Dataset Information

A pragmatic method for electronic medical record-based observational studies: developing an electronic medical records retrieval system for clinical research.

ABSTRACT: OBJECTIVE:The use of electronic medical record (EMR) data is necessary to improve clinical research efficiency. However, it is not easy to identify patients who meet research eligibility criteria and collect the necessary information from EMRs because the data collection process must integrate various techniques, including the development of a data warehouse and translation of eligibility criteria into computable criteria. This research aimed to demonstrate an electronic medical records retrieval system (ERS) and an example of a hospital-based cohort study that identified both patients and exposure with an ERS. We also evaluated the feasibility and usefulness of the method. DESIGN:The system was developed and evaluated. PARTICIPANTS:In total, 800 000 cases of clinical information stored in EMRs at our hospital were used. PRIMARY AND SECONDARY OUTCOME MEASURES:The feasibility and usefulness of the ERS, the method to convert text from eligible criteria to computable criteria, and a confirmation method to increase research data accuracy. RESULTS:To comprehensively and efficiently collect information from patients participating in clinical research, we developed an ERS. To create the ERS database, we designed a multidimensional data model optimised for patient identification. We also devised practical methods to translate narrative eligibility criteria into computable parameters. We applied the system to an actual hospital-based cohort study performed at our hospital and converted the test results into computable criteria. Based on this information, we identified eligible patients and extracted data necessary for confirmation by our investigators and for statistical analyses with our ERS. CONCLUSIONS:We propose a pragmatic methodology to identify patients from EMRs who meet clinical research eligibility criteria. Our ERS allowed for the efficient collection of information on the eligibility of a given patient, reduced the labour required from the investigators and improved the reliability of the results.

SUBMITTER: Yamamoto K

PROVIDER: S-EPMC3533000 | biostudies-literature | 2012

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A pragmatic method for electronic medical record-based observational studies: developing an electronic medical records retrieval system for clinical research.

Yamamoto Keiichi K Sumi Eriko E Yamazaki Toru T Asai Keita K Yamori Masashi M Teramukai Satoshi S Bessho Kazuhisa K Yokode Masayuki M Fukushima Masanori M

BMJ open 20121031 6

<h4>Objective</h4>The use of electronic medical record (EMR) data is necessary to improve clinical research efficiency. However, it is not easy to identify patients who meet research eligibility criteria and collect the necessary information from EMRs because the data collection process must integrate various techniques, including the development of a data warehouse and translation of eligibility criteria into computable criteria. This research aimed to demonstrate an electronic medical records ...[more]

PMID: 23117567

Similar Datasets

Project description:BACKGROUND:Chronic kidney disease (CKD) is a common chronic condition and a rising public health issue with increased morbidity and mortality, even at an early stage. Primary care has a pivotal role in the early detection and in the integrated management of CKD which should be of high quality. The quality of care for CKD can be assessed using quality indicators (QIs) and if these QIs are extractable from the electronic medical record (EMR) of the general physician, the number of patients whose quality of care can be evaluated, could increase vastly. Therefore the aim of this study is to develop QIs which are evidence based, EMR extractable and which can be used as a framework to automate quality assessment. METHODS:We used a Rand-modified Delphi method to develop QIs for CKD in primary care. A questionnaire was designed by extracting recommendations from international guidelines based on the SMART principle and the EMR extractability. A multidisciplinary expert panel, including patients, individually scored the recommendations for measuring high quality care on a 9-point Likert scale. The results were analyzed based on the median Likert score, prioritization and agreement. Subsequently, the recommendations were discussed in a consensus meeting for their in- or exclusion. After a final appraisal by the panel members this resulted in a core set of recommendations, which were then transformed into QIs. RESULTS:A questionnaire composed of 99 recommendations was extracted from 10 international guidelines. The consensus meeting resulted in a core set of 36 recommendations that were translated into 36 QIs. This final set consists of QIs concerning definition & classification, screening, diagnosis, management consisting of follow up, treatment & vaccination, medication & patient safety and referral to a specialist. It were mostly the patients participating in the panel who stressed the importance of the QIs concerning medication & patient safety and a timely referral to a specialist. CONCLUSION:This study provides a set of 36 EMR extractable QIs for measuring the quality of primary care for CKD. These QIs can be used as a framework to automate quality assessment for CKD in primary care.

Dataset Information

A pragmatic method for electronic medical record-based observational studies: developing an electronic medical records retrieval system for clinical research.

Publications

A pragmatic method for electronic medical record-based observational studies: developing an electronic medical records retrieval system for clinical research.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets