Use Of Electronic Medical Record Data Research Articles

Use of Sequential Hot-Deck Imputation for Missing Health Care Systems Data for Population Health Research.

Electronic medical record (EMR) data present many opportunities for population health research. The use of EMR data for population risk models can be impeded by the high proportion of missingness in key patient variables. Common approaches like complete case analysis and multiple imputation may not be appropriate for some population health initiatives that require a single, complete analytic data set. In this study, we demonstrate a sequential hot-deck imputation (HDI) procedure to address missingness in a set of cardiometabolic measures in an EMR data set. We assessed the performance of sequential HDI within the individual variables and a commonly used composite risk score. A data set of cardiometabolic measures based on EMR data from 2 large urban hospitals was used to create a benchmark data set with simulated missingness. Sequential HDI was applied, and the resulting data were used to calculate atherosclerotic cardiovascular disease risk scores. The performance of the imputation approach was assessed using a set of metrics to evaluate the distribution and validity of the imputed data. Of the 567,841 patients, 65% had at least 1 missing cardiometabolic measure. Sequential HDI resulted in the distribution of variables and risk scores that reflected those in the simulated data while retaining correlation. When stratified by age and sex, risk scores were plausible and captured patterns expected in the general population. The use of sequential HDI was shown to be a suitable approach to multivariate missingness in EMR data. Sequential HDI could benefit population health research by providing a straightforward, computationally nonintensive approach to missing EMR data that results in a single analytic data set.

Use of EMR data to validate performance of machine learning prognostic models for patients with cancer.

512 Background: Machine learning models that predict survival time for patients with cancer can be useful in the clinic. Validating their performance in deployment is important but challenging, because the only timely source of follow-up/death data is the electronic medical record (EMR), which is known to under-capture deaths resulting in informative censoring. We examined whether validation using EMR data can distinguish between low- and high-quality models, by using gold-standard Cancer Registry data to calculate the true model performance level. Methods: This was a retrospective study of 3417 patients diagnosed with metastatic cancer from 2008-2014, who were either diagnosed with cancer or received their initial treatment at our institution. We used regularized logistic regression on 525 demographic and diagnostic features to predict whether each patient survived for >1 year following diagnosis. We trained the model on high-quality Cancer Registry data from 2197 patients diagnosed between 2008 and 2012 and tested it on two sets of patients. The first set consisted of 1184 patients who had Cancer Registry data evaluable for the 1 year survival endpoint (not lost to follow-up prior to 1 year). Of these, 335 were marked as lost to follow-up prior to 1 year in the EMR dataset, leaving 849 patients evaluable for the endpoint using the EMR data. Results: The model using all available features yielded an AUC of 0.760 when validating with Cancer Registry data (n=1184), and an AUC of 0.771 with EMR data (n=849). When using fewer features, observed AUC dropped for both validation data sources. When using Cancer Registry validation data the model was found to be well calibrated, but with EMR validation data the model incorrectly appeared to systematically underpredict survival. Conclusions: EMR data was useful for validation of model discrimination, but not calibration.

The current and potential uses of Electronic Medical Record (EMR) data for primary health care performance measurement in the Canadian context: a qualitative analysis

BackgroundElectronic Medical Records (EMRs) are a rich data source to measure and improve quality of care. As Canadian primary health care (PHC) EMRs mature, there is increasing potential use of EMR data for performance measurement. This study identifies and describes current uses of EMR data for performance measurement and considerations to further its potential in the Canadian context.MethodsWe applied a qualitative case study design and descriptive assessment in three phases, consulting multiple data sources including scientific and grey literature, system leaders (n = 41), and clinician/researchers (n = 20). Phases included a multimethod approach to identify initiatives using EMR data for performance measurement across Canadian jurisdictions; in-depth review of current initiatives identified from a healthcare performance intelligence lens; and triangulation and thematic analysis across data sources to explore considerations for advancing performance measurement uses of EMR data in the Canadian context.ResultsSix initiatives of EMR data use for performance measurement were identified: one multi-jurisdictional; five jurisdiction-specific in the provinces of British Columbia, Manitoba and Ontario. EMR data uses were predominately for micro-level PHC physician and team performance improvement, with some use for meso-level organization/network-wide improvement. Indicator sets varied in number, though shared emphasis on chronic disease management and prevention/screening and to a lesser extent medication management. Key considerations for governing, resourcing and implementing EMR data for performance measurement were identified.ConclusionsThe extent of EMR data use for performance measurement varies across Canada. To further its potential, pan-Canadian data and privacy standards, performance intelligence competencies and renewed core PHC indicators should be prioritized. Experiences across countries, coupled with increasing momentum for performance measurement using real-world data, should be leveraged to avoid unnecessarily slow progress in Canada and abroad.

