Attitudes of Healthcare Professionals and Egyptian Users toward the Use of Health Chatbots in Analyzing Health Data and Personalizing Treatment Recommendations

The study aims to investigate the trends of healthcare specialists and Egyptian users, regarding the use of health chatbots in analyzing health data and the personalization of therapeutic recommendations. Additionally, the potential for the future success of AI in this area is assessed, while proposing ways to enhance the media's role in shaping trends towards the adoption of these technologies, as the world is currently experiencing health challenges.
The study employed “Cognitive Computing Theory” that provides a transformative approach to AI, with the aim of simulating human cognitive processes. It is an analytical descriptive study, employing Mixed Methods Research, to conclude generalizable scientific findings, and to adequately explain the phenomenon under consideration. It also employed a semi-experimental approach through the One-Group Posttest-Only Design. Data were collected using a semi-structured in-depth interview tool and a questionnaire, in conjunction with big data analysis tools.
The study sample consisted of two parts:

A convenience sample of (16) individuals comprising (8) doctors from various specialties and (8) specialists in Flow Cytometry Analysis and Biotechnology, which enhanced the accuracy of the assessments regarding the performance of the health chatbot.
A purposive sample of (100) individuals for the quasi-experimental study, ensuring diverse demographic characteristics and prior experience with chatbots.

The findings of the study showed that:

A large number of respondents agreed to employ health robots in disease prediction and health data analysis, along with an increased confidence rate in robots’ role in improving the quality of health care.
Rapid response, accurate diagnosis, and ease of use are the most influential factors in respondents’ adoption of health chatbots, while language interaction has a limited impact.
Younger age groups (18–30 years) trusted chatbots more than older groups, and lower-income groups exhibited greater trust compared to higher-income groups.
The SVM model achieved the highest accuracy in predicting the adoption of health robots at 90%, outperforming the decision tree model and the hybrid model. This means that the SVM model is more effective and accurate when classifying users.

The Study recommended that:

Media should be employed to raise awareness about the benefits of health chatbots, with an emphasis on rapid response, accurate diagnosis, and improved health care quality to boost users’ confidence in these technologies.
Health institutions need to adopt effective digital communication campaigns to educate the public about the operational mechanisms and practical benefits of health chatbots, showcasing real-life success stories.