Automated detection of substance use information from electronic health records for a pediatric population

ObjectiveSubstance use screening in adolescence is unstandardized and often documented in clinical notes, rather than in structured electronic health records (EHRs). The objective of this study was to integrate logic rules with state-of-the-art natural language processing (NLP) and machine learning technologies to detect substance use information from both structured and unstructured EHR data.Materials and MethodsPediatric patients (10-20 years of age) with any encounter between July 1, 2012, and October 31, 2017, were included (n = 3890 patients; 19 478 encounters). EHR data were extracted at each encounter, manually reviewed for substance use (alcohol, tobacco, marijuana, opiate, any use), and coded as lifetime use, current use, or family use. Logic rules mapped structured EHR indicators to screening results. A knowledge-based NLP system and a deep learning model detected substance use information from unstructured clinical narratives. System performance was evaluated using positive predictive value, sensitivity, negative predictive value, specificity, and area under the receiver-operating characteristic curve (AUC).ResultsThe dataset included 17 235 structured indicators and 27 141 clinical narratives. Manual review of clinical narratives captured 94.0% of positive screening results, while structured EHR data captured 22.0%. Logic rules detected screening results from structured data with 1.0 and 0.99 for sensitivity and specificity, respectively. The knowledge-based system detected substance use information from clinical narratives with 0.86, 0.79, and 0.88 for AUC, sensitivity, and specificity, respectively. The deep learning model further improved detection capacity, achieving 0.88, 0.81, and 0.85 for AUC, sensitivity, and specificity, respectively. Finally, integrating predictions from structured and unstructured data achieved high detection capacity across all cases (0.96, 0.85, and 0.87 for AUC, sensitivity, and specificity, respectively).ConclusionsIt is feasible to detect substance use screening and results among pediatric patients using logic rules, NLP, and machine learning technologies.     

PheMap: a multi-resource knowledge base for high-throughput phenotyping within electronic health records

ObjectiveDeveloping algorithms to extract phenotypes from electronic health records (EHRs) can be challenging and time-consuming. We developed PheMap, a high-throughput phenotyping approach that leverages multiple independent, online resources to streamline the phenotyping process within EHRs.Materials and MethodsPheMap is a knowledge base of medical concepts with quantified relationships to phenotypes that have been extracted by natural language processing from publicly available resources. PheMap searches EHRs for each phenotype’s quantified concepts and uses them to calculate an individual’s probability of having this phenotype. We compared PheMap to clinician-validated phenotyping algorithms from the Electronic Medical Records and Genomics (eMERGE) network for type 2 diabetes mellitus (T2DM), dementia, and hypothyroidism using 84 821 individuals from Vanderbilt Univeresity Medical Center''s BioVU DNA Biobank. We implemented PheMap-based phenotypes for genome-wide association studies (GWAS) for T2DM, dementia, and hypothyroidism, and phenome-wide association studies (PheWAS) for variants in FTO, HLA-DRB1, and TCF7L2. ResultsIn this initial iteration, the PheMap knowledge base contains quantified concepts for 841 disease phenotypes. For T2DM, dementia, and hypothyroidism, the accuracy of the PheMap phenotypes were >97% using a 50% threshold and eMERGE case-control status as a reference standard. In the GWAS analyses, PheMap-derived phenotype probabilities replicated 43 of 51 previously reported disease-associated variants for the 3 phenotypes. For 9 of the 11 top associations, PheMap provided an equivalent or more significant P value than eMERGE-based phenotypes. The PheMap-based PheWAS showed comparable or better performance to a traditional phecode-based PheWAS. PheMap is publicly available online.ConclusionsPheMap significantly streamlines the process of extracting research-quality phenotype information from EHRs, with comparable or better performance to current phenotyping approaches.     

