Telehealth Implementation Response to COVID-19 in the OneFlorida+ Clinical Research Network: Perspectives of Clinicians and Health Systems Leaders.

Introduction: The COVID-19 pandemic forced health systems worldwide to make rapid adjustments to patient care. Nationwide stay-at-home mandates and public health concerns increased demand for telehealth to maintain patients' continuity of care. These circumstances permitted observation of telehealth implementation in real-world settings at a large scale. This study aimed to understand clinician and health system leader (HSL) experiences in expanding, implementing, and sustaining telehealth during COVID-19 in the OneFlorida+ clinical research network. Methods: We conducted semistructured videoconference interviews with 5 primary care providers, 7 specialist providers, and 12 HSLs across 7 OneFlorida+ health systems and settings. Interviews were audiorecorded, transcribed, and summarized using deductive team-based template coding. We then used matrix analysis to organize the qualitative data and identify inductive themes. Results: Rapid telehealth implementation occurred even among sites with low readiness, facilitated by responsive planning, shifts in resource allocation, and training. Common hurdles in routine telehealth use, including technical and reimbursement issues, were also barriers to telehealth implementation. Acceptability of telehealth was influenced by benefits such as the providers' ability to view a patient's home environment and the availability of tools to enhance patient education. Lower acceptability stemmed from the inability to conduct physical examinations during the shutdown. Conclusions: This study identified a broad range of barriers, facilitators, and strategies for implementing telehealth within large clinical research networks. The findings can contribute to optimizing the effectiveness of telehealth implementation in similar settings, and point toward promising directions for telehealth provider training to improve acceptability and promote sustainability.

A Subjective Culture Approach to Cancer Prevention: Rural Black and White Adults' Perceptions of Using Virtual Health Assistants to Promote Colorectal Cancer Screening.

In the US, Black adults are less likely than White adults to be screened for colorectal cancer (CRC). This study uses a subjective culture approach to describe and compare perceptions of a CRC screening intervention delivered via virtual health assistants (VHAs) among rural Black and White study participants. We analyzed 28 focus groups with Black (n = 85) and White (n = 69) adults aged 50-73. Participants, largely recruited through community engagement efforts, tested the VHA intervention on mobile phones provided by the research team. Moderated discussions were recorded, transcribed, and analyzed using thematic analysis. All groups preferred the VHA to be friendly. Other important cues included trustworthiness, authority, and expertise. Black participants expressed a preference for receiving information about their CRC risk from the VHA compared with White adults. Black participants also expressed the importance of sharing the intervention and the CRC screening messages with younger members of their networks, including family members who could benefit from screening messages before reaching the recommended age for screening. The key similarities and differences between Black and White adults' perceptions of the intervention that were identified in this study can help inform future efforts to develop effective communication strategies and reduce cancer screening inequities.

Impact of the Early Phase of the COVID-19 Pandemic on US Healthcare Workers: Results from the HERO Registry.

BACKGROUND: The HERO registry was established to support research on the impact of the COVID-19 pandemic on US healthcare workers. OBJECTIVE: Describe the COVID-19 pandemic experiences of and effects on individuals participating in the HERO registry. DESIGN: Cross-sectional, self-administered registry enrollment survey conducted from April 10 to July 31, 2020. SETTING: Participants worked in hospitals (74.4%), outpatient clinics (7.4%), and other settings (18.2%) located throughout the nation. PARTICIPANTS: A total of 14,600 healthcare workers. MAIN MEASURES: COVID-19 exposure, viral and antibody testing, diagnosis of COVID-19, job burnout, and physical and emotional distress. KEY RESULTS: Mean age was 42.0 years, 76.4% were female, 78.9% were White, 33.2% were nurses, 18.4% were physicians, and 30.3% worked in settings at high risk for COVID-19 exposure (e.g., ICUs, EDs, COVID-19 units). Overall, 43.7% reported a COVID-19 exposure and 91.3% were exposed at work. Just 3.8% in both high- and low-risk settings experienced COVID-19 illness. In regression analyses controlling for demographics, professional role, and work setting, the risk of COVID-19 illness was higher for Black/African-Americans (aOR 2.32, 99% CI 1.45, 3.70, p < 0.01) and Hispanic/Latinos (aOR 2.19, 99% CI 1.55, 3.08, p < 0.01) compared with Whites. Overall, 41% responded that they were experiencing job burnout. Responding about the day before they completed the survey, 53% of participants reported feeling tired a lot of the day, 51% stress, 41% trouble sleeping, 38% worry, 21% sadness, 19% physical pain, and 15% anger. On average, healthcare workers reported experiencing 2.4 of these 7 distress feelings a lot of the day. CONCLUSIONS: Healthcare workers are at high risk for COVID-19 exposure, but rates of COVID-19 illness were low. The greater risk of COVID-19 infection among race/ethnicity minorities reported in the general population is also seen in healthcare workers. The HERO registry will continue to monitor changes in healthcare worker well-being during the pandemic. TRIAL REGISTRATION: ClinicalTrials.gov identifier NCT04342806.

Key changes to improve social presence of a virtual health assistant promoting colorectal cancer screening informed by a technology acceptance model.

BACKGROUND: Understanding how older, minoritized patients attend to cues when interacting with web-based health messages may provide opportunities to improve engagement with novel health technologies. We assess acceptance-promoting and acceptance-inhibiting cues of a web-based, intervention promoting colorectal cancer (CRC) screening with a home stool test among Black women. MATERIALS AND METHODS: Focus group and individual interview data informed iterative changes to a race- and gender-concordant virtual health assistant (VHA). A user-centered design approach was used across 3 iterations to identify changes needed to activate cues described as important; such as portraying authority and expertise. Questionnaire data were analyzed using non-parametric tests for perceptions of cues. Analysis was guided by the Technology Acceptance Model. RESULTS: Perceptions of interactivity, social presence, expertise, and trust were important cues in a VHA-delivered intervention promoting CRC screening. Features of the web-based platform related to ease of navigation and use were also discussed. Participant comments varied across the 3 iterations and indicated acceptance of or a desire to improve source cues for subsequent iterations. We highlight the specific key changes made at each of three iterative versions of the interactive intervention in conjunction with user perception of changes. DISCUSSION: Virtual agents can be adapted to better meet patient expectations such as being a trustworthy and expert source. Across three evolving versions of a Black, VHA, cues for social presence were particularly important. Social presence cues helped patients engage with CRC screening messages delivered in this novel digital context. CONCLUSIONS: When using a VHA to disseminate health information, cues associated with acceptability can be leveraged and adapted as needed for diverse audiences. Patient characteristics (age, identity, health status) are important to note as they may affect perceptions of a novel health technologies ease of use and relevancy according to the leading models.

Tailoring virtual human-delivered interventions: A digital intervention promoting colorectal cancer screening for Black women.

OBJECTIVE: Despite efforts to reduce cancer disparities, Black women remain underrepresented in cancer research. Virtual health assistants (VHAs) are one promising digital technology for communicating health messages and promoting health behaviors to diverse populations. This study describes participant responses to a VHA-delivered intervention promoting colorectal cancer (CRC) screening with a home-stool test. METHODS: We recruited 53 non-Hispanic Black women 50 to 73 years old to participate in focus groups and think-aloud interviews and test a web-based intervention delivered by a race- and gender-concordant VHA. A user-centered design approach prioritized modifications to three successive versions of the intervention based on participants' comments. RESULTS: Participants identified 26 cues relating to components of the VHA's credibility, including trustworthiness, expertise, and authority. Comments on early versions revealed preferences for communicating with a human doctor and negative critiques of the VHA's appearance and movements. Modifications to specific cues improved the user experience, and participants expressed increased willingness to engage with later versions of the VHA and the screening messages it delivered. Informed by the Modality, Agency, Interactivity, Navigability Model, we present a framework for developing credible VHA-delivered cancer screening messages. CONCLUSIONS: VHAs provide a systematic way to deliver health information. A culturally sensitive intervention designed for credibility promoted user interest in engaging with guideline-concordant CRC screening messages. We present strategies for effectively using cues to engage audiences with health messages, which can be applied to future research in varying contexts.

Prodromal clinical, demographic, and socio-ecological correlates of asthma in adults: a 10-year statewide big data multi-domain analysis.

