Combining information from a clinical data warehouse and a pharmaceutical database to generate a framework to detect comorbidities in electronic health records

BackgroundMedical coding is used for a variety of activities, from observational studies to hospital billing. However, comorbidities tend to be under-reported by medical coders. The aim of this study was to develop an algorithm to detect comorbidities in electronic health records (EHR) by using a clinical data warehouse (CDW) and a knowledge database.MethodsWe enriched the Theriaque pharmaceutical database with the French national Comorbidities List to identify drugs associated with at least one major comorbid condition and diagnoses associated with a drug indication. Then, we compared the drug indications in the Theriaque database with the ICD-10 billing codes in EHR to detect potentially missing comorbidities based on drug prescriptions. Finally, we improved comorbidity detection by matching drug prescriptions and laboratory test results. We tested the obtained algorithm by using two retrospective datasets extracted from the Rennes University Hospital (RUH) CDW. The first dataset included all adult patients hospitalized in the ear, nose, throat (ENT) surgical ward between October and December 2014 (ENT dataset). The second included all adult patients hospitalized at RUH between January and February 2015 (general dataset). We reviewed medical records to find written evidence of the suggested comorbidities in current or past stays.ResultsAmong the 22,132 Common Units of Dispensation (CUD) codes present in the Theriaque database, 19,970 drugs (90.2%) were associated with one or several ICD-10 diagnoses, based on their indication, and 11,162 (50.4%) with at least one of the 4878 comorbidities from the comorbidity list. Among the 122 patients of the ENT dataset, 75.4% had at least one drug prescription without corresponding ICD-10 code. The comorbidity diagnoses suggested by the algorithm were confirmed in 44.6% of the cases. Among the 4312 patients of the general dataset, 68.4% had at least one drug prescription without corresponding ICD-10 code. The comorbidity diagnoses suggested by the algorithm were confirmed in 20.3% of reviewed cases.ConclusionsThis simple algorithm based on combining accessible and immediately reusable data from knowledge databases, drug prescriptions and laboratory test results can detect comorbidities.

[1]  H. Prokosch,et al.  Perspectives for Medical Informatics , 2009, Methods of Information in Medicine.

[2]  Yuan Luo,et al.  Identifying patient smoking status from medical discharge records. , 2008, Journal of the American Medical Informatics Association : JAMIA.

[3]  Thomas Agoritsas,et al.  Activity-Based Funding of Hospitals and Its Impact on Mortality, Readmission, Discharge Destination, Severity of Illness, and Volume of Care: A Systematic Review and Meta-Analysis , 2014, PloS one.

[4]  Elias Mossialos,et al.  Regulating Pharmaceuticals In Europe: Striving For Efficiency, Equity And Quality , 2004 .

[5]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[6]  Matthew R. Sydes,et al.  Technical Brief: Using Implicit Information to Identify Smoking Status in Smoke-blind Medical Discharge Summaries , 2008, J. Am. Medical Informatics Assoc..

[7]  Marc Cuggia,et al.  Semantic integration of medication data into the EHOP Clinical Data Warehouse , 2015, MIE.

[8]  L. Hacin,et al.  Évaluation des bases de données médicamenteuses disponibles sur le marché français , 2013 .

[9]  Jay Urbain,et al.  Mining heart disease risk factors in clinical text with named entity recognition and distributional semantic models , 2015, J. Biomed. Informatics.

[10]  Rob Koeling,et al.  Optimising the use of electronic health records to estimate the incidence of rheumatoid arthritis in primary care: what information is hidden in free text? , 2013, BMC Medical Research Methodology.

[11]  André Happe,et al.  Roogle: An Information Retrieval Engine for Clinical Data Warehouse , 2011, MIE.

[12]  Melissa A. Basford,et al.  Validation of electronic medical record-based phenotyping algorithms: results and lessons learned from the eMERGE network. , 2013, Journal of the American Medical Informatics Association : JAMIA.

