scholarly journals Constructing Epidemiologic Cohorts from Electronic Health Record Data

2021 ◽  
Vol 18 (24) ◽  
pp. 13193
Author(s):  
Brent A. Williams
Keyword(s):  
The United States ◽  
Epidemiologic Studies ◽  
Healthcare Organizations ◽  
Electronic Health Record Data ◽  
Patient Health ◽  
Electronic Health ◽  
Cost Efficient ◽  
Future Outcomes ◽  
The Cost ◽  
Study Participants

In the United States, electronic health records (EHR) are increasingly being incorporated into healthcare organizations to document patient health and services rendered. EHRs serve as a vast repository of demographic, diagnostic, procedural, therapeutic, and laboratory test data generated during the routine provision of health care. The appeal of using EHR data for epidemiologic research is clear: EHRs generate large datasets on real-world patient populations in an easily retrievable form permitting the cost-efficient execution of epidemiologic studies on a wide array of topics. Constructing epidemiologic cohorts from EHR data involves as a defining feature the development of data machinery, which transforms raw EHR data into an epidemiologic dataset from which appropriate inference can be drawn. Though data machinery includes many features, the current report focuses on three aspects of machinery development of high salience to EHR-based epidemiology: (1) selecting study participants; (2) defining “baseline” and assembly of baseline characteristics; and (3) follow-up for future outcomes. For each, the defining features and unique challenges with respect to EHR-based epidemiology are discussed. An ongoing example illustrates key points. EHR-based epidemiology will become more prominent as EHR data sources continue to proliferate. Epidemiologists must continue to improve the methods of EHR-based epidemiology given the relevance of EHRs in today’s healthcare ecosystem.

scholarly journals Utilization of Deep Learning for Subphenotype Identification in Sepsis-Associated Acute Kidney Injury

2020 ◽  
Vol 15 (11) ◽  
pp. 1557-1565 ◽  
Author(s):  
Kumardeep Chaudhary ◽  
Akhil Vaid ◽  
Áine Duffy ◽  
Ishan Paranjpe ◽  
Suraj Jaladanki ◽  
...  
Keyword(s):  
Intensive Care ◽  
Deep Learning ◽  
Medical Information ◽  
Tertiary Care ◽  
Vital Signs ◽  
Kidney Injury ◽  
The United States ◽  
Care Hospital ◽  
Electronic Health Record Data ◽  
Electronic Health

Background and objectivesSepsis-associated AKI is a heterogeneous clinical entity. We aimed to agnostically identify sepsis-associated AKI subphenotypes using deep learning on routinely collected data in electronic health records.Design, setting, participants, & measurementsWe used the Medical Information Mart for Intensive Care III database, which consists of electronic health record data from intensive care units in a tertiary care hospital in the United States. We included patients ≥18 years with sepsis who developed AKI within 48 hours of intensive care unit admission. We then used deep learning to utilize all available vital signs, laboratory measurements, and comorbidities to identify subphenotypes. Outcomes were mortality 28 days after AKI and dialysis requirement.ResultsWe identified 4001 patients with sepsis-associated AKI. We utilized 2546 combined features for K-means clustering, identifying three subphenotypes. Subphenotype 1 had 1443 patients, and subphenotype 2 had 1898 patients, whereas subphenotype 3 had 660 patients. Subphenotype 1 had the lowest proportion of liver disease and lowest Simplified Acute Physiology Score II scores compared with subphenotypes 2 and 3. The proportions of patients with CKD were similar between subphenotypes 1 and 3 (15%) but highest in subphenotype 2 (21%). Subphenotype 1 had lower median bilirubin levels, aspartate aminotransferase, and alanine aminotransferase compared with subphenotypes 2 and 3. Patients in subphenotype 1 also had lower median lactate, lactate dehydrogenase, and white blood cell count than patients in subphenotypes 2 and 3. Subphenotype 1 also had lower creatinine and BUN than subphenotypes 2 and 3. Dialysis requirement was lowest in subphenotype 1 (4% versus 7% [subphenotype 2] versus 26% [subphenotype 3]). The mortality 28 days after AKI was lowest in subphenotype 1 (23% versus 35% [subphenotype 2] versus 49% [subphenotype 3]). After adjustment, the adjusted odds ratio for mortality for subphenotype 3, with subphenotype 1 as a reference, was 1.9 (95% confidence interval, 1.5 to 2.4).ConclusionsUtilizing routinely collected laboratory variables, vital signs, and comorbidities, we were able to identify three distinct subphenotypes of sepsis-associated AKI with differing outcomes.

