Validation of electronic medical record-based phenotyping algorithms: results and lessons learned from the eMERGE network

Large language models facilitate the generation of electronic health record phenotyping algorithms

OBJECTIVES

Phenotyping is a core task in observational health research utilizing electronic health records (EHRs). Developing an accurate algorithm demands substantial input from domain experts, involving extensive literature review and evidence synthesis. This burdensome process limits scalability and delays knowledge discovery. We investigate the potential for leveraging large language models (LLMs) to enhance the efficiency of EHR phenotyping by generating high-quality algorithm drafts.

MATERIALS AND METHODS

We prompted four LLMs-GPT-4 and GPT-3.5 of ChatGPT, Claude 2, and Bard-in October 2023, asking them to generate executable phenotyping algorithms in the form of SQL queries adhering to a common data model (CDM) for three phenotypes (ie, type 2 diabetes mellitus, dementia, and hypothyroidism). Three phenotyping experts evaluated the returned algorithms across several critical metrics. We further implemented the top-rated algorithms and compared them against clinician-validated phenotyping algorithms from the Electronic Medical Records and Genomics (eMERGE) network.

RESULTS

GPT-4 and GPT-3.5 exhibited significantly higher overall expert evaluation scores in instruction following, algorithmic logic, and SQL executability, when compared to Claude 2 and Bard. Although GPT-4 and GPT-3.5 effectively identified relevant clinical concepts, they exhibited immature capability in organizing phenotyping criteria with the proper logic, leading to phenotyping algorithms that were either excessively restrictive (with low recall) or overly broad (with low positive predictive values).

CONCLUSION

GPT versions 3.5 and 4 are capable of drafting phenotyping algorithms by identifying relevant clinical criteria aligned with a CDM. However, expertise in informatics and clinical experience is still required to assess and further refine generated algorithms.

A general framework for developing computable clinical phenotype algorithms

OBJECTIVE

To present a general framework providing high-level guidance to developers of computable algorithms for identifying patients with specific clinical conditions (phenotypes) through a variety of approaches, including but not limited to machine learning and natural language processing methods to incorporate rich electronic health record data.

MATERIALS AND METHODS

Drawing on extensive prior phenotyping experiences and insights derived from 3 algorithm development projects conducted specifically for this purpose, our team with expertise in clinical medicine, statistics, informatics, pharmacoepidemiology, and healthcare data science methods conceptualized stages of development and corresponding sets of principles, strategies, and practical guidelines for improving the algorithm development process.

RESULTS

We propose 5 stages of algorithm development and corresponding principles, strategies, and guidelines: (1) assessing fitness-for-purpose, (2) creating gold standard data, (3) feature engineering, (4) model development, and (5) model evaluation.

DISCUSSION AND CONCLUSION

This framework is intended to provide practical guidance and serve as a basis for future elaboration and extension.

Variations in Electronic Health Record-Based Definitions of Diabetic Retinopathy Cohorts: A Literature Review and Quantitative Analysis

Purpose

Use of the electronic health record (EHR) has motivated the need for data standardization. A gap in knowledge exists regarding variations in existing terminologies for defining diabetic retinopathy (DR) cohorts. This study aimed to review the literature and analyze variations regarding codified definitions of DR.

Design

Literature review and quantitative analysis.

Subjects

Published manuscripts.

Methods

Four graders reviewed PubMed and Google Scholar for peer-reviewed studies. Studies were included if they used codified definitions of DR (e.g., billing codes). Data elements such as author names, publication year, purpose, data set type, and DR definitions were manually extracted. Each study was reviewed by ≥ 2 authors to validate inclusion eligibility. Quantitative analyses of the codified definitions were then performed to characterize the variation between DR cohort definitions.

Main Outcome Measures

Number of studies included and numeric counts of billing codes used to define codified cohorts.

Results

In total, 43 studies met the inclusion criteria. Half of the included studies used datasets based on structured EHR data (i.e., data registries, institutional EHR review), and half used claims data. All but 1 of the studies used billing codes such as the International Classification of Diseases 9th or 10th edition (ICD-9 or ICD-10), either alone or in addition to another terminology for defining disease. Of the 27 included studies that used ICD-9 and the 20 studies that used ICD-10 codes, the most common codes used pertained to the full spectrum of DR severity. Diabetic retinopathy complications (e.g., vitreous hemorrhage) were also used to define some DR cohorts.

Conclusions

Substantial variations exist among codified definitions for DR cohorts within retrospective studies. Variable definitions may limit generalizability and reproducibility of retrospective studies. More work is needed to standardize disease cohorts.

Financial Disclosures

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Using electronic health records for clinical pharmacology research: Challenges and considerations

Electronic health records (EHRs) contain a vast array of phenotypic data on large numbers of individuals, often collected over decades. Due to the wealth of information, EHR data have emerged as a powerful resource to make first discoveries and identify disparities in our healthcare system. While the number of EHR-based studies has exploded in recent years, most of these studies are directed at associations with disease rather than pharmacotherapeutic outcomes, such as drug response or adverse drug reactions. This is largely due to challenges specific to deriving drug-related phenotypes from the EHR. There is great potential for EHR-based discovery in clinical pharmacology research, and there is a critical need to address specific challenges related to accurate and reproducible derivation of drug-related phenotypes from the EHR. This review provides a detailed evaluation of challenges and considerations for deriving drug-related data from EHRs. We provide an examination of EHR-based computable phenotypes and discuss cutting-edge approaches to map medication information for clinical pharmacology research, including medication-based computable phenotypes and natural language processing. We also discuss additional considerations such as data structure, heterogeneity and missing data, rare phenotypes, and diversity within the EHR. By further understanding the complexities associated with conducting clinical pharmacology research using EHR-based data, investigators will be better equipped to design thoughtful studies with more reproducible results. Progress in utilizing EHRs for clinical pharmacology research should lead to significant advances in our ability to understand differential drug response and predict adverse drug reactions.

Identifying prehospital trauma patients from ambulance patient care records; comparing two methods using linked data in New South Wales, Australia

BACKGROUND

Linked datasets for trauma system monitoring should ideally follow patients from the prehospital scene to hospital admission and post-discharge. Having a well-defined cohort when using administrative datasets is essential because they must capture the representative population. Unlike hospital electronic health records (EHR), ambulance patient-care records lack access to sources beyond immediate clinical notes. Relying on a limited set of variables to define a study population might result in missed patient inclusion. We aimed to compare two methods of identifying prehospital trauma patients: one using only those documented under a trauma protocol and another incorporating additional data elements from ambulance patient care records.

METHODS

We analyzed data from six routinely collected administrative datasets from 2015 to 2018, including ambulance patient-care records, aeromedical data, emergency department visits, hospitalizations, rehabilitation outcomes, and death records. Three prehospital trauma cohorts were created: an Extended-T-protocol cohort (patients transported under a trauma protocol and/or patients with prespecified criteria from structured data fields), T-protocol cohort (only patients documented as transported under a trauma protocol) and non-T-protocol (extended-T-protocol population not in the T-protocol cohort). Patient-encounter characteristics, mortality, clinical and post-hospital discharge outcomes were compared. A conservative p-value of 0.01 was considered significant RESULTS: Of 1 038 263 patient-encounters included in the extended-T-population 814 729 (78.5 %) were transported, with 438 893 (53.9 %) documented as a T-protocol patient. Half (49.6 %) of the non-T-protocol sub-cohort had an International Classification of Disease 10th edition injury or external cause code, indicating 79644 missed patients when a T-protocol-only definition was used. The non-T-protocol sub-cohort also identified additional patients with intubation, prehospital blood transfusion and positive eFAST. A higher proportion of non-T protocol patients than T-protocol patients were admitted to the ICU (4.6% vs 3.6 %), ventilated (1.8% vs 1.3 %), received in-hospital transfusion (7.9 vs 6.8 %) or died (1.8% vs 1.3 %). Urgent trauma surgery was similar between groups (1.3% vs 1.4 %).

CONCLUSION

The extended-T-population definition identified 50 % more admitted patients with an ICD-10-AM code consistent with an injury, including patients with severe trauma. Developing an EHR phenotype incorporating multiple data fields of ambulance-transported trauma patients for use with linked data may avoid missing these patients.

Designing target trials using electronic health records: A case study of second-line disease-modifying anti-rheumatic drugs and cardiovascular disease outcomes in patients with rheumatoid arthritis

BACKGROUND

Emulation of the "target trial" (TT), a hypothetical pragmatic randomized controlled trial (RCT), using observational data can be used to mitigate issues commonly encountered in comparative effectiveness research (CER) when randomized trials are not logistically, ethically, or financially feasible. However, cardiovascular (CV) health research has been slow to adopt TT emulation. Here, we demonstrate the design and analysis of a TT emulation using electronic health records to study the comparative effectiveness of the addition of a disease-modifying anti-rheumatic drug (DMARD) to a regimen of methotrexate on CV events among rheumatoid arthritis (RA) patients.

METHODS

We used data from an electronic medical records-based cohort of RA patients from Northwestern Medicine to emulate the TT. Follow-up began 3 months after initial prescription of MTX (2000-2020) and included all available follow-up through June 30, 2020. Weighted pooled logistic regression was used to estimate differences in CVD risk and survival. Cloning was used to handle immortal time bias and weights to improve baseline and time-varying covariate imbalance.

RESULTS

We identified 659 eligible people with RA with average follow-up of 46 months and 31 MACE events. The month 24 adjusted risk difference for MACE comparing initiation vs non-initiation of a DMARD was -1.47% (95% confidence interval [CI]: -4.74, 1.95%), and the marginal hazard ratio (HR) was 0.72 (95% CI: 0.71, 1.23). In analyses subject to immortal time bias, the HR was 0.62 (95% CI: 0.29-1.44).

CONCLUSION

In this sample, we did not observe evidence of differences in risk of MACE, a finding that is compatible with previously published meta-analyses of RCTs. Thoughtful application of the TT framework provides opportunities to conduct CER in observational data. Benchmarking results of observational analyses to previously published RCTs can lend credibility to interpretation.

Optimising the use of electronic medical records for large scale research in psychiatry

The explosion and abundance of digital data could facilitate large-scale research for psychiatry and mental health. Research using so-called "real world data"-such as electronic medical/health records-can be resource-efficient, facilitate rapid hypothesis generation and testing, complement existing evidence (e.g. from trials and evidence-synthesis) and may enable a route to translate evidence into clinically effective, outcomes-driven care for patient populations that may be under-represented. However, the interpretation and processing of real-world data sources is complex because the clinically important 'signal' is often contained in both structured and unstructured (narrative or "free-text") data. Techniques for extracting meaningful information (signal) from unstructured text exist and have advanced the re-use of routinely collected clinical data, but these techniques require cautious evaluation. In this paper, we survey the opportunities, risks and progress made in the use of electronic medical record (real-world) data for psychiatric research.

Automated HIV Case Identification from the MIMIC-IV Database

Automatic HIV phenotyping is needed for HIV research based on electronic health records (EHRs). MIMIC-IV, an extension of MIMIC-III, contains more than 520,000 hospital admissions and has become a valuable EHR database for secondary medical research. However, there was no prior phenotyping algorithm to extract HIV cases from MIMIC-IV, which requires a comprehensive knowledge of the database. Moreover, previous HIV phenotyping algorithms did not consider the new HIV-1/HIV-2 antibody differentiation immunoassay tests that MIMIC-IV contains. Our work provided insight into the structure and data elements in MIMIC-IV and proposed a new HIV phenotyping algorithm to fill in these gaps. The results included MIMIC-IV's data tables and elements used, 1,781 and 1,843 HIV cases from MIMIC-IV's versions 0.4 and 2.1, respectively, and summary statistics of these two HIV case cohorts. They could be used for the development of statistical and machine learning models in future studies about the disease.

Human-centered Design of a Health Recommender System for Orthopaedic Shoulder Treatment

Background

Rich data on diverse patients and their treatments and outcomes within Electronic Health Record (EHR) systems can be used to generate real world evidence. A health recommender system (HRS) framework can be applied to a decision support system application to generate data summaries for similar patients during the clinical encounter to assist physicians and patients in making evidence-based shared treatment decisions.

Objective

A human-centered design (HCD) process was used to develop a HRS for treatment decision support in orthopaedic medicine, the Informatics Consult for Individualized Treatment (I-C-IT). We also evaluate the usability and utility of the system from the physician's perspective, focusing on elements of utility and shared decision-making in orthopaedic medicine.

Methods

The HCD process for I-C-IT included 6 steps across three phases of analysis, design, and evaluation. A team of informaticians and comparative effectiveness researchers directly engaged with orthopaedic surgeon subject matter experts in a collaborative I-C-IT prototype design process. Ten orthopaedic surgeons participated in a mixed methods evaluation of the I-C-IT prototype that was produced.

Results

The HCD process resulted in a prototype system, I-C-IT, with 14 data visualization elements and a set of design principles crucial for HRS for decision support. The overall standard system usability scale (SUS) score for the I-C-IT Webapp prototype was 88.75 indicating high usability. In addition, utility questions addressing shared decision-making found that 90% of orthopaedic surgeon respondents either strongly agreed or agreed that I-C-IT would help them make data informed decisions with their patients.

Conclusion

The HCD process produced an HRS prototype that is capable of supporting orthopaedic surgeons and patients in their information needs during clinical encounters. Future research should focus on refining I-C-IT by incorporating patient feedback in future iterative cycles of system design and evaluation.

Development and validation of a race-agnostic computable phenotype for kidney health in adult hospitalized patients

Standard race adjustments for estimating glomerular filtration rate (GFR) and reference creatinine can yield a lower acute kidney injury (AKI) and chronic kidney disease (CKD) prevalence among African American patients than non-race adjusted estimates. We developed two race-agnostic computable phenotypes that assess kidney health among 139,152 subjects admitted to the University of Florida Health between 1/2012-8/2019 by removing the race modifier from the estimated GFR and estimated creatinine formula used by the race-adjusted algorithm (race-agnostic algorithm 1) and by utilizing 2021 CKD-EPI refit without race formula (race-agnostic algorithm 2) for calculations of the estimated GFR and estimated creatinine. We compared results using these algorithms to the race-adjusted algorithm in African American patients. Using clinical adjudication, we validated race-agnostic computable phenotypes developed for preadmission CKD and AKI presence on 300 cases. Race adjustment reclassified 2,113 (8%) to no CKD and 7,901 (29%) to a less severe CKD stage compared to race-agnostic algorithm 1 and reclassified 1,208 (5%) to no CKD and 4,606 (18%) to a less severe CKD stage compared to race-agnostic algorithm 2. Of 12,451 AKI encounters based on race-agnostic algorithm 1, race adjustment reclassified 591 to No AKI and 305 to a less severe AKI stage. Of 12,251 AKI encounters based on race-agnostic algorithm 2, race adjustment reclassified 382 to No AKI and 196 (1.6%) to a less severe AKI stage. The phenotyping algorithm based on refit without race formula performed well in identifying patients with CKD and AKI with a sensitivity of 100% (95% confidence interval [CI] 97%-100%) and 99% (95% CI 97%-100%) and a specificity of 88% (95% CI 82%-93%) and 98% (95% CI 93%-100%), respectively. Race-agnostic algorithms identified substantial proportions of additional patients with CKD and AKI compared to race-adjusted algorithm in African American patients. The phenotyping algorithm is promising in identifying patients with kidney disease and improving clinical decision-making.

Improving reporting standards for phenotyping algorithm in biomedical research: 5 fundamental dimensions

INTRODUCTION

Phenotyping algorithms enable the interpretation of complex health data and definition of clinically relevant phenotypes; they have become crucial in biomedical research. However, the lack of standardization and transparency inhibits the cross-comparison of findings among different studies, limits large scale meta-analyses, confuses the research community, and prevents the reuse of algorithms, which results in duplication of efforts and the waste of valuable resources.

RECOMMENDATIONS

Here, we propose five independent fundamental dimensions of phenotyping algorithms-complexity, performance, efficiency, implementability, and maintenance-through which researchers can describe, measure, and deploy any algorithms efficiently and effectively. These dimensions must be considered in the context of explicit use cases and transparent methods to ensure that they do not reflect unexpected biases or exacerbate inequities.

Use of noisy labels as weak learners to identify incompletely ascertainable outcomes: A Feasibility study with opioid-induced respiratory depression

Objective

Assigning outcome labels to large observational data sets in a timely and accurate manner, particularly when outcomes are rare or not directly ascertainable, remains a significant challenge within biomedical informatics. We examined whether noisy labels generated from subject matter experts' heuristics using heterogenous data types within a data programming paradigm could provide outcomes labels to a large, observational data set. We chose the clinical condition of opioid-induced respiratory depression for our use case because it is rare, has no administrative codes to easily identify the condition, and typically requires at least some unstructured text to ascertain its presence.

Materials and methods

Using de-identified electronic health records of 52,861 post-operative encounters, we applied a data programming paradigm (implemented in the Snorkel software) for the development of a machine learning classifier for opioid-induced respiratory depression. Our approach included subject matter experts creating 14 labeling functions that served as noisy labels for developing a probabilistic Generative model. We used probabilistic labels from the Generative model as outcome labels for training a Discriminative model on the source data. We evaluated performance of the Discriminative model with a hold-out test set of 599 independently-reviewed patient records.

Results

The final Discriminative classification model achieved an accuracy of 0.977, an F1 score of 0.417, a sensitivity of 1.0, and an AUC of 0.988 in the hold-out test set with a prevalence of 0.83% (5/599).

Discussion

All of the confirmed Cases were identified by the classifier. For rare outcomes, this finding is encouraging because it reduces the number of manual reviews needed by excluding visits/patients with low probabilities.

Conclusion

Application of a data programming paradigm with expert-informed labeling functions might have utility for phenotyping clinical phenomena that are not easily ascertainable from highly-structured data.

Large Language Models Facilitate the Generation of Electronic Health Record Phenotyping Algorithms

Objectives

Phenotyping is a core task in observational health research utilizing electronic health records (EHRs). Developing an accurate algorithm demands substantial input from domain experts, involving extensive literature review and evidence synthesis. This burdensome process limits scalability and delays knowledge discovery. We investigate the potential for leveraging large language models (LLMs) to enhance the efficiency of EHR phenotyping by generating high-quality algorithm drafts.

Materials and Methods

We prompted four LLMs-GPT-4 and GPT-3.5 of ChatGPT, Claude 2, and Bard-in October 2023, asking them to generate executable phenotyping algorithms in the form of SQL queries adhering to a common data model (CDM) for three phenotypes (i.e., type 2 diabetes mellitus, dementia, and hypothyroidism). Three phenotyping experts evaluated the returned algorithms across several critical metrics. We further implemented the top-rated algorithms and compared them against clinician-validated phenotyping algorithms from the Electronic Medical Records and Genomics (eMERGE) network.

Results

GPT-4 and GPT-3.5 exhibited significantly higher overall expert evaluation scores in instruction following, algorithmic logic, and SQL executability, when compared to Claude 2 and Bard. Although GPT-4 and GPT-3.5 effectively identified relevant clinical concepts, they exhibited immature capability in organizing phenotyping criteria with the proper logic, leading to phenotyping algorithms that were either excessively restrictive (with low recall) or overly broad (with low positive predictive values).

Conclusion

GPT versions 3.5 and 4 are capable of drafting phenotyping algorithms by identifying relevant clinical criteria aligned with a CDM. However, expertise in informatics and clinical experience is still required to assess and further refine generated algorithms.

Use of Noisy Labels as Weak Learners to Identify Incompletely Ascertainable Outcomes: A Feasibility Study with Opioid-Induced Respiratory Depression

Objective

Assigning outcome labels to large observational data sets in a timely and accurate manner, particularly when outcomes are rare or not directly ascertainable, remains a significant challenge within biomedical informatics. We examined whether noisy labels generated from subject matter experts' heuristics using heterogenous data types within a data programming paradigm could provide outcomes labels to a large, observational data set. We chose the clinical condition of opioid-induced respiratory depression for our use case because it is rare, has no administrative codes to easily identify the condition, and typically requires at least some unstructured text to ascertain its presence.

Materials and Methods

Using de-identified electronic health records of 52,861 post-operative encounters, we applied a data programming paradigm (implemented in the Snorkel software) for the development of a machine learning classifier for opioid-induced respiratory depression. Our approach included subject matter experts creating 14 labeling functions that served as noisy labels for developing a probabilistic Generative model. We used probabilistic labels from the Generative model as outcome labels for training a Discriminative model on the source data. We evaluated performance of the Discriminative model with a hold-out test set of 599 independently-reviewed patient records.

Results

The final Discriminative classification model achieved an accuracy of 0.977, an F1 score of 0.417, a sensitivity of 1.0, and an AUC of 0.988 in the hold-out test set with a prevalence of 0.83% (5/599).

Discussion

All of the confirmed Cases were identified by the classifier. For rare outcomes, this finding is encouraging because it reduces the number of manual reviews needed by excluding visits/patients with low probabilities.

Conclusion

Application of a data programming paradigm with expert-informed labeling functions might have utility for phenotyping clinical phenomena that are not easily ascertainable from highly-structured data.

Developing a Framework to Infer Opioid Use Disorder Severity From Clinical Notes to Inform Natural Language Processing Methods: Characterization Study

BACKGROUND

Information regarding opioid use disorder (OUD) status and severity is important for patient care. Clinical notes provide valuable information for detecting and characterizing problematic opioid use, necessitating development of natural language processing (NLP) tools, which in turn requires reliably labeled OUD-relevant text and understanding of documentation patterns.

OBJECTIVE

To inform automated NLP methods, we aimed to develop and evaluate an annotation schema for characterizing OUD and its severity, and to document patterns of OUD-relevant information within clinical notes of heterogeneous patient cohorts.

METHODS

We developed an annotation schema to characterize OUD severity based on criteria from the Diagnostic and Statistical Manual of Mental Disorders, 5th edition. In total, 2 annotators reviewed clinical notes from key encounters of 100 adult patients with varied evidence of OUD, including patients with and those without chronic pain, with and without medication treatment for OUD, and a control group. We completed annotations at the sentence level. We calculated severity scores based on annotation of note text with 18 classes aligned with criteria for OUD severity and determined positive predictive values for OUD severity.

RESULTS

The annotation schema contained 27 classes. We annotated 1436 sentences from 82 patients; notes of 18 patients (11 of whom were controls) contained no relevant information. Interannotator agreement was above 70% for 11 of 15 batches of reviewed notes. Severity scores for control group patients were all 0. Among noncontrol patients, the mean severity score was 5.1 (SD 3.2), indicating moderate OUD, and the positive predictive value for detecting moderate or severe OUD was 0.71. Progress notes and notes from emergency department and outpatient settings contained the most and greatest diversity of information. Substance misuse and psychiatric classes were most prevalent and highly correlated across note types with high co-occurrence across patients.

CONCLUSIONS

Implementation of the annotation schema demonstrated strong potential for inferring OUD severity based on key information in a small set of clinical notes and highlighting where such information is documented. These advancements will facilitate NLP tool development to improve OUD prevention, diagnosis, and treatment.

LATTE: Label-efficient incident phenotyping from longitudinal electronic health records

Electronic health record (EHR) data are increasingly used to support real-world evidence studies but are limited by the lack of precise timings of clinical events. Here, we propose a label-efficient incident phenotyping (LATTE) algorithm to accurately annotate the timing of clinical events from longitudinal EHR data. By leveraging the pre-trained semantic embeddings, LATTE selects predictive features and compresses their information into longitudinal visit embeddings through visit attention learning. LATTE models the sequential dependency between the target event and visit embeddings to derive the timings. To improve label efficiency, LATTE constructs longitudinal silver-standard labels from unlabeled patients to perform semi-supervised training. LATTE is evaluated on the onset of type 2 diabetes, heart failure, and relapses of multiple sclerosis. LATTE consistently achieves substantial improvements over benchmark methods while providing high prediction interpretability. The event timings are shown to help discover risk factors of heart failure among patients with rheumatoid arthritis.

Algorithmic identification of persons with dementia for research recruitment: ethical considerations

Underdiagnosis, misdiagnosis, and patterns of social inequality that translate into unequal access to health systems all pose barriers to identifying and recruiting diverse and representative populations into research on Alzheimer's disease and Alzheimer's disease related dementias. In response, some have turned to algorithms to identify patients living with dementia using information that is associated with this condition but that is not as specific as a diagnosis. This paper explains six ethical issues associated with the use of such algorithms including the generation of new, sensitive, identifiable medical information for research purposes without participant consent, issues of justice and equity, risk, and ethical communication. It concludes with a discussion of strategies for addressing these issues and prompting valuable research.

State-Level Hypertension Prevalence and Control Among Adults in the U.S

INTRODUCTION

Improving hypertension control is a national priority. Electronic health record data have the potential to augment traditional surveillance systems. This study aimed to assess hypertension prevalence and control at the state level using a previously established electronic health record-based phenotype for hypertension.

METHODS

Adult patients (N=11,031,368) were included from the IQVIA ambulatory electronic medical record-U.S. 2019 data set. IQVIA ambulatory electronic medical record comprises electronic health records from >100,000 providers and includes patients from every U.S. state and Washington DC. Authors compared hypertension prevalence and control estimates against those from the Behavioral Risk Factor Surveillance System 2019. Results were age-standardized and stratified by state and sociodemographic characteristics. Statistical analyses were conducted in 2022-2023.

RESULTS

IQVIA ambulatory electronic medical record-U.S. patients had a median age of 55 years, and 56.7% were women. Overall age-standardized hypertension prevalence was higher in IQVIA ambulatory electronic medical record-U.S. (35.0%) than in the Behavioral Risk Factor Surveillance System (29.7%), however, state-level geographic patterns were similar, with the highest burden in the South and Appalachia. Similar patterns were also observed by sociodemographic characteristics in both data sets: hypertension prevalence was higher in older age groups (than younger), men (than women), and Black patients (than other races). Hypertension control varied widely across states: among states with >1% data coverage, control rates were lowest in Nevada (51.1%), Washington DC (52.0%), and Mississippi (55.2%); highest in Kansas (73.4%), New Jersey (72.3%), and Iowa (71.9%).

CONCLUSIONS

This study provided the first-ever estimates of hypertension control for all states and Washington DC. Electronic health record-based surveillance could support hypertension prevention and control efforts at the state level.

Scalable and interpretable alternative to chart review for phenotype evaluation using standardized structured data from electronic health records

OBJECTIVES

Chart review as the current gold standard for phenotype evaluation cannot support observational research on electronic health records and claims data sources at scale. We aimed to evaluate the ability of structured data to support efficient and interpretable phenotype evaluation as an alternative to chart review.

MATERIALS AND METHODS

We developed Knowledge-Enhanced Electronic Profile Review (KEEPER) as a phenotype evaluation tool that extracts patient's structured data elements relevant to a phenotype and presents them in a standardized fashion following clinical reasoning principles. We evaluated its performance (interrater agreement, intermethod agreement, accuracy, and review time) compared to manual chart review for 4 conditions using randomized 2-period, 2-sequence crossover design.

RESULTS

Case ascertainment with KEEPER was twice as fast compared to manual chart review. 88.1% of the patients were classified concordantly using charts and KEEPER, but agreement varied depending on the condition. Missing data and differences in interpretation accounted for most of the discrepancies. Pairs of clinicians agreed in case ascertainment in 91.2% of the cases when using KEEPER compared to 76.3% when using charts. Patient classification aligned with the gold standard in 88.1% and 86.9% of the cases respectively.

CONCLUSION

Structured data can be used for efficient and interpretable phenotype evaluation if they are limited to relevant subset and organized according to the clinical reasoning principles. A system that implements these principles can achieve noninferior performance compared to chart review at a fraction of time.

Performance Characteristics of a Rule-Based Electronic Health Record Algorithm to Identify Patients with Gross and Microscopic Hematuria

BACKGROUND

Two million patients per year are referred to urologists for hematuria, or blood in the urine. The American Urological Association recently adopted a risk-stratified hematuria evaluation guideline to limit multi-phase computed tomography to individuals at highest risk of occult malignancy.

OBJECTIVES

To understand population-level hematuria evaluations, we developed an algorithm to accurately identify hematuria cases from electronic health records (EHRs).

METHODS

We used International Classification of Diseases (ICD)-9/ICD-10 diagnosis codes, urine color, and urine microscopy values to identify hematuria cases and to differentiate between gross and microscopic hematuria. Using an iterative process, we refined the ICD-9 algorithm on a gold standard, chart-reviewed cohort of 3,094 hematuria cases, and the ICD-10 algorithm on a 300 patient cohort. We applied the algorithm to Geisinger patients ≥35 years (n = 539,516) and determined performance by conducting chart review (n = 500).

RESULTS

After applying the hematuria algorithm, we identified 51,500 hematuria cases and 488,016 clean controls. Of the hematuria cases, 11,435 were categorized as gross, 26,658 as microscopic, 12,562 as indeterminate, and 845 were uncategorized. The positive predictive value (PPV) of identifying hematuria cases using the algorithm was 100% and the negative predictive value (NPV) was 99%. The gross hematuria algorithm had a PPV of 100% and NPV of 99%. The microscopic hematuria algorithm had lower PPV of 78% and NPV of 100%.

CONCLUSION

We developed an algorithm utilizing diagnosis codes and urine laboratory values to accurately identify hematuria and categorize as gross or microscopic in EHRs. Applying the algorithm will help researchers to understand patterns of care for this common condition.

Identification of opioid use disorder using electronic health records: Beyond diagnostic codes

BACKGROUND

We used structured and unstructured electronic health record (EHR) data to develop and validate an approach to identify moderate/severe opioid use disorder (OUD) that includes individuals without prescription opioid use or chronic pain, an underrepresented population.

METHODS

Using electronic diagnosis grouper text from EHRs of ~1 million patients (2012-2020), we created indicators of OUD-with "tiers" indicating OUD likelihood-combined with OUD medication (MOUD) orders. We developed six sub-algorithms with varying criteria (multiple vs single MOUD orders, multiple vs single tier 1 indicators, tier 2 indicators, tier 3 and 4 indicators). Positive predictive values (PPVs) were calculated based on chart review to determine OUD status and severity. We compared demographic and clinical characteristics of cases identified by the sub-algorithms.

RESULTS

In total, 14,852 patients met criteria for one of the sub-algorithms. Five sub-algorithms had PPVs ≥0.90 for any severity OUD; four had PPVs ≥0.90 for moderate/severe OUD. Demographic and clinical characteristics differed substantially between groups. Of identified OUD cases, 31.3% had no past opioid analgesic orders, 79.7% lacked evidence of chronic prescription opioid use, and 43.5% lacked a chronic pain diagnosis.

DISCUSSION

Incorporating unstructured data with MOUD orders yielded an approach that adequately identified moderate/severe OUD, identified unique demographic and clinical sub-groups, and included individuals without prescription opioid use or chronic pain, whose OUD may stem from illicit opioids. Findings show that incorporating unstructured data strengthens EHR algorithms for identifying OUD and suggests approaches limited to populations with prescription opioid use or chronic pain exclude many individuals with OUD.

Estimated Prevalence and Testing for Albuminuria in US Adults at Risk for Chronic Kidney Disease

Importance

Albuminuria testing is crucial for guiding evidence-based treatments to mitigate chronic kidney disease (CKD) progression and cardiovascular morbidity, but it is widely underutilized among persons with or at risk for CKD.

Objective

To estimate the extent of albuminuria underdetection from lack of testing and evaluate its association with CKD treatment in a large US cohort of patients with hypertension or diabetes.

Design, Setting, and Participants

This cohort study examined adults with hypertension or diabetes, using data from the 2007 to 2018 National Health and Nutrition Examination Surveys (NHANES) and the Optum deidentified electronic health record (EHR) data set of diverse US health care organizations. Analyses were conducted from October 31, 2022, to May 19, 2023.

Main Outcomes and Measures

Using NHANES as a nationally representative sample, a logistic regression model was developed to estimate albuminuria (urine albumin-creatinine ratio ≥30 mg/g). This model was then applied to active outpatients in the EHR from January 1, 2017, to December 31, 2018. The prevalence of albuminuria among those with and without albuminuria testing during this period was estimated. A multivariable logistic regression was used to examine associations between having albuminuria testing and CKD therapies within the subsequent year (prescription for angiotensin-converting enzyme inhibitor [ACEi] or angiotensin II receptor blocker [ARB], prescription for sodium-glucose cotransporter 2 inhibitor [SGLT2i], and blood pressure control to less than 130/80 mm Hg or less than 140/90 mm Hg on the latest outpatient measure).

Results

The total EHR study population included 192 108 patients (mean [SD] age, 60.3 [15.1] years; 185 589 [96.6%] with hypertension; 50 507 [26.2%] with diabetes; mean [SD] eGFR, 84 [21] mL/min/1.73 m2). There were 33 629 patients (17.5%) who had albuminuria testing; of whom 11 525 (34.3%) had albuminuria. Among 158 479 patients who were untested, the estimated albuminuria prevalence rate was 13.4% (n = 21 231). Thus, only 35.2% (11 525 of 32 756) of the projected population with albuminuria had been tested. Albuminuria testing was associated with higher adjusted odds of receiving ACEi or ARB treatment (OR, 2.39 [95% CI, 2.32-2.46]), SGLT2i treatment (OR, 8.22 [95% CI, 7.56-8.94]), and having blood pressure controlled to less than 140/90 mm Hg (OR, 1.20 [95% CI, 1.16-1.23]).

Conclusions and Relevance

In this cohort study of patients with hypertension or diabetes, it was estimated that approximately two-thirds of patients with albuminuria were undetected due to lack of testing. These results suggest that improving detection of CKD with albuminuria testing represents a substantial opportunity to optimize care delivery for reducing CKD progression and cardiovascular complications.

A metadata framework for computational phenotypes

With the burgeoning development of computational phenotypes, it is increasingly difficult to identify the right phenotype for the right tasks. This study uses a mixed-methods approach to develop and evaluate a novel metadata framework for retrieval of and reusing computational phenotypes. Twenty active phenotyping researchers from 2 large research networks, Electronic Medical Records and Genomics and Observational Health Data Sciences and Informatics, were recruited to suggest metadata elements. Once consensus was reached on 39 metadata elements, 47 new researchers were surveyed to evaluate the utility of the metadata framework. The survey consisted of 5-Likert multiple-choice questions and open-ended questions. Two more researchers were asked to use the metadata framework to annotate 8 type-2 diabetes mellitus phenotypes. More than 90% of the survey respondents rated metadata elements regarding phenotype definition and validation methods and metrics positively with a score of 4 or 5. Both researchers completed annotation of each phenotype within 60 min. Our thematic analysis of the narrative feedback indicates that the metadata framework was effective in capturing rich and explicit descriptions and enabling the search for phenotypes, compliance with data standards, and comprehensive validation metrics. Current limitations were its complexity for data collection and the entailed human costs.

A platform for phenotyping disease progression and associated longitudinal risk factors in large-scale EHRs, with application to incident diabetes complications in the UK Biobank

Objective

Modern healthcare data reflect massive multi-level and multi-scale information collected over many years. The majority of the existing phenotyping algorithms use case-control definitions of disease. This paper aims to study the time to disease onset and progression and identify the time-varying risk factors that drive them.

Materials and Methods

We developed an algorithmic approach to phenotyping the incidence of diseases by consolidating data sources from the UK Biobank (UKB), including primary care electronic health records (EHRs). We focused on defining events, event dates, and their censoring time, including relevant terms and existing phenotypes, excluding generic, rare, or semantically distant terms, forward-mapping terminology terms, and expert review. We applied our approach to phenotyping diabetes complications, including a composite cardiovascular disease (CVD) outcome, diabetic kidney disease (DKD), and diabetic retinopathy (DR), in the UKB study.

Results

We identified 49 049 participants with diabetes. Among them, 1023 had type 1 diabetes (T1D), and 40 193 had type 2 diabetes (T2D). A total of 23 833 diabetes subjects had linked primary care records. There were 3237, 3113, and 4922 patients with CVD, DKD, and DR events, respectively. The risk prediction performance for each outcome was assessed, and our results are consistent with the prediction area under the ROC (receiver operating characteristic) curve (AUC) of standard risk prediction models using cohort studies.

Discussion and Conclusion

Our publicly available pipeline and platform enable streamlined curation of incidence events, identification of time-varying risk factors underlying disease progression, and the definition of a relevant cohort for time-to-event analyses. These important steps need to be considered simultaneously to study disease progression.

Familial Hypercholesterolemia in the Electronic Medical Records and Genomics Network: Prevalence, Penetrance, Cardiovascular Risk, and Outcomes After Return of Results

BACKGROUND

The implications of secondary findings detected in large-scale sequencing projects remain uncertain. We assessed prevalence and penetrance of pathogenic familial hypercholesterolemia (FH) variants, their association with coronary heart disease (CHD), and 1-year outcomes following return of results in phase III of the electronic medical records and genomics network.

METHODS

Adult participants (n=18 544) at 7 sites were enrolled in a prospective cohort study to assess the clinical impact of returning results from targeted sequencing of 68 actionable genes, including LDLR, APOB, and PCSK9. FH variant prevalence and penetrance (defined as low-density lipoprotein cholesterol >155 mg/dL) were estimated after excluding participants enrolled on the basis of hypercholesterolemia. Multivariable logistic regression was used to estimate the odds of CHD compared to age- and sex-matched controls without FH-associated variants. Process (eg, referral to a specialist or ordering new tests), intermediate (eg, new diagnosis of FH), and clinical (eg, treatment modification) outcomes within 1 year after return of results were ascertained by electronic health record review.

RESULTS

The prevalence of FH-associated pathogenic variants was 1 in 188 (69 of 13,019 unselected participants). Penetrance was 87.5%. The presence of an FH variant was associated with CHD (odds ratio, 3.02 [2.00-4.53]) and premature CHD (odds ratio, 3.68 [2.34-5.78]). At least 1 outcome occurred in 92% of participants; 44% received a new diagnosis of FH and 26% had treatment modified following return of results.

CONCLUSIONS

In a multisite cohort of electronic health record-linked biobanks, monogenic FH was prevalent, penetrant, and associated with presence of CHD. Nearly half of participants with an FH-associated variant received a new diagnosis of FH and a quarter had treatment modified after return of results. These results highlight the potential utility of sequencing electronic health record-linked biobanks to detect FH.

Trends and opportunities in computable clinical phenotyping: A scoping review

Identifying patient cohorts meeting the criteria of specific phenotypes is essential in biomedicine and particularly timely in precision medicine. Many research groups deliver pipelines that automatically retrieve and analyze data elements from one or more sources to automate this task and deliver high-performing computable phenotypes. We applied a systematic approach based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines to conduct a thorough scoping review on computable clinical phenotyping. Five databases were searched using a query that combined the concepts of automation, clinical context, and phenotyping. Subsequently, four reviewers screened 7960 records (after removing over 4000 duplicates) and selected 139 that satisfied the inclusion criteria. This dataset was analyzed to extract information on target use cases, data-related topics, phenotyping methodologies, evaluation strategies, and portability of developed solutions. Most studies supported patient cohort selection without discussing the application to specific use cases, such as precision medicine. Electronic Health Records were the primary source in 87.1 % (N = 121) of all studies, and International Classification of Diseases codes were heavily used in 55.4 % (N = 77) of all studies, however, only 25.9 % (N = 36) of the records described compliance with a common data model. In terms of the presented methods, traditional Machine Learning (ML) was the dominant method, often combined with natural language processing and other approaches, while external validation and portability of computable phenotypes were pursued in many cases. These findings revealed that defining target use cases precisely, moving away from sole ML strategies, and evaluating the proposed solutions in the real setting are essential opportunities for future work. There is also momentum and an emerging need for computable phenotyping to support clinical and epidemiological research and precision medicine.

Using a gradient boosted model for case ascertainment from free-text veterinary records

Case ascertainment for prevalence and incidence studies from veterinary clinical data poses a major challenge because medical notes are not consistently structured or complete. Using natural language processing (NLP) and machine learning, this study aimed to obtain accurate case recognition for feline upper respiratory tract infections (primarily caused by viruses such as feline herpes virus (FHV-1) and feline calici virus (FCV), and bacteria such as Chlamydophila felis, Mycoplasma felis and Bordetella bronchiseptica using retrospective electronic veterinary records from the Royal Society for Prevention of Cruelty to Animals, Queensland (RSPCA Qld). Data cleaning and NLP on eight years of free-text veterinary records from RSPCA Queensland was carried out to derive text-based predictors. The NLP steps included sorting records by length of stay, vectorising, tokenising and spell checking against a bespoke veterinary database. A gradient boosted model (GBM) was trained to predict the probability of each animal having a diagnosis of upper respiratory infection. A manually annotated dataset was used for training the algorithm to learn dominant patterns between predictors (frequencies of n-grams) and responses (manual binary case classification). The GBM's performance was tested against an out of sample validation dataset, and model agnostics were used to interrogate the model's learning process. The GBM used patient-level frequencies of 1250 unique n-grams as predictor variables and was able to predict the probability of cases in the validation dataset with an accuracy of 0.95 (95% CI 0.92, 0.97) and F1 score of 0.96. Predictors that exerted the highest influence on the model included frequencies of "doxycycline", "flu", "sneezing", "doxybrom" and "ocular". The trained GBM was deployed on the full dataset spanning eight years, comprising 60,258 clinical entries. The prevalence in the full dataset was predicted to be 23.59%, which is in line with domain expertise from practicing veterinarians at the shelter. Case ascertainment is a crucial step for further epidemiological study of cat flu. Ultimately, this tool can be extended to other clinical procedures, conditions, and diseases such as intensive care treatment due to snake bites and tick paralysis, physical injuries such as orthopaedic fractures or chest injuries and labour-intensive infectious diseases like parvovirus, canine cough, and ringworm, all of which require prolonged quarantine and care.

Evaluating resources composing the PheMAP knowledge base to enhance high-throughput phenotyping

OBJECTIVE

A previous study, PheMAP, combined independent, online resources to enable high-throughput phenotyping (HTP) using electronic health records (EHRs). However, online resources offer distinct quality descriptions of diseases which may affect phenotyping performance. We aimed to evaluate the phenotyping performance of single resource-based PheMAPs and investigate an optimized strategy for HTP.

MATERIALS AND METHODS

We compared how each resource produced top-ranked concept unique identifiers (CUIs) by term frequency-inverse document frequency with Jaccard matrices comparing single resources and the original PheMAP. We correlated top-ranked concepts from each resource to features used in established Phenotype KnowledgeBase (PheKB) algorithms for hypothyroidism, type II diabetes mellitus (T2DM), and dementias. Using resources separately, we calculated multiple phenotype risk scores for individuals from Vanderbilt University Medical Center's BioVU DNA Biobank and compared phenotyping performance against rule-based eMERGE algorithms. Lastly, we implemented an ensemble strategy which classified patient case/control status based upon PheMAP resource agreement.

RESULTS

Jaccard similarity matrices indicate that the similarity of CUIs comprising single resource-based PheMAPs varies. Single resource-based PheMAPs generated from MedlinePlus and MedicineNet outperformed others but only encompass 81.6% of overall disease phenotypes. We propose the PheMAP-Ensemble which provides higher average accuracy and precision than the combined average accuracy and precision of single resource-based PheMAPs. While offering complete phenotype coverage, PheMAP-Ensemble significantly increases phenotyping recall compared to the original iteration.

CONCLUSIONS

Resources comprising the PheMAP produce different phenotyping performance when implemented individually. The ensemble method significantly improves the quality of PheMAP by fully utilizing dissimilar resources to capture accurate phenotyping data from EHRs.

Characterizing variability of electronic health record-driven phenotype definitions

OBJECTIVE

The aim of this study was to analyze a publicly available sample of rule-based phenotype definitions to characterize and evaluate the variability of logical constructs used.

MATERIALS AND METHODS

A sample of 33 preexisting phenotype definitions used in research that are represented using Fast Healthcare Interoperability Resources and Clinical Quality Language (CQL) was analyzed using automated analysis of the computable representation of the CQL libraries.

RESULTS

Most of the phenotype definitions include narrative descriptions and flowcharts, while few provide pseudocode or executable artifacts. Most use 4 or fewer medical terminologies. The number of codes used ranges from 5 to 6865, and value sets from 1 to 19. We found that the most common expressions used were literal, data, and logical expressions. Aggregate and arithmetic expressions are the least common. Expression depth ranges from 4 to 27.

DISCUSSION

Despite the range of conditions, we found that all of the phenotype definitions consisted of logical criteria, representing both clinical and operational logic, and tabular data, consisting of codes from standard terminologies and keywords for natural language processing. The total number and variety of expressions are low, which may be to simplify implementation, or authors may limit complexity due to data availability constraints.

CONCLUSIONS

The phenotype definitions analyzed show significant variation in specific logical, arithmetic, and other operators but are all composed of the same high-level components, namely tabular data and logical expressions. A standard representation for phenotype definitions should support these formats and be modular to support localization and shared logic.

Evaluation of the portability of computable phenotypes with natural language processing in the eMERGE network

The electronic Medical Records and Genomics (eMERGE) Network assessed the feasibility of deploying portable phenotype rule-based algorithms with natural language processing (NLP) components added to improve performance of existing algorithms using electronic health records (EHRs). Based on scientific merit and predicted difficulty, eMERGE selected six existing phenotypes to enhance with NLP. We assessed performance, portability, and ease of use. We summarized lessons learned by: (1) challenges; (2) best practices to address challenges based on existing evidence and/or eMERGE experience; and (3) opportunities for future research. Adding NLP resulted in improved, or the same, precision and/or recall for all but one algorithm. Portability, phenotyping workflow/process, and technology were major themes. With NLP, development and validation took longer. Besides portability of NLP technology and algorithm replicability, factors to ensure success include privacy protection, technical infrastructure setup, intellectual property agreement, and efficient communication. Workflow improvements can improve communication and reduce implementation time. NLP performance varied mainly due to clinical document heterogeneity; therefore, we suggest using semi-structured notes, comprehensive documentation, and customization options. NLP portability is possible with improved phenotype algorithm performance, but careful planning and architecture of the algorithms is essential to support local customizations.

Assess the documentation of cognitive tests and biomarkers in electronic health records via natural language processing for Alzheimer's disease and related dementias

BACKGROUND

Cognitive tests and biomarkers are the key information to assess the severity and track the progression of Alzheimer's' disease (AD) and AD-related dementias (AD/ADRD), yet, both are often only documented in clinical narratives of patients' electronic health records (EHRs). In this work, we aim to (1) assess the documentation of cognitive tests and biomarkers in EHRs that can be used as real-world endpoints, and (2) identify, extract, and harmonize the different commonly used cognitive tests from clinical narratives using natural language processing (NLP) methods into categorical AD/ADRD severity.

METHODS

We developed a rule-based NLP pipeline to extract the cognitive tests and biomarkers from clinical narratives in AD/ADRD patients' EHRs. We aggregated the extracted results to the patient level and harmonized the cognitive test scores into severity categories using cutoffs determined based on both relevant literature and domain knowledge of AD/ADRD clinicians.

RESULTS

We identified an AD/ADRD cohort of 48,912 patients from the University of Florida (UF) Health system and identified 7 measurements (6 cognitive tests and 1 biomarker) that are frequently documented in our data. Our NLP pipeline achieved an overall F1-score of 0.9059 across the 7 measurements. Among the 6 cognitive tests, we were able to harmonize 4 cognitive test scores into severity categories, and the population characteristics of patients with different severity were described. We also identified several factors related to the availability of their documentation in EHRs.

CONCLUSION

This study demonstrates that our NLP pipelines can extract cognitive tests and biomarkers of AD/ADRD accurately for downstream studies. Although, the documentation of cognitive tests and biomarkers in EHRs appears to be low, RWD is still an important resource for AD/ADRD research. Nevertheless, providing standardized approach to document cognitive tests and biomarkers in EHRS are also warranted.

Machine learning approaches for electronic health records phenotyping: a methodical review

OBJECTIVE

Accurate and rapid phenotyping is a prerequisite to leveraging electronic health records for biomedical research. While early phenotyping relied on rule-based algorithms curated by experts, machine learning (ML) approaches have emerged as an alternative to improve scalability across phenotypes and healthcare settings. This study evaluates ML-based phenotyping with respect to (1) the data sources used, (2) the phenotypes considered, (3) the methods applied, and (4) the reporting and evaluation methods used.

MATERIALS AND METHODS

We searched PubMed and Web of Science for articles published between 2018 and 2022. After screening 850 articles, we recorded 37 variables on 100 studies.

RESULTS

Most studies utilized data from a single institution and included information in clinical notes. Although chronic conditions were most commonly considered, ML also enabled the characterization of nuanced phenotypes such as social determinants of health. Supervised deep learning was the most popular ML paradigm, while semi-supervised and weakly supervised learning were applied to expedite algorithm development and unsupervised learning to facilitate phenotype discovery. ML approaches did not uniformly outperform rule-based algorithms, but deep learning offered a marginal improvement over traditional ML for many conditions.

DISCUSSION

Despite the progress in ML-based phenotyping, most articles focused on binary phenotypes and few articles evaluated external validity or used multi-institution data. Study settings were infrequently reported and analytic code was rarely released.

CONCLUSION

Continued research in ML-based phenotyping is warranted, with emphasis on characterizing nuanced phenotypes, establishing reporting and evaluation standards, and developing methods to accommodate misclassified phenotypes due to algorithm errors in downstream applications.

Federated Learning in Health care Using Structured Medical Data

The success of machine learning-based studies is largely subjected to accessing a large amount of data. However, accessing such data is typically not feasible within a single health system/hospital. Although multicenter studies are the most effective way to access a vast amount of data, sharing data outside the institutes involves legal, business, and technical challenges. Federated learning (FL) is a newly proposed machine learning framework for multicenter studies, tackling data-sharing issues across participant institutes. The promise of FL is simple. FL facilitates multicenter studies without losing data access control and allows the construction of a global model by aggregating local models trained from participant institutes. This article reviewed recently published studies that utilized FL in clinical studies with structured medical data. In addition, challenges and open questions in FL in clinical studies with structured medical data were discussed.

Sex Differences in Health Conditions Associated with Sexual Assault in a Large Hospital Population

Introduction

Sexual assault is an urgent public health concern with both immediate and long-lasting health consequences, affecting 44% of women and 25% of men during their lifetimes. Large studies are needed to understand the unique healthcare needs of this patient population.

Methods

We mined clinical notes to identify patients with a history of sexual assault in the electronic health record (EHR) at Vanderbilt University Medical Center (VUMC), a large university hospital in the Southeastern USA, from 1989 to 2021 (N = 3,376,424). Using a phenome-wide case-control study, we identified diagnoses co-occurring with disclosures of sexual assault. We performed interaction tests to examine whether sex modified any of these associations. Association analyses were restricted to a subset of patients receiving regular care at VUMC (N = 833,185).

Results

The phenotyping approach identified 14,496 individuals (0.43%) across the VUMC-EHR with documentation of sexual assault and achieved a positive predictive value of 93.0% (95% confidence interval = 85.6-97.0%), determined by manual patient chart review. Out of 1,703 clinical diagnoses tested across all subgroup analyses, 465 were associated with sexual assault. Sex-by-trauma interaction analysis revealed 55 sex-differential associations and demonstrated increased odds of psychiatric diagnoses in male survivors.

Discussion

This case-control study identified associations between disclosures of sexual assault and hundreds of health conditions, many of which demonstrated sex-differential effects. The findings of this study suggest that patients who have experienced sexual assault are at risk for developing wide-ranging medical and psychiatric comorbidities and that male survivors may be particularly vulnerable to developing mental illness.

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.

Prescription Patterns of Cardiovascular- and Kidney-Protective Therapies Among Patients With Type 2 Diabetes and Chronic Kidney Disease

OBJECTIVE

To assess the prevalence and correlates of prescription of sodium-glucose cotransporter 2 inhibitors (SGLT2i) and/or glucagon-like peptide 1 receptor agonists (GLP1-RA) in individuals with type 2 diabetes mellitus (T2DM) with and without chronic kidney disease (CKD).

RESEARCH DESIGN AND METHODS

This was a cross-sectional analyses of SGLT2i and GLP1-RA prescriptions from 1 January 2019 to 31 December 2020 in the Veterans Health Administration System. The likelihood of prescriptions was examined by the presence or absence of CKD and by predicted risks of atherosclerotic cardiovascular disease (ASCVD) and end-stage kidney disease (ESKD).

RESULTS

Of 1,197,880 adults with T2DM, SGLT2i and GLP1-RA were prescribed to 11% and 8% of patients overall, and to 12% and 10% of those with concomitant CKD, respectively. In adjusted models, patients with severe albuminuria were less likely to be prescribed SGLT2i or GLP1-RA versus nonalbuminuric patients with CKD, with odds ratios (ORs) of 0.91 (95% CI 0.89, 0.93) and 0.97 (0.94, 1.00), respectively. Patients with a 10-year ASCVD risk >20% (vs. <5%), had lower odds of SGLT2i use (OR 0.66 [0.61, 0.71]) and GLP1-RA prescription (OR 0.55 [0.52, 0.59]). A 5-year ESKD risk >5%, compared with <1%, was associated with lower likelihood of SGLT2i prescription (OR 0.63 [0.59, 0.67]) but higher likelihood of GLP1-RA prescription (OR 1.53 [1.46, 1.61]).

CONCLUSIONS

Among a large cohort of patients with T2DM, prescription of SGLT2i and GLP1-RA was low in those with CKD. We observed a "risk-treatment paradox," whereby patients with higher risk of adverse outcomes were less likely to receive these therapies.

A semi-supervised adaptive Markov Gaussian embedding process (SAMGEP) for prediction of phenotype event times using the electronic health record

While there exist numerous methods to identify binary phenotypes (i.e. COPD) using electronic health record (EHR) data, few exist to ascertain the timings of phenotype events (i.e. COPD onset or exacerbations). Estimating event times could enable more powerful use of EHR data for longitudinal risk modeling, including survival analysis. Here we introduce Semi-supervised Adaptive Markov Gaussian Embedding Process (SAMGEP), a semi-supervised machine learning algorithm to estimate phenotype event times using EHR data with limited observed labels, which require resource-intensive chart review to obtain. SAMGEP models latent phenotype states as a binary Markov process, and it employs an adaptive weighting strategy to map timestamped EHR features to an embedding function that it models as a state-dependent Gaussian process. SAMGEP's feature weighting achieves meaningful feature selection, and its predictions significantly improve AUCs and F1 scores over existing approaches in diverse simulations and real-world settings. It is particularly adept at predicting cumulative risk and event counting process functions, and is robust to diverse generative model parameters. Moreover, it achieves high accuracy with few (50-100) labels, efficiently leveraging unlabeled EHR data to maximize information gain from costly-to-obtain event time labels. SAMGEP can be used to estimate accurate phenotype state functions for risk modeling research.

