Patient Experience and Artificial Intelligence

Oriane Simon
Share this article

Understanding the Patient Experience

What is the Patient Experience?

A patient's experience is not limited to the treatments received and the procedures performed, but also includes all the extra-medical elements: connection with the people around him or her, well-being and dignity, listening to the staff, etc. According to the Beryl Institute's definition, the Patient Experience is the sum of all the interactions, shaped by the culture of an organization, that influence the perception of the patient across the continuum of care.

Why talk about Patient Experience?

If the Patient Experience is today one of the subjects at the heart of the reflections of the Health actors, it is in particular because it can have a direct influence on the evolution of the patient's health.

Indeed, several studies [1][2] report that, in an ICU patient, "anxiety and fear are very present during the stay" with "anxious symptoms (...) in 52% [of the patients in the panel at the time of discharge from an ICU]". The study also mentions post-traumatic stress in 33% of families of patients in intensive care.

More generally, the care of a patient in a medical or medico-social structure involves several stress factors, such as noise [3] to which many sleep disorders are linked [4]. According to these studies [5][6], they can "negatively influence the course of the operation and the postoperative phase" to the point of "requiring the administration of more medication", and also "the hospital stay can be prolonged due to complications, or delayed healing or recovery".

How to improve the Patient Experience?

Patients are among the best placed to talk about their experiences. Therefore, a major lever for better understanding their expectations is to collect and analyze their feedback. There are two complementary types of approach: those that rely on close collaboration with a small number of committed users (patient-trackers, patient-experts, etc.) and those that wish to make use of the opinions of as many people as possible (questionnaires, listening centers, suggestion boxes, etc.). At the center, the user representative has the task of bringing the opinion of the patients of the establishment.

These approaches are complementary, and are worth combining. The remainder of this article focuses on approaches based on exploiting the opinions of as many people as possible, with a particular emphasis on free expression - verbatim - which is a much richer means of expression for the patient than the score alone. Indeed, a patient who, for example, gives his room a mark of 3 out of 10 does not explain the source of his dissatisfaction, which can be multiple: the availability of blankets, the noise in the middle of the night, the comfort of the mattress... Textual feedback, on the other hand, enables us to precisely identify the subject of concern, so that we can effectively remedy it. Having said that, it's far more laborious to analyze textual (qualitative) feedback than quantitative (ratings): this is where Artificial Intelligence comes in, and more specifically the dedicated branch of Natural Language Processing.

At Better World, we are passionate about the new opportunities offered by advances in Artificial Intelligence to help institutions collect and analyze feedback from patients and professionals in healthcare settings.

Collect feedback

The importance of the number

The first step in leveraging the Voice of the Patient is to collect their opinions. The greater the volume of feedback collected and the size of the contributing population, the more analysis will provide a reliable and comprehensive view of the Patient Experience and pinpoint areas to focus efforts. Maintaining this collection over time and storing the history will help monitor changes in patient satisfaction and measure the impact of actions taken.

The proposed devices

In this sense, since 2016, the French National Authority for Health (HAS) has been offering a national device whose role is to measure the satisfaction and experience of hospitalized patients as part of an approach to improving the quality of care: e-Satis. The system takes the form of a series of multiple-choice questions for patients, supplemented by free expression, which allows them to give a rich textual opinion on areas of satisfaction or areas for improvement. The HAS offers institutions an analysis of the answers to the multiple choice questions.

In addition to this system, there are a number of initiatives carried out by facilities or available on the web: "in-house" surveys, complaints forms, suggestion boxes, interviews, opinion sites, forums... There is no shortage of possibilities. What's more, the democratization of QR codes generated by the health crisis has also enabled innovative initiatives to emerge; at Better World, we now offer facilities a continuous listening platform that patients can access at any time during their stay by flashing QR codes placed around the facility.

Analyzing verbatim reports

Classify the content

Each Patient Experience is unique and multi-dimensional at the same time, so each review can concern several aspects of a stay or consultation: the comfort of the room, the quality of the staff's attention, the management of the file, the reception of relatives, etc. Identifying and classifying these elements manually in each verbatim can be a long and time-consuming process, but a branch of Artificial Intelligence, and more specifically Automated Language Processing, is dedicated to this type of problem. Thus, it is possible to automatically segment the opinions to extract the topics discussed, saving precious time for operational teams and giving visibility on the topics of concern to patients to all actors of the institution.

Analyze the feeling

Knowing what subject the patient is expressing himself on is certainly interesting, but in order to really understand his expectations, it is still necessary to be able to say whether the patient is expressing himself positively or negatively on this subject: this is the role of the sentiment analysis, which, once again, calls upon Artificial Intelligence techniques. Following this step, the segments of verbatims identified during the classification are enriched with a sentiment score that quantifies the patient's satisfaction and provides the institution with a detailed understanding of the patient's feelings on each subject.

Share results and create action plans

Following the analysis phase, the unit results obtained can be aggregated at the desired level (department, establishment, group, etc.) to provide satisfaction indicators with the desired precision, quantifiable over time, offering a detailed understanding of patients' feelings on each of the subjects addressed.

Automating the analysis of verbatims provides reliable, rich and easily exploitable results over time. It thus contributes to the dissemination of the quality culture at all levels of the organization, making the subject of the Patient Experience a central strategic work axis for the teams.

Caregivers: drivers of this transition

Developing symmetrical listening between patients and professionals has several quickly identifiable interests: holistic understanding of the experience in the institution, centralization of everyone's ideas to improve the health system, contribution to a dynamic of commitment and participation...

Beyond these subjects, there is a strong correlation between the improvement of the quality of life at work of professionals and that of the Patient Experience, as we will detail in a future article: the care given to the well-being of teams has a proven impact [5][6] on the quality of patient care. Thus, the improvement of this well-being is today a major challenge, which requires work on several key issues, including: the perceived disconnect between the theoretical functioning of the hospital and the reality in the field, the feeling of a lack of appreciation of work, etc. A study conducted by SHAM (Société Hospitalière d'Assurances Mutuelles) reveals that 35% of the professionals in the health and medico-social sector questioned said they were "discouraged" when they started their working day, 34% "stressed" and 33% "exhausted".

Better World is the French leader in listening to patient verbatimwith 600+ healthcare and medical-social establishments covered.

[1] Pochard F (2010) Recognizing and treating psychological suffering in the ICU. Réanimation 19:236-42

[2] De Miranda S, Chaize M, Megarbane B, et al (2009) Consequences of ICU admission on COPD patients and families. Am J Respir Crit Care Med 179:A5474

[3] Kalfon P, Mimoz O, Auquier P, et al (2010) Development and validation of a questionnaire for quantitative assessment of perceived discomforts in critically ill patients. Intensive care Med 10:1751-8

[4] Rotondi AJ, Chelluri L, Sirio C, et al (2002) Patients' recollections of stressful experiences while receiving prolonged mechanical ventilation in an intensive care unit. Crit Care Med 30:746-52

[5] Krohne, de Bruin, Mohiyeddini, Breimer & Schäfer, 2000 and de Bruin, 2001, Higgins et al, 2009, Stirling, 2006

Let's build your Action Plan

Get personalized and actionable advice to build your action plans and track your target audience's satisfaction over time.


You have a project in mind?

Please fill out this form, and a member of our team will contact you within 24 hours.

Thank you! Our teams will get back to you as soon as possible!
Oops! Sorry, a problem has occurred.