Validating drug repurposing signals using electronic health records: a case study of metformin associated with reduced cancer mortality

Objectives Drug repurposing, which finds new indications for existing drugs, has received great attention recently. The goal of our work is to assess the feasibility of using electronic health records (EHRs) and automated informatics methods to efficiently validate a recent drug repurposing association of metformin with reduced cancer mortality.Methods By linking two large EHRs from Vanderbilt University Medical Center and Mayo Clinic to their tumor registries, we constructed a cohort including 32 415 adults with a cancer diagnosis at Vanderbilt and 79 258 cancer patients at Mayo from 1995 to 2010. Using automated informatics methods, we further identified type 2 diabetes patients within the cancer cohort and determined their drug exposure information, as well as other covariates such as smoking status. We then estimated HRs for all-cause mortality and their associated 95% CIs using stratified Cox proportional hazard models. HRs were estimated according to metformin exposure, adjusted for age at diagnosis, sex, race, body mass index, tobacco use, insulin use, cancer type, and non-cancer Charlson comorbidity index.Results Among all Vanderbilt cancer patients, metformin was associated with a 22% decrease in overall mortality compared to other oral hypoglycemic medications (HR 0.78; 95% CI 0.69 to 0.88) and with a 39% decrease compared to type 2 diabetes patients on insulin only (HR 0.61; 95% CI 0.50 to 0.73). Diabetic patients on metformin also had a 23% improved survival compared with non-diabetic patients (HR 0.77; 95% CI 0.71 to 0.85). These associations were replicated using the Mayo Clinic EHR data. Many site-specific cancers including breast, colorectal, lung, and prostate demonstrated reduced mortality with metformin use in at least one EHR.Conclusions EHR data suggested that the use of metformin was associated with decreased mortality after a cancer diagnosis compared with diabetic and non-diabetic cancer patients not on metformin, indicating its potential as a chemotherapeutic regimen. This study serves as a model for robust and inexpensive validation studies for drug repurposing signals using EHR data.     

The quality of social determinants data in the electronic health record: a systematic review

ObjectiveThe aim of this study was to collect and synthesize evidence regarding data quality problems encountered when working with variables related to social determinants of health (SDoH).Materials and MethodsWe conducted a systematic review of the literature on social determinants research and data quality and then iteratively identified themes in the literature using a content analysis process.ResultsThe most commonly represented quality issue associated with SDoH data is plausibility (n = 31, 41%). Factors related to race and ethnicity have the largest body of literature (n = 40, 53%). The first theme, noted in 62% (n = 47) of articles, is that bias or validity issues often result from data quality problems. The most frequently identified validity issue is misclassification bias (n = 23, 30%). The second theme is that many of the articles suggest methods for mitigating the issues resulting from poor social determinants data quality. We grouped these into 5 suggestions: avoid complete case analysis, impute data, rely on multiple sources, use validated software tools, and select addresses thoughtfully.DiscussionThe type of data quality problem varies depending on the variable, and each problem is associated with particular forms of analytical error. Problems encountered with the quality of SDoH data are rarely distributed randomly. Data from Hispanic patients are more prone to issues with plausibility and misclassification than data from other racial/ethnic groups.ConclusionConsideration of data quality and evidence-based quality improvement methods may help prevent bias and improve the validity of research conducted with SDoH data.     

Exploring the relationship between electronic health records and provider burnout: A systematic review

