Rapid identification of inflammatory arthritis and associated adverse events following immune checkpoint therapy: a machine learning approach.

Introduction: Immune checkpoint inhibitor-induced inflammatory arthritis (ICI-IA) poses a major clinical challenge to ICI therapy for cancer, with 13% of cases halting ICI therapy and ICI-IA being difficult to identify for timely referral to a rheumatologist. The objective of this study was to rapidly identify ICI-IA patients in clinical data and assess associated immune-related adverse events (irAEs) and risk factors. Methods: We conducted a retrospective study of the electronic health records (EHRs) of 89 patients who developed ICI-IA out of 2451 cancer patients who received ICI therapy at Northwestern University between March 2011 to January 2021. Logistic regression and random forest machine learning models were trained on all EHR diagnoses, labs, medications, and procedures to identify ICI-IA patients and EHR codes indicating ICI-IA. Multivariate logistic regression was then used to test associations between ICI-IA and cancer type, ICI regimen, and comorbid irAEs. Results: Logistic regression and random forest models identified ICI-IA patients with accuracies of 0.79 and 0.80, respectively. Key EHR features from the random forest model included ICI-IA relevant features (joint pain, steroid prescription, rheumatoid factor tests) and features suggesting comorbid irAEs (thyroid function tests, pruritus, triamcinolone prescription). Compared to 871 adjudicated ICI patients who did not develop arthritis, ICI-IA patients had higher odds of developing cutaneous (odds ratio [OR]=2.66; 95% Confidence Interval [CI] 1.63-4.35), endocrine (OR=2.09; 95% CI 1.15-3.80), or gastrointestinal (OR=2.88; 95% CI 1.76-4.72) irAEs adjusting for demographics, cancer type, and ICI regimen. Melanoma (OR=1.99; 95% CI 1.08-3.65) and renal cell carcinoma (OR=2.03; 95% CI 1.06-3.84) patients were more likely to develop ICI-IA compared to lung cancer patients. Patients on nivolumab+ipilimumab were more likely to develop ICI-IA compared to patients on pembrolizumab (OR=1.86; 95% CI 1.01-3.43). Discussion: Our machine learning models rapidly identified patients with ICI-IA in EHR data and elucidated clinical features indicative of comorbid irAEs. Patients with ICI-IA were significantly more likely to also develop cutaneous, endocrine, and gastrointestinal irAEs during their clinical course compared to ICI therapy patients without ICI-IA.

Natural language processing to identify lupus nephritis phenotype in electronic health records.

BACKGROUND: Systemic lupus erythematosus (SLE) is a rare autoimmune disorder characterized by an unpredictable course of flares and remission with diverse manifestations. Lupus nephritis, one of the major disease manifestations of SLE for organ damage and mortality, is a key component of lupus classification criteria. Accurately identifying lupus nephritis in electronic health records (EHRs) would therefore benefit large cohort observational studies and clinical trials where characterization of the patient population is critical for recruitment, study design, and analysis. Lupus nephritis can be recognized through procedure codes and structured data, such as laboratory tests. However, other critical information documenting lupus nephritis, such as histologic reports from kidney biopsies and prior medical history narratives, require sophisticated text processing to mine information from pathology reports and clinical notes. In this study, we developed algorithms to identify lupus nephritis with and without natural language processing (NLP) using EHR data from the Northwestern Medicine Enterprise Data Warehouse (NMEDW). METHODS: We developed five algorithms: a rule-based algorithm using only structured data (baseline algorithm) and four algorithms using different NLP models. The first NLP model applied simple regular expression for keywords search combined with structured data. The other three NLP models were based on regularized logistic regression and used different sets of features including positive mention of concept unique identifiers (CUIs), number of appearances of CUIs, and a mixture of three components (i.e. a curated list of CUIs, regular expression concepts, structured data) respectively. The baseline algorithm and the best performing NLP algorithm were externally validated on a dataset from Vanderbilt University Medical Center (VUMC). RESULTS: Our best performing NLP model incorporated features from both structured data, regular expression concepts, and mapped concept unique identifiers (CUIs) and showed improved F measure in both the NMEDW (0.41 vs 0.79) and VUMC (0.52 vs 0.93) datasets compared to the baseline lupus nephritis algorithm. CONCLUSION: Our NLP MetaMap mixed model improved the F-measure greatly compared to the structured data only algorithm in both internal and external validation datasets. The NLP algorithms can serve as powerful tools to accurately identify lupus nephritis phenotype in EHR for clinical research and better targeted therapies.

