Robert J. Carroll

Evaluating phecodes, clinical classification software, and ICD-9-CM codes for phenome-wide association studies in the electronic health record

Objective To compare three groupings of Electronic Health Record (EHR) billing codes for their ability to represent clinically meaningful phenotypes and to replicate known genetic associations. The three tested coding systems were the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes, the Agency for Healthcare Research and Quality Clinical Classification Software for ICD-9-CM (CCS), and manually curated “phecodes” designed to facilitate phenome-wide association studies (PheWAS) in EHRs. Methods and materials We selected 100 disease phenotypes and compared the ability of each coding system to accurately represent them without performing additional groupings. The 100 phenotypes included 25 randomly-chosen clinical phenotypes pursued in prior genome-wide association studies (GWAS) and another 75 common disease phenotypes mentioned across free-text problem lists from 189,289 individuals. We then evaluated the performance of each coding system to replicate known associations for 440 SNP-phenotype pairs. Results Out of the 100 tested clinical phenotypes, phecodes exactly matched 83, compared to 53 for ICD-9-CM and 32 for CCS. ICD-9-CM codes were typically too detailed (requiring custom groupings) while CCS codes were often not granular enough. Among 440 tested known SNP-phenotype associations, use of phecodes replicated 153 SNP-phenotype pairs compared to 143 for ICD-9-CM and 139 for CCS. Phecodes also generally produced stronger odds ratios and lower p-values for known associations than ICD-9-CM and CCS. Finally, evaluation of several SNPs via PheWAS identified novel potential signals, some seen in only using the phecode approach. Among them, rs7318369 in PEPD was associated with gastrointestinal hemorrhage. Conclusion Our results suggest that the phecode groupings better align with clinical diseases mentioned in clinical practice or for genomic studies. ICD-9-CM, CCS, and phecode groupings all worked for PheWAS-type studies, though the phecode groupings produced superior results.

Clinical Pharmacogenetics of Cytochrome P450-Associated Drugs in Children

Cytochrome P450 (CYP) enzymes are commonly involved in drug metabolism, and genetic variation in the genes encoding CYPs are associated with variable drug response. While genotype-guided therapy has been clinically implemented in adults, these associations are less well established for pediatric patients. In order to understand the frequency of pediatric exposures to drugs with known CYP interactions, we compiled all actionable drug–CYP interactions with a high level of evidence using Clinical Pharmacogenomic Implementation Consortium (CPIC) data and surveyed 10 years of electronic health records (EHR) data for the number of children exposed to CYP-associated drugs. Subsequently, we performed a focused literature review for drugs commonly used in pediatrics, defined as more than 5000 pediatric patients exposed in the decade-long EHR cohort. There were 48 drug–CYP interactions with a high level of evidence in the CPIC database. Of those, only 10 drugs were commonly used in children (ondansetron, oxycodone, codeine, omeprazole, lansoprazole, sertraline, amitriptyline, citalopram, escitalopram, and risperidone). For these drugs, reports of the drug–CYP interaction in cohorts including children were sparse. There are adequate data for implementation of genotype-guided therapy for children for three of the 10 commonly used drugs (codeine, omeprazole and lansoprazole). For the majority of commonly used drugs with known CYP interactions, more data are required to support pharmacogenomic implementation in children.

Phenotype risk scores identify patients with unrecognized Mendelian disease patterns

Hidden effects of Mendelian inheritance Identifying the determinate factors of genetic disease has been quite successful for Mendelian inheritance of large-effect pathogenic variants. In these cases, two non- or low-functioning genes contribute to disease. However, Mendelian effects of lesser strength have generally been ignored when looking at genomic consequences in human health. Bastarache et al. used electronic records to identify the phenotypic effects of previously unidentified Mendelian variations. Their analysis suggests that individuals with undiagnosed Mendelian diseases may be more prevalent in the general population than assumed. Because of this, genetic analysis may be able to assist clinicians in arriving at a diagnosis. Science, this issue p. 1233 Electronic health records coupled with exome sequencing identify disease phenotypes linked to Mendelian inheritance. Genetic association studies often examine features independently, potentially missing subpopulations with multiple phenotypes that share a single cause. We describe an approach that aggregates phenotypes on the basis of patterns described by Mendelian diseases. We mapped the clinical features of 1204 Mendelian diseases into phenotypes captured from the electronic health record (EHR) and summarized this evidence as phenotype risk scores (PheRSs). In an initial validation, PheRS distinguished cases and controls of five Mendelian diseases. Applying PheRS to 21,701 genotyped individuals uncovered 18 associations between rare variants and phenotypes consistent with Mendelian diseases. In 16 patients, the rare genetic variants were associated with severe outcomes such as organ transplants. PheRS can augment rare-variant interpretation and may identify subsets of patients with distinct genetic causes for common diseases.

Developing Electronic Health Record Algorithms that Accurately Identify Patients with Systemic Lupus Erythematosus

To study systemic lupus erythematosus (SLE) in the electronic health record (EHR), we must accurately identify patients with SLE. Our objective was to develop and validate novel EHR algorithms that use International Classification of Diseases, Ninth Revision (ICD‐9), Clinical Modification codes, laboratory testing, and medications to identify SLE patients.