Study on the semi-supervised learning-based patient similarity from heterogeneous electronic medical records

BackgroundA new learning-based patient similarity measurement was proposed to measure patients’ similarity for heterogeneous electronic medical records (EMRs) data.MethodsWe first calculated feature-level similarities according to the features’ attributes. A domain expert provided patient similarity scores of 30 randomly selected patients. These similarity scores and feature-level similarities for 30 patients comprised the labeled sample set, which was used for the semi-supervised learning algorithm to learn the patient-level similarities for all patients. Then we used the k-nearest neighbor (kNN) classifier to predict four liver conditions. The predictive performances were compared in four different situations. We also compared the performances between personalized kNN models and other machine learning models. We assessed the predictive performances by the area under the receiver operating characteristic curve (AUC), F1-score, and cross-entropy (CE) loss.ResultsAs the size of the random training samples increased, the kNN models using the learned patient similarity to select near neighbors consistently outperformed those using the Euclidean distance to select near neighbors (all P values < 0.001). The kNN models using the learned patient similarity to identify the top k nearest neighbors from the random training samples also had a higher best-performance (AUC: 0.95 vs. 0.89, F1-score: 0.84 vs. 0.67, and CE loss: 1.22 vs. 1.82) than those using the Euclidean distance. As the size of the similar training samples increased, which composed the most similar samples determined by the learned patient similarity, the performance of kNN models using the simple Euclidean distance to select the near neighbors degraded gradually. When exchanging the role of the Euclidean distance, and the learned patient similarity in selecting the near neighbors and similar training samples, the performance of the kNN models gradually increased. These two kinds of kNN models had the same best-performance of AUC 0.95, F1-score 0.84, and CE loss 1.22. Among the four reference models, the highest AUC and F1-score were 0.94 and 0.80, separately, which were both lower than those for the simple and similarity-based kNN models.ConclusionsThis learning-based method opened an opportunity for similarity measurement based on heterogeneous EMR data and supported the secondary use of EMR data.

Prospect of Artificial Intelligence Based on Electronic Medical Record.

With the advent of the big data era, the interest of the international community is focusing on increasing the utilization of medical big data. Many hospitals are attempting to increase the efficiency of their operations and patient management by adopting artificial intelligence (AI) technology that enables the use of electronic medical record (EMR) data. EMR includes information about a patient's health history, such as diagnoses, medicines, tests, allergies, immunizations, treatment plans, personalized medical care, and improvement of medical quality and safety. EMR data can also be used for AI-based new drug development. In particular, it is effective to develop AI that can predict the occurrence of specific diseases or provide individualized customized treatments by classifying the individualized characteristics of patients. In order to improve performance of artificial intelligence research using EMR data, standardization and refinement of data are essential. In addition, since EMR data deal with sensitive personal information of patients, it is also vital to protect the patient's privacy. There are already various supports for the use of EMR data in the Korean government, and researchers are encouraged to be proactive.