National Cost Efficiency of Supported Employees With Intellectual Disabilities: 2002 to 2007

2010 ◽  
Vol 115 (1) ◽  
pp. 19-29 ◽  
Author(s):  
Robert E. Cimera
Keyword(s):  
United States ◽  
Intellectual Disabilities ◽  
Vocational Rehabilitation ◽  
Cost Efficiency ◽  
The United States ◽  
Considerable Variability ◽  
United State ◽  
Cost Efficient ◽  
The Cost

Abstract The cost efficiency of supported employees with intellectual disabilities who were served by vocational rehabilitation agencies throughout the United State from 2002 to 2007 was explored. Findings indicate that, on average, supported employees with intellectual disabilities were cost-efficient from the taxpayers' perspective regardless of whether they had secondary disabilities. In addition, no changes in cost efficiency were found during the period investigated. The data, however, did demonstrate considerable variability in cost efficiency throughout the United States and its territories.

scholarly journals Predicting mortality among patients with liver cirrhosis in electronic health records with machine learning

PLoS ONE ◽  
2021 ◽  
Vol 16 (8) ◽  
pp. e0256428
Author(s):  
Aixia Guo ◽  
Nikhilesh R. Mazumder ◽  
Daniela P. Ladner ◽  
Randi E. Foraker
Keyword(s):  
Machine Learning ◽  
Liver Cirrhosis ◽  
Deep Learning ◽  
Risk Prediction ◽  
The United States ◽  
Meld Score ◽  
Receiver Operating Curve ◽  
Categorical Variables ◽  
Electronic Health Record Data ◽  
Electronic Health

Objective Liver cirrhosis is a leading cause of death and effects millions of people in the United States. Early mortality prediction among patients with cirrhosis might give healthcare providers more opportunity to effectively treat the condition. We hypothesized that laboratory test results and other related diagnoses would be associated with mortality in this population. Our another assumption was that a deep learning model could outperform the current Model for End Stage Liver disease (MELD) score in predicting mortality. Materials and methods We utilized electronic health record data from 34,575 patients with a diagnosis of cirrhosis from a large medical center to study associations with mortality. Three time-windows of mortality (365 days, 180 days and 90 days) and two cases with different number of variables (all 41 available variables and 4 variables in MELD-NA) were studied. Missing values were imputed using multiple imputation for continuous variables and mode for categorical variables. Deep learning and machine learning algorithms, i.e., deep neural networks (DNN), random forest (RF) and logistic regression (LR) were employed to study the associations between baseline features such as laboratory measurements and diagnoses for each time window by 5-fold cross validation method. Metrics such as area under the receiver operating curve (AUC), overall accuracy, sensitivity, and specificity were used to evaluate models. Results Performance of models comprising all variables outperformed those with 4 MELD-NA variables for all prediction cases and the DNN model outperformed the LR and RF models. For example, the DNN model achieved an AUC of 0.88, 0.86, and 0.85 for 90, 180, and 365-day mortality respectively as compared to the MELD score, which resulted in corresponding AUCs of 0.81, 0.79, and 0.76 for the same instances. The DNN and LR models had a significantly better f1 score compared to MELD at all time points examined. Conclusion Other variables such as alkaline phosphatase, alanine aminotransferase, and hemoglobin were also top informative features besides the 4 MELD-Na variables. Machine learning and deep learning models outperformed the current standard of risk prediction among patients with cirrhosis. Advanced informatics techniques showed promise for risk prediction in patients with cirrhosis.

