Multi-ancestry genome- and phenome-wide association studies of diverticular disease in electronic health records with natural language processing enriched phenotyping algorithm.

OBJECTIVE: Diverticular disease (DD) is one of the most prevalent conditions encountered by gastroenterologists, affecting ~50% of Americans before the age of 60. Our aim was to identify genetic risk variants and clinical phenotypes associated with DD, leveraging multiple electronic health record (EHR) data sources of 91,166 multi-ancestry participants with a Natural Language Processing (NLP) technique. MATERIALS AND METHODS: We developed a NLP-enriched phenotyping algorithm that incorporated colonoscopy or abdominal imaging reports to identify patients with diverticulosis and diverticulitis from multicenter EHRs. We performed genome-wide association studies (GWAS) of DD in European, African and multi-ancestry participants, followed by phenome-wide association studies (PheWAS) of the risk variants to identify their potential comorbid/pleiotropic effects in clinical phenotypes. RESULTS: Our developed algorithm showed a significant improvement in patient classification performance for DD analysis (algorithm PPVs ≥ 0.94), with up to a 3.5 fold increase in terms of the number of identified patients than the traditional method. Ancestry-stratified analyses of diverticulosis and diverticulitis of the identified subjects replicated the well-established associations between ARHGAP15 loci with DD, showing overall intensified GWAS signals in diverticulitis patients compared to diverticulosis patients. Our PheWAS analyses identified significant associations between the DD GWAS variants and circulatory system, genitourinary, and neoplastic EHR phenotypes. DISCUSSION: As the first multi-ancestry GWAS-PheWAS study, we showcased that heterogenous EHR data can be mapped through an integrative analytical pipeline and reveal significant genotype-phenotype associations with clinical interpretation. CONCLUSION: A systematic framework to process unstructured EHR data with NLP could advance a deep and scalable phenotyping for better patient identification and facilitate etiological investigation of a disease with multilayered data.

A case study evaluating the portability of an executable computable phenotype algorithm across multiple institutions and electronic health record environments.

Electronic health record (EHR) algorithms for defining patient cohorts are commonly shared as free-text descriptions that require human intervention both to interpret and implement. We developed the Phenotype Execution and Modeling Architecture (PhEMA, http://projectphema.org) to author and execute standardized computable phenotype algorithms. With PhEMA, we converted an algorithm for benign prostatic hyperplasia, developed for the electronic Medical Records and Genomics network (eMERGE), into a standards-based computable format. Eight sites (7 within eMERGE) received the computable algorithm, and 6 successfully executed it against local data warehouses and/or i2b2 instances. Blinded random chart review of cases selected by the computable algorithm shows PPV ≥90%, and 3 out of 5 sites had >90% overlap of selected cases when comparing the computable algorithm to their original eMERGE implementation. This case study demonstrates potential use of PhEMA computable representations to automate phenotyping across different EHR systems, but also highlights some ongoing challenges.

Screening Colonoscopy Withdrawal Time Threshold for Adequate Proximal Serrated Polyp Detection Rate.

INTRODUCTION: For adequate adenoma detection rate (ADR), guidelines recommend a mean withdrawal time (MWT) of ≥ 6 min. ADR has been shown to correlate strongly with proximal serrated polyp detection rate (PSP-DR), which is another suggested quality measure for screening colonoscopy. However, the impact of directly measured withdrawal time on PSP-DR has not been rigorously studied. We examined the relationship between MWT to ADR and PSP-DR, with the aim of identifying a functional threshold withdrawal time associated with both increased ADR and PSP-DR. METHODS: This was a retrospective study of endoscopy and pathology data from average-risk screening colonoscopy examinations performed at a large system with six endoscopy laboratories. A natural language processing tool was used to determine polyp location and histology. ADR and PSP-DR were calculated for each endoscopist. MWT was calculated from colonoscopy examinations in which no polyps were resected. RESULTS: In total, 31,558 colonoscopy examinations were performed, of which 10,196 were average-risk screening colonoscopy examinations with cecal intubation and adequate prep by 24 gastroenterologists. When assessing the statistical significance of increasing MWT by minute, the first significant time mark for PSP-DR was at 11 min at a rate of 14.2% (p = 0.01). There was a significant difference comparing aggregated MWT < 11 min compared to ≥ 11 min looking at the rates of adenomas [OR 1.65 (1.09-2.51)] and proximal serrated polyps [OR 1.81 (1.06-3.08)]. While ADR linearly correlated well with MWT (R = 0.76, p < 0.001), the linear relationship with PSP-DR was less robust (R = 0.42, p = 0.043). CONCLUSION: In this large cohort of average-risk screening colonoscopy, a MWT of 11 min resulted in a statistically significant increase in both ADR and PSP-DR. Our data suggest that a longer withdrawal time may be required to meet both quality metrics.

A Framework for Leveraging "Big Data" to Advance Epidemiology and Improve Quality: Design of the VA Colonoscopy Collaborative.

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.

Regional Variability of Repeat Esophagogastroduodenoscopy Use in the National Veteran Population.

BACKGROUND: Esophagogastroduodenoscopy (EGD) procedures are performed frequently to evaluate gastrointestinal disease and symptoms. AIM: To determine regional practice variability of repeat EGDs in a national population. METHODS: The study sample included US Veterans with an outpatient index EGD from 1/1/2008 to 12/2010. We determined risk of repeat endoscopy from 1/2008 to 10/1/2014. A logistic regression model was used to assess the association between the odds of repeated EGD and patient demographics, ICD diagnostic codes, and geographic region. Multivariable logistic regression was performed to obtain the adjusted odds ratio and predicted probabilities of repeat EGDs by region. RESULTS: A total of 202,086 patients had an index endoscopy from 1/2008 to 12/2010. Unique patients with an index endoscopy were predominantly male (93.2%), white (72.8%), and on average 61 years. A total of 58,469 patients (28.9%) had one or more repeat EGDs, accounting for 103,253 repeat procedures through 10/2014. ICD-9-CM codes associated with increased risk of repeat procedures were Barrett's esophagus (OR 3.6, 95% CI 3.5-3.7), dysphagia (OR 1.3, 95% CI 1.2-1.3), ulcer (OR 1.3, 95% CI 2.2-2.4), stricture (OR 1.8, 95% CI 1.7-1.9), and esophageal varices (OR 2.8, 95% CI 2.7-3.0). There was a significant difference in the probability of repeat EGD by VA region, with the Midwest region having the highest probability (31.2%) and Southeast the lowest probability (27.3%). This difference would account for 400 more EGD procedures per 10,000 Veterans, after adjusting for patient demographics and diagnosis codes. CONCLUSIONS: Regional practice variability accounts for a substantial volume of repeat EGD procedures, regardless of patient characteristics and associated diagnoses.

Natural Language Processing for EHR-Based Pharmacovigilance: A Structured Review.

The goal of pharmacovigilance is to detect, monitor, characterize and prevent adverse drug events (ADEs) with pharmaceutical products. This article is a comprehensive structured review of recent advances in applying natural language processing (NLP) to electronic health record (EHR) narratives for pharmacovigilance. We review methods of varying complexity and problem focus, summarize the current state-of-the-art in methodology advancement, discuss limitations and point out several promising future directions. The ability to accurately capture both semantic and syntactic structures in clinical narratives becomes increasingly critical to enable efficient and accurate ADE detection. Significant progress has been made in algorithm development and resource construction since 2000. Since 2012, statistical analysis and machine learning methods have gained traction in automation of ADE mining from EHR narratives. Current state-of-the-art methods for NLP-based ADE detection from EHRs show promise regarding their integration into production pharmacovigilance systems. In addition, integrating multifaceted, heterogeneous data sources has shown promise in improving ADE detection and has become increasingly adopted. On the other hand, challenges and opportunities remain across the frontier of NLP application to EHR-based pharmacovigilance, including proper characterization of ADE context, differentiation between off- and on-label drug-use ADEs, recognition of the importance of polypharmacy-induced ADEs, better integration of heterogeneous data sources, creation of shared corpora, and organization of shared-task challenges to advance the state-of-the-art.

Adverse events after surgery for nonmalignant colon polyps are common and associated with increased length of stay and costs.

BACKGROUND AND AIMS: Endoscopic resection (ER) is a safe and effective treatment for nonmalignant complex colorectal polyps (complex polyps). Surgical resection (SR) remains prevalent despite limited outcomes data. We aimed to evaluate SR outcomes for complex polyps and compare SR outcomes to those of ER. METHODS: We performed a single-center, retrospective, cohort study of all patients undergoing SR (2003-2013) and ER (2011-2013) for complex polyps. We excluded patients with invasive carcinoma from the SR cohort. Primary outcomes were 12-month adverse event (AE) rate, length of stay (LOS), and costs. SR outcomes over a 3-year period (2011-2013) were compared with the overlapping ER cohort. RESULTS: Over the 11-year period, 359 patients (mean [± SD] age 64 ± 11 years) underwent SR (58% laparoscopic) for complex polyps. In total, 17% experienced an AE, and 3% required additional surgery; 12-month mortality was 1%. Including readmissions, median LOS was 5 days (IQR 4-7 days), and costs were $14,528. When an AE occurred, costs ($25,557 vs $14,029; P < .0001) and LOS (11 vs 5 days; P < .0001) significantly increased. From 2011 to 2013, 198 patients were referred for ER, and 73 underwent primary SR (70% laparoscopic). There was a lower AE rate for ER versus primary SR (10% vs 18%; P = .09). ER costs (including rescue SR, when required) were lower than those of primary SR ($2152 vs $15,264; P < .0001). CONCLUSIONS: AEs occur in approximately one-sixth of patients after SR for complex polyps. ER-accounting for rescue SR caused by malignancy, AEs, or incomplete resection-is associated with markedly lower costs than SR. These data should be used when counseling patients about treatment options for complex polyps.

