Search

Introduction We examined mutational testing of the epidermal growth factor gene ( EGFR ) and erlotinib treatment among veterans diagnosed with non–small-cell lung cancer in the United States Department of Veterans Affairs (VA). Our objectives were to identify the prevalence of clinically actionable EGFR mutations, to determine whether testing and treatment were guideline concordant, to evaluate the impact of testing and treatment on survival, and to estimate the rate of testing. Patients and Methods Test results were linked to electronic health records from VA Corporate Data Warehouse and the VA Central Cancer Registry. We analyzed patient demographic and clinical characteristics, prevalence of EGFR mutations, and timing of EGFR mutational testing and erlotinib treatment based on pharmacy records. Overall survival was assessed by Kaplan-Meier analysis. Results Among 973 patients tested at 70 VA medical centers between 2011 and 2013, 64 (7%) had sensitizing EGFR mutations, 694 (71%) were EGFR wild type, and 168 (17%) had clinically insignificant polymorphisms or variants of unknown significance. Results were not documented in 47 tests (5%). Erlotinib administration was in agreement with test results in 843 cases (87%). Conclusion Veterans have a much lower rate of sensitizing EGFR mutations than the reported average of 10% to 15%, which correlates with a high rate of smoking among veterans. This may partially explain clinicians' reluctance to prescribe EGFR testing, which results in underutilization. Although test results appear to have influenced erlotinib treatment decisions, we documented a substantial number of cases where treatment was not applied in accordance with clinical guidelines, potentially resulting in worse outcomes and unnecessary cost.

Background Current clinical guidelines recommend epidermal growth factor receptor (EGFR) mutational testing in patients with metastatic non-small cell lung cancer (NSCLC) to predict the benefit of the tyrosine kinase inhibitor erlotinib as first-line treatment. Proteomic (VeriStrat) testing is recommended for patients with EGFR negative or unknown status when erlotinib is being considered. Departure from this clinical algorithm can increase costs and may result in worse outcomes. We examined EGFR and proteomic testing among patients with NSCLC within the Department of Veterans Affairs (VA). We explored adherence to guidelines and the impact of test results on treatment decisions and cost of care. Methods Proteomic and EGFR test results from 2013 to 2015 were merged with VA electronic health records and pharmacy data. Chart reviews were conducted. Cases were categorized based on the appropriateness of testing and treatment. Results Of the 69 patients with NSCLC who underwent proteomic testing, 33 (48%) were EGFR-negative and 36 (52%) did not have documented EGFR status. We analyzed 138 clinical decisions surrounding EGFR/proteomic testing and erlotinib treatment. Most decisions (105, or 76%) were concordant with clinical practice guidelines. However, for 24 (17%) decisions documentation of testing or justification of treatment was inadequate, and 9 (7%) decisions represented clear departures from guidelines. Conclusion EGFR testing, the least expensive clinical intervention analyzed in this study, was significantly underutilized or undocumented. The records of more than half of the patients lacked information on EGFR status. Our analysis illustrated several clinical scenarios where the timing of proteomic testing and erlotinib diverged from the recommended algorithm, resulting in excessive costs of care with no documented improvements in health outcomes.

Multiple sclerosis (MS), a central nervous system disease in which nerve signals are disrupted by scarring and demyelination, is classified into phenotypes depending on the patterns of cognitive or physical impairment progression: relapsing-remitting MS (RRMS), primary-progressive MS (PPMS), secondary-progressive MS (SPMS), or progressive-relapsing MS (PRMS). The phenotype is important in managing the disease and determining appropriate treatment. The ICD-9-CM code 340.0 is uninformative about MS phenotype, which increases the difficulty of studying the effects of phenotype on disease.To identify MS phenotype using natural language processing (NLP) techniques on progress notes and other clinical text in the electronic medical record (EMR).Patients with at least 2 ICD-9-CM codes for MS (340.0) from 1999 through 2010 were identified from nationwide EMR data in the Department of Veterans Affairs. Clinical experts were interviewed for possible keywords and phrases denoting MS phenotype in order to develop a data dictionary for NLP. For each patient, NLP was used to search EMR clinical notes, since the first MS diagnosis date for these keywords and phrases. Presence of phenotype-related keywords and phrases were analyzed in context to remove mentions that were negated (e.g., "not relapsing-remitting") or unrelated to MS (e.g., "RR" meaning "respiratory rate"). One thousand mentions of MS phenotype were validated, and all records of 150 patients were reviewed for missed mentions.There were 7,756 MS patients identified by ICD-9-CM code 340.0. MS phenotype was identified for 2,854 (36.8%) patients, with 1,836 (64.3%) of those having just 1 phenotype mentioned in their EMR clinical notes: 1,118 (39.2%) RRMS, 325 (11.4%) PPMS, 374 (13.1%) SPMS, and 19 (0.7%) PRMS. A total of 747 patients (26.2%) had 2 phenotypes, the most common being 459 patients (16.1%) with RRMS and SPMS. A total of 213 patients (7.5%) had 3 phenotypes, and 58 patients (2.0%) had 4 phenotypes mentioned in their EMR clinical notes. Positive predictive value of phenotype identification was 93.8% with sensitivity of 94.0%.Phenotype was documented for slightly more than one third of MS patients, an important but disappointing finding that sets a limit on studying the effects of phenotype on MS in general. However, for cases where the phenotype was documented, NLP accurately identified the phenotypes. Having multiple phenotypes documented is consistent with disease progression. The most common misidentification was because of ambiguity while clinicians were trying to determine phenotype. This study brings attention to the need for care providers to document MS phenotype more consistently and provides a solution for capturing phenotype from clinical text.This study was funded by Anolinx and F. Hoffman-La Roche. Nelson serves as a consultant for Anolinx. Kamauu is owner of Anolinx, which has received multiple research grants from pharmaceutical and biotechnology companies. LaFleur has received a Novartis grant for ongoing work. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the U.S. government. Study concept and design were contributed by Butler, LaFleur, Kamauu, DuVall, and Nelson. DuVall collected the data, and interpretation was performed by Nelson, DuVall, and Kamauu, along with Butler, LaFleur, and Knippenberg. The manuscript was written primarily by Nelson, along with Knippenberg and assisted by the other authors, and revised by Knippenberg, Nelson, and DuVall, along with the other authors.

Objective To assess a large national sample of bladder cancer pathology reports to determine if they contained the components necessary for clinical decision-making. Methods We examined a random sample of 507 bladder cancer pathology reports from the national Department of Veterans Affairs Corporate Data Warehouse to assess whether each included information on the 4 report components explicitly recommended by the College of American Pathologists' protocol for the examination of such specimens: histology, grade, presence vs absence of muscularis propria in the specimen, and microscopic extent. We then assessed variation in the proportion of reports lacking at least 1 component across Department of Veterans Affairs facilities. Results One hundred eight of 507 reports (21%) lacked at least 1 of the 4 components, with microscopic extent and presence vs absence of muscularis propria in the specimen most commonly missing (each in 11% of reports). There was wide variation across facilities in the proportion of reports lacking at least 1 component, ranging from 0% to 80%. Conclusion One-fifth of bladder cancer pathology reports lack information needed for clinical decision-making. The wide variation in incomplete report rates across facilities implies that some facilities already have implemented best practices assuring complete reporting whereas others have room for improvement. Future work to better understand barriers and facilitators of complete reporting may lead to interventions that improve bladder cancer care.