A Framework for Automating Psychiatric Distress Screening in Ophthalmology Clinics Using an EHR-Derived AI Algorithm

Purpose

In patients with ophthalmic disorders, psychosocial risk factors play an important role in morbidity and mortality. Proper and early psychiatric screening can result in prompt intervention and mitigate its impact. Because screening is resource intensive, we developed a framework for automating screening using an electronic health record (EHR)-derived artificial intelligence (AI) algorithm.

Methods

Subjects came from the Duke Ophthalmic Registry, a retrospective EHR database for the Duke Eye Center. Inclusion criteria included at least two encounters and a minimum of 1 year of follow-up. Presence of distress was defined at the encounter level using a computable phenotype. Risk factors included available EHR history. At each encounter, risk factors were used to discriminate psychiatric status. Model performance was evaluated using area under the receiver operating characteristic (ROC) curve and area under the precision-recall curve (PR AUC). Variable importance was presented using odds ratios (ORs).

Results

Our cohort included 358,135 encounters from 40,326 patients with an average of nine encounters per patient over 4 years. The ROC and PR AUC were 0.91 and 0.55, respectively. Of the top 25 predictors, the majority were related to existing distress, but some indicated stressful conditions, including chemotherapy (OR = 1.36), esophageal disorders (OR = 1.31), central pain syndrome (OR = 1.25), and headaches (OR = 1.24).

Conclusions

Psychiatric distress in ophthalmology patients can be monitored passively using an AI algorithm trained on existing EHR data.

Translational Relevance

When paired with an effective referral and treatment program, such algorithms may improve health outcomes in ophthalmology.

Sex hormones, body mass index, and related comorbidities associated with developing Sjögren's disease: a nested case-control study

OBJECTIVE

Sjögren's disease (SjD), a highly female predominant systemic autoimmune disease, peaks in perimenopause. Prior studies lack details on timing or type of sex hormone exposure. We examined SjD risk using endogenous and exogenous hormone exposure and related comorbidities.

METHODS

We performed a retrospective case-control study of adult women, nested within a population cohort. Cases had SjD diagnosed by a rheumatology provider or two SjD diagnoses from a non-rheumatology provider with a positive anti-SSA antibody or salivary gland biopsy. Cases were age-matched to three SjD-free controls. We calculated modified composite estrogen scores (mCES) and collected demographics, comorbidities, and endogenous and exogenous hormone exposures. Risk ratios were adjusted for demographics.

RESULTS

Of 546 SjD cases and 1637 age-matched controls, mCES was not significantly associated with SjD in adjusted models. The top individual hormone exposures associated with SjD included estrogen replacement therapy (ERT; RR 1.78 [95% CI 1.47-2.14]), polycystic ovarian syndrome (1.65 [1.28-2.12]), and hysterectomy without bilateral oophorectomy (1.51 [1.13-2.03]). We identified comorbidities preceding SjD including fibromyalgia, pulmonary disease, diabetes, lymphoma, osteoporosis, peripheral vascular disease, and renal disease. Taking comorbidities into account, we developed a predictive model for SjD that included fibromyalgia (2.50 [1.93-3.25]), osteoporosis (1.84 [1.27-2.66]), hormone replacement therapy (HRT) (1.61 [1.22-2.12]), diabetes (0.27 [0.13-0.50]), and body mass index (BMI) (0.97 [0.95-0.99]).

CONCLUSIONS

We report a novel algorithm to improve identifying patients at risk for SjD and describe sex hormone association with SjD. Finally, we report new comorbidities associated with SjD decrease, BMI and diabetes, and increase, lymphoma and osteoporosis.. Key Points •Given female predominance and typical perimenopausal onset, sex hormones should be considered when studying comorbidities in Sjögren's disease. •The top exposures associated with developing Sjögren's disease included fibromyalgia, osteoporosis, and use of hormone replacement therapy. Possible protective factors included prior diabetes and higher body mass index. •We used our newly identified exposures to generate a predictive algorithm, which has potential to improve diagnosis and pathogenic insights into Sjögren's disease.

Not all phenotypes are created equal: covariates of success in e-phenotype specification

BACKGROUND

Electronic (e)-phenotype specification by noninformaticist investigators remains a challenge. Although validation of each patient returned by e-phenotype could ensure accuracy of cohort representation, this approach is not practical. Understanding the factors leading to successful e-phenotype specification may reveal generalizable strategies leading to better results.

MATERIALS AND METHODS

Noninformaticist experts (n = 21) were recruited to produce expert-mediated e-phenotypes using i2b2 assisted by a honest data-broker and a project coordinator. Patient- and visit-sets were reidentified and a random sample of 20 charts matching each e-phenotype was returned to experts for chart-validation. Attributes of the queries and expert characteristics were captured and related to chart-validation rates using generalized linear regression models.

RESULTS

E-phenotype validation rates varied according to experts' domains and query characteristics (mean = 61%, range 20-100%). Clinical domains that performed better included infectious, rheumatic, neonatal, and cancers, whereas other domains performed worse (psychiatric, GI, skin, and pulmonary). Match-rate was negatively impacted when specification of temporal constraints was required. In general, the increase in e-phenotype specificity contributed positively to match-rate.

DISCUSSIONS AND CONCLUSIONS

Clinical experts and informaticists experience a variety of challenges when building e-phenotypes, including the inability to differentiate clinical events from patient characteristics or appropriately configure temporal constraints; a lack of access to available and quality data; and difficulty in specifying routes of medication administration. Biomedical query mediation by informaticists and honest data-brokers in designing e-phenotypes cannot be overstated. Although tools such as i2b2 may be widely available to noninformaticists, successful utilization depends not on users' confidence, but rather on creating highly specific e-phenotypes.

Association of Race and Ethnicity With Prescription of SGLT2 Inhibitors and GLP1 Receptor Agonists Among Patients With Type 2 Diabetes in the Veterans Health Administration System

IMPORTANCE

Novel therapies for type 2 diabetes can reduce the risk of cardiovascular disease and chronic kidney disease progression. The equitability of these agents' prescription across racial and ethnic groups has not been well-evaluated.

OBJECTIVE

To investigate differences in the prescription of sodium-glucose cotransporter-2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP-1 RA) among adult patients with type 2 diabetes by racial and ethnic groups.

DESIGN, SETTING, AND PARTICIPANTS

Cross-sectional analysis of data from the US Veterans Health Administration's Corporate Data Warehouse. The sample included adult patients with type 2 diabetes and at least 2 primary care clinic visits from January 1, 2019, to December 31, 2020.

EXPOSURES

Self-identified race and self-identified ethnicity.

MAIN OUTCOMES AND MEASURES

The primary outcomes were prevalent SGLT2i or GLP-1 RA prescription, defined as any active prescription during the study period.

RESULTS

Among 1 197 914 patients (mean age, 68 years; 96% men; 1% American Indian or Alaska Native, 2% Asian, Native Hawaiian, or Other Pacific Islander, 20% Black or African American, 71% White, and 7% of Hispanic or Latino ethnicity), 10.7% and 7.7% were prescribed an SGLT2i or a GLP-1 RA, respectively. Prescription rates for SGLT2i and GLP-1 RA, respectively, were 11% and 8.4% among American Indian or Alaska Native patients; 11.8% and 8% among Asian, Native Hawaiian, or Other Pacific Islander patients; 8.8% and 6.1% among Black or African American patients; and 11.3% and 8.2% among White patients, respectively. Prescription rates for SGLT2i and GLP-1 RA, respectively, were 11% and 7.1% among Hispanic or Latino patients and 10.7% and 7.8% among non-Hispanic or Latino patients. After accounting for patient- and system-level factors, all racial groups had significantly lower odds of SGLT2i and GLP-1 RA prescription compared with White patients. Black patients had the lowest odds of prescription compared with White patients (adjusted odds ratio, 0.72 [95% CI, 0.71-0.74] for SGLT2i and 0.64 [95% CI, 0.63-0.66] for GLP-1 RA). Patients of Hispanic or Latino ethnicity had significantly lower odds of prescription (0.90 [95% CI, 0.88-0.93] for SGLT2i and 0.88 [95% CI, 0.85-0.91] for GLP-1 RA) compared with non-Hispanic or Latino patients.

CONCLUSIONS AND RELEVANCE

Among patients with type 2 diabetes in the Veterans Health Administration system during 2019 and 2020, prescription rates of SGLT2i and GLP-1 RA medications were low, and individuals of several different racial groups and those of Hispanic ethnicity had statistically significantly lower odds of receiving prescriptions for these medications compared with individuals of White race and non-Hispanic ethnicity. Further research is needed to understand the mechanisms underlying these differences in rates of prescribing and the potential relationship with differences in clinical outcomes.

