Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems?
Abstract Clinical prediction models (CPMs) have become fundamental for risk stratification across healthcare. The CPM pipeline (development, validation, deployment, and impact assessment) is commonly viewed as a one-time activity, with model updating rarely considered and done in a somewhat ad hoc m...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2021-01-01
|
| Series: | Diagnostic and Prognostic Research |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s41512-020-00090-3 |
| _version_ | 1818821661835657216 |
|---|---|
| author | David A. Jenkins Glen P. Martin Matthew Sperrin Richard D. Riley Thomas P. A. Debray Gary S. Collins Niels Peek |
| author_facet | David A. Jenkins Glen P. Martin Matthew Sperrin Richard D. Riley Thomas P. A. Debray Gary S. Collins Niels Peek |
| author_sort | David A. Jenkins |
| collection | DOAJ |
| description | Abstract Clinical prediction models (CPMs) have become fundamental for risk stratification across healthcare. The CPM pipeline (development, validation, deployment, and impact assessment) is commonly viewed as a one-time activity, with model updating rarely considered and done in a somewhat ad hoc manner. This fails to address the fact that the performance of a CPM worsens over time as natural changes in populations and care pathways occur. CPMs need constant surveillance to maintain adequate predictive performance. Rather than reactively updating a developed CPM once evidence of deteriorated performance accumulates, it is possible to proactively adapt CPMs whenever new data becomes available. Approaches for validation then need to be changed accordingly, making validation a continuous rather than a discrete effort. As such, “living” (dynamic) CPMs represent a paradigm shift, where the analytical methods dynamically generate updated versions of a model through time; one then needs to validate the system rather than each subsequent model revision. |
| first_indexed | 2024-12-18T23:11:44Z |
| format | Article |
| id | doaj.art-d7bdafe1d9784bc08311853e561aad05 |
| institution | Directory Open Access Journal |
| issn | 2397-7523 |
| language | English |
| last_indexed | 2024-12-18T23:11:44Z |
| publishDate | 2021-01-01 |
| publisher | BMC |
| record_format | Article |
| series | Diagnostic and Prognostic Research |
| spelling | doaj.art-d7bdafe1d9784bc08311853e561aad052022-12-21T20:48:20ZengBMCDiagnostic and Prognostic Research2397-75232021-01-01511710.1186/s41512-020-00090-3Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems?David A. Jenkins0Glen P. Martin1Matthew Sperrin2Richard D. Riley3Thomas P. A. Debray4Gary S. Collins5Niels Peek6Division of Informatics, Imaging and Data Science, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester Academic Health Science CentreDivision of Informatics, Imaging and Data Science, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester Academic Health Science CentreDivision of Informatics, Imaging and Data Science, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester Academic Health Science CentreCentre for Prognosis Research, School of Primary, Community and Social Care, Keele UniversityJulius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht UniversityCentre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of OxfordDivision of Informatics, Imaging and Data Science, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester Academic Health Science CentreAbstract Clinical prediction models (CPMs) have become fundamental for risk stratification across healthcare. The CPM pipeline (development, validation, deployment, and impact assessment) is commonly viewed as a one-time activity, with model updating rarely considered and done in a somewhat ad hoc manner. This fails to address the fact that the performance of a CPM worsens over time as natural changes in populations and care pathways occur. CPMs need constant surveillance to maintain adequate predictive performance. Rather than reactively updating a developed CPM once evidence of deteriorated performance accumulates, it is possible to proactively adapt CPMs whenever new data becomes available. Approaches for validation then need to be changed accordingly, making validation a continuous rather than a discrete effort. As such, “living” (dynamic) CPMs represent a paradigm shift, where the analytical methods dynamically generate updated versions of a model through time; one then needs to validate the system rather than each subsequent model revision.https://doi.org/10.1186/s41512-020-00090-3Clinical prediction modelsDynamic modelValidationModel updatingModel developmentLearning health system |
| spellingShingle | David A. Jenkins Glen P. Martin Matthew Sperrin Richard D. Riley Thomas P. A. Debray Gary S. Collins Niels Peek Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems? Diagnostic and Prognostic Research Clinical prediction models Dynamic model Validation Model updating Model development Learning health system |
| title | Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems? |
| title_full | Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems? |
| title_fullStr | Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems? |
| title_full_unstemmed | Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems? |
| title_short | Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems? |
| title_sort | continual updating and monitoring of clinical prediction models time for dynamic prediction systems |
| topic | Clinical prediction models Dynamic model Validation Model updating Model development Learning health system |
| url | https://doi.org/10.1186/s41512-020-00090-3 |
| work_keys_str_mv | AT davidajenkins continualupdatingandmonitoringofclinicalpredictionmodelstimefordynamicpredictionsystems AT glenpmartin continualupdatingandmonitoringofclinicalpredictionmodelstimefordynamicpredictionsystems AT matthewsperrin continualupdatingandmonitoringofclinicalpredictionmodelstimefordynamicpredictionsystems AT richarddriley continualupdatingandmonitoringofclinicalpredictionmodelstimefordynamicpredictionsystems AT thomaspadebray continualupdatingandmonitoringofclinicalpredictionmodelstimefordynamicpredictionsystems AT garyscollins continualupdatingandmonitoringofclinicalpredictionmodelstimefordynamicpredictionsystems AT nielspeek continualupdatingandmonitoringofclinicalpredictionmodelstimefordynamicpredictionsystems |