PACS Failure Mode and Effects

2010 ◽  
pp. 193-206
Author(s):  
Carrison K.S. Tong ◽  
Eric T.T. Wong
Keyword(s):  
Failure Mode ◽  
Medical Errors ◽  
Medical Error ◽  
Low Cost ◽  
The United States ◽  
Drug Dose ◽  
Professional Organizations ◽  
Healthcare Organizations ◽  
Leadership Standards ◽  
The Cost

There are some medical errors for which preventability is rarely questioned. These include medical errors such as wrong site surgery, wrong procedure, wrong patient operations (Seiden & Barach, 2006; Michaels et al., 2007; Lee et al., 2007), wrong drug/dose/duration (Pugh et al., 2005) or incompatible organ transplantation (Cook et al., 2007). Less preventable medical errors include judgment type errors such as case studies reported in journals, where one or more experts review the treatment decisions of a clinician and conclude that the clinician’s judgment was incorrect (Lukela et al., 2005). Many healthcare managers first heard about Failure Mode and Effects Analysis FMEA when Joint Commission on Accreditation of Healthcare Organizations (JCAHO) released its Leadership Standards and Elements of Performance Guidelines in July 2002 (JCAHO, 2002). The purpose of performing an FMEA for JCAHO was to identify where and when possible system failures could occur and to prevent those problems before they happen. If a particular, failure could not be prevented, then the goal would be to prevent the issue from affecting healthcare organizations in the accreditation process. FMEA is a tool that when performed adequately, can reduce the risk of preventable medical errors. Hospitals in the United States that are accredited by JCAHO are required to perform at least one FMEA each year. The main output of FMEA is a series of mitigations, each of which is some process change implemented to reduce the risk of error. Because resources are limited, implementing all mitigations is not possible so the challenge is to find the set of mitigations that provides the highest reduction in risk for the least cost. Hence, preventability may be viewed in terms of the cost and effectiveness of mitigation. A low-cost and effective mitigation is associated with a highly preventable medical error, whereas a high-cost and or less effective mitigation is associated with a less preventable medical error. Currently AAPM TG 100 (2007) is reviewing reports from previous task groups and from several professional organizations. This group is also reviewing ISO guidelines in an effort develop a suitable general QA approach that “balances patient safety and quality versus resources commonly available and strikes a good balance between prescriptiveness and flexibility.” The TG 100 initiative identifies three industrial engineering–based tools as potential components of a QA management system in radiation therapy and FMEA is one of them.

scholarly journals Use of electronic health records to support a public health response to the COVID-19 pandemic in the United States: a perspective from 15 academic medical centers

2020 ◽  
Author(s):  
Subha Madhavan ◽  
Lisa Bastarache ◽  
Jeffrey S Brown ◽  
Atul J Butte ◽  
David A Dorr ◽  
...  
Keyword(s):  
Public Health ◽  
Clinical Care ◽  
Population Level ◽  
The United States ◽  
Socioeconomic Indicators ◽  
Healthcare Organizations ◽  
Public Health Response ◽  
Public Health Decision ◽  
Health Decision ◽  
Electronic Health

Abstract Our goal is to summarize the collective experience of 15 organizations in dealing with uncoordinated efforts that result in unnecessary delays in understanding, predicting, preparing for, containing, and mitigating the COVID-19 pandemic in the US. Response efforts involve the collection and analysis of data corresponding to healthcare organizations, public health departments, socioeconomic indicators, as well as additional signals collected directly from individuals and communities. We focused on electronic health record (EHR) data, since EHRs can be leveraged and scaled to improve clinical care, research, and to inform public health decision-making. We outline the current challenges in the data ecosystem and the technology infrastructure that are relevant to COVID-19, as witnessed in our 15 institutions. The infrastructure includes registries and clinical data networks to support population-level analyses. We propose a specific set of strategic next steps to increase interoperability, overall organization, and efficiencies

scholarly journals Broadening the reach of the FDA Sentinel system: A roadmap for integrating electronic health record data in a causal analysis framework

