Search

BACKGROUND: The mechanism for recording International Classification of Diseases (ICD) and diagnosis related groups (DRG) codes in a patients chart is through a certified medical coder who manually reviews the medical record at the completion of an admission. High-acuity ICD codes justify DRG modifiers, indicating the need for escalated hospital resources. In this manuscript, we demonstrate that value of rules-based computer algorithms that audit for omission of administrative codes and quantifying the downstream effects with regard to financial impacts and demographic findings did not indicate significant disparities. METHODS: All study data were acquired via the UCLA Department of Anesthesiology and Perioperative Medicines Perioperative Data Warehouse. The DataMart is a structured reporting schema that contains all the relevant clinical data entered into the EPIC (EPIC Systems, Verona, WI) electronic health record. Computer algorithms were created for eighteen disease states that met criteria for DRG modifiers. Each algorithm was run against all hospital admissions with completed billing from 2019. The algorithms scanned for the existence of disease, appropriate ICD coding, and DRG modifier appropriateness. Secondarily, the potential financial impact of ICD omissions was estimated by payor class and an analysis of ICD miscoding was done by ethnicity, sex, age, and financial class. RESULTS: Data from 34,104 hospital admissions were analyzed from January 1, 2019, to December 31, 2019. 11,520 (32.9%) hospital admissions were algorithm positive for a disease state with no corresponding ICD code. 1,990 (5.8%) admissions were potentially eligible for DRG modification/upgrade with an estimated lost revenue of $22,680,584.50. ICD code omission rates compared against reference groups (private payors, Caucasians, middle-aged patients) demonstrated significant p-values

BACKGROUND: Outcome measures are crucial to support a treat-to-target approach to rheumatoid arthritis (RA) care, yet their integration into clinical practice remains inconsistent. We developed an Electronic Heath Record-integrated, patient-facing side-car application to display RA outcomes (disease activity, functional status, pain scores), medications, and lab results during clinical visits (RA PRO Dashboard). The study aimed to evaluate patient perceptions and attitudes towards the implementation of a novel patient-facing dashboard during clinical visits using a mixed-methods approach. METHODS: RA patients whose clinicians used the dashboard at least once during their clinical visit were invited to complete a survey regarding its usefulness in care. We also conducted semi-structured interviews with a subset of patients to assess their perceptions of the dashboard. The interviews were transcribed verbatim and analyzed thematically using deductive and inductive techniques. Emerging themes and subthemes were organized into four domains of the Ecological Model of Health. RESULTS: Out of 173 survey respondents, 79% were interested in seeing the dashboard again at a future visit, 71% felt it improved their understanding of their disease, and 65% believed it helped with decision-making about their RA care. Many patients reported that the dashboard helped them discuss their RA symptoms (76%) and medications (72%) with their clinician. Interviews with 29 RA patients revealed 10 key themes: the dashboard was perceived as a valuable visual tool that improved patients understanding of RA outcome measures, enhanced their involvement in care, and increased their trust in clinicians and the clinic. Common reported limitations included concerns about reliability of RA outcome questionnaires for some RA patients and inconsistent collection and explanation of these measures by clinicians. CONCLUSIONS: In both the quantitative and qualitative components of the study, patients reported that the dashboard improved their understanding of their RA, enhanced patient-clinician communication, supported shared decision-making, and increased patient engagement in care. These findings support the use of dashboards or similar data visualization tools in RA care and can be used in future interventions to address challenges in data collection and patient education.

The release of GPT-4 and other large language models (LLMs) has the potential to transform healthcare. However, existing research evaluating LLM performance on real-world clinical notes is limited. Here, we conduct a highly-powered study to determine whether LLMs can provide clinical recommendations for three tasks (admission status, radiological investigation(s) request status, and antibiotic prescription status) using clinical notes from the Emergency Department. We randomly selected 10,000 Emergency Department visits to evaluate the accuracy of zero-shot, GPT-3.5-turbo- and GPT-4-turbo-generated clinical recommendations across four different prompting strategies. We found that both GPT-4-turbo and GPT-3.5-turbo performed poorly compared to a resident physician, with accuracy scores 8% and 24%, respectively, lower than physician on average. Both LLMs tended to be overly cautious in its recommendations, with high sensitivity at the cost of specificity. Our findings demonstrate that, while early evaluations of the clinical use of LLMs are promising, LLM performance must be significantly improved before their deployment as decision support systems for clinical recommendations and other complex tasks.

ObjectiveAlthough supervised machine learning is popular for information extraction from clinical notes, creating large annotated datasets requires extensive domain expertise and is time-consuming. Meanwhile, large language models (LLMs) have demonstrated promising transfer learning capability. In this study, we explored whether recent LLMs could reduce the need for large-scale data annotations.Materials and methodsWe curated a dataset of 769 breast cancer pathology reports, manually labeled with 12 categories, to compare zero-shot classification capability of the following LLMs: GPT-4, GPT-3.5, Starling, and ClinicalCamel, with task-specific supervised classification performance of 3 models: random forests, long short-term memory networks with attention (LSTM-Att), and the UCSF-BERT model.ResultsAcross all 12 tasks, the GPT-4 model performed either significantly better than or as well as the best supervised model, LSTM-Att (average macro F1-score of 0.86 vs 0.75), with advantage on tasks with high label imbalance. Other LLMs demonstrated poor performance. Frequent GPT-4 error categories included incorrect inferences from multiple samples and from history, and complex task design, and several LSTM-Att errors were related to poor generalization to the test set.DiscussionOn tasks where large annotated datasets cannot be easily collected, LLMs can reduce the burden of data labeling. However, if the use of LLMs is prohibitive, the use of simpler models with large annotated datasets can provide comparable results.ConclusionsGPT-4 demonstrated the potential to speed up the execution of clinical NLP studies by reducing the need for large annotated datasets. This may increase the utilization of NLP-based variables and outcomes in clinical studies.

AIMS: Heart failure (HF) is a clinical syndrome with no definitive diagnostic tests. HF registries are often based on manual reviews of medical records of hospitalized HF patients identified using International Classification of Diseases (ICD) codes. However, most HF patients are not hospitalized, and manual review of big electronic health record (EHR) data is not practical. The US Department of Veterans Affairs (VA) has the largest integrated healthcare system in the nation, and an estimated 1.5 million patients have ICD codes for HF (HF ICD-code universe) in their VA EHR. The objective of our study was to develop artificial intelligence (AI) models to phenotype HF in these patients. METHODS AND RESULTS: The model development cohort (n = 20 000: training, 16 000; validation 2000; testing, 2000) included 10 000 patients with HF and 10 000 without HF who were matched by age, sex, race, inpatient/outpatient status, hospital, and encounter date (within 60 days). HF status was ascertained by manual chart reviews in VAs External Peer Review Program for HF (EPRP-HF) and non-HF status was ascertained by the absence of ICD codes for HF in VA EHR. Two clinicians annotated 1000 random snippets with HF-related keywords and labelled 436 as HF, which was then used to train and test a natural language processing (NLP) model to classify HF (positive predictive value or PPV, 0.81; sensitivity, 0.77). A machine learning (ML) model using linear support vector machine architecture was trained and tested to classify HF using EPRP-HF as cases (PPV, 0.86; sensitivity, 0.86). From the HF ICD-code universe, we randomly selected 200 patients (gold standard cohort) and two clinicians manually adjudicated HF (gold standard HF) in 145 of those patients by chart reviews. We calculated NLP, ML, and NLP + ML scores and used weighted F scores to derive their optimal threshold values for HF classification, which resulted in PPVs of 0.83, 0.77, and 0.85 and sensitivities of 0.86, 0.88, and 0.83, respectively. HF patients classified by the NLP + ML model were characteristically and prognostically similar to those with gold standard HF. All three models performed better than ICD code approaches: one principal hospital discharge diagnosis code for HF (PPV, 0.97; sensitivity, 0.21) or two primary outpatient encounter diagnosis codes for HF (PPV, 0.88; sensitivity, 0.54). CONCLUSIONS: These findings suggest that NLP and ML models are efficient AI tools to phenotype HF in big EHR data to create contemporary HF registries for clinical studies of effectiveness, quality improvement, and hypothesis generation.

