When artificial intelligence facilitates the analysis of comments from hospitalized patients

When artificial intelligence facilitates the analysis of comments from hospitalized patients

2024-10-31 11:55:00

The High Authority for Health (HAS) has developed a new tool for providing healthcare establishments with all the results of the national system for measuring the satisfaction and experience of hospitalized patients. Among the new features, an artificial intelligence algorithm automatically classifies patient verbatims to facilitate their analysis. This tool has been made available free of charge to all health establishments since the end of September.

e-Satis, a national system allowing patients to express themselves and establishments to improve

Since 2015, HAS has been piloting a national system for measuring patient satisfaction and experience, e-Satis (1). It makes it possible to integrate the patient point of view into the processes of improving the quality and safety of care.

Concretely, an online survey is carried out among all patients after hospitalization, with questionnaires adapted according to the type of hospitalization (ambulatory surgery, other short stays, rehabilitation stays). The questions relate to the perception of reception, care, hotel services and exit. The questionnaires are essentially composed of structured, so-called “closed” questions. At the end, patients can also write a free comment on their hospitalization, called “verbatim”.

Until now, the HAS provided quantitative results to establishments based on responses to closed questions but did not provide an analysis of the verbatim statements. Establishments could download all of the free comments, but it was difficult to analyze them without a suitable tool to sort and visualize them.

Or, as shown a national study published by HAS in 2022 (2), This qualitative feedback provides valuable information for implementing actions to improve the quality of care, in addition to the quantitative results. This study made it possible, from the analysis of more than 2 million verbatims from e-Satis, to obtain a classification of the themes mainly reported by patients. : the importance of a fluid, coordinated and reassuring care circuit; the desire for hotel quality more adapted to the needs of the patient and more personalized, and, above all, the importance of professionalism assessed through the quality of the information transmitted and human contact. The quality of human relationships, essential for patients, was considered a prerequisite for these 3 elements.

A new tool for reporting e-Satis results, which integrates an artificial intelligence algorithm to more easily analyze patient comments

In response to health establishments requesting optimization of the restitution of e-Satis results, HAS, in partnership with the Hospices Civils de Lyon, initiated the development of a new tool at the end of 2022, offering graphic visualizations and advanced analytics capabilities. A working group composed of user representatives and healthcare quality professionals was formed to participate in the design. The group’s particular objective was to restore all the results from the questionnaires, by coherently linking the quantitative scores and the qualitative assessments expressed by patients.

A classification composed of themes and sub-themes has been defined to analyze the verbatims. This classification includes the main dimensions reported by patients in the e-Satis surveys and identified in the 2022 national study. An automatic language processing algorithm (artificial intelligence) was then developed to automatically classify the verbatims into these themes, and thus offer filter and analysis functionalities to health establishments in the application.

With the provision of this new tool, HAS encourages professionals to rely on the wealth of feedback contained in patients’ verbatim statements, and health establishments to increase their participation in the e-Satis system in a objective of improving the service provided.

After a pilot phase carried out this summer, this new tool is made available free of charge in September. It is intended for health establishments concerned by the e-Satis system and for health professionals who work there; to user representatives and professionals interested in analyzing the patient experience.

A “HAS quality meeting” will be organized online on Tuesday, November 5, with the participation of HCL, in order to discuss with users of this tool. To find out more and register, you can consult the dedicated page.

(1) Find out more on the national e-Satis system

(2) Read the National report “Experience of hospitalized patients in France: national analysis of free comments on the e-Satis system.

1730638967
#artificial #intelligence #facilitates #analysis #comments #hospitalized #patients

**Interview with Dr. Émilie‍ Roche, Project Lead at HAS**

**Interviewer:** Welcome, Dr. Roche! Today,​ we’re excited to‌ dive into the new e-Satis tool developed by the High⁢ Authority for Health. ‍Can you explain the motivation behind⁣ this project?

**Dr. Roche:** Thank ‌you for having me! The motivation for developing the e-Satis tool stems from a growing need to better understand patient experiences in healthcare facilities. Traditionally, while we had quantitative data from⁢ structured questionnaires, we lacked the means to effectively analyze ⁤the rich qualitative feedback provided by patients in ⁤their comments. Our ⁢aim was to integrate ​these insights to‍ help hospitals improve the quality and safety of care.

**Interviewer:** That’s fascinating!‍ Can you elaborate on how the new AI algorithm enhances this process?

**Dr. Roche:** Absolutely. The AI ⁤algorithm we implemented classifies patient ‌verbatims into predefined themes and sub-themes, which allows healthcare establishments to ‍easily visualize and analyze the feedback. This classification helps⁤ identify key areas where improvements can be made, tailored to patient needs. It⁤ transforms raw data into actionable ‌insights.

**Interviewer:** It sounds like this could significantly‌ impact healthcare​ quality. ⁤What⁢ are some‍ key themes you’ve observed in patient feedback?

**Dr. Roche:** Based on our analysis, some prominent themes include the necessity for a coordinated and reassuring care circuit, a desire for personalized hotel services, and the importance of professionalism—particularly the quality of information and human interactions. These insights ‍underscore that the human ​aspect of care is‌ vital for patients.

**Interviewer:** How has the response been from healthcare establishments since launching⁣ the new tool?

**Dr. Roche:** The response has been overwhelmingly positive! Since its launch, many healthcare establishments have expressed ⁣gratitude for the‍ ability to analyze patient feedback⁤ more comprehensively. The visualizations and detailed analytics⁢ help them understand not just what scores patients give, but the nuances behind those scores.

**Interviewer:** This tool is available free of charge, correct? How important was that decision?

**Dr. Roche:** Yes,‍ making it available for free was essential for us. We want to ensure that all healthcare facilities, regardless of their resources, can benefit from these insights. Our priority is to enhance patient care across the board,‌ and accessibility plays a key role‍ in achieving that.

**Interviewer:** Thank you, Dr. Roche, for sharing your insights on this⁤ important⁣ project. It’s encouraging to see technology driving improvements in ⁣patient ⁣care.

**Dr. Roche:** Thank you for the opportunity to discuss our work! We believe that by harnessing tools like AI, we can genuinely elevate the patient experience‍ in ​healthcare.

Leave a Replay