Maudsley National Health Service Research Articles

HbA1c recording in patients following a first diagnosis of serious mental illness: the South London and Maudsley Biomedical Research Centre case register

ObjectivesTo investigate factors associated with the recording of glycated haemoglobin (HbA1c) in people with first diagnoses of serious mental illness (SMI) in a large mental healthcare provider, and factors associated...

Development of a Corpus Annotated With Mentions of Pain in Mental Health Records: Natural Language Processing Approach.

Pain is a widespread issue, with 20% of adults (1 in 5) experiencing it globally. A strong association has been demonstrated between pain and mental health conditions, and this association is known to exacerbate disability and impairment. Pain is also known to be strongly related to emotions, which can lead to damaging consequences. As pain is a common reason for people to access health care facilities, electronic health records (EHRs) are a potential source of information on this pain. Mental health EHRs could be particularly beneficial since they can show the overlap of pain with mental health. Most mental health EHRs contain the majority of their information within the free-text sections of the records. However, it is challenging to extract information from free text. Natural language processing (NLP) methods are therefore required to extract this information from the text. This research describes the development of a corpus of manually labeled mentions of pain and pain-related entities from the documents of a mental health EHR database, for use in the development and evaluation of future NLP methods. The EHR database used, Clinical Record Interactive Search, consists of anonymized patient records from The South London and Maudsley National Health Service Foundation Trust in the United Kingdom. The corpus was developed through a process of manual annotation where pain mentions were marked as relevant (ie, referring to physical pain afflicting the patient), negated (ie, indicating absence of pain), or not relevant (ie, referring to pain affecting someone other than the patient, or metaphorical and hypothetical mentions). Relevant mentions were also annotated with additional attributes such as anatomical location affected by pain, pain character, and pain management measures, if mentioned. A total of 5644 annotations were collected from 1985 documents (723 patients). Over 70% (n=4028) of the mentions found within the documents were annotated as relevant, and about half of these mentions also included the anatomical location affected by the pain. The most common pain character was chronic pain, and the most commonly mentioned anatomical location was the chest. Most annotations (n=1857, 33%) were from patients who had a primary diagnosis of mood disorders (International Classification of Diseases-10th edition, chapter F30-39). This research has helped better understand how pain is mentioned within the context of mental health EHRs and provided insight into the kind of information that is typically mentioned around pain in such a data source. In future work, the extracted information will be used to develop and evaluate a machine learning-based NLP application to automatically extract relevant pain information from EHR databases.

Associations of remote mental healthcare with clinical outcomes: a natural language processing enriched electronic health record data study protocol

IntroductionPeople often experience significant difficulties in receiving mental healthcare due to insufficient resources, stigma and lack of access to care. Remote care technology has the potential to overcome these barriers...

Predictors of falls and fractures leading to hospitalisation in 36 101 people with affective disorders: a large representative cohort study

ObjectivesTo investigate predictors of falls and fractures leading to hospitalisation in people with affective disorders.DesignCohort study.SettingThe South London and Maudsley National Health Service (NHS) Foundation Trust (SLaM) Biomedical Research Centre...

Using natural language processing to extract self-harm and suicidality data from a clinical sample of patients with eating disorders: a retrospective cohort study

ObjectivesThe objective of this study was to determine risk factors for those diagnosed with eating disorders who report self-harm and suicidality.Design and settingThis study was a retrospective cohort study within...

Real-World Clinical Outcomes Two Years After Transition to Psychosis in Individuals at Clinical High Risk: Electronic Health Record Cohort Study.