OBJECTIVES: This study aimed to develop a machine learning (ML) model to predict disengagement from HIV care, high viral load or death among people living with HIV (PLHIV) with the goal of enabling proactive support interventions in Tanzania. The algorithm addressed common challenges when applying ML to electronic medical record (EMR) data: (1) imbalanced outcome distribution; (2) heterogeneity across multisite EMR data and (3) evolving virological suppression thresholds. DESIGN: Observational study using a national EMR database. SETTING: Conducted in two regions in Tanzania, using data from the National HIV Care database. PARTICIPANTS: The study included over 6 million HIV care visit records from 295 961 PLHIV in two regions in Tanzanias National HIV Care database from January 2015 to May 2023. RESULTS: Our ML model effectively identified PLHIV at increased risk of adverse outcomes. Key predictors included past disengagement from care, antiretroviral therapy (ART) status (which tracks a patients engagement with ART across visits), age and time on ART. The downsampling approach we implemented effectively managed imbalanced data to reduce prediction bias. Site-specific algorithms performed better compared with a universal approach, highlighting the importance of tailoring ML models to local contexts. A sensitivity analysis confirmed the models robustness to changes in viral load suppression thresholds. CONCLUSIONS: ML models leveraging large-scale databases of patient data offer significant potential to identify PLHIV for interventions to enhance engagement in HIV care in resource-limited settings. Tailoring algorithms to local contexts and flexibility towards evolving clinical guidelines are essential for maximising their impact.

RATIONALE: Research studies typically quantify acute respiratory exacerbation episodes (AECOPD) among people with chronic obstructive pulmonary disease (COPD) based on self-report elicited by survey questionnaire. However, AECOPD quantification by self-report could be inaccurate, potentially rendering it an imprecise tool for identification of those with exacerbation tendency. OBJECTIVE: Determine the agreement between self-reported and health records-documented quantification of AECOPD and their association with airway inflammation. METHODS: We administered a questionnaire to elicit the incidence and severity of respiratory exacerbations in the three years preceding the survey among current or former heavy smokers with or without diagnosis of COPD. We then examined electronic health records (EHR) of those with COPD and those without (tobacco-exposed persons with preserved spirometry or TEPS) to determine whether the documentation of the three-year incidence of moderate to very severe respiratory exacerbations was consistent with self-report using Kappa Interrater statistic. A subgroup of participants also underwent bronchoalveolar lavage (BAL) to quantify their airway inflammatory cells. We further used multivariable regressions analysis to estimate the association between respiratory exacerbations and BAL inflammatory cell composition with adjustment for covariates including age, sex, height, weight, smoking status (current versus former) and burden (pack-years). RESULTS: Overall, a total of 511 participants completed the questionnaire, from whom 487 had EHR available for review. Among the 222 participants with COPD (70 ± 7 years-old; 96% male; 70 ± 38 pack-years smoking; 42% current smoking), 57 (26%) reported having any moderate to very severe AECOPD (m/s-AECOPD) while 66 (30%) had EHR documentation of m/s-AECOPD. However, 42% of those with EHR-identified m/s-AECOPD had none by self-report, and 33% of those who reported m/s-AECOPD had none by EHR, suggesting only moderate agreement (Cohens Kappa = 0.47 ± 0.07; P

BACKGROUND: The consent protocol is now a critical part in the overall orchestration of clinical research. We aimed to demonstrate the feasibility of an Ethereum-based informed consent system, which includes an immutable and automated channel of consent matching, to simultaneously assure patient privacy and increase the efficiency of researchers data access. METHOD: We simulated a multi-site scenario, each assigned 10000 consent records. A consent record contained one patients data-sharing preference with regards to seven data categories. We developed a blockchain-based infrastructure with a smart contract to record consents on-chain, and to query consenting patients corresponding to specific criteria. We measured our systems recording efficiency against a baseline design and verified accuracy by testing an exhaustive list of possible queries. RESULTS: Our method achieved ∼3-4% lead with an average insertion speed of ∼2 s per record per node on either a 3-, 4- or 5-node network, and 100 % accuracy. It also outperformed other solutions in external validation. DISCUSSION: The speed we achieved is reasonable in a real-world system under the realistic assumption that patients may not change their minds too frequently, with the added benefit of immutability. Furthermore, the per-insertion time did improve slightly as the number of network nodes increased, attesting to the benefit of node parallelism as it suggests no attrition of insertion efficiency due to scale of nodes. CONCLUSIONS: Our work confirms the technical feasibility of a blockchain-based consent mechanism, assuring patients with an immutable audit trail, and providing researchers with an efficient way to reach their cohorts.

BackgroundOne challenge for primary care providers caring for patients with nonalcoholic fatty liver disease is to identify those at the highest risk for clinically significant liver disease.AimTo derive a risk stratification tool using variables from structured electronic health record (EHR) data for use in populations which are disproportionately affected with obesity and diabetes.MethodsWe used data from 344 participants who underwent Fibroscan examination to measure liver fat and liver stiffness measurement [LSM]. Using two approaches, multivariable logistic regression and random forest classification, we assessed risk factors for any hepatic fibrosis (LSM > 7 kPa) and significant hepatic fibrosis (> 8 kPa). Possible predictors included data from the EHR for age, gender, diabetes, hypertension, FIB-4, body mass index (BMI), LDL, HDL, and triglycerides.ResultsOf 344 patients (56.4% women), 34 had any hepatic fibrosis, and 15 significant hepatic fibrosis. Three variables (BMI, FIB-4, diabetes) were identified from both approaches. When we used variable cut-offs defined by Youden's index, the final model predicting any hepatic fibrosis had an AUC of 0.75 (95% CI 0.67-0.84), NPV of 91.5% and PPV of 40.0%. The final model with variable categories based on standard clinical thresholds (i.e., BMI ≥ 30 kg/m2; FIB-4 ≥ 1.45) had lower discriminatory ability (AUC 0.65), but higher PPV (50.0%) and similar NPV (91.3%). We observed similar findings for predicting significant hepatic fibrosis.ConclusionsOur results demonstrate that standard thresholds for clinical risk factors/biomarkers may need to be modified for greater discriminatory ability among populations with high prevalence of obesity and diabetes.

OBJECTIVE: The study aimed to develop natural language processing (NLP) algorithms to automate extracting patient-centred breast cancer treatment outcomes from clinical notes in electronic health records (EHRs), particularly for women from under-represented populations. METHODS: The study used clinical notes from 2010 to 2021 from a tertiary hospital in the USA. The notes were processed through various NLP techniques, including vectorisation methods (term frequency-inverse document frequency (TF-IDF), Word2Vec, Doc2Vec) and classification models (support vector classification, K-nearest neighbours (KNN), random forest (RF)). Feature selection and optimisation through random search and fivefold cross-validation were also conducted. RESULTS: The study annotated 100 out of 1000 clinical notes, using 970 notes to build the text corpus. TF-IDF and Doc2Vec combined with RF showed the highest performance, while Word2Vec was less effective. RF classifier demonstrated the best performance, although with lower recall rates, suggesting more false negatives. KNN showed lower recall due to its sensitivity to data noise. DISCUSSION: The study highlights the significance of using NLP in analysing clinical notes to understand breast cancer treatment outcomes in under-represented populations. The TF-IDF and Doc2Vec models were more effective in capturing relevant information than Word2Vec. The study observed lower recall rates in RF models, attributed to the datasets imbalanced nature and the complexity of clinical notes. CONCLUSION: The study developed high-performing NLP pipeline to capture treatment outcomes for breast cancer in under-represented populations, demonstrating the importance of document-level vectorisation and ensemble methods in clinical notes analysis. The findings provide insights for more equitable healthcare strategies and show the potential for broader NLP applications in clinical settings.

OBJECTIVE: This study evaluates regularization variants in logistic regression (L1, L2, ElasticNet, Adaptive L1, Adaptive ElasticNet, Broken adaptive ridge [BAR], and Iterative hard thresholding [IHT]) for discrimination and calibration performance, focusing on both internal and external validation. MATERIALS AND METHODS: We use data from 5 US claims and electronic health record databases and develop models for various outcomes in a major depressive disorder patient population. We externally validate all models in the other databases. We use a train-test split of 75%/25% and evaluate performance with discrimination and calibration. Statistical analysis for difference in performance uses Friedmans test and critical difference diagrams. RESULTS: Of the 840 models we develop, L1 and ElasticNet emerge as superior in both internal and external discrimination, with a notable AUC difference. BAR and IHT show the best internal calibration, without a clear external calibration leader. ElasticNet typically has larger model sizes than L1. Methods like IHT and BAR, while slightly less discriminative, significantly reduce model complexity. CONCLUSION: L1 and ElasticNet offer the best discriminative performance in logistic regression for healthcare predictions, maintaining robustness across validations. For simpler, more interpretable models, L0-based methods (IHT and BAR) are advantageous, providing greater parsimony and calibration with fewer features. This study aids in selecting suitable regularization techniques for healthcare prediction models, balancing performance, complexity, and interpretability.