ObjectiveStress and burnout due to electronic health record (EHR) technology has become a focus for burnout intervention. The aim of this study is to systematically review the relationship between EHR use and provider burnout.Materials and MethodsA systematic literature search was performed on PubMed, EMBASE, PsychInfo, ACM Digital Library in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement. Inclusion criterion was original research investigating the association between EHR and provider burnout. Studies that did not measure the association objectively were excluded. Study quality was assessed using the Medical Education Research Study Quality Instrument. Qualitative synthesis was also performed.ResultsTwenty-six studies met inclusion criteria. The median sample size of providers was 810 (total 20 885; 44% male; mean age 53 [range, 34-56] years). Twenty-three (88%) studies were cross-sectional studies and 3 were single-arm cohort studies measuring pre- and postintervention burnout prevalence. Burnout was assessed objectively with various validated instruments. Insufficient time for documentation (odds ratio [OR], 1.40-5.83), high inbox or patient call message volumes (OR, 2.06-6.17), and negative perceptions of EHR by providers (OR, 2.17-2.44) were the 3 most cited EHR-related factors associated with higher rates of provider burnout that was assessed objectively.ConclusionsThe included studies were mostly observational studies; thus, we were not able to determine a causal relationship. Currently, there are few studies that objectively assessed the relationship between EHR use and provider burnout. The 3 most cited EHR factors associated with burnout were confirmed and should be the focus of efforts to improve EHR-related provider burnout.     

Use of unstructured text in prognostic clinical prediction models: a systematic review

ObjectiveThis systematic review aims to assess how information from unstructured text is used to develop and validate clinical prognostic prediction models. We summarize the prediction problems and methodological landscape and determine whether using text data in addition to more commonly used structured data improves the prediction performance.Materials and MethodsWe searched Embase, MEDLINE, Web of Science, and Google Scholar to identify studies that developed prognostic prediction models using information extracted from unstructured text in a data-driven manner, published in the period from January 2005 to March 2021. Data items were extracted, analyzed, and a meta-analysis of the model performance was carried out to assess the added value of text to structured-data models.ResultsWe identified 126 studies that described 145 clinical prediction problems. Combining text and structured data improved model performance, compared with using only text or only structured data. In these studies, a wide variety of dense and sparse numeric text representations were combined with both deep learning and more traditional machine learning methods. External validation, public availability, and attention for the explainability of the developed models were limited.ConclusionThe use of unstructured text in the development of prognostic prediction models has been found beneficial in addition to structured data in most studies. The text data are source of valuable information for prediction model development and should not be neglected. We suggest a future focus on explainability and external validation of the developed models, promoting robust and trustworthy prediction models in clinical practice.     

CancerBERT: a cancer domain-specific language model for extracting breast cancer phenotypes from electronic health records

ObjectiveAccurate extraction of breast cancer patients’ phenotypes is important for clinical decision support and clinical research. This study developed and evaluated cancer domain pretrained CancerBERT models for extracting breast cancer phenotypes from clinical texts. We also investigated the effect of customized cancer-related vocabulary on the performance of CancerBERT models.Materials and MethodsA cancer-related corpus of breast cancer patients was extracted from the electronic health records of a local hospital. We annotated named entities in 200 pathology reports and 50 clinical notes for 8 cancer phenotypes for fine-tuning and evaluation. We kept pretraining the BlueBERT model on the cancer corpus with expanded vocabularies (using both term frequency-based and manually reviewed methods) to obtain CancerBERT models. The CancerBERT models were evaluated and compared with other baseline models on the cancer phenotype extraction task.ResultsAll CancerBERT models outperformed all other models on the cancer phenotyping NER task. Both CancerBERT models with customized vocabularies outperformed the CancerBERT with the original BERT vocabulary. The CancerBERT model with manually reviewed customized vocabulary achieved the best performance with macro F1 scores equal to 0.876 (95% CI, 0.873–0.879) and 0.904 (95% CI, 0.902–0.906) for exact match and lenient match, respectively.ConclusionsThe CancerBERT models were developed to extract the cancer phenotypes in clinical notes and pathology reports. The results validated that using customized vocabulary may further improve the performances of domain specific BERT models in clinical NLP tasks. The CancerBERT models developed in the study would further help clinical decision support.     

