NeuroBlu, an electronic health record (EHR) trusted research environment (TRE) to support mental healthcare analytics with real-world data

Optimising the use of electronic medical records for large scale research in psychiatry

The explosion and abundance of digital data could facilitate large-scale research for psychiatry and mental health. Research using so-called "real world data"-such as electronic medical/health records-can be resource-efficient, facilitate rapid hypothesis generation and testing, complement existing evidence (e.g. from trials and evidence-synthesis) and may enable a route to translate evidence into clinically effective, outcomes-driven care for patient populations that may be under-represented. However, the interpretation and processing of real-world data sources is complex because the clinically important 'signal' is often contained in both structured and unstructured (narrative or "free-text") data. Techniques for extracting meaningful information (signal) from unstructured text exist and have advanced the re-use of routinely collected clinical data, but these techniques require cautious evaluation. In this paper, we survey the opportunities, risks and progress made in the use of electronic medical record (real-world) data for psychiatric research.

A review of information sources and analysis methods for data driven decision aids in child and adolescent mental health services

OBJECTIVE

Clinical data analysis relies on effective methods and appropriate data. Recognizing distinctive clinical services and service functions may lead to improved decision-making. Our first objective is to categorize analytical methods, data sources, and algorithms used in current research on information analysis and decision support in child and adolescent mental health services (CAMHS). Our secondary objective is to identify the potential for data analysis in different clinical services and functions in which data-driven decision aids can be useful.

MATERIALS AND METHODS

We searched related studies in Science Direct and PubMed from 2018 to 2023(Jun), and also in ACM (Association for Computing Machinery) Digital Library, DBLP (Database systems and Logic Programming), and Google Scholar from 2018 to 2021. We have reviewed 39 studies and extracted types of analytical methods, information content, and information sources for decision-making.

RESULTS

In order to compare studies, we developed a framework for characterizing health services, functions, and data features. Most data sets in reviewed studies were small, with a median of 1,550 patients and 46,503 record entries. Structured data was used for all studies except two that used textual clinical notes. Most studies used supervised classification and regression. Service and situation-specific data analysis dominated among the studies, only two studies used temporal, or process features from the patient data. This paper presents and summarizes the utility, but not quality, of the studies according to the care situations and care providers to identify service functions where data-driven decision aids may be relevant.

CONCLUSIONS

Frameworks identifying services, functions, and care processes are necessary for characterizing and comparing electronic health record (EHR) data analysis studies. The majority of studies use features related to diagnosis and assessment and correspondingly have utility for intervention planning and follow-up. Profiling the disease severity of referred patients is also an important application area.

Psychometric evaluation and linking of the PHQ-9, QIDS-C, and VQIDS-C in a real-world population with major depressive disorder

Purpose

Major depressive disorder (MDD) is a leading cause of disability worldwide. An accurate assessment of depressive symptomology is crucial for clinical management and research. This study assessed the convergent validity, reliability, and total scale score interconversion across the 9-item Patient Health Questionnaire (PHQ-9) self-report, the 16-item Quick Inventory of Depressive Symptomatology-clinician report (QIDS-C) (two widely used clinical ratings) and the 5-item Very Brief Quick Inventory of Depressive Symptoms-clinician report (VQIDS-C), which evaluate the core features of MDD.

Patients and Methods

This study leveraged electronic health record (EHR)-derived, de-identified data from the NeuroBlu Database (Version 23R1), a longitudinal behavioural health real-world platform. Classical Test Theory (CTT) and Item Response Theory (IRT) analyses were used to evaluate the reliability, validity of, and conversions between the scales. The Test Information Function (TIF) was calculated for each scale, with greater test information reflecting higher precision and reliability in measuring depressive symptomology. IRT was also used to generate conversion tables so that total scores on each scale could be compared to the other.

Results

The study sample (n = 2,156) had an average age of 36.4 years (standard deviation [SD] = 13.0) and 59.7% were female. The mean depression scores for the PHQ-9, QIDS-C, and VQIDS-C were 12.9 (SD = 6.6), 12.0 (SD = 4.9), and 6.18 (SD = 3.2), respectively. The Cronbach's alpha coefficients for PHQ-9, QIDS-C, and VQIDS-C were 0.9, 0.8, and 0.7, respectively, suggesting acceptable internal consistency. PHQ-9 (TIF = 30.3) demonstrated the best assessment of depressive symptomology, followed by QIDS-C (TIF = 25.8) and VQIDS-C (TIF = 17.7).

Conclusion

Overall, PHQ-9, QIDS-C, and VQIDS-C appear to be reliable and convertible measures of MDD symptomology within a US-based adult population in a real-world clinical setting.