Objectives: To identify prodromal correlates of asthma as compared to chronic obstructive pulmonary disease and allied-conditions (COPDAC) using a multi domain analysis of socio-ecological, clinical, and demographic domains.Methods: This is a retrospective case-risk-control study using data from Florida's statewide Healthcare Cost and Utilization Project (HCUP). Patients were grouped into three groups: asthma, COPDAC (without asthma), and neither asthma nor COPDAC. To identify socio-ecological, clinical, demographic, and clinical predictors of asthma and COPDAC, we used univariate analysis, feature ranking by bootstrapped information gain ratio, multivariable logistic regression with LogitBoost selection, decision trees, and random forests.Results: A total of 141,729 patients met inclusion criteria, of whom 56,052 were diagnosed with asthma, 85,677 with COPDAC, and 84,737 with neither asthma nor COPDAC. The multi-domain approach proved superior in distinguishing asthma versus COPDAC and non-asthma/non-COPDAC controls (area under the curve (AUROC) 84%). The best domain to distinguish asthma from COPDAC without controls was prior clinical diagnoses (AUROC 82%). Ranking variables from all the domains found the most important predictors for the asthma versus COPDAC and controls were primarily socio-ecological variables, while for asthma versus COPDAC without controls, demographic and clinical variables such as age, CCI, and prior clinical diagnoses, scored better.Conclusions: In this large statewide study using a machine learning approach, we found that a multi-domain approach with demographics, clinical, and socio-ecological variables best predicted an asthma diagnosis. Future work should focus on integrating machine learning-generated predictive models into clinical practice to improve early detection of those common respiratory diseases.

Optimizing identification of resistant hypertension: Computable phenotype development and validation.

PURPOSE: Computable phenotypes are constructed to utilize data within the electronic health record (EHR) to identify patients with specific characteristics; a necessary step for researching a complex disease state. We developed computable phenotypes for resistant hypertension (RHTN) and stable controlled hypertension (HTN) based on the National Patient-Centered Clinical Research Network (PCORnet) common data model (CDM). The computable phenotypes were validated through manual chart review. METHODS: We adapted and refined existing computable phenotype algorithms for RHTN and stable controlled HTN to the PCORnet CDM in an adult HTN population from the OneFlorida Clinical Research Consortium (2015-2017). Two independent reviewers validated the computable phenotypes through manual chart review of 425 patient records. We assessed precision of our computable phenotypes through positive predictive value (PPV) and test validity through interrater reliability (IRR). RESULTS: Among the 156 730 HTN patients in our final dataset, the final computable phenotype algorithms identified 24 926 patients with RHTN and 19 100 with stable controlled HTN. The PPV for RHTN in patients randomly selected for validation of the final algorithm was 99.1% (n = 113, CI: 95.2%-99.9%). The PPV for stable controlled HTN in patients randomly selected for validation of the final algorithm was 96.5% (n = 113, CI: 91.2%-99.0%). IRR analysis revealed a raw percent agreement of 91% (152/167) with Cohen's kappa statistic = 0.87. CONCLUSIONS: We constructed and validated a RHTN computable phenotype algorithm and a stable controlled HTN computable phenotype algorithm. Both algorithms are based on the PCORnet CDM, allowing for future application to epidemiological and drug utilization based research.

Geospatial Analyses of Pain Intensity and Opioid Unit Doses Prescribed on the Day of Discharge Following Orthopedic Surgery.

OBJECTIVE: Inappropriate opioid prescribing after surgery contributes to opioid use disorder and risk of opioid overdose. In this cross-sectional analysis of orthopedic surgical patients, we examined the role of patient location on postoperative pain intensity and opioids prescribed on hospital discharge. METHODS: We used geospatial analyses to characterize spatial patterns of mean pain intensity on the day of discharge (PiDoD) and opioid units prescribed on the day of discharge (OuPoD), as well as the effect of regional social deprivation on these outcomes. RESULTS: At a 500-km radius from the surgery site, the Global Moran's I for PiDoD (2.71 × 10-3, variance = 1.67 × 10-6, P = 0.012) and OuPoD (2.19 × 10-3, SD = 1.87, variance = 1.66 × 10-6, P = 0.03) suggested significant spatial autocorrelation within each outcome. Local indicators of spatial autocorrelation, including local Moran's I, Local Indicator of Spatial Autocorrelation cluster maps, and Getis-Ord Gi* statistics, further demonstrated significant, specific regions of clustering both OuPoD and PiDoD. These spatial patterns were associated with spatial regions of area deprivation. CONCLUSIONS: Our results suggest that the outcomes of pain intensity and opioid doses prescribed exhibit varying degrees of clustering of patient locations of residence, at both global and local levels. This indicates that a given patient's pain intensity on discharge is related to the pain intensity of nearby individuals. Similar interpretations exist for OuPoD, although the relative locations of hot spots of opioids dispensed in a geographic area appear to differ from those of hot spots of pain intensity on discharge.

