Under-specification as the source of ambiguity and vagueness in narrative phenotype algorithm definitions.

INTRODUCTION: Currently, one of the commonly used methods for disseminating electronic health record (EHR)-based phenotype algorithms is providing a narrative description of the algorithm logic, often accompanied by flowcharts. A challenge with this mode of dissemination is the potential for under-specification in the algorithm definition, which leads to ambiguity and vagueness. METHODS: This study examines incidents of under-specification that occurred during the implementation of 34 narrative phenotyping algorithms in the electronic Medical Record and Genomics (eMERGE) network. We reviewed the online communication history between algorithm developers and implementers within the Phenotype Knowledge Base (PheKB) platform, where questions could be raised and answered regarding the intended implementation of a phenotype algorithm. RESULTS: We developed a taxonomy of under-specification categories via an iterative review process between two groups of annotators. Under-specifications that lead to ambiguity and vagueness were consistently found across narrative phenotype algorithms developed by all involved eMERGE sites. DISCUSSION AND CONCLUSION: Our findings highlight that under-specification is an impediment to the accuracy and efficiency of the implementation of current narrative phenotyping algorithms, and we propose approaches for mitigating these issues and improved methods for disseminating EHR phenotyping algorithms.

Unfolding of hidden white blood cell count phenotypes for gene discovery using latent class mixed modeling.

Resting-state white blood cell (WBC) count is a marker of inflammation and immune system health. There is evidence that WBC count is not fixed over time and there is heterogeneity in WBC trajectory that is associated with morbidity and mortality. Latent class mixed modeling (LCMM) is a method that can identify unobserved heterogeneity in longitudinal data and attempts to classify individuals into groups based on a linear model of repeated measurements. We applied LCMM to repeated WBC count measures derived from electronic medical records of participants of the National Human Genetics Research Institute (NHRGI) electronic MEdical Record and GEnomics (eMERGE) network study, revealing two WBC count trajectory phenotypes. Advancing these phenotypes to GWAS, we found genetic associations between trajectory class membership and regions on chromosome 1p34.3 and chromosome 11q13.4. The chromosome 1 region contains CSF3R, which encodes the granulocyte colony-stimulating factor receptor. This protein is a major factor in neutrophil stimulation and proliferation. The association on chromosome 11 contain genes RNF169 and XRRA1; both involved in the regulation of double-strand break DNA repair.

Probing the Virtual Proteome to Identify Novel Disease Biomarkers.

BACKGROUND: Proteomic approaches allow measurement of thousands of proteins in a single specimen, which can accelerate biomarker discovery. However, applying these technologies to massive biobanks is not currently feasible because of the practical barriers and costs of implementing such assays at scale. To overcome these challenges, we used a "virtual proteomic" approach, linking genetically predicted protein levels to clinical diagnoses in >40 000 individuals. METHODS: We used genome-wide association data from the Framingham Heart Study (n=759) to construct genetic predictors for 1129 plasma protein levels. We validated the genetic predictors for 268 proteins and used them to compute predicted protein levels in 41 288 genotyped individuals in the Electronic Medical Records and Genomics (eMERGE) cohort. We tested associations for each predicted protein with 1128 clinical phenotypes. Lead associations were validated with directly measured protein levels and either low-density lipoprotein cholesterol or subclinical atherosclerosis in the MDCS (Malmö Diet and Cancer Study; n=651). RESULTS: In the virtual proteomic analysis in eMERGE, 55 proteins were associated with 89 distinct diagnoses at a false discovery rate q<0.1. Among these, 13 associations involved lipid (n=7) or atherosclerosis (n=6) phenotypes. We tested each association for validation in MDCS using directly measured protein levels. At Bonferroni-adjusted significance thresholds, levels of apolipoprotein E isoforms were associated with hyperlipidemia, and circulating C-type lectin domain family 1 member B and platelet-derived growth factor receptor-ß predicted subclinical atherosclerosis. Odds ratios for carotid atherosclerosis were 1.31 (95% CI, 1.08-1.58; P=0.006) per 1-SD increment in C-type lectin domain family 1 member B and 0.79 (0.66-0.94; P=0.008) per 1-SD increment in platelet-derived growth factor receptor-ß. CONCLUSIONS: We demonstrate a biomarker discovery paradigm to identify candidate biomarkers of cardiovascular and other diseases.

LPA Variants Are Associated With Residual Cardiovascular Risk in Patients Receiving Statins.

BACKGROUND: Coronary heart disease (CHD) is a leading cause of death globally. Although therapy with statins decreases circulating levels of low-density lipoprotein cholesterol and the incidence of CHD, additional events occur despite statin therapy in some individuals. The genetic determinants of this residual cardiovascular risk remain unknown. METHODS: We performed a 2-stage genome-wide association study of CHD events during statin therapy. We first identified 3099 cases who experienced CHD events (defined as acute myocardial infarction or the need for coronary revascularization) during statin therapy and 7681 controls without CHD events during comparable intensity and duration of statin therapy from 4 sites in the Electronic Medical Records and Genomics Network. We then sought replication of candidate variants in another 160 cases and 1112 controls from a fifth Electronic Medical Records and Genomics site, which joined the network after the initial genome-wide association study. Finally, we performed a phenome-wide association study for other traits linked to the most significant locus. RESULTS: The meta-analysis identified 7 single nucleotide polymorphisms at a genome-wide level of significance within the LPA/PLG locus associated with CHD events on statin treatment. The most significant association was for an intronic single nucleotide polymorphism within LPA/PLG (rs10455872; minor allele frequency, 0.069; odds ratio, 1.58; 95% confidence interval, 1.35-1.86; P=2.6×10-10). In the replication cohort, rs10455872 was also associated with CHD events (odds ratio, 1.71; 95% confidence interval, 1.14-2.57; P=0.009). The association of this single nucleotide polymorphism with CHD events was independent of statin-induced change in low-density lipoprotein cholesterol (odds ratio, 1.62; 95% confidence interval, 1.17-2.24; P=0.004) and persisted in individuals with low-density lipoprotein cholesterol ≤70 mg/dL (odds ratio, 2.43; 95% confidence interval, 1.18-4.75; P=0.015). A phenome-wide association study supported the effect of this region on coronary heart disease and did not identify noncardiovascular phenotypes. CONCLUSIONS: Genetic variations at the LPA locus are associated with CHD events during statin therapy independently of the extent of low-density lipoprotein cholesterol lowering. This finding provides support for exploring strategies targeting circulating concentrations of lipoprotein(a) to reduce CHD events in patients receiving statins.

Meta-Analysis of Genome-Wide Association Studies for Abdominal Aortic Aneurysm Identifies Four New Disease-Specific Risk Loci.

RATIONALE: Abdominal aortic aneurysm (AAA) is a complex disease with both genetic and environmental risk factors. Together, 6 previously identified risk loci only explain a small proportion of the heritability of AAA. OBJECTIVE: To identify additional AAA risk loci using data from all available genome-wide association studies. METHODS AND RESULTS: Through a meta-analysis of 6 genome-wide association study data sets and a validation study totaling 10 204 cases and 107 766 controls, we identified 4 new AAA risk loci: 1q32.3 (SMYD2), 13q12.11 (LINC00540), 20q13.12 (near PCIF1/MMP9/ZNF335), and 21q22.2 (ERG). In various database searches, we observed no new associations between the lead AAA single nucleotide polymorphisms and coronary artery disease, blood pressure, lipids, or diabetes mellitus. Network analyses identified ERG, IL6R, and LDLR as modifiers of MMP9, with a direct interaction between ERG and MMP9. CONCLUSIONS: The 4 new risk loci for AAA seem to be specific for AAA compared with other cardiovascular diseases and related traits suggesting that traditional cardiovascular risk factor management may only have limited value in preventing the progression of aneurysmal disease.

Facilitating phenotype transfer using a common data model.

