In:
International journal of health sciences, Universidad Tecnica de Manabi, ( 2022-06-18), p. 5048-5061
Abstract:
Despite suicide's prominence as a worldwide killer, research into the phenomenon has been slowed by several obstacles. First, self-reports of suicidal thoughts and intentions have been almost exclusively used in previous attempts to predict suicide. This is problematic since there are well-known reporting biases in self-reported data. It is particularly troublesome in the case of suicide since many persons are driven to conceal their suicidal intentions to avoid being hospitalised. The Bayesian models were created by employing a retrospective cohort technique. Predictors of future reported suicide conduct were established using electronic health record information from a large healthcare database encompassing 15 years (1998-2012. The effectiveness of the models was evaluated retroactively by utilizing a separate testing data set. Suicide risk assessment may benefit from longitudinal EHR data, often available in clinical settings. Automated risk screening tools may improve prediction beyond what is possible for human clinicians by considering the whole phenotypic range of the EHR.
Type of Medium:
Online Resource
ISSN:
2550-696X
,
2550-6978
DOI:
10.53730/ijhs.v6nS4.2022
DOI:
10.53730/ijhs.v6nS4.9271
Language:
Unknown
Publisher:
Universidad Tecnica de Manabi
Publication Date:
2022
