Implementing a chatbot on Facebook to reach and collect data from thousands of health care providers: PharmindBot as a case

The world is moving fast toward digital transformation as we live in the artificial intelligence (AI) era. The COVID-19 pandemic accelerates this movement. Chatbots were used successfully to help researchers collect data for research purposes. To implement a chatbot on the Facebook platform to estab...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of the American Pharmacists Association 2023-09, Vol.63 (5), p.1634-1642.e3
Hauptverfasser: Alkoudmani, Ramez M., Ooi, Guat See, Tan, Mei Lan
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The world is moving fast toward digital transformation as we live in the artificial intelligence (AI) era. The COVID-19 pandemic accelerates this movement. Chatbots were used successfully to help researchers collect data for research purposes. To implement a chatbot on the Facebook platform to establish connections with health care professionals who had subscribed to the chatbot, provide medical and pharmaceutical educational content, and collect data for online pharmacy research projects. Facebook was chosen because it has billions of daily active users, which offers a massive potential audience for research projects. The chatbot was successfully implemented on the Facebook platform following 3 consecutive steps. Firstly, the ChatPion script was installed on the Pharmind website to establish the chatbot system. Secondly, the PharmindBot application was developed on Facebook. Finally, the PharmindBot app was integrated with the chatbot system. The chatbot responds automatically to public comments and sends subscribers private responses using AI. The chatbot collected quantitative and qualitative data with minimal costs. The chatbot's auto-reply function was tested using a post published on a specific page on Facebook. Testers were asked to leave predefined keywords to test its functionality. The chatbot's ability to collect and save data was tested by asking testers to fill out an online survey within Facebook Messenger for quantitative data and answer predefined questions for qualitative data. The chatbot was tested on 1000 subscribers who interacted with it. Almost all testers (n = 990, 99%) obtained a successful private reply from the chatbot after sending a predefined keyword. Also, the chatbot replied privately to almost all public comments (n = 985, 98.5%) which helped to increase the organic reach and to establish a connection with the chatbot subscribers. No missing data were found when the chatbot was used to collect quantitative and qualitative data. The chatbot reached thousands of health care professionals and provided them with automated responses. At a low cost, the chatbot was able to gather both qualitative and quantitative data without relying on Facebook ads to reach the intended audience. The data collection was efficient and effective. Using chatbots by pharmacy and medical researchers will help do more feasible online studies using AI to advance health care research.
ISSN:1544-3191
1544-3450
DOI:10.1016/j.japh.2023.06.007