JMIR Publications

ABSTRACT

Background:

According to the United States Food and Drug Administration (FDA) Center for Biologics Evaluation and Research, healthcare systems have been experiencing blood transfusion overuse. Patients are given more blood than required which makes the patients prone to immunological reactions, including hemolysis and acute lung injury, and also to circulatory volume overuse and acute heart failure.

Objective:

To minimize the overuse of blood product transfusions, a proprietary artificial intelligence-based Blood Utilization Calculator (AI-BUC) was developed and integrated into a US hospital’s electronic health record. This qualitative study explores how clinicians perceived this AI-based decision support system.

Methods:

We interviewed ten clinicians (BUC users) until the data saturation point was reached. The interviews were conducted over a virtual platform and were recorded. The Audio-Visual recordings were then anonymously transcribed verbatim. We used an inductive-deductive thematic analysis to analyze the transcripts and applied predetermined themes to the data (deductive), and consecutively identified new themes as they emerged in the data (inductive).

Results:

We identified three themes (a) workload and usability, (b) safety, and (c) clinical decision-making. Clinicians acknowledged the usefulness of BUC for the general inpatient patient population. However, their perception regarding workload and decision-making were not consistent. Some clinicians reported safety concerns with BUC and viewed the technology as unsuitable for critical patients.

Conclusions:

The findings of this study highlight the importance of usability and user-centered design. Different clinicians had different needs from the BUC, and the fact that the system was not designed to cover all patient types, hindered its use in the hospital. AI such as the BUC, if not designed for individual users at the department level, will not be utilized as intended.