Stereoscopic Integrated Imaging Goggles for Multimodal Intraoperative Image Guidance.

We have developed novel stereoscopic wearable multimodal intraoperative imaging and display systems entitled Integrated Imaging Goggles for guiding surgeries. The prototype systems offer real time stereoscopic fluorescence imaging and color reflectance imaging capacity, along with in vivo handheld microscopy and ultrasound imaging. With the Integrated Imaging Goggle, both wide-field fluorescence imaging and in vivo microscopy are provided. The real time ultrasound images can also be presented in the goggle display. Furthermore, real time goggle-to-goggle stereoscopic video sharing is demonstrated, which can greatly facilitate telemedicine. In this paper, the prototype systems are described, characterized and tested in surgeries in biological tissues ex vivo. We have found that the system can detect fluorescent targets with as low as 60 nM indocyanine green and can resolve structures down to 0.25 mm with large FOV stereoscopic imaging. The system has successfully guided simulated cancer surgeries in chicken. The Integrated Imaging Goggle is novel in 4 aspects: it is (a) the first wearable stereoscopic wide-field intraoperative fluorescence imaging and display system, (b) the first wearable system offering both large FOV and microscopic imaging simultaneously,

Anatomic and advanced adenoma detection rates as quality metrics determined via natural language processing.

OBJECTIVES: The objectives of this study were to use an open-source natural language-processing tool (NLP) to accurately assess total, anatomic (left and right colon), and advanced adenoma detection rates (ADRs) and to determine how these metrics differ between high- and low-performing endoscopists. METHODS: An NLP tool was developed using the Apache Unstructured Information Management Architecture and queried all procedure records for screening colonoscopies performed in patients aged 50-75 years at a single institution from April 1998 to December 2013. Validation was performed on 200 procedures and associated pathology reports. The total, left colon, right colon, and advanced ADRs were calculated and physicians were stratified by total ADR (<20% and ≥20%). Comparisons of colonoscopy characteristics and ADR comparisons (advanced, left, right, and right/left ratio) were determined by t-tests and Wilcoxon rank-sum tests. RESULTS: The total ADR for 34,998 screening colonoscopies from 1998 to 2013 was 20.3%, as determined via NLP. The institutional left and right colon ADRs were 10.1% and 12.5%, respectively. The overall advanced ADR was 4.4%. Endoscopists with total ADRs ≥20% had higher left (12.4%) and right colon (16.4%) ADRs than endoscopists with ADRs <20% (left ADR=5.6%, right ADR=5.8%). Endoscopists with ADRs ≥20% had higher individual right/left ADR ratios than those with low ADRs (1.4 (interquartile range (IQR) 0.4) vs. 1.0 (IQR 0.4), P=0.02). There was a moderate positive correlation between advanced ADR detection and both right (Spearman's rho=0.5, P=0.05) and left colon (Spearman's rho=0.4, P=0.03) ADRs. CONCLUSIONS: Institutions should consider the use of anatomic and advanced ADRs determined via natural language processing as a refined measure of colonoscopy quality. The ability to continuously monitor and provide feedback on colonoscopy quality metrics may encourage endoscopists to refine technique, resulting in overall improvements in adenoma detection.

Comparison of traditional trigger tool to data warehouse based screening for identifying hospital adverse events.

BACKGROUND: Research supports medical record review using screening triggers as the optimal method to detect hospital adverse events (AE), yet the method is labour-intensive. METHOD: This study compared a traditional trigger tool with an enterprise data warehouse (EDW) based screening method to detect AEs. We created 51 automated queries based on 33 traditional triggers from prior research, and then applied them to 250 randomly selected medical patients hospitalised between 1 September 2009 and 31 August 2010. Two physicians each abstracted records from half the patients using a traditional trigger tool and then performed targeted abstractions for patients with positive EDW queries in the complementary half of the sample. A third physician confirmed presence of AEs and assessed preventability and severity. RESULTS: Traditional trigger tool and EDW based screening identified 54 (22%) and 53 (21%) patients with one or more AE. Overall, 140 (56%) patients had one or more positive EDW screens (total 366 positive screens). Of the 137 AEs detected by at least one method, 86 (63%) were detected by a traditional trigger tool, 97 (71%) by EDW based screening and 46 (34%) by both methods. Of the 11 total preventable AEs, 6 (55%) were detected by traditional trigger tool, 7 (64%) by EDW based screening and 2 (18%) by both methods. Of the 43 total serious AEs, 28 (65%) were detected by traditional trigger tool, 29 (67%) by EDW based screening and 14 (33%) by both. CONCLUSIONS: We found relatively poor agreement between traditional trigger tool and EDW based screening with only approximately a third of all AEs detected by both methods. A combination of complementary methods is the optimal approach to detecting AEs among hospitalised patients.