RESUMEN
BACKGROUND AND OBJECTIVE: To develop the COVid Veteran (COVet) score for clinical deterioration in Veterans hospitalized with COVID-19 and further validate this model in both Veteran and non-Veteran samples. No such score has been derived and validated while incorporating a Veteran sample. DERIVATION COHORT: Adults (age ≥ 18 yr) hospitalized outside the ICU with a diagnosis of COVID-19 for model development to the Veterans Health Administration (VHA) (n = 80 hospitals). VALIDATION COHORT: External validation occurred in a VHA cohort of 34 hospitals, as well as six non-Veteran health systems for further external validation (n = 21 hospitals) between 2020 and 2023. PREDICTION MODEL: eXtreme Gradient Boosting machine learning methods were used, and performance was assessed using the area under the receiver operating characteristic curve and compared with the National Early Warning Score (NEWS). The primary outcome was transfer to the ICU or death within 24 hours of each new variable observation. Model predictor variables included demographics, vital signs, structured flowsheet data, and laboratory values. RESULTS: A total of 96,908 admissions occurred during the study period, of which 59,897 were in the Veteran sample and 37,011 were in the non-Veteran sample. During external validation in the Veteran sample, the model demonstrated excellent discrimination, with an area under the receiver operating characteristic curve of 0.88. This was significantly higher than NEWS (0.79; p < 0.01). In the non-Veteran sample, the model also demonstrated excellent discrimination (0.86 vs. 0.79 for NEWS; p < 0.01). The top three variables of importance were eosinophil percentage, mean oxygen saturation in the prior 24-hour period, and worst mental status in the prior 24-hour period. CONCLUSIONS: We used machine learning methods to develop and validate a highly accurate early warning score in both Veterans and non-Veterans hospitalized with COVID-19. The model could lead to earlier identification and therapy, which may improve outcomes.
Asunto(s)
COVID-19 , Aprendizaje Automático , Veteranos , Humanos , COVID-19/diagnóstico , COVID-19/epidemiología , Masculino , Femenino , Persona de Mediana Edad , Veteranos/estadística & datos numéricos , Anciano , Medición de Riesgo/métodos , Estados Unidos/epidemiología , Hospitalización/estadística & datos numéricos , Adulto , Unidades de Cuidados Intensivos , Curva ROC , Estudios de CohortesRESUMEN
Objectives: To develop and externally validate machine learning models using structured and unstructured electronic health record data to predict postoperative acute kidney injury (AKI) across inpatient settings. Materials and Methods: Data for adult postoperative admissions to the Loyola University Medical Center (2009-2017) were used for model development and admissions to the University of Wisconsin-Madison (2009-2020) were used for validation. Structured features included demographics, vital signs, laboratory results, and nurse-documented scores. Unstructured text from clinical notes were converted into concept unique identifiers (CUIs) using the clinical Text Analysis and Knowledge Extraction System. The primary outcome was the development of Kidney Disease Improvement Global Outcomes stage 2 AKI within 7 days after leaving the operating room. We derived unimodal extreme gradient boosting machines (XGBoost) and elastic net logistic regression (GLMNET) models using structured-only data and multimodal models combining structured data with CUI features. Model comparison was performed using the receiver operating characteristic curve (AUROC), with Delong's test for statistical differences. Results: The study cohort included 138â389 adult patient admissions (mean [SD] age 58 [16] years; 11â506 [8%] African-American; and 70â826 [51%] female) across the 2 sites. Of those, 2959 (2.1%) developed stage 2 AKI or higher. Across all data types, XGBoost outperformed GLMNET (mean AUROC 0.81 [95% confidence interval (CI), 0.80-0.82] vs 0.78 [95% CI, 0.77-0.79]). The multimodal XGBoost model incorporating CUIs parameterized as term frequency-inverse document frequency (TF-IDF) showed the highest discrimination performance (AUROC 0.82 [95% CI, 0.81-0.83]) over unimodal models (AUROC 0.79 [95% CI, 0.78-0.80]). Discussion: A multimodality approach with structured data and TF-IDF weighting of CUIs increased model performance over structured data-only models. Conclusion: These findings highlight the predictive power of CUIs when merged with structured data for clinical prediction models, which may improve the detection of postoperative AKI.
RESUMEN
Patients desire greater control over sharing their digital health data. Consent2Share (C2S) is an open-source consent tool offered by SAMHA and the VA to support granular data sharing (GDS) options that align with patient preferences and data privacy regulations. The need to validate this tool exists. We pilot tested C2S with 199 English and Spanish-speaking patients with behavioral health conditions (BHCs) and patient guardians. Data were analyzed using mixed methodology. All participants desired granular control over the sharing of their health data. Most participants (87%) were highly interested in using a tool that offered granular options for executing data sharing decisions, with over half (55%) indicated that being able to specify the data type, data recipient, and data use purpose made them more willing to share their medical records. Majority (83%) indicated that the supported data type sharing categories satisfied their data-sharing privacy preferences. Majority (87%) also reported that knowing the purpose of data use made them more comfortable in sharing. Some participants (28%) accessed the education materials provided on data type sharing options. Patients want granular choices when sharing medical records. Consent2Share and its supported data type sharing categories are adequate to capture patients' data sharing preferences. Further development is needed before deployment in clinical environments.