BACKGROUND: Language in nonmedical data sets is known to transmit human-like biases when used in natural language processing (NLP) algorithms that can reinforce disparities. It is unclear if NLP algorithms of medical notes could lead to similar transmissions of biases. RESEARCH QUESTION: Can we identify implicit bias in clinical notes, and are biases stable across time and geography? STUDY DESIGN AND METHODS: To determine whether different racial and ethnic descriptors are similar contextually to stigmatizing language in ICU notes and whether these relationships are stable across time and geography, we identified notes on critically ill adults admitted to the University of California, San Francisco (UCSF), from 2012 through 2022 and to Beth Israel Deaconess Hospital (BIDMC) from 2001 through 2012. Because word meaning is derived largely from context, we trained unsupervised word-embedding algorithms to measure the similarity (cosine similarity) quantitatively of the context between a racial or ethnic descriptor (eg, African-American) and a stigmatizing target word (eg, nonco-operative) or group of words (violence, passivity, noncompliance, nonadherence). RESULTS: In UCSF notes, Black descriptors were less likely to be similar contextually to violent words compared with White descriptors. Contrastingly, in BIDMC notes, Black descriptors were more likely to be similar contextually to violent words compared with White descriptors. The UCSF data set also showed that Black descriptors were more similar contextually to passivity and noncompliance words compared with Latinx descriptors. INTERPRETATION: Implicit bias is identifiable in ICU notes. Racial and ethnic group descriptors carry different contextual relationships to stigmatizing words, depending on when and where notes were written. Because NLP models seem able to transmit implicit bias from training data, use of NLP algorithms in clinical prediction could reinforce disparities. Active debiasing strategies may be necessary to achieve algorithmic fairness when using language models in clinical research.

Acute COVID-19 infection can be followed by diverse clinical manifestations referred to as Post Acute Sequelae of SARS-CoV2 Infection (PASC). Studies have shown an increased risk of being diagnosed with new-onset psychiatric disease following a diagnosis of acute COVID-19. However, it was unclear whether non-psychiatric PASC-associated manifestations (PASC-AMs) are associated with an increased risk of new-onset psychiatric disease following COVID-19. A retrospective electronic health record (EHR) cohort study of 2,391,006 individuals with acute COVID-19 was performed to evaluate whether non-psychiatric PASC-AMs are associated with new-onset psychiatric disease. Data were obtained from the National COVID Cohort Collaborative (N3C), which has EHR data from 76 clinical organizations. EHR codes were mapped to 151 non-psychiatric PASC-AMs recorded 28-120 days following SARS-CoV-2 diagnosis and before diagnosis of new-onset psychiatric disease. Association of newly diagnosed psychiatric disease with age, sex, race, pre-existing comorbidities, and PASC-AMs in seven categories was assessed by logistic regression. There were significant associations between a diagnosis of any psychiatric disease and five categories of PASC-AMs with odds ratios highest for neurological, cardiovascular, and constitutional PASC-AMs with odds ratios of 1.31, 1.29, and 1.23 respectively. Secondary analysis revealed that the proportions of 50 individual clinical features significantly differed between patients diagnosed with different psychiatric diseases. Our study provides evidence for association between non-psychiatric PASC-AMs and the incidence of newly diagnosed psychiatric disease. Significant associations were found for features related to multiple organ systems. This information could prove useful in understanding risk stratification for new-onset psychiatric disease following COVID-19. Prospective studies are needed to corroborate these findings.

BackgroundIschemic cardiomyopathy is the leading cause of heart failure (HF). Most patients do not undergo coronary assessment after HF diagnosis. There are no randomized clinical trials of coronary assessment after HF diagnosis.MethodsUsing an electronic health record cohort of all individuals with HF within the San Francisco Health Network from 2001 to 2019, we identified factors associated with coronary assessment. Then, we studied the association of coronary assessment within 30 days of HF diagnosis with all-cause mortality and a composite of mortality and emergent angiography using a target trial emulation observational comparative-effectiveness approach. Target trial emulation is an approach to causal inference based on creating a hypothetical randomized clinical trial protocol and using observational data to emulate the protocol. We used propensity scores for covariate adjustment. We used national death records to improve the ascertainment of mortality and included falsification end points for the cause of death.ResultsAmong 14 829 individuals with HF (median, 62 years old; 5855 [40%] women), 3987 (26.9%) ever completed coronary assessment, with 2467/13 301 (18.5%) with unknown coronary artery disease status at HF diagnosis assessed. Women, older individuals, and people without stable housing were less likely to complete coronary assessment. Among 5972 eligible persons of whom 627 underwent early elective coronary assessment, coronary assessment was associated with lower mortality (hazard ratio, 0.84 [95% CI, 0.72-0.97]; P=0.025), reduced risk of the composite outcome (hazard ratio, 0.86 [95% CI, 0.73-1.00]), higher rates of revascularization (odds ratio, 7.6 [95% CI, 5.4-10.6]), and higher use of medical therapy (odds ratio, 2.5 [95% CI, 1.7-3.6]), but not the falsification end points.ConclusionsIn a safety-net population, disparities in coronary assessment after HF diagnosis are not fully explained by coronary artery disease risk factors. Early coronary assessment is associated with improved HF outcomes possibly related to higher rates of revascularization and guideline-directed medical therapy but with low certainty that this finding is not attributable to unmeasured confounding.

Tobacco use disorder (TUD) is the most prevalent substance use disorder in the world. Genetic factors influence smoking behaviours and although strides have been made using genome-wide association studies to identify risk variants, most variants identified have been for nicotine consumption, rather than TUD. Here we leveraged four US biobanks to perform a multi-ancestral meta-analysis of TUD (derived via electronic health records) in 653,790 individuals (495,005 European, 114,420 African American and 44,365 Latin American) and data from UK Biobank (ncombined = 898,680). We identified 88 independent risk loci; integration with functional genomic tools uncovered 461 potential risk genes, primarily expressed in the brain. TUD was genetically correlated with smoking and psychiatric traits from traditionally ascertained cohorts, externalizing behaviours in children and hundreds of medical outcomes, including HIV infection, heart disease and pain. This work furthers our biological understanding of TUD and establishes electronic health records as a source of phenotypic information for studying the genetics of TUD.

BACKGROUND: Improving shared decision-making using a treat-to-target approach, including the use of clinical outcome measures, is important to providing high quality care for rheumatoid arthritis (RA). We developed an Electronic Health Record (EHR) integrated, patient-facing sidecar dashboard application that displays RA outcomes, medications, and lab results for use during clinical visits (RA PRO dashboard). The purpose of this study was to assess clinician perceptions and experiences using the dashboard in a university rheumatology clinic. METHODS: We conducted focus group (FG) discussions with clinicians who had access to the dashboard as part of a randomized, stepped-wedge pragmatic trial. FGs explored clinician perceptions towards the usability, acceptability, and usefulness of the dashboard. FG data were analyzed thematically using deductive and inductive techniques; generated themes were categorized into the domains of the Technology Acceptance Model (TAM). RESULTS: 3 FG discussions were conducted with a total of 13 clinicians. Overall, clinicians were enthusiastic about the dashboard and expressed the usefulness of visualizing RA outcome trajectories in a graphical format for motivating patients, enhancing patient understanding of their RA outcomes, and improving communication about medications. Major themes that emerged from the FG analysis as barriers to using the dashboard included inconsistent collection of RA outcomes leading to sparse data in the dashboard and concerns about explaining RA outcomes, especially to patients with fibromyalgia. Other challenges included time constraints and technical difficulties refreshing the dashboard to display real-time data. Methods for integrating the dashboard into the visit varied: some clinicians used the dashboard at the beginning of the visit as they documented RA outcomes; others used it at the end to justify changes to therapy; and a few shared it only with stable patients. CONCLUSIONS: The study provides valuable insights into clinicians perceptions and experiences with the RA PRO dashboard. The dashboard showed promise in enhancing patient-clinician communication, shared decision-making, and overall acceptance among clinicians. Addressing challenges related to data collection, education, and tailoring dashboard use to specific patient populations will be crucial for maximizing its potential impact on RA care. Further research and ongoing improvements in dashboard design and implementation are warranted to ensure its successful integration into routine clinical practice.

ObjectiveBlockchain has emerged as a potential data-sharing structure in healthcare because of its decentralization, immutability, and traceability. However, its use in the biomedical domain is yet to be investigated comprehensively, especially from the aspects of implementation and evaluation, by existing blockchain literature reviews. To address this, our review assesses blockchain applications implemented in practice and evaluated with quantitative metrics.Materials and methodsThis systematic review adapts the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework to review biomedical blockchain papers published by August 2023 from 3 databases. Blockchain application, implementation, and evaluation metrics were collected and summarized.ResultsFollowing screening, 11 articles were included in this review. Articles spanned a range of biomedical applications including COVID-19 medical data sharing, decentralized internet of things (IoT) data storage, clinical trial management, biomedical certificate storage, electronic health record (EHR) data sharing, and distributed predictive model generation. Only one article demonstrated blockchain deployment at a medical facility.DiscussionEthereum was the most common blockchain platform. All but one implementation was developed with private network permissions. Also, 8 articles contained storage speed metrics and 6 contained query speed metrics. However, inconsistencies in presented metrics and the small number of articles included limit technological comparisons with each other.ConclusionWhile blockchain demonstrates feasibility for adoption in healthcare, it is not as popular as currently existing technologies for biomedical data management. Addressing implementation and evaluation factors will better showcase blockchain's practical benefits, enabling blockchain to have a significant impact on the health sector.