Uncovering exposures responsible for birth season - disease effects: a global study

Abstract Objective Birth month and climate impact lifetime disease risk, while the underlying exposures remain largely elusive. We seek to uncover distal risk factors underlying these relationships by probing the relationship between global exposure variance and disease risk variance by birth season. Material and Methods This study utilizes electronic health record data from 6 sites representing 10.5 million individuals in 3 countries (United States, South Korea, and Taiwan). We obtained birth month–disease risk curves from each site in a case-control manner. Next, we correlated each birth month–disease risk curve with each exposure. A meta-analysis was then performed of correlations across sites. This allowed us to identify the most significant birth month–exposure relationships supported by all 6 sites while adjusting for multiplicity. We also successfully distinguish relative age effects (a cultural effect) from environmental exposures. Results Attention deficit hyperactivity disorder was the only identified relative age association. Our methods identified several culprit exposures that correspond well with the literature in the field. These include a link between first-trimester exposure to carbon monoxide and increased risk of depressive disorder (Ru2009=u20090.725, confidence interval [95% CI], 0.529-0.847), first-trimester exposure to fine air particulates and increased risk of atrial fibrillation (Ru2009=u20090.564, 95% CI, 0.363-0.715), and decreased exposure to sunlight during the third trimester and increased risk of type 2 diabetes mellitus (Ru2009=u2009−0.816, 95% CI, −0.5767, −0.929). Conclusion A global study of birth month–disease relationships reveals distal risk factors involved in causal biological pathways that underlie them.

Phenome-wide association studies uncover a novel association of increased atrial fibrillation in males with systemic lupus erythematosus

Phenome‐wide association studies (PheWAS) scan across billing codes in the electronic health record (EHR) and re‐purpose clinical EHR data for research. In this study, we examined whether PheWAS could function as an EHR‐based discovery tool for systemic lupus erythematosus (SLE) and identified novel clinical associations in male versus female patients with SLE.

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.

Evaluating statistical approaches to leverage large clinical datasets for uncovering therapeutic and adverse medication effects

MotivationnPhenome-wide association studies (PheWAS) have been used to discover many genotype-phenotype relationships and have the potential to identify therapeutic and adverse drug outcomes using longitudinal data within electronic health records (EHRs). However, the statistical methods for PheWAS applied to longitudinal EHR medication data have not been established.nnnResultsnIn this study, we developed methods to address two challenges faced with reuse of EHR for this purpose: confounding by indication, and low exposure and event rates. We used Monte Carlo simulation to assess propensity score (PS) methods, focusing on two of the most commonly used methods, PS matching and PS adjustment, to address confounding by indication. We also compared two logistic regression approaches (the default of Wald versus Firths penalized maximum likelihood, PML) to address complete separation due to sparse data with low exposure and event rates. PS adjustment resulted in greater power than PS matching, while controlling Type I error at 0.05. The PML method provided reasonable P-values, even in cases with complete separation, with well controlled Type I error rates. Using PS adjustment and the PML method, we identify novel latent drug effects in pediatric patients exposed to two common antibiotic drugs, ampicillin and gentamicin.nnnAvailability and implementationnR packages PheWAS and EHR are available at https://github.com/PheWAS/PheWAS and at CRAN (https://www.r-project.org/), respectively. The R script for data processing and the main analysis is available at https://github.com/choileena/EHR.nnnSupplementary informationnSupplementary data are available at Bioinformatics online.

Phenome-wide association study identifies marked increased in burden of comorbidities in African Americans with systemic lupus erythematosus.

BackgroundAfrican Americans with systemic lupus erythematosus (SLE) have increased renal disease compared to Caucasians, but differences in other comorbidities have not been well-described. We used an electronic health record (EHR) technique to test for differences in comorbidities in African Americans compared to Caucasians with SLE.MethodsWe used a de-identified EHR with 2.8 million subjects to identify SLE cases using a validated algorithm. We performed phenome-wide association studies (PheWAS) comparing African American to Caucasian SLE cases and African American SLE cases to matched non-SLE controls. Controls were age, sex, and race matched to SLE cases. For multiple testing, a false discovery rate (FDR) p value of 0.05 was used.ResultsWe identified 270 African Americans and 715 Caucasians with SLE and 1425 matched African American controls. Compared to Caucasians with SLE adjusting for age and sex, African Americans with SLE had more comorbidities in every organ system. The most striking included hypertension odds ratio (OR) = 4.25, FDR p = 5.49 × 10− 15; renal dialysis OR = 10.90, FDR p = 8.75 × 10− 14; and pneumonia OR = 3.57, FDR p = 2.32 × 10− 8. Compared to the African American matched controls without SLE, African Americans with SLE were more likely to have comorbidities in every organ system. The most significant codes were renal and cardiac, and included renal failure (OR = 9.55, FDR p = 2.26 × 10− 40) and hypertensive heart and renal disease (OR = 8.08, FDR p = 1.78 × 10− 22). Adjusting for race, age, and sex in a model including both African American and Caucasian SLE cases and controls, SLE was independently associated with renal, cardiovascular, and infectious diseases (all p <u20090.01).ConclusionsAfrican Americans with SLE have an increased comorbidity burden compared to Caucasians with SLE and matched controls. This increase in comorbidities in African Americans with SLE highlights the need to monitor for cardiovascular and infectious complications.