 16 ].…”