Effect of Artıfıcıal Intellıgence Based Mobıle Vırtual Assıstant (AI)

Effect of Artıfıcıal Intellıgence Based Mobıle Vırtual Assıstant Developed For Indıvıduals Wıth Dıabetes On Cost Effıcıency, Hospıtal Admıssıon Rate, Self-Care And Hypoglycemıa

Aim: This study was conducted experimentally to examine the effect of artificial intelligence-based mobile virtual assistant developed for individuals with diabetes on cost, hospitalization rate, self-care and hypoglycemia.

Methods: The research is multi-stage and designed as three stages in itself. According to this; development of the mobile application in the first and second stages and adding artificial intelligence to the application as a project; In the third stage, it was planned to examine the effect of the application on the variables and scales. The data of the study were collected between June 2022 and June 2023 in the Endocrinology Polyclinic of two private hospitals in Izmir and a diabetes association where individuals with diabetes were registered. Power 0.80 was determined by using NCSS PAS statistical software from the population of the research; The minimum number of samples to be included in the study was calculated as n:122 and they were divided into two as intervention and control groups by randomization. The research sample was carried out as intervention (n:60) and control (n:60) lastly due to death and cost. Five data collection tools were used, namely "Individual Introduction Form", "Diabetes Self-Care Scale", "Hypoglycemia Confidence Scale", "Mobile Application Opinion Form" and "Cost Table". An artificial intelligence-based mobile virtual assistant application was applied to the individuals with diabetes in the intervention group, and the data were collected three times, at the 0th, 6th and 12th months, and the costs were recorded. The standard outpatient trainings, which are currently applied, continued to be given to individuals with diabetes in the control group, the data were collected twice, at the beginning (0. month) and 12. months, and the costs were recorded. In the evaluation of the data, number, percentage, arithmetic mean, standard deviation, minimum and maximum median were calculated. Among the variables, chi-square, Kruskal Wallis, Mann Whitney U test and t test were used.

Study Overview

• Status: Completed
• Conditions: Diabetes
• Intervention / Treatment: Other: artificial intelligence based mobile application

• Study Type: Observational
• Enrollment (Actual): 1

Contacts and Locations

Study Locations

• Turkey
  • Izmır, Turkey, 35620
    • Izmir Katip Çelebi University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Probability Sample

Study Population

• The patient has a perception disorder and psychiatric disorder that prevents communication,
• Desire to leave the research,
• His death during the research,
• Being pregnant,
• Individuals with severe retinopathy and neuropathy that prevent smartphone use were excluded from the study.

Description

Inclusion Criteria:

• Having been diagnosed with Type 1 and Type 2 diabetes at least six months ago, according to the criteria of the American Diabetes Association (ADA),
• Using insulin for at least six months,
• Being between the ages of 18- 65,
• Being able to read and write and speak Turkish,
• Having an Android phone and being able to use mobile applications,
• To volunteer to participate in the study.

Exclusion Criteria:

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
experimental group; Artificial intelligence-based mobile application initiative was implemented for diabetes patients	Other: artificial intelligence based mobile application; Artificial intelligence-based mobile application developed by me that includes diabetes education for individuals with diabetes.
control group; no intervention was applied

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diabetes Self-Care Scale	The score consists of 35 items and is a 4-point Likert type, the lowest acceptable score is 92 and the highest score is 140. As the score increases, self-care increases	12 months
Hypoglycemic Confidence Scale	The scale consists of 9 items and is a 4-point Likert type. There is no cut-off value, the average score is used.	12 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
hospitalization rate	percentage	12 months

Collaborators and Investigators

• Sponsor: Izmir Tinaztepe University

Study record dates

Study Major Dates

• Study Start (Actual): June 10, 2022
• Primary Completion (Actual): January 10, 2023
• Study Completion (Actual): August 7, 2023

Study Registration Dates

• First Submitted: August 10, 2023
• First Submitted That Met QC Criteria: October 5, 2023
• First Posted (Actual): October 12, 2023

Study Record Updates

• Last Update Posted (Actual): October 12, 2023
• Last Update Submitted That Met QC Criteria: October 5, 2023
• Last Verified: October 1, 2023

More Information

Terms related to this study

• Keywords: Diabetes; artificial intelligence; hospitalization; mobile application; cost; self-care
• Additional Relevant MeSH Terms: Glucose Metabolism Disorders; Metabolic Diseases; Endocrine System Diseases; Diabetes Mellitus
• Other Study ID Numbers: ikçü (Other Identifier: Katip Çelebi University)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No