Marcela Tabares Tabares, Edgar León Landa, Consuelo Vélez Álvarez, Paula Andrea Carmona Gallego and Joshua Bernal Salcedo
Adv. Artif. Intell. Mach. Learn., 4 (3):2665-2686
Marcela Tabares Tabares : Universidad de Caldas
Edgar León Landa : Universidad Internacional de la Rioja (UNIR), España
Consuelo Vélez Álvarez : Grupo Promoción de la Salud y Prevención de la Enfermedad, Universidad de Caldas, Colombia.
Paula Andrea Carmona Gallego : Grupo Telesalud, Universidad de Caldas, Colombia.
Joshua Bernal Salcedo : d Fundación para el Fomento de la Investigación Sanitario y Biomédica de la Comunidad Valenciana Unidad Mixta de Imagen Biomédica FISABIO-CIPF Valencia, España.
DOI: https://dx.doi.org/10.54364/AAIML.2024.43155
Article History: Received on: 17-Jul-24, Accepted on: 21-Sep-24, Published on: 28-Sep-24
Corresponding Author: Marcela Tabares Tabares
Email: marcela.tabares@ucaldas.edu.co
Citation: Marcela Tabares Tabares, Edgar León Landa, Consuelo Vélez Álvarez, Paula Carmona Gallego, Joshua Bernal Salcedo. (2024). Can Chatbots Alleviate Depression? Results of a Systematic Review. Adv. Artif. Intell. Mach. Learn., 4 (3 ):2665-2686
Purpose: This
article systematically reviews the impact of chatbots on the treatment of
depression, evaluating studies published between January 2019 and April 2024.
Depression, exacerbated by factors such as the COVID-19 pandemic, requires
accessible and effective treatments. Chatbots, using artificial intelligence
and natural language processing, emerge as accessible alternatives, offering
interventions based on cognitive-behavioral therapy (CBT). Methods: Using
databases such as Cochrane Library, PubMed, Scopus, and ScienceDirect, 321
articles were identified, of which 12 met the inclusion criteria. These studies
evaluated changes in depression symptoms using validated instruments in
individuals who interacted with chatbots. Results: The results indicate
that chatbots can significantly reduce depression symptoms, although their
effectiveness varies depending on the design and implementation of the
intervention. A lack of gender balance and variations in sample sizes and
intervention durations, ranging from one to 16 weeks, were found.
Methodological limitations include selection bias and lack of clear information
on random allocation and blinding. Conclusions: The bibliometric
analysis highlights the interrelation of key terms such as "chatbot,"
"depression," and "cognitive-behavioral therapy,"
underscoring the importance of advanced technologies in developing these tools.
Despite their potential, chatbots should not be considered a substitute for
professional treatment in severe cases. This study suggests that chatbots are
promising for psychological support, but more research is needed to optimize
their clinical effectiveness and user acceptance.