BACKGROUND: Implementing clinical phenotypes across a network is labor intensive and potentially error prone. Use of a common data model may facilitate the process. METHODS: Electronic Medical Records and Genomics (eMERGE) sites implemented the Observational Health Data Sciences and Informatics (OHDSI) Observational Medical Outcomes Partnership (OMOP) Common Data Model across their electronic health record (EHR)-linked DNA biobanks. Two previously implemented eMERGE phenotypes were converted to OMOP and implemented across the network. RESULTS: It was feasible to implement the common data model across sites, with laboratory data producing the greatest challenge due to local encoding. Sites were then able to execute the OMOP phenotype in less than one day, as opposed to weeks of effort to manually implement an eMERGE phenotype in their bespoke research EHR databases. Of the sites that could compare the current OMOP phenotype implementation with the original eMERGE phenotype implementation, specific agreement ranged from 100% to 43%, with disagreements due to the original phenotype, the OMOP phenotype, changes in data, and issues in the databases. Using the OMOP query as a standard comparison revealed differences in the original implementations despite starting from the same definitions, code lists, flowcharts, and pseudocode. CONCLUSION: Using a common data model can dramatically speed phenotype implementation at the cost of having to populate that data model, though this will produce a net benefit as the number of phenotype implementations increases. Inconsistencies among the implementations of the original queries point to a potential benefit of using a common data model so that actual phenotype code and logic can be shared, mitigating human error in reinterpretation of a narrative phenotype definition.

Epistatic Gene-Based Interaction Analyses for Glaucoma in eMERGE and NEIGHBOR Consortium.

Primary open angle glaucoma (POAG) is a complex disease and is one of the major leading causes of blindness worldwide. Genome-wide association studies have successfully identified several common variants associated with glaucoma; however, most of these variants only explain a small proportion of the genetic risk. Apart from the standard approach to identify main effects of variants across the genome, it is believed that gene-gene interactions can help elucidate part of the missing heritability by allowing for the test of interactions between genetic variants to mimic the complex nature of biology. To explain the etiology of glaucoma, we first performed a genome-wide association study (GWAS) on glaucoma case-control samples obtained from electronic medical records (EMR) to establish the utility of EMR data in detecting non-spurious and relevant associations; this analysis was aimed at confirming already known associations with glaucoma and validating the EMR derived glaucoma phenotype. Our findings from GWAS suggest consistent evidence of several known associations in POAG. We then performed an interaction analysis for variants found to be marginally associated with glaucoma (SNPs with main effect p-value <0.01) and observed interesting findings in the electronic MEdical Records and GEnomics Network (eMERGE) network dataset. Genes from the top epistatic interactions from eMERGE data (Likelihood Ratio Test i.e. LRT p-value <1e-05) were then tested for replication in the NEIGHBOR consortium dataset. To replicate our findings, we performed a gene-based SNP-SNP interaction analysis in NEIGHBOR and observed significant gene-gene interactions (p-value <0.001) among the top 17 gene-gene models identified in the discovery phase. Variants from gene-gene interaction analysis that we found to be associated with POAG explain 3.5% of additional genetic variance in eMERGE dataset above what is explained by the SNPs in genes that are replicated from previous GWAS studies (which was only 2.1% variance explained in eMERGE dataset); in the NEIGHBOR dataset, adding replicated SNPs from gene-gene interaction analysis explain 3.4% of total variance whereas GWAS SNPs alone explain only 2.8% of variance. Exploring gene-gene interactions may provide additional insights into many complex traits when explored in properly designed and powered association studies.

Identification of Four Novel Loci in Asthma in European American and African American Populations.

RATIONALE: Despite significant advances in knowledge of the genetic architecture of asthma, specific contributors to the variability in the burden between populations remain uncovered. OBJECTIVES: To identify additional genetic susceptibility factors of asthma in European American and African American populations. METHODS: A phenotyping algorithm mining electronic medical records was developed and validated to recruit cases with asthma and control subjects from the Electronic Medical Records and Genomics network. Genome-wide association analyses were performed in pediatric and adult asthma cases and control subjects with European American and African American ancestry followed by metaanalysis. Nominally significant results were reanalyzed conditioning on allergy status. MEASUREMENTS AND MAIN RESULTS: The validation of the algorithm yielded an average of 95.8% positive predictive values for both cases and control subjects. The algorithm accrued 21,644 subjects (65.83% European American and 34.17% African American). We identified four novel population-specific associations with asthma after metaanalyses: loci 6p21.31, 9p21.2, and 10q21.3 in the European American population, and the PTGES gene in African Americans. TEK at 9p21.2, which encodes TIE2, has been shown to be involved in remodeling the airway wall in asthma, and the association remained significant after conditioning by allergy. PTGES, which encodes the prostaglandin E synthase, has also been linked to asthma, where deficient prostaglandin E2 synthesis has been associated with airway remodeling. CONCLUSIONS: This study adds to understanding of the genetic architecture of asthma in European Americans and African Americans and reinforces the need to study populations of diverse ethnic backgrounds to identify shared and unique genetic predictors of asthma.