Noninvasive Ventilation Is Interrupted Frequently and Mostly Used at Night in the Pediatric Intensive Care Unit

Noninvasive ventilation (NIV) is commonly used to support children with respiratory failure, but detailed patterns of real-world use are lacking. The aim of our study was to describe use patterns of NIV via electronic medical record (EMR) data. We performed a retrospective electronic chart review in a tertiary care pediatric ICU in the United States. Subjects admitted to the pediatric ICU from 2014 to 2017 who were mechanically ventilated were included in the study. The median number of discrete device episodes, defined as a time on support without interruption, was 20 (interquartile range [IQR] 8-49) per subject. The median duration of bi-level positive airway pressure (BPAP) support prior to interruption was 6.3 h (IQR 2.4-10.4); the median duration of CPAP was 6 h (IQR 2.1-10.4). Interruptions to BPAP had a median duration of 6.3 h (IQR 2-15.5); interruptions to CPAP had a median duration of 8.6 h (IQR 2.2-16.8). Use of NIV followed a diurnal pattern, with 44% of BPAP and 42% of CPAP subjects initiating support between 7:00 pm and midnight, and 49% of BPAP and 46% of CPAP subjects stopping support between 5:00 am and 10:00 am. NIV was frequently interrupted, and initiation and discontinuation of NIV follows a diurnal pattern. Use of EMR data collected for routine clinical care allowed the analysis of granular details of typical use patterns. Understanding NIV use patterns may be particularly important to understanding the burden of pediatric ICU bed utilization for nocturnal NIV. To our knowledge, this is the first study to examine in detail the use of pediatric NIV and to define diurnal use and frequent interruptions to support.

Assessment of Traceability Implementation of a Cross-Institutional Secure Data Collection System Based on Distributed Standardized EMR Storage.

This paper presents results concerning the traceability of a secure data collection infrastructure based on the Bloom Filter involving standardized electronic medical record (EMR) storage. The objective of this infrastructure is to facilitate rapid secondary use of exported EMR data in cross-patient or cross-institutional analyses based on the Standardized Structured Medical Information eXchange (SS-MIX), Japan's domestic standard for EMR export. Secondary use of EMR data should be based on the principle of patient consent. Therefore, securing the traceability of patient EMR data is important for achieving reliable data collection systems for secondary use. Blockchain is a strong candidate, and we evaluated its performance using several implementations. As shown by the evaluation, it is difficult to realize the method of recording the query history of secondary use on the blockchain. We thus propose a method for recording index information of distributed log data.

Exploiting the concept level feature for enhanced name entity recognition in Chinese EMRs

The accumulation and explosive growth of the electronic medical records (EMRs) make the name entity recognition (NER) technologies become critical for the meaningful use of EMR data and then the practice of evidence-based medicine. The dominate NER approaches use the distributed representation of the words and characters to build deep learning-based NER models. However, for the task of biomedical named entity recognition, there are a large amount of complicated medical terminologies that are composed of multiple words. Splitting these terminologies to learn the word and character embeddings might cause semantic ambiguities. In this paper, we treat each medical terminology as a concept and propose a concept-enhanced named entity recognition model (CNER), where the features from three different granularities (i.e., concept, word, and character) are combined together for bio-NER. The extensive experiments are conducted on two real-world corpora: fully labeled corpus and partially labeled corpus. CNER achieves the highest F1 score (fully labeled corpus: precision = 88.23, recall = 88.29, and F1 = 88.26; partially labeled corpus: precision = 87.03, recall = 88.19, and F1 = 87.61) by outperforming the baseline CW-BLSTM-CRF approach for 0.58% and 1.15% respectively, which demonstrates the effectiveness of the proposed approach.

The Chang Gung Research Database-A multi-institutional electronic medical records database for real-world epidemiological studies in Taiwan.