Facilitation or Hindrance: Physicians' Perception on Best Practice Alerts (BPA) Usage in an Electronic Health Record System.

A qualitative study was conducted to examine physicians' perception on best practice alerts (BPAs) usage under the theoretical framework of technology acceptance model (TAM). In particular, 20 face-to-face in-depth interviews were conducted from September 2016 to February 2017 to collect data. Four themes emerged from the current set of data: support, de-escalation, management, and enhancement are physicians' perceived motivations of using BPAs; interactivity, timing, and interface design are key factors that influence physicians' perceived experiences of using BPAs; alert fatigue and inaccurate alters are major challenges and issues faced by physicians when using BPAs; and, control and team approach are physicians' perceptions on the future improvement of BPAs. This study not only offers a detailed description of physicians' perception of BPAs, revealing the rich meanings associated with this phenomenon, but also enriches our understanding of the TAM in the context of BPAs by uncovering the key dimensions of abstract constructs in the model.

Big data hurdles in precision medicine and precision public health.

BACKGROUND: Nowadays, trendy research in biomedical sciences juxtaposes the term 'precision' to medicine and public health with companion words like big data, data science, and deep learning. Technological advancements permit the collection and merging of large heterogeneous datasets from different sources, from genome sequences to social media posts or from electronic health records to wearables. Additionally, complex algorithms supported by high-performance computing allow one to transform these large datasets into knowledge. Despite such progress, many barriers still exist against achieving precision medicine and precision public health interventions for the benefit of the individual and the population. MAIN BODY: The present work focuses on analyzing both the technical and societal hurdles related to the development of prediction models of health risks, diagnoses and outcomes from integrated biomedical databases. Methodological challenges that need to be addressed include improving semantics of study designs: medical record data are inherently biased, and even the most advanced deep learning's denoising autoencoders cannot overcome the bias if not handled a priori by design. Societal challenges to face include evaluation of ethically actionable risk factors at the individual and population level; for instance, usage of gender, race, or ethnicity as risk modifiers, not as biological variables, could be replaced by modifiable environmental proxies such as lifestyle and dietary habits, household income, or access to educational resources. CONCLUSIONS: Data science for precision medicine and public health warrants an informatics-oriented formalization of the study design and interoperability throughout all levels of the knowledge inference process, from the research semantics, to model development, and ultimately to implementation.

OC-2-KB: integrating crowdsourcing into an obesity and cancer knowledge base curation system.

BACKGROUND: There is strong scientific evidence linking obesity and overweight to the risk of various cancers and to cancer survivorship. Nevertheless, the existing online information about the relationship between obesity and cancer is poorly organized, not evidenced-based, of poor quality, and confusing to health information consumers. A formal knowledge representation such as a Semantic Web knowledge base (KB) can help better organize and deliver quality health information. We previously presented the OC-2-KB (Obesity and Cancer to Knowledge Base), a software pipeline that can automatically build an obesity and cancer KB from scientific literature. In this work, we investigated crowdsourcing strategies to increase the number of ground truth annotations and improve the quality of the KB. METHODS: We developed a new release of the OC-2-KB system addressing key challenges in automatic KB construction. OC-2-KB automatically extracts semantic triples in the form of subject-predicate-object expressions from PubMed abstracts related to the obesity and cancer literature. The accuracy of the facts extracted from scientific literature heavily relies on both the quantity and quality of the available ground truth triples. Thus, we incorporated a crowdsourcing process to improve the quality of the KB. RESULTS: We conducted two rounds of crowdsourcing experiments using a new corpus with 82 obesity and cancer-related PubMed abstracts. We demonstrated that crowdsourcing is indeed a low-cost mechanism to collect labeled data from non-expert laypeople. Even though individual layperson might not offer reliable answers, the collective wisdom of the crowd is comparable to expert opinions. We also retrained the relation detection machine learning models in OC-2-KB using the crowd annotated data and evaluated the content of the curated KB with a set of competency questions. Our evaluation showed improved performance of the underlying relation detection model in comparison to the baseline OC-2-KB. CONCLUSIONS: We presented a new version of OC-2-KB, a system that automatically builds an evidence-based obesity and cancer KB from scientific literature. Our KB construction framework integrated automatic information extraction with crowdsourcing techniques to verify the extracted knowledge. Our ultimate goal is a paradigm shift in how the general public access, read, digest, and use online health information.

An ontology-guided semantic data integration framework to support integrative data analysis of cancer survival.

BACKGROUND: Cancer is the second leading cause of death in the United States, exceeded only by heart disease. Extant cancer survival analyses have primarily focused on individual-level factors due to limited data availability from a single data source. There is a need to integrate data from different sources to simultaneously study as much risk factors as possible. Thus, we proposed an ontology-based approach to integrate heterogeneous datasets addressing key data integration challenges. METHODS: Following best practices in ontology engineering, we created the Ontology for Cancer Research Variables (OCRV) adapting existing semantic resources such as the National Cancer Institute (NCI) Thesaurus. Using the global-as-view data integration approach, we created mapping axioms to link the data elements in different sources to OCRV. Implemented upon the Ontop platform, we built a data integration pipeline to query, extract, and transform data in relational databases using semantic queries into a pooled dataset according to the downstream multi-level Integrative Data Analysis (IDA) needs. RESULTS: Based on our use cases in the cancer survival IDA, we created tailored ontological structures in OCRV to facilitate the data integration tasks. Specifically, we created a flexible framework addressing key integration challenges: (1) using a shared, controlled vocabulary to make data understandable to both human and computers, (2) explicitly modeling the semantic relationships makes it possible to compute and reason with the data, (3) linking patients to contextual and environmental factors through geographic variables, (4) being able to document the data manipulation and integration processes clearly in the ontologies. CONCLUSIONS: Using an ontology-based data integration approach not only standardizes the definitions of data variables through a common, controlled vocabulary, but also makes the semantic relationships among variables from different sources explicit and clear to all users of the same datasets. Such an approach resolves the ambiguity in variable selection, extraction and integration processes and thus improve reproducibility of the IDA.

Towards a privacy preserving cohort discovery framework for clinical research networks.

BACKGROUND: The last few years have witnessed an increasing number of clinical research networks (CRNs) focused on building large collections of data from electronic health records (EHRs), claims, and patient-reported outcomes (PROs). Many of these CRNs provide a service for the discovery of research cohorts with various health conditions, which is especially useful for rare diseases. Supporting patient privacy can enhance the scalability and efficiency of such processes; however, current practice mainly relies on policy, such as guidelines defined in the Health Insurance Portability and Accountability Act (HIPAA), which are insufficient for CRNs (e.g., HIPAA does not require encryption of data - which can mitigate insider threats). By combining policy with privacy enhancing technologies we can enhance the trustworthiness of CRNs. The goal of this research is to determine if searchable encryption can instill privacy in CRNs without sacrificing their usability. METHODS: We developed a technique, implemented in working software to enable privacy-preserving cohort discovery (PPCD) services in large distributed CRNs based on elliptic curve cryptography (ECC). This technique also incorporates a block indexing strategy to improve the performance (in terms of computational running time) of PPCD. We evaluated the PPCD service with three real cohort definitions: (1) elderly cervical cancer patients who underwent radical hysterectomy, (2) oropharyngeal and tongue cancer patients who underwent robotic transoral surgery, and (3) female breast cancer patients who underwent mastectomy) with varied query complexity. These definitions were tested in an encrypted database of 7.1 million records derived from the publically available Healthcare Cost and Utilization Project (HCUP) Nationwide Inpatient Sample (NIS). We assessed the performance of the PPCD service in terms of (1) accuracy in cohort discovery, (2) computational running time, and (3) privacy afforded to the underlying records during PPCD. RESULTS: The empirical results indicate that the proposed PPCD can execute cohort discovery queries in a reasonable amount of time, with query runtime in the range of 165-262s for the 3 use cases, with zero compromise in accuracy. We further show that the search performance is practical because it supports a highly parallelized design for secure evaluation over encrypted records. Additionally, our security analysis shows that the proposed construction is resilient to standard adversaries. CONCLUSIONS: PPCD services can be designed for clinical research networks. The security construction presented in this work specifically achieves high privacy guarantees by preventing both threats originating from within and beyond the network.

