Electronic Patient Record Data Research Articles

Using text‐mining techniques in electronic patient records to identify ADRs from medicine use

This literature review included studies that use text-mining techniques in narrative documents stored in electronic patient records (EPRs) to investigate ADRs. We searched PubMed, Embase, Web of Science and International Pharmaceutical Abstracts without restrictions from origin until July 2011. We included empirically based studies on text mining of electronic patient records (EPRs) that focused on detecting ADRs, excluding those that investigated adverse events not related to medicine use. We extracted information on study populations, EPR data sources, frequencies and types of the identified ADRs, medicines associated with ADRs, text-mining algorithms used and their performance. Seven studies, all from the United States, were eligible for inclusion in the review. Studies were published from 2001, the majority between 2009 and 2010. Text-mining techniques varied over time from simple free text searching of outpatient visit notes and inpatient discharge summaries to more advanced techniques involving natural language processing (NLP) of inpatient discharge summaries. Performance appeared to increase with the use of NLP, although many ADRs were still missed. Due to differences in study design and populations, various types of ADRs were identified and thus we could not make comparisons across studies. The review underscores the feasibility and potential of text mining to investigate narrative documents in EPRs for ADRs. However, more empirical studies are needed to evaluate whether text mining of EPRs can be used systematically to collect new information about ADRs.

Securing EPR Data Using Cryptography and Image Watermarking

This paper demonstrates new methodology to improve security and avoid data overlapping between patients records which are defined as Electronic Patient Records (EPR), a combination of digital watermarking techniques and cryptography are used to ensure the non-separation of EPR and medical images during communications within open networks. The EPR data is encrypted, by a symmetric key algorithm based on an Elliptic Curve Cryptosystem (ECC), and inserted in liberated zone of the Least Significant Bit plan (LSB) of the medical image by compressing the original one using the Huffman coding. The proposed method improves security issues and reduces the computation cost related to data encryption and decryption.

Incidence and prevalence of “diagnosed OCD” in a primary care, treatment seeking, population

Objective. To obtain valid and accurate estimates of the incidence and prevalence of OCD in a treatment-seeking primary care population and to compare these estimates with estimates from epidemiological community studies. Methods. A retrospective cohort study (1996–2007) was conducted in a GP research database with longitudinal electronic patient record data of 800,000 patients throughout The Netherlands. OCD was ascertained and classified by systematic review of computerized longitudinal medical records. Age and gender specific incidence rates were calculated per calendar year as the number of newly diagnosed cases per 100 person years. Results. Among 577,085 eligible patients, 346 patients were newly diagnosed with OCD resulting in a 1-year treatment-seeking incidence of 0.016% (95% CI: 0.014–0.018). Across the entire study period, a total of 780 patients had a clinical diagnosis of OCD resulting in a treatment-seeking prevalence of 0.14% (95% CI: 0.126–0.145). The incidence rate was highest among women and between the age of 20 and 29. No significant changes over time were observed. Conclusions. The incidence rate and prevalence of OCD in treatment-seeking GP patients are at least 3 times lower than estimates known from the most conservative epidemiological community studies, suggesting that OCD may be under recognised and under treated.

O914 From cervical cancer screening to cervical dysplasia management – electronic patient record data on over 20000 patients from year 2007 in East-Tallinn Central Hospital

International Journal of Gynecology & ObstetricsVolume 107, Issue S2 p. S353-S354 Free communication (oral) presentations O914 From cervical cancer screening to cervical dysplasia management – electronic patient record data on over 20000 patients from year 2007 in East-Tallinn Central Hospital F. Szirko, F. SzirkoSearch for more papers by this authorL. Tammemae, L. TammemaeSearch for more papers by this authorE. Kaik, E. KaikSearch for more papers by this author F. Szirko, F. SzirkoSearch for more papers by this authorL. Tammemae, L. TammemaeSearch for more papers by this authorE. Kaik, E. KaikSearch for more papers by this author First published: 20 November 2009 https://doi.org/10.1016/S0020-7292(09)61287-XAboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat No abstract is available for this article. Volume107, IssueS2Abstracts of XIX FIGO World Congress of Gynecology and ObstetricsOctober 2009Pages S353-S354 RelatedInformation

Altered thermoregulation in ambulatory schizophrenia patients: A naturalistic study