The objective of this study is to describe the 2-year real-world clinical outcomes after transition to psychosis in patients at clinical high-risk. The study used the clinical electronic health record cohort study including all patients receiving a first index primary diagnosis of nonorganic International Classification of Diseases (ICD)-10 psychotic disorder within the early psychosis pathway in the South London and Maudsley (SLaM) National Health Service (NHS) Trust from 2001 to 2017. Outcomes encompassed: cumulative probability (at 3, 6, 12, and 24 months) of receiving a first (1) treatment with antipsychotic, (2) informal admission, (3) compulsory admission, and (4) treatment with clozapine and (5) numbers of days spent in hospital (at 12 and 24 months) in patients transitioning to psychosis from clinical high-risk services (Outreach and Support in south London; OASIS) compared to other first-episode groups. Analyses included logistic and 0-inflated negative binomial regressions. In the study, 1561 patients were included; those who had initially been managed by OASIS and had subsequently transitioned to a first episode of psychosis (n = 130) were more likely to receive antipsychotic medication (at 3, 6, and 24 months; all P < .023), to be admitted informally (at all timepoints, all P < .004) and on a compulsory basis (at all timepoints, all P < .013), and to have spent more time in hospital (all timepoints, all P < .007) than first-episode patients who were already psychotic when seen by the OASIS service (n = 310), or presented to early intervention services (n = 1121). The likelihood of receiving clozapine was similar across all groups (at 12/24 months, all P < .101). Transition to psychosis from a clinical high-risk state is associated with severe real-world clinical outcomes. Prevention of transition to psychosis should remain a core target of future research. The study protocol was registered on www.researchregistry.com; researchregistry5039).

Linking two administrative datasets about looked after children: testing feasibility and enhancing understanding

Background with rationaleEvery year all 32 local authorities in Scotland provide information on looked after children in their area to the Scottish Government. This forms the basis for the annual Children Looked After Statistics (CLAS). Information is also collected by Scottish Children’s Reporter Administration (SCRA) on all children who are involved in the Children’s Hearings System. Until now these two data sets had never been linked.&#x0D; Main Aim To test the feasibility and success of the linkage on the basis that these datasets had not previously been linked, and if linkage was possible, use this data to enhance our understanding of the child and process factors associated with pathways to permanence or lack of permanence.&#x0D; Methods/ApproachVeterans were identified using the South London and Maudsley Biomedical Research Centre (SLaM) case register – a database holding secondary mental health care electronic records for the South London and Maudsley National Health Service Trust of 300,000 patients. We developed two methods. An NLP and machine learning tool were developed to automatically evaluate personal history statements written by clinicians.&#x0D; ResultsFor the first time, as part of the Permanently Progressing? Building secure futures for children in Scotland study, these two data sets were linked safely and successfully for 1,000 children who became looked after in 2012-13 when they were aged five and under.&#x0D; The linkage provided important new information for practitioners and policymakers. In this presentation we will focus on the key findings, such as what it told us about previous referrals and methodological insights regarding these data sets and their linkage.&#x0D; ConclusionThe data linkage process was complex and time-consuming but possible. The data we were able to link provided valuable information that enhanced our understanding of child and process factors.

Childhood cognitive ability and the use of long-term care in later life

Background As the population ages the demand for care is predicted to increase. Previous studies have reported that individuals with poorer post-morbid cognitive ability are at higher risk of entering long-term care, both institutionalised care and home-based care. Given that post-morbid cognitive ability is sensitive to the type and severity of morbidity, it remains unclear whether higher cognitive ability, as a trait-level measure of individual differences, contributes to care usage. Some success has been observed using non-cognitive early-life circumstances such as socioeconomic circumstances as pre-morbid predictors of care risk. However, the contribution of early-life cognitive ability has yet to be examined.&#x0D; Main Aim We investigate the association between early-life circumstances, particularly cognitive ability, and the risk of entry into long-term care in later life (age 65+).&#x0D; Methods Veterans were identified using the South London and Maudsley Biomedical Research Centre (SLaM) case register – a database holding secondary mental health care electronic records for the South London and Maudsley National Health Service Trust of 300,000 patients. We developed two methods. An NLP and machine learning tool were developed to automatically evaluate personal history statements written by clinicians.&#x0D; Results This study uses a large sample of individuals born in Scotland in 1936 and who took part in the Scottish Mental Survey 1947. It links research data from childhood to routinely-collected administrative and healthcare records from across the life course.&#x0D; Conclusion We demonstrate the importance of early-life factors for predicting care usage in later life and how this role differs between types of long-term care. The implications of the results for research and policy will be discussed.

