Medical Phenome Research Articles

Disentangling the phenotypic patterns of hypertension and chronic hypotension

Objective2017 blood pressure (BP) categories focus on cardiac risk. We hypothesize that studying the balance between mechanisms that increase or decrease BP across the medical phenome will lead to new insights. We devised a classifier that uses BP measures to assign individuals to mutually exclusive categories centered in the upper (Htn), lower (Hotn) and middle (Naf) zones of the BP spectrum; and examined the epidemiologic and phenotypic patterns of these BP-categories. MethodsWe classified a cohort of 832,560 deidentified electronic health records by BP-category; compared the frequency of BP-categories and four subtypes of Htn and Hotn by sex and age-decade; visualized the distributions of systolic, diastolic, mean arterial and pulse pressures stratified by BP-category; and ran Phenome-wide Association Studies (PheWAS) for Htn and Hotn. We paired knowledgebases for hypertension and hypotension and computed aggregate knowledgebase status (KB-status) indicating known associations. We assessed alignment of PheWAS results with KB-status for phecodes in the knowledgebase, and paired PheWAS correlations with KB-status to surface phenotypic patterns. ResultsBP-categories represent distinct distributions within the multimodal distributions of systolic and diastolic pressure. They are centered in the upper, lower, and middle zones of mean arterial pressure and provide a different signal than pulse pressure. For phecodes in the knowledgebase, 85% of positive correlations align with KB-status. Phenotypic patterns for Htn and Hotn overlap for several phecodes and are separate for others. Our analysis suggests five candidates for hypothesis testing research, two where the prevalence of the association with Htn or Hotn may be under appreciated, three where mechanisms that increase and decrease blood pressure may be affecting one another’s expression. ConclusionPairedPheWAS methods may open a phenome-wide path to disentangling hypertension and chronic hypotension. Our classifier provides a starting point for assigning individuals to BP-categories representing the upper, lower, and middle zones of the BP spectrum. 4.7 % of individuals matching 2017 BP categories for normal, elevated BP or isolated hypertension, have diastolic pressure < 60. Research is needed to fine-tune the classifier, provide external validation, evaluate the clinical significance of diastolic pressure < 60, and test the candidate hypotheses.

Comorbidity Profiles of Posttraumatic Stress Disorder Across the Medical Phenome

BackgroundPrevious epidemiological research has linked posttraumatic stress disorder (PTSD) with specific physical health problems, but the comprehensive landscape of medical conditions associated with PTSD remains uncharacterized. Electronic health records provide an opportunity to overcome clinical knowledge gaps and uncover associations with biological relevance that potentially vary by sex. MethodsPTSD was defined among biobank participants (N = 145,959) in 3 major healthcare systems using 2 ICD code-based definitions: broad (≥1 PTSD or acute stress codes vs. 0; ncases = 16,706) and narrow (≥2 PTSD codes vs. 0; ncases = 3325). Using a phenome-wide association study design, we tested associations between each PTSD definition and all prevalent disease umbrella categories, i.e., phecodes. We also conducted sex-stratified phenome-wide association study analyses including a sex × diagnosis interaction term in each logistic regression. ResultsA substantial number of phecodes were significantly associated with PTSDNarrow (61%) and PTSDBroad (83%). While the strongest associations were shared between the 2 definitions, PTSDBroad captured 334 additional phecodes not significantly associated with PTSDNarrow and exhibited a wider range of significantly associated phecodes across various categories, including respiratory, genitourinary, and circulatory conditions. Sex differences were observed in that PTSDBroad was more strongly associated with osteoporosis, respiratory failure, hemorrhage, and pulmonary heart disease among male patients and with urinary tract infection, acute pharyngitis, respiratory infections, and overweight among female patients. ConclusionsThis study provides valuable insights into a diverse range of comorbidities associated with PTSD, including both known and novel associations, while highlighting the influence of sex differences and the impact of defining PTSD using electronic health records.

