Justin Starren's research while affiliated with University of Illinois at Chicago and other places

Introduction The rapid development of artificial intelligence (AI) in healthcare has exposed the unmet need for growing a multidisciplinary workforce that can collaborate effectively in the learning health systems. Maximizing the synergy among multiple teams is critical for Collaborative AI in Healthcare. Methods We have developed a series of data, tools, and educational resources for cultivating the next generation of multidisciplinary workforce for Collaborative AI in Healthcare. We built bulk‐natural language processing pipelines to extract structured information from clinical notes and stored them in common data models. We developed multimodal AI/machine learning (ML) tools and tutorials to enrich the toolbox of the multidisciplinary workforce to analyze multimodal healthcare data. We have created a fertile ground to cross‐pollinate clinicians and AI scientists and train the next generation of AI health workforce to collaborate effectively. Results Our work has democratized access to unstructured health information, AI/ML tools and resources for healthcare, and collaborative education resources. From 2017 to 2022, this has enabled studies in multiple clinical specialties resulting in 68 peer‐reviewed publications. In 2022, our cross‐discipline efforts converged and institutionalized into the Center for Collaborative AI in Healthcare. Conclusions Our Collaborative AI in Healthcare initiatives has created valuable educational and practical resources. They have enabled more clinicians, scientists, and hospital administrators to successfully apply AI methods in their daily research and practice, develop closer collaborations, and advanced the institution‐level learning health system.

Objective Data from DNA genotyping via a 96-SNP panel in a study of 25,015 clinical samples were utilized for quality control and tracking of sample identity in a clinical sequencing network. The study aimed to demonstrate the value of both the precise SNP tracking and the utility of the panel for predicting the sex-by-genotype of the participants, to identify possible sample mix-ups. Results Precise SNP tracking showed no sample swap errors within the clinical testing laboratories. In contrast, when comparing predicted sex-by-genotype to the provided sex on the test requisition, we identified 110 inconsistencies from 25,015 clinical samples (0.44%), that had occurred during sample collection or accessioning. The genetic sex predictions were confirmed using additional SNP sites in the sequencing data or high-density genotyping arrays. It was determined that discrepancies resulted from clerical errors (49.09%), samples from transgender participants (3.64%) and stem cell or bone marrow transplant patients (7.27%) along with undetermined sample mix-ups (40%) for which sample swaps occurred prior to arrival at genome centers, however the exact cause of the events at the sampling sites resulting in the mix-ups were not able to be determined.

We investigate risk factors for severe COVID-19 in persons living with HIV (PWH), including among racialized PWH, using the U.S. population-sampled National COVID Cohort Collaborative (N3C) data released from January 1, 2020 to October 10, 2022. We defined severe COVID-19 as hospitalized with invasive mechanical ventilation, extracorporeal membrane oxygenation, discharge to hospice or death. We used machine learning methods to identify highly ranked, uncorrelated factors predicting severe COVID-19, and used multivariable logistic regression models to assess the associations of these variables with severe COVID-19 in several models, including race-stratified models. There were 3 241 627 individuals with incident COVID-19 cases and 81 549 (2.5%) with severe COVID-19, of which 17 445 incident COVID-19 and 1 020 (5.8%) severe cases were among PWH. The top highly ranked factors of severe COVID-19 were age, congestive heart failure (CHF), dementia, renal disease, sodium concentration, smoking status, and sex. Among PWH, age and sodium concentration were important predictors of COVID-19 severity, and the effect of sodium concentration was more pronounced in Hispanics (aOR 4.11 compared to aOR range: 1.47–1.88 for Black, White, and Other non-Hispanics). Dementia, CHF, and renal disease was associated with higher odds of severe COVID-19 among Black, Hispanic, and Other non-Hispanics PWH, respectively. Our findings suggest that the impact of factors, especially clinical comorbidities, predictive of severe COVID-19 among PWH varies by racialized groups, highlighting a need to account for race and comorbidity burden when assessing the risk of PWH developing severe COVID-19.

Background Patient-reported outcome (PRO) measures have become an essential component of quality measurement, quality improvement, and capturing the voice of the patient in clinical care. In 2004, the National Institutes of Health endorsed the importance of PROs by initiating the Patient-Reported Outcomes Measurement Information System (PROMIS), which leverages computer-adaptive tests (CATs) to reduce patient burden while maintaining measurement precision. Historically, PROMIS CATs have been used in a large number of research studies outside the electronic health record (EHR), but growing demand for clinical use of PROs requires creative information technology solutions for integration into the EHR. Objectives This paper describes the introduction of PROMIS CATs into the Epic Systems EHR at a large academic medical center using a tight integration; we describe the process of creating a secure, automatic connection between the application programming interface (API) which scores and selects CAT items and Epic. Methods The overarching strategy was to make CATs appear indistinguishable from conventional measures to clinical users, patients, and the EHR software itself. We implemented CATs in Epic without compromising patient data security by creating custom middleware software within the organization's existing middleware framework. This software communicated between the Assessment Center API for item selection and scoring and Epic for item presentation and results. The middleware software seamlessly administered CATs alongside fixed-length, conventional PROs while maintaining the display characteristics and functions of other Epic measures, including automatic display of PROMIS scores in the patient's chart. Pilot implementation revealed differing workflows for clinicians using the software. Results The middleware software was adopted in 27 clinics across the hospital system. In the first 2 years of hospital-wide implementation, 793 providers collected 70,446 PROs from patients using this system. Conclusion This project demonstrated the importance of regular communication across interdisciplinary teams in the design and development of clinical software. It also demonstrated that implementation relies on buy-in from clinical partners as they integrate new tools into their existing clinical workflow.

