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

TY - JOUR

T1 - Associations of remote mental healthcare with clinical outcomes: a natural language processing enriched electronic health record data study protocol

AU - Ahmed, Muhammad

AU - Kornblum, Daisy

AU - Oliver, Dominic

AU - Fusar-Poli, Paolo

AU - Patel, Rashmi

N1 - Funding Information: RP has received funding from an NIHR Advanced Fellowship (NIHR301690), a Medical Research Council (MRC) Health Data Research UK Fellowship (MR/S003118/1) and an Academy of Medical Sciences Starter Grant for Clinical Lecturers (SGL015/1020) supported by the Wellcome Trust, MRC, British Heart Foundation, Arthritis Research UK, the Royal College of Physicians and Diabetes UK. Funding Information: This project is funded by the National Institute for Health Research (Advanced Fellowship), grant number NIHR301690. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. The funders had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript and decision to submit the manuscript for publication. Publisher Copyright: © 2023 BMJ Publishing Group. All rights reserved.

PY - 2023/2/10

Y1 - 2023/2/10

N2 - INTRODUCTION: People 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 by reducing travel time and increasing frequency of contact with patients. However, the safe delivery of remote mental healthcare requires evidence on which aspects of care are suitable for remote delivery and which are better served by in-person care. We aim to investigate clinical and demographic associations with remote mental healthcare in a large electronic health record (EHR) dataset and the degree to which remote care is associated with differences in clinical outcomes using natural language processing (NLP) derived EHR data. METHODS AND ANALYSIS: Deidentified EHR data, derived from the South London and Maudsley (SLaM) National Health Service Foundation Trust Biomedical Research Centre (BRC) Case Register, will be extracted using the Clinical Record Interactive Search tool for all patients receiving mental healthcare between 1 January 2019 and 31 March 2022. First, data on a retrospective, longitudinal cohort of around 80 000 patients will be analysed using descriptive statistics to investigate clinical and demographic associations with remote mental healthcare and multivariable Cox regression to compare clinical outcomes of remote versus in-person assessments. Second, NLP models that have been previously developed to extract mental health symptom data will be applied to around 5 million documents to analyse the variation in content of remote versus in-person assessments. ETHICS AND DISSEMINATION: The SLaM BRC Case Register and Clinical Record Interactive Search (CRIS) tool have received ethical approval as a deidentified dataset (including NLP-derived data from unstructured free text documents) for secondary mental health research from Oxfordshire REC C (Ref: 18/SC/0372). The study has received approval from the SLaM CRIS Oversight Committee. Study findings will be disseminated through peer-reviewed, open access journal articles and service user and carer advisory groups.

AB - INTRODUCTION: People 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 by reducing travel time and increasing frequency of contact with patients. However, the safe delivery of remote mental healthcare requires evidence on which aspects of care are suitable for remote delivery and which are better served by in-person care. We aim to investigate clinical and demographic associations with remote mental healthcare in a large electronic health record (EHR) dataset and the degree to which remote care is associated with differences in clinical outcomes using natural language processing (NLP) derived EHR data. METHODS AND ANALYSIS: Deidentified EHR data, derived from the South London and Maudsley (SLaM) National Health Service Foundation Trust Biomedical Research Centre (BRC) Case Register, will be extracted using the Clinical Record Interactive Search tool for all patients receiving mental healthcare between 1 January 2019 and 31 March 2022. First, data on a retrospective, longitudinal cohort of around 80 000 patients will be analysed using descriptive statistics to investigate clinical and demographic associations with remote mental healthcare and multivariable Cox regression to compare clinical outcomes of remote versus in-person assessments. Second, NLP models that have been previously developed to extract mental health symptom data will be applied to around 5 million documents to analyse the variation in content of remote versus in-person assessments. ETHICS AND DISSEMINATION: The SLaM BRC Case Register and Clinical Record Interactive Search (CRIS) tool have received ethical approval as a deidentified dataset (including NLP-derived data from unstructured free text documents) for secondary mental health research from Oxfordshire REC C (Ref: 18/SC/0372). The study has received approval from the SLaM CRIS Oversight Committee. Study findings will be disseminated through peer-reviewed, open access journal articles and service user and carer advisory groups.

UR - http://www.scopus.com/inward/record.url?scp=85147894391&partnerID=8YFLogxK

U2 - 10.1136/bmjopen-2022-067254

DO - 10.1136/bmjopen-2022-067254

M3 - Article

SN - 2044-6055

VL - 13

SP - e067254

JO - BMJ Open

JF - BMJ Open

IS - 2

M1 - e067254

ER -