[13]  William K. Thompson,et al.  Comparison of traditional trigger tool to data warehouse based screening for identifying hospital adverse events , 2012, BMJ quality & safety.

[14]  Serguei V. S. Pakhomov,et al.  Identifying Family History and Substance Use Associations for Adult Epilepsy from the Electronic Health Record , 2016, CRI.

[15]  Régis Beuscart,et al.  Adverse drug events with hyperkalaemia during inpatient stays: evaluation of an automated method for retrospective detection in hospital databases , 2014, BMC Medical Informatics and Decision Making.

[16]  Kevin B. Johnson,et al.  Presenting Multiple Drug Alerts in an Ambulatory Electronic Prescribing System , 2014, Applied Clinical Informatics.

[17]  Christopher M. Burkle,et al.  Ten common questions (and their answers) about off-label drug use. , 2012, Mayo Clinic proceedings.

[18]  Patrick Taffé,et al.  Improved accuracy of co-morbidity coding over time after the introduction of ICD-10 administrative data , 2011, BMC health services research.

[19]  Christian Lovis,et al.  Model Formulation: A Risk Analysis Method to Evaluate the Impact of a Computerized Provider Order Entry System on Patient Safety , 2008, J. Am. Medical Informatics Assoc..

[20]  M.-C. Husson.  Thériaque® : base de données indépendante sur le médicament, outil de bon usage pour les professionnels de santé , 2008 .

[21]  Griffin M. Weber,et al.  Serving the enterprise and beyond with informatics for integrating biology and the bedside (i2b2) , 2010, J. Am. Medical Informatics Assoc..

[22]  H. Quan,et al.  Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data , 2005, Medical care.

[23]  Özlem Uzuner,et al.  Viewpoint Paper: Recognizing Obesity and Comorbidities in Sparse Data , 2009, J. Am. Medical Informatics Assoc..

[24]  W. King,et al.  The effect of computerized physician order entry on medication errors and adverse drug events in pediatric inpatients. , 2003, Pediatrics.

[25]  Dan Wang,et al.  Similarity-based prediction for Anatomical Therapeutic Chemical classification of drugs by integrating multiple data sources , 2015, Bioinform..

[26]  B. Édouard,et al.  [Assessment of pharmaceutical databases available in France]. , 2013, Annales pharmaceutiques francaises.

[27]  Zhi Zheng,et al.  Current diagnosis and management of malignant gastrocolic fistulas: a single surgical unit's experience. , 2014, International journal of clinical and experimental medicine.

[28]  Eneida A. Mendonça,et al.  Relational machine learning for electronic health record-driven phenotyping , 2014, J. Biomed. Informatics.

[29]  M. Wojczynski,et al.  Definition of phenotype. , 2008, Advances in genetics.

[30]  Joseph T. Hanlon,et al.  The Medication Appropriateness Index at 20: Where It Started, Where It Has Been, and Where It May Be Going , 2013, Drugs & Aging.

[31]  C. Chute,et al.  Electronic Medical Records for Genetic Research: Results of the eMERGE Consortium , 2011, Science Translational Medicine.

[32]  Annunziata Faustini,et al.  The Reliability of Hospital and Pharmaceutical Data to Assess Prevalent Cases of Chronic Obstructive Pulmonary Disease , 2012, COPD.

[33]  Xin Liu,et al.  An automatic system to identify heart disease risk factors in clinical texts over time , 2015, J. Biomed. Informatics.

[34]  John Spertus,et al.  AHA/ACCF Secondary Prevention and Risk Reduction Therapy for Patients with Coronary and other Atherosclerotic Vascular Disease: 2011 update: a guideline from the American Heart Association and American College of Cardiology Foundation. , 2011, Circulation.

[35]  Matthias Eikermann,et al.  Codifying healthcare – big data and the issue of misclassification , 2015, BMC Anesthesiology.

[36]  N. Barber,et al.  Regulating Pharmaceuticals in Europe: Striving for Efficiency, Equity and Quality , 2005, Quality and Safety in Health Care.