Machine‐learning based classification of Frontotemporal dementia in electronic health records for genetic discovery

AbstractBackgroundFrontotemporal dementia (FTD) is a neurodegenerative syndrome characterized by distinct subtypes, each with unique clinical presentations and varying neuropathologic features. Although FTD is highly heritable, it is frequently undiagnosed and genetic causes of FTD are rarely determined, limiting what is known about the medical phenome associated with FTD. Here, we perform a combination of innovative strategies to improve detection of FTD in a large electronic health record (EHR) using machine‐learning (ML) and uncover genetic factors for FTD.Method123 FTD cases and 615 matched controls were identified via diagnostic codes and confirmed by manual expert review of medical notes in Vanderbilt University Medical Center’s (VUMC) EHR. We then applied a feature selection algorithm which calculated the frequency for each phecode within a randomly generated and demographically matched control set. A p‐value denoting the enrichment of each phecode was calculated based on the observed phecode frequency relative to the permutation‐derived null distribution. 28 phecodes were enriched in FTD cases compared to controls. Top predictors included “Neurological disorders”, “Memory loss”, Aphasia”, and “Parkinson’s disorder”. We then developed a classification and regression tree model that leveraged enriched phecodes to label an undiagnosed case set in VUMC’s EHR‐linked DNA databank, BioVU. The model was trained on 96 cases and 481 controls, with 28 cases and 139 controls reserved for testing.ResultOf the testing set, 19 of the FTD cases were classified as cases and 5 were misclassified as controls. 135 controls were classified as such and 4 were misclassified as cases, indicating 82% PPV. Application of this model in BioVU resulted in an expanded case set of inferred FTD, which is currently being leveraged in genomic analyses to identify causal variants. We have identified a genomic segment on chromosome 1 (146.9MB‐147.3MB) spanning ACP6 and GJA5 that is shared in significant excess among FTD cases. These genes have been associated with intellectual disability, aphasia, and autism.ConclusionOur results demonstrate both the potential of machine learning‐based approaches for case acquisition in dementia subtypes that are often underdiagnosed in medical record databases and illustrate how inferred cases can be leveraged for novel gene discovery.

Next-generation phenotyping: introducing phecodeX for enhanced discovery research in medical phenomics.

Phecodes are widely used and easily adapted phenotypes based on International Classification of Diseases codes. The current version of phecodes (v1.2) was designed primarily to study common/complex diseases diagnosed in adults; however, there are numerous limitations in the codes and their structure. Here, we present phecodeX, an expanded version of phecodes with a revised structure and 1,761 new codes. PhecodeX adds granularity to phenotypes in key disease domains that are under-represented in the current phecode structure-including infectious disease, pregnancy, congenital anomalies, and neonatology-and is a more robust representation of the medical phenome for global use in discovery research. phecodeX is available at https://github.com/PheWAS/phecodeX.

Scanning the medical phenome to identify new diagnoses after recovery from COVID-19 in a US cohort.

ObjectiveCOVID-19 survivors are at risk for long-term health effects, but assessing the sequelae of COVID-19 at large scales is challenging. High-throughput methods to efficiently identify new medical problems arising after acute medical events using the electronic health record (EHR) could improve surveillance for long-term consequences of acute medical problems like COVID-19.Materials and MethodsWe augmented an existing high-throughput phenotyping method (PheWAS) to identify new diagnoses occurring after an acute temporal event in the EHR. We then used the temporal-informed phenotypes to assess development of new medical problems among COVID-19 survivors enrolled in an EHR cohort of adults tested for COVID-19 at Vanderbilt University Medical Center.ResultsThe study cohort included 186 105 adults tested for COVID-19 from March 5, 2020 to November 1, 2021; of which 30 088 (16.2%) tested positive. Median follow-up after testing was 412 days (IQR 274–528). Our temporal-informed phenotyping was able to distinguish phenotype chapters based on chronicity of their constituent diagnoses. PheWAS with temporal-informed phenotypes identified increased risk for 43 diagnoses among COVID-19 survivors during outpatient follow-up, including multiple new respiratory, cardiovascular, neurological, and pregnancy-related conditions. Findings were robust to sensitivity analyses, and several phenotypic associations were supported by changes in outpatient vital signs or laboratory tests from the pretesting to postrecovery period.ConclusionTemporal-informed PheWAS identified new diagnoses affecting multiple organ systems among COVID-19 survivors. These findings can inform future efforts to enable longitudinal health surveillance for survivors of COVID-19 and other acute medical conditions using the EHR.