Performance of an electronic health record-based phenotype algorithm to identify community associated methicillin-resistant Staphylococcus aureus cases and controls for genetic association studies.

BACKGROUND: Community associated methicillin-resistant Staphylococcus aureus (CA-MRSA) is one of the most common causes of skin and soft tissue infections in the United States, and a variety of genetic host factors are suspected to be risk factors for recurrent infection. Based on the CDC definition, we have developed and validated an electronic health record (EHR) based CA-MRSA phenotype algorithm utilizing both structured and unstructured data. METHODS: The algorithm was validated at three eMERGE consortium sites, and positive predictive value, negative predictive value and sensitivity, were calculated. The algorithm was then run and data collected across seven total sites. The resulting data was used in GWAS analysis. RESULTS: Across seven sites, the CA-MRSA phenotype algorithm identified a total of 349 cases and 7761 controls among the genotyped European and African American biobank populations. PPV ranged from 68 to 100% for cases and 96 to 100% for controls; sensitivity ranged from 94 to 100% for cases and 75 to 100% for controls. Frequency of cases in the populations varied widely by site. There were no plausible GWAS-significant (p < 5 E -8) findings. CONCLUSIONS: Differences in EHR data representation and screening patterns across sites may have affected identification of cases and controls and accounted for varying frequencies across sites. Future work identifying these patterns is necessary.

SeqHBase: a big data toolset for family based sequencing data analysis.

BACKGROUND: Whole-genome sequencing (WGS) and whole-exome sequencing (WES) technologies are increasingly used to identify disease-contributing mutations in human genomic studies. It can be a significant challenge to process such data, especially when a large family or cohort is sequenced. Our objective was to develop a big data toolset to efficiently manipulate genome-wide variants, functional annotations and coverage, together with conducting family based sequencing data analysis. METHODS: Hadoop is a framework for reliable, scalable, distributed processing of large data sets using MapReduce programming models. Based on Hadoop and HBase, we developed SeqHBase, a big data-based toolset for analysing family based sequencing data to detect de novo, inherited homozygous, or compound heterozygous mutations that may contribute to disease manifestations. SeqHBase takes as input BAM files (for coverage at every site), variant call format (VCF) files (for variant calls) and functional annotations (for variant prioritisation). RESULTS: We applied SeqHBase to a 5-member nuclear family and a 10-member 3-generation family with WGS data, as well as a 4-member nuclear family with WES data. Analysis times were almost linearly scalable with number of data nodes. With 20 data nodes, SeqHBase took about 5 secs to analyse WES familial data and approximately 1 min to analyse WGS familial data. CONCLUSIONS: These results demonstrate SeqHBase's high efficiency and scalability, which is necessary as WGS and WES are rapidly becoming standard methods to study the genetics of familial disorders.

Relational machine learning for electronic health record-driven phenotyping.

OBJECTIVE: Electronic health records (EHR) offer medical and pharmacogenomics research unprecedented opportunities to identify and classify patients at risk. EHRs are collections of highly inter-dependent records that include biological, anatomical, physiological, and behavioral observations. They comprise a patient's clinical phenome, where each patient has thousands of date-stamped records distributed across many relational tables. Development of EHR computer-based phenotyping algorithms require time and medical insight from clinical experts, who most often can only review a small patient subset representative of the total EHR records, to identify phenotype features. In this research we evaluate whether relational machine learning (ML) using inductive logic programming (ILP) can contribute to addressing these issues as a viable approach for EHR-based phenotyping. METHODS: Two relational learning ILP approaches and three well-known WEKA (Waikato Environment for Knowledge Analysis) implementations of non-relational approaches (PART, J48, and JRIP) were used to develop models for nine phenotypes. International Classification of Diseases, Ninth Revision (ICD-9) coded EHR data were used to select training cohorts for the development of each phenotypic model. Accuracy, precision, recall, F-Measure, and Area Under the Receiver Operating Characteristic (AUROC) curve statistics were measured for each phenotypic model based on independent manually verified test cohorts. A two-sided binomial distribution test (sign test) compared the five ML approaches across phenotypes for statistical significance. RESULTS: We developed an approach to automatically label training examples using ICD-9 diagnosis codes for the ML approaches being evaluated. Nine phenotypic models for each ML approach were evaluated, resulting in better overall model performance in AUROC using ILP when compared to PART (p=0.039), J48 (p=0.003) and JRIP (p=0.003). DISCUSSION: ILP has the potential to improve phenotyping by independently delivering clinically expert interpretable rules for phenotype definitions, or intuitive phenotypes to assist experts. CONCLUSION: Relational learning using ILP offers a viable approach to EHR-driven phenotyping.

