Related Experiment Video
Updated: Aug 4, 2025
06:21
Concept Development and Use of an Automated Food Intake and Eating Behavior Assessment Method
Published on: February 19, 2021
5.8K
Collecting Food and Drink Intake Data With Voice Input: Development, Usability, and Acceptability Study.
Louise A C Millard1,2, Laura Johnson1,2,3,4, Samuel R Neaves1,2
1Medical Research Council (MRC) Integrative Epidemiology Unit, University of Bristol, Bristol, United Kingdom.
JMIR Mhealth and Uhealth
|March 31, 2023
Summary
Voice-based systems like Amazon Alexa show potential for real-time data collection in epidemiology studies. While some interaction issues exist, participants found the system acceptable for future research.
Area of Science:
- Epidemiology
- Human-Computer Interaction
- Health Informatics
Background:
- Traditional epidemiological data collection relies on infrequent questionnaires, potentially leading to errors and incomplete information.
- Voice-based systems offer continuous, real-time self-reported data collection, a novel approach for epidemiological studies.
- The use of voice-based systems like Amazon Alexa for collecting epidemiological data has not been previously evaluated.
Purpose of the Study:
- To assess the technical feasibility of using Amazon Alexa for collecting participant data in research.
- To investigate participant acceptability of voice-based data collection methods.
- To conduct an initial evaluation of the validity of data collected via Alexa, using food and drink intake as an example.
Main Methods:
- Recruited 45 participants (staff and students) at the University of Bristol for a 7-day study.
- Participants reported food and drink intake via Amazon Alexa and a parallel web-based form.
- Collected demographic information and participant feedback on their experience and system acceptability.
Main Results:
- Out of 37 participants with valid data, 81% were aged 20-39, and 62% were female.
- Alexa entries matched corresponding web entries for food and drink information in 60.1% of cases.
- Participants reported frequent interjections from Alexa, particularly during food/drink entry (49%), but 80% expressed willingness to use voice systems in future research.
Conclusions:
- Despite interaction challenges due to Alexa's conversational nature and interjections, participants showed a high willingness to engage in future research using this technology.
- The study demonstrates the technical feasibility and participant acceptability of using voice-based systems for data collection in epidemiological research.
- Further research is recommended to explore less conversational interfaces for voice-based data collection systems in scientific studies.