Privacy-preserving record linkage across disparate institutions and datasets to enable a learning health system: The national COVID cohort collaborative (N3C) experience.

Introduction: Research driven by real-world clinical data is increasingly vital to enabling learning health systems, but integrating such data from across disparate health systems is challenging. As part of the NCATS National COVID Cohort Collaborative (N3C), the N3C Data Enclave was established as a centralized repository of deidentified and harmonized COVID-19 patient data from institutions across the US. However, making this data most useful for research requires linking it with information such as mortality data, images, and viral variants. The objective of this project was to establish privacy-preserving record linkage (PPRL) methods to ensure that patient-level EHR data remains secure and private when governance-approved linkages with other datasets occur. Methods: Separate agreements and approval processes govern N3C data contribution and data access. The Linkage Honest Broker (LHB), an independent neutral party (the Regenstrief Institute), ensures data linkages are robust and secure by adding an extra layer of separation between protected health information and clinical data. The LHB's PPRL methods (including algorithms, processes, and governance) match patient records using "deidentified tokens," which are hashed combinations of identifier fields that define a match across data repositories without using patients' clear-text identifiers. Results: These methods enable three linkage functions: Deduplication, Linking Multiple Datasets, and Cohort Discovery. To date, two external repositories have been cross-linked. As of March 1, 2023, 43 sites have signed the LHB Agreement; 35 sites have sent tokens generated for 9 528 998 patients. In this initial cohort, the LHB identified 135 037 matches and 68 596 duplicates. Conclusion: This large-scale linkage study using deidentified datasets of varying characteristics established secure methods for protecting the privacy of N3C patient data when linked for research purposes. This technology has potential for use with registries for other diseases and conditions.

Multiarterial vs Single-Arterial Coronary Surgery: 10-Year Follow-up of 1 Million Patients.

BACKGROUND: Although many options exist for multivessel coronary revascularization, controversy persists over whether multiarterial grafting (MAG) confers a survival advantage over single-arterial grafting (SAG) with saphenous vein in coronary artery bypass grafting (CABG). This study sought to compare longitudinal survival between patients undergoing MAG and those undergoing SAG. METHODS: All patients undergoing isolated CABG with ≥2 bypass grafts in The Society of Thoracic Surgeons Adult Cardiac Surgery Database (2008-2019) were linked to the National Death Index. Risk adjustment was performed using inverse probability weighting and multivariable modeling. The primary end point was longitudinal survival. Subpopulation analyses were performed and volume thresholds were analyzed to determine optimal benefit. RESULTS: A total of 1,021,632 patients underwent isolated CABG at 1108 programs (100,419 MAG [9.83%]; 920,943 SAG [90.17%]). Median follow-up was 5.30 years (range, 0-12 years). After risk adjustment, all characteristics were well balanced. At 10 years, MAG was associated with improved unadjusted (hazard ratio, 0.59; 95% CI 0.58-0.61) and adjusted (hazard ratio, 0.86; 95% CI, 0.85-0.88) 10-year survival. Center volume of ≥10 MAG cases/year was associated with benefit. MAG was associated with an overall survival advantage over SAG in all subgroups, including stable coronary disease, acute coronary syndrome, and acute infarction. Survival was equivalent to that with SAG for patients age ≥80 years and those with severe heart failure, renal failure, peripheral vascular disease, or obesity. Only patients with a body mass index ≥40 kg/m2 had superior survival with SAG. CONCLUSIONS: Multiarterial CABG is associated with superior long-term survival and should be the surgical multivessel revascularization strategy of choice for patients with a body mass index of less than 40 kg/m2.

Genetic variation in the human leukocyte antigen region confers susceptibility to Clostridioides difficile infection.