Assessing the Quality of Mobile Exercise Apps Based on the American College of Sports Medicine Guidelines: A Reliable and Valid Scoring Instrument.

BACKGROUND: Regular physical activity can not only help with weight management, but also lower cardiovascular risks, cancer rates, and chronic disease burden. Yet, only approximately 20% of Americans currently meet the physical activity guidelines recommended by the US Department of Health and Human Services. With the rapid development of mobile technologies, mobile apps have the potential to improve participation rates in exercise programs, particularly if they are evidence-based and are of sufficient content quality. OBJECTIVE: The goal of this study was to develop and test an instrument, which was designed to score the content quality of exercise program apps with respect to the exercise guidelines set forth by the American College of Sports Medicine (ACSM). METHODS: We conducted two focus groups (N=14) to elicit input for developing a preliminary 27-item scoring instruments based on the ACSM exercise prescription guidelines. Three reviewers who were no sports medicine experts independently scored 28 exercise program apps using the instrument. Inter- and intra-rater reliability was assessed among the 3 reviewers. An expert reviewer, a Fellow of the ACSM, also scored the 28 apps to create criterion scores. Criterion validity was assessed by comparing nonexpert reviewers' scores to the criterion scores. RESULTS: Overall, inter- and intra-rater reliability was high with most coefficients being greater than .7. Inter-rater reliability coefficients ranged from .59 to .99, and intra-rater reliability coefficients ranged from .47 to 1.00. All reliability coefficients were statistically significant. Criterion validity was found to be excellent, with the weighted kappa statistics ranging from .67 to .99, indicating a substantial agreement between the scores of expert and nonexpert reviewers. Finally, all apps scored poorly against the ACSM exercise prescription guidelines. None of the apps received a score greater than 35, out of a possible maximal score of 70. CONCLUSIONS: We have developed and presented valid and reliable scoring instruments for exercise program apps. Our instrument may be useful for consumers and health care providers who are looking for apps that provide safe, progressive general exercise programs for health and fitness.

Analysis of the accuracy of weight loss information search engine results on the internet.

OBJECTIVES: We systematically identified and evaluated the quality and comprehensiveness of online information related to weight loss that users were likely to access. METHODS: We evaluated the content quality, accessibility of the information, and author credentials for Web sites in 2012 that were identified from weight loss specific queries that we generated. We scored the content with respect to available evidence-based guidelines for weight loss. RESULTS: One hundred three Web sites met our eligibility criteria (21 commercial, 52 news/media, 7 blogs, 14 medical, government, or university, and 9 unclassified sites). The mean content quality score was 3.75 (range=0-16; SD=2.48). Approximately 5% (4.85%) of the sites scored greater than 8 (of 12) on nutrition, physical activity, and behavior. Content quality score varied significantly by type of Web site; the medical, government, or university sites (mean=4.82, SD=2.27) and blogs (mean=6.33, SD=1.99) had the highest scores. Commercial (mean=2.37, SD=2.60) or news/media sites (mean=3.52, SD=2.31) had the lowest scores (analysis of variance P<.005). CONCLUSIONS: The weight loss information that people were likely to access online was often of substandard quality because most comprehensive and quality Web sites ranked too low in search results.

Development of a Probabilistic Boolean network (PBN) to model intraoperative blood pressure management.

BACKGROUND: Blood pressure is a vital sign for organ perfusion that anesthesiologists measure and modulate during surgery. However, current decision-making processes rely heavily on clinicians' experience, which can lead to variability in treatment across surgeries. With the advent of machine learning, we can now create models to predict the outcomes of interventions and guide perioperative decision-making. The first step in this process involves translating the clinical decision-making process into a framework understood by an algorithm. Probabilistic Boolean networks (PBNs) provide an information-rich approach to this problem. A PBN trends toward a steady state, and its decisions are easily understood via its Boolean predictor functions. We hypothesize that a PBN can be developed that corrects hemodynamic instability in patients by selecting clinical interventions to maintain blood pressure within a given range. METHODS: Data on patients over the age of 65 undergoing surgery with general anesthesia from 2018 to 2020 were drawn from the UF Health PRECEDE data set with IRB approval (IRB201700747). Parameters examined included heart rate, blood pressure, and frequency of medications given 15 min after anesthetic induction and 15 min before awakening. The medication frequency data were truncated into a 66/33 split for the training and validation set used in the PBN. The model was coded using Python 3 and evaluated by comparing the frequency of medications chosen by the program to the values in the testing set via linear regression analysis. RESULTS: The network developed successfully models a hemodynamically unstable patient and corrects the imbalance by administering medications. This is evidenced by the model achieving a stable, steady state matrix in all iterations. However, the model's ability to emulate clinical drug selection was variable. It was successful with its use of vasodilator selection but struggled with the appropriate selection of vasopressors. CONCLUSIONS: The PBN has demonstrated the ability to choose appropriate interventions based on a patient's current vitals. Additional work must be done to have the network emulate the frequency at which drugs are selected from in clinical practice. In its current state, the model provides an understanding of how a PBN behaves in the context of correcting hemodynamic instability and can aid in developing more robust models in the future.