ObjectiveAccurate identification of lupus nephritis (LN) cases is essential for patient management, research and public health initiatives. However, LN diagnosis codes in electronic health records (EHRs) are underused, hindering efficient identification. We investigated the current performance of International Classification of Diseases (ICD) codes, 9th and 10th editions (ICD9/10), for identifying prevalent LN, and developed scoring systems to increase identification of LN that are adaptable to settings with and without LN ICD codes.MethodsTraining and test sets derived from EHR data from a large health system. An external set comprised data from the EHR of a second large health system. Adults with ICD9/10 codes for SLE were included. LN cases were ascertained through manual chart reviews conducted by rheumatologists. Two definitions of LN were used: strict (definite LN) and inclusive (definite, potential or diagnostic uncertainty). Gradient boosting models including structured EHR fields were used for predictor selection. Two logistic regression-based scoring systems were developed ('LN-Code' included LN ICD codes and 'LN-No Code' did not), calibrated and validated using standard performance metrics.ResultsA total of 4152 patients from University of California San Francisco Medical Center and 370 patients from Zuckerberg San Francisco General Hospital and Trauma Center met the eligibility criteria. Mean age was 50 years, 87% were female. LN diagnosis codes demonstrated low sensitivity (43-73%) but high specificity (92-97%). LN-Code achieved an area under the curve (AUC) of 0.93 and a sensitivity of 0.88 for identifying LN using the inclusive definition. LN-No Code reached an AUC of 0.91 and a sensitivity of 0.95 (0.97 for the strict definition). Both scoring systems had good external validity, calibration and performance across racial and ethnic groups.ConclusionsThis study quantified the underutilisation of LN diagnosis codes in EHRs and introduced two adaptable scoring systems to enhance LN identification. Further validation in diverse healthcare settings is essential to ensure their broader applicability.

ImportanceIncreasing influenza vaccination rates is a public health priority. One method recommended by the US Centers for Disease Control and Prevention and others is for health systems to send reminders nudging patients to be vaccinated.ObjectiveTo evaluate and compare the effect of electronic health record (EHR)-based patient portal reminders vs text message reminders on influenza vaccination rates across a health system.Design, setting, and participantsThis 3-arm randomized clinical trial was conducted from September 7, 2022, to April 30, 2023, among primary care patients within the University of California, Los Angeles (UCLA) health system.InterventionsArm 1 received standard of care. The health system sent monthly reminder messages to patients due for an influenza vaccine by portal (arm 2) or text (arm 3). Arm 2 had a 2 × 2 nested design, with fixed vs responsive monthly reminders and preappointment vs no preappointment reminders. Arm 3 had 1 × 2 design, with preappointment vs no preappointment reminders. Preappointment reminders for eligible patients were sent 24 and 48 hours before scheduled primary care visits. Fixed reminders (in October, November, and December) involved identical messages via portal or text. Responsive portal reminders involved a September message asking patients about their plans for vaccination, with a follow-up reminder if the response was affirmative but the patient was not yet vaccinated.Main outcomes and measuresThe primary outcome was influenza vaccination by April 30, 2023, obtained from the UCLA EHR, including vaccination from pharmacies and other sources.ResultsA total of 262 085 patients (mean [SD] age, 45.1 [20.7] years; 237 404 [90.6%] adults; 24 681 [9.4%] children; 149 349 [57.0%] women) in 79 primary care practices were included (87 257 in arm 1, 87 478 in arm 2, and 87 350 in arm 3). At the entire primary care population level, none of the interventions improved influenza vaccination rates. All groups had rates of approximately 47%. There was no statistical or clinically significant improvement following portal vs text, preappointment reminders vs no preappointment reminders (portal and text reminders combined), or responsive vs fixed monthly portal reminders.Conclusions and relevanceAt the population level, neither portal nor text reminders for influenza vaccination were effective. Given that vaccine hesitancy may be a major reason for the lack of impact of portal or text reminders, more intensive interventions by health systems are needed to raise influenza vaccination coverage levels.Trial registrationClinicalTrials.gov Identifier: NCT05525494.

Three years after the start of the SARS-CoV-2 virus (COVID-19) pandemic, its effects continue to affect society and COVID-19 vaccination campaigns continue to be a topic of controversy and inconsistent practice. After experiencing spikes in COVID-19 cases, our University of California Davis Health Division of Hospital Medicine sought to understand the reasons underlying the low COVID-19 vaccination rates in our county and find approaches to improve the number of vaccinations among adults admitted to the inpatient setting. This quality improvement project aimed to increase COVID-19 primary and booster vaccine efforts through a multi-pronged approach of increased collaboration with specialised staff and optimisation of use of our electronic health record system.Our key interventions focused on developing a visual reminder of COVID-19 vaccine status using the functionality of our electronic medical record (EMR), standardising documentation of COVID-19 vaccine status and enhancing team-based vaccination discussions through team huddles and partnering with inpatient care coordinators. While our grassroots approach enhanced COVID-19 vaccination rates in the inpatient setting and had additional benefits such as increased collaboration among teams, system-level efforts often made a greater impact at our healthcare centre. For other institutions interested in increasing COVID-19 vaccination rates, our top three recommendations include integrating vaccination into pre-existing workflows, optimising EMR functionality and increasing vaccine accessibility in the inpatient setting.

OBJECTIVES: This study sought to capture current digital health company experiences integrating with electronic health records (EHRs), given new federally regulated standards-based application programming interface (API) policies. MATERIALS AND METHODS: We developed and fielded a survey among companies that develop solutions enabling human interaction with an EHR API. The survey was developed by the University of California San Francisco in collaboration with the Office of the National Coordinator for Health Information Technology, the California Health Care Foundation, and ScaleHealth. The instrument contained questions pertaining to experiences with API integrations, barriers faced during API integrations, and API-relevant policy efforts. RESULTS: About 73% of companies reported current or previous use of a standards-based EHR API in production. About 57% of respondents indicated using both standards-based and proprietary APIs to integrate with an EHR, and 24% worked about equally with both APIs. Most companies reported use of the Fast Healthcare Interoperability Resources standard. Companies reported that standards-based APIs required on average less burden than proprietary APIs to establish and maintain. However, companies face barriers to adopting standards-based APIs, including high fees, lack of realistic clinical testing data, and lack of data elements of interest or value. DISCUSSION: The industry is moving toward the use of standardized APIs to streamline data exchange, with a majority of digital health companies using standards-based APIs to integrate with EHRs. However, barriers persist. CONCLUSION: A large portion of digital health companies use standards-based APIs to interoperate with EHRs. Continuing to improve the resources for digital health companies to find, test, connect, and use these APIs without special effort will be crucial to ensure future technology robustness and durability.

OBJECTIVES: Although guidelines recommend antiviral therapy for outpatients with COVID-19 who are at high risk of progressing to severe conditions, such as older adults, many patients do not receive appropriate treatment. Little is known, however, about the physician factors associated with the prescription of guideline-recommended antiviral therapy for patients with COVID-19. DESIGN: A cross-sectional study. SETTING: Data including outpatient visits in primary care clinics in Japan from April to August 2023. PARTICIPANTS: We analysed 30 953 outpatients aged ≥65 years treated with COVID-19 (mean (SD) age, 75.0 (7.6) years; 17 652 women (57.0%)) in 1394 primary care clinics. OUTCOME MEASURES: The primary outcome was the prescription of guideline-recommended antivirals (ie, nirmatrelvir-ritonavir or molnupiravir), adjusted for patient characteristics, months of visits and regions. RESULTS: Antiviral prescriptions were concentrated among a small proportion of physicians; for example, the top 10% of physicians that had the largest number of nirmatrelvir-ritonavir prescriptions accounted for 92.4% of all nirmatrelvir-ritonavir prescriptions. After adjusting for potential confounders, physicians with higher patient volumes were more likely to prescribe guideline-recommended antivirals to their patients (adjusted OR (aOR) for high vs low volume, 1.76; 95% CI 1.31 to 2.38; adjusted p

Here, we summarize the proceedings of the inaugural Artificial Intelligence in Primary Immune Deficiencies conference, during which experts and advocates gathered to advance research into the applications of artificial intelligence (AI), machine learning, and other computational tools in the diagnosis and management of inborn errors of immunity (IEIs). The conference focused on the key themes of expediting IEI diagnoses, challenges in data collection, roles of natural language processing and large language models in interpreting electronic health records, and ethical considerations in implementation. Innovative AI-based tools trained on electronic health records and claims databases have discovered new patterns of warning signs for IEIs, facilitating faster diagnoses and enhancing patient outcomes. Challenges in training AIs persist on account of data limitations, especially in cases of rare diseases, overlapping phenotypes, and biases inherent in current data sets. Furthermore, experts highlighted the significance of ethical considerations, data protection, and the necessity for open science principles. The conference delved into regulatory frameworks, equity in access, and the imperative for collaborative efforts to overcome these obstacles and harness the transformative potential of AI. Concerted efforts to successfully integrate AI into daily clinical immunology practice are still needed.