The Chang Gung Research Database (CGRD), the largest multi-institutional electronic medical records (EMR) collection in Taiwan, provides good access for researchers to efficiently use the standardized patient-level data. This study evaluates the capacity and representativeness of the CGRD to promote secondary use of EMR data for clinical research with more accurate estimates. The National Health Insurance Research Database (NHIRD) which covers over 99.9% of the Taiwanese population served as the comparator in this study. We compare the data components of the CGRD with the NHIRD, including records for health care facilities, patients, diagnoses, drugs, and procedures. Using the chi-square test, we compared the distributions of age categories and sex of patients, and the rates of their health conditions between NHIRD and CGRD based on the year 2015. The CGRD contains more clinical information such as pathological and laboratory results than the NHIRD. The CGRD includes 6.1% of outpatients and 10.2% of hospitalized patients from the NHIRD. We found the CGRD includes more elderly outpatients (23.5% vs 12.5%) and pediatric inpatients (19.7% vs 14.4%) compared with the NHIRD. We found patients' sex distributions were similar between CGRD and NHIRD, but coverage rates of severe conditions, such as cancer, were higher than other health conditions in CGRD. The CGRD could serve as the basis for accurate estimates in medical studies. However, researchers should pay special attention to selection biases since patients' characteristics from CGRD differ from those of the national database.

Pharmacist use of the electronic medical record to identify adults at risk for anaphylaxis without epinephrine for self-administration

ObjectivesTo describe an innovative pharmacist-led approach, with the use of electronic medical record (EMR) data, to identify patients at risk of anaphylaxis in need of epinephrine auto-injector (EAI) for self-administration. SettingAn urban free health care center for an uninsured indigent adult population in Pittsburgh, PA. Practice innovationIn this pilot service, patients with allergy history fields containing the words “anaphylaxis,” “nut,” “bee,” or “shellfish” were screened for inclusion. Practice descriptionPatients were identified with the use of a report generated by the EMR vendor and were contacted via telephone by a clinical pharmacist. Using the patient-reported clinical history related to anaphylaxis and allergies, the pharmacist assessed the need for EAI, counseled the patient, and provided physician referral when appropriate. EvaluationThe service was evaluated with the use of a cross-sectional study of patients with electronic records at the health center during the time period studied. Data obtained from patient interviews were used to classify patients who were candidates for EAI and to assess prescribing and access to EAI. Demographic data and outcomes related to the availability of non-expired EAI were collected and analyzed using descriptive statistics. ResultsNinety-five patients were identified as potential candidates for the service, and 20 patients were able to be contacted via telephone. Fourteen participants provided consent for their data to be used in the program evaluation. A treatment gap likely existed for 11 of the 14 participants (79%) owing to history of anaphylaxis (57%) or history of systemic reaction (29%). The most common indication for EAI was anaphylaxis in response to bee stings. ConclusionThe findings from this project demonstrate that pharmacist use of EMR data, coupled with patient interview, may be an effective means for identifying treatment gaps in the long-term management of anaphylaxis and has the potential to ultimately improve management of anaphylaxis in patients within the community.

Personalized mortality prediction driven by electronic medical data and a patient similarity metric.

BackgroundClinical outcome prediction normally employs static, one-size-fits-all models that perform well for the average patient but are sub-optimal for individual patients with unique characteristics. In the era of digital healthcare, it is feasible to dynamically personalize decision support by identifying and analyzing similar past patients, in a way that is analogous to personalized product recommendation in e-commerce. Our objectives were: 1) to prove that analyzing only similar patients leads to better outcome prediction performance than analyzing all available patients, and 2) to characterize the trade-off between training data size and the degree of similarity between the training data and the index patient for whom prediction is to be made.Methods and FindingsWe deployed a cosine-similarity-based patient similarity metric (PSM) to an intensive care unit (ICU) database to identify patients that are most similar to each patient and subsequently to custom-build 30-day mortality prediction models. Rich clinical and administrative data from the first day in the ICU from 17,152 adult ICU admissions were analyzed. The results confirmed that using data from only a small subset of most similar patients for training improves predictive performance in comparison with using data from all available patients. The results also showed that when too few similar patients are used for training, predictive performance degrades due to the effects of small sample sizes. Our PSM-based approach outperformed well-known ICU severity of illness scores. Although the improved prediction performance is achieved at the cost of increased computational burden, Big Data technologies can help realize personalized data-driven decision support at the point of care.ConclusionsThe present study provides crucial empirical evidence for the promising potential of personalized data-driven decision support systems. With the increasing adoption of electronic medical record (EMR) systems, our novel medical data analytics contributes to meaningful use of EMR data.

