Leveraging AI/Technology in Survey Design, Deployment, and Analysis
Author(s)
Jordana Schmier, MA1, Sayeli Jayade, MPH2;
1OPEN Health, Director, New York City, NY, USA, 2OPEN Health, New York City, NY, USA
1OPEN Health, Director, New York City, NY, USA, 2OPEN Health, New York City, NY, USA
OBJECTIVES: Automation or artificial intelligence (AI) may alleviate certain challenges in survey research in the healthcare field, specifically in survey design, testing and refinement, and distribution/data collection. Currently there is no summary of either the technologies that are in use in healthcare survey research or the extent to which these technologies have been successful in improving quality or efficiency. Our objective was to identify and describe the current use of AI/technology in these focused areas of survey research.
METHODS: Targeted literature search of PubMed, Google Scholar, and Google to identify uses of automation, artificial intelligence, and machine learning in market research between 2020-2024. Findings were organized according to technology used, its role in survey research, and whether they are used in healthcare. Each was then evaluated using a framework that considered opportunities, feasibility, and risks for use in healthcare.
RESULTS: We identified multiple technologies used in each of the three areas of survey research. In survey design, machine learning algorithms have been used to analyze past survey data to identify possible sources of bias and to refine items. AI has also been used to develop instructions. Further, web accessibility evaluation tools have been used to expand survey participation. AI has a multifaceted role in survey testing and refinement, being used to assess aspects of programming such as skip logics and branching and for adaptive testing to create individualized surveys based on participants’ responses to previous questions in the survey. Examples of AI/automation in survey distribution and data collection include the use of chatbots to improve survey delivery, real-time checks for fraud, and requesting communication preferences to implement effective reminders.
CONCLUSIONS: We identified successful examples of AI and automation in market research. While some of these have been widely adopted in market research, their acceptability and feasibility in healthcare research is not yet established.
METHODS: Targeted literature search of PubMed, Google Scholar, and Google to identify uses of automation, artificial intelligence, and machine learning in market research between 2020-2024. Findings were organized according to technology used, its role in survey research, and whether they are used in healthcare. Each was then evaluated using a framework that considered opportunities, feasibility, and risks for use in healthcare.
RESULTS: We identified multiple technologies used in each of the three areas of survey research. In survey design, machine learning algorithms have been used to analyze past survey data to identify possible sources of bias and to refine items. AI has also been used to develop instructions. Further, web accessibility evaluation tools have been used to expand survey participation. AI has a multifaceted role in survey testing and refinement, being used to assess aspects of programming such as skip logics and branching and for adaptive testing to create individualized surveys based on participants’ responses to previous questions in the survey. Examples of AI/automation in survey distribution and data collection include the use of chatbots to improve survey delivery, real-time checks for fraud, and requesting communication preferences to implement effective reminders.
CONCLUSIONS: We identified successful examples of AI and automation in market research. While some of these have been widely adopted in market research, their acceptability and feasibility in healthcare research is not yet established.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
MSR15
Topic
Methodological & Statistical Research
Topic Subcategory
Survey Methods
Disease
No Additional Disease & Conditions/Specialized Treatment Areas