Background. Schizophrenia patients may exhibit alterations in core/body temperature. Hence, we intended to examine the potential existence of thermoregulatory abnormalities in ambulatory schizophrenia patients. Methods. Anonymous electronic patient record data of the Leumit Health Fund (Israel) were screened for all schizophrenia patients who have no other apparent chronic co-morbidity (mental or non-mental) and had their oral temperature assessed during routine follow-ups (Schiz-rFUs) or for various transitory infectious/inflammatory processes (Schiz-Infect) during the years 1999–2005 (n = 535). The comparison group consisted of a comparable sample (n = 560) of healthy subjects (Control-rFUs and Control-Infect). Results. The sub-group of Schiz-rFUs (n = 216) exhibited significantly lower mean oral temperature compared to the matched group of Control-rFUs (n = 140) (36.72±0.54 vs. 36.94±0.64°C, respectively; P<0.05). There was no significant difference in mean oral temperatures between the Schiz-Infect (n = 319) and the Control-Infect (n = 420) (37.32±0.92 vs. 37.28±0.98°C, respectively; NS). Conclusions. Ambulatory schizophrenia patients without a concomitant infectious/inflammatory process exhibit altered thermoregulation manifested by a substantial (about 0.2°C) and significantly lower oral temperature compared to healthy comparison subjects as well as a potential exaggerated increase in oral temperature during transitory infectious/inflammatory processes. The relevance of these phenomena to the pathophysiology of schizophrenia as well as the potential immune-mediated pathologies in schizophrenia merit further investigation.

Electronic Patient Record data as proxy of GPs' thoughts.

Data currently available in primary care Electronic Patient Records (EPR) can potentially be used to study quality of care. In this paper we investigate to which extend these data can reflect GPs' "thoughts" that are an important issue when considering GPs' practice and quality improvement cycle. Within the Resoprim project, we mainly used the consolidated data of three software systems, 26 practices, 1,554 hypertensive patients and 1,977 contacts. Extracted data from the EPR were: some diagnoses, some drugs, referral events, marital status, some parameters (smoking status, height, weight, blood pressure). As "gold standard" of GPs' thoughts we used an electronic questionnaire at the end of each contact. Measures of missing and incoherent values were used to assess our "gold standard". Sensitivity, positive predictive values, correctness and global completeness were used to measure the quality of the automatic extracted data (our proxy). For the "gold standard", the global percentage of missing values is 1.88% and of incoherent values is 3.92%. For most of the practices, the PPV or the correctness of automatic extracted drugs and automatic extracted parameters is high (>95%). The PPV of automatic extracted diagnoses is variable (42.1% to 94.9%). The sensitivity of automatic extracted diagnoses and drugs is lower than 67%. For most of the practices the sensitivity of automatic extracted parameters (excl. smoking status) is higher than 95%. The global completeness of height and weight is lower than 76%. Referrals are badly recorded or extracted. Currently in Belgium, without additional investigations, databases built on data extracted from EPRs can hardly be considered as good proxies of what is thought or known by the GPs. To use them as proxies, we should at least develop tools such as electronic questionnaires to calibrate them. As priority, we suggest an improvement of the extraction procedure design, of the current software interfaces and of the quality control of the extraction modules in order to improve respectively the extracted drugs sensitivity, the global completeness of extracted parameters and the PPV of extracted diagnoses. Training GPs could also be helpful.

A Computer Scientist's Reactions to NPfIT

This paper contains a set of personal views relating to NHS Connecting for Health's National Programme for IT (NPfIT), and in particular its Care Records Service, written from the point of view of a computer scientist, not a medical informatics expert. The principal points made are as follows: Centralisation: Pulling lots of data together (for individual patients and then for large patient populations) harms safety and privacy – it is one byproduct of excessive use of identification when in fact all that is usually needed is authentication. Large centralized data storage facilities can be useful for reliability, but risk exchanging lots of small failures for a lesser number of much larger failures. A much more decentralised approach to electronic patient record (EPR) data and its storage should be investigated. Evolutionary acquisition: Specifying, implementing, deploying and evaluating a sequence of ever more complete IT systems is the best way of ending up with well-accepted and well-trusted systems – especially when this process is controlled by the stakeholders who are most directly involved, rather than by some distant central bureaucracy. Thus authority as well as responsibility should be left with hospital and general practitioner trusts to acquire IT systems that suit their environments and priorities – subject to adherence to minimal interoperability constraints – and to use centralized services (e.g., for system support and back-up) as if and when they choose. Socio-technical issues: Ill-chosen imposed medical IT systems impede patient care, are resisted, result in lots of accidental faults, and lose user support and trust. All these points are attested to by rigorous studies involving expertise from the social sciences (psychology, ethnography, etc.) as well as by technical (medical and computer) experts – much more attention needs to be paid to such studies, and more such studies encouraged. Constructive reviews: A constructive expert review, working closely with Connecting for Health, could be very helpful, but should be evidently independent and open and thus essentially different in nature to past and current inquiries. A review of this nature could not just recommend appropriate changes of plan, and speed progress. It could also contribute to the vital task of helping to restore the trust and confidence of the public and the media in the programme and in the government officials involved.