Clostridioides difficile (C. diff.) infection (CDI) is a leading cause of hospital acquired diarrhea in North America and Europe and a major cause of morbidity and mortality. Known risk factors do not fully explain CDI susceptibility, and genetic susceptibility is suggested by the fact that some patients with colons that are colonized with C. diff. do not develop any infection while others develop severe or recurrent infections. To identify common genetic variants associated with CDI, we performed a genome-wide association analysis in 19,861 participants (1349 cases; 18,512 controls) from the Electronic Medical Records and Genomics (eMERGE) Network. Using logistic regression, we found strong evidence for genetic variation in the DRB locus of the MHC (HLA) II region that predisposes individuals to CDI (P > 1.0 × 10-14; OR 1.56). Altered transcriptional regulation in the HLA region may play a role in conferring susceptibility to this opportunistic enteric pathogen.

Identification and assessment of classification criteria attributes for systemic lupus erythematosus in a regional medical record data network.

OBJECTIVE: To assess the application and utility of algorithms designed to detect features of SLE in electronic health record (EHR) data in a multisite, urban data network. METHODS: Using the Chicago Area Patient-Centered Outcomes Research Network (CAPriCORN), a Clinical Data Research Network (CDRN) containing data from multiple healthcare sites, we identified patients with at least one positively identified criterion from three SLE classification criteria sets developed by the American College of Rheumatology (ACR) in 1997, the Systemic Lupus International Collaborating Clinics (SLICC) in 2012, and the European Alliance of Associations for Rheumatology and the ACR in 2019 using EHR-based algorithms. To measure the algorithms' performance in this data setting, we first evaluated whether the number of clinical encounters for SLE was associated with a greater quantity of positively identified criteria domains using Poisson regression. We next quantified the amount of SLE criteria identified at a single healthcare institution versus all sites to assess the amount of SLE-related information gained from implementing the algorithms in a CDRN. RESULTS: Patients with three or more SLE encounters were estimated to have documented 2.77 (2.73 to 2.80) times the number of positive SLE attributes from the 2012 SLICC criteria set than patients without an SLE encounter via Poisson regression. Patients with three or more SLE-related encounters and with documented care from multiple institutions were identified with more SLICC criteria domains when data were included from all CAPriCORN sites compared with a single site (p<0.05). CONCLUSIONS: The positive association observed between amount of SLE-related clinical encounters and the number of criteria domains detected suggests that the algorithms used in this study can be used to help describe SLE features in this data environment. This work also demonstrates the benefit of aggregating data across healthcare institutions for patients with fragmented care.

An exploratory survey about using ChatGPT in education, healthcare, and research.

OBJECTIVE: ChatGPT is the first large language model (LLM) to reach a large, mainstream audience. Its rapid adoption and exploration by the population at large has sparked a wide range of discussions regarding its acceptable and optimal integration in different areas. In a hybrid (virtual and in-person) panel discussion event, we examined various perspectives regarding the use of ChatGPT in education, research, and healthcare. MATERIALS AND METHODS: We surveyed in-person and online attendees using an audience interaction platform (Slido). We quantitatively analyzed received responses on questions about the use of ChatGPT in various contexts. We compared pairwise categorical groups with a Fisher's Exact. Furthermore, we used qualitative methods to analyze and code discussions. RESULTS: We received 420 responses from an estimated 844 participants (response rate 49.7%). Only 40% of the audience had tried ChatGPT. More trainees had tried ChatGPT compared with faculty. Those who had used ChatGPT were more interested in using it in a wider range of contexts going forwards. Of the three discussed contexts, the greatest uncertainty was shown about using ChatGPT in education. Pros and cons were raised during discussion for the use of this technology in education, research, and healthcare. DISCUSSION: There was a range of perspectives around the uses of ChatGPT in education, research, and healthcare, with still much uncertainty around its acceptability and optimal uses. There were different perspectives from respondents of different roles (trainee vs faculty vs staff). More discussion is needed to explore perceptions around the use of LLMs such as ChatGPT in vital sectors such as education, healthcare and research. Given involved risks and unforeseen challenges, taking a thoughtful and measured approach in adoption would reduce the likelihood of harm.

Differences in Health Professionals' Engagement With Electronic Health Records Based on Inpatient Race and Ethnicity.