Extracting seizure frequency from epilepsy clinic notes: a machine reading approach to natural language processing

ObjectiveSeizure frequency and seizure freedom are among the most important outcome measures for patients with epilepsy. In this study, we aimed to automatically extract this clinical information from unstructured text in clinical notes. If successful, this could improve clinical decision-making in epilepsy patients and allow for rapid, large-scale retrospective research.Materials and MethodsWe developed a finetuning pipeline for pretrained neural models to classify patients as being seizure-free and to extract text containing their seizure frequency and date of last seizure from clinical notes. We annotated 1000 notes for use as training and testing data and determined how well 3 pretrained neural models, BERT, RoBERTa, and Bio_ClinicalBERT, could identify and extract the desired information after finetuning.ResultsThe finetuned models (BERT_FT, Bio_ClinicalBERT_FT, and RoBERTa_FT) achieved near-human performance when classifying patients as seizure free, with BERT_FT and Bio_ClinicalBERT_FT achieving accuracy scores over 80%. All 3 models also achieved human performance when extracting seizure frequency and date of last seizure, with overall F₁ scores over 0.80. The best combination of models was Bio_ClinicalBERT_FT for classification, and RoBERTa_FT for text extraction. Most of the gains in performance due to finetuning required roughly 70 annotated notes.Discussion and ConclusionOur novel machine reading approach to extracting important clinical outcomes performed at or near human performance on several tasks. This approach opens new possibilities to support clinical practice and conduct large-scale retrospective clinical research. Future studies can use our finetuning pipeline with minimal training annotations to answer new clinical questions.     

Bias and fairness assessment of a natural language processing opioid misuse classifier: detection and mitigation of electronic health record data disadvantages across racial subgroups

ObjectivesTo assess fairness and bias of a previously validated machine learning opioid misuse classifier.Materials & MethodsTwo experiments were conducted with the classifier’s original (n = 1000) and external validation (n = 53 974) datasets from 2 health systems. Bias was assessed via testing for differences in type II error rates across racial/ethnic subgroups (Black, Hispanic/Latinx, White, Other) using bootstrapped 95% confidence intervals. A local surrogate model was estimated to interpret the classifier’s predictions by race and averaged globally from the datasets. Subgroup analyses and post-hoc recalibrations were conducted to attempt to mitigate biased metrics.ResultsWe identified bias in the false negative rate (FNR = 0.32) of the Black subgroup compared to the FNR (0.17) of the White subgroup. Top features included “heroin” and “substance abuse” across subgroups. Post-hoc recalibrations eliminated bias in FNR with minimal changes in other subgroup error metrics. The Black FNR subgroup had higher risk scores for readmission and mortality than the White FNR subgroup, and a higher mortality risk score than the Black true positive subgroup (P < .05).DiscussionThe Black FNR subgroup had the greatest severity of disease and risk for poor outcomes. Similar features were present between subgroups for predicting opioid misuse, but inequities were present. Post-hoc mitigation techniques mitigated bias in type II error rate without creating substantial type I error rates. From model design through deployment, bias and data disadvantages should be systematically addressed.ConclusionStandardized, transparent bias assessments are needed to improve trustworthiness in clinical machine learning models.     

Measurement of clinical documentation burden among physicians and nurses using electronic health records: a scoping review

Background ObjectiveElectronic health records (EHRs) are linked with documentation burden resulting in clinician burnout. While clear classifications and validated measures of burnout exist, documentation burden remains ill-defined and inconsistently measured. We aim to conduct a scoping review focused on identifying approaches to documentation burden measurement and their characteristics.Materials and MethodsBased on Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) Extension for Scoping Reviews (ScR) guidelines, we conducted a scoping review assessing MEDLINE, Embase, Web of Science, and CINAHL from inception to April 2020 for studies investigating documentation burden among physicians and nurses in ambulatory or inpatient settings. Two reviewers evaluated each potentially relevant study for inclusion/exclusion criteria.ResultsOf the 3482 articles retrieved, 35 studies met inclusion criteria. We identified 15 measurement characteristics, including 7 effort constructs: EHR usage and workload, clinical documentation/review, EHR work after hours and remotely, administrative tasks, cognitively cumbersome work, fragmentation of workflow, and patient interaction. We uncovered 4 time constructs: average time, proportion of time, timeliness of completion, activity rate, and 11 units of analysis. Only 45.0% of studies assessed the impact of EHRs on clinicians and/or patients and 40.0% mentioned clinician burnout.DiscussionStandard and validated measures of documentation burden are lacking. While time and effort were the core concepts measured, there appears to be no consensus on the best approach nor degree of rigor to study documentation burden.ConclusionFurther research is needed to reliably operationalize the concept of documentation burden, explore best practices for measurement, and standardize its use.     

Building longitudinal medication dose data using medication information extracted from clinical notes in electronic health records

ObjectiveTo develop an algorithm for building longitudinal medication dose datasets using information extracted from clinical notes in electronic health records (EHRs).Materials and MethodsWe developed an algorithm that converts medication information extracted using natural language processing (NLP) into a usable format and builds longitudinal medication dose datasets. We evaluated the algorithm on 2 medications extracted from clinical notes of Vanderbilt’s EHR and externally validated the algorithm using clinical notes from the MIMIC-III clinical care database.ResultsFor the evaluation using Vanderbilt’s EHR data, the performance of our algorithm was excellent; F1-measures were ≥0.98 for both dose intake and daily dose. For the external validation using MIMIC-III, the algorithm achieved F1-measures ≥0.85 for dose intake and ≥0.82 for daily dose.DiscussionOur algorithm addresses the challenge of building longitudinal medication dose data using information extracted from clinical notes. Overall performance was excellent, but the algorithm can perform poorly when incorrect information is extracted by NLP systems. Although it performed reasonably well when applied to the external data source, its performance was worse due to differences in the way the drug information was written. The algorithm is implemented in the R package, “EHR,” and the extracted data from Vanderbilt’s EHRs along with the gold standards are provided so that users can reproduce the results and help improve the algorithm.ConclusionOur algorithm for building longitudinal dose data provides a straightforward way to use EHR data for medication-based studies. The external validation results suggest its potential for applicability to other systems.     

Development of a predictive model for retention in HIV care using natural language processing of clinical notes

ObjectiveAdherence to a treatment plan from HIV-positive patients is necessary to decrease their mortality and improve their quality of life, however some patients display poor appointment adherence and become lost to follow-up (LTFU). We applied natural language processing (NLP) to analyze indications towards or against LTFU in HIV-positive patients’ notes.Materials and MethodsUnstructured lemmatized notes were labeled with an LTFU or Retained status using a 183-day threshold. An NLP and supervised machine learning system with a linear model and elastic net regularization was trained to predict this status. Prevalence of characteristics domains in the learned model weights were evaluated.ResultsWe analyzed 838 LTFU vs 2964 Retained notes and obtained a weighted F1 mean of 0.912 via nested cross-validation; another experiment with notes from the same patients in both classes showed substantially lower metrics. “Comorbidities” were associated with LTFU through, for instance, “HCV” (hepatitis C virus) and likewise “Good adherence” with Retained, represented with “Well on ART” (antiretroviral therapy).DiscussionMentions of mental health disorders and substance use were associated with disparate retention outcomes, however history vs active use was not investigated. There remains further need to model transitions between LTFU and being retained in care over time.ConclusionWe provided an important step for the future development of a model that could eventually help to identify patients who are at risk for falling out of care and to analyze which characteristics could be factors for this. Further research is needed to enhance this method with structured electronic medical record fields.     