Identifying Military Veterans in a Clinical Research Database using Natural Language Processing

BackgroundThere is a lack of quantitative evidence concerning United Kingdom veterans who access secondary mental health care. This is mainly due to a person&amp;#8217;s veteran status not being routinely collected when they enter the health care system.&#x0D; Main AimThe study aimed to develop an NLP approach for identifying veterans accessing secondary mental health care services using National Health Service electronic health records.&#x0D; MethodsVeterans were identified using the South London and Maudsley Biomedical Research Centre (SLaM) case register – a database holding secondary mental health care electronic records for the South London and Maudsley National Health Service Trust of 300,000 patients. We developed two methods. An NLP and machine learning tool were developed to automatically evaluate personal history statements written by clinicians.&#x0D; ResultsThis study showed that it was possible to identify veterans using the NLP and machine learning approach on a sub-set of 4,200 patients. The automatic machine learning method was able to identify 270 veterans representing an accuracy of 97.2%. It is estimated to take between 6 to 16 minutes to manually search patient history statements whereas the automatic machine learning method took only one minute to run.&#x0D; ConclusionWe have shown that it is possible to identify veterans using NLP combined with machine learning. This work contributes towards the development of a more comprehensive picture of veterans who are accessing secondary mental health care services in the UK. It represents a first step in identifying veterans from one dataset and we hope that future research can inform the possibility of deploying the methods nationally. Despite our success in the current work, the tools are tailored to the SLaM dataset and future work is needed to develop a more agnostic framework.&#x0D; FundingForces in Mind Trust

Transdiagnostic Individualized Clinically Based Risk Calculator for the Detection of Individuals at Risk and the Prediction of Psychosis: Model Refinement Including Nonlinear Effects of Age.

Background: The first rate-limiting step for primary indicated prevention of psychosis is the detection of young people who may be at risk. The ability of specialized clinics to detect individuals at risk for psychosis is limited. A clinically based, individualized, transdiagnostic risk calculator has been developed and externally validated to improve the detection of individuals at risk in secondary mental health care. This calculator employs core sociodemographic and clinical predictors, including age, which is defined in linear terms. Recent evidence has suggested a nonlinear impact of age on the probability of psychosis onset. Aim: To define at a meta-analytical level the function linking age and probability of psychosis onset. To incorporate this function in a refined version of the transdiagnostic risk calculator and to test its prognostic performance, compared to the original specification. Design: Secondary analyses on a previously published meta-analysis and clinical register-based cohort study based on 2008–2015 routine secondary mental health care in South London and Maudsley (SLaM) National Health Service (NHS) Foundation Trust. Participants: All patients receiving a first index diagnosis of non-organic/non-psychotic mental disorder within SLaM NHS Trust in the period 2008–2015. Main outcome measure: Prognostic accuracy (Harrell’s C). Results: A total of 91,199 patients receiving a first index diagnosis of non-organic and non-psychotic mental disorder within SLaM NHS Trust were included in the derivation (33,820) or external validation (54,716) datasets. The mean follow-up was 1,588 days. The meta-analytical estimates showed that a second-degree fractional polynomial model with power (−2, −1: age1 = age−2 and age2 = age−1) was the best-fitting model (P < 0.001). The refined model that included this function showed an excellent prognostic accuracy in the external validation (Harrell’s C = 0.805, 95% CI from 0.790 to 0.819), which was statistically higher than the original model, although of modest magnitude (Harrell’s C change = 0.0136, 95% CIs from 0.006 to 0.021, P < 0.001). Conclusions: The use of a refined version of the clinically based, individualized, transdiagnostic risk calculator, which allows for nonlinearity in the association between age and risk of psychosis onset, may offer a modestly improved prognostic performance. This calculator may be particularly useful in young individuals at risk of developing psychosis who access secondary mental health care.