Identifying features of risk periods for suicide attempts using document frequency and language use in electronic health records

Background

Individualising mental healthcare at times when a patient is most at risk of suicide involves shifting research emphasis from static risk factors to those that may be modifiable with interventions. Currently, risk assessment is based on a range of extensively reported stable risk factors, but critical to dynamic suicide risk assessment is an understanding of each individual patient's health trajectory over time. The use of electronic health records (EHRs) and analysis using machine learning has the potential to accelerate progress in developing early warning indicators.

Setting

EHR data from the South London and Maudsley NHS Foundation Trust (SLaM) which provides secondary mental healthcare for 1.8 million people living in four South London boroughs.

Objectives

To determine whether the time window proximal to a hospitalised suicide attempt can be discriminated from a distal period of lower risk by analysing the documentation and mental health clinical free text data from EHRs and (i) investigate whether the rate at which EHR documents are recorded per patient is associated with a suicide attempt; (ii) compare document-level word usage between documents proximal and distal to a suicide attempt; and (iii) compare n-gram frequency related to third-person pronoun use proximal and distal to a suicide attempt using machine learning.

Methods

The Clinical Record Interactive Search (CRIS) system allowed access to de-identified information from the EHRs. CRIS has been linked with Hospital Episode Statistics (HES) data for Admitted Patient Care. We analysed document and event data for patients who had at some point between 1 April 2006 and 31 March 2013 been hospitalised with a HES ICD-10 code related to attempted suicide (X60-X84; Y10-Y34; Y87.0/Y87.2).

Findings

n = 8,247 patients were identified to have made a hospitalised suicide attempt. Of these, n = 3,167 (39.8%) of patients had at least one document available in their EHR prior to their first suicide attempt. N = 1,424 (45.0%) of these patients had been "monitored" by mental healthcare services in the past 30 days. From 60 days prior to a first suicide attempt, there was a rapid increase in the monitoring level (document recording of the past 30 days) increasing from 35.1 to 45.0%. Documents containing words related to prescribed medications/drugs/overdose/poisoning/addiction had the highest odds of being a risk indicator used proximal to a suicide attempt (OR 1.88; precision 0.91 and recall 0.93), and documents with words citing a care plan were associated with the lowest risk for a suicide attempt (OR 0.22; precision 1.00 and recall 1.00). Function words, word sequence, and pronouns were most common in all three representations (uni-, bi-, and tri-gram).

Conclusion

EHR documentation frequency and language use can be used to distinguish periods distal from and proximal to a suicide attempt. However, in our study 55.0% of patients with documentation, prior to their first suicide attempt, did not have a record in the preceding 30 days, meaning that there are a high number who are not seen by services at their most vulnerable point.

Big Data in Health Care: An Interprofessional Course

BACKGROUND

The widespread adoption of the electronic health record (EHR) has resulted in vast repositories of EHR big data that are being used to identify patterns and correlations that translate into data-informed health care decision making.

PROBLEM

Health care professionals need the skills necessary to navigate a digitized, data-rich health care environment as big data plays an increasingly integral role in health care.

APPROACH

Faculty incorporated the concept of big data in an asynchronous online course allowing an interprofessional mix of students to analyze EHR big data on over a million patients.

OUTCOMES

Students conducted a descriptive analysis of cohorts of patients with selected diagnoses and presented their findings.

CONCLUSIONS

Students collaborated with an interprofessional team to analyze EHR big data on selected variables. The teams used data visualization tools to describe an assigned diagnosis patient population.

Associations of comorbid substance use disorders with clinical outcomes in schizophrenia using electronic health record data

BACKGROUND AND HYPOTHESIS

Schizophrenia and comorbid substance use disorders (SUDs) are associated with poor treatment outcomes but differences between the associations of different SUDs with clinical outcomes are poorly characterized. This study examines the associations of comorbid SUDs with clinical outcomes in schizophrenia using a largescale electronic health record (EHR) database.

DESIGN

Real-world data (RWD) analysis using the NeuroBlu database; de-identified EHR data were analysed. Multivariable logistic regression, Poisson and CoxPH models were used to compare the associations of specific comorbid SUDs with outcome variables.

RESULTS

Comorbid SUD was significantly different on all outcome measures compared to no SUD (U = 1.44e7-1.81e7, all ps < .001), except number of unique antipsychotics (U = 1.61e7, p = .43). Cannabis (OR = 1.58, p < .001) and polysubstance (OR = 1.22, p = .007) use disorders were associated with greater CGI-S. Cannabis (IRR = 1.13, p = .003) and polysubstance (IRR = 1.08, p = .003) use disorders were associated with greater number of unique antipsychotics prescribed, while cocaine (HR = 1.87, p < .001), stimulants (HR = 1.64, p = .024), and polysubstance (HR = 1.46, p < .001) use disorders were associated with a shorter time to antipsychotic discontinuation. Conversely, alcohol use (IRR = 0.83, p < .001), cocaine use (IRR = 0.61, p < .001), opioid use (IRR = 0.61, p < .001), stimulant use (IRR = 0.57, p < .001) and polysubstance use (IRR = 0.87, p < .001) disorders were associated fewer inpatient days.

