Predicting which patients with cancer will see a psychiatrist or counsellor from their initial oncology consultation document using natural language processing.

BACKGROUND: Patients with cancer often have unmet psychosocial needs. Early detection of who requires referral to a counsellor or psychiatrist may improve their care. This work used natural language processing to predict which patients will see a counsellor or psychiatrist from a patient's initial oncology consultation document. We believe this is the first use of artificial intelligence to predict psychiatric outcomes from non-psychiatric medical documents. METHODS: This retrospective prognostic study used data from 47,625 patients at BC Cancer. We analyzed initial oncology consultation documents using traditional and neural language models to predict whether patients would see a counsellor or psychiatrist in the 12 months following their initial oncology consultation. RESULTS: Here, we show our best models achieved a balanced accuracy (receiver-operating-characteristic area-under-curve) of 73.1% (0.824) for predicting seeing a psychiatrist, and 71.0% (0.784) for seeing a counsellor. Different words and phrases are important for predicting each outcome. CONCLUSION: These results suggest natural language processing can be used to predict psychosocial needs of patients with cancer from their initial oncology consultation document. Future research could extend this work to predict the psychosocial needs of medical patients in other settings.

Patients with cancer often need support for their mental health. Early detection of who requires referral to a counsellor or psychiatrist may improve their care. This study trained a type of artificial intelligence (AI) called natural language processing to read the consultation report an oncologist writes after they first see a patient to predict which patients will see a counsellor or psychiatrist. The AI predicted this with performance similar to other uses of AI in mental health, and used different words and phrases to predict who would see a psychiatrist compared to seeing a counsellor. We believe this is the first use of AI to predict mental health outcomes from medical documents written by clinicians outside of mental health. This study suggests this type of AI can predict the mental health needs of patients with cancer from this widely-available document.

Reduction in Substance-Related Composite Harm Scores Through Street Soccer.

Introduction Street soccer makes the sport accessible to people affected by homelessness or precarious housing. There is overwhelming evidence that exercise improves physical and mental health. In addition, sport facilitates positive peer pressure that leads to beneficial life changes. Method To examine participants' accounts of the effects of street soccer in a sample of socially disadvantaged players from Western Canada, we collected 73 cross-sectional self-reports of life changes via a questionnaire. The questionnaire included questions on social, mental, and physical health, including substance use. This allowed the calculation of a modified composite harm score. Results Participants reported improved physical (46% of participants) and mental (43% of participants) health, reduced cigarette (50% of smokers), alcohol (45% of users), cannabis (42% of users), and other non-prescribed drug use, increased number of friends (88% of participants), improved housing (60% of participants), increased income (19% of participants), increased community medical supports (40% of participants), and decreased conflicts with police (47% of those with prior recent conflict). Perceived reductions in substance use were supported by significant changes in composite harm score. Conclusion Street soccer appears to promote improved physical, mental, and social health among people affected by homelessness or precarious housing, with reduction in substance use likely to be a key factor. This work builds upon past qualitative research showing the benefits of street soccer and supports future research which may help elucidate the mechanisms by which street soccer has beneficial effects.

Predicting the Survival of Patients With Cancer From Their Initial Oncology Consultation Document Using Natural Language Processing.

Importance: Predicting short- and long-term survival of patients with cancer may improve their care. Prior predictive models either use data with limited availability or predict the outcome of only 1 type of cancer. Objective: To investigate whether natural language processing can predict survival of patients with general cancer from a patient's initial oncologist consultation document. Design, Setting, and Participants: This retrospective prognostic study used data from 47 625 of 59 800 patients who started cancer care at any of the 6 BC Cancer sites located in the province of British Columbia between April 1, 2011, and December 31, 2016. Mortality data were updated until April 6, 2022, and data were analyzed from update until September 30, 2022. All patients with a medical or radiation oncologist consultation document generated within 180 days of diagnosis were included; patients seen for multiple cancers were excluded. Exposures: Initial oncologist consultation documents were analyzed using traditional and neural language models. Main Outcomes and Measures: The primary outcome was the performance of the predictive models, including balanced accuracy and receiver operating characteristics area under the curve (AUC). The secondary outcome was investigating what words the models used. Results: Of the 47 625 patients in the sample, 25 428 (53.4%) were female and 22 197 (46.6%) were male, with a mean (SD) age of 64.9 (13.7) years. A total of 41 447 patients (87.0%) survived 6 months, 31 143 (65.4%) survived 36 months, and 27 880 (58.5%) survived 60 months, calculated from their initial oncologist consultation. The best models achieved a balanced accuracy of 0.856 (AUC, 0.928) for predicting 6-month survival, 0.842 (AUC, 0.918) for 36-month survival, and 0.837 (AUC, 0.918) for 60-month survival, on a holdout test set. Differences in what words were important for predicting 6- vs 60-month survival were found. Conclusions and Relevance: These findings suggest that models performed comparably with or better than previous models predicting cancer survival and that they may be able to predict survival using readily available data without focusing on 1 cancer type.

