Dataset Information

Novel Method to Flag Cardiac Implantable Device Infections by Integrating Text Mining With Structured Data in the Veterans Health Administration's Electronic Medical Record.

ABSTRACT: Importance:Health care-associated infections (HAIs) are preventable, harmful, and costly; however, few resources are dedicated to infection surveillance of nonsurgical procedures, particularly cardiovascular implantable electronic device (CIED) procedures. Objective:To develop a method that includes text mining of electronic clinical notes to reliably and efficiently measure HAIs for CIED procedures. Design, Setting, and Participants:In this multicenter, national cohort study using electronic medical record data for patients undergoing CIED procedures in Veterans Health Administration (VA) facilities for fiscal years (FYs) 2016 and 2017, an algorithm to flag cases with a true CIED-related infection based on structured (eg, microbiology orders, vital signs) and free text diagnostic and therapeutic data (eg, procedure notes, discharge summaries, microbiology results) was developed and validated. Procedure data were divided into development and validation data sets. Criterion validity (ie, positive predictive validity [PPV], sensitivity, and specificity) was assessed via criterion-standard manual medical record review. Exposures:CIED procedure. Main Outcomes and Measures:The concordance between medical record review and the study algorithm with respect to the presence or absence of a CIED infection. CIED infection in the algorithm included 90-day mortality, congestive heart failure and nonmetastatic tumor comorbidities, CIED or surgical site infection International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes, antibiotic treatment of Staphylococci, a microbiology test of a cardiac specimen, and text documentation of infection in specific clinical notes (eg, cardiology, infectious diseases, inpatient discharge summaries). Results:The algorithm sample consisted of 19?212 CIED procedures; 15?077 patients (78.5%) were White individuals, 1487 (15.5%) were African American; 18?766 (97.7%) were men. The mean (SD) age in our sample was 71.8 (10.6) years. The infection detection threshold of predicted probability was set to greater than 0.10 and the algorithm flagged 276 of 9606 (2.9%) cases in the development data set (9606 procedures); PPV in this group was 41.4% (95% CI, 31.6%-51.8%). In the validation set (9606 procedures), at predicted probability 0.10 or more the algorithm PPV was 43.5% (95% CI, 37.1%-50.2%), and overall sensitivity and specificity were 94.4% (95% CI, 88.2%-97.9%) and 48.8% (95% CI, 42.6%-55.1%), respectively. Conclusions and Relevance:The findings of this study suggest that the method of combining structured and text data in VA electronic medical records can be used to expand infection surveillance beyond traditional boundaries to include outpatient and procedural areas.

SUBMITTER: Mull HJ

PROVIDER: S-EPMC7506515 | biostudies-literature | 2020 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Novel Method to Flag Cardiac Implantable Device Infections by Integrating Text Mining With Structured Data in the Veterans Health Administration's Electronic Medical Record.

Mull Hillary J HJ Stolzmann Kelly L KL Shin Marlena H MH Kalver Emily E Schweizer Marin L ML Branch-Elliman Westyn W

JAMA network open 20200901 9

<h4>Importance</h4>Health care-associated infections (HAIs) are preventable, harmful, and costly; however, few resources are dedicated to infection surveillance of nonsurgical procedures, particularly cardiovascular implantable electronic device (CIED) procedures.<h4>Objective</h4>To develop a method that includes text mining of electronic clinical notes to reliably and efficiently measure HAIs for CIED procedures.<h4>Design, setting, and participants</h4>In this multicenter, national cohort stu ...[more]