We investigated type 2 diabetes (T2D) genetic susceptibility in a multi-ethnic meta-analysis of 228,499 cases and 1,178,783 controls in the Million Veteran Program (MVP) and other biobanks. We identified 558 autosomal and 10 X-chromosome T2D-associated variants, of which 286 autosomal and 7 X-chromosome variants were previously unreported. Ancestry-specific analyses identified 25 additional novel T2D-susceptibility variants. Transcriptome-wide association analysis detected 3,568 T2D-associations with T2D-colocalized genetically predicted gene expression of 804 genes in 52 tissues, of which 687 are novel. Fifty-four of these genes are known to interact with FDA-approved drugs and chemical compounds. T2D polygenic risk score was strongly associated with increased the risk of T2D-related retinopathy, and additionally showed evidence for association with chronic kidney disease (CKD), neuropathy, and peripheral artery disease (PAD). We investigated the genetic etiology of T2D-related vascular outcomes in the MVP and observed statistical SNP-T2D interactions at 13 variants, including 3 for coronary heart disease, 1 for PAD, 2 for stroke, 4 for retinopathy, 2 for CKD, and 1 for neuropathy. Our findings may identify potential novel therapeutic targets for T2D and genomic pathways that link T2D and its vascular outcomes.

We investigated type 2 diabetes (T2D) genetic susceptibility via multi-ancestry meta-analysis of 228,499 cases and 1,178,783 controls in the Million Veteran Program (MVP), DIAMANTE, Biobank Japan and other studies. We report 568 associations, including 286 autosomal, 7 X-chromosomal and 25 identified in ancestry-specific analyses that were previously unreported. Transcriptome-wide association analysis detected 3,568 T2D associations with genetically predicted gene expression in 687 novel genes; of these, 54 are known to interact with FDA-approved drugs. A polygenic risk score (PRS) was strongly associated with increased risk of T2D-related retinopathy and modestly associated with chronic kidney disease (CKD), peripheral artery disease (PAD) and neuropathy. We investigated the genetic etiology of T2D-related vascular outcomes in the MVP and observed statistical SNP-T2D interactions at 13 variants, including coronary heart disease (CHD), CKD, PAD and neuropathy. These findings may help to identify potential therapeutic targets for T2D and genomic pathways that link T2D to vascular outcomes.

Background & aimsGiven ongoing challenges in non-invasive non-alcoholic liver disease (NAFLD) diagnosis, we sought to validate an ALT-based NAFLD phenotype using measures readily available in electronic health records (EHRs) and population-based studies by leveraging the clinical and genetic data in the Million Veteran Program (MVP), a multi-ethnic mega-biobank of US Veterans.MethodsMVP participants with alanine aminotransferases (ALT) >40 units/L for men and >30 units/L for women without other causes of liver disease were compared to controls with normal ALT. Genetic variants spanning eight NAFLD risk or ALT-associated loci (LYPLAL1, GCKR, HSD17B13, TRIB1, PPP1R3B, ERLIN1, TM6SF2, PNPLA3) were tested for NAFLD associations with sensitivity analyses adjusting for metabolic risk factors and alcohol consumption. A manual EHR review assessed performance characteristics of the NAFLD phenotype with imaging and biopsy data as gold standards. Genetic associations with advanced fibrosis were explored using FIB4, NAFLD Fibrosis Score and platelet counts.ResultsAmong 322,259 MVP participants, 19% met non-invasive criteria for NAFLD. Trans-ethnic meta-analysis replicated associations with previously reported genetic variants in all but LYPLAL1 and GCKR loci (PConclusionsWe validate a simple, non-invasive ALT-based NAFLD phenotype using EHR data by leveraging previously established NAFLD risk-associated genetic polymorphisms.

Chronic kidney disease (CKD), defined by low estimated glomerular filtration rate (eGFR), contributes to global morbidity and mortality. Here we conduct a transethnic Genome-Wide Association Study of eGFR in 280,722 participants of the Million Veteran Program (MVP), with replication in 765,289 participants from the Chronic Kidney Disease Genetics (CKDGen) Consortium. We identify 82 previously unreported variants, confirm 54 loci, and report interesting findings including association of the sickle cell allele of betaglobin among non-Hispanic blacks. Our transcriptome-wide association study of kidney function in healthy kidney tissue identifies 36 previously unreported and nine known genes, and maps gene expression to renal cell types. In a Phenome-Wide Association Study in 192,868 MVP participants using a weighted genetic score we detect associations with CKD stages and complications and kidney stones. This investigation reinterprets the genetic architecture of kidney function to identify the gene, tissue, and anatomical context of renal homeostasis and the clinical consequences of dysregulation., Persistently low levels of estimated glomerular filtration rate (eGFR) are a biomarker of chronic kidney disease. Here, the authors reinterpret the genetic architecture of kidney function across ancestries, to identify not only genes, but the tissue and anatomical contexts of renal homeostasis.

We demonstrate the utility of concept lexicon expansion and evaluation using enriched samples of patients and documents with sexual orientation as a use case for rare event detection in electronic medical records. Using this approach, we found 7 additional words and 21 misspellings beyond our initial set of five seed words. We can use the expanded vocabulary to further develop a full natural language processing system to identify instances where sexual orientation is documented.

Hospital systems frequently implement quality measures to quantify healthcare processes and patient outcomes. One such measure that has previously been used is the Surgical Care Improvement Project (SCIP) quality measure of perioperative beta blocker continuation, SCIP-Card-2. The SCIP-Card-2 measure requires resource-intensive medical chart abstraction, limiting its application to a small sample of eligible patients. This paper describes a natural language processing (NLP) system for automatic extraction of SCIP-Card-2 quality measures in clinical text notes.

The Department of Defense (DoD) and Department of Veterans Affairs (VA) Infrastructure for Clinical Intelligence (DaVINCI) creates an electronic network between the two United States federal agencies that provides a consolidated view of electronic medical record data for both service members and Veterans. This inter-agency collaboration has created new opportunities for supporting transitions in clinical care, reporting to Congress, and longitudinal research.

In 2015, the VA Informatics and Computing Infrastructure, a resource center of the Department of Veterans Affairs, began to transform parts of its Corporate Data Warehouse (CDW) into the Observational Medical Outcomes Partnership) Common Data Model for use by its research and operations communities. Using the hierarchical relationships within the clinical vocabularies in OMOP we found differences in visits, disease prevalence, and medications prescribed between male and female veterans seen between VA fiscal years 2000-17.

Aim: Describe the characteristics of providers ordering, patients receiving, and clinical impact of a psychotropic pharmacogenetic test on veteran care. Patients & methods: Observational cohort study linking veterans' laboratory results to electronic health record data. Changes in psychotropic medication prescribing were measured as a function of test results. Results: A total of 38 providers tested 181 veterans between 10/6/2014 and 2/1/2018. Prescriptions for medications with severe gene–drug interactions decreased; however, 11 such medications were used after testing. For 43 patients, documentation of the results was missing. Conclusion: Most prescribing decisions were congruent with test results, but in a nontrivial number of cases, prescribers appeared not to act on the results. Poor result documentation impeded the potential of results to inform clinical care.

Background Canadian Cardiovascular Society (CCS) angina severity classification is associated with mortality, myocardial infarction, and coronary revascularization in clinical trial and registry data. The objective of this study was to determine associations between CCS class and all‐cause mortality and healthcare utilization, using natural language processing to extract CCS classifications from clinical notes. Methods and Results In this retrospective cohort study of veterans in the United States with stable angina from January 1, 2006, to December 31, 2013, natural language processing extracted CCS classifications. Veterans with a prior diagnosis of coronary artery disease were excluded. Outcomes included all‐cause mortality (primary), all‐cause and cardiovascular‐specific hospitalizations, coronary revascularization, and 1‐year healthcare costs. Of 299 577 veterans identified, 14 216 (4.7%) had ≥1 CCS classification extracted by natural language processing. The mean age was 66.6±9.8 years, 99% of participants were male, and 81% were white. During a median follow‐up of 3.4 years, all‐cause mortality rates were 4.58, 4.60, 6.22, and 6.83 per 100 person‐years for CCS classes I, II, III, and IV, respectively. Multivariable adjusted hazard ratios for all‐cause mortality comparing CCS II, III, and IV with those in class I were 1.05 (95% CI, 0.95–1.15), 1.33 (95% CI, 1.20–1.47), and 1.48 (95% CI, 1.25–1.76), respectively. The multivariable hazard ratio comparing CCS IV with CCS I was 1.20 (95% CI, 1.09–1.33) for all‐cause hospitalization, 1.25 (95% CI, 0.96–1.64) for acute coronary syndrome hospitalizations, 1.00 (95% CI, 0.80–1.26) for heart failure hospitalizations, 1.05 (95% CI, 0.88–1.25) for atrial fibrillation hospitalizations, 1.92 (95% CI, 1.40–2.64) for percutaneous coronary intervention, and 2.51 (95% CI, 1.99–3.16) for coronary artery bypass grafting surgery. Conclusions Natural language processing–extracted CCS classification was positively associated with all‐cause mortality and healthcare utilization, demonstrating the prognostic importance of anginal symptom assessment and documentation., See Editorial Mehta and Bradley

Background Longer time intervals from presentation with hematuria to bladder cancer diagnosis have been reported among women compared with men. Despite women being the fastest growing cohort within the Department of Veterans Affairs, little is known about women veterans with bladder cancer. Our objectives were to quantify the time from hematuria to bladder cancer diagnosis in Department of Veterans Affairs and assess differences between sexes. Methods This was a retrospective cohort study of patients diagnosed with bladder cancer from 2001 to 2016. Included were patients with hematuria for fewer than 365 days before a bladder cancer diagnosis and who had a record of diagnostic cystoscopy after hematuria but before diagnosis. We evaluated the number of days from hematuria to diagnostic cystoscopy (clinical appraisal), cystoscopy to bladder cancer diagnosis (surgical appraisal), and hematuria to bladder cancer diagnosis (total diagnostic appraisal). We used quantile regression models to separately evaluate the effect of sex on the three appraisal intervals. Results Data from 213 women and 24,295 men were analyzed. The median clinical appraisal time was 78 days for women and 72 for men (p = .49). The median surgical appraisal time was 32 days for women and 33 for men (p = .74). The median total diagnostic appraisal time was 135 days for women and 130 for men (p = .71). Multivariable analyses showed no differences between men and women for any of the three appraisal intervals. Conclusions The majority of time from hematuria to bladder cancer diagnosis is spent in clinical appraisal, but little difference was observed between men and women in Department of Veterans Affairs.

BACKGROUND Although evidence-based psychotherapies (EBPs) for posttraumatic stress disorder (PTSD) were implemented starting in 2005 in the veterans health administration (VHA), the largest national healthcare system in the U.S., the rate of initiation (uptake) and prevalence of these treatments in each calendar year have not been determined. We aimed to elucidate two metrics of EBP utilization, uptake and prevalence, following implementation. METHODS Cohort study of Iraq and Afghanistan veterans in VHA (N = 181,620) with a PTSD diagnosis and ≥1 psychotherapy-coded outpatient visit from 2001 to 2014. Using natural language processing techniques, annual and cumulative uptake and prevalence rates from 2001 to 2014 were calculated for each of the two EBPs for PTSD, cognitive processing therapy (CPT) and prolonged exposure (PE) therapy. RESULTS Annual uptake of CPT increased during most years, reaching a maximum of 11.1%. Annual uptake of PE showed little change until 2008 and then increased, reaching a maximum of 4.4%. The annual prevalence of CPT increased throughout the study, reaching a maximum of 14.6%. The annual prevalence of PE increased to a maximum of 5.0% in 2010, but then flattened and declined. Annual uptake of minimally adequate CPT increased a to maximum of 5% in 2014. Annual uptake of minimally adequate PE increased to a maximum of 1.2% in 2010. The cumulative prevalence of CPT was 19.9% and cumulative prevalence for PE was 7.5%. CONCLUSIONS Access to EBPs for PTSD modestly increased for Iraq and Afghanistan veterans after nationwide implementation efforts. Further expanding the reach to veterans is critical, given low rates of minimally adequate EBPs for PTSD.

Introduction Janssen received reports of needle detachments for Risperdal® CONSTA® and, in response, redesigned the kit. Objective The study objective was to estimate the rate of Risperdal® CONSTA® needle detachments prior to and after the introduction of a redesigned kit. Methods This retrospective study used record abstraction in the US Department of Veterans Affairs (VA). The 3 phases included: (1) a pilot study for methods evaluation in a sample of 6 hospitals with previously reported detachments; (2) a baseline study to ascertain the baseline detachment rate; and (3) a follow-up study to ascertain the rate for the redesigned kit. Administrative codes and natural language processing with clinical review were used to identify detachments. Results Pilot: we identified a subset of spontaneously reported detachments and several previously unreported events. In the baseline study (original device), from January through December 2013, 22 needle detachments were identified among 47,934 administrations of the drug in a census of administrations in the VA; an incidence of 0.0459%. In the follow-up study (redesigned device), from December 2015 through December 2016, there were 14 reported detachments in 41,819 injections, 0.0335%. This represents a reduction of 27% from the baseline. Conclusion This approach enabled us to identify needle detachments we would not have otherwise found (“solicited”). However, it likely resulted in incomplete outcome ascertainment. While this may have resulted in lower overall rates, it did not bias the comparison of the baseline and follow-up studies. The results showed that the redesigned Risperdal® CONSTA® kit reduced the incidence of needle detachment events in the VA. Funding Janssen Pharmaceuticals, Inc.

International audience; In this trans-ethnic multi-omic study, we reinterpret the genetic architecture of blood pressure to identify genes, tissues, phenomes and medication contexts of blood pressure homeostasis. We discovered 208 novel common blood pressure SNPs and 53 rare variants in genome-wide association studies of systolic, diastolic and pulse pressure in up to 776,078 participants from the Million Veteran Program (MVP) and collaborating studies, with analysis of the blood pressure clinical phenome in MVP. Our transcriptome-wide association study detected 4,043 blood pressure associations with genetically predicted gene expression of 840 genes in 45 tissues, and mouse renal single-cell RNA sequencing identified upregulated blood pressure genes in kidney tubule cells.

Background: Familial hypercholesterolemia (FH) is characterized by inherited high levels of LDL-C (low-density lipoprotein cholesterol) and premature coronary heart disease. Over a thousand low-frequency variants in LDLR, APOB, and PCSK9 have been implicated in FH, but few have been examined at the population level. We aim to estimate the phenotypic effects of a subset of FH variants on LDL-C and clinical outcomes among 331 107 multiethnic participants. Methods: We examined the individual and collective association between putatively pathogenic FH variants included on the Million Veteran Program biobank array and the maximum LDL-C level over an interval of 15 years (maxLDL). We assessed the collective effect on clinical outcomes by leveraging data from 61.7 million clinical encounters. Results: We found 8 out of 16 putatively pathogenic FH variants with ≥30 observed carriers to be significantly associated with elevated maxLDL (9.4–80.2 mg/dL). Phenotypic effects were similar for European Americans and African Americans, despite substantial differences in carrier frequencies. Based on observed effects on maxLDL, we identified a total of 748 carriers (1:443) who had elevated maxLDL (36.5±1.4 mg/dL; P =1.2×10 –152 ), and higher prevalence of clinical diagnoses related to hypercholesterolemia and coronary heart disease in a phenome-wide scan. Adjusted for maxLDL, FH variants collectively associated with higher prevalence of coronary heart disease (odds ratio, 1.59; 95% CI, 1.36–1.86, P =1.1×10 –8 ) but not peripheral artery disease. Conclusions: The distribution and phenotypic effects of putatively pathogenic FH variants were heterogeneous within and across variants. More robust evidence of genotype-phenotype associations of FH variants in multiethnic populations is needed to accurately infer at-risk individuals from genetic screening.

This cohort study compares suicide mortality among sexual minority veterans with suicide mortality in the general US and veteran populations., Key Points Question How does suicide mortality among sexual minority veterans compare with suicide mortality in the general US and veteran populations? Findings In this cohort study of 96 893 sexual minority veterans, risk of death from suicide was higher among sexual minority veterans compared with the general US population and the general veteran population. Meaning These findings suggest that sexual minority veterans have an increased risk of suicide mortality and that understanding whether and to what extent prevention efforts reach this population should be the focus of future research., Importance Identification of subgroups at greatest risk for suicide mortality is essential for prevention efforts and targeting interventions. Sexual minority individuals may have an increased risk for suicide compared with heterosexual individuals, but a lack of sufficiently powered studies with rigorous methods for determining sexual orientation has limited the knowledge on this potential health disparity. Objective To investigate suicide mortality among sexual minority veterans using Veterans Health Administration (VHA) electronic health record data. Design, Setting, and Participants This retrospective population-based cohort study used data on 8.1 million US veterans enrolled in the VHA after fiscal year 1999 that were obtained from VHA electronic health records from October 1, 1999 to September 30, 2017. Data analysis was carried out from March 1, 2020 to October 31, 2020. Exposure Veterans with documentation of a minority sexual orientation. Documentation of sexual minority status was obtained through natural language processing of clinical notes and extraction of structured administrative data for sexual orientation in VHA electronic health records. Main Outcomes and Measures Suicide mortality rate using data on the underlying cause of death obtained from the National Death Index. Crude and age-adjusted mortality rates were calculated for all-cause death and death from suicide among sexual minority veterans compared with the general US population and the general population of veterans. Results Among the 96 893 veterans with at least 1 sexual minority documentation in the electronic health record, the mean (SD) age was 46 (16) years, 68% were male, and 70% were White. Of the 12 591 total deaths, 3.5% were from suicide. Veterans had a significantly higher rate of mortality from suicide (standardized mortality ratio, 4.50; 95% CI, 4.13-4.99) compared with the general US population. Suicide was the fifth leading cause of death in 2017 among sexual minority veterans (3.8% of deaths) and the tenth leading cause of death in the general US population (1.7% of deaths). The crude suicide rate among sexual minority veterans (82.5 per 100 000 person-years) was higher than the rate in the general veteran population (37.7 per 100 000 person-years). Conclusions and Relevance The results of this population-based cohort study suggest that sexual minority veterans have a greater risk for suicide than the general US population and the general veteran population. Further research is needed to determine whether and how suicide prevention efforts reach sexual minority veterans.

Prolonged exposure therapy (PE) is an effective treatment for posttraumatic stress disorder (PTSD). Identifying metrics of treatment response can guide treatment delivery. The median effective dose represents the number of sessions at which there is a 50% probability of clinically meaningful improvement (i.e., 10-point reduction in PTSD checklist). The goal of the current study was to investigate the median effective dose of PE. We identified a cohort of Iraq and Afghanistan war veterans who received psychotherapy for PTSD in the Veterans Health Administration between 2001 and 2017. From this cohort, 10,234 veterans who received PE (as identified using natural language processing) and had ≥2 PTSD symptom measures were included in analyses. To determine how the number of PE sessions and covariates affected clinically meaningful improvement, we utilized a Cox proportional hazards regression, followed by Kaplan-Meier curves to determine the median effective dose. The median effective dose of PE was four sessions. Although some covariates were found to be statistically significant predictors of clinically meaningful improvement (e.g., age, gender, PTSD medications, and depressive disorder comorbidity), these effects were small. Clinicians and patients should consider evaluating treatment response after four sessions to determine preliminary effectiveness of PE.

In 2013 binge-eating disorder (BED) was recognized as a formal diagnosis, but was historically included under the diagnosis code for eating disorder not otherwise specified (EDNOS). This study compared the characteristics and use of treatment modalities in BED patients to those with EDNOS without BED (EDNOS-only) and to matched-patients with no eating disorders (NED).Patients were identified for this study from electronic health records in the Department of Veterans Affairs from 2000 to 2011. Patients with BED were identified using natural language processing and patients with EDNOS-only were identified by ICD-9 code (307.50). First diagnosis defined index date for these groups. NED patients were frequency matched to BED patients up to 4:1, as available, on age, sex, BMI, depression, and index month encounter. Baseline characteristics and use of treatment modalities during the post-index year were compared using t-tests or chi-square tests.There were 593 BED, 1354 EDNOS-only, and 1895 matched-NED patients identified. Only 68 patients with BED had an EDNOS diagnosis. BED patients were younger (48.7 vs. 49.8years, p=0.04), more were male (72.2% vs. 62.8%, p0.001) and obese (BMI 40.2 vs. 37.0, p0.001) than EDNOS-only patients. In the follow-up period fewer BED (68.0%) than EDNOS-only patients (87.6%, p0.001), but more BED than NED patients (51.9%, p0.001) used at least one treatment modality.The characteristics of BED patients were different from those with EDNOS-only and NED as was their use of treatment modalities. These differences highlight the need for a separate identifier of BED.

60 Background: Veterans may benefit from promising innovations in treatments for mPCa. The Veterans Affairs (VA) and Prostate Cancer Foundation (PCF) leadership issued a challenge to identify, in real time, the national census of Veterans receiving care for mPCa. Administrative diagnostic and procedural coding do not accurately identify the risk status or disease state of prostate cancer (PCa). This study reports the development and validation of NLP tools deployed on clinical notes to identify risk status or disease state. Methods: Using diagnosis and histology codes, we queried the VA Corporate Data Warehouse to identify Veterans with prostate cancer. We included structured laboratory tests, medications, procedures, and surgeries related to prostate cancer diagnosis or treatment in the analysis. Using structured data, we identified 1000 likely mPCa cases and controls. Medical records were reviewed to confirm status and to extract term dictionaries related to cancer, anatomy, metastasis, and other diagnostic concepts. We went through several iterations of testing to refine and validate the NLP tool on a limited set of known cases and controls. We deployed the tool on all cancer, urology, pathology, and radiation oncology notes. Results: The NLP system was able to identify the patients' history of metastatic disease with 0.975 precision and 0.828 recall. Among the 1,081,137 Veterans with prostate cancer, NLP identified 63,222 (5.8%) with mPCa. There are 16,282 Veterans alive with mPCa. Mean age of diagnosis was 67 and 8,847 (54.3%) were diagnosed in the VA. Demographics were: White 9,756 (60%), Black 4,466 (27%), and other 2,060 (13%). Conclusions: NLP is a reliable tool for identifying Veterans who may benefit from novel innovations in mPCa diagnosis and treatment.[Table: see text]