Long term outcomes of acute and transient psychotic disorders: The missed opportunity of preventive interventions

BackgroundAcute and transient psychotic disorders (ATPD) are characterized by an acute onset and a remitting course, and overlap with subgroups of the clinical high-risk state for psychosis. The long-term course and outcomes of ATPD are not completely clear. MethodsElectronic health record-based retrospective cohort study, including all patients who received a first index diagnosis of ATPD (F23, ICD-10) within the South London and Maudsley (SLaM) National Health Service Trust, between 1 st April 2006 and 15th June 2017. The primary outcome was risk of developing persistent psychotic disorders, defined as the development of any ICD-10 diagnoses of non-organic psychotic disorders. Cumulative risk of psychosis onset was estimated through Kaplan-Meier failure functions (non-competing risks) and Greenwood confidence intervals. ResultsA total of 3074 patients receiving a first index diagnosis of ATPD (F23, ICD-10) within SLaM were included. The mean follow-up was 1495 days. After 8-year, 1883 cases (61.26%) retained the index diagnosis of ATPD; the remaining developed psychosis. The cumulative incidence (Kaplan-Meier failure function) of risk of developing any ICD-10 non-organic psychotic disorder was 16.10% at 1-year (95%CI 14.83–17.47%), 28.41% at 2-year (95%CI 26.80–30.09%), 33.96% at 3-year (95% CI 32.25–35.75%), 36.85% at 4-year (95%CI 35.07–38.69%), 40.99% at 5-year (95% CI 39.12–42.92%), 42.58% at 6-year (95%CI 40.67–44.55%), 44.65% at 7-year (95% CI 42.66–46.69%), and 46.25% at 8-year (95% CI 44.17–48.37%). The cumulative risk of schizophrenia-spectrum disorder at 8-year was 36.14% (95% CI 34.09–38.27%). ConclusionsIndividuals with ATPD have a very high risk of developing persistent psychotic disorders and may benefit from early detection and preventive treatments to improve their outcomes.

T122. UNMET NEEDS IN PATIENTS WITH ACUTE TRANSIENT PSYCHOTIC DISORDERS (ATPD): ANALYSIS OF PATHWAYS TO CARE: AN 8 YEARS FOLLOW-UP STUDY

BackgroundAcute and transient psychotic disorders (ATPDs) constitute a highly heterogeneous category of brief psychotic disorders. The long-term course and outcomes of ATPDs is not completely clear, with more than half of patients initially diagnosed with ATPDs shifting towards other psychotic spectrum diagnoses. Uncertainties in the real-world clinical care of these patients is further complicated by the diagnostic overlap with the Brief Limited Intermittent Psychotic Symptoms (BLIPS). Thus, patients with similar diagnostic features may either be recommended conventional antipsychotic treatment (if diagnosed with ATPD and according to the current guidelines for first episode psychosis - FEP) or be contraindicated antipsychotic treatment and receive psychological therapies (recommended for BLIPS cases).Given the complexity of the clinical presentation, admission to highly specialized services for early intervention in psychotic disorders (EIP) should represent the best therapeutic pathway for these patients. However, it is not known how many individuals with ATPDs are effectively detected and treated by EIP services.This study aims at overcoming such a gap in knowledge by describing the pathways to care of patients with ATPDs and the treatments received across eight follow-up time-points (3, 6, 12, 18, 24, 48, 72, and 96 months).MethodsElectronic health record-based retrospective cohort study including all patients who received a first index diagnosis of ATPD (F23, ICD-10) within the South London and Maudsley (SLaM) National Health Service Trust, between 1st April 2006 and 15th June 2017. Sociodemographic and clinical characteristics were analyzed using one-way ANOVA and Tukey post-hoc tests for continuous variables and chi square test for categorical variables. Logistic regression analyses were used to investigate the association between sociodemographic characteristics and detection/treatment by EIP.ResultsA total of 3074 patients receiving a first index diagnosis of ATPD (F23, ICD-10) within SLaM were included. The mean follow-up was 1495 days. After 8-year, 1883 cases (61.26%) retained the index diagnosis of ATPD; the remaining developed schizophrenia (23.8%), affective-spectrum psychoses (4.8%), and other psychotic disorders.Only 7.5 % of ATPDs was detected and treated by Early intervention in Psychosis services (EIP). The remaining quote of patients were treated with general mental health services (91.5%).Active treatment by EIS was more common among males, caucasian, and younger individuals (odds ratio (OR) = 1.35, 95% CI 1.01–1.7, P<0.001; OR = 0.6, 95% CI 0.46–0.78, P<0.001; and OR = 0.91, 95% CI 0.90–0.92, P<0.001, respectively).Almost half of the total sample (48.5%) was in treatment with antipsychotic medications after 1 year of follow-up. This proportion dropped to 25% after 3 years of active treatment.Less than 1% of ATPDs were offered psychotherapy interventions at any of the 8 time points of interest.DiscussionThe present study shows that the largest majority of individuals with ATPD (91.5%) is never detected and treated by the EIP services, which should be the best clinical option for these patients.This suggests that they are receiving neither the best first-episode care nor the best preventative care. Efforts should be done to promote outreach campaigns in general mental health services to persuade clinicians referring these patients to local EIP services, with the aim of providing the best possible care.