PMID: 32955571

Similar Datasets

Project description:BACKGROUND:Antimicrobial prophylaxis is an evidence-proven strategy for reducing procedure-related infections; however, measuring this key quality metric typically requires manual review, due to the way antimicrobial prophylaxis is documented in the electronic medical record (EMR). Our objective was to electronically measure compliance with antimicrobial prophylaxis using both structured and unstructured data from the Veterans Health Administration (VA) EMR. We developed this methodology for cardiac device implantation procedures. METHODS:With clinician input and review of clinical guidelines, we developed a list of antimicrobial names recommended for the prevention of cardiac device infection. We trained the algorithm using existing fiscal year (FY) 2008-15 data from the VA Clinical Assessment Reporting and Tracking-Electrophysiology (CART-EP), which contains manually determined information about antimicrobial prophylaxis. We merged CART-EP data with EMR data and programmed statistical software to flag an antimicrobial orders or drug fills from structured data fields in the EMR and hits on text string searches of antimicrobial names documented in clinician's notes. We iteratively tested combinations of these data elements to optimize an algorithm to accurately classify antimicrobial use. The final algorithm was validated in a national cohort of VA cardiac device procedures from FY2016-2017. Discordant cases underwent expert manual review to identify reasons for algorithm misclassification. RESULTS:The CART-EP dataset included 2102 procedures at 38 VA facilities with manually identified antimicrobial prophylaxis in 2056 cases (97.8%). The final algorithm combining structured EMR fields and text note search results correctly classified 2048 of the CART-EP cases (97.4%). In the validation sample, the algorithm measured compliance with antimicrobial prophylaxis in 16,606 of 18,903 cardiac device procedures (87.8%). Misclassification was due to EMR documentation issues, such as antimicrobial prophylaxis documented only in hand-written clinician notes in a format that cannot be electronically searched. CONCLUSIONS:We developed a methodology with high accuracy to measure guideline concordant use of antimicrobial prophylaxis before cardiac device procedures using data fields present in modern EMRs. This method can replace manual review in quality measurement in the VA and other healthcare systems with EMRs; further, this method could be adapted to measure compliance in other procedural areas where antimicrobial prophylaxis is recommended.

Project description:BackgroundThe electronic medical record (EMR) contains a rich source of information that could be harnessed for epidemic surveillance. We asked if structured EMR data could be coupled with computerized processing of free-text clinical entries to enhance detection of acute respiratory infections (ARI).MethodologyA manual review of EMR records related to 15,377 outpatient visits uncovered 280 reference cases of ARI. We used logistic regression with backward elimination to determine which among candidate structured EMR parameters (diagnostic codes, vital signs and orders for tests, imaging and medications) contributed to the detection of those reference cases. We also developed a computerized free-text search to identify clinical notes documenting at least two non-negated ARI symptoms. We then used heuristics to build case-detection algorithms that best combined the retained structured EMR parameters with the results of the text analysis.Principal findingsAn adjusted grouping of diagnostic codes identified reference ARI patients with a sensitivity of 79%, a specificity of 96% and a positive predictive value (PPV) of 32%. Of the 21 additional structured clinical parameters considered, two contributed significantly to ARI detection: new prescriptions for cough remedies and elevations in body temperature to at least 38°C. Together with the diagnostic codes, these parameters increased detection sensitivity to 87%, but specificity and PPV declined to 95% and 25%, respectively. Adding text analysis increased sensitivity to 99%, but PPV dropped further to 14%. Algorithms that required satisfying both a query of structured EMR parameters as well as text analysis disclosed PPVs of 52-68% and retained sensitivities of 69-73%.ConclusionStructured EMR parameters and free-text analyses can be combined into algorithms that can detect ARI cases with new levels of sensitivity or precision. These results highlight potential paths by which repurposed EMR information could facilitate the discovery of epidemics before they cause mass casualties.