Sex-Stratified Relationships Between PTSD Phenotypes and the Medical Phenome Using Electronic Health Records

Prior epidemiological research has linked PTSD with specific physical health problems, but the comprehensive landscape of medical conditions associated with PTSD remains uncharacterized. Electronic health records (EHR) provide an opportunity to overcome prior clinical knowledge gaps and uncover associations with biological relevance that potentially vary by sex.

Integration of genetic, transcriptomic, and clinical data provides insight into 16p11.2 and 22q11.2 CNV genes

BackgroundDeletions and duplications of the multigenic 16p11.2 and 22q11.2 copy number variant (CNV) regions are associated with brain-related disorders including schizophrenia, intellectual disability, obesity, bipolar disorder, and autism spectrum disorder (ASD). The contribution of individual CNV genes to each of these identified phenotypes is unknown, as well as the contribution of these CNV genes to other potentially subtler health implications for carriers. Hypothesizing that DNA copy number exerts most effects via impacts on RNA expression, we attempted a novel in silico fine-mapping approach in non-CNV carriers using both GWAS and biobank data.MethodsWe first asked whether gene expression level in any individual gene in the CNV region alters risk for a known CNV-associated behavioral phenotype(s). Using transcriptomic imputation, we performed association testing for CNV genes within large genotyped cohorts for schizophrenia, IQ, BMI, bipolar disorder, and ASD. Second, we used a biobank containing electronic health data to compare the medical phenome of CNV carriers to controls within 700,000 individuals in order to investigate the full spectrum of health effects of the CNVs. Third, we used genotypes for over 48,000 individuals within the biobank to perform phenome-wide association studies between imputed expressions of individual 16p11.2 and 22q11.2 genes and over 1500 health traits.ResultsUsing large genotyped cohorts, we found individual genes within 16p11.2 associated with schizophrenia (TMEM219, INO80E, YPEL3), BMI (TMEM219, SPN, TAOK2, INO80E), and IQ (SPN), using conditional analysis to identify upregulation of INO80E as the driver of schizophrenia, and downregulation of SPN and INO80E as increasing BMI. We identified both novel and previously observed over-represented traits within the electronic health records of 16p11.2 and 22q11.2 CNV carriers. In the phenome-wide association study, we found seventeen significant gene-trait pairs, including psychosis (NPIPB11, SLX1B) and mood disorders (SCARF2), and overall enrichment of mental traits.ConclusionsOur results demonstrate how integration of genetic and clinical data aids in understanding CNV gene function and implicates pleiotropy and multigenicity in CNV biology.

Medical phenome of musicians: an investigation of health records collected on 9803 musically active individuals.

Previous studies suggest that musicians may be at higher risk for a set of medical problems; however, this literature has been limited by relatively small sample sizes, self‐reports, and lack of controls. To address such limitations, we examined trends in the medical care of musicians in an Electronic Health Record database. On the basis of a collection of keywords and regular expressions in the patients’ clinical notes, we identified 9803 “musicians” that we matched for sex, median age (across the medical record), ethnicity, race, the length of record, and the number of visits with 49,015 controls. We fitted 1263 logistic regression models to determine whether the phenotype was correlated with musicianship. Two hundred fifty‐seven phenotypes were more prevalent in musicians than controls after Bonferroni adjustment (P < 7.6 × 10–6), including diseases of the larynx and vocal cords (OR = 2.32 (95% CI: 2.25–2.40)), and hearing loss (OR = 1.36 (95% CI: 1.32–1.39)). Fifteen phenotypes were significantly more prevalent in controls than musicians, including coronary atherosclerosis (OR = 0.91 (95% CI: 0.89–0.94)). Although being a musician was related to many occupational health problems, we identified protective effects of musicianship in which certain disorders were less common in musicians than in controls, indicating that active musical engagement could have health benefits analogous to athletic engagement.

A Phenome-Wide Association Study (PheWAS) of COVID-19 Outcomes by Race Using the Electronic Health Records Data in Michigan Medicine.

Background: We performed a phenome-wide association study to identify pre-existing conditions related to Coronavirus disease 2019 (COVID-19) prognosis across the medical phenome and how they vary by race. Methods: The study is comprised of 53,853 patients who were tested/diagnosed for COVID-19 between 10 March and 2 September 2020 at a large academic medical center. Results: Pre-existing conditions strongly associated with hospitalization were renal failure, pulmonary heart disease, and respiratory failure. Hematopoietic conditions were associated with intensive care unit (ICU) admission/mortality and mental disorders were associated with mortality in non-Hispanic Whites. Circulatory system and genitourinary conditions were associated with ICU admission/mortality in non-Hispanic Blacks. Conclusions: Understanding pre-existing clinical diagnoses related to COVID-19 outcomes informs the need for targeted screening to support specific vulnerable populations to improve disease prevention and healthcare delivery.

Scanning the Medical Phenome to Identify New Medical Diagnoses After Recovery From COVID-19 in a US Cohort

Background: COVID-19 survivors are at risk for long-term health effects, but population-level data remain limited. High-throughput methods to efficiently identify new medical problems among COVID-19 survivors using large electronic health record (EHR) databases could improve surveillance of COVID-19’s long-term consequences.Methods: Retrospective cohort study using a research-enabled EHR registry of all adults who underwent SARS-CoV-2 testing at Vanderbilt University Medical Center from March 5, 2020 to May 1, 2021. We extracted every new diagnosis code entered into the EHR following recovery from infection or after an equivalent time period for patients who tested negative, and assessed development of over 1,000 disease outcomes using a phenome-wide association study (PheWAS) approach.Findings: We included 148,710 adults tested for SARS-CoV-2 , of which 24,203 (16·3%) tested positive. Median follow-up was 275 days (IQR 198–364). After correction for multiple testing (p ≤ 4·89×10-5), COVID-19 survivors had increased risk for 24 clinical phenotypes during outpatient follow-up, including new diagnoses related to abnormal pulmonary function, cardiac arrhythmias, myocardial inflammation, and neurological impairment. Findings were robust to multiple sensitivity analyses and several of these phenotypic associations were detectable in the cohort as early as January 2021.Interpretation: In this regional US registry, survival from COVID-19 was associated with new diagnoses affecting multiple organ systems, which may reflect late effects of inflammation and vascular injury or the sequelae of severe illness among hospitalized survivors. These findings can inform efforts for longitudinal health surveillance of COVID-19 survivors.Funding: This study was supported in part by the American Thoracic Society (VEK), and National Institutes of Health continuing education grant NIH T15 LM007450 (VEK), and research grants NIH K01 HL15775-01 (VEK), NIH U01 HG011166-01S1 (JFP), and NIH R01 GM139891-01 (WQW).Declaration of Interest: The authors declare no competing financial or competing interests.Ethical Approval: This study was conducted with approval from the Vanderbilt University Institutional Review Board (study approval numbers: #200512, #200731) under a waiver of informed consent

Phenotype risk scores (PheRS) for pancreatic cancer using time-stamped electronic health record data: Discovery and validation in two large biobanks

BackgroundTraditional methods for disease risk prediction and assessment, such as diagnostic tests using serum, urine, blood, saliva or imaging biomarkers, have been important for identifying high-risk individuals for many diseases, leading to early detection and improved survival. For pancreatic cancer, traditional methods for screening have been largely unsuccessful in identifying high-risk individuals in advance of disease progression leading to high mortality and poor survival. Electronic health records (EHR) linked to genetic profiles provide an opportunity to integrate multiple sources of patient information for risk prediction and stratification. We leverage a constellation of temporally associated diagnoses available in the EHR to construct a summary risk score, called a phenotype risk score (PheRS), for identifying individuals at high-risk for having pancreatic cancer. The proposed PheRS approach incorporates the time with respect to disease onset into the prediction framework. We combine and contrast the PheRS with more well-known measures of inherited susceptibility, namely, the polygenic risk scores (PRS) for prediction of pancreatic cancer. MethodologyWe first calculated pairwise, unadjusted associations between pancreatic cancer diagnosis and all possible other diagnoses across the medical phenome. We call these pairwise associations co-occurrences. After accounting for cross-phenotype correlations, the multivariable association estimates from a subset of relatively independent diagnoses were used to create a weighted sum PheRS. We constructed time-restricted risk scores using data from 38,359 participants in the Michigan Genomics Initiative (MGI) based on the diagnoses contained in the EHR at 0, 1, 2, and 5 years prior to the target pancreatic cancer diagnosis. The PheRS was assessed for predictability in the UK Biobank (UKB). We tested the relative contribution of PheRS when added to a model containing a summary measure of inherited genetic susceptibility (PRS) plus other covariates like age, sex, smoking status, drinking status, and body mass index (BMI). ResultsOur exploration of co-occurrence patterns identified expected associations while also revealing unexpected relationships that may warrant closer attention. Solely using the pancreatic cancer PheRS at 5 years before the target diagnoses yielded an AUC of 0.60 (95% CI = [0.58, 0.62]) in UKB. A larger predictive model including PheRS, PRS, and the covariates at the 5-year threshold achieved an AUC of 0.74 (95% CI = [0.72, 0.76]) in UKB. We note that PheRS does contribute independently in the joint model. Finally, scores at the top percentiles of the PheRS distribution demonstrated promise in terms of risk stratification. Scores in the top 2% were 10.20 (95% CI = [9.34, 12.99]) times more likely to identify cases than those in the bottom 98% in UKB at the 5-year threshold prior to pancreatic cancer diagnosis. ConclusionsWe developed a framework for creating a time-restricted PheRS from EHR data for pancreatic cancer using the rich information content of a medical phenome. In addition to identifying hypothesis-generating associations for future research, this PheRS demonstrates a potentially important contribution in identifying high-risk individuals, even after adjusting for PRS for pancreatic cancer and other traditional epidemiologic covariates. The methods are generalizable to other phenotypic traits.

Phenome-wide Burden of Copy-Number Variation in the UK Biobank

Copy-number variations (CNVs) represent a significant proportion of the genetic differences between individuals and many CNVs associate causally with syndromic disease and clinical outcomes. Here, we characterize the landscape of copy-number variation and their phenome-wide effects in a sample of 472,228 array-genotyped individuals from the UK Biobank. In addition to population-level selection effects against genic loci conferring high mortality, we describe genetic burden from potentially pathogenic and previously uncharacterized CNV loci across more than 3,000 quantitative and dichotomous traits, with separate analyses for common and rare classes of variation. Specifically, we highlight the effects of CNVs at two well-known syndromic loci 16p11.2 and 22q11.2, previously uncharacterized variation at 9p23, and several genic associations in the context of acute coronary artery disease and high body mass index. Our data constitute a deeply contextualized portrait of population-wide burden of copy-number variation, as well as a series of dosage-mediated genic associations across the medical phenome.