Identification of Alzheimer's disease (AD) onset risk can facilitate interventions before irreversible disease progression. We demonstrate that electronic health records from the University of California, San Francisco, followed by knowledge networks (for example, SPOKE) allow for (1) prediction of AD onset and (2) prioritization of biological hypotheses, and (3) contextualization of sex dimorphism. We trained random forest models and predicted AD onset on a cohort of 749 individuals with AD and 250,545 controls with a mean area under the receiver operating characteristic of 0.72 (7 years prior) to 0.81 (1 day prior). We further harnessed matched cohort models to identify conditions with predictive power before AD onset. Knowledge networks highlight shared genes between multiple top predictors and AD (for example, APOE, ACTB, IL6 and INS). Genetic colocalization analysis supports AD association with hyperlipidemia at the APOE locus, as well as a stronger female AD association with osteoporosis at a locus near MS4A6A. We therefore show how clinical data can be utilized for early AD prediction and identification of personalized biological hypotheses.

Advance care planning (ACP) has been recognized as crucial by patients, families, and clinicians; however, different definitions and measurements have led to inconsistencies in practice and mixed evidence in the literature. This narrative review explores ACPs evolution, innovations, and outcomes using thematic analysis to synthesize data from randomized controlled trials, reviews, and editorials. Key findings include (1) ACP has evolved over the past several decades from a sole focus on code status and advance directive (AD) forms to a continuum of care planning over the life course focused on tailored preparation for patients and surrogate decision-makers and (2) ACP measurement has evolved from traditional outcome metrics, such as AD completion, to a comprehensive outcomes framework that includes behavior change theory, systems, implementation science, and a focus on surrogate outcomes. Since the recent development of an ACP consensus definition and outcomes framework, high-quality trials have reported mainly positive outcomes for interventions, especially for surrogates, which aligns with the patient desire to relieve decision-making burden for loved ones. Additionally, measurement of clinically meaningful ACP information, including documented goals of care discussions, is increasingly being integrated into electronic health records (EHR), and emerging, real-time assessments and natural language processing are enhancing ACP evaluation. To make things easier for patients, families, and care teams, clinicians and researchers can use and disseminate these evolved definitions; provide patients validated, easy-to-use tools that prime patients for conversations and decrease health disparities; use easy-to-access clinician training and simple scripts for interdisciplinary team members; and document patients values and preferences in the medical record to capture clinically meaningful ACP so this information is available at the point of care. Future efforts should focus on efficient implementation, expanded reimbursement options, and seamless integration of EHR documentation to ensure ACPs continued evolution to better serve patients and their care partners.

OBJECTIVE: Map multimorbidity-weighted index (MWI) conditions to International Classification of Diseases, 10th Revision (ICD-10), expand the conditions and codes to develop a new ICD-10-coded MWI (MWI-ICD10) and updated MWI-ICD9, and assess their consistency. DESIGN: Population-based retrospective cohort. SETTING: Large medical centre between 2013 and 2017. PARTICIPANTS: Adults ≥18 years old with encounters in each of 4 years (2013, 2014, 2016, 2017). MAIN OUTCOME MEASURES: MWI conditions mapped to ICD-10 codes, and additional conditions and codes added to produce a new MWI-ICD10 and updated MWI-ICD9. We compared the prevalence of ICD-coded MWI conditions within the ICD-9 era (2013-2014), within the ICD-10 era (2016-2017) and across the ICD-9-ICD-10 transition in 2015 (washout period) among adults present in both sets of comparison years. We computed the prevalence and change in prevalence of conditions when using MWI-ICD10 versus MWI-ICD9. RESULTS: 88 175 adults met inclusion criteria. Participants were 60.8% female, 50.5% white, with mean age 54.7±17.3 years and baseline MWI-ICD9 4.47±6.02 (range 0-64.33). Of 94 conditions, 65 had

BACKGROUND: Despite efforts to view electronic health records (EHR) data through an equity lens, crucial contextual information regarding patients social environments remains limited. Integrating EHR data and Geographic Information Systems (GIS) technology can give deeper insights into the relationships between patients social environments, health outcomes, and geographic factors. This study aims to identify regions with the fastest and slowest access to outpatient physical therapy services using bivariate choropleth maps to provide contextual insights that may contribute to health disparity in access. METHODS: This was a retrospective cohort study of patients access timelines for the first visit to outpatient physical therapy services (n = 10,363). The three timelines evaluated were (1) referral-to-scheduled appointment time, (2) scheduled appointment to first visit time, and (3) referral to first visit time. Hot and coldspot analyses (CI 95%) determined the fastest and slowest access times with patient-level characteristics and bivariate choropleth maps that were developed to visualize associations between access patterns and disadvantaged areas using Area Deprivation Index scores. Data were collected between January 1, 2016 and January 1, 2020. EHR data were geocoded via GIS technology to calculate geospatial statistics (Gi∗ statistic from ArcGIS Pro) in an urban area. RESULTS: Statistically significant differences were found for all three access timelines between coldspot (i.e., fast access group) and hotspot (i.e., slow access group) comparisons (p

ImportanceProcedural proficiency is a core competency for graduate medical education; however, procedural reporting often relies on manual workflows that are duplicative and generate data whose validity and accuracy are difficult to assess. Failure to accurately gather these data can impede learner progression, delay procedures, and negatively impact patient safety.ObjectiveTo examine accuracy and procedure logging completeness of a system that extracts procedural data from an electronic health record system and uploads these data securely to an application used by many residency programs for accreditation.Design, setting, and participantsThis quality improvement study of all emergency medicine resident physicians at University of California, San Diego Health was performed from May 23, 2023, to June 25, 2023.ExposuresAutomated system for procedure data extraction and upload to a residency management software application.Main outcomes and measuresThe number of procedures captured by the automated system when running silently compared with manually logged procedures in the same timeframe, as well as accuracy of the data upload.ResultsForty-seven residents participated in the initial silent assessment of the extraction component of the system. During a 1-year period (May 23, 2022, to May 7, 2023), 4291 procedures were manually logged by residents, compared with 7617 procedures captured by the automated system during the same period, representing a 78% increase. During assessment of the upload component of the system (May 8, 2023, to June 25, 2023), a total of 1353 procedures and patient encounters were evaluated, with the system operating with a sensitivity of 97.4%, specificity of 100%, and overall accuracy of 99.5%.Conclusions and relevanceIn this quality improvement study of emergency medicine resident physicians, an automated system demonstrated that reliance on self-reported procedure logging resulted in significant procedural underreporting compared with the use of data obtained at the point of performance. Additionally, this system afforded a degree of reliability and validity heretofore absent from the usual after-the-fact procedure logging workflows while using a novel application programming interface-based approach. To our knowledge, this system constitutes the first generalizable implementation of an automated solution to a problem that has existed in graduate medical education for decades.

ObjectivesThe number of surgeries performed in the United States has increased over the past two decades, with a shift to the ambulatory setting. Perioperative complications and mortality pose significant health care burdens. Inadequate preoperative assessment and documentation contribute to communication failure and poor patient outcomes. The aim of this quality improvement project was to design and implement a preoperative evaluation documentation template that not only improved communication during the perioperative pathway but also enhanced the overall user experience.MethodsWe implemented a revamped evidence-based documentation template in the electronic medical records of a health care organization across three internal medicine clinics on the downtown campus and seven satellite family medicine clinics. A pre- and postintervention design was used to assess the template utilization rate and clinician satisfaction.ResultsThe preoperative template utilization rate increased from 51.2% at baseline to 66.5% after the revamped template "went live" (p

BackgroundElectronic health record (EHR)-based clinical decision support is a scalable way to help standardize clinical care. Clinical decision support systems have not been extensively investigated in cirrhosis management. Human-centered design (HCD) is an approach that engages with potential users in intervention development. In this study, we applied HCD to design the features and interface for a clinical decision support system for cirrhosis management, called CirrhosisRx.MethodsWe conducted technical feasibility assessments to construct a visual blueprint that outlines the basic features of the interface. We then convened collaborative-design workshops with generalist and specialist clinicians. We elicited current workflows for cirrhosis management, assessed gaps in existing EHR systems, evaluated potential features, and refined the design prototype for CirrhosisRx. At the conclusion of each workshop, we analyzed recordings and transcripts.ResultsWorkshop feedback showed that the aggregation of relevant clinical data into 6 cirrhosis decompensation domains (defined as common inpatient clinical scenarios) was the most important feature. Automatic inference of clinical events from EHR data, such as gastrointestinal bleeding from hemoglobin changes, was not accepted due to accuracy concerns. Visualizations for risk stratification scores were deemed not necessary. Lastly, the HCD co-design workshops allowed us to identify the target user population (generalists).ConclusionsThis is one of the first applications of HCD to design the features and interface for an electronic intervention for cirrhosis management. The HCD process altered features, modified the design interface, and likely improved CirrhosisRx's overall usability. The finalized design for CirrhosisRx proceeded to development and production and will be tested for effectiveness in a pragmatic randomized controlled trial. This work provides a model for the creation of other EHR-based interventions in hepatology care.