Evaluating ChatGPT's moral competence in health care-related ethical problems.

Objectives: Artificial intelligence tools such as Chat Generative Pre-trained Transformer (ChatGPT) have been used for many health care-related applications; however, there is a lack of research on their capabilities for evaluating morally and/or ethically complex medical decisions. The objective of this study was to assess the moral competence of ChatGPT. Materials and methods: This cross-sectional study was performed between May 2023 and July 2023 using scenarios from the Moral Competence Test (MCT). Numerical responses were collected from ChatGPT 3.5 and 4.0 to assess individual and overall stage scores, including C-index and overall moral stage preference. Descriptive analysis and 2-sided Student's t-test were used for all continuous data. Results: A total of 100 iterations of the MCT were performed and moral preference was found to be higher in the latter Kohlberg-derived arguments. ChatGPT 4.0 was found to have a higher overall moral stage preference (2.325 versus 1.755) when compared to ChatGPT 3.5. ChatGPT 4.0 was also found to have a statistically higher C-index score in comparison to ChatGPT 3.5 (29.03 ± 11.10 versus 19.32 ± 10.95, P =.0000275). Discussion: ChatGPT 3.5 and 4.0 trended towards higher moral preference for the latter stages of Kohlberg's theory for both dilemmas with C-indices suggesting medium moral competence. However, both models showed moderate variation in C-index scores indicating inconsistency and further training is recommended. Conclusion: ChatGPT demonstrates medium moral competence and can evaluate arguments based on Kohlberg's theory of moral development. These findings suggest that future revisions of ChatGPT and other large language models could assist physicians in the decision-making process when encountering complex ethical scenarios.

Racial disparities in septic shock mortality: a retrospective cohort study.

Background: Patients with septic shock have the highest risk of death from sepsis, however, racial disparities in mortality outcomes in this cohort have not been rigorously investigated. Our objective was to describe the association between race/ethnicity and mortality in patients with septic shock. Methods: Our study is a retrospective cohort study of adult patients in the OneFlorida Data Trust (Florida, United States of America) admitted with septic shock between January 2012 and July 2018. We identified patients as having septic shock if they received vasopressors during their hospital encounter and had either an explicit International Classification of Disease (ICD) code for sepsis, or had an infection ICD code and received intravenous antibiotics. Our primary outcome was 90-day mortality. Our secondary outcome was in-hospital mortality. Multiple logistic regression with Least Absolute Shrinkage and Selection Operator (LASSO) for variable selection was used to assess associations. Findings: There were 13,932 patients with septic shock in our cohort. The mean age was 61 years (SD 16), 68% of the cohort identified as White (n = 9419), 28% identified as Black (n = 3936), 2% (n = 294) identified as Hispanic ethnicity, and 2% as other races not specified in the previous groups (n = 283). In our logistic regression model for 90-day mortality, patients identified as Black had 1.57 times the odds of mortality (95% CI 1.07-2.29, p = 0.02) compared to White patients. Other significant predictors included mechanical ventilation (OR 3.66, 95% CI 3.35-4.00, p < 0.01), liver disease (OR 1.75, 95% CI 1.59-1.93, p < 0.01), laboratory components of the Sequential Organ Failure Assessment score (OR 1.18, 95% CI 1.16-1.21, p < 0.01), lactate (OR 1.10, 95% CI 1.08-1.12, p < 0.01), congestive heart failure (OR 1.19, 95% CI 1.10-1.30, p < 0.01), human immunodeficiency virus (OR 1.35, 95% CI 1.04-1.75, p = 0.03), age (OR 1.04, 95% CI 1.04-1.04, p < 0.01), and the interaction between age and race (OR 0.99, 95% CI 0.99-1.00, p < 0.01). Among younger patients (<45 years), patients identified as Black accounted for a higher proportion of the deaths. Results were similar in the in-hospital mortality model. Interpretation: In this retrospective study of septic shock patients, we found that patients identified as Black had higher odds of mortality compared to patients identified as non-Hispanic White. Our findings suggest that the greatest disparities in mortality are among younger Black patients with septic shock. Funding: National Institutes of Health National Center for Advancing Translational Sciences (1KL2TR001429); National Institute of Health National Institute of General Medical Sciences (1K23GM144802).