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Rishi J. Desai ◽  
Michael E. Matheny ◽  
Kevin Johnson ◽  
Keith Marsolo ◽  
Lesley H. Curtis ◽  
...  
Keyword(s):  
Large Scale ◽  
Data Science ◽  
Claims Data ◽  
The United States ◽  
Medical Product ◽  
Product Safety ◽  
Electronic Health Record Data ◽  
Data Infrastructure ◽  
Safety Surveillance ◽  
Electronic Health

AbstractThe Sentinel System is a major component of the United States Food and Drug Administration’s (FDA) approach to active medical product safety surveillance. While Sentinel has historically relied on large quantities of health insurance claims data, leveraging longitudinal electronic health records (EHRs) that contain more detailed clinical information, as structured and unstructured features, may address some of the current gaps in capabilities. We identify key challenges when using EHR data to investigate medical product safety in a scalable and accelerated way, outline potential solutions, and describe the Sentinel Innovation Center’s initiatives to put solutions into practice by expanding and strengthening the existing system with a query-ready, large-scale data infrastructure of linked EHR and claims data. We describe our initiatives in four strategic priority areas: (1) data infrastructure, (2) feature engineering, (3) causal inference, and (4) detection analytics, with the goal of incorporating emerging data science innovations to maximize the utility of EHR data for medical product safety surveillance.

scholarly journals Health and Economics of Lifestyle Medicine Strategies

2020 ◽  
Vol 14 (3) ◽  
pp. 274-277
Author(s):  
Dee W. Edington ◽  
Wayne N. Burton ◽  
Alyssa B. Schultz
Keyword(s):  
Health Care ◽  
Health Care Cost ◽  
Health Management ◽  
Employee Performance ◽  
The United States ◽  
Lifestyle Medicine ◽  
Patient Health ◽  
Evidenced Based ◽  
The Cost ◽  
The University

The cost of medical care in the United States is increasing at an unsustainable rate. The lifestyle medicine (LM) approach is essential to influence the root causes of the growing chronic disease burden. LM addresses health risk factors in primary, secondary, and tertiary prevention of developing disease rather than limiting resources and medical expenditures on acute care and reacting to illness, injury, and disease. Employers have much to gain financially from such an approach due to their status as the payer of health care costs for their employees, and as the recipient of productivity gains among their employees. This article discusses LM programs delivered at the worksite, including important findings from the University of Michigan Health Management Research Center. Examples of evidenced-based population LM interventions are summarized for physical activity, weight management, and nutrition programs that address chronic diseases such as cardiovascular disease, cancer, and diabetes mellitus. These approaches have the potential to reduce health care cost trends, increase employee performance/productivity, and improve patient health outcomes.

scholarly journals Consumer-Mediated Data Exchange for Research: Current State of US Law, Technology, and Trust (Preprint)

2018 ◽  
Author(s):  
Yiscah Bracha ◽  
Jacqueline Bagwell ◽  
Robert Furberg ◽  
Jonathan S Wald
Keyword(s):  
Data Acquisition ◽  
Data Exchange ◽  
Strong Support ◽  
The United States ◽  
Institutional Review Boards ◽  
Electronic Health Record Data ◽  
Current State ◽  
Institutional Review ◽  
Research Organizations ◽  
Study Participants

UNSTRUCTURED A compendium of US laws and regulations offers increasingly strong support for the concept that researchers can acquire the electronic health record data that their studies need directly from the study participants using technologies and processes called consumer-mediated data exchange. This data acquisition method is particularly valuable for studies that need complete longitudinal electronic records for all their study participants who individually and collectively receive care from multiple providers in the United States. In such studies, it is logistically infeasible for the researcher to receive necessary data directly from each provider, including providers who may not have the capability, capacity, or interest in supporting research. This paper is a tutorial to inform the researcher who faces these data acquisition challenges about the opportunities offered by consumer-mediated data exchange. It outlines 2 approaches and reviews the current state of provider- and consumer-facing technologies that are necessary to support each approach. For one approach, the technology is developed and estimated to be widely available but could raise trust concerns among research organizations or their institutional review boards because of the current state of US law applicable to consumer-facing technologies. For the other approach, which does not elicit the same trust concerns, the necessary technology is emerging and a pilot is underway. After reading this paper, the researcher who has not been following these developments should have a good understanding of the legal, regulatory, technology, and trust issues surrounding consumer-mediated data exchange for research, with an awareness of what is potentially possible now, what is not possible now, and what could change in the future. The researcher interested in trying consumer-mediated data exchange will also be able to anticipate and respond to an anticipated barrier: the trust concerns that their own organizations could raise.