Design patterns for the development of electronic health record-driven phenotype extraction algorithms.

BACKGROUND: Design patterns, in the context of software development and ontologies, provide generalized approaches and guidance to solving commonly occurring problems, or addressing common situations typically informed by intuition, heuristics and experience. While the biomedical literature contains broad coverage of specific phenotype algorithm implementations, no work to date has attempted to generalize common approaches into design patterns, which may then be distributed to the informatics community to efficiently develop more accurate phenotype algorithms. METHODS: Using phenotyping algorithms stored in the Phenotype KnowledgeBase (PheKB), we conducted an independent iterative review to identify recurrent elements within the algorithm definitions. We extracted and generalized recurrent elements in these algorithms into candidate patterns. The authors then assessed the candidate patterns for validity by group consensus, and annotated them with attributes. RESULTS: A total of 24 electronic Medical Records and Genomics (eMERGE) phenotypes available in PheKB as of 1/25/2013 were downloaded and reviewed. From these, a total of 21 phenotyping patterns were identified, which are available as an online data supplement. CONCLUSIONS: Repeatable patterns within phenotyping algorithms exist, and when codified and cataloged may help to educate both experienced and novice algorithm developers. The dissemination and application of these patterns has the potential to decrease the time to develop algorithms, while improving portability and accuracy.

Practical challenges in integrating genomic data into the electronic health record.

Genetic testing has had limited impact on routine clinical care. Widespread adoption of electronic health records presents a promising means of disseminating genetic testing into diverse care settings. Practical challenges to integration of genomic data into electronic health records include size and complexity of genetic test results, inadequate use of standards for clinical and genetic data, and limitations in electronic health record capacity to store and analyze genetic data. Related challenges include uncertainty in the interpretation of regulatory requirements for return of results, and privacy concerns specific to genetic testing. Successful integration of genomic data may require significant redesign of existing electronic health record systems.

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.

Ambiguous and Incomplete: Natural Language Processing Reveals Problematic Reporting Styles in Thyroid Ultrasound Reports.

OBJECTIVE: Natural language processing (NLP) systems convert unstructured text into analyzable data. Here, we describe the performance measures of NLP to capture granular details on nodules from thyroid ultrasound (US) reports and reveal critical issues with reporting language. METHODS: We iteratively developed NLP tools using clinical Text Analysis and Knowledge Extraction System (cTAKES) and thyroid US reports from 2007 to 2013. We incorporated nine nodule features for NLP extraction. Next, we evaluated the precision, recall, and accuracy of our NLP tools using a separate set of US reports from an academic medical center (A) and a regional health care system (B) during the same period. Two physicians manually annotated each test-set report. A third physician then adjudicated discrepancies. The adjudicated "gold standard" was then used to evaluate NLP performance on the test-set. RESULTS: A total of 243 thyroid US reports contained 6,405 data elements. Inter-annotator agreement for all elements was 91.3%. Compared with the gold standard, overall recall of the NLP tool was 90%. NLP recall for thyroid lobe or isthmus characteristics was: laterality 96% and size 95%. NLP accuracy for nodule characteristics was: laterality 92%, size 92%, calcifications 76%, vascularity 65%, echogenicity 62%, contents 76%, and borders 40%. NLP recall for presence or absence of lymphadenopathy was 61%. Reporting style accounted for 18% errors. For example, the word "heterogeneous" interchangeably referred to nodule contents or echogenicity. While nodule dimensions and laterality were often described, US reports only described contents, echogenicity, vascularity, calcifications, borders, and lymphadenopathy, 46, 41, 17, 15, 9, and 41% of the time, respectively. Most nodule characteristics were equally likely to be described at hospital A compared with hospital B. CONCLUSIONS: NLP can automate extraction of critical information from thyroid US reports. However, ambiguous and incomplete reporting language hinders performance of NLP systems regardless of institutional setting. Standardized or synoptic thyroid US reports could improve NLP performance.