Importance: US health professionals devote a large amount of effort to engaging with patients' electronic health records (EHRs) to deliver care. It is unknown whether patients with different racial and ethnic backgrounds receive equal EHR engagement. Objective: To investigate whether there are differences in the level of health professionals' EHR engagement for hospitalized patients according to race or ethnicity during inpatient care. Design, Setting, and Participants: This cross-sectional study analyzed EHR access log data from 2 major medical institutions, Vanderbilt University Medical Center (VUMC) and Northwestern Medicine (NW Medicine), over a 3-year period from January 1, 2018, to December 31, 2020. The study included all adult patients (aged ≥18 years) who were discharged alive after hospitalization for at least 24 hours. The data were analyzed between August 15, 2022, and March 15, 2023. Exposures: The actions of health professionals in each patient's EHR were based on EHR access log data. Covariates included patients' demographic information, socioeconomic characteristics, and comorbidities. Main Outcomes and Measures: The primary outcome was the quantity of EHR engagement, as defined by the average number of EHR actions performed by health professionals within a patient's EHR per hour during the patient's hospital stay. Proportional odds logistic regression was applied based on outcome quartiles. Results: A total of 243 416 adult patients were included from VUMC (mean [SD] age, 51.7 [19.2] years; 54.9% female and 45.1% male; 14.8% Black, 4.9% Hispanic, 77.7% White, and 2.6% other races and ethnicities) and NW Medicine (mean [SD] age, 52.8 [20.6] years; 65.2% female and 34.8% male; 11.7% Black, 12.1% Hispanic, 69.2% White, and 7.0% other races and ethnicities). When combining Black, Hispanic, or other race and ethnicity patients into 1 group, these patients were significantly less likely to receive a higher amount of EHR engagement compared with White patients (adjusted odds ratios, 0.86 [95% CI, 0.83-0.88; P < .001] for VUMC and 0.90 [95% CI, 0.88-0.92; P < .001] for NW Medicine). However, a reduction in this difference was observed from 2018 to 2020. Conclusions and Relevance: In this cross-sectional study of inpatient EHR engagement, the findings highlight differences in how health professionals distribute their efforts to patients' EHRs, as well as a method to measure these differences. Further investigations are needed to determine whether and how EHR engagement differences are correlated with health care outcomes.

Machine learning identification of patient clusters for cutaneous melanoma.

BACKGROUND: Cutaneous melanoma is a cancer arising in melanocyte skin cells and is the deadliest form of skin cancer worldwide. Although some risk factors are known, accurate prediction of disease progression and probability for metastasis are difficult to ascertain, given the complexity of the disease and the absence of reliable predictive markers. Since early detection and treatment are essential to enhance survival, this study utilizing machine learning (ML) aims to further delineate additional risk factors associated with cutaneous melanoma. METHODS: A Bayesian Gaussian Mixture ML model was created with data from 2056 patients diagnosed with cutaneous melanoma and then used to group the patients into six Clusters based on a Silhouette Score analysis. A t-distributed stochastic neighbor embedding (t-SNE) model was used to visualize the six Clusters. RESULTS: Statistical analysis revealed that Cluster 4 showed a significantly higher rate of metastatic disease, as well as higher Breslow depth at diagnosis, compared to the other five Clusters. Compared to the other five Clusters, patients represented in Cluster 4 also had lower healthcare utilization, fewer dermatology clinic visits, fewer primary care providers, and less frequent colonoscopies and mammograms, and were more likely to smoke and less likely to have a prior diagnosis of basal cell carcinoma. CONCLUSIONS: This study uncovers gaps in healthcare utilization of services among patient groups with cutaneous melanoma as well as possible implications for management of disease progression. Data-driven analyses emphasize the importance of routine clinic visits to dermatologists and/or primary care physicians (PCPs) for early detection and management of cutaneous melanoma. The findings from this study demonstrate that unsupervised ML methodology may serve to define the best candidate patients to benefit from enhanced dermatology/primary care which, in turn, is expected to improve outcomes for cutaneous melanoma.

The All of Us Data and Research Center: Creating a Secure, Scalable, and Sustainable Ecosystem for Biomedical Research.