34.2 IMPROVING THE DETECTION OF INDIVIDUALS AT RISK OF PSYCHOSIS IN SECONDARY MENTAL HEALTH CARE

BackgroundThe overall effect of At Risk Mental State (ARMS) services for the detection of individuals who will develop psychosis in secondary mental health care is undetermined. The objective of the study presented in this lecture is to measure the proportion of individuals with a first episode of psychosis detected by ARMS services in secondary mental health services, and to develop and externally validate a practical web-based individualized risk calculator tool for the transdiagnostic prediction of psychosis in secondary mental health care.MethodsClinical register-based cohort study. Patients were drawn from electronic, real-world, real-time clinical records relating to 2008 to 2015 routine secondary mental health care in the South London and the Maudsley National Health Service Foundation Trust. The study included all patients receiving a first index diagnosis of nonorganic and nonpsychotic mental disorder within the South London and the Maudsley National Health Service Foundation Trust in the period between January 1, 2008, and December 31, 2015. Data analysis began on September 1, 2016.The main outcome is risk of development of nonorganic International Statistical Classification of Diseases and Related Health Problems, Tenth Revision psychotic disorders.ResultsA total of 91 199 patients receiving a first index diagnosis of nonorganic and nonpsychotic mental disorder within South London and the Maudsley National Health Service Foundation Trust were included in the derivation (n=33 820) or external validation (n=54 716) data sets. The mean age was 32.97 years, 50.88% were men, and 61.05% were white race/ethnicity. The mean follow-up was 1588 days. The overall 6-year risk of psychosis in secondary mental health care was 3.02 (95% CI, 2.88–3.15), which is higher than the 6-year risk in the local general population (0.62). Compared with the ARMS designation, all of the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnoses showed a lower risk of psychosis, with the exception of bipolar mood disorders (similar risk) and brief psychotic episodes (higher risk). The ARMS designation accounted only for a small proportion of transitions to psychosis (n=52 of 1001; 5.19% in the derivation data set), indicating the need for transdiagnostic prediction of psychosis in secondary mental health care. A prognostic risk stratification model based on preselected variables, including index diagnosis, age, sex, age by sex, and race/ethnicity, was developed and externally validated, showing good performance and potential clinical usefulness.DiscussionThis lecture will introduce a new online individualized risk calculator which can be of clinical usefulness for the transdiagnostic prediction of psychosis in secondary mental health care. The risk calculator can help to identify those patients at risk of developing psychosis who require an ARMS assessment and specialized care. The use of this calculator may eventually facilitate the implementation of an individualized provision of preventive focused interventions and improve outcomes of first episode psychosis.

Training peers to treat Ebola centre workers with anxiety and depression in Sierra Leone.