Cataract research using electronic health records.

BACKGROUND: The eMERGE (electronic MEdical Records and Genomics) network, funded by the National Human Genome Research Institute, is a national consortium formed to develop, disseminate, and apply approaches to research that combine DNA biorepositories with electronic health record (EHR) systems for large-scale, high-throughput genetic research. Marshfield Clinic is one of five sites in the eMERGE network and primarily studied: 1) age-related cataract and 2) HDL-cholesterol levels. The purpose of this paper is to describe the approach to electronic evaluation of the epidemiology of cataract using the EHR for a large biobank and to assess previously identified epidemiologic risk factors in cases identified by electronic algorithms. METHODS: Electronic algorithms were used to select individuals with cataracts in the Personalized Medicine Research Project database. These were analyzed for cataract prevalence, age at cataract, and previously identified risk factors. RESULTS: Cataract diagnoses and surgeries, though not type of cataract, were successfully identified using electronic algorithms. Age specific prevalence of both cataract (22% compared to 17.2%) and cataract surgery (11% compared to 5.1%) were higher when compared to the Eye Diseases Prevalence Research Group. The risk factors of age, gender, diabetes, and steroid use were confirmed. CONCLUSIONS: Using electronic health records can be a viable and efficient tool to identify cataracts for research. However, using retrospective data from this source can be confounded by historical limits on data availability, differences in the utilization of healthcare, and changes in exposures over time.

Medical records-based chronic kidney disease phenotype for clinical care and "big data" observational and genetic studies.

Chronic Kidney Disease (CKD) represents a slowly progressive disorder that is typically silent until late stages, but early intervention can significantly delay its progression. We designed a portable and scalable electronic CKD phenotype to facilitate early disease recognition and empower large-scale observational and genetic studies of kidney traits. The algorithm uses a combination of rule-based and machine-learning methods to automatically place patients on the staging grid of albuminuria by glomerular filtration rate ("A-by-G" grid). We manually validated the algorithm by 451 chart reviews across three medical systems, demonstrating overall positive predictive value of 95% for CKD cases and 97% for healthy controls. Independent case-control validation using 2350 patient records demonstrated diagnostic specificity of 97% and sensitivity of 87%. Application of the phenotype to 1.3 million patients demonstrated that over 80% of CKD cases are undetected using ICD codes alone. We also demonstrated several large-scale applications of the phenotype, including identifying stage-specific kidney disease comorbidities, in silico estimation of kidney trait heritability in thousands of pedigrees reconstructed from medical records, and biobank-based multicenter genome-wide and phenome-wide association studies.

Pharmacogenomic clinical decision support design and multi-site process outcomes analysis in the eMERGE Network.

To better understand the real-world effects of pharmacogenomic (PGx) alerts, this study aimed to characterize alert design within the eMERGE Network, and to establish a method for sharing PGx alert response data for aggregate analysis. Seven eMERGE sites submitted design details and established an alert logging data dictionary. Six sites participated in a pilot study, sharing alert response data from their electronic health record systems. PGx alert design varied, with some consensus around the use of active, post-test alerts to convey Clinical Pharmacogenetics Implementation Consortium recommendations. Sites successfully shared response data, with wide variation in acceptance and follow rates. Results reflect the lack of standardization in PGx alert design. Standards and/or larger studies will be necessary to fully understand PGx impact. This study demonstrated a method for sharing PGx alert response data and established that variation in system design is a significant barrier for multi-site analyses.

Machine Learning Assisted Discovery of Novel Predictive Lab Tests Using Electronic Health Record Data.

Epidemiological studies identifying biological markers of disease state are valuable, but can be time-consuming, expensive, and require extensive intuition and expertise. Furthermore, not all hypothesized markers will be borne out in a study, suggesting that higher quality initial hypotheses are crucial. In this work, we propose a high-throughput pipeline to produce a ranked list of high-quality hypothesized marker laboratory tests for diagnoses. Our pipeline generates a large number of candidate lab-diagnosis hypotheses derived from machine learning models, filters and ranks them according to their potential novelty using text mining, and corroborate final hypotheses with logistic regression analysis. We test our approach on a large electronic health record dataset and the PubMed corpus, and find several promising candidate hypotheses.