The All of Us Research Program's Data and Research Center (DRC) was established to help acquire, curate, and provide access to one of the world's largest and most diverse datasets for precision medicine research. Already, over 500,000 participants are enrolled in All of Us, 80% of whom are underrepresented in biomedical research, and data are being analyzed by a community of over 2,300 researchers. The DRC created this thriving data ecosystem by collaborating with engaged participants, innovative program partners, and empowered researchers. In this review, we first describe how the DRC is organized to meet the needs of this broad group of stakeholders. We then outline guiding principles, common challenges, and innovative approaches used to build the All of Us data ecosystem. Finally, we share lessons learned to help others navigate important decisions and trade-offs in building a modern biomedical data platform.

Geographic Disparities in Re-triage Destinations Among Seriously Injured Californians.

Objective: To quantify geographic disparities in sub-optimal re-triage of seriously injured patients in California. Summary of Background Data: Re-triage is the emergent transfer of seriously injured patients from the emergency departments of non-trauma and low-level trauma centers to, ideally, high-level trauma centers. Some patients are re-triaged to a second non-trauma or low-level trauma center (sub-optimal) instead of a high-level trauma center (optimal). Methods: This was a retrospective observational cohort study of seriously injured patients, defined by an Injury Severity Score > 15, re-triaged in California (2009-2018). Re-triages within one day of presentation to the sending center were considered. The sub-optimal re-triage rate was quantified at the state, regional trauma coordinating committees (RTCC), local emergency medical service agencies, and sending center level. A generalized linear mixed-effects regression quantified the association of sub-optimality with the RTCC of the sending center. Geospatial analyses demonstrated geographic variations in sub-optimal re-triage rates and calculated alternative re-triage destinations. Results: There were 8,882 re-triages of seriously injured patients and 2,680 (30.2 %) were sub-optimal. Sub-optimally re-triaged patients had 1.5 higher odds of transfer to a third short-term acute care hospital and 1.25 increased odds of re-admission within 60 days from discharge. The sub-optimal re-triage rates increased from 29.3 % in 2009 to 38.6 % in 2018. 56.0 % of non-trauma and low-level trauma centers had at least one sub-optimal re-triage. The Southwest RTCC accounted for the largest proportion (39.8 %) of all sub-optimal re-triages in California. Conclusion: High population density geographic areas experienced higher sub-optimal re-triage rates.

Significance of Intraoperative Medication Data and Predictive Model Selection for Predicting Postoperative First-Time Atrial Fibrillation.

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in clinical practice and has a well-established association with coronary artery bypass graft (CABG) surgery. Being able to predict post-operative AF (POAF) may improve surgical outcomes. This study retrospectively assembled a large cohort of 3,807 first-time CABG patients with no prior AF to study factors that contribute to occurrence of POAF, in addition to testing models that may predict its incidence. Several clinical features with established relevance to POAF were extracted from the EHR, along with a record of medications administered intra-operatively. Tests of performance with logistic regression, decision tree, and neural network predictive models showed slight improvements when incorporating medication information. Analysis of the clinical and medications data indicate that there may be effects contributing to POAF incidence captured in the medication administration records. Our results show that improved predictive performance is achievable by incorporating a record of medications administered intra-operatively.

Multi-ancestry genome- and phenome-wide association studies of diverticular disease in electronic health records with natural language processing enriched phenotyping algorithm.

OBJECTIVE: Diverticular disease (DD) is one of the most prevalent conditions encountered by gastroenterologists, affecting ~50% of Americans before the age of 60. Our aim was to identify genetic risk variants and clinical phenotypes associated with DD, leveraging multiple electronic health record (EHR) data sources of 91,166 multi-ancestry participants with a Natural Language Processing (NLP) technique. MATERIALS AND METHODS: We developed a NLP-enriched phenotyping algorithm that incorporated colonoscopy or abdominal imaging reports to identify patients with diverticulosis and diverticulitis from multicenter EHRs. We performed genome-wide association studies (GWAS) of DD in European, African and multi-ancestry participants, followed by phenome-wide association studies (PheWAS) of the risk variants to identify their potential comorbid/pleiotropic effects in clinical phenotypes. RESULTS: Our developed algorithm showed a significant improvement in patient classification performance for DD analysis (algorithm PPVs ≥ 0.94), with up to a 3.5 fold increase in terms of the number of identified patients than the traditional method. Ancestry-stratified analyses of diverticulosis and diverticulitis of the identified subjects replicated the well-established associations between ARHGAP15 loci with DD, showing overall intensified GWAS signals in diverticulitis patients compared to diverticulosis patients. Our PheWAS analyses identified significant associations between the DD GWAS variants and circulatory system, genitourinary, and neoplastic EHR phenotypes. DISCUSSION: As the first multi-ancestry GWAS-PheWAS study, we showcased that heterogenous EHR data can be mapped through an integrative analytical pipeline and reveal significant genotype-phenotype associations with clinical interpretation. CONCLUSION: A systematic framework to process unstructured EHR data with NLP could advance a deep and scalable phenotyping for better patient identification and facilitate etiological investigation of a disease with multilayered data.

