Gender differences in clinical presentation and illicit substance use during first episode psychosis: a natural language processing, electronic case register study

Abstract

ObjectiveTo determine whether gender differences in symptom presentation at first episode psychosis (FEP) remain even when controlling for substance use, age and ethnicity, using natural language processing applied to electronic health records (EHRs).Design, setting and participantsData were extracted from EHRs of 3350 people (62% male patients) who had presented to the South London and Maudsley NHS Trust with a FEP between 1 April 2007 and 31 March 2017. Logistic regression was used to examine gender differences in the presentation of positive, negative, depressive, mania and disorganisation symptoms.Exposure(s) (for observational studies)Gender (male vs female).Main outcome(s) and measure(s)Presence of positive, negative, depressive, mania and disorganisation symptoms at initial clinical presentation.ResultsEight symptoms were significantly more prevalent in men (poverty of thought, negative symptoms, social withdrawal, poverty of speech, aggression, grandiosity, paranoia and agitation). Conversely, tearfulness, low energy, reduced appetite, low mood, pressured speech, mood instability, flight of ideas, guilt, mutism, insomnia, poor concentration, tangentiality and elation were more prevalent in women than men. Negative symptoms were more common among men (OR 1.85, 95% CI 1.33 to 2.62) and depressive and manic symptoms more common among women (OR 0.30, 95% CI 0.26 to 0.35). After adjustment for illicit substance use, the strength of associations between gender and negative, manic and depression symptoms increased, whereas gender differences in aggression, agitation, paranoia and grandiosity became insignificant.ConclusionsThere are clear gender differences in the clinical presentation of FEP. Our findings suggest that gender can have a substantial influence on the nature of clinical presentation in people with psychosis, and that this is only partly explained by exposure to illicit substance use.

Best Practices on Big Data Analytics to Address Sex-Specific Biases in our Understanding of the Etiology, Diagnosis and Prognosis of Diseases

“…Differences in symptom presentation at first episode psychosis (FEP) have been studied using NLP and ML in EHRs (44) and diagnosis and clinical manifestations of COVID (45) and tinnitus (46), Friedreich’s ataxia (FRDA) (47) and Alzheimer’s disease (48). Using the data output from medical devices, sex-specific differences were examined for gait kinematics in patients with knee osteoarthritis (49) and heart rate variability (50).…”

“…The data sources for the studies were varied, with most utilizing information from EHRs or other clinical databases. Within these studies, the majority only included structured data in their models (47, 48, 49, 50, 58, 59, 60, 61, 62) while a few incorporated both structured and unstructured data from these resources (45, 43, 44, 63). Another source of data was surveys, questionnaires, or interviews, where again the majority of studies used only structured data in their models (46, 53, 54, 51, 57), with only one study applying NLP methods and used both structured and unstructured data (64).…”

“…One study used topic modelling to identify disease disclosure (52). Two of the included studies (2/22, 9.1%) applied NLP techniques only to extract information from the free text of EHRs (44, 63). The remaining studies (5/22, 22.7%) used both NLP to identify and extract information and machine learning methods for classification or prediction (43, 41, 45, 52, 64).…”

“…Most of the studies explored the role of sex-specific differences in the creation of classification or prediction models for a variety of conditions including, Alzheimer’s disease (48), autism(51), cancer (59), cardiovascular disease (47, 60, 63), musculoskeletal disorders (57), substance use (44,55, 58), mental illness (43, 52, 62, 64) obesity/physical activity (41, 54) and tinnitus (46). Within these studies, nine developed separate models for males and females finding these models were better able to discriminate sex-specific determinants of the trait of interest.…”

Psychosis in Women: Time for Personalized Treatment

“…It has been shown that gender differences at presentation are independent of age and ethnicity. Furthermore, there is evidence that after adjusting for illicit substance use, negative symptoms still remain more prominent in men while manic and depressive symptoms became even more prominent in women, thus suggesting the idea that gender differences in the clinical presentation of psychosis and in symptom domains may reflect specific and identifiable sex differences in pathophysiology [29].…”