The incidence of complex regional pain syndrome: A population-based study

The complex regional pain syndrome (CRPS) is a painful disorder that can occur in an extremity after any type of injury, or even spontaneously. Data on the incidence of CRPS are scarce and mostly hospital based. Therefore the size of the problem and its burden on health care and society are unknown. The objective of the present study was to estimate the incidence of CRPS in the general population. A retrospective cohort study was conducted during 1996–2005 in the Integrated Primary Care Information (IPCI) project, a general practice research database with electronic patient record data from 600,000 patients throughout the Netherlands. Potential CRPS cases were identified by a sensitive search algorithm including synonyms and abbreviations for CRPS. Subsequently, cases were validated by electronic record review, supplemented with original specialist letters and information from an enquiry of general practitioners. The estimated overall incidence rate of CRPS was 26.2 per 100,000 person years (95% CI: 23.0–29.7). Females were affected at least three times more often than males (ratio: 3.4). The highest incidence occurred in females in the age category of 61–70 years. The upper extremity was affected more frequently than the lower extremity and a fracture was the most common precipitating event (44%). The observed incidence rate of CRPS is more as four times higher than the incidence rate observed in the only other population-based study, performed in Olmsted County, USA. Postmenopausal woman appeared to be at the highest risk for the development of CRPS.

Web-based electronic patient records for collaborative medical applications

We developed a web-based system to interactively display electronic patient records (EPR), such as DICOM images, graphics, and structure reports and therapy records, for intranet and internet collaborative medical applications. This system has three major components, a C/S (client/server) architecture for EPR data acquisition and authoring, and a Web B/S architecture for data delivering. The Web viewer of this system integrates multi-media display modules and remote control module together to provide interactive EPR display and manipulation functions for collaborative applications. We have successfully used this system two times to provide teleconsultation for severe acute respiratory syndrome (SARS) patients in Shanghai Infection Hospital and Xinhua Hospital. During the consultation, both the physicians in infection control area and the experts outside the control area could use this system interactively to manipulate and navigate the EPR objects of the SARS patients to facilitate a more precise diagnosis. This paper gives a new approach to create and manage image-based EPR from actual patient records, and also presents a novel method to use Web technology and DICOM standard to build an open architecture for collaborative medical applications. The system can be used for both intranet and internet medical applications such as tele-diagnosis, teleconsultation, and distant learning.

Design and implementation considerations for a personalized patient education system in burn care

Patient education is a significant factor in the provision of health care services, contributing to improved disease management and health care outcomes. Personalization has been suggested as a means for increasing patient education effectiveness and computer-based approaches have been explored as a possible means of achieving this goal. The success and capabilities of the resulting applications have been restricted by the absence of a direct link to patient data and the reliance on locally produced written material, which is expensive to produce, update and tailor. In our research project STructured Evaluated Personalized Patient Support (STEPPS), we are investigating the potential of a novel strategy for personalized or tailored patient education, based on the integration of electronic patient record data and material derived from online health information resources. In this paper we present an overview of the pertinent technical issues and the way we have addressed them in the context of our development work in the domain of burn care. Further, we discuss how the choices made in the design of the system interrelate with the considerations for its implementation in health care practice settings.

The integration of medical images with the electronic patient record and their web-based distribution 1

Medical images are currently created digitally and stored in the radiology department’s picture archiving and communication system. Reports are usually stored in the electronic patient record of other information systems, such as the radiology information system (RIS) and the hospital information system (HIS). But high-quality services can only be provided if electronic patient record data is integrated with digital images in picture archiving and communication systems. Clinicians should be able to access both systems’ data in an integrated and consistent way as part of their regular working environment, whether HIS or RIS. Also, this system should allow for teleconferencing with other users, eg, for consultation with a specialist in the radiology department. This article describes a web-based solution that integrates the digital images of picture archiving and communication systems with electronic patient record/HIS/RIS data and has built-in teleconferencing functionality. This integration has been successfully tested using three different commercial RIS and HIS products.

