Back

Influence of AI-based decision support on shared decision making in hemodialysis: a Wizard of Oz experiment

Osmanodja, B.; Samhammer, D.; Roller, R.; Beck, S.; Burchardt, A.; Duettmann, W.; Dabrock, P.; Gerndt, S.; Moeller, S.; Ronicke, S.; Budde, K.

2023-01-14 health informatics
10.1101/2023.01.12.23284468 medRxiv
Show abstract

Artificial Intelligence-based decision support systems (AI-DSS) to improve hemodialysis therapy are currently under development. However, the influence of AI-DSS on shared decision making (SDM) in hemodialysis patients has not been studied so far. We performed a Wizard of Oz experiment, using a sham AI-DSS suggesting ultrafiltration volume at the beginning of each dialysis session. We performed 10 patient interviews in 5 patients, and investigated views towards AI, different aspects of the SDM, and the influence of an AI-DSS on SDM in a real-life scenario. Five main topics were identified: (1) the patient as self-determined, (2) role of the medical staff, (3) other forms of automation, (4) attitude towards AI, (5) needs and preferences for future use of AI. The patients describe novel AI-DSS in dialysis as an opportunity to promote self-determination and to ensure more efficient therapy. At the same time, they describe the special relationship with the nursing staff and physicians, who, from the patients point of view, must be given control over an AI-DSS. This study provides first evidence regarding the influence of AI-DSS on SDM in the context of hemodialysis. Further studies should focus on other aspects that require SDM such as initiation of dialysis, and selection of dialysis modality.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.