Classifying early infant feeding status from clinical notes using natural language processing and machine learning.

The objective of this study is to develop and evaluate natural language processing (NLP) and machine learning models to predict infant feeding status from clinical notes in the Epic electronic health records system. The primary outcome was the classification of infant feeding status from clinical notes using Medical Subject Headings (MeSH) terms. Annotation of notes was completed using TeamTat to uniquely classify clinical notes according to infant feeding status. We trained 6 machine learning models to classify infant feeding status: logistic regression, random forest, XGBoost gradient descent, k-nearest neighbors, and support-vector classifier. Model comparison was evaluated based on overall accuracy, precision, recall, and F1 score. Our modeling corpus included an even number of clinical notes that was a balanced sample across each class. We manually reviewed 999 notes that represented 746 mother-infant dyads with a mean gestational age of 38.9 weeks and a mean maternal age of 26.6 years. The most frequent feeding status classification present for this study was exclusive breastfeeding [n = 183 (18.3%)], followed by exclusive formula bottle feeding [n = 146 (14.6%)], and exclusive feeding of expressed mother's milk [n = 102 (10.2%)], with mixed feeding being the least frequent [n = 23 (2.3%)]. Our final analysis evaluated the classification of clinical notes as breast, formula/bottle, and missing. The machine learning models were trained on these three classes after performing balancing and down sampling. The XGBoost model outperformed all others by achieving an accuracy of 90.1%, a macro-averaged precision of 90.3%, a macro-averaged recall of 90.1%, and a macro-averaged F1 score of 90.1%. Our results demonstrate that natural language processing can be applied to clinical notes stored in the electronic health records to classify infant feeding status. Early identification of breastfeeding status using NLP on unstructured electronic health records data can be used to inform precision public health interventions focused on improving lactation support for postpartum patients.

Clinical Decision Support Systems for Palliative Care Management: A Scoping Review.

CONTEXT: With the expansion of palliative care services in clinical settings, clinical decision support systems (CDSSs) have become increasingly crucial for assisting bedside nurses and other clinicians in improving the quality of care to patients with life-limiting health conditions. OBJECTIVES: To characterize palliative care CDSSs and explore end-users' actions taken, adherence recommendations, and clinical decision time. METHODS: The CINAHL, Embase, and PubMed databases were searched from inception to September 2022. The review was developed following the preferred reporting items for systematic reviews and meta-analyses extension for scoping reviews guidelines. Qualified studies were described in tables and assessed the level of evidence. RESULTS: A total of 284 abstracts were screened, and 12 studies comprised the final sample. The CDSSs selected focused on identifying patients who could benefit from palliative care based on their health status, making referrals to palliative care services, and managing medications and symptom control. Despite the variability of palliative CDSSs, all studies reported that CDSSs assisted clinicians in becoming more informed about palliative care options leading to better decisions and improved patient outcomes. Seven studies explored the impact of CDSSs on end-user adherence. Three studies revealed high adherence to recommendations while four had low adherence. Lack of feature customization and trust in guideline-based in the initial stages of feasibility and usability testing were evident, limiting the usefulness for nurses and other clinicians. CONCLUSION: This study demonstrated that implementing palliative care CDSSs can assist nurses and other clinicians in improving the quality of care for palliative patients. The studies' different methodological approaches and variations in palliative CDSSs made it challenging to compare and validate the applicability under which CDSSs are effective. Further research utilizing rigorous methods to evaluate the impact of clinical decision support features and guideline-based actions on clinicians' adherence and efficiency is recommended.