Background: Several clinical trials support the efficacy of music therapy (MT) for improving outcomes in hospitalized patients, but few studies have evaluated the real-world delivery and integration of MT across multiple medical centers. This article describes the rationale, design, and population characteristics of a retrospective study examining the delivery and integration of MT within a large health system. Methods: A retrospective electronic health record (EHR) review was conducted of hospitalized patients seen by and/or referred to MT between January 2017 and July 2020. MT was provided across ten medical centers, including an academic medical center, a freestanding cancer center, and eight community hospitals. Discrete demographic, clinical, and MT treatment and referral characteristics were extracted from the EHR, cleaned, and organized using regular expressions functions, and they were summarized using descriptive statistics. Results: The MT team (average 11.6 clinical fulltime equivalent staff/year) provided 14,261 sessions to 7378 patients across 9091 hospitalizations. Patients were predominantly female (63.7%), White (54.3%) or Black/African American (44.0%), 63.7 ± 18.5 years of age at admission, and insured under Medicare (51.1%), Medicaid (18.1%), or private insurance (14.2%). Patients hospitalizations (median length of stay: 5 days) were primarily for cardiovascular (11.8%), respiratory (9.9%), or musculoskeletal (8.9%) conditions. Overall, 39.4% of patients hospital admissions included a mental health diagnosis, and 15.4% were referred to palliative care. Patients were referred by physicians (34.7%), nurses (29.4%), or advanced practice providers (24.7%) for coping (32.0%), anxiety reduction (20.4%), or pain management (10.1%). Therapists provided sessions to patients discharged from medical/surgical (74.5%), oncology (18.4%), or intensive care (5.8%) units. Conclusions: This retrospective study indicates that MT can be integrated across a large health system for addressing the needs of socioeconomically diverse patients. However, future research is needed to assess MTs impact on health care utilization (i.e., length of stay and rates of readmission) and immediate patient-reported outcomes.

The use of a stratified psychiatry approach that combines electronic health records (EHR) data with machine learning (ML) is one potentially fruitful path toward rapidly improving precision treatment in clinical practice. This strategy, however, requires confronting pervasive methodological flaws as well as deficiencies in transparency and reporting in the current conduct of ML-based studies for treatment prediction. EHR data shares many of the same data quality issues as other types of data used in ML prediction, plus some unique challenges. To fully leverage EHR datas power for patient stratification, increased attention to data quality and collection of patient-reported outcome data is needed.

The accurate extraction of surgical data from electronic health records (EHRs), particularly operative notes through manual chart review (MCR), is complex, crucial, and time-intensive, limited by human error due to fatigue and the level of training. This study aimed to develop and validate a novel Natural Language Processing (NLP) algorithm integrated with a Large Language Model (LLM; GPT4-Turbo) to automate the extraction of spinal surgery data from EHRs. The algorithm employed a two-stage approach. Initially, a rule-based NLP framework reviewed and classified candidate segments from the text, preserving their reference segments. These segments were then verified in the second stage through the LLM. The primary outcomes of this study were the accurate extraction of surgical data, including the type of surgery, levels operated, number of disks removed, and presence of intraoperative incidental durotomies. Secondary objectives explored time efficiency, tokenization lengths, and costs. The performance of the algorithm was assessed across two validation databases, analyzing metrics such as accuracy, sensitivity, discrimination, F1-score, and precision, with 95% confidence intervals calculated using percentile-based bootstrapping. The NLP + LLM algorithm markedly outperformed all performance metrics, demonstrating significant improvements in time and cost efficiency. These results suggest the potential for widespread adoption of this technology. [ABSTRACT FROM AUTHOR]

BACKGROUND: Prescribing non-opioid pain medications, such as non-steroidal anti-inflammatory (NSAIDs) medications, has been shown to reduce pain and decrease opioid use, but it is unclear how to effectively encourage multimodal pain medication prescribing for hospitalised patients. Therefore, the aim of this study is to evaluate the effect of prechecking non-opioid pain medication orders on clinician prescribing of NSAIDs among hospitalised adults. METHODS: This was a cluster randomised controlled trial of adult (≥18 years) hospitalised patients admitted to three hospital sites under one quaternary hospital system in the USA from 2 March 2022 to 3 March 2023. A multimodal pain order panel was embedded in the admission order set, with NSAIDs prechecked in the intervention group. The intervention group could uncheck the NSAID order. The control group had access to the same NSAID order. The primary outcome was an increase in NSAID ordering. Secondary outcomes include NSAID administration, inpatient pain scores and opioid use and prescribing and relevant clinical harms including acute kidney injury, new gastrointestinal bleed and in-hospital death. RESULTS: Overall, 1049 clinicians were randomised. The study included 6239 patients for a total of 9595 encounters. Both NSAID ordering (36 vs 43%, p

OBJECTIVES: Low-value esophagogastroduodenoscopies (EGDs) for uncomplicated gastro-oesophageal reflux disease (GERD) can harm patients and raise patient and payer costs. We developed an electronic health record (EHR) eMeasure to detect low-value EGDs. DESIGN: Retrospective cohort of 518 adult patients diagnosed with GERD who underwent initial EGD between 1 January 2019 and 31 December 2019. SETTING: Outpatient primary care and gastroenterology clinics at a large, urban, academic health centre. PARTICIPANTS: Adult primary care patients at the University of California Los Angeles who underwent initial EGD for GERD in 2019. MAIN OUTCOME MEASURES: EGD appropriateness criteria were based on the American College of Gastroenterology 2012 guidelines. An initial EGD was considered low-value if it lacked a documented guideline-based indication, including alarm symptoms (eg, iron-deficiency anaemia); failure of an 8-week proton pump inhibitor trial or elevated Barretts oesophagus risk. We performed manual chart review on a random sample of 204 patients as a gold standard of the eMeasures validity. We estimated EGD costs using Medicare physician and facility fee rates. RESULTS: Among 518 initial EGDs performed (mean age 53 years; 54% female), the eMeasure identified 81 (16%) as low-value. The eMeasures sensitivity was 42% (95% CI 22 to 61) and specificity was 93% (95% CI 89 to 96). Stratifying across clinics, 62 (74.6%) low-value EGDs originated from 2 (12.5%) out of 16 clinics. Total cost for 81 low-value EGDs was approximately US$75 573, including US$14 985 in patients out-of-pocket costs. CONCLUSIONS: We developed a highly specific eMeasure that showed that low-value EGDs occurred frequently in our healthcare system and were concentrated in a minority of clinics. These results can inform future QI efforts at our institution, such as best practice alerts for the ordering physician. Moreover, this open-source eMeasure has a much broader potential impact, as it can be integrated into any EHR and improve medical decision-making at the point of care.

MOTIVATION: Methods for concept recognition (CR) in clinical texts have largely been tested on abstracts or articles from the medical literature. However, texts from electronic health records (EHRs) frequently contain spelling errors, abbreviations, and other nonstandard ways of representing clinical concepts. RESULTS: Here, we present a method inspired by the BLAST algorithm for biosequence alignment that screens texts for potential matches on the basis of matching k-mer counts and scores candidates based on conformance to typical patterns of spelling errors derived from 2.9 million clinical notes. Our method, the Term-BLAST-like alignment tool (TBLAT) leverages a gold standard corpus for typographical errors to implement a sequence alignment-inspired method for efficient entity linkage. We present a comprehensive experimental comparison of TBLAT with five widely used tools. Experimental results show an increase of 10% in recall on scientific publications and 20% increase in recall on EHR records (when compared against the next best method), hence supporting a significant enhancement of the entity linking task. The method can be used stand-alone or as a complement to existing approaches. AVAILABILITY AND IMPLEMENTATION: Fenominal is a Java library that implements TBLAT for named CR of Human Phenotype Ontology terms and is available at https://github.com/monarch-initiative/fenominal under the GNU General Public License v3.0.