Security of health care records in Belgium: Application in a University Hospital

Access rules to electronic patient records (EPR) have been issued by the Belgian Council of Physicians. Access to identifiable data of the EPR is restricted to anyone responsible for diagnosis, treatment and continuity of care of the patient. By delegation, associated personnel, like secretaries, can also be authorised to have access. A new perspective is given by the availability in 2003 of a national identification card allowing electronic signature of patients. It could not only authorise but also forbid some accesses. A law in 2002 gives right to patients to access to their own record. Health personnel can also be identified by cards but the system is not yet implemented. In the meantime, local measures have been made. We describe practical solutions that have been taken as priorities in a University Hospital. It was felt more important to allow access to lifesaving EPR data than to restrict its access by too strictly theoretical rules. A pilot study (S3 project) is also in progress for interinstitutional communication in Belgium, using the unique identification number of the patient and a “third server”.

Discriminant Patterns of Consistency in Health Data Dictionaries: Implications for Evidence-Based Medicine

An increasing amount of the medical information in health provider organizations is derived from electronic patient records (EPRs). As such, the quality of information maintained by healthcare organizations becomes a key component of the healthcare delivery process, with a corresponding need for consistent information collection and management methods. The objective of this study was to examine and compare adoption rates of EPR data reliability within clinical support systems, identifying regional variation across the USA. In a nationwide study of all accredited US health information managers, reported levels of data dictionary adoption in electronic patient records were examined. Results show that the majority of health information managers have widely disparate policies and procedures related to data consistency. Despite increasing reliance on consistency of data for comparative purposes, most organizations have not exhibited regular adoption of data dictionaries to uniformly define system-based information within their organizations. This occurs in an era where government-mandated data standards require greater uniformity of medical information. Utilization of EPR data in evidence-based medical decision making must be undertaken with caution, especially when outcomes or other decision support data are compared across states or regions.

Web-based distribution of radiological images from PACS to EPR

Medical images are currently created digitally and stored in the radiology department's picture archiving and communication system (PACS). Reports are usually stored in the electronic patient record (EPR) of other information systems, such as the radiology information system (RIS) or the hospital information system (HIS). But high-quality service can only be provided if the EPR data is integrated with the PACS digital images. The clinician should be able to access both systems' data in an integrated and consistent way as part of the regular working environment, the HIS or the RIS. Additionally, this system should allow for teleconferences with other users, e.g., for consultations with a specialist in the radiology department. This paper describes a web-based solution that integrates the digital images of the PACS, the EPR/HIS/RIS data and a built-in teleconferencing functionality. The integration has been successfully tested with three different commercial RIS and HIS products.

Systematic review of scope and quality of electronic patient record data in primary care

Objective To systematically review measures of data quality in electronic patient records (EPRs) in primary care.Design Systematic review of English language publications, 1980-2001.Data sources Bibliographic searches of medical databases, specialist...

Data quality probes—exploiting and improving the quality of electronic patient record data and patient care

Increasing reliance is being placed on electronic medical records to support clinical care and achieve improved quality standards. In order for clinical information systems (CIS) to deliver excellence the data within it needs to be complete, consistent and accurate. This capture of data is critical but forms only part of the procedure in delivering quality health care during the clinician-patient encounter. A number of processes are involved in this encounter, each of which has to be performed flawlessly to deliver a perfect outcome. This paper outlines a method of assessing the quality of these processes involved in healthcare provision and data quality within a CIS. It proposes the principle of Data Quality Probes (DQP) to assess the performance of the whole encounter system. The main feature of this is the generation of a query which clinical knowledge predicts should not retrieve any cases in a system performing flawlessly. Any cases retrieved (which fail the DQP) indicate an error in either data quality or clinical judgment. This approach is applied practically within the paradigm of a UK family practice testing the hypothesis that a series DQPs can provide a valuable method for monitoring both the data accuracy of a CIS and the provision of quality patient care.

Assessment of the possibility to classify patients according to cholesterol guideline screening criteria using routinely recorded electronic patient record data

Computerised decision support systems (CDSS) can be categorised as either being inquisitive or non-inquisitive. The non-inquisitive system uses routinely entered electronic patient data, to generate patient specific feedback based on guidelines. The Dutch College of General Practitioners' (DCGP) cholesterol guideline classifies patients into risk groups, eligible for screening. The availability of sufficient routinely recorded electronic patient data to classify patients according to the DCGP cholesterol guideline is unknown. To assess whether it is possible to classify patients according to the screening criteria of the DCGP cholesterol guideline, using data routinely recorded by general practitioners. We analysed the DCGP cholesterol guideline to identify selection criteria for screening. These selection criteria were subsequently used to create a cohort of patient records eligible for screening in the Integrated Primary Care Information (IPCI) Database. We calculated incidence and prevalence of risk factors and selected patient records for active management according to the identified screening selection criteria. 145866 valid patient records were selected for classification. In the retrieved records 9741 (13.6%) males and 5756 (7.8%) females were identified for active management according to the selection criteria of the DCGP cholesterol guideline. The classification of patients into risk groups, eligible for screening, according to the criteria of the DCGP cholesterol guideline using routinely recorded electronic patient data is feasible. Care should be taken when using only diagnostic codes, as it gives higher than expected incidence and prevalence of risk factors. Based on these findings we are currently building Cholgate, a non-inquisitive decision support system for cholesterol management.