Understanding death with limited experience in life: dying children's and adolescents' understanding of their own terminal illness and death.

PURPOSE OF REVIEW: An up-to-date summary of the literature on children's and adolescents' understanding of their own terminal illness and death. RECENT FINDINGS: Clinicians still find it difficult to speak with pediatric patients about death even though guidelines for facilitating communication on the topic exist. As a result, pediatric patients are less likely to develop a clear understanding of their illness and there is a disconnect between clinicians and parents about prognosis, even when clinicians have concluded there is no longer possibility for cure. Insufficient communication and poor understanding may increase the risk of patients feeling isolated, mistrustful and anxious, and deprive them of a role model who can communicate about painful issues or share difficult feelings. Despite these complexities, young people often show remarkable resiliency in the face of death and want to get the most out of the remaining time they have. SUMMARY: In addition to these most recent findings, this review examines the challenges in researching this topic, obstacles to patients receiving information about prognosis, and how physical symptoms affect patients' ability to develop an understanding. It also reviews sources of insight into pediatric patients' understanding including the development of concepts of death, fears about their own death, legal interpretations of what patients understand, and how terminally ill young people continue to treasure life. It concludes by addressing ways clinicians can use the knowledge we have to communicate well with dying children and adolescents and their families.

Error processing-associated event-related potentials in schizophrenia and unaffected siblings.

A reduction in the error-related negativity (ERN), a response-locked event-related potential (ERP) observed when participants commit errors during processing of stimuli, is a well-replicated cerebral abnormality in schizophrenia. However, the extent to which this abnormality reflects susceptibility to schizophrenia rather than overt change in behavioural state is unclear. As the unaffected siblings of individuals with schizophrenia are at an increased genetic risk, this study examines whether they display abnormality of the ERN similar to that observed in individuals with schizophrenia. ERPs were recorded from 29 individuals with schizophrenia, 36 unaffected siblings and 35 healthy control participants while they performed a simple Go/No-Go task. Group differences in the ERN and also in the error positivity (Pe), a response-locked positive component that follows the ERN, were investigated. Reductions of ERN amplitudes were found in both individuals with schizophrenia and siblings. No significant abnormalities were observed in Pe. The finding of reduced ERN amplitudes in siblings without prodromal symptoms supports the hypothesis that the abnormality is not a consequence of behavioural disturbance, and that it is a trait marker for susceptibility to schizophrenia, rather than being a result of illness or medication.

Electrophysiological indices of abnormal error-processing in adolescents with attention deficit hyperactivity disorder (ADHD).

BACKGROUND: Impaired cognitive control has been frequently observed in children and young people with attention deficit hyperactivity disorder (ADHD) and might underlie the excessive hyperactivity and impulsivity in this population. We investigated behavioural and electrophysiological indices relevant to one domain of cognitive control; namely error processing. METHODS: Adolescents aged 14 to 17 with ADHD (n = 23) and a typically developing control group (HC; n = 19) performed a visual go/no-go task. Electro-encephalography (EEG) data were collected simultaneously and response-locked error trials were averaged to derive two event-related potentials, the error-related negativity (ERN) and error positivity (Pe). Evoked theta power and inter-trial phase coherence (ITC) were measured in two time windows ('early' and 'late') equivalent to those used for detection of the ERN and Pe. RESULTS: Analysis revealed normal ERN amplitude and a statistical trend for smaller Pe amplitude at a fronto-central electrode site in the ADHD group. The group also showed significant reductions in late evoked theta power and early and late theta ITC. Relationships between behavioural measures and ITC were different between groups, particularly for post-error slowing, a measure of strategic response adjustment on trials immediately following an error. CONCLUSIONS: The results reveal abnormalities in behavioural and electrophysiological indices of error processing in adolescents with ADHD and suggest that ITC is more sensitive than traditional ERP measures to error-processing abnormalities.

Low-frequency EEG oscillations associated with information processing in schizophrenia.

Numerous studies have described attenuated event-related potential (ERP) component amplitudes in schizophrenia (e.g., P300, Mismatch Negativity (MMN), Error Negativity/Error-Related Negativity (Ne/ERN)). Functional magnetic resonance imaging (fMRI) studies have typically shown decreased recruitment of diverse brain areas during performance of tasks that elicit the above ERP components. Recent research suggests that phase-resetting of slow-oscillations (e.g., in the delta and theta bands) underlies the potentials observed in ERP averages. Several studies have reported that slow-oscillations are increased in amplitude in people with schizophrenia at rest. Few studies have examined event-related low-frequency oscillations in schizophrenia. We examined event-related evoked and induced delta and theta activity in 17 people with schizophrenia and 17 healthy controls in two go/no-go task variants. We analyzed stimulus-related and response-related oscillations associated with correct-hits, correct-rejects and false-alarms. Our results reveal a pattern of reduced delta and theta activity for task-relevant events in schizophrenia. The findings indicate that while low-frequency oscillations are increased in amplitude at rest, they are not coordinated effectively in schizophrenia during various information processing tasks including target-detection, response-inhibition and error-detection. This slow-oscillation coordination abnormality may help explain the decreased recruitment of brain areas seen in fMRI studies.

Different psychological effects of cannabis use in adolescents at genetic high risk for schizophrenia and with attention deficit/hyperactivity disorder (ADHD).

BACKGROUND: Controversy exists regarding whether young people at risk for schizophrenia are at increased risk of adverse mental effects of cannabis use. METHODS: We examined cannabis use and mental health functioning in three groups of young people aged 14-21; 36 non-psychotic siblings of adolescents with schizophrenia (genetic high risk group), 25 adolescents with attention deficit hyperactivity disorder (ADHD) and 72 healthy controls. The groups were sub-divided into 'users' and 'non-users' of cannabis based on how often they had used cannabis previously. Mental health functioning was quantified by creating a composite index derived from scores on the Schizotypal Personality Questionnaire (SPQ), Strengths and Difficulties Questionnaire (SDQ) and Global Assessment of Function (GAF). RESULTS: A significant positive association between cannabis use and mental health disturbance was confined to young people at genetic high risk for schizophrenia. To determine whether the relationship was specific to particular dimensions of mental health function, a second composite index was created based on scores from the SPQ Disorganisation and SDQ hyperactivity-inattention sub-scales. Again, there was a significant positive association between cannabis use and factor scores which was specific to the genetic high risk group. There was a trend for this association to be negative in the ADHD group (p=0.07). CONCLUSIONS: The findings support the view that young people at genetic high risk for schizophrenia are particularly vulnerable to mental health problems associated with cannabis use. Further research is needed to investigate the basis of relationships between cannabis and mental health in genetically vulnerable individuals.

Event-related potentials in adolescents with schizophrenia and their siblings: a comparison with attention-deficit/hyperactivity disorder.

BACKGROUND: Identifying trait markers specific to schizophrenia might uncover mechanisms underlying illness susceptibility. Previous research shows the N2 and P3 event-related potentials are abnormal in schizophrenia; specificity of these potential trait markers has not been well established. METHODS: Electroencephalogram data were recorded from four adolescent groups: early-onset schizophrenia patients (SZ; n = 30); non-psychotic siblings of schizophrenia patients (SZ-SIB; n = 36); healthy control subjects (HC; n = 36); a neurodevelopmental attention-deficit/hyperactivity disorder (ADHD) comparison group (n = 27), during auditory oddball and visual go/no-go tasks. The P3 was measured to targets in the oddball task. The N2 and P3 were measured to go and no-go stimuli in the go/no-go task. RESULTS: Compared with the HC group, the SZ and SZ-SIB groups showed significantly reduced auditory oddball P3 amplitude. Visual P3 amplitude was significantly reduced in the SZ group for no-go stimuli and the SZ-SIB group for go and no-go stimuli. The P3 amplitude in the ADHD group was not significantly reduced for either paradigm. The SZ and ADHD groups showed significantly reduced N2 amplitude in the go/no-go task; the SZ-SIB group was not significantly different from the HC group. CONCLUSIONS: Results revealed reduced P3 amplitude in schizophrenia patients and adolescent non-psychotic siblings in an auditory oddball and a visual go/no-go task. The SZ-SIB and ADHD groups showed a different ERP profile when each was compared with the HC group: siblings showed reduced P3 amplitude in both tasks and normal N2 in the go/no-go task; the opposite pattern was observed in the ADHD group.

