We tested the ability of our natural language processing (NLP) algorithm to identify delirium episodes in a large-scale study using real-world clinical notes.

We used the Rochester Epidemiology Project to identify persons ≥ 65 years who were hospitalized between 2011 and 2017. We identified all persons with an International Classification of Diseases code for delirium within ±14 days of a hospitalization. We independently applied our NLP algorithm to all clinical notes for this same population. We calculated rates using number of delirium episodes as the numerator and number of hospitalizations as the denominator. Rates were estimated overall, by demographic characteristics, and by year of episode, and differences were tested using Poisson regression.

In total, 14,255 persons had 37,554 hospitalizations between 2011 and 2017. The code-based delirium rate was 3.02 per 100 hospitalizations (95% CI: 2.85, 3.20). The NLP-based rate was 7.36 per 100 (95% CI: 7.09, 7.64). Rates increased with age (both p < 0.0001). Code-based rates were higher in men compared to women (p = 0.03), but NLP-based rates were similar by sex (p = 0.89). Code-based rates were similar by race and ethnicity, but NLP-based rates were higher in the White population compared to the Black and Asian populations (p = 0.001). Both types of rates increased significantly over time (both p values < 0.001).

The NLP algorithm identified more delirium episodes compared to the ICD code method. However, NLP may still underestimate delirium cases because of limitations in real-world clinical notes, including incomplete documentation, practice changes over time, and missing clinical notes in some time periods.

The study aimed to examine variations in the use of International Classification of Diseases, Tenth Edition (ICD-10) diagnostic categories for mental and behavioural disorders across countries, regions and income levels using data from the online World Psychiatric Association (WPA)-World Health Organization (WHO) Global Survey that examined the attitudes of psychiatrists towards the classification of mental disorders.

A survey was sent to 46 psychiatric societies which are members of WPA. A total of 4887 psychiatrists participated in the survey, which asked about their use of classification, their preferred system and the categories that were used most frequently.

The majority (70.1%) of participating psychiatrists (out of 4887 psychiatrists) reported using the ICD-10 the most and using at least one diagnostic category once a week. Nine out of 44 diagnostic categories were considerably variable in terms of frequency of use across countries. These were: emotionally unstable personality disorder, borderline type; dissociative (conversion) disorder; somatoform disorders; obsessive–compulsive disorder (OCD); mental and behavioural disorders due to the use of alcohol; adjustment disorder; mental and behavioural disorders due to the use of cannabinoids; dementia in Alzheimer's disease; and acute and transient psychotic disorder. The frequency of use for these nine categories was examined across WHO regions and income levels. The most striking differences across WHO regions were found for five out of these nine categories. For dissociative (conversion) disorder, use was highest for the WHO Eastern Mediterranean Region (EMRO) and non-existent for the WHO African Region. For mental and behavioural disorders due to the use of alcohol, use was lowest for EMRO. For mental and behavioural disorders due to the use of cannabinoids, use was lowest for the WHO European Region and the WHO Western Pacific Region. For OCD and somatoform disorders, use was lowest for EMRO and the WHO Southeast Asian Region. Differences in the frequency of use across income levels were statistically significant for all categories except for mental and behavioural disorders due to the use of alcohol. The most striking variations were found for acute and transient psychotic disorder, which was reported to be more commonly used among psychiatrists from countries with lower income levels.

The differences in frequency of use reported in the current study show that cross-cultural variations in psychiatric practice exist. However, whether these differences are due to the variations in prevalence, treatment-seeking behaviour and other factors, such as psychiatrist and patient characteristics as a result of culture, cannot be determined based on the findings of the study. Further research is needed to examine whether these variations are culturally determined and how that would affect the cross-cultural applicability of ICD-10 diagnostic categories.