Following the 2014 Ebola virus disease (EVD) outbreak in West Africa, the UK Department for International Development funded South London and Maudsley National Health Service (NHS) to develop a psychological intervention that ex-Ebola Treatment Centre (ETC) staff could be trained to deliver to their peers to improve mental health in Sierra Leone. The two key aims were to assess the feasibility of training a national team to deliver a cognitive behavioural therapy (CBT)-based group intervention, and to evaluate the effectiveness of the overall intervention within this population. UK clinicians travelled to Sierra Leone to train a small team of ex-ETC staff in a three-phased CBT-based intervention. Standardised clinical measures, as well as bespoke measures, were applied with participants through the intervention to assess changes in mental health symptomology, and the effectiveness of the intervention. The results found improvements across all factors of mental health in the bespoke measure from phase 1 to phase 3. Additionally, the majority of standardised clinical measures showed improvements between phase 2 and the start of phase 3, and pre- and post-phase 3. Overall, the findings suggest that it is possible to train staff from ETCs to deliver effective CBT interventions to peers. The implications of these results are discussed, including suggestions for future research and clinical intervention implementation within this population. The limitations of this research are also addressed.

What is the impact of research champions on integrating research in mental health clinical practice? A quasiexperimental study in South London, UK

ObjectivesKey challenges for mental health healthcare professionals to implement research alongside clinical activity have been highlighted, such as insufficient time to apply research skills and lack of support and resources....

Effect of implementation of a smoke-free policy on physical violence in a psychiatric inpatient setting: an interrupted time series analysis

Smoke-free policies are important to protect health and reduce health inequalities. A major barrier to policy implementation in psychiatric hospitals is staff concern that physical violence will increase. We aimed to assess the effect of implementing a comprehensive smoke-free policy on rates of physical assaults in a large UK mental health organisation. We did an interrupted time series analysis of incident reports of physical assault 30 months before and 12 months after the implementation of the policy in the inpatient wards of South London and Maudsley National Health Service Foundation Trust, London, UK. We used a quasi-Poisson generalised additive mixed model to model the monthly incidence of physical assaults as a function of several explanatory variables. 4550 physical assaults took place between April 1, 2012, and Sept 30, 2015; 225 (4·9%) of which were smoking-related. After adjustment for temporal and seasonal trends and key confounders (sex, age, schizophrenia or related disorders, or having been sectioned under the Mental Health Act), there was a 39% reduction in the number of physical assaults per month after the policy introduction compared with beforehand (incidence rate ratio 0·61, 95% CI 0·53-0·70; p<0·0001). Introduction of a comprehensive smoke-free policy appeared to reduce the incidence of physical assaults. Adequately resourced smoke-free policies could be part of broader violence reduction strategies in psychiatric settings. National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care South London (King's College Hospital NHS Foundation Trust).

The London Contribution to Neurology and Psychiatry: The second British Symposium on the History of Neurology and Psychiatry

Following the first British Symposium on the History of Neurology and Psychiatry on 9 July 2014 at the Institute of Psychiatry, Psychology and Neuroscience/King's College London1 a second Symposium was...

Development and Validation of a Clinically Based Risk Calculator for the Transdiagnostic Prediction of Psychosis

The overall effect of At Risk Mental State (ARMS) services for the detection of individuals who will develop psychosis in secondary mental health care is undetermined. To measure the proportion of individuals with a first episode of psychosis detected by ARMS services in secondary mental health services, and to develop and externally validate a practical web-based individualized risk calculator tool for the transdiagnostic prediction of psychosis in secondary mental health care. Clinical register-based cohort study. Patients were drawn from electronic, real-world, real-time clinical records relating to 2008 to 2015 routine secondary mental health care in the South London and the Maudsley National Health Service Foundation Trust. The study included all patients receiving a first index diagnosis of nonorganic and nonpsychotic mental disorder within the South London and the Maudsley National Health Service Foundation Trust in the period between January 1, 2008, and December 31, 2015. Data analysis began on September 1, 2016. Risk of development of nonorganic International Statistical Classification of Diseases and Related Health Problems, Tenth Revision psychotic disorders. A total of 91 199 patients receiving a first index diagnosis of nonorganic and nonpsychotic mental disorder within South London and the Maudsley National Health Service Foundation Trust were included in the derivation (n = 33 820) or external validation (n = 54 716) data sets. The mean age was 32.97 years, 50.88% were men, and 61.05% were white race/ethnicity. The mean follow-up was 1588 days. The overall 6-year risk of psychosis in secondary mental health care was 3.02 (95% CI, 2.88-3.15), which is higher than the 6-year risk in the local general population (0.62). Compared with the ARMS designation, all of the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnoses showed a lower risk of psychosis, with the exception of bipolar mood disorders (similar risk) and brief psychotic episodes (higher risk). The ARMS designation accounted only for a small proportion of transitions to psychosis (n = 52 of 1001; 5.19% in the derivation data set), indicating the need for transdiagnostic prediction of psychosis in secondary mental health care. A prognostic risk stratification model based on preselected variables, including index diagnosis, age, sex, age by sex, and race/ethnicity, was developed and externally validated, showing good performance and potential clinical usefulness. This online individualized risk calculator can be of clinical usefulness for the transdiagnostic prediction of psychosis in secondary mental health care. The risk calculator can help to identify those patients at risk of developing psychosis who require an ARMS assessment and specialized care. The use of this calculator may eventually facilitate the implementation of an individualized provision of preventive focused interventions and improve outcomes of first episode psychosis.

Predictors of care home and hospital admissions and their costs for older people with Alzheimer's disease: findings from a large London case register

ObjectivesTo examine links between clinical and other characteristics of people with Alzheimer's disease living in the community, likelihood of care home or hospital admission, and associated costs.DesignObservational data extracted from...

Deconstructing Pretest Risk Enrichment to Optimize Prediction of Psychosis in Individuals at Clinical High Risk

Pretest risk estimation is routinely used in clinical medicine to inform further diagnostic testing in individuals with suspected diseases. To our knowledge, the overall characteristics and specific determinants of pretest risk of psychosis onset in individuals undergoing clinical high risk (CHR) assessment are unknown. To investigate the characteristics and determinants of pretest risk of psychosis onset in individuals undergoing CHR assessment and to develop and externally validate a pretest risk stratification model. Clinical register-based cohort study. Individuals were drawn from electronic, real-world, real-time clinical records relating to routine mental health care of CHR services in South London and the Maudsley National Health Service Trust in London, United Kingdom. The study included nonpsychotic individuals referred on suspicion of psychosis risk and assessed by the Outreach and Support in South London CHR service from 2002 to 2015. Model development and validation was performed with machine-learning methods based on Least Absolute Shrinkage and Selection Operator for Cox proportional hazards model. Pretest risk of psychosis onset in individuals undergoing CHR assessment. Predictors included age, sex, age × sex interaction, race/ethnicity, socioeconomic status, marital status, referral source, and referral year. A total of 710 nonpsychotic individuals undergoing CHR assessment were included. The mean age was 23 years. Three hundred ninety-nine individuals were men (56%), their race/ethnicity was heterogenous, and they were referred from a variety of sources. The cumulative 6-year pretest risk of psychosis was 14.55% (95% CI, 11.71% to 17.99%), confirming substantial pretest risk enrichment during the recruitment of individuals undergoing CHR assessment. Race/ethnicity and source of referral were associated with pretest risk enrichment. The predictive model based on these factors was externally validated, showing moderately good discrimination and sufficient calibration. It was used to stratify individuals undergoing CHR assessment into 4 classes of pretest risk (6-year): low, 3.39% (95% CI, 0.96% to 11.56%); moderately low, 11.58% (95% CI, 8.10% to 16.40%); moderately high, 23.69% (95% CI, 16.58% to 33.20%); and high, 53.65% (95% CI, 36.78% to 72.46%). Significant risk enrichment occurs before individuals are assessed for a suspected CHR state. Race/ethnicity and source of referral are associated with pretest risk enrichment in individuals undergoing CHR assessment. A stratification model can identify individuals at differential pretest risk of psychosis. Identification of these subgroups may inform outreach campaigns and subsequent testing and eventually optimize psychosis prediction.