Brain potentials implicate temporal lobe abnormalities in criminal psychopaths.

Psychopathy is associated with abnormalities in attention and orienting. However, few studies have examined the neural systems underlying these processes. To address this issue, the authors recorded event-related potentials (ERPs) while 80 incarcerated men, classified as psychopathic or nonpsychopathic via the Hare Psychopathy Checklist--Revised (R. D. Hare, 1991, 2003), completed an auditory oddball task. Consistent with hypotheses, processing of targets elicited larger frontocentral negativities (N550) in psychopaths than in nonpsychopaths. Psychopaths also showed an enlarged N2 and reduced P3 during target detection. Similar ERP modulations have been reported in patients with amygdala and temporal lobe damage. The data are interpreted as supporting the hypothesis that psychopathy may be related to dysfunction of the paralimbic system--a system that includes parts of the temporal and frontal lobes.

State dependent changes in error monitoring in schizophrenia.

The aim of this study was to determine if error negativity/error-related negativity (N(e)/ERN), error positivity (P(e)), correct response negativity (CRN) or correct response positivity (P(c)) amplitude are influenced by state changes in schizophrenia. Event-related potentials (ERPs) were recorded from nine schizophrenic patients while they performed a simple go/no-go task during the early stages of an acute episode and again following 6 weeks of treatment with antipsychotics. ERPs were also recorded from nine healthy participants while they performed the same task. Response-locked potentials were computed for errors of commission and for correct responses. Scores for reality distortion syndrome, psychomotor poverty syndrome and disorganization syndrome were determined for the schizophrenic participants before and after treatment using the Signs and Symptoms of Psychotic Illness (SSPI) Scale. N(e)/ERN amplitude was significantly reduced, compared with that in healthy participants, in the schizophrenic patients when acutely ill, and increased significantly following treatment. N(e)/ERN amplitude remained significantly larger in the healthy group than in the patients with schizophrenia after treatment. This study suggests that N(e)/ERN and CRN amplitude are modulated by clinical state in schizophrenia and provides further support to findings that decreased N(e)/ERN amplitude is a potentially useful trait marker for schizophrenia, while P(c) and P(e) amplitude are not abnormal.

Abnormal processing of speech during oddball target detection in schizophrenia.

Healthy subjects show increased activation in left temporal lobe regions in response to speech sounds compared to complex nonspeech sounds. Abnormal lateralization of speech-processing regions in the temporal lobes has been posited to be a cardinal feature of schizophrenia. Event-related fMRI was used to test the hypothesis that schizophrenic patients would show an abnormal pattern of hemispheric lateralization when detecting speech compared with complex nonspeech sounds in an auditory oddball target-detection task. We predicted that differential activation for speech in the vicinity of the superior temporal sulcus would be greater in schizophrenic patients than in healthy subjects in the right hemisphere, but less in patients than in healthy subjects in the left hemisphere. Fourteen patients with schizophrenia (selected from an outpatient population, 2 females, 12 males, mean age 35.1 years) and 29 healthy subjects (8 females, 21 males, mean age 29.3 years) were scanned while they performed an auditory oddball task in which the oddball stimuli were either speech sounds or complex nonspeech sounds. Compared to controls, individuals with schizophrenia showed greater differential activation between speech and nonspeech in right temporal cortex, left superior frontal cortex, and the left temporal-parietal junction. The magnitude of the difference in the left temporal parietal junction was significantly correlated with severity of disorganized thinking. This study supports the hypothesis that aberrant functional lateralization of speech processing is an underlying feature of schizophrenia and suggests the magnitude of the disturbance in speech-processing circuits may be associated with severity of disorganized thinking.

