Medical and public health guidelines help healthcare providers choose the best possible treatment. However, these guidelines are often based on clinical research, while experiential knowledge could make them more relevant and applicable. Interviewing patient representatives is time-consuming and often impractical. Therefore, in a joint project between the Athena Institute, the Social AI group at Vrije Universiteit Amsterdam, and the RIVM (National Institute for Public Health and the Environment), Lösch investigated how digital methods can help map the experiences of patients, healthcare providers, and citizens.
Social media
“We analyzed experiences shared on social media, forums, and other online platforms,” says Lösch. “It turns out that these digital methods, such as artificial intelligence, can be a valuable tool for incorporating experiential knowledge into healthcare.”
Lösch and her colleagues used natural language processing — processing, analyzing, and generating language using AI — to collect and analyze online experiences shared by patients and healthcare providers. They did this together with guideline developers for COVID-19 vaccinations, scabies, and the quality standard for transgender care.
Integrating into guidelines
Gathering experiential knowledge alone is not enough. Therefore, the scientists combined digital methods with field research, interviews, participatory workshops, and observations during the guideline development process. Lösch: “This mix of methods allowed me not only to collect data but also to understand how this knowledge is used in practice and where the bottlenecks lie in integrating experiential knowledge into guidelines.”
Scabies
A concrete example of improved guidelines can be found in the case of scabies. Direct patient input led to practical adjustments in the recommendations. Lösch: “In the future, these digital methods can also be used in other guideline development processes, especially in situations where traditional forms of participation are difficult to implement, such as in crisis situations or with infectious diseases where patient organizations are often lacking.”
Lösch will defend her PhD research on September 9.