Project description:UnlabelledOver one-third of patients treated in the Veterans Health Administration (VHA) are obese. VHA introduced the MOVE! Weight Management Program for Veterans in 2006 to provide comprehensive weight management services. An evolving, periodic evaluation using the RE-AIM framework (reach, effectiveness, adoption, implementation, and maintenance) has been conducted to gauge success and opportunities for improvement. Key metrics were identified in each RE-AIM dimension. Data were compiled over fiscal years (FY) 2006 through 2010 from a variety of sources including VHA administrative and clinical databases, electronic medical record reviews, and an annual, structured VHA facility self-report. REACH: Screening for obesity and offering weight management treatment to eligible patients increased from 66% to 95% over the past 3 years. MOVE! is currently provided at every VHA hospital facility and at over one-half of VHA community-based outpatient clinics. The percent of eligible patients who participate in at least one weight management visit has doubled since implementation began but has stabilized at 10 to 12%.EffectivenessAbout 18.6% of the 31,854 patients with available weight data who participated in at least two treatment visits between Jul 1, 2008 and Sep 30, 2009 had at least a 5% body weight loss by 6 months as did almost one-third of those who participated in more intense and sustained treatment. By contrast, only 12.5% of a comparison group of patients matched on age, gender, body mass index (BMI) class, and comorbidity status who were not treated with MOVE! had at least a 5% body weight loss. ADOPTION: The median full-time staff equivalent providing weight management services at each facility has increased over time and was 1.76 in FY 2010.ImplementationStaff from multiple disciplines typically provide MOVE!-related care although not all disciplines are involved with providing care at every facility. Group-based treatment has become increasingly utilized, and in FY 2010 it represented 72% of all MOVE!-related visits. Intensity of treatment has increased from an average of 3.6 visits per patient per year in FY 2007 to 4.6 in FY 2010, but more than half of patients have two visits or less. Almost all facilities now report the consistent use of key evidence-based behavioral strategies with patients. MAINTENANCE: While participation in MOVE! by patients continues to grow each year, facility self-reported program staffing and space/equipment challenges are potential barriers to long-term program maintenance. Evidence-based weight management treatment can be delivered at VHA medical centers and community-based outpatient clinics, but the REACH remains limited after several years of implementation. Intense and sustained treatment with MOVE! results in a modest positive impact on short-term weight loss outcomes, but a relatively small proportion of patients engage in this level of care. Increasing reach, improving effectiveness of care, and keeping patients engaged in treatment are areas for future policy, practice, and research.

Project description:The rapidly growing amount of publicly available information from biomedical research is readily accessible on the Internet, providing a powerful resource for predictive biomolecular modeling. The accumulated data on experimentally determined structures transformed structure prediction of proteins and protein complexes. Instead of exploring the enormous search space, predictive tools can simply proceed to the solution based on similarity to the existing, previously determined structures. A similar major paradigm shift is emerging due to the rapidly expanding amount of information, other than experimentally determined structures, which still can be used as constraints in biomolecular structure prediction. Automated text mining has been widely used in recreating protein interaction networks, as well as in detecting small ligand binding sites on protein structures. Combining and expanding these two well-developed areas of research, we applied the text mining to structural modeling of protein-protein complexes (protein docking). Protein docking can be significantly improved when constraints on the docking mode are available. We developed a procedure that retrieves published abstracts on a specific protein-protein interaction and extracts information relevant to docking. The procedure was assessed on protein complexes from Dockground (http://dockground.compbio.ku.edu). The results show that correct information on binding residues can be extracted for about half of the complexes. The amount of irrelevant information was reduced by conceptual analysis of a subset of the retrieved abstracts, based on the bag-of-words (features) approach. Support Vector Machine models were trained and validated on the subset. The remaining abstracts were filtered by the best-performing models, which decreased the irrelevant information for ~ 25% complexes in the dataset. The extracted constraints were incorporated in the docking protocol and tested on the Dockground unbound benchmark set, significantly increasing the docking success rate.

Dataset Information

Novel Method to Flag Cardiac Implantable Device Infections by Integrating Text Mining With Structured Data in the Veterans Health Administration's Electronic Medical Record.

Publications

Novel Method to Flag Cardiac Implantable Device Infections by Integrating Text Mining With Structured Data in the Veterans Health Administration's Electronic Medical Record.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets