Connecticut VA Healthcare System

West Haven, CT, United States

Connecticut VA Healthcare System

West Haven, CT, United States
SEARCH FILTERS
Time filter
Source Type

Tsai J.,Indiana University – Purdue University Indianapolis | Tsai J.,Connecticut VA Healthcare System | Bond G.R.,Indiana University – Purdue University Indianapolis | Salyers M.P.,Indiana University – Purdue University Indianapolis | And 4 more authors.
Community Mental Health Journal | Year: 2010

Housing is a crucial issue for adults with severe mental illness and co-occurring substance use disorders, as this population is particularly susceptible to housing instability and homelessness. We interviewed 40 adults with dual disorders, living in either supervised or independent housing arrangements, to examine housing preferences, decision making processes surrounding housing choices, and perceived barriers to housing. We found that many clients indicated their housing preferences had changed over time, and some clients related housing preferences to recovery. Although the majority of clients preferred independent housing, many also described benefits of supervised housing. Clients' current living situations appeared to be driven primarily by treatment provider recommendations and availability of housing. Common barriers to obtaining desired housing were lack of income and information. These findings have implications for supported housing models and approaches to providing housing for clients. © 2009 Springer Science+Business Media, LLC.


Garla V.,Yale University | Re III. V.L.,University of Pennsylvania | Dorey-Stein Z.,University of Pennsylvania | Kidwai F.,Connecticut VA Healthcare System | And 8 more authors.
Journal of the American Medical Informatics Association | Year: 2011

Background: Open-source clinical natural-languageprocessing (NLP) systems have lowered the barrier to the development of effective clinical document classification systems. Clinical natural-languageprocessing systems annotate the syntax and semantics of clinical text; however, feature extraction and representation for document classification pose technical challenges. Methods: The authors developed extensions to the clinical Text Analysis and Knowledge Extraction System (cTAKES) that simplify feature extraction, experimentation with various feature representations, and the development of both rule and machine-learning based document classifiers. The authors describe and evaluate their system, the Yale cTAKES Extensions (YTEX), on the classification of radiology reports that contain findings suggestive of hepatic decompensation. Results and discussion: The F 1-Score of the system for the retrieval of abdominal radiology reports was 96%, and was 79%, 91%, and 95% for the presence of liver masses, ascites, and varices, respectively. The authors released YTEX as open source, available at http://code.google.com/p/ytex.


Weisbord S.D.,VA Pittsburgh Healthcare System | Weisbord S.D.,University of Pittsburgh | Weisbord S.D.,Center for Health Equity Research and Promotion | Gallagher M.,George Institute for Global Health | And 20 more authors.
Clinical Journal of the American Society of Nephrology | Year: 2013

Contrast-induced AKI (CI-AKI) is a common condition associated with serious, adverse outcomes. CI-AKI may be preventable because its risk factors are well characterized and the timing of renal insult is commonly known in advance. Intravenous (IV) fluids and N-acetylcysteine (NAC) are two of the most widely studied preventive measures for CI-AKI. Despite a multitude of clinical trials and meta-analyses, the most effective type of IV fluid (sodium bicarbonate versus sodium chloride) and the benefit of NAC remain unclear. Careful review of published trials of these interventions reveals design limitations that contributed to their inconclusive findings. Such design limitations include the enrollment of small numbers of patients, increasing the risk for type I and type II statistical errors; the use of surrogate primary endpoints defined by small increments in serum creatinine, which are associated with, but not necessarily causally related to serious, adverse, patient-centered outcomes; and the inclusion of low-risk patients with intact baseline kidney function, yielding low event rates and reduced generalizability to a higher-risk population. The Prevention of Serious Adverse Events following Angiography (PRESERVE) trial is a randomized, double-blind, multicenter trial that will enroll 8680 high-risk patients undergoing coronary or noncoronary angiography to compare the effectiveness of IV isotonic sodium bicarbonate versus IV isotonic sodium chloride and oral NAC versus oral placebo for the prevention of serious, adverse outcomes associated with CI-AKI. This article discusses key methodological issues of past trials investigating IV fluids and NAC and how they informed the design of the PRESERVE trial. © 2013 by the American Society of Nephrology.


Vollset S.E.,Norwegian Institute of Public Health | Clarke R.,University of Oxford | Lewington S.,University of Oxford | Ebbing M.,University of Bergen | And 19 more authors.
The Lancet | Year: 2013

Background Some countries fortify flour with folic acid to prevent neural tube defects but others do not, partly because of concerns about possible cancer risks. We aimed to assess any effects on site-specific cancer rates in the randomised trials of folic acid supplementation, at doses higher than those from fortification. Methods In these meta-analyses, we sought all trials completed before 2011 that compared folic acid versus placebo, had scheduled treatment duration at least 1 year, included at least 500 participants, and recorded data on cancer incidence. We obtained individual participant datasets that included 49 621 participants in all 13 such trials (ten trials of folic acid for prevention of cardiovascular disease [n=46 969] and three trials in patients with colorectal adenoma [n=2652]). All these trials were evenly randomised. The main outcome was incident cancer (ignoring non-melanoma skin cancer) during the scheduled treatment period (among participants who were still free of cancer). We compared those allocated folic acid with those allocated placebo, and used log-rank analyses to calculate the cancer incidence rate ratio (RR). Findings During a weighted average scheduled treatment duration of 5•2 years, allocation to folic acid quadrupled plasma concentrations of folic acid (57•3 nmol/L for the folic acid groups vs 13•5 nmol/L for the placebo groups), but had no significant effect on overall cancer incidence (1904 cancers in the folic acid groups vs 1809 cancers in the placebo groups, RR 1•06, 95% CI 0•99-1•13, p=0•10). There was no trend towards greater effect with longer treatment. There was no significant heterogeneity between the results of the 13 individual trials (p=0•23), or between the two overall results in the cadiovascular prevention trials and the adenoma trials (p=0•13). Moreover, there was no significant effect of folic acid supplementation on the incidence of cancer of the large intestine, prostate, lung, breast, or any other specific site. Interpretation Folic acid supplementation does not substantially increase or decrease incidence of site-specific cancer during the first 5 years of treatment. Fortification of flour and other cereal products involves doses of folic acid that are, on average, an order of magnitude smaller than the doses used in these trials. Funding British Heart Foundation, Medical Research Council, Cancer Research UK, Food Standards Agency.


Garla V.N.,Yale University | Brandt C.,Yale University | Brandt C.,Connecticut VA Healthcare System
BMC Bioinformatics | Year: 2012

Background: Semantic similarity measures estimate the similarity between concepts, and play an important role in many text processing tasks. Approaches to semantic similarity in the biomedical domain can be roughly divided into knowledge based and distributional based methods. Knowledge based approaches utilize knowledge sources such as dictionaries, taxonomies, and semantic networks, and include path finding measures and intrinsic information content (IC) measures. Distributional measures utilize, in addition to a knowledge source, the distribution of concepts within a corpus to compute similarity; these include corpus IC and context vector methods. Prior evaluations of these measures in the biomedical domain showed that distributional measures outperform knowledge based path finding methods; but more recent studies suggested that intrinsic IC based measures exceed the accuracy of distributional approaches. Limitations of previous evaluations of similarity measures in the biomedical domain include their focus on the SNOMED CT ontology, and their reliance on small benchmarks not powered to detect significant differences between measure accuracy. There have been few evaluations of the relative performance of these measures on other biomedical knowledge sources such as the UMLS, and on larger, recently developed semantic similarity benchmarks.Results: We evaluated knowledge based and corpus IC based semantic similarity measures derived from SNOMED CT, MeSH, and the UMLS on recently developed semantic similarity benchmarks. Semantic similarity measures based on the UMLS, which contains SNOMED CT and MeSH, significantly outperformed those based solely on SNOMED CT or MeSH across evaluations. Intrinsic IC based measures significantly outperformed path-based and distributional measures. We released all code required to reproduce our results and all tools developed as part of this study as open source, available under http://code.google.com/p/ytex. We provide a publicly-accessible web service to compute semantic similarity, available under http://informatics.med.yale.edu/ytex.web/.Conclusions: Knowledge based semantic similarity measures are more practical to compute than distributional measures, as they do not require an external corpus. Furthermore, knowledge based measures significantly and meaningfully outperformed distributional measures on large semantic similarity benchmarks, suggesting that they are a practical alternative to distributional measures. Future evaluations of semantic similarity measures should utilize benchmarks powered to detect significant differences in measure accuracy. © 2012 Garla and Brandt; licensee BioMed Central Ltd.


Tsai J.,Indiana University – Purdue University Indianapolis | Tsai J.,Connecticut VA Healthcare System
Journal of Dual Diagnosis | Year: 2010

The current study examined whether individuals with dual diagnoses in different types of housing experience different levels of hope and whether hope is related to certain housing characteristics. A total of 87 participants (65 in residential programs and 22 in independent apartments) responded to questionnaires about hope and current housing arrangements. Hope did not vary by housing type or housing characteristics. Clients in group housing may have as much hope as clients in apartments. Replication and future study is needed to better understand the relationship between housing and hope. Copyright © Taylor & Francis Group, LLC.


Garla V.N.,Yale University | Brandt C.,Connecticut VA Healthcare System
Proceedings - 2012 IEEE 2nd Conference on Healthcare Informatics, Imaging and Systems Biology, HISB 2012 | Year: 2012

Motivation: Word Sense Disambiguation (WSD) methods automatically assign an unambiguous concept to an ambiguous term based on context, and are important to many text processing tasks. In this study, we developed and evaluated a knowledge-based WSD method that uses semantic similarity measures derived from the Unified Medical Language System (UMLS), and we evaluated the contribution of WSD to clinical text classification. Results: We evaluated our system on biomedical WSD datasets; our system compares favorably to other knowledge-based methods. We evaluated the contribution of our WSD system to clinical document classification on the 2007 Computational Medicine Challenge corpus. Machine learning classifiers trained on disambiguated concepts significantly outperformed those trained using all concepts. Availability: We integrated our WSD system with MetaMap and cTAKES, two popular biomedical natural language processing systems. We released all code required to reproduce our results and all tools developed as part of this study as open source, available under http://code.google.com/p/ytex. © 2012 IEEE.


Garla V.N.,Yale University | Brandt C.,Yale University | Brandt C.,Connecticut VA Healthcare System
Journal of the American Medical Informatics Association | Year: 2013

Background: Word sense disambiguation (WSD) methods automatically assign an unambiguous concept to an ambiguous term based on context, and are important to many text-processing tasks. In this study we developed and evaluated a knowledge-based WSD method that uses semantic similarity measures derived from the Unified Medical Language System (UMLS) and evaluated the contribution of WSD to clinical text classification. Methods: We evaluated our system on biomedical WSD datasets and determined the contribution of our WSD system to clinical document classification on the 2007 Computational Medicine Challenge corpus. Results: Our system compared favorably with other knowledge-based methods. Machine learning classifiers trained on disambiguated concepts significantly outperformed those trained using all concepts. Conclusions: We developed a WSD system that achieves high disambiguation accuracy on standard biomedical WSD datasets and showed that our WSD system improves clinical document classification. Data sharing: We integrated our WSD system with MetaMap and the clinical Text Analysis and Knowledge Extraction System, two popular biomedical natural language processing systems. All codes required to reproduce our results and all tools developed as part of this study are released as open source, available under http://code.google.com/p/ytex.


Garla V.N.,Yale University | Brandt C.,Connecticut VA Healthcare System | Brandt C.,Yale University
Journal of Biomedical Informatics | Year: 2012

In this study we present novel feature engineering techniques that leverage the biomedical domain knowledge encoded in the Unified Medical Language System (UMLS) to improve machine-learning based clinical text classification. Critical steps in clinical text classification include identification of features and passages relevant to the classification task, and representation of clinical text to enable discrimination between documents of different classes. We developed novel information-theoretic techniques that utilize the taxonomical structure of the Unified Medical Language System (UMLS) to improve feature ranking, and we developed a semantic similarity measure that projects clinical text into a feature space that improves classification. We evaluated these methods on the 2008 Integrating Informatics with Biology and the Bedside (I2B2) obesity challenge. The methods we developed improve upon the results of this challenge's top machine-learning based system, and may improve the performance of other machine-learning based clinical text classification systems. We have released all tools developed as part of this study as open source, available at http://code.google.com/p/ytex. © 2012 Elsevier Inc..


Garla V.,Yale University | Taylor C.,Connecticut VA Healthcare System | Brandt C.,Yale University | Brandt C.,Connecticut VA Healthcare System
Journal of Biomedical Informatics | Year: 2013

Objective: To compare linear and Laplacian SVMs on a clinical text classification task; to evaluate the effect of unlabeled training data on Laplacian SVM performance. Background: The development of machine-learning based clinical text classifiers requires the creation of labeled training data, obtained via manual review by clinicians. Due to the effort and expense involved in labeling data, training data sets in the clinical domain are of limited size. In contrast, electronic medical record (EMR) systems contain hundreds of thousands of unlabeled notes that are not used by supervised machine learning approaches. Semi-supervised learning algorithms use both labeled and unlabeled data to train classifiers, and can outperform their supervised counterparts. Methods: We trained support vector machines (SVMs) and Laplacian SVMs on a training reference standard of 820 abdominal CT, MRI, and ultrasound reports labeled for the presence of potentially malignant liver lesions that require follow up (positive class prevalence 77%). The Laplacian SVM used 19,845 randomly sampled unlabeled notes in addition to the training reference standard. We evaluated SVMs and Laplacian SVMs on a test set of 520 labeled reports. Results: The Laplacian SVM trained on labeled and unlabeled radiology reports significantly outperformed supervised SVMs (Macro-F1 0.773 vs. 0.741, Sensitivity 0.943 vs. 0.911, Positive Predictive value 0.877 vs. 0.883). Performance improved with the number of labeled and unlabeled notes used to train the Laplacian SVM (pearson's ρ= 0.529 for correlation between number of unlabeled notes and macro-F1 score). These results suggest that practical semi-supervised methods such as the Laplacian SVM can leverage the large, unlabeled corpora that reside within EMRs to improve clinical text classification. © 2013 Elsevier Inc.

Loading Connecticut VA Healthcare System collaborators
Loading Connecticut VA Healthcare System collaborators