Mapping the Delirium Literature Through Probabilistic Topic Modeling and Network Analysis: A Computational Scoping...

Area of Science:

• Medical Informatics
• Psychiatry
• Bibliometrics

Background:

• Delirium is an acute confusional state linked to increased morbidity and mortality in medically ill patients.
• It is a core topic in consultation psychiatry, recognized in professional competencies and educational materials.

Purpose of the Study:

• To conduct a computational scoping review of the delirium literature.
• To identify the overall trends and evolution of delirium research from 1995 to 2015.

Main Methods:

• Algorithmic analysis of 3591 delirium research articles indexed in MEDLINE (1995-2015).
• Utilized network analysis of citation Medical Subject Headings (MeSH) tags and probabilistic topic modeling of abstracts.
• Compared publication trends and topic distribution between psychiatric and non-psychiatric journals.

Main Results:

• Delirium publications increased annually from 40 (1995) to 420 (2015).
• Treatment articles (806) outnumbered prevention articles (432).
• Identified key research themes including delirium in geriatric, critically ill, palliative, and postsurgical patients, diagnostic criteria, and risk factors, with uneven distribution across journal types.

Conclusions:

• The scientific literature on delirium is substantial and growing, with a notable increase in non-psychiatric publications.
• Algorithmic identification of subtopics reveals distinct research clusters within the delirium field.
• The distribution of these subtopics varies significantly between psychiatric and non-psychiatric journals.