ODP168 Can Virtual Assistants Help our Diverse Population of Diabetic Patients?

Abstract

Abstract Background: Diabetes affects 34 million in the US and an additional 88 million are prediabetic. 1 According to the CDC, Diabetes affects Hispanic Americans (22%) and Asian Americans (19%) more than the Caucasian (12%) population. 2 Many of these patients do not speak English. In addition, 55% of US adults did not graduate from high school. 3 Health literacy is a barrier to health care. There are 4.2 billion Virtual Assistant AI devices, e. g. Amazon Echo Dot and Google Home Mini, which are used worldwide. Virtual assistant devices are used in nursing homes and assisted living facilities. 4 Purpose: To evaluate Virtual Assistant Devices, Alexa and Google, as a educational tool for our multicultural Diabetic Population Methods: Evaluation of Amazon Echo Dot (Alexa) and Google Home Mini (Google) with voice activation in English, Spanish and Mandarin was performed. Alexa did not have Mandarin Chinese speaking capability. Responses from both devices were recorded and then transcribed to text using Google Docs "speech to text" tool and assistance from native Mandarin Chinese and Spanish speakers. Transcribed English language responses were pasted into WebFx.com and readabilityformulas.com for readability analysis via Flesch-Kincaid Reading Ease (FRE), Flesch-Kincaid Grade Level Score (FGL), Simple Measure of Gobbledygook Index (SI), and the Dale-Chall Adjusted Grade Level (DCGL). Verbal Questions: 1) I have diabetes- what do I do? 2) I have Type 1 Diabetes Mellitus - What do I do? 3) I have Type 2 Diabetes - What do I do? 4) Does diabetes cause me to have problems seeing? 5) Does diabetes cause kidney problems? 6) Does diabetes cause problems with my feet? Results: Word Counts: English: Google (36.7 words +/- 10.4), Alexa (39.2 words +/- 25.2) Spanish: Google (31.2 words +/- 13.6), Alexa (45.2 words +/- 27.9) Chinese: Google (73.3 words +/- 31.6), Alexa (NA) No significant differences between the number of words for English vs. Chinese responses or English vs. Spanish responses. English Responses Readability: Avg FRE: Google (51.3 +/-12. 0), Alexa (71.1 +/- 9.9) (higher = easier reading) Avg FGL: Google (9.8 grade +/-1.7), Alexa (6.5 grade +/- 0.8) (higher = higher grade level) Avg SI: Google (9.5 +/-1.5), Alexa (6.5 +/-0.9) (higher = harder reading) Avg DCGL: Google (College), Alexa (Grades 9-10) Conclusion: Digital Voice Assistant Devices are helpful to patients who are visually impaired, have poor mobility or need language translation. Under physician guidance, these devices provide diabetes education for our patients and for the lay public.