Factors associated with nurse well-being in relation to electronic health record use: A systematic review

ObjectiveAlthough nurses comprise the largest group of health professionals and electronic health record (EHR) user base, it is unclear how EHR use has affected nurse well-being. This systematic review assesses the multivariable (ie, organizational, nurse, and health information technology [IT]) factors associated with EHR-related nurse well-being and identifies potential improvements recommended by frontline nurses.Materials and MethodsWe searched MEDLINE, Embase, CINAHL, PsycINFO, ProQuest, and Web of Science for literature reporting on EHR use, nurses, and well-being. A quality appraisal was conducted using a previously developed tool.ResultsOf 4583 articles, 12 met inclusion criteria. Two-thirds of the studies were deemed to have a moderate or low risk of bias. Overall, the studies primarily focused on nurse- and IT-level factors, with 1 study examining organizational characteristics. That study found worse nurse well-being was associated with EHRs compared with paper charts. Studies on nurse-level factors suggest that personal digital literacy is one modifiable factor to improving well-being. Additionally, EHRs with integrated displays were associated with improved well-being. Recommendations for improving EHRs suggested IT-, organization-, and policy-level solutions to address the complex nature of EHR-related nurse well-being.ConclusionsThe overarching finding from this synthesis reveals a critical need for multifaceted interventions that better organize, manage, and display information for clinicians to facilitate decision making. Our study also suggests that nurses have valuable insight into ways to reduce EHR-related burden. Future research is needed to test multicomponent interventions that address these complex factors and use participatory approaches to engage nurses in intervention development.     

Electronic phenotyping of health outcomes of interest using a linked claims-electronic health record database: Findings from a machine learning pilot project

ObjectiveClaims-based algorithms are used in the Food and Drug Administration Sentinel Active Risk Identification and Analysis System to identify occurrences of health outcomes of interest (HOIs) for medical product safety assessment. This project aimed to apply machine learning classification techniques to demonstrate the feasibility of developing a claims-based algorithm to predict an HOI in structured electronic health record (EHR) data.Materials and MethodsWe used the 2015-2019 IBM MarketScan Explorys Claims-EMR Data Set, linking administrative claims and EHR data at the patient level. We focused on a single HOI, rhabdomyolysis, defined by EHR laboratory test results. Using claims-based predictors, we applied machine learning techniques to predict the HOI: logistic regression, LASSO (least absolute shrinkage and selection operator), random forests, support vector machines, artificial neural nets, and an ensemble method (Super Learner).ResultsThe study cohort included 32 956 patients and 39 499 encounters. Model performance (positive predictive value [PPV], sensitivity, specificity, area under the receiver-operating characteristic curve) varied considerably across techniques. The area under the receiver-operating characteristic curve exceeded 0.80 in most model variations.DiscussionFor the main Food and Drug Administration use case of assessing risk of rhabdomyolysis after drug use, a model with a high PPV is typically preferred. The Super Learner ensemble model without adjustment for class imbalance achieved a PPV of 75.6%, substantially better than a previously used human expert-developed model (PPV = 44.0%).ConclusionsIt is feasible to use machine learning methods to predict an EHR-derived HOI with claims-based predictors. Modeling strategies can be adapted for intended uses, including surveillance, identification of cases for chart review, and outcomes research.     

Facilitating pharmacogenetic studies using electronic health records and natural-language processing: a case study of warfarin

Objective

DNA biobanks linked to comprehensive electronic health records systems are potentially powerful resources for pharmacogenetic studies. This study sought to develop natural-language-processing algorithms to extract drug-dose information from clinical text, and to assess the capabilities of such tools to automate the data-extraction process for pharmacogenetic studies.

Materials and methods

A manually validated warfarin pharmacogenetic study identified a cohort of 1125 patients with a stable warfarin dose, in which 776 patients were managed by Coumadin Clinic physicians, and the remaining 349 patients were managed by their providers. The authors developed two algorithms to extract weekly warfarin doses from both data sets: a regular expression-based program for semistructured Coumadin Clinic notes; and an advanced weekly dose calculator based on an existing medication information extraction system (MedEx) for narrative providers'' notes. The authors then conducted an association analysis between an automatically extracted stable weekly dose of warfarin and four genetic variants of VKORC1 and CYP2C9 genes. The performance of the weekly dose-extraction program was evaluated by comparing it with a gold standard containing manually curated weekly doses. Precision, recall, F-measure, and overall accuracy were reported. Associations between known variants in VKORC1 and CYP2C9 and warfarin stable weekly dose were performed with linear regression adjusted for age, gender, and body mass index.

