By Dr Alice Le Bonniec, Behavioural Insights & Change Specialist, National Screening Service
We’ve published new research about designing and testing a chatbot-based cervical screening decision aid. It was designed to support women facing socioeconomic disadvantage in France to make decisions about taking part in cervical screening, including HPV self-sampling. We developed a user-friendly tool that women could use to ask questions about cervical cancer screening, in their own language and at their own pace.
We worked with the World Health Organization’s International Agency for Research on Cancer (IARC) on this research. The project was led by Dr Farida Selmouni, Public Health Officer at IARC.
Why we did the study
Women from lower-income or underserved communities are often less likely to take part in cervical screening. A lack of tailored, accessible information can be a barrier to taking part. Women may not understand screening options, may worry about doing things incorrectly, or face language, cultural, or literacy barriers.
We wanted to design a tool that could help bridge that information gap and empower women to make decisions about cervical screening, especially HPV self-sampling.
What we did
We developed a chatbot called ‘AppDate-You’ aimed at helping women to make informed decisions about cervical cancer screening,
We used a user-centred design approach. We got input from women and health professionals from the start. We refined the tool based on feedback.
Participants in the study included:
- women aged 30 to 65 years living in disadvantaged areas in the Occitanie region of France
- health professionals - GPs, midwives, gynaecologists - working with this population group.
We developed and tested the chatbot in 3 phases:
- Exploratory phase - interviews and focus groups to understand barriers, needs, and preferences.
- Prototype testing - trial the chatbot version with women and health professionals to get feedback.
- Field testing - test the refined chatbot with women in real-world settings to assess feasibility, usability and acceptability.
We built the chatbot to work on WhatsApp and Facebook Messenger, so there was no need for women to install a separate app.
The tool:
- supports text and voice interactions
- includes multimedia content, for example videos and infographics
- aims to be simple to use, making it easier for people with low digital literacy.
What we found
Feedback and usability
- Users generally found the chatbot easy to use, friendly and informative, including those with limited experience using digital tools.
- The integration into WhatsApp was especially appreciated - many already use it so it lowered the barrier to trying the tool.
- Some limitations emerged: sometimes medical terms were not clear enough and responses were sometimes repeated.
Interest in self-sampling
- In the field testing, 80% (12 out of 15) of women expressed interest in HPV self-sampling after interacting with the chatbot.
- Participants engaged with the chatbot for about 10 minutes every day, asking questions and looking at videos and infographics.
Surprising insights and barriers
- Some women who were older or assumed to be less tech-savvy were willing to use the tool, challenging assumptions about digital literacy.
- Concerns remained about women’s confidence in doing self-sampling correctly, fears of injury, and preference for professional consultation in some cases.
- Health professionals had doubts about reach - whether the tool would reach those most disadvantaged or with low literacy.
Conclusion
This study shows that designing a chatbot-based decision aid with users in mind can lead to a tool that is acceptable, feasible, and valued by its intended audience. The positive feedback and interest in HPV self-sampling suggest such tools may help reduce barriers to cervical cancer screening among underserved women.
Next steps
This is an early step. The sample was small and outcomes like actual screening uptake were not measured in this study. We’re planning a larger study to see whether using this chatbot actually increases screening participation.
If successful, chatbots like this could become a model for how digital tools help support decision-making and access to healthcare.
- Le Bonniec A, Sauvaget C, Lucas E, Nassiri A, Selmouni F; Design and Validation of a Chatbot-Based Cervical Cancer Screening Decision Aid for Women Experiencing Socioeconomic Disadvantage: User-Centered Approach Study; JMIR Cancer 2025;11:e70251; doi: 10.2196/70251