Electronic medical record data to identify variables associated with a fibromyalgia diagnosis: importance of health care resource utilization.

BackgroundDiagnosis of fibromyalgia (FM) is often challenging. Identifying factors associated with an FM diagnosis may guide health care providers in implementing appropriate diagnostic and management strategies.MethodsThis retrospective study used the de-identified Humedica electronic medical record (EMR) database to identify variables associated with an FM diagnosis. Cases (n=4,296) were subjects ≥18 years old with ≥2 International Classification of Diseases, Ninth Revision (ICD-9) codes for FM (729.1) ≥30 days apart during 2012, associated with an integrated delivery network, with ≥1 encounter with a health care provider in 2011 and 2012. Controls without FM (no-FM; n=583,665) did not have the ICD-9 codes for FM. Demographic, clinical, and health care resource utilization variables were extracted from structured EMR data. Univariate analysis identified variables showing significant differences between the cohorts based on odds ratios (ORs).ResultsConsistent with FM epidemiology, FM subjects were predominantly female (78.7% vs 64.5%; P<0.0001) and slightly older (mean age 53.3 vs 52.7 years; P=0.0318). Relative to the no-FM cohort, the FM cohort was characterized by a higher prevalence of nearly all evaluated comorbidities; the ORs suggested a higher likelihood of an FM diagnosis (P<0.0001), especially for musculoskeletal and neuropathic pain conditions (OR 3.1 for each condition). Variables potentially associated with an FM diagnosis included higher levels of use of specific health care resources including emergency-room visits, outpatient visits, hospitalizations, and medications. Units used per subject for emergency-room visits, outpatient visits, hospitalizations, and medications were also significantly higher in the FM cohort (P<0.0001), confirming resource utilization as an important variable associated with an FM diagnosis.ConclusionSignificant differences between the FM and no-FM cohorts were observed for nearly all the demographic, clinical, and health care resource variables, suggesting an association with FM diagnosis. These results also support use of EMR data for identifying variables associated with FM, which may help in the diagnosis and management of this condition.

Improving the power of genetic association tests with imperfect phenotype derived from electronic medical records.

To reduce costs and improve clinical relevance of genetic studies, there has been increasing interest in performing such studies in hospital-based cohorts by linking phenotypes extracted from electronic medical records (EMRs) to genotypes assessed in routinely collected medical samples. A fundamental difficulty in implementing such studies is extracting accurate information about disease outcomes and important clinical covariates from large numbers of EMRs. Recently, numerous algorithms have been developed to infer phenotypes by combining information from multiple structured and unstructured variables extracted from EMRs. Although these algorithms are quite accurate, they typically do not provide perfect classification due to the difficulty in inferring meaning from the text. Some algorithms can produce for each patient a probability that the patient is a disease case. This probability can be thresholded to define case-control status, and this estimated case-control status has been used to replicate known genetic associations in EMR-based studies. However, using the estimated disease status in place of true disease status results in outcome misclassification, which can diminish test power and bias odds ratio estimates. We propose to instead directly model the algorithm-derived probability of being a case. We demonstrate how our approach improves test power and effect estimation in simulation studies, and we describe its performance in a study of rheumatoid arthritis. Our work provides an easily implemented solution to a major practical challenge that arises in the use of EMR data, which can facilitate the use of EMR infrastructure for more powerful, cost-effective, and diverse genetic studies.

Application of electronic medical record data for health outcomes research: a review of recent literature

Electronic medical records (EMRs) have become a common source of data for outcomes research. This review discusses trends in EMR data use for outcomes research as well as strengths and limitations, and likely future developments to help optimize value and use of EMR data for outcomes research. EMR-based studies reporting treatment outcomes published between 2007 and 2012 were predominantly from the USA and Europe. There has been a substantial increase in the number of EMR-based outcomes studies published from 2007–2008 (n = 28) to 2010–2011 (n = 55). Many studies evaluated biometric and laboratory test outcomes in common chronic conditions. However, researchers are expanding the scope of evaluated diseases and outcomes using advanced techniques, such as natural language processing and linking EMRs to other patient-level data to overcome issues with missing data or data that cannot be accessed using standard queries. These advances will help to expand the scope and sophistication of outcomes research in the coming years.