Genes In The Context of The Medical Phenome: The Challenges, Opportunities, and Pitfalls of Psychiatric Genetics Using Biobanks and Registries

Overall Abstract The theme of the 2017 WCPG Meeting, “Genes in Context,” highlights the essential insights that become apparent when individual risk genes are considered in the context of polygenic background, epigenetic modifications in the cell, and environmental conditions and development. Similarly, the large-scale collection of genetic material integrated with the increasing availability of expansive phenotype information (“phenome”) from electronic medical records (EMR) allows a broader context in which to study the genetics of neuropsychiatric phenotypes. Identifying correlated phenotypes and assessing the broader role of implicated variants/genes are only a few of the many ways in which these studies could provide a unique lens through which to interpret genetic findings. The potential for these datasets to scale (e.g. 500k individuals for UK biobank) and the comprehensiveness of the captured phenome offers opportunities that are difficult to achieve in disease focused cohort studies. Initial work on these extraordinary datasets is already yielding exciting new findings with promising translational implications. Rigorous analysis of biobank data requires integrating multiple disciplines including genomics, epidemiology, informatics, ethics and the clinical domain of interest. We have assembled a diverse roster of researchers who bring expertise in these areas as speakers in this session. Co-chair Dr. Douglas Ruderfer, will introduce the session by presenting the opportunities, challenges and some initial successes of utilizing these resources. Next, Dr. Lea Davis will describe development of EMR-based algorithms for detection of neurodevelopmental disorders, results of an EMR-enabled biomarker discovery approach using Vanderbilt’s 270,000 patient biobank (BioVU), and community-engaged efforts to develop a more complete understanding of genomic privacy concerns related to participation in biobanks. Dr. Gerome Breen will then discuss general applicability of using the 500k-individual UK Biobank, along with details of a study in anorexia nervosa. Dr. Beate St. Pourcain will describe new results from her work examining genetic factors that contribute to the developmental trajectory of behavioral traits using the Avon Longitudinal Study of Parents and Children (ALSPAC) data set, and highlight implications when assessing trait-disorder overlap. Finally, Dr. Zhengming Chen will describe results of ongoing PheWAS of functional variants to assess causal associations (e.g. alcohol) in the 500K-person Kadoorie Biobank. Following the speakers, Dr. Jordan Smoller will lead a discussion focused on efforts currently underway to combine biobank resources through the psychEMERGE consortium and future resources available from the NIH All of Us project. Topics for discussion include genome-informed machine learning approaches to prediction models, structural equation models using genetic relationship data in cohorts, challenges to algorithmic portability, and opportunities for immediate translation of genomic findings. Symposium attendees will become familiar with (a) potential resources that are available to them, (b) strategies for maximizing the utility of such resources, (c) challenges in utilizing these resources, and (d) ethical issues related to genomic privacy in the context of biobanks. Finally, attendees will learn about novel discoveries emerging from biobank and registry resources in the United States, United Kingdom, and China.

Phenome-wide association studies: a new method for functional genomics in humans.

In experimental physiological research, a common study design for examining the functional role of a gene or a genetic variant is to introduce that genetic variant into a model organism (such as yeast or mouse) and then to search for phenotypic consequences. The development of DNA biobanks linked to dense phenotypic information enables such an experiment to be applied to human subjects in the form of a phenome-wide association study (PheWAS). The PheWAS paradigm takes advantage of a curated medical phenome, often derived from electronic health records, to search for associations between 'input functions' and phenotypes in an unbiased fashion. The most commonly studied input function to date has been single nucleotide polymorphisms (SNPs), but other inputs, such as sets of SNPs or a disease or drug exposure, are now being explored to probe the genetic and phenotypic architecture of human traits. Potential outcomes of these approaches include defining subsets of complex diseases (that can then be targeted by specific therapies) and drug repurposing.