Comparing natural language processing representations of disease sequences for prediction in the electronic healthcare record

Natural language processing (NLP) is increasingly being applied to obtain unsupervised representations of electronic healthcare record (EHR) data, but their performance for the prediction of clinical endpoints remains unclear. Here we use primary care EHRs from 6,286,233 people with Multiple Long-Term Conditions in England to generate vector representations of sequences of disease development using two input strategies (212 disease categories versus 9,462 diagnostic codes) and different NLP algorithms (Latent Dirichlet Allocation, doc2vec and two transformer models designed for EHRs). We also develop a new transformer architecture, named EHR-BERT, which incorporates socio-demographic information. We then compare use of each of these representations to predict mortality, healthcare use and new disease diagnosis. We find that representations generated using disease categories perform similarly to those using diagnostic codes, suggesting models can equally manage smaller or larger vocabularies. Sequence-based algorithms perform consistently better than bag-of-words methods, with the highest performance for EHR-BERT. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement This work was funded through a clinical PhD fellowship from the Wellcome Trust. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: Data access to CPRD and ethical approval was granted by the CPRD Research Data Governance Process on 28th April 2022 (Protocol reference: 22\_001818) and with linkage to HES data on 6th March 2023 (Protocol reference: 22\_002481). I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes This study uses patient data which is not publicly available but can be requested for users meeting certain requirements: https://cprd.com/research-applications. Codes, including the Medcode to disease mapping are available from https://tbeaney.github.io/MMclustering/