MixEHR-Guided: A guided multi-modal topic modeling approach for large-scale automatic phenotyping using the electronic health record

Electronic Health Records (EHRs) contain rich clinical data collected at the point of the care, and their increasing adoption offers exciting opportunities for clinical informatics, disease risk prediction, and personalized treatment recommendation. However, effective use of EHR data for research and clinical decision support is often hampered by a lack of reliable disease labels. To compile gold-standard labels, researchers often rely on clinical experts to develop rule-based phenotyping algorithms from billing codes and other surrogate features. This process is tedious and error-prone due to recall and observer biases in how codes and measures are selected, and some phenotypes are incompletely captured by a handful of surrogate features. To address this challenge, we present a novel automatic phenotyping model called MixEHR-Guided (MixEHR-G), a multimodal hierarchical Bayesian topic model that efficiently models the EHR generative process by identifying latent phenotype structure in the data. Unlike existing topic modeling algorithms wherein the inferred topics are not identifiable, MixEHR-G uses prior information from informative surrogate features to align topics with known phenotypes. We applied MixEHR-G to an openly-available EHR dataset of 38,597 intensive care patients (MIMIC-III) in Boston, USA and to administrative claims data for a population-based cohort (PopHR) of 1.3 million people in Quebec, Canada. Qualitatively, we demonstrate that MixEHR-G learns interpretable phenotypes and yields meaningful insights about phenotype similarities, comorbidities, and epidemiological associations. Quantitatively, MixEHR-G outperforms existing unsupervised phenotyping methods on a phenotype label annotation task, and it can accurately estimate relative phenotype prevalence functions without gold-standard phenotype information. Altogether, MixEHR-G is an important step towards building an interpretable and automated phenotyping system using EHR data.

New Insights into Clinical and Mechanistic Heterogeneity of the Acute Respiratory Distress Syndrome: Summary of the Aspen Lung Conference 2021

Clinical and molecular heterogeneity are common features of human disease. Understanding the basis for heterogeneity has led to major advances in therapy for many cancers and pulmonary diseases such as cystic fibrosis and asthma. Although heterogeneity of risk factors, disease severity, and outcomes in survivors are common features of the acute respiratory distress syndrome (ARDS), many challenges exist in understanding the clinical and molecular basis for disease heterogeneity and using heterogeneity to tailor therapy for individual patients. This report summarizes the proceedings of the 2021 Aspen Lung Conference, which was organized to review key issues related to understanding clinical and molecular heterogeneity in ARDS. The goals were to review new information about ARDS phenotypes, to explore multicellular and multisystem mechanisms responsible for heterogeneity, and to review how best to account for clinical and molecular heterogeneity in clinical trial design and assessment of outcomes. The report concludes with recommendations for future research to understand the clinical and basic mechanisms underlying heterogeneity in ARDS to advance the development of new treatments for this life-threatening critical illness.

ReviewR: a light-weight and extensible tool for manual review of clinical records

Objectives

Manual record review is a crucial step for electronic health record (EHR)-based research, but it has poor workflows and is error prone. We sought to build a tool that provides a unified environment for data review and chart abstraction data entry.

Materials and Methods

ReviewR is an open-source R Shiny application that can be deployed on a single machine or made available to multiple users. It supports multiple data models and database systems, and integrates with the REDCap API for storing abstraction results.

Results

We describe 2 real-world uses and extensions of ReviewR. Since its release in April 2021 as a package on CRAN it has been downloaded 2204 times.

Discussion and Conclusion

ReviewR provides an easily accessible review interface for clinical data warehouses. Its modular, extensible, and open source nature afford future expansion by other researchers.

Palliative Care and Documented Suicide: Association Among Veterans With High Mortality Risk

CONTEXT

Palliative care consultations (PCCs) are associated with reduced physical and psychological symptoms that are related to suicide risk. Little is known, however, about the association between PCCs and death from suicide among patients at high risk of short-term mortality.

OBJECTIVE

To examine the association between the number of PCCs and documentation of suicide in a cohort of Veterans at high risk of short-term mortality, before and after accounting for Veterans' sociodemographic characteristics and clinical conditions.

METHODS

An observational cohort study was conducted using linked Veterans Affairs clinical and administrative databases for 580,620 decedents with high risk of one-year mortality. Logistic regression models were used to examine the association between number of PCCs and documentation of suicide.

RESULTS

Higher percentages of Veterans who died by suicide were diagnosed with chronic pulmonary disease as well as mental health/substance use conditions compared with Veterans who died from other causes. In adjusted models, one PCC in the 90 days prior to death was significantly associated with a 71% decrease in the odds of suicide (OR = 0.29, 95% CI = 0.23-0.37, P < 0.001) and two or more PCCs were associated with a 78% decrease (OR = 0.22, 95% CI = 0.15-0.33, P < 0.001). Associated "number needed to be exposed" estimates suggest that 421 Veterans in this population would need to receive at least one PCC to prevent one suicide.

CONCLUSION

While acknowledging the importance of specialized mental health care in reducing suicide among high-risk patients, interventions delivered in the context of PCCs may also play a role.

Design and validation of a FHIR-based EHR-driven phenotyping toolbox

OBJECTIVES

To develop and validate a standards-based phenotyping tool to author electronic health record (EHR)-based phenotype definitions and demonstrate execution of the definitions against heterogeneous clinical research data platforms.

MATERIALS AND METHODS

We developed an open-source, standards-compliant phenotyping tool known as the PhEMA Workbench that enables a phenotype representation using the Fast Healthcare Interoperability Resources (FHIR) and Clinical Quality Language (CQL) standards. We then demonstrated how this tool can be used to conduct EHR-based phenotyping, including phenotype authoring, execution, and validation. We validated the performance of the tool by executing a thrombotic event phenotype definition at 3 sites, Mayo Clinic (MC), Northwestern Medicine (NM), and Weill Cornell Medicine (WCM), and used manual review to determine precision and recall.

RESULTS

An initial version of the PhEMA Workbench has been released, which supports phenotype authoring, execution, and publishing to a shared phenotype definition repository. The resulting thrombotic event phenotype definition consisted of 11 CQL statements, and 24 value sets containing a total of 834 codes. Technical validation showed satisfactory performance (both NM and MC had 100% precision and recall and WCM had a precision of 95% and a recall of 84%).

CONCLUSIONS

We demonstrate that the PhEMA Workbench can facilitate EHR-driven phenotype definition, execution, and phenotype sharing in heterogeneous clinical research data environments. A phenotype definition that integrates with existing standards-compliant systems, and the use of a formal representation facilitates automation and can decrease potential for human error.

Novel Analysis Methods to Mine Immune-Mediated Phenotypes and Find Genetic Variation Within the Electronic Health Record (Roadmap for Phenotype to Genotype: Immunogenomics)

The field of immunogenomics has the opportunity for accelerated genetic discovery aided by the maturation of electronic health records (EHRs) linked to DNA biobanks. Novel analysis methods in deep phenotyping of EHR data allow the full realization of the paired and increasingly dense genetic/phenotypic information available. This enables researchers to uncover genetic risk factors for the prevention and optimal treatment of immune-mediated diseases and immune-mediated adverse drug reactions. This article reviews the background of EHRs linked to DNA biobanks, potential applications to immunogenomic discovery, and current and emerging techniques in EHR-based deep phenotyping.

PhenoDEF: a corpus for annotating sentences with information of phenotype definitions in biomedical literature

Background

Adverse events induced by drug-drug interactions are a major concern in the United States. Current research is moving toward using electronic health record (EHR) data, including for adverse drug events discovery. One of the first steps in EHR-based studies is to define a phenotype for establishing a cohort of patients. However, phenotype definitions are not readily available for all phenotypes. One of the first steps of developing automated text mining tools is building a corpus. Therefore, this study aimed to develop annotation guidelines and a gold standard corpus to facilitate building future automated approaches for mining phenotype definitions contained in the literature. Furthermore, our aim is to improve the understanding of how these published phenotype definitions are presented in the literature and how we annotate them for future text mining tasks.

Results

Two annotators manually annotated the corpus on a sentence-level for the presence of evidence for phenotype definitions. Three major categories (inclusion, intermediate, and exclusion) with a total of ten dimensions were proposed characterizing major contextual patterns and cues for presenting phenotype definitions in published literature. The developed annotation guidelines were used to annotate the corpus that contained 3971 sentences: 1923 out of 3971 (48.4%) for the inclusion category, 1851 out of 3971 (46.6%) for the intermediate category, and 2273 out of 3971 (57.2%) for exclusion category. The highest number of annotated sentences was 1449 out of 3971 (36.5%) for the “Biomedical & Procedure” dimension. The lowest number of annotated sentences was 49 out of 3971 (1.2%) for “The use of NLP”. The overall percent inter-annotator agreement was 97.8%. Percent and Kappa statistics also showed high inter-annotator agreement across all dimensions.

Conclusions

The corpus and annotation guidelines can serve as a foundational informatics approach for annotating and mining phenotype definitions in literature, and can be used later for text mining applications.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13326-022-00272-6.

Opportunities and challenges for the use of common controls in sequencing studies

Genome-wide association studies using large-scale genome and exome sequencing data have become increasingly valuable in identifying associations between genetic variants and disease, transforming basic research and translational medicine. However, this progress has not been equally shared across all people and conditions, in part due to limited resources. Leveraging publicly available sequencing data as external common controls, rather than sequencing new controls for every study, can better allocate resources by augmenting control sample sizes or providing controls where none existed. However, common control studies must be carefully planned and executed as even small differences in sample ascertainment and processing can result in substantial bias. Here, we discuss challenges and opportunities for the robust use of common controls in high-throughput sequencing studies, including study design, quality control and statistical approaches. Thoughtful generation and use of large and valuable genetic sequencing data sets will enable investigation of a broader and more representative set of conditions, environments and genetic ancestries than otherwise possible.