A social network of hospital acquired infection built from electronic medical record data

Social networks have been used in the study of outbreaks of infectious diseases, including in small group settings such as individual hospitals. Collecting the data needed to create such networks, however, can be time consuming, costly, and error prone. We sought to create a social network of hospital inpatients using electronic medical record (EMR) data already collected for other purposes, for use in simulating outbreaks of nosocomial infections. We used the EMR data warehouse of a tertiary academic hospital to model contact among inpatients. Patient-to-patient contact due to shared rooms was inferred from admission-discharge-transfer data, and contact with healthcare workers was inferred from clinical documents. Contacts were used to generate a social network, which was then used to conduct probabilistic simulations of nosocomial outbreaks of methicillin-resistant Staphylococcus aureus and influenza. Simulations of infection transmission across the network reflected the staffing and patient flow practices of the hospital. Simulations modeling patient isolation, increased hand hygiene, and staff vaccination showed a decrease in the spread of infection. We developed a method of generating a social network of hospital inpatients from EMR data. This method allows the derivation of networks that reflect the local hospital environment, obviate the need for simulated or manually collected data, and can be updated in near real time. Inpatient social networks represent a novel secondary use of EMR data, and can be used to simulate nosocomial infections. Future work should focus on prospective validation of the simulations, and adapting such networks to other tasks.

Feasibility of studying brain morphology in major depressive disorder with structural magnetic resonance imaging and clinical data from the electronic medical record: A pilot study

For certain research questions related to long-term outcomes or to rare disorders, designing prospective studies is impractical or prohibitively expensive. Such studies could instead utilize clinical and magnetic resonance imaging data (MRI) collected as part of routine clinical care, stored in the electronic medical record (EMR). Using major depressive disorder (MDD) as a disease model, we examined the feasibility of studying brain morphology and associations with remission using clinical and MRI data exclusively drawn from the EMR. Advanced automated tools were used to select MDD patients and controls from the EMR who had brain MRI data, but no diagnosed brain pathology. MDD patients were further assessed for remission status by review of clinical charts. Twenty MDD patients (eight full-remitters, six partial-remitters, and six non-remitters), and 15 healthy control subjects met all study criteria for advanced morphometric analyses. Compared to controls, MDD patients had significantly smaller right rostral-anterior cingulate volume, and level of non-remission was associated with smaller left hippocampus and left rostral-middle frontal gyrus volume. The use of EMR data for psychiatric research may provide a timely and cost-effective approach with the potential to generate large study samples reflective of the real population with the illness studied.

Using your electronic medical record for research: a primer for avoiding pitfalls

In Canada, use of electronic medical records (EMRs) among primary health care (PHC) providers is relatively low. However, it appears that EMRs will eventually become more ubiquitous in PHC. This represents an important development in the use of health care information technology as well as a potential new source of PHC data for research. However, care in the use of EMR data is required. Four years ago, researchers at the Centre for Studies in Family Medicine, The University of Western Ontario created an EMR-based research project, called Deliver Primary Health Care Information. Implementing this project led us to two conclusions about using PHC EMR data for research: first, additional time is required for providers to undertake EMR training and to standardize the way data are entered into the EMR and second, EMRs are designed for clinical care, not research. Based on these experiences, we offer our thoughts about how EMRs may, nonetheless, be used for research. Family physician researchers who intend to use EMR data to answer timely questions relevant to practice should evaluate the possible impact of the four questions raised by this paper: (i) why are EMR data different?; (ii) how do you extract data from an EMR?; (iii) where are the data stored? and (iv) what is the data quality? In addition, consideration needs to be given to the complexity of the research question since this can have an impact on how easily issues of using EMR data for research can be overcome.