CONCLUSION

Comorbid SUDs were generally associated with greater CGI-S and poorer clinical outcomes in patients with schizophrenia. Treatment strategies should target not only schizophrenia symptoms but also comorbid SUD to improve management of both conditions.

Comparative effectiveness study of paliperidone palmitate 6-month with a real-world external comparator arm of paliperidone palmitate 1-month or 3-month in patients with schizophrenia

Background

The paliperidone palmitate 6-month (PP6M) long-acting injectable formulation is currently the longest dosing interval available for schizophrenia treatment.

Objective

To compare treatment outcomes between a real-world external comparator arm (ECA; NeuroBlu database) and the PP6M open-label extension (OLE) clinical trial arm.

Methods

The ECA comprised patients receiving PP 1-month (PP1M) or PP 3-month (PP3M) for ⩾12 months without a relapse. The PP6M OLE arm included patients with PP1M treatment prior to randomization who completed the 12-month double-blind PP6M study on either PP3M or PP6M relapse-free. Inverse probability treatment weighting (IPTW) was used to study time-to-relapse (primary outcome) and change in Clinical Global Impressions-Severity (CGI-S) score (secondary outcome).

Results

At 24 months, 3.9% (7/178) of patients in the PP6M cohort experienced a relapse versus 15.6% (26/167) in the ECA. Time-to-relapse was longer in the PP6M cohort versus the ECA at 12-, 18-, and 24-months across the different weighting methods; median time-to-relapse was not reached in both cohorts. Hazard ratio (HR) for relapse was significantly lower for the PP6M cohort versus the ECA throughout the duration of the study [HR at 24 months: 0.18 (95% CI: 0.08-0.42), p < 0.001]. At 24 months, change in CGI-S score for the PP6M cohort was 0.76 points lower than the ECA (p < 0.001). Results were similar in a sensitivity analysis using propensity score matching (PSM); IPTW resulted in larger sample sizes in balanced dataset than PSM.

Conclusion

Consistent findings across weighting and matching methods suggest PP6M efficacy in reducing and delaying relapses and long-term symptom control compared to PP1M/PP3M in usual-care settings. Additional confounds, such as greater illness severity and more frequent comorbidities and comedications in the ECA, were not fully controlled by the applied statistical methods. Future real-world studies directly comparing PP6M with PP3M/PP1M and adjusting for these confounders are warranted.

Early trajectory of clinical global impression as a transdiagnostic predictor of psychiatric hospitalisation: a retrospective cohort study

BACKGROUND

Identifying patients most at risk of psychiatric hospitalisation is crucial to improving service provision and patient outcomes. Existing predictors focus on specific clinical scenarios and are not validated with real-world data, limiting their translational potential. This study aimed to determine whether early trajectories of Clinical Global Impression Severity are predictors of 6 month risk of hospitalisation.

METHODS

This retrospective cohort study used data from the NeuroBlu database, an electronic health records network from 25 US mental health-care providers. Patients with an ICD-9 or ICD-10 code of major depressive disorder, bipolar disorder, generalised anxiety disorder, post-traumatic stress disorder, schizophrenia or schizoaffective disorder, ADHD, or personality disorder were included. Using this cohort, we assessed whether clinical severity and instability (operationalised using Clinical Global Impression Severity measurements) during a 2-month period were predictors of psychiatric hospitalisation within the next 6 months.

FINDINGS

36 914 patients were included (mean age 29·7 years [SD 17·5]; 21 156 [57·3%] female, 15 748 [42·7%] male; 20 559 [55·7%] White, 4842 [13·1%] Black or African American, 286 [0·8%] Native Hawaiian or other Pacific Islander, 300 [0·8%] Asian, 139 [0·4%] American Indian or Alaska Native, 524 (1·4%) other or mixed race, and 10 264 [27·8%] of unknown race). Clinical severity and instability were independent predictors of risk of hospitalisation (adjusted hazard ratio [HR] 1·09, 95% CI 1·07-1·10 for every SD increase in instability; 1·11, 1·09-1·12 for every SD increase in severity; p<0·0001 for both). These associations were consistent across all diagnoses, age groups, and in both males and females, as well as in several robustness analyses, including when clinical severity and clinical instability were based on the Patient Health Questionnaire-9 rather than Clinical Global Impression Severity measurements. Patients in the top half of the cohort for both clinical severity and instability were at an increased risk of hospitalisation compared with those in the bottom half along both dimensions (HR 1·45, 95% CI 1·39-1·52; p<0·0001).