scholarly journals Lymphocyte levels following COVID-19 vaccine BNT162b2

2021 ◽  
Author(s):  
Oren Miron ◽  
Nadav Davidovitch
Keyword(s):  
Clinical Trial ◽  
The United States ◽  
Rank Test ◽  
Blood Levels ◽  
Electronic Health Record Data ◽  
Medical Reason ◽  
Dose Administration ◽  
Blood Lymphocytes ◽  
Before And After ◽  
Electronic Health

Introduction: The BNT162b2 vaccine has been shown to be effective in reducing the incidence, severity and mortality of Coronavirus Disease 2019 (COVID-19). The clinical trial report of BNT162b2 suggested that the mechanism of BNT162b2 includes lymphocytes migration from the blood to the lymph nodes, and that it relates to the clinical trial finding of decreased blood lymphocytes in the 3 days following dose-1 of BNT162b2. A decrease in blood lymphocytes was also shown in the second day after dose-1 BNT162b2 in another study, and in studies of BNT162b1 and other mRNA vaccines. The BNT162b2 clinical trial also found that lymphocytes were normal in the 6-8 days following dose-1 and dose-2, but it did not test lymphocytes in the 3 days following dose-2. Our study aims to estimate the lymphocytes in the 3 days following dose-2 using existing electronic health records, to help improve the understanding of the BNT162b2 dose-2 mechanism.Methods: We extracted values of lymphocyte blood tests and BNT162b2 immunization from Electronic Health Record data including diagnosis, procedures, labs, vitals, medications and histories sourced from participating members of the Healthjump network, which is situated in the United States. Absolute lymphocytes were calculated as 10^3/mm3 (thousands per cubic millimeter), and vaccines were extracted from December 10th 2020 to September 30th 2021 based on the CXV code 208, the CPT code 91300, and the vaccine description from a database of 900 thousand vaccinations. We included BNT162b2 first dose administration and second dose BNT162b2 administrations that were done 21 days after a BNT162b2 first dose administration to resemble the clinical trial protocol (excluding dose-2 before or after day 21). We calculated the median lymphocyte values in each of the 14 days before and after the vaccination to determine its lowest median value, and we also compared the lymphocyte values in the 3 days before and after the administration using Wilcox rank test (significance at p<0.05). Results: We extracted 13,329 records, with a mean age of 59 years. The median lymphocyte in the 14 days before the 1st and 2nd dose was 1.85 (10^3/mm3). For the first dose, the lowest median value was 1.75 (10^3/mm3) and it was reached 3 days after administration, while for the second dose the lowest median value was 1.35 (10^3/mm3) and it was reached 2 days after administration. The lymphocyte value in the 3 days after dose-2 administration were significantly higher than those in the 3 days before the vaccination (p<0.01). Discussion: Our analysis suggests that lymphocyte blood levels have a temporary decrease in the 3 days following the second dose of BNT162b, which resembles the decrease reported after the first dose in the clinical trial. This could relate to the high increase in antibodies that is often found following the second dose. The main limitation of the study is that the lymphocyte tests were done for a medical reason, such as an annual exam or to diagnose a disease, unlike the clinical trial that tested each participant. Future vaccine studies could examine blood lymphocytes in the 3 days following the second dose to verify this finding and further examine the dose-2 mechanism and its effect.

Sign in / Sign up

Export Citation Format

Share Document