Asunto(s)
Registros Médicos , Tecnología , Humanos , Proyectos Piloto , Privacidad , Difusión de la InformaciónRESUMEN
Objectives: Learn how substance use disorder (SUD) treatment organizations view and respond to changes in confidentiality and disclosure law following adoption of CARES Act Sec. 3221 and prior to promulgation of revised implementing regulations. Methods: Online survey followed by informal interviews. Representatives of SUD organizations reported their degree of awareness of Sec. 3221 provisions and their organizations' views on amendments to disclosure practices; current and future changes of organizational policies; difficulties anticipated in implementing new rules; and preferences for resources. Results: Forty informant surveys on 30 organizations completed. Participants (62.5 percent) indicated being somewhat knowledgeable about Sec. 3221. Evenly divided positive and concerned views on Sec. 3221 reflect tension between preserving confidentiality of patient records and improving coordination of care. Most (76.7 percent) reported organizational discussions on Sec. 3221. Some (30 percent) identified changes to make in near future. Over a third expected few or no barriers to implementing changes to privacy and disclosure practices, while most (64.7 percent) expected hindrances including complexity and tensions in the law, staff education, cost, technological adjustments, and changes in the ways SUD organizations interact with external organizations and individuals. To overcome barriers noted, participants expressed desire (66.7 percent) for teaching tools such as webinars and templates to follow. Conclusions: SUD treatment organizations began thinking of and planning for proposed changes well before expected implementation of Sec. 3221. Their concerns reflected practicalities of implementation, determining content of law, and wondering about the extent to which it solves problems (improving coordination of care among various providers) or endangered other goals (protecting confidentiality of SUD patient records).
Asunto(s)
Confidencialidad , Humanos , Encuestas y CuestionariosRESUMEN
OBJECTIVES: Title 42 Code of Federal Regulations Part 2 (42 CFR Part 2 or Part 2) was enacted in 1975 to protect patients receiving treatment for substance use disorders. A scoping review on Part 2 characterized published work. METHODS: Seven databases were searched. Studies were categorized based on date of publication, primary objectives, methods, and findings. A thematic analysis was conducted using article titles and abstracts. RESULTS: Of the 36 studies identified, the majority (78%) were opinions or legal reviews, (22%) employed quantitative and/or qualitative methods to study the impact of Part 2, and over half (58%) were published in the past 5 years. Only 8% of studies reported stakeholder involvement in Part 2 studies and no study included patients or patient advocates. No study discussed the efforts to align Part 2 with Health Insurance Portability and Accountability Act or provided evidence of the benefit of this alignment. Three main themes emerged from the review: care delivery (46.6%), law and ethics (27.6%), and technology (25.9%). There were no studies focusing on stigma and its effect on substance use treatment and Part 2. CONCLUSIONS: Despite the significance of Part 2, the literature is sparse. It is time to initiate a new era of scholarly research that focuses on the impact of statutes and policies that govern substance use disorder data sharing. Specifically, we recommend research on Part 2 and Health Insurance Portability and Accountability Act alignment, the effects of Part 2 on patients, as well as providers and other stakeholder perceptions on the regulation.
Asunto(s)
Trastornos Relacionados con Sustancias , Humanos , Difusión de la Información , Trastornos Relacionados con Sustancias/terapiaRESUMEN
OBJECTIVES: This article aims to evaluate adult type 1 diabetes mellitus (T1DM) self-management behaviors (SMBs) related to exercise and alcohol on a survey versus a smartphone app to compare self-reported and self-tracked SMBs, and examine inter- and intrapatient variability. METHODS: Adults with T1DM on insulin pump therapy were surveyed about their alcohol, meal, and exercise SMBs. For 4 weeks, participants self-tracked their alcohol, meal, and exercise events, and their SMBs corresponding with these events via an investigator-developed app. Descriptive statistics and generalized linear mixed-effect models were used to analyze the data RESULTS: Thirty-five participants self-tracked over 5,000 interactions using the app. Variability in how participants perceived the effects of exercise and alcohol on their blood glucose was observed. The congruity between SMBs self-reported on the survey and those self-tracked with the app was measured as mean (SD). The lowest congruity was for alcohol and exercise with 61.9% (22.7) and 66.4% (20.2), respectively. Congruity was higher for meals with 80.9% (21.0). There was significant daily intra- and interpatient variability in SMBs related to preprandial bolusing: recommended bolus, p < 0.05; own bolus choice, p < 0.01; and recommended basal adjustment, p < 0.01. CONCLUSION: This study highlights the variability in intra- and interpatient SMBs obtained through the use of a survey and app. The outcomes of this study indicate that clinicians could use both one-time and every-day assessment tools to assess SMBs related to meals. For alcohol and exercise, further research is needed to understand the best assessment method for SMBs. Given this degree of patient variability, there is a need for an educational intervention that goes beyond the traditional "one-size-fits-all" approach of diabetes management to target individualized treatment barriers.
