Predicting high-cost care in a mental health setting
BackgroundThe density of information in digital health records offers new potential opportunities for automated prediction of cost-relevant outcomes.AimsWe investigated the extent to which routinely recorded data held in the electronic health record (EHR) predict priority service outcomes and whethe...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Cambridge University Press
2020-01-01
|
| Series: | BJPsych Open |
| Subjects: | |
| Online Access: | https://www.cambridge.org/core/product/identifier/S2056472419000966/type/journal_article |
| _version_ | 1827995721165438976 |
|---|---|
| author | Craig Colling Mizanur Khondoker Rashmi Patel Marcella Fok Robert Harland Matthew Broadbent Paul McCrone Robert Stewart |
| author_facet | Craig Colling Mizanur Khondoker Rashmi Patel Marcella Fok Robert Harland Matthew Broadbent Paul McCrone Robert Stewart |
| author_sort | Craig Colling |
| collection | DOAJ |
| description | BackgroundThe density of information in digital health records offers new potential opportunities for automated prediction of cost-relevant outcomes.AimsWe investigated the extent to which routinely recorded data held in the electronic health record (EHR) predict priority service outcomes and whether natural language processing tools enhance the predictions. We evaluated three high priority outcomes: in-patient duration, readmission following in-patient care and high service cost after first presentation.MethodWe used data obtained from a clinical database derived from the EHR of a large mental healthcare provider within the UK. We combined structured data with text-derived data relating to diagnosis statements, medication and psychiatric symptomatology. Predictors of the three different clinical outcomes were modelled using logistic regression with performance evaluated against a validation set to derive areas under receiver operating characteristic curves.ResultsIn validation samples, the full models (using all available data) achieved areas under receiver operating characteristic curves between 0.59 and 0.85 (in-patient duration 0.63, readmission 0.59, high service use 0.85). Adding natural language processing-derived data to the models increased the variance explained across all clinical scenarios (observed increase in r2 = 12–46%).ConclusionsEHR data offer the potential to improve routine clinical predictions by utilising previously inaccessible data. Of our scenarios, prediction of high service use after initial presentation achieved the highest performance. |
| first_indexed | 2024-04-10T05:00:47Z |
| format | Article |
| id | doaj.art-dea00952a29e43049ca1eb2c29f559bb |
| institution | Directory Open Access Journal |
| issn | 2056-4724 |
| language | English |
| last_indexed | 2024-04-10T05:00:47Z |
| publishDate | 2020-01-01 |
| publisher | Cambridge University Press |
| record_format | Article |
| series | BJPsych Open |
| spelling | doaj.art-dea00952a29e43049ca1eb2c29f559bb2023-03-09T12:28:54ZengCambridge University PressBJPsych Open2056-47242020-01-01610.1192/bjo.2019.96Predicting high-cost care in a mental health settingCraig Colling0https://orcid.org/0000-0001-5178-0383Mizanur Khondoker1Rashmi Patel2Marcella Fok3Robert Harland4Matthew Broadbent5Paul McCrone6Robert Stewart7Applied Clinical Informatics Lead, SLaM Biomedical Research Center, South London & Maudsley Foundation NHS Trust, UKSenior Lecturer in Medical Statistics, University of East Anglia, Norwich Medical School, UKMRC UKRI Health Data Research UK Fellow, Department of Psychosis Studies, Institute of Psychiatry, Psychology and Neuroscience, Kings College London; and South London & Maudsley Foundation NHS Trust, UKVisiting Researcher, Psychological Medicine, Institute of Psychiatry, Psychology and Neuroscience, Kings College London; and Central and North West London NHS Foundation Trust, UKClinical Director of Psychosis, Psychosis CAG, South London & Maudsley Foundation NHS Trust, UKInformatics Lead, SLaM Biomedical Research Center, South London & Maudsley Foundation NHS Trust, UKProfessor of Health Economics, School of Health Science, Institute of Psychiatry, Psychology and Neuroscience, Kings College London, UKProfessor of Psychiatric Epidemiology and Clinical Informatics, Psychological Medicine, Institute of Psychiatry, Psychology and Neuroscience, Kings College London; and South London & Maudsley Foundation NHS Trust, UKBackgroundThe density of information in digital health records offers new potential opportunities for automated prediction of cost-relevant outcomes.AimsWe investigated the extent to which routinely recorded data held in the electronic health record (EHR) predict priority service outcomes and whether natural language processing tools enhance the predictions. We evaluated three high priority outcomes: in-patient duration, readmission following in-patient care and high service cost after first presentation.MethodWe used data obtained from a clinical database derived from the EHR of a large mental healthcare provider within the UK. We combined structured data with text-derived data relating to diagnosis statements, medication and psychiatric symptomatology. Predictors of the three different clinical outcomes were modelled using logistic regression with performance evaluated against a validation set to derive areas under receiver operating characteristic curves.ResultsIn validation samples, the full models (using all available data) achieved areas under receiver operating characteristic curves between 0.59 and 0.85 (in-patient duration 0.63, readmission 0.59, high service use 0.85). Adding natural language processing-derived data to the models increased the variance explained across all clinical scenarios (observed increase in r2 = 12–46%).ConclusionsEHR data offer the potential to improve routine clinical predictions by utilising previously inaccessible data. Of our scenarios, prediction of high service use after initial presentation achieved the highest performance.https://www.cambridge.org/core/product/identifier/S2056472419000966/type/journal_articleDigital health recordsmental health servicepredictionnatural language processing |
| spellingShingle | Craig Colling Mizanur Khondoker Rashmi Patel Marcella Fok Robert Harland Matthew Broadbent Paul McCrone Robert Stewart Predicting high-cost care in a mental health setting BJPsych Open Digital health records mental health service prediction natural language processing |
| title | Predicting high-cost care in a mental health setting |
| title_full | Predicting high-cost care in a mental health setting |
| title_fullStr | Predicting high-cost care in a mental health setting |
| title_full_unstemmed | Predicting high-cost care in a mental health setting |
| title_short | Predicting high-cost care in a mental health setting |
| title_sort | predicting high cost care in a mental health setting |
| topic | Digital health records mental health service prediction natural language processing |
| url | https://www.cambridge.org/core/product/identifier/S2056472419000966/type/journal_article |
| work_keys_str_mv | AT craigcolling predictinghighcostcareinamentalhealthsetting AT mizanurkhondoker predictinghighcostcareinamentalhealthsetting AT rashmipatel predictinghighcostcareinamentalhealthsetting AT marcellafok predictinghighcostcareinamentalhealthsetting AT robertharland predictinghighcostcareinamentalhealthsetting AT matthewbroadbent predictinghighcostcareinamentalhealthsetting AT paulmccrone predictinghighcostcareinamentalhealthsetting AT robertstewart predictinghighcostcareinamentalhealthsetting |