INTERPRETATION

Clinical instability and severity are independent predictors of future risk of hospitalisation, across diagnoses, age groups, and in both males and females. These findings could help clinicians make prognoses and screen patients who are most likely to benefit from intensive interventions, as well as help health-care providers plan service provisions by adding additional detail to risk prediction tools that incorporate other risk factors.

FUNDING

National Institute for Health and Care Research, National Institute for Health and Care Research Oxford Health Biomedical Research Centre, Medical Research Council, Academy of Medical Sciences, and Holmusk.

A Computational Platform to Support the Detection, Follow-up, and Epidemiological Surveillance of Mental Health and Substance Use Disorders: Protocol for a Development and Evaluation Study

BACKGROUND

According to the World Health Organization, approximately 15% of the global population is affected by mental health or substance use disorders. These conditions contribute significantly to the global disease burden, which has worsened because of the direct and indirect effects of COVID-19. In Mexico, a quarter of the population between the ages of 18 and 65 years who reside in urban areas present a mental health condition. The presence of a mental or substance abuse disorder is behind a significant percentage of suicidal behaviors in Mexico, where only 1 in 5 of those who have these disorders receive any treatment.

OBJECTIVE

This study aims to develop, deploy, and evaluate a computational platform to support the early detection and intervention of mental and substance use disorders in secondary and high schools as well as primary care units. The platform also aims to facilitate monitoring, treatment, and epidemiological surveillance ultimately helping specialized health units at the secondary level of care.

METHODS

The development and evaluation of the proposed computational platform will run during 3 stages. In stage 1, the identification of the functional and user requirements and the implementation of the modules to support the screening, follow-up, treatment, and epidemiological surveillance will be performed. In stage 2, the initial deployment of the screening module will be carried out in a set of secondary and high schools, as well as the deployment of the modules to support the follow-up, treatment, and epidemiological surveillance processes in primary and secondary care health units. In parallel, during stage 2, patient applications to support early interventions and continuous monitoring will also be developed. Finally, during stage 3, the deployment of the complete platform will be performed jointly with a quantitative and qualitative evaluation.

RESULTS

The screening process has started, and 6 schools have been currently enrolled. As of February 2023, a total of 1501 students have undergone screening, and the referral of those students presenting a risk in mental health or substance use to primary care units has also started. The development, deployment, and evaluation of all the modules of the proposed platform are expected to be completed by late 2024.

CONCLUSIONS

The expected results of this study are to impact a better integration between the different levels of health care, from early detection to follow-up and epidemiological surveillance of mental and substance use disorders contributing to reducing the gap in the attention to these problems in the community.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/44607.

Psychiatric healthcare resource utilization following initiation of aripiprazole once-monthly: a retrospective real-world study

OBJECTIVES

This observational retrospective real-world study examined changes in healthcare resource utilization (HCRU) pre- and post-initiation of aripiprazole once-monthly (AOM 400) in patients with schizophrenia or bipolar I disorder.

METHODS

Electronic health record-derived, de-identified data from the NeuroBlu Database (2013-2020) were used to identify patients ≥18 years with schizophrenia (n = 222) or bipolar I disorder (n = 129) who were prescribed AOM 400, and had visit data within 3, 6, 9, or 12 months pre- and post-initial AOM 400 prescription. Rates of inpatient hospitalization, emergency department visits, inpatient readmissions, and average length of stay were examined and compared over 3, 6, 9, and 12 months pre-/post-AOM 400 using a McNemar test.

RESULTS

Statistically significant differences were seen in both schizophrenia and bipolar I disorder patient cohorts pre- and post-AOM 400 in inpatient hospitalization rates (p < .001 all time points, both cohorts) and 30-day readmission per patient rates (p < .001 all time points, both cohorts). Statistically significant improvement in mean length of stay was observed in both cohorts at all time points, except for at six months in patients with schizophrenia. Emergency department visit rates were significantly lower after AOM 400 initiation for both cohorts at all time points (p < .001).

CONCLUSIONS

A reduction in the rate of hospitalizations, emergency department visits, 30-day readmissions, and average length-of-stay was observed for patients diagnosed with either schizophrenia or bipolar I disorder, which suggests a positive effect of AOM 400 treatment on HCRU outcomes and is supportive of earlier analyses from different data sources.

[Effect of antibiotics on D-serotonin-reactive structures]

Experiments on the isolated organs showed that ampicillin and levomycetin have pronounced D-antiserotoninergic effects; antagonism of antibodies and serotonin was found to be of competitive type. At an increase in levomycetin dosage D-antiserotoninergic effect was followed by the spasmolytic effect. Kefzol and benzylpenicillin failed to show any D-antiserotonin-ergic properties.