Background and aimsNonalcoholic fatty liver disease (NAFLD) has reached pandemic proportions. Early detection can identify at-risk patients who can be linked to hepatology care. The vibration-controlled transient elastography (VCTE) controlled attenuation parameter (CAP) is biopsy validated to diagnose hepatic steatosis (HS). We aimed to develop a novel clinical predictive algorithm for HS using the CAP score at a Veterans' Affairs hospital.MethodsWe identified 403 patients in the Greater Los Angeles VA Healthcare System with valid VCTEs during 1/2018-6/2020. Patients with alcohol-associated liver disease, genotype 3 hepatitis C, any malignancies, or liver transplantation were excluded. Linear regression was used to identify predictors of NAFLD. To identify a CAP threshold for HS detection, receiver operating characteristic analysis was applied using liver biopsy, MRI, and ultrasound as the gold standards.ResultsThe cohort was racially/ethnically diverse (26% Black/African American; 20% Hispanic). Significant positive predictors of elevated CAP score included diabetes, cholesterol, triglycerides, BMI, and self-identifying as Hispanic. Our predictions of CAP scores using this model strongly correlated (r = 0.61, p

Abstract The accurate extraction of surgical data from electronic health records (EHRs), particularly operative notes through manual chart review (MCR), is complex, crucial, and time-intensive, limited by human error due to fatigue and the level of training. This study aimed to develop and validate a novel Natural Language Processing (NLP) algorithm integrated with a Large Language Model (LLM; GPT4-Turbo) to automate the extraction of spinal surgery data from EHRs. The algorithm employed a two-stage approach. Initially, a rule-based NLP framework reviewed and classified candidate segments from the text, preserving their reference segments. These segments were then verified in the second stage through the LLM. The primary outcomes of this study were the accurate extraction of surgical data, including the type of surgery, levels operated, number of disks removed, and presence of intraoperative incidental durotomies. Secondary objectives explored time efficiency, tokenization lengths, and costs. The performance of the algorithm was assessed across two validation databases, analyzing metrics such as accuracy, sensitivity, discrimination, F1-score, and precision, with 95% confidence intervals calculated using percentile-based bootstrapping. The NLP + LLM algorithm markedly outperformed all performance metrics, demonstrating significant improvements in time and cost efficiency. These results suggest the potential for widespread adoption of this technology.

BACKGROUND: Comorbidities and symptoms in metastatic breast cancer patients impact treatment decisions and influence prognosis and quality of life. The objective of this study is to examine the concordance between physician documentation and patient reports of comorbidities and symptoms to understand their comparative effectiveness. METHODS: New patients with metastatic breast cancer completed an electronic intake survey assessing patient health history and symptoms. Physician documentation across 54 comorbidities and 42 symptoms was abstracted from notes for the corresponding clinic visits between November 2016 and March 2020. Concordance between patient reports and medical records for each condition and hazards ratios for each patient versus physician reported comorbidity and symptom were assessed. RESULTS: A total of 168 patients were included in the analysis (age, median = 56 years, range = 29-86 years; 131 white [78.9%]). Twenty-three of 54 comorbidities had a moderate to high level of agreement between patients and physicians (κ ≥ 0.40). Physicians documented higher numbers of comorbidities that can be objectively measured which also had higher concordance (e.g., diabetes [κ = 0.83] and hypertension [κ = 0.79]) while patients reported higher numbers of comorbidities that are more subjective which also had lower concordance (anxiety [κ = 0.30], GERD [κ = 0.36]). One physician-documented and two patient-reported comorbidities were significantly associated with survival (p

BackgroundLung cancer screening (LCS) is recommended for individuals at high risk due to age and smoking history after a shared decision-making conversation. However, little is known about best strategies for incorporating shared decision-making, especially in a busy primary care setting.ObjectiveTo develop a novel tool, Lung Cancer Assessment of Risk and Education (LungCARE) to guide LCS decisions among eligible primary care patients.DesignPilot cluster randomized controlled trial of LungCARE versus usual care.ParticipantsPatients of providers in a university primary care clinic, who met criteria for LCS.InterventionProviders were randomized to LungCARE intervention or control. LungCARE participants completed a computer tablet-based video assessment of lung cancer educational needs in the waiting room prior to a primary care visit. Patient and provider both received a summary handout of patient concerns and responses.Main measuresAll eligible patients completed baseline interviews by telephone. One week after the index visit, participants completed a follow-up telephone survey that assessed patient-physician discussion of LCS, referral to and scheduling of LCS, as well as LCS knowledge and acceptability of LungCARE. Two months after index visit, we reviewed patients' electronic health records (EHRs) for evidence of a shared decision-making conversation and referral to and receipt of LCS.Key resultsA total of 66 participants completed baseline and follow-up visits (34: LungCARE; 32: usual care). Mean age was 65.9 (± 6.0). Based on EHR review, compared to usual care, LungCARE participants were more likely to have discussed LCS with their physicians (56% vs 25%; p = 0.04) and to be referred to LCS (44% vs 13%; p

PURPOSE: Using a novel, electronic health record (EHR)-based approach, to estimate the prevalence of prescription medication use at 2, 4, and 6 months postpartum among lactating individuals. METHODS: We utilized automated EHR data from a US health system that records infant feeding information at well-child visits. We linked mothers who received prenatal care to their infants born May 2018-June 2019, and we required infants to have ≥1 well-child visit between 31 and 90 days of life (i.e., 2-month well-child visit with a ±1 month window). Mothers were classified as lactating at the 2-month well-child visit if their infant received breast milk at the 2-month well-child visit. For subsequent well-child visits at 4 and 6 months, mothers were considered lactating if their infant was still receiving breast milk. RESULTS: We identified 6013 mothers meeting inclusion criteria, and 4158 (69.2%) were classified as lactating at the 2-month well-child visit. Among those classified as lactating, the most common medication classes dispensed around the 2-month well-child visit were oral progestin contraceptives (19.1%), selective serotonin reuptake inhibitors (8.8%), first generation cephalosporins (4.3%), thyroid hormones (3.5%), nonsteroidal anti-inflammatory agents (3.4%), penicillinase-resistant penicillins (3.1%), topical corticosteroids (2.9%), and oral imidazole-related antifungals (2.0%). The most common medication classes were similar around the 4 and 6-month well-child visits although prevalence estimates were often lower. CONCLUSIONS: Progestin-only contraceptives, antidepressants, and antibiotics were the most dispensed medications among lactating mothers. With routine collection of breastfeeding information, mother-infant linked EHR data may overcome limitations in previous studies of medication utilization during lactation. These data should be considered for studies of medication safety during lactation given the need for human safety data.

BACKGROUND: For patients who may permanently or temporarily lose their ability to communicate preferences, advance care planning is a critical mechanism to guide medical decision-making but is currently underused among surgical patients. METHODS: A resident-led quality improvement project, including education and performance measurement, was conducted on an emergency general surgery service to increase the completion of inpatient advance care planning notes using a specialized template in the electronic health record. Advance care planning documentation was defined as either preadmission advance care planning documentation (eg, advance directive) or inpatient advance care planning (use of the electronic health record template). Data from patients admitted to the emergency general surgery service for 12+ hours were analyzed, and baseline data (July 2020 to June 2021) were compared with data from the intervention period (July 2021 to June 2022). The chart review evaluated the content of the inpatient advance care planning documentation from the intervention period. RESULTS: The frequency of inpatient advance care planning documentation increased (9.3%, n = 56 to 16.6%, n = 92, P < .001) with a greater contribution of inpatient advance care planning notes by the surgery team (16.7% to 55.4%) in the intervention period. Content analysis indicated that 79.0% of inpatient advance care planning notes listed preferences for life-sustaining therapy, 78.3% listed surrogacy, 57.3% listed overall health goals, and 50.3% listed treatment goals specific to the surgical encounter. CONCLUSION: Although a resident-led quality improvement project contributed to greater adoption of standardized inpatient advance care planning documentation on an emergency general surgery service, progress was slow, and integration into standard work was not achieved. Future efforts are needed to better understand the integration of essential advance care planning elements into workflows and to establish inclusive educational programming to prepare all team members for conducting and documenting advance care planning conversations.

Aim: To evaluate treatment effects of eteplirsen among patients with Duchenne muscular dystrophy. Methods: Using real-world claims and electronic medical record data, this retrospective comparative analysis assessed eteplirsen-treated and control cohorts matched by age, disease progression state, and pre-index period healthcare resource utilization. Poisson regression was used to evaluate eteplirsen effects on healthcare resource utilization outcomes. Results: Eteplirsen was associated with statistically significant reductions in rates of hospital encounters (31%), emergency room visits (31%), need for pulmonary management (33%), cardiac management (21%), tracheostomy (86%), and assisted ventilation (39%) versus the control group. Other assessed outcomes favored eteplirsen numerically but did not all reach statistical significance. Conclusion: Eteplirsen-treated patients had reduced rates of multiple healthcare resource utilization measures versus matched controls.