An exploratory survey about using ChatGPT in education, healthcare, and research.

Objective ChatGPT is the first large language model (LLM) to reach a large, mainstream audience. Its rapid adoption and exploration by the population at large has sparked a wide range of discussions regarding its acceptable and optimal integration in different areas. In a hybrid (virtual and in-person) panel discussion event, we examined various perspectives regarding the use of ChatGPT in education, research, and healthcare. Materials and Methods We surveyed in-person and online attendees using an audience interaction platform (Slido). We quantitatively analyzed received responses on questions about the use of ChatGPT in various contexts. We compared pairwise categorical groups with Fisher's Exact. Furthermore, we used qualitative methods to analyze and code discussions. Results We received 420 responses from an estimated 844 participants (response rate 49.7%). Only 40% of the audience had tried ChatGPT. More trainees had tried ChatGPT compared with faculty. Those who had used ChatGPT were more interested in using it in a wider range of contexts going forwards. Of the three discussed contexts, the greatest uncertainty was shown about using ChatGPT in education. Pros and cons were raised during discussion for the use of this technology in education, research, and healthcare. Discussion There was a range of perspectives around the uses of ChatGPT in education, research, and healthcare, with still much uncertainty around its acceptability and optimal uses. There were different perspectives from respondents of different roles (trainee vs faculty vs staff). More discussion is needed to explore perceptions around the use of LLMs such as ChatGPT in vital sectors such as education, healthcare and research. Given involved risks and unforeseen challenges, taking a thoughtful and measured approach in adoption would reduce the likelihood of harm.

Is more better? The impact of implementing more interventions for hypertension control in a practice facilitation study for small- and medium-sized practices.

Interventions for blood pressure (BP) control have positive effects on outcomes for patients with hypertension. Research on these effects in small- and medium-sized practices is limited. Our retrospective analysis used data from Healthy Hearts in the Heartland (H3), a research program conducted in 2016-2018 as part of the Agency for Healthcare Research and Quality's EvidenceNOW initiative, to examine the impact of implementing more interventions for BP control in these settings. Thirty-eight H3 practices met inclusion criteria and were assigned to an implementer group (high or low) based on the number of interventions implemented with the support of a practice facilitator during the study. Practices in the high-implementer group implemented a mean of 2.2 additional interventions relative to the low-implementer group. Groups were compared on two measures of BP control: (1) mean percentage of hypertensive patients with a most recent BP below 140/90, and (2) mean systolic and diastolic BP of hypertensive patients. In the first measure, practices in the high-implementer group had greater improvement between baseline and the end of the study. Among the 10,150 patients included in the second measure, reductions in mean SBP and DBP were greater for the high-implementer group. These outcomes show that implementing additional interventions had a positive association with measures of BP control, though clinical significance was unknown or limited. Future research is needed to understand the impact of interventions for BP control in small- and medium-sized practices, including the interactions among intervention implementation, practice facilitation, and practice and patient characteristics.

Comparison of State-of-the-Art Neural Network Survival Models with the Pooled Cohort Equations for Cardiovascular Disease Risk Prediction.