Rostral anterior cingulate cortex dysfunction during error processing in schizophrenia.

Previous research has demonstrated that patients with schizophrenia have an impaired ability to monitor erroneous responses to stimuli internally. Event-related potential (ERP) studies of error-eliciting tasks indicate that, in healthy adults, the commission of an erroneous response is associated with a fronto-centrally distributed negative voltage component termed the error negativity (Ne) or error-related negativity (ERN). In patients with schizophrenia, the Ne/ERN elicited by errors of commission (EoC) is reduced in amplitude compared with that elicited in healthy participants. Functional MRI (fMRI) studies and source localization analyses of ERP data in healthy participants suggest that EoC are associated with activity in the rostral anterior cingulate cortex (ACC). Using event-related fMRI, we examined the brain activity associated with EoC in a group of 10 patients with schizophrenia and 16 matched healthy participants. Patients were stable, partially remitted, medicated out-patients recruited from the community. Participants performed a Go/NoGo task variant that was shown previously to elicit a reduced Ne/ERN during EoC in patients with schizophrenia relative to healthy participants, as well as robust rostral ACC activation during EoC in healthy participants. Patients with schizophrenia were characterized by relative underactivity in the rostral ACC compared with healthy participants. There was also evidence for more widespread underactivity in the limbic system. In contrast to these regions of relative hypoactivity, patients with schizophrenia demonstrated hyperactivity relative to healthy participants in bilateral parietal cortex during both EoC and correctly rejected NoGo trials. Our results are consistent with previous ERP research demonstrating functional abnormalities during error processing in schizophrenia. In light of the role of the rostral ACC and other limbic structures in mediating affective and motivational behaviour, our results suggest there may be a disturbed affective or motivational response to the commission of errors in schizophrenia.

Thought and Language Index: an instrument for assessing thought and language in schizophrenia.

BACKGROUND: Subtle formal thought disorders are difficult to quantify. Their relationship to florid thought disorder is unknown. AIMS: To assess the interrater reliability, sensitivity and factor structure of a new assessment instrument, the Thought and Language Index (TLI), and to determine if minor aberrations detectable in the speech of healthy individuals are related to the more severe formal thought disorders characteristic of schizophrenia. METHOD: Interrater reliability was evaluated by determining the intraclass correlation for the ratings by five assessors. Factor analysis of the TLI scores of 87 patients was performed, and TLI scores in matched patients and controls were compared. RESULTS: The intraclass correlation was good for individual TLI items, and excellent for sub-scale scores. Factor analysis identified three groups of approximately orthogonal disorders. Mild speech aberrations were observed in healthy participants and in patients with schizophrenia. The prevalence of mild aberrations was correlated with the prevalence of definite formal thought disorders. CONCLUSIONS: The TLI is reliable and capable of detecting subtle disorders. Some mild aberrations occurring in the speech of healthy individuals appear to be attenuated forms of the florid disorders characteristic of schizophrenia.

Error-related negativity and correct response negativity in schizophrenia.

OBJECTIVES: To examine error-related negativity (ERN) and correct response negativity (CRN) in schizophrenia in light of two previous conflicting reports, and to determine their relation to disorganization, psychomotor poverty and reality distortion. METHODS: Event-related potentials were recorded from 21 schizophrenic and 21 control participants who performed a simple go/no-go task. Response-locked potentials were computed for errors of commission and for correct-hits. Scores for reality distortion syndrome, psychomotor poverty syndrome and disorganization syndrome were determined for each schizophrenic participant using the Signs and Symptoms of Psychotic Illness (SSPI) scale. RESULTS: ERN produced during error trials and CRN produced during correct trials were significantly larger in the control participant group than in the schizophrenic participant group. In the schizophrenic patients, ERN amplitude was negatively correlated with psychomotor poverty syndrome score and CRN amplitude was negatively correlated with disorganization syndrome score. CONCLUSIONS: Decreased ERN and CRN in the schizophrenic participant group suggests abnormal internal behavior monitoring in schizophrenic patients. Patients with high disorganization symptoms may employ an abnormal strategy for comparing actual response outcome with desired response outcome, while patients with psychomotor poverty may be less emotionally responsive to errors.