Assessment of the implementation of a chatbot-based screening for burnout and COVID-19 symptoms among residents during the pandemic

Página do item simplificado
dc.creatorBruno Nascimento Moreira
dc.creatorAlexandre Sampaio Moura
dc.creatorAleida Nazareth Soares
dc.creatorZilma Silveira Nogueira Reis
dc.creatorRosa Malena Delbone
dc.date.accessioned2025-07-22T16:24:42Z
dc.date.accessioned2025-09-08T23:57:45Z
dc.date.available2025-07-22T16:24:42Z
dc.date.issued2023
dc.format.mimetypepdf
dc.identifier.doihttp://dx.doi.org/10.4300/JGME-D-22-00920.1
dc.identifier.issn1949-8357
dc.identifier.urihttps://hdl.handle.net/1843/83740
dc.languageeng
dc.publisherUniversidade Federal de Minas Gerais
dc.relation.ispartofJournal of graduate medical education
dc.rightsAcesso Aberto
dc.subjectCOVID-19 (Doença)
dc.subjectBurnout (Psicologia)
dc.subjectPandemia
dc.subjectSaúde pública
dc.subjectInteligência artificial
dc.titleAssessment of the implementation of a chatbot-based screening for burnout and COVID-19 symptoms among residents during the pandemic
dc.typeArtigo de periódico
local.citation.epage381
local.citation.issue3
local.citation.spage378
local.citation.volume15
local.description.resumoBackground: Early identification of COVID-19 symptoms and burnout among residents is essential for proper management. Digital assistants might help in the large-scale screening of residents. Objective: To assess the implementation of a chatbot for tele-screening emotional exhaustion and COVID-19 among residents at a hospital in Brazil. Methods: From August to October 2020, a chatbot sent participants’ phones a daily question about COVID-19 symptoms and a weekly question about emotional exhaustion. After 8 weeks, the residents answered the Maslach Burnout Inventory-Human Services Survey (MBI-HSS). The primary outcome was the reliability of the chatbot in identifying suspect cases of COVID-19 and burnout. Results: Among the 489 eligible residents, 174 (35.6%) agreed to participate. The chatbot identified 61 positive responses for COVID-19 symptoms, and clinical suspicion was confirmed in 9 residents. User error in the first weeks was the leading cause (57.7%, 30 of 52) of nonconfirmed suspicion. The chatbot failed to identify 3 participants with COVID-19 due to nonresponse. Twelve of 118 (10.2%) participants who answered the MBI-HSS were characterized as having burnout by the MBI-HHS. Two of them were identified as at risk by the chatbot and 8 never answered the emotional exhaustion screening question. Conversely, among the 19 participants identified as at risk for emotional exhaustion by the chatbot, 2 (10.5%) were classified with burnout, and 5 (26.3%) as overextended based on MBI-HHS scores. Conclusions: The chatbot was able to identify residents suspected of having COVID-19 and those at risk for burnout. Nonresponse was the leading cause of failure in identifying those at risk.
local.identifier.orcidhttps://orcid.org/0000-0002-4818-5425
local.identifier.orcidhttps://orcid.org/0000-0002-2671-3661
local.identifier.orcidhttps://orcid.org/0000-0001-6374-9295
local.identifier.orcidhttps://orcid.org/0000-0001-7740-8408
local.publisher.countryBrasil
local.publisher.departmentMED - DEPARTAMENTO DE PROPEDÊUTICA COMPLEMENTAR
local.publisher.departmentMED - DEPARTAMENTO DE SAÚDE MENTAL
local.publisher.initialsUFMG
local.url.externahttps://meridian.allenpress.com/jgme/article/15/3/378/493664/Assessment-of-the-Implementation-of-a-Chatbot

Arquivos

Pacote original

Agora exibindo 1 - 1 de 1
Imagem de Miniatura
Nome:
Assessment of the Implementation of a Chatbot-Based Screening for Burnout and COVID-19 Symptoms Among Residents During the Pandemic.pdf
Tamanho:
110.69 KB
Formato:
Adobe Portable Document Format
Baixar

Licença do pacote

Agora exibindo 1 - 1 de 1
Imagem de Miniatura
Nome:
License.txt
Tamanho:
1.99 KB
Formato:
Plain Text
Descrição:
Baixar

Coleções

Artigo de Periódico

Logo UFMG