Variation in EPR clinical support systems: data, decisions and disparity

In an evidence-based healthcare environment, the quality of electronic patient record (EPR) data maintained on clinical support systems becomes a critical factor in the decision making process, with a corresponding need for error management through automated data quality assurance practices. This paper reports on a study, the objective of which was to examine and compare adoption rates of EPR data reliability within clinical support systems, and associated regional variation across the US. Data were collected by mail and telephone surveys nationwide across the US. The participants in the surveys were all accredited US health information managers. The outcome measure was reported levels of automated data edit methodologies in electronic patient records. Findings suggest inconsistent adoption of automated data quality management in electronic patient records. This low overall adoption of automated edits was accompanied by significant variation across geographic regions and key organizational variables. The authors concluded that utilization of EPR data in evidence-based medical decision making must be undertaken with caution, especially when outcomes or other decision support data are compared across states or regions.

Towards personalized Internet health information: the STEPPS architecture.

STEPPS (STructured Evaluated Personalized Patient Support) proposes a strategy for integration of electronic patient records with Internet health-related content and its consequent use in personalized information retrieval for patient education. The application domain is the post-discharge support of burn patients in the Netherlands. We developed an electronic patient record interface for structured data collection in burn care. The system's thesaurus was projected to UMLS terms and the corresponding codes were incorporated in our software. A list of topics central to burn patient education was identified and a collection of related Web pages was compiled using meta-search software (Copernic2001Pro). The HTML pages were filed into catalogues by the Collexis indexing-matching software, using the UMLS Metathesaurus as indexing vocabulary. The bilingual (English and Dutch) structured data interface is currently used to create a database of retrospective patient data. Each patient's profile, i.e. set of characteristics employed to personalize information retrieval, can be automatically extracted. We have assembled a collection of more than 2500 Internet pages containing relevant information for burn patients. When patient data is available, the Collexis matching engine will accept the patient's profile as input and retrieve the most relevant HTML documents available in the catalogues. We have addressed some basic issues around the technical feasibility of linking electronic patient record data to online content. Although the functionality of STEPPS is not yet optimal, it contributes to the efforts towards improved relevance of information retrieval. Electronic patient record applications in conjunction with Internet resources can give a significant boost to the availability of tailored health education material. In this context, quality assurance of online health information is an indispensable element.

Tendo Achillis bursitis (winter heel).

<h3>Introduction</h3> Following the publication of the GOSH digital strategy, electronic patient record (EPR) data has been recognised as an ever-growing cornerstone of the hospital infrastructure. Because EPR data remains the property of the patient and the NHS, access to it is rightly controlled by stringent governance processes which introduce a hurdle for researchers requiring the use of these data. In-short, these restrictions can make it challenging to run short-term clinical research projects, to publish reproducible results without compromising patient anonymity, and for researchers to train on complex EPR data before applying to access it. <h3>Methods</h3> The GOSH DRE team have worked with collaborators in UCL and NHS Digital to develop generative statistical and deep learning (AI) models that learn the structure and statistical properties of EPR data. These models have the capability to generate synthetic EPR data without reproducing individual patient records. To facilitate this work, de-identified EPR data from all patients treated at GOSH between January 2016 and January 2019 were extracted from the DRE data lake and modelled using the PyTorch, TensorFlow, and Scikit learn Python libraries. <h3>Results</h3> A range of privacy-preserving models were developed on admissions, ward movements, demographics, drug administrations, laboratory tests, vital signs, and microbiological isolate datasets. On inspection, each model was capable of generating realistic EPR data. A method for automatically evaluating each model’s output was developed through the use of stringent statistical similarity and disclosure control metrics that are automatically applied to generated data. <h3>Discussion</h3> Integrating the models into a programmatic interface has enabled the production of realistic, consistent EPR records that are not related to any single patient, yet generate high-quality data in the same format as the clinical data made available through the GOSH Digital Research Environment (DRE). This will enable future data-driven developments in GOSH, DRIVE, and the DRE.