Introduction Integrating social and environmental determinants of health (SEDoH) into enterprise-wide clinical workflows and decision-making is one of the most important and challenging aspects of improving health equity. We engaged domain experts to develop a SEDoH informatics maturity model (SIMM) to help guide organizations to address technical, operational, and policy gaps. Methods We established a core expert group consisting of developers, informaticists, and subject matter experts to identify different SIMM domains and define maturity levels. The candidate model (v0.9) was evaluated by 15 informaticists at a Center for Data to Health community meeting. After incorporating feedback, a second evaluation round for v1.0 collected feedback and self-assessments from 35 respondents from the National COVID Cohort Collaborative, the Center for Leading Innovation and Collaboration’s Informatics Enterprise Committee, and a publicly available online self-assessment tool. Results We developed a SIMM comprising seven maturity levels across five domains: data collection policies, data collection methods and technologies, technology platforms for analysis and visualization, analytics capacity, and operational and strategic impact. The evaluation demonstrated relatively high maturity in analytics and technological capacity, but more moderate maturity in operational and strategic impact among academic medical centers. Changes made to the tool in between rounds improved its ability to discriminate between intermediate maturity levels. Conclusion The SIMM can help organizations identify current gaps and next steps in improving SEDoH informatics. Improving the collection and use of SEDoH data is one important component of addressing health inequities.

Clinical data sharing combined with deep learning, and soon quantum computing, has the potential to radically accelerate research, improve healthcare, and lower costs. Unfortunately, those tools also make it much easier to use the data in ways that can harm patients. This article will argue that the vast amounts of data collected by data brokers, combined with advances in computing, have made reidentification a serious risk for any clinical data that is shared openly. The new NIH data sharing policy acknowledges this new reality by directing researchers to consider controlled access for any individual-level data. The clinical data sharing community will be well-advised to follow the lead of the physics and astronomy communities and create a “walled garden” approach to data sharing. While the investment will be significant, this approach provides the optimal combination of both access and privacy.

Objective The effects of Coronavirus disease 2019 (COVID-19) infection and altered processes of care on nonelective coronary artery bypass grafting (CABG) outcomes remain unknown. We hypothesized that patients with COVID-19 infection would have longer hospital lengths of stay and greater mortality compared with COVID-negative patients, but that these outcomes would not differ between COVID-negative and pre-COVID controls. Methods The National COVID Cohort Collaborative 2020-2022 was queried for adult patients undergoing CABG. Patients were divided into COVID-negative, COVID-active, and COVID-convalescent groups. Pre-COVID control patients were drawn from the National Surgical Quality Improvement Program database. Adjusted analysis of the 3 COVID groups was performed via generalized linear models. Results A total of 17,293 patients underwent nonelective CABG, including 16,252 COVID-negative, 127 COVID-active, 367 COVID-convalescent, and 2254 pre-COVID patients. Compared to pre-COVID patients, COVID-negative patients had no difference in mortality, whereas COVID-active patients experienced increased mortality. Mortality and pneumonia were higher in COVID-active patients compared to COVID-negative and COVID-convalescent patients. Adjusted analysis demonstrated that COVID-active patients had higher in-hospital mortality, 30- and 90-day mortality, and pneumonia compared to COVID-negative patients. COVID-convalescent patients had a shorter length of stay but a higher rate of renal impairment. Conclusions Traditional care processes were altered during the COVID-19 pandemic. Our data show that nonelective CABG in patients with active COVID-19 is associated with significantly increased rates of mortality and pneumonia. The equivalent mortality in COVID-negative and pre-COVID patients suggests that pandemic-associated changes in processes of care did not impact CABG outcomes. Additional research into optimal timing of CABG after COVID infection is warranted.

Objectives To define pregnancy episodes and estimate gestational age within electronic health record (EHR) data from the National COVID Cohort Collaborative (N3C). Materials and Methods We developed a comprehensive approach, named Hierarchy and rule-based pregnancy episode Inference integrated with Pregnancy Progression Signatures (HIPPS), and applied it to EHR data in the N3C (January 1, 2018–April 7, 2022). HIPPS combines: (1) an extension of a previously published pregnancy episode algorithm, (2) a novel algorithm to detect gestational age-specific signatures of a progressing pregnancy for further episode support, and (3) pregnancy start date inference. Clinicians performed validation of HIPPS on a subset of episodes. We then generated pregnancy cohorts based on gestational age precision and pregnancy outcomes for assessment of accuracy and comparison of COVID-19 and other characteristics. Results We identified 628 165 pregnant persons with 816 471 pregnancy episodes, of which 52.3% were live births, 24.4% were other outcomes (stillbirth, ectopic pregnancy, abortions), and 23.3% had unknown outcomes. Clinician validation agreed 98.8% with HIPPS-identified episodes. We were able to estimate start dates within 1 week of precision for 475 433 (58.2%) episodes. 62 540 (7.7%) episodes had incident COVID-19 during pregnancy. Discussion HIPPS provides measures of support for pregnancy-related variables such as gestational age and pregnancy outcomes based on N3C data. Gestational age precision allows researchers to find time to events with reasonable confidence. Conclusion We have developed a novel and robust approach for inferring pregnancy episodes and gestational age that addresses data inconsistency and missingness in EHR data.