Asunto(s)
Conducta , Diabetes Mellitus Tipo 1/diagnóstico , Automanejo , Consumo de Bebidas Alcohólicas/epidemiología , Glucemia/análisis , Diabetes Mellitus Tipo 1/sangre , Ejercicio Físico , Femenino , Humanos , Masculino , Comidas , Persona de Mediana Edad , Aplicaciones Móviles , Autoinforme , Teléfono Inteligente , Encuestas y CuestionariosRESUMEN
OBJECTIVE: It is difficult to assess self-management behaviors (SMBs) and incorporate them into a personalized self-care plan. We aimed to develop and apply SMB phenotyping algorithms from data collected by diabetes devices and a mobile health (mHealth) application to create patient-specific SMBs reports to guide individualized interventions. Follow-up interventions aimed to understand patient's reasoning behind discovered SMB choices. METHODS: This study deals with adults on continuous subcutaneous insulin infusion using a continuous glucose monitor (CGM) who self-tracked SMBs with an mHealth application for 1 month. Patient-generated data were quantified and an SMB report was designed and populated for each participant. A diabetes educator used the report to conduct personalized, data-driven educational interventions. Thematic analysis of the intervention was conducted. RESULTS: Twenty-two participants recorded 118 alcohol, 251 exercise, 2,661 meal events, and 1,900 photos. A patient-specific SMB report was created from this data and used to conduct the educational intervention. High variability of SMB was observed between patients. There was variability in the percentage of alcohol events accompanied by a blood glucose check, median 79% (38-100% range), and frequency of changing the bolus waveform, median 11 (7-95 range). Interventions confirmed variability of SMBs. Main emerging themes from thematic analysis were: challenges and barriers, motivators, current SMB techniques, and future plans to improve glycemic control. CONCLUSION: The ability to quantify SMBs and understand patients' rationale may help improve diabetes self-care and related outcomes. This study describes our first steps in piloting a patient-specific diabetes educational intervention, as opposed to the current "one size fits all" approach.
RESUMEN
OBJECTIVE: Sensitive health information possesses risks, such as stigma and discrimination, when disclosed. Few studies have used a patient's own electronic health records (EHRs) to explore what types of information are considered sensitive andhow such perceptions affect data sharing preferences. After a systematic literature review, we designed and piloted a mixed-method approach that employs an individual's own records to assess content sensitivity and preferences for granular data sharing for care and research. METHODS: A systematic literature review of methodologies employed to assess data sharing willingness and perceptions on data sensitivity was conducted. A methodology was designed to organize and categorize sensitive health information from EHRs. Patients were asked permission to access their EHRs, including those available through the state's health information exchange. A semi-structured interview script with closed card sorting was designed and personalized to each participant's own EHRs using 30 items from each patient record. This mixed method combines the quantitative outcomes from the card sorting exercises with themes captured from interview audio recording analysis. RESULTS: Eight publications on patients' perspectives on data sharing and sensitivity were found. Based on our systematic review, the proposed method meets a need to use EHRs to systematize the study of data privacy issues. Twenty-five patients with behavioral health conditions, English and Spanish-speaking, were recruited. On average, participants recognized 82.7% of the 30 items from their own EHRs. Participants considered mental health (76.0%), sexual and reproductive health (75.0%) and alcohol use and alcoholism (50.0%) sensitive information. Participants were willing to share information related to other addictions (100.0%), genetic data (95.8%) and general physical health information (90.5%). CONCLUSION: The findings indicate diversity in patient views on EHR sensitivity and data sharing preferences and the need for more granular and patient-centered electronic consent mechanisms to accommodate patient needs. More research is needed to validate the generalizability of the proposed methodology.
Asunto(s)
Difusión de la Información , Privacidad , Registros Electrónicos de Salud , Humanos , Percepción , Proyectos de InvestigaciónRESUMEN
BACKGROUND: The goal of this study was to assess patient perspectives and satisfaction with the MiniMed 670G insulin pump. Those participants who used the pump as part of a hybrid closed loop were also asked to provide their views on the automatic feature (auto mode). METHODS: Adults with type 1 diabetes mellitus using the Medtronic™ 670G pump were asked about their experience with the device using a semi-structured survey developed by the research team. Responses were quantified to identify emergent themes. RESULTS: Seventeen participants used the pump as part of a hybrid closed loop system, while four participants used the pump in combination with a nonintegrated continuous glucose monitoring system. Overall, participants indicated a high level of satisfaction with the pump (14/21) mostly because of improvements in blood glucose (BG) control (15/21). Least liked features were physical design and structure (6/21), frequency of user input (5/21), alert frequency (4/21), and difficulty of use (3/21). Those using the hybrid closed loop were satisfied with the auto mode feature (11/17), mostly because of improvements in BG control (9/17). The least liked features of the auto mode technology were that blood glucose levels remained elevated (5/17) and the frequency of alerts (4/17). CONCLUSION: Participants indicated a high level of satisfaction with the pump and its auto mode featured mostly because of improvements in BG control. They also pointed out some key aspects of the device that are of potential clinical or commercial relevance. Additional research is needed to further evaluate users' perspectives on this new device.