Healthcare Workers’ Experience with an Artificial Intelligence Virtual Assistant for Plastic Surgery Patients

Abstract

PURPOSE: Artificial intelligence virtual assistants (AIVA) are interactive voice response systems capable of understanding unconstrained speech using machine learning and natural language processing. Despite their use in the healthcare setting,1-3 the implementational aspects of these AIVAs are seldom studied. We created an AIVA capable of answering ten of the most frequently asked questions (FAQs) by Plastic Surgery patients regarding their procedure.4 We describe the results of an implementation-focused survey conducted on healthcare workers (HCWs) at our institution regarding the acceptability, adoption, and feasibility of the AIVA. METHOD: The HCWs were instructed on the topics the AIVA could answer and were then provided with a phone to call an office number linked to the AIVA. Once the HCWs had used the AIVA, they answered the survey. The survey consisted of 13 questions and was conducted on 17 HCWs from our institution’s Division of Plastic Surgery. It was answered using a 5-point Likert scale model ranging from ‘Completely Agree’ to ‘Completely Disagree.’ For analytical purposes, the positive items in the scale were merged into ‘Agree,’ and the negative items were combined into ‘Disagree.’ The remaining item was defined as ‘Neutral.’ Descriptive statistics were performed using R version 4.1.2. RESULTS: All the HCWs were Nurse Practitioners or Physician Assistants. The average age was 40.1±12.07 (Min. 24, Max. 60). There were 15 female (88.24%) and two male (11.76%). Out of 17 surveys, 15 were complete (88.24%). All interviewees agreed that this AIVA would be useful for patients. Individually, more than 75% of HCWs agreed that the AIVA could positively impact their work time (88.23%) and workflow (76.47%), be useful for appointment scheduling (88.24%), and serve to answer pre- (100%) and postoperative questions (94.12%). Furthermore, between 50% and 75% of HCWs thought the AIVA could positively impact their workload (70.58%), allow them to spend better time with patients (70.58%), work more efficiently (64.71%), and spend more time with new patients rather than follow-ups (52.94%). Only 17.65% thought it would bring more revenue to the institution. Lastly, 76.47% of HCWs considered that an AIVA would be necessary within the next ten years, and 81.25% considered this an environmentally friendly technology. CONCLUSION: These results suggest that HCWs in direct contact with patients consider that implementing an AIVA to answer FAQs would help improve their work environment and benefit patients pre- and postoperatively. REFERENCES: 1. Pauletto S, Balentine B, Pidcock C, et al. Exploring expressivity and emotion with artificial voice and speech technologies. Logoped Phoniatr Vocol. 2013;38(3):115-125. 2. Ireland D, Atay C, Liddle J, et al. Hello Harlie: Enabling Speech Monitoring Through Chat-Bot Conversations. Stud Health Technol Inform. 2016;227:55-60. 3. Sato A, Haneda E, Suganuma N, Narimatsu H. Preliminary Screening for Hereditary Breast and Ovarian Cancer Using a Chatbot Augmented Intelligence Genetic Counselor: Development and Feasibility Study. JMIR Form Res. 2021;5(2):e25184. 4. Boczar D, Sisti A, Oliver JD, et al. Artificial Intelligent Virtual Assistant for Plastic Surgery Patient’s Frequently Asked Questions: A Pilot Study. Ann Plast Surg. 2020;84(4):e16-e21.

Ähnliche Arbeiten