BACKGROUND: The Pooled Cohort Equations (PCEs) are race- and sex-specific Cox proportional hazards (PH)-based models used for 10-year atherosclerotic cardiovascular disease (ASCVD) risk prediction with acceptable discrimination. In recent years, neural network models have gained increasing popularity with their success in image recognition and text classification. Various survival neural network models have been proposed by combining survival analysis and neural network architecture to take advantage of the strengths from both. However, the performance of these survival neural network models compared to each other and to PCEs in ASCVD prediction is unknown. METHODS: In this study, we used 6 cohorts from the Lifetime Risk Pooling Project (with 5 cohorts as training/internal validation and one cohort as external validation) and compared the performance of the PCEs in 10-year ASCVD risk prediction with an all two-way interactions Cox PH model (Cox PH-TWI) and three state-of-the-art neural network survival models including Nnet-survival, Deepsurv, and Cox-nnet. For all the models, we used the same 7 covariates as used in the PCEs. We fitted each of the aforementioned models in white females, white males, black females, and black males, respectively. We evaluated models' internal and external discrimination power and calibration. RESULTS: The training/internal validation sample comprised 23216 individuals. The average age at baseline was 57.8 years old (SD = 9.6); 16% developed ASCVD during average follow-up of 10.50 (SD = 3.02) years. Based on 10 × 10 cross-validation, the method that had the highest C-statistics was Deepsurv (0.7371) for white males, Deepsurv and Cox PH-TWI (0.7972) for white females, PCE (0.6981) for black males, and Deepsurv (0.7886) for black females. In the external validation dataset, Deepsurv (0.7032), Cox-nnet (0.7282), PCE (0.6811), and Deepsurv (0.7316) had the highest C-statistics for white male, white female, black male, and black female population, respectively. Calibration plots showed that in 10 × 10 validation, all models had good calibration in all race and sex groups. In external validation, all models overestimated the risk for 10-year ASCVD. CONCLUSIONS: We demonstrated the use of the state-of-the-art neural network survival models in ASCVD risk prediction. Neural network survival models had similar if not superior discrimination and calibration compared to PCEs.

Supermarket Proximity and Risk of Hypertension, Diabetes, and CKD: A Retrospective Cohort Study.

RATIONALE & OBJECTIVE: Living in environments with low access to food may increase the risk of chronic diseases. We investigated the association of household distance to the nearest supermarket (as a measure of food access) with the incidence of hypertension, diabetes, and chronic kidney disease (CKD) in a metropolitan area of the United States. STUDY DESIGN: Retrospective cohort study. SETTING & PARTICIPANTS: 777,994 individuals without hypertension, diabetes, or CKD at baseline within the HealthLNK Data Repository, which contains electronic health records from 7 health care institutions in Chicago, Illinois. EXPOSURE: Zip code-level average distance between households and nearest supermarket. OUTCOME: Incidence of hypertension, diabetes, and CKD based on presence of ICD-9 code and/or blood pressure≥140/90mm Hg, hemoglobin A1c≥6.5%, and eGFR<60mL/min/1.73m2, respectively. ANALYTICAL APPROACH: Average distance to nearest supermarket was aggregated from street-level metrics for 56 Chicagoland zip codes. The cumulative incidence of hypertension, diabetes, and CKD from 2007-2012 was calculated for each zip code in patients free of these diseases in 2006. Spatial analysis of food access and disease incidence was performed using bivariate local indicator of spatial association (BiLISA) maps and bivariate local Moran I statistics. The relationship between supermarket access and outcomes was analyzed using logistic regression. RESULTS: Of 777,994 participants, 408,608 developed hypertension, 51,380 developed diabetes, and 56,365 developed CKD. There was significant spatial overlap between average distance to supermarket and incidence of hypertension and diabetes but not CKD. Zip codes with large average supermarket distances and high incidence of hypertension and diabetes were clustered in southern and western neighborhoods. Models adjusted only for neighborhood factors (zip code-level racial composition, access to vehicles, median income) revealed significant associations between zip code-level average distance to supermarket and chronic disease incidence. Relative to tertile 1 (shortest distance), ORs in tertiles 2 and 3, respectively, were 1.27 (95% CI, 1.23-1.30) and 1.38 (95% CI, 1.33-1.43) for diabetes, 1.03 (95% CI, 1.02-1.05) and 1.04 (95% CI, 1.02-1.06) for hypertension, and 1.18 (95% CI, 1.15-1.21) and 1.33 (95% CI, 1.29-1.37) for CKD. Models adjusted for demographic factors and health insurance showed significant and positive association with greater odds of incident diabetes (tertile 2: 1.29 [95% CI, 1.26-1.33]; tertile 3: 1.35 [95% CI, 1.31-1.39]) but lesser odds of hypertension (tertile 2: 0.95 [95% CI, 0.94-0.97]; tertile 3: 0.91 [95% CI, 0.89-0.92]) and CKD (tertile 2: 0.80 [95% CI, 0.78-0.82]; tertile 3: 0.73 [95% CI, 0.72-0.76]). After adjusting for both neighborhood and individual covariates, supermarket distance remained significantly associated with greater odds of diabetes and lesser odds of hypertension, but there was no significant association with CKD. LIMITATIONS: Unmeasured neighborhood and social confounding variables, zip code-level analysis, and limited individual-level information. CONCLUSIONS: There are significant disparities in supermarket proximity and incidence of hypertension, diabetes, and CKD in Chicago, Illinois. The relationship between supermarket access and chronic disease is largely explained by individual- and neighborhood-level factors.