Gestational diabetes mellitus (GDM) is a major pregnancy complication affecting approximately 14.0% of pregnancies around the world. Air pollution exposure, particularly exposure to PM2.5, has become a major environmental issue affecting health, especially for vulnerable pregnant women. Associations between PM2.5 exposure and adverse birth outcomes are generally assumed to be the same throughout a large geographical area. However, the effects of air pollution on health can very spatially in subpopulations. Such spatially varying effects are likely due to a wide range of contextual neighborhood and individual factors that are spatially correlated, including SES, demographics, exposure to housing characteristics and due to different composition of particulate matter from different emission sources. This combination of elevated environmental hazards in conjunction with socioeconomic-based disparities forms what has been described as a double jeopardy for marginalized sub-populations. In this manuscript our analysis combines both an examination of spatially varying effects of a) unit-changes in exposure and examines effects of b) changes from current exposure levels down to a fixed compliance level, where compliance levels correspond to the Air Quality Standards (AQS) set by the U.S. Environmental Protection Agency (EPA) and World Health Organization (WHO) air quality guideline values. Results suggest that exposure reduction policies should target certain hotspot areas where size and effects of potential reductions will reap the greatest rewards in terms of health benefits, such as areas of southeast Los Angeles County which experiences high levels of PM2.5 exposures and consist of individuals who may be particularly vulnerable to the effects of air pollution on the risk of GDM.

OBJECTIVE: This study assessed contributing factors associated with dental adverse events (AEs). METHODS: Seven electronic health record-based triggers were deployed identifying potential AEs at 2 dental institutions. From 4106 flagged charts, 2 reviewers examined 439 charts selected randomly to identify and classify AEs using our dental AE type and severity classification systems. Based on information captured in the electronic health record, we analyzed harmful AEs to assess potential contributing factors; harmful AEs were defined as those that resulted in temporary moderate to severe harm, required hospitalization, or resulted in permanent moderate to severe harm. We classified potential contributing factors according to (1) who was involved (person), (2) what were they doing (tasks), (3) what tools/technologies were they using (tools/technologies), (4) where did the event take place (environment), (5) what organizational conditions contributed to the event? (organization), (6) patient (including parents), and (7) professional-professional collaboration. A blinded panel of dental experts conducted a second review to confirm the presence of an AE. RESULTS: Fifty-nine cases had 1 or more harmful AEs. Pain occurred most frequently (27.1%), followed by nerve injury (16.9%), hard tissue injury (15.2%), and soft tissue injury (15.2%). Forty percent of the cases were classified as temporary not moderate to severe harm. Person (training, supervision, and fatigue) was the most common contributing factor (31.5%), followed by patient (noncompliance, unsafe practices at home, low health literacy, 17.1%), and professional-professional collaboration (15.3%). CONCLUSIONS: Pain was the most common harmful AE identified. Person, patient, and professional-professional collaboration were the most frequently assessed factors associated with harmful AEs.

INTRODUCTION: Vaccine safety surveillance commonly includes a serial testing approach with a sensitive method for signal generation and specific method for signal validation. The extent to which serial testing in real-world studies improves or hinders overall performance in terms of sensitivity and specificity remains unknown. METHODS: We assessed the overall performance of serial testing using three administrative claims and one electronic health record database. We compared type I and II errors before and after empirical calibration for historical comparator, self-controlled case series (SCCS), and the serial combination of those designs against six vaccine exposure groups with 93 negative control and 279 imputed positive control outcomes. RESULTS: The historical comparator design mostly had fewer type II errors than SCCS. SCCS had fewer type I errors than the historical comparator. Before empirical calibration, the serial combination increased specificity and decreased sensitivity. Type II errors mostly exceeded 50%. After empirical calibration, type I errors returned to nominal; sensitivity was lowest when the methods were combined. CONCLUSION: While serial combination produced fewer false-positive signals compared with the most specific method, it generated more false-negative signals compared with the most sensitive method. Using a historical comparator design followed by an SCCS analysis yielded decreased sensitivity in evaluating safety signals relative to a one-stage SCCS approach. While the current use of serial testing in vaccine surveillance may provide a practical paradigm for signal identification and triage, single epidemiological designs should be explored as valuable approaches to detecting signals.

BACKGROUND: Timely discharge to post-acute care (PAC) settings, such as skilled nursing facilities, requires early identification of eligible patients. We sought to develop and internally validate a model which predicts a patients likelihood of requiring PAC based on information obtained in the first 24 h of hospitalization. METHODS: This was a retrospective observational cohort study. We collected clinical data and commonly used nursing assessments from the electronic health record (EHR) for all adult inpatient admissions at our academic tertiary care center from September 1, 2017 to August 1, 2018. We performed a multivariable logistic regression to develop the model from the derivation cohort of the available records. We then evaluated the capability of the model to predict discharge destination on an internal validation cohort. RESULTS: Age (adjusted odds ratio [AOR], 1.04 [per year]; 95% Confidence Interval [CI], 1.03 to 1.04), admission to the intensive care unit (AOR, 1.51; 95% CI, 1.27 to 1.79), admission from the emergency department (AOR, 1.53; 95% CI, 1.31 to 1.78), more home medication prescriptions (AOR, 1.06 [per medication count increase]; 95% CI 1.05 to 1.07), and higher Morse fall risk scores at admission (AOR, 1.03 [per unit increase]; 95% CI 1.02 to 1.03) were independently associated with higher likelihood of being discharged to PAC facility. The c-statistic of the model derived from the primary analysis was 0.875, and the model predicted the correct discharge destination in 81.2% of the validation cases. CONCLUSIONS: A model that utilizes baseline clinical factors and risk assessments has excellent model performance in predicting discharge to a PAC facility.

ObjectiveOccupational health (OH) documentation has traditionally been separate from health system electronic health records (EHRs), but this can create patient safety and care continuity challenges. Herein, we describe outcomes and challenges of such integration including how one health system managed compliance with laws, regulations, and ethical principles concerning digital privacy.MethodsOccupational health integration with the enterprise EHR at the University of California San Diego Health was started in June 2021 and completed in December 2021.ResultsIntegrating with the enterprise EHR allowed for a secure telehealth system, faster visit times, digitization of questionnaires medical clearance forms, and improved reporting capabilities.ConclusionsIntegration and interoperability are fundamental building blocks to any OH EHR solution and will allow for evaluation of worker population trends, and targeted interventions to improve worker health status.

ObjectivesAs the real-world electronic health record (EHR) data continue to grow exponentially, novel methodologies involving artificial intelligence (AI) are becoming increasingly applied to enable efficient data-driven learning and, ultimately, to advance healthcare. Our objective is to provide readers with an understanding of evolving computational methods and help in deciding on methods to pursue.Target audienceThe sheer diversity of existing methods presents a challenge for health scientists who are beginning to apply computational methods to their research. Therefore, this tutorial is aimed at scientists working with EHR data who are early entrants into the field of applying AI methodologies.ScopeThis manuscript describes the diverse and growing AI research approaches in healthcare data science and categorizes them into 2 distinct paradigms, the bottom-up and top-down paradigms to provide health scientists venturing into artificial intelligent research with an understanding of the evolving computational methods and help in deciding on methods to pursue through the lens of real-world healthcare data.

BackgroundAlzheimer's Disease (AD) is a complex clinical phenotype with unprecedented social and economic tolls on an ageing global population. Real-world data (RWD) from electronic health records (EHRs) offer opportunities to accelerate precision drug development and scale epidemiological research on AD. A precise characterization of AD cohorts is needed to address the noise abundant in RWD.MethodsWe conducted a retrospective cohort study to develop and test computational models for AD cohort identification using clinical data from 8 Massachusetts healthcare systems. We mined temporal representations from EHR data using the transitive sequential pattern mining algorithm (tSPM) to train and validate our models. We then tested our models against a held-out test set from a review of medical records to adjudicate the presence of AD. We trained two classes of Machine Learning models, using Gradient Boosting Machine (GBM), to compare the utility of AD diagnosis records versus the tSPM temporal representations (comprising sequences of diagnosis and medication observations) from electronic medical records for characterizing AD cohorts.FindingsIn a group of 4985 patients, we identified 219 tSPM temporal representations (i.e., transitive sequences) of medical records for constructing the best classification models. The models with sequential features improved AD classification by a magnitude of 3-16 percent over the use of AD diagnosis codes alone. The computed cohort included 663 patients, 35 of whom had no record of AD. Six groups of tSPM sequences were identified for characterizing the AD cohorts.InterpretationWe present sequential patterns of diagnosis and medication codes from electronic medical records, as digital markers of Alzheimer's Disease. Classification algorithms developed on sequential patterns can replace standard features from EHRs to enrich phenotype modelling.FundingNational Institutes of Health: the National Institute on Aging (RF1AG074372) and the National Institute of Allergy and Infectious Diseases (R01AI165535).