Results

The authors'' evaluation showed that the MedEx-based system could determine patients'' warfarin weekly doses with 99.7% recall, 90.8% precision, and 93.8% accuracy. Using the automatically extracted weekly doses of warfarin, the authors successfully replicated the previous known associations between warfarin stable dose and genetic variants in VKORC1 and CYP2C9.     

Electronic health records and clinician burnout: A story of three eras

ObjectiveThe study sought to provide physicians, informaticians, and institutional policymakers with an introductory tutorial about the history of medical documentation, sources of clinician burnout, and opportunities to improve electronic health records (EHRs). We now have unprecedented opportunities in health care, with the promise of new cures, improved equity, greater sensitivity to social and behavioral determinants of health, and data-driven precision medicine all on the horizon. EHRs have succeeded in making many aspects of care safer and more reliable. Unfortunately, current limitations in EHR usability and problems with clinician burnout distract from these successes. A complex interplay of technology, policy, and healthcare delivery has contributed to our current frustrations with EHRs. Fortunately, there are opportunities to improve the EHR and health system. A stronger emphasis on improving the clinician’s experience through close collaboration by informaticians, clinicians, and vendors can combine with specific policy changes to address the causes of burnout.Target audienceThis tutorial is intended for clinicians, informaticians, policymakers, and regulators, who are essential participants in discussions focused on improving clinician burnout. Learners in biomedicine, regardless of clinical discipline, also may benefit from this primer and review.ScopeWe include (1) an overview of medical documentation from a historical perspective; (2) a summary of the forces converging over the past 20 years to develop and disseminate the modern EHR; and (3) future opportunities to improve EHR structure, function, user base, and time required to collect and extract information.     

Reporting of demographic data and representativeness in machine learning models using electronic health records

ObjectiveThe development of machine learning (ML) algorithms to address a variety of issues faced in clinical practice has increased rapidly. However, questions have arisen regarding biases in their development that can affect their applicability in specific populations. We sought to evaluate whether studies developing ML models from electronic health record (EHR) data report sufficient demographic data on the study populations to demonstrate representativeness and reproducibility.Materials and MethodsWe searched PubMed for articles applying ML models to improve clinical decision-making using EHR data. We limited our search to papers published between 2015 and 2019.ResultsAcross the 164 studies reviewed, demographic variables were inconsistently reported and/or included as model inputs. Race/ethnicity was not reported in 64%; gender and age were not reported in 24% and 21% of studies, respectively. Socioeconomic status of the population was not reported in 92% of studies. Studies that mentioned these variables often did not report if they were included as model inputs. Few models (12%) were validated using external populations. Few studies (17%) open-sourced their code. Populations in the ML studies include higher proportions of White and Black yet fewer Hispanic subjects compared to the general US population.DiscussionThe demographic characteristics of study populations are poorly reported in the ML literature based on EHR data. Demographic representativeness in training data and model transparency is necessary to ensure that ML models are deployed in an equitable and reproducible manner. Wider adoption of reporting guidelines is warranted to improve representativeness and reproducibility.     

Let the left hand know what the right is doing: a vision for care coordination and electronic health records

Despite the potential for electronic health records to help providers coordinate care, the current marketplace has failed to provide adequate solutions. Using a simple framework, we describe a vision of information technology capabilities that could substantially improve four care coordination activities: identifying collaborators, contacting collaborators, collaborating, and monitoring. Collaborators can include any individual clinician, caregiver, or provider organization involved in care for a given patient. This vision can be used to guide the development of care coordination tools and help policymakers track and promote their adoption.