Comparison of Sodium-Glucose Cotransporter-2 Inhibitor and Glucagon-Like Peptide-1 Receptor Agonist Prescribing in Patients With Diabetes Mellitus With and Without Cardiovascular Disease.

Sodium-glucose cotransporter-2 inhibitors (SGLT2is) and glucagon-like peptide-1 receptor agonists (GLP1-RAs) reduce cardiovascular events and mortality in patients with type 2 diabetes mellitus (T2DM). We sought to describe trends in prescribing for SGLT2is and GLP1-RAs in diverse care settings, including (1) the outpatient clinics of a midwestern integrated health system and (2) small- and medium-sized community-based primary care practices and health centers in 3 midwestern states. We included adults with T2DM and ≥1 outpatient clinic visit. The outcomes of interest were annual active prescription rates for SGLT2is and GLP1-RAs (separately). In the integrated health system, 22,672 patients met the case definition of T2DM. From 2013 to 2019, the overall prescription rate for SGLT2is increased from 1% to 15% (absolute difference [AD] 14%, 95% confidence interval [CI] 13% to 15%, p <0.01). The GLP1-RA prescription rate was stable at 10% (AD 0%, 95% CI -1% to 1%, p = 0.9). In community-based primary care practices, 43,340 patients met the case definition of T2DM. From 2013 to 2017, the SGLT2i prescription rate increased from 3% to 7% (AD 4%, 95% CI 3% to 6%, p <0.01), whereas the GLP1-RA prescription rate was stable at 2% to 3% (AD 1%, 95% CI -1 to 1%, p = 0.40). In a fully adjusted regression model, non-Hispanic Black patients had lower odds of SGLT2i or GLP1-RA prescription (odds ratio 0.56, 95% CI 0.34 to 0.89, p = 0.016). In conclusion, the increase in prescription rates was greater for SGLT2is than for GLP1-RAs in patients with T2DM in a large integrated medical center and community primary care practices. Overall, prescription rates for eligible patients were low, and racial disparities were observed.

Characterizing variability of electronic health record-driven phenotype definitions.

OBJECTIVE: The aim of this study was to analyze a publicly available sample of rule-based phenotype definitions to characterize and evaluate the variability of logical constructs used. MATERIALS AND METHODS: A sample of 33 preexisting phenotype definitions used in research that are represented using Fast Healthcare Interoperability Resources and Clinical Quality Language (CQL) was analyzed using automated analysis of the computable representation of the CQL libraries. RESULTS: Most of the phenotype definitions include narrative descriptions and flowcharts, while few provide pseudocode or executable artifacts. Most use 4 or fewer medical terminologies. The number of codes used ranges from 5 to 6865, and value sets from 1 to 19. We found that the most common expressions used were literal, data, and logical expressions. Aggregate and arithmetic expressions are the least common. Expression depth ranges from 4 to 27. DISCUSSION: Despite the range of conditions, we found that all of the phenotype definitions consisted of logical criteria, representing both clinical and operational logic, and tabular data, consisting of codes from standard terminologies and keywords for natural language processing. The total number and variety of expressions are low, which may be to simplify implementation, or authors may limit complexity due to data availability constraints. CONCLUSIONS: The phenotype definitions analyzed show significant variation in specific logical, arithmetic, and other operators but are all composed of the same high-level components, namely tabular data and logical expressions. A standard representation for phenotype definitions should support these formats and be modular to support localization and shared logic.