394 Background: It is essential to study real-world evidence of novel molecular diagnostic tests to measure their impact on clinical decision-making and health outcomes. Studying test utilization within the Department of Veterans Affairs (VA) informs about clinical validity in a racially diverse population. This study analyzed the impact of DNA promoter methylation (ConfirmDx), on the diagnostic workup of Veterans. Methods: The VA Corporate Data Warehouse (CDW) was used to identify all incident prostate biopsies and cancers from 1/2014-10/2019. We captured patient clinical characteristics, including number and date of biopsies, date of cancer diagnosis, dates and values of PSA tests, comorbidities, demographics, whether the patient underwent DNA promoter methylation testing, and the results of that test. Results: Each year an average of 25,299 biopsies are conducted in the VA. Between 2014 and 2019, 151,794 biopsies were performed. 98,155 (70%) Veterans underwent first-time biopsies and 54,447 (55%) were positive for carcinoma. 8,062 (8.2%) underwent multiple biopsies, among whom 3,589 (44%) were subsequently diagnosed with cancer. 34 Veterans underwent 5 or more biopsies before being diagnosed. 33,878 (35%) did not undergo another biopsy in the study period. 328 (0.3%) of the 98,155 Veterans underwent the DNA promoter methylation test. 55 (17%) of these Veterans were eligible for Medicare coverage of the test. Of the 328 tested for DNA methylation, 217 (66%) were minorities. 132/328 (40%) tested positive for carcinoma and 168/328 (51%) were negative. Conclusions: Most Veterans referred for prostate biopsy have detectable cancer, and a large number are minorities with potential increased prostate cancer risk. DNA promoter methylation may improve care by reducing the need for multiple biopsies, limiting exposure to infection, and reducing time to diagnosis and treatment.

Little is known about predictors of initiation and completion of evidence-based psychotherapy (EBP) for posttraumatic stress disorder (PTSD), with most data coming from small cohort studies and post-hoc analyses of clinical trials. We examined patient and treatment factors associated with initiation and completion of EBP for PTSD in a large longitudinal cohort. We conducted a national, retrospective cohort study of all Iraq and Afghanistan War veterans who had a post-deployment PTSD diagnosis from 10/01-9/15 at a Veterans Health Administration facility and had at least one coded post-deployment psychotherapy visit. We examined utilization of PE and CPT (individual or group) during any 24-week period. We used ordered logistic, logistic, and Cox proportional hazards regressions to examine variables associated with EBP initiation, early termination, and completion, and time to completion. Over a 15-year period, of 265,566 veterans with PTSD, 22.8% initiated an EBP, and only 9.1% completed treatment. Completers did so about three years after their initial mental health visit. Factors positively associated with EBP completion included military sexual trauma, older age, race/ethnicity (i.e., African-American race for PE), combat, and multiple deployments. The VHA has become timelier in delivering EBP for PTSD, and several subgroups are more likely to complete EBP.

Natural Language Processing (NLP) holds potential for patient care and clinical research, but a gap exists between promise and reality. While some studies have demonstrated portability of NLP systems across multiple sites, challenges remain. Strategies to mitigate these challenges can strive for complex NLP problems using advanced methods (hard-to-reach fruit), or focus on simple NLP problems using practical methods (low-hanging fruit). This paper investigates a practical strategy for NLP portability using extraction of left ventricular ejection fraction (LVEF) as a use case. We used a tool developed at the Department of Veterans Affair (VA) to extract the LVEF values from free-text echocardiograms in the MIMIC-III database. The approach showed an accuracy of 98.4%, sensitivity of 99.4%, a positive predictive value of 98.7%, and F-score of 99.0%. This experience, in which a simple NLP solution proved highly portable with excellent performance, illustrates the point that simple NLP applications may be easier to disseminate and adapt, and in the short term may prove more useful, than complex applications.

Introduction: PAD is a leading cause of cardiovascular morbidity and mortality. Previously published GWAS have been limited by small sample sizes and have only identified 3 genome-wide significant (P-8 ) risk loci to date. Hypothesis: DNA sequence variants affecting multiple biological pathways are associated with PAD risk. Methods: Using electronic health record data, we identified individuals with and without clinical PAD from the 353,323 Million Veteran Program (MVP) participants genotyped on a customized Affymetrix Biobank array. We tested 32 million genotyped and imputed DNA variants for association with clinical PAD separately in participants of European (EUR), African (AFR), and Hispanic (HIS) ancestry using logistic regression models controlling for age, sex and population structure, and then performed trans-ethnic meta-analysis. The results were replicated with data from the UK Biobank. Results: We identified 31,307 individuals (24,009 EUR, 5,373 AFR, 1,925 HIS) with, and 211,753 individuals without, PAD. Following meta-analysis and replication, there were 19 genome-wide significant risk loci associated with PAD. We replicated a known association at 9p21 (P=4.3 x10 -39 ), and identified several novel loci associated with PAD that were previously known to be associated with atherosclerosis ( LPA , HDAC9 ), diabetes ( TCF7L2 ), lipid levels ( LPL , CELSR2 ), and tobacco use ( CHRNA3 ). We also identified a novel association with PAD for the Factor V Leiden (FVL) mutation (OR 1.20, P=1.6x10 -12 ), which remained significant after controlling for venous thromboembolism (OR 1.10, P=8.7x10 -4 ). Sensitivity analysis demonstrated FVL is associated with increasing risk estimates for PAD severity (claudication OR 1.19, P=0.0012; rest pain OR 1.41, P=0.004; tissue loss OR 1.57, P=7x10 -9 ). Conclusions: Using the MVP, we assembled the largest reported cohort of individuals with clinical PAD and genetic data worldwide. Our data replicate known causal risk factors and identify a novel association for FVL and its putative role for thrombosis in the development of clinical PAD.

Objective: To describe a framework for leveraging big data for research and quality improvement purposes and demonstrate implementation of the framework for design of the Department of Veterans Affairs (VA) Colonoscopy Collaborative.Methods: We propose that research utilizing large-scale electronic health records (EHRs) can be approached in a 4 step framework: 1) Identify data sources required to answer research question; 2) Determine whether variables are available as structured or free-text data; 3) Utilize a rigorous approach to refine variables and assess data quality; 4) Create the analytic dataset and perform analyses. We describe implementation of the framework as part of the VA Colonoscopy Collaborative, which aims to leverage big data to 1) prospectively measure and report colonoscopy quality and 2) develop and validate a risk prediction model for colorectal cancer (CRC) and high-risk polyps.Results: Examples of implementation of the 4 step framework are provided. To date, we have identified 2,337,171 Veterans who have undergone colonoscopy between 1999 and 2014. Median age was 62 years, and 4.6 percent (n = 106,860) were female. We estimated that 2.6 percent (n = 60,517) had CRC diagnosed at baseline. An additional 1 percent (n = 24,483) had a new ICD-9 code-based diagnosis of CRC on follow up.Conclusion: We hope our framework may contribute to the dialogue on best practices to ensure high quality epidemiologic and quality improvement work. As a result of implementation of the framework, the VA Colonoscopy Collaborative holds great promise for 1) quantifying and providing novel understandings of colonoscopy outcomes, and 2) building a robust approach for nationwide VA colonoscopy quality reporting.

Background: This study estimated the risk of infection-related hospitalizations and death in patients with and without multiple sclerosis (MS). Methods: We identified adults with MS in the US Department of Veterans Affairs (VA) system between 1999 and 2010. Each veteran with MS was matched, on age and sex, with up to four veterans without MS. Multivariable Cox proportional hazards regression models were performed to assess the influence of MS on the development of serious and fatal infections. Results: The cohort included 7743 veterans with MS and 30,972 veterans without MS. Mean (SD) age was 53.8 (13.3) years, and 80.8% were male. The incidence per 1000 person-years of overall serious infections was 19.2 (95% confidence interval [CI], 17.6–20.8) for those with MS and 10.3 (95% CI, 9.8–10.9) for those without MS. Fatal infection incidence rates were 1.2 (95% CI, 0.8–1.7) for patients with MS and 0.5 (95% CI, 0.3–0.6) for patients without MS. Regression models showed that veterans with MS were at greater risk for overall serious (hazard ratio [HR] = 1.52, P < .01) and fatal (HR = 1.85, P = .03) infections and serious respiratory (HR = 1.31, P = .01), urinary tract (HR = 4.44, P < .01), and sepsis-related infections (HR = 2.56, P < .01). Conclusions: This study provides evidence that VA patients with MS are more likely than those without MS to be hospitalized and die of infection.

Objective The objective of this study was to compare the one-year healthcare costs and utilization of patients with binge-eating disorder (BED) to patients with eating disorder not otherwise specified without BED (EDNOS-only) and to matched patients without an eating disorder (NED). Methods A natural language processing (NLP) algorithm identified adults with BED from clinical notes in the Department of Veterans Affairs (VA) electronic health record database from 2000 to 2011. Patients with EDNOS-only were identified using ICD-9 code (307.50) and those with NLP-identified BED were excluded. First diagnosis date defined the index date for both groups. Patients with NED were randomly matched 4:1, as available, to patients with BED on age, sex, BMI, depression diagnosis, and index month. Patients with cost data (2005–2011) were included. Total healthcare, inpatient, outpatient, and pharmacy costs were examined. Generalized linear models were used to compare total one-year healthcare costs while adjusting for baseline patient characteristics. Results There were 257 BED, 743 EDNOS-only, and 823 matched NED patients identified. The mean (SD) total unadjusted one-year costs, in 2011 US dollars, were $33,716 ($38,928) for BED, $37,052 ($40,719) for EDNOS-only, and $19,548 ($35,780) for NED patients. When adjusting for patient characteristics, BED patients had one-year total healthcare costs $5,589 higher than EDNOS-only (p = 0.06) and $18,152 higher than matched NED patients (p

There are limitations to using administrative data to identify postoperative venous thromboembolism (VTE). We used a novel approach to quantify postoperative VTE events among Department of Veterans Affairs (VA) surgical patients during 2005-2010.We used VA administrative data to exclude patients with VTE during 12 months prior to surgery. We identified probable postoperative VTE events within 30 and 90 days post-surgery in three settings: 1) pre-discharge inpatient, using a VTE diagnosis code and a pharmacy record for anticoagulation; 2) post-discharge inpatient, using a VTE diagnosis code followed by a pharmacy record for anticoagulation within 7 days; and 3) outpatient, using a VTE diagnosis code and either anticoagulation or a therapeutic procedure code with natural language processing (NLP) to confirm acute VTE in clinical notes.Among 468,515 surgeries without prior VTE, probable VTEs were documented within 30 and 90 days in 3,931 (0.8%) and 5,904 (1.3%), respectively. Of probable VTEs within 30 or 90 days post-surgery, 47.8% and 62.9%, respectively, were diagnosed post-discharge. Among post-discharge VTE diagnoses, 86% resulted in a VA hospital readmission. Fewer than 25% of outpatient records with both VTE diagnoses and anticoagulation prescriptions were confirmed by NLP as acute VTE events.More than half of postoperative VTE events were diagnosed post-discharge; analyses of surgical discharge records are inadequate to identify postoperative VTE. The NLP results demonstrate that the combination of VTE diagnoses and anticoagulation prescriptions in outpatient administrative records cannot be used to validly identify postoperative VTE events.

Objective To evaluate physician adherence with gout quality indicators (QIs) for medication use and monitoring, and behavioral modification (BM). Methods Gout patients were assessed for the QIs as follows: QI 1: initial allopurinol dosage

Objective To examine the association of serum lipids, inflammation and seropositivity on coronary heart disease (CHD) and stroke in patients with rheumatoid arthritis (RA). Methods The incidence of hospitalised myocardial infarction (MI) or stroke was calculated in a cohort of patients with RA receiving care within the national Veterans Health Administration from 1998 to 2011. Cox proportional hazard models were used to examine the association between these outcomes and low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), C reactive protein (CRP) and erythrocyte sedimentation rate (ESR) as time-varying variables, divided into quintiles. Results There were 37 568 patients with RA in the cohort with mean age of 63 years (SD 12.1); 90% were men. There was a no clear association between LDL-C and CHD/stroke. Compared with lower HDL-C (2.17 mg/dL (vs CRP 47 mm/h compared with

During the dual-therapy era, many patients with chronic hepatitis C discontinued therapy for reasons other than lack of efficacy (non-LOE). We determined whether selected patient characteristics predicted non-LOE discontinuation using national databases of U.S. veterans with Genotypes 1-4. We identified U.S. veterans in the Veterans Health Administration system in 2004-2009 who had hepatitis C-confirming RNA laboratory results and initiated therapy with pegylated interferon and ribavirin. We used a rule to classify patients who discontinued pegylated interferon early, based on pharmacy refill and viral response data. Multivariate Cox regression was used to identify predictors of non-LOE discontinuation. Of 321,238 patients with a hepatitis C International Classification of Diseases, Ninth Revision, code, 15,297 (4.8%) met all inclusion criteria. Non-LOE discontinuers comprised 30.3% of patients. For Genotypes 1-4, the predictors (adjusted hazard ratio) of greatest magnitude were comorbidities of myocardial infarction/congestive heart failure (1.36), renal disease (1.34), and platelets 100/mm or more (1.38). For Genotypes 2 and 3, predictors of greatest magnitude were Black race (1.30), myocardial infarction/congestive heart failure (1.84), albumin 3.5 mg/dl or more (1.65), sleep aid use (1.32), and poor persistence with antidepressants (1.31) and antihypertensive agents (1.37). Our study suggests that many host factors may have contributed to non-LOE dual-therapy discontinuation in veterans and may possibly predict non-LOE discontinuation in triple therapy.

Background In order to investigate the mechanisms of cardiovascular disease in HIV infected and uninfected patients, an analysis of echocardiogram reports is required for a large longitudinal multi-center study. Implementation A natural language processing system using a dictionary lookup, rules, and patterns was developed to extract heart function measurements that are typically recorded in echocardiogram reports as measurement-value pairs. Curated semantic bootstrapping was used to create a custom dictionary that extends existing terminologies based on terms that actually appear in the medical record. A novel disambiguation method based on semantic constraints was created to identify and discard erroneous alternative definitions of the measurement terms. The system was built utilizing a scalable framework, making it available for processing large datasets. Results The system was developed for and validated on notes from three sources: general clinic notes, echocardiogram reports, and radiology reports. The system achieved F-scores of 0.872, 0.844, and 0.877 with precision of 0.936, 0.982, and 0.969 for each dataset respectively averaged across all extracted values. Left ventricular ejection fraction (LVEF) is the most frequently extracted measurement. The precision of extraction of the LVEF measure ranged from 0.968 to 1.0 across different document types. Conclusions This system illustrates the feasibility and effectiveness of a large-scale information extraction on clinical data. New clinical questions can be addressed in the domain of heart failure using retrospective clinical data analysis because key heart function measurements can be successfully extracted using natural language processing. Electronic supplementary material The online version of this article (doi:10.1186/s12872-017-0580-8) contains supplementary material, which is available to authorized users.

Objective To take the first step toward assembling population-based cohorts of patients with bladder cancer with longitudinal pathology data, we developed and validated a natural language processing (NLP) engine that abstracts pathology data from full-text pathology reports. Methods Using 600 bladder pathology reports randomly selected from the Department of Veterans Affairs, we developed and validated an NLP engine to abstract data on histology, invasion (presence vs absence and depth), grade, the presence of muscularis propria, and the presence of carcinoma in situ. Our gold standard was based on an independent review of reports by 2 urologists, followed by adjudication. We assessed the NLP performance by calculating the accuracy, the positive predictive value, and the sensitivity. We subsequently applied the NLP engine to pathology reports from 10,725 patients with bladder cancer. Results When comparing the NLP output to the gold standard, NLP achieved the highest accuracy (0.98) for the presence vs the absence of carcinoma in situ. Accuracy for histology, invasion (presence vs absence), grade, and the presence of muscularis propria ranged from 0.83 to 0.96. The most challenging variable was depth of invasion (accuracy 0.68), with an acceptable positive predictive value for lamina propria (0.82) and for muscularis propria (0.87) invasion. The validated engine was capable of abstracting pathologic characteristics for 99% of the patients with bladder cancer. Conclusion NLP had high accuracy for 5 of 6 variables and abstracted data for the vast majority of the patients. This now allows for the assembly of population-based cohorts with longitudinal pathology data.

© 2017 American Medical Association. All rights reserved. Importance: With improved survival, heart failure (HF) has become a major complication for individuals with human immunodeficiency virus (HIV) infection. It is unclear if this risk extends to different types of HF in the antiretroviral therapy (ART) era. Determining whether HIV infection is associated with HF with reduced ejection fraction (HFrEF), HF with preserved ejection fraction (HFpEF), or both is critical because HF types differ with respect to underlying mechanism, treatment, and prognosis. Objectives: To investigate whether HIV infection increases the risk of future HFrEF and HFpEF and to assess if this risk varies by sociodemographic and HIV-specific factors. Design, Setting, and Participants: This study evaluated 98 015 participants without baseline cardiovascular disease from the Veterans Aging Cohort Study, an observational cohort of HIV-infected veterans and uninfected veterans matched by age, sex, race/ethnicity, and clinical site, enrolled on or after April 1, 2003, and followed up through September 30, 2012. The dates of the analysis were October 2015 to November 2016. Exposure: Human immunodeficiency virus infection. Main Outcomes and Measures: Outcomes included HFpEF (EF≥50%), borderline HFpEF (EF 40%-49%), HFrEF (EF

Background: Stable angina is estimated to affect more than 10 million Americans and is the presenting symptom in half of patients diagnosed with coronary disease. Documentation of angina severity resides as unstructured data and is often unavailable in large datasets. We used natural language processing (NLP) to identify Canadian Cardiovascular Society (CCS) angina class and determine the association with all-cause mortality in an integrated health system’s electronic records (EHR). Methods: We performed a historic cohort study using national Veterans Health Administration data between 1/1/06 and 12/31/13. Veterans with incident stable angina were identified by ICD-9-CM codes. We developed an NLP tool to extract CCS class from free text notes. Risk ratios (RR) for all-cause mortality at one year associated with CCS class were calculated using Poisson regression. Results: There were 299,577 Veterans with angina, of which 14,216 had at least one CCS class extracted via NLP. Mean age was 66.6 years, 98% were male sex, and 82% were white. Diabetes increased with CCS class, but other comorbidities were stable (Table). There were 719 deaths at one year follow-up. The adjusted RR for all-cause mortality at one-year comparing Class III to Class I and Class IV to Class I was 1.40 (95% CI 1.16 - 1.68) and 1.52 (95% CI 1.13 - 2.04), respectively. Conclusion: NLP-derived CCS class was independently associated with one year all-cause mortality. Its application may be limited by inadequate EHR documentation of angina severity.

Objective The comparative risk of infection associated with non–anti–tumor necrosis factor (anti-TNF) biologic agents is not well established. Our objective was to compare risk for hospitalized infections between anti-TNF and non–anti-TNF biologic agents in US veterans with rheumatoid arthritis (RA). Methods Using 1998–2011 data from the US Veterans Health Administration, we studied RA patients initiating rituximab, abatacept, or anti-TNF therapy. Exposure was based upon days supplied (injections) or usual dosing intervals (infusions). Treatment episodes were defined as new biologic agent use. Hazard ratios (HRs) and 95% confidence intervals (95% CIs) for hospitalization for a bacterial infection were estimated from Cox proportional hazards models, adjusting for potential confounders. Results Among 3,152 unique RA patients contributing 4,158 biologic treatment episodes to rituximab (n = 596), abatacept (n = 451), and anti-TNF agents (n = 3,111), the patient mean age was 60 years and 87% were male. The most common infections were pneumonia (37%), skin/soft tissue (22%), urinary tract (9%), and bacteremia/sepsis (7%). Hospitalized infection rates per 100 person-years were 4.4 (95% CI 3.1–6.4) for rituximab, 2.8 (95% CI 1.7–4.7) for abatacept, and 3.0 (95% CI 2.5–3.5) for anti-TNF. Compared to etanercept, the adjusted rate of hospitalized infection was not different for adalimumab (HR 1.4, 95% CI 0.9–2.2), abatacept (HR 1.1, 95% CI 0.6–2.1), or rituximab (HR 1.4, 0.8–2.6), although it was increased for infliximab (HR 2.3, 95% CI 1.3–4.0). Infection risk was greater for those taking prednisone >7.5 mg/day (HR 1.8, 95% CI 1.3–2.7) and in the highest quartile of C-reactive protein (HR 2.3, 95% CI 1.4–3.8) and erythrocyte sedimentation rate (HR 4.1, 95% CI 2.3–7.2) compared to the lowest quartile. Conclusion In older, predominantly male US veterans with RA, the risk of hospitalized bacterial infections associated with rituximab or abatacept was similar to etanercept.

Binge eating disorder (BED) does not have an International Classification of Diseases, 9th or 10th edition code, but is included under ‘eating disorder not otherwise specified’ (EDNOS). This historical cohort study identified patients with clinician-diagnosed BED from electronic health records (EHR) in the Department of Veterans Affairs between 2000 and 2011 using natural language processing (NLP) and compared their characteristics to patients identified by EDNOS diagnosis codes. NLP identified 1487 BED patients with classification accuracy of 91.8% and sensitivity of 96.2% compared to human review. After applying study inclusion criteria, 525 patients had NLP-identified BED only, 1354 had EDNOS only, and 68 had both BED and EDNOS. Patient characteristics were similar between the groups. This is the first study to use NLP as a method to identify BED patients from EHR data and will allow further epidemiological study of patients with BED in systems with adequate clinical notes.

In observational studies, higher uric acid levels are associated with metabolic syndrome, diabetes, and kidney disease.The objective of this study is to examine whether reduction of plasma uric acid with febuxostat, a xanthine oxido reductase inhibitor, impacts adipose tissue oxidative stress, adipokines, and markers of systemic inflammation or kidney fibrosis.This was a double-blinded randomized controlled trial.Academic university setting was used.Overweight or obese adults with hyperuricemia and type 2 diabetic nephropathy were included.Adipose tissue thiobarbituric acid reducing substances (TBARS) and adiponectin concentrations and urinary transforming growth factor-β (TGF-β) were primary endpoints. Plasma C-reactive protein, high molecular weight-adiponectin, interleukin-6, tumor necrosis factor-α, and TBARS and albuminuria were among predefined secondary endpoints.Participants were randomly assigned to febuxostat (n = 40) or matching placebo (n = 40) and followed for 24 weeks.Baseline plasma uric acid levels were 426 ± 83 µmol/L; 95% completed the study. Estimated glomerular filtration rate (eGFR) declined from 54 ± 17 mL/min/1.73 mRelatively modest sample size and short duration of follow-up.In this population with progressive diabetic nephropathy, febuxostat effectively reduced plasma uric acid. However, no detectable effects were observed for the prespecified primary or secondary endpoints.The study was registered in clinicaltrials.gov (NCT01350388).Dans les études observationnelles, des taux élevés d’acide urique sont associés à un syndrome métabolique, au diabète et à l’insuffisance rénale.Cette étude visait à déterminer si l’utilisation de febuxostat, un inhibiteur de la xanthine oxydoréductase, pour réduire le taux d’acide urique dans le plasma avait une incidence sur le stress oxydatif du tissu adipeux, les adipokines, les marqueurs de l’inflammation systémique ou sur la fibrose kystique.Il s’agit d’un essai à double insu, randomisé et contrôlé.L’étude s’est effectuée en contexte universitaire.Les participants à cette étude étaient des adultes obèses ou en surpoids, présentant une hyperuricémie ainsi qu’une néphropathie diabétique de type 2.Les critères principaux incluaient la concentration d’adiponectine et de substances réagissant avec l’acide thiobarbiturique (TBARS) dans les tissus adipeux, de même que le facteur de croissance transformant urinaire (TGF-β). Les critères secondaires incluaient les protéines C-réactives du plasma, l’adiponectine de poids moléculaire élevé, l’interleukine-6, le facteur de nécrose tumorale alpha, les TBARS ainsi que l’albuminurie.On a prescrit, de façon aléatoire, du febuxostat (n = 40) ou un placebo (n = 40) aux participants, et ces derniers ont été suivis sur une période de 24 semaines.Les valeurs initiales d’acide urique dans le plasma se situaient à 426 ± 83 µmol/L. La grande majorité des participants (95%) a complété l’étude. Le débit de filtration glomérulaire estimé a chuté de 54 ± 17 mL/min/1,73 mLa taille relativement modeste de l’échantillon, de même que la courte durée du suivi constituent les limites de l’étude.Dans la population observée, soit des patients atteints de néphropathie diabétique progressive, l’administration de fébuxostat a réduit de façon efficace les taux plasmatiques d’acide urique. Par ailleurs, aucun effet apparent n’a été observé sur les critères primaires et secondaires préétablis.Cette étude a été enregistrée sur clinicaltrials.gov (NCT01350388).

We assessed implementation of precision medicine within the Veterans Health Administration. We analyzed the use of interleukin-28B (IL28B) pharmacogenetic test, which predicts interferon-α treatment response in patients with hepatitis C. Patients with favorable CC genotype exhibit a two-fold higher response than patients with less favorable genotypes (CT and TT). Linking IL28B tests to Veterans Health Administration clinical data, we analyzed test use, concordance with guidelines, subsequent interferon-α treatment, and site variations. From January 2011 until December 2013, 3,529 Veterans underwent IL28B testing. There were 2,988 (85%) tests linked to electronic health record data. Demographics in those with clinical data: 97% male, mean age 59 years (SD 6.8), race: White 50%, Black 43%, Hispanic 3%, and other 3%. In Whites, the favorable CC genotype was three-fold more prevalent (541, 36%) than in Blacks (151, 12%). Analysis of IL28B timing revealed 2,373 (79%) Veterans were appropriately tested before interferon-α, 49 (2%) tested concurrent with interferon-α, and 566 (19%) tested post interferon-α treatment. Of the 630 treatment-naïve Veterans with CC genotype, only 144 (23%) initiated interferon-α treatment post-testing. Although 35% of IL28B tests overall did not influence care, the majority of tests were guideline concordant and clinically useful. IL28B testing varied substantially by site and state.

Guideline-concordant cancer care is a priority within the Department of Veterans Affairs (VA). In 2009, the VA expanded its capacity to treat breast cancer patients within VA medical centers (VAMCs). We sought to determine whether male and female Veterans diagnosed with breast cancer received BRCA testing as recommended by the National Comprehensive Cancer Network (NCCN) guidelines on Genetic/Familial High-Risk Assessment in Breast and Ovarian Cancer (v. 1.2010-1.2012). Using the 2011-2012 VA Central Cancer Registry and BRCA test orders from Myriad Genetics, we conducted a retrospective study. The outcome variable was a recommendation for genetic counseling or BRCA testing, determined by chart review. Independent variables expected to predict testing included region, site of care, and patient characteristics. We performed descriptive analysis of all patients and conducted multivariable logistic regression on patients who sought care at VAMCs that offered BRCA testing. Of the 462 Veterans who met NCCN testing criteria, 126 (27 %) received guideline-concordant care, either a referral for counseling or actual testing. No BRCA testing was recommended in 49 (50 %) VAMCs that provide cancer treatment. Surprisingly, patients with second primary breast cancer were less likely to be referred/tested (OR 0.39; CI 0.17, 0.89; p = 0.025). For patients under age 51, a yearly increase in age decreased likelihood of referral or testing (OR 0.85; CI 0.76, 0.94; p

Introduction: Substantial amounts of clinically significant information are contained only within the narrative of the clinical notes in electronic medical records. v3NLP Framework is a set of best of breed functionalities developed to transform this information into structured data for use in quality improvement, research, population health surveillance, and decision support.Background: MetaMap, cTAKES and similar well known NLP tools do not have sufficient scalability out of the box. v3NLP Framework evolved out of the necessity to scale these tools up and provide a framework to customize and tune techniques to fit a variety of tasks, including document classification, tuned concept extraction for specific conditions, patient classification, and information retrieval.Innovation: Beyond scalability, several v3NLP Framework developed projects have been efficacy tested and benchmarked. While v3NLP Framework includes annotators, pipelines and applications, the functionalities enable developers to create novel annotators, put annotators into pipelines and scaled applications.Discussion: v3NLP Framework has been successfully utilized in many projects including general concept extraction, risk factors for homelessness among veterans, and identification of mentions of the presence of an indwelling urinary catheter. Projects as diverse as predicting colonization with methicillin resistant Staphylococcus aureus and extracting references to military sexual trauma are being built using v3NLP Framework components.Conclusion: v3NLP Framework is a set of functionalities and components that provide Java developers the ability to create novel annotators, place annotators into pipelines, and applications to extract concepts from clinical text. There are scale-up and scale-out functionalities to process large numbers of records.

Automatic clustering of Web pages helps a number of information retrieval tasks, such as improving user interfaces, collection clustering, introducing diversity in search results, etc. Typically, Web page clustering algorithms use only features extracted from the page-text. However, the advent of social-bookmarking Web sites, such as StumbleUpon.com and Delicious.com, has led to a huge amount of user-generated content such as the social tag information that is associated with the Web pages. In this article, we present a subspace based feature extraction approach that leverages the social tag information to complement the page-contents of a Web page for extracting beter features, with the goal of improved clustering performance. In our approach, we consider page-text and tags as two separate views of the data, and learn a shared subspace that maximizes the correlation between the two views. Any clustering algorithm can then be applied in this subspace. We then present an extension that allows our approach to be applicable even if the Web page corpus is only partially tagged, that is, when the social tags are present for not all, but only for a small number of Web pages. We compare our subspace based approach with a number of baselines that use tag information in various other ways, and show that the subspace based approach leads to improved performance on the Web page clustering task. We also discuss some possible future work including an active learning extension that can help in choosing which Web pages to get tags for, if we only can get the social tags for only a small number of Web pages.

Objectives Left ventricular ejection fraction (EF) is a key component of heart failure quality measures used within the Department of Veteran Affairs (VA). Our goals were to build a natural language processing system to extract the EF from free-text echocardiogram reports to automate measurement reporting and to validate the accuracy of the system using a comparison reference standard developed through human review. This project was a Translational Use Case Project within the VA Consortium for Healthcare Informatics. Materials and methods We created a set of regular expressions and rules to capture the EF using a random sample of 765 echocardiograms from seven VA medical centers. The documents were randomly assigned to two sets: a set of 275 used for training and a second set of 490 used for testing and validation. To establish the reference standard, two independent reviewers annotated all documents in both sets; a third reviewer adjudicated disagreements. Results System test results for document-level classification of EF of