RESUMO
OBJECTIVE: Current estimates of the incidence of tachyarrhythmias in infants rely on clinical documentation and may not reflect the true rate in the general population. Our aim was to describe the epidemiology of tachyarrhythmia detected in a large cohort of infants using direct-to-consumer heart rate (HR) monitoring. STUDY DESIGN: Data were collected from Owlet Smart Sock devices used in infants in the US with birthdates between February 2017 and February 2019. We queried the HR data for episodes of tachyarrhythmia (HR of ≥240 bpm for >60 seconds). RESULTS: The study included 100 949 infants (50.8% male) monitored for more than 200 million total hours. We identified 5070 episodes of tachyarrhythmia in 2508 infants. The cumulative incidence of tachyarrhythmia in our cohort was 2.5% over the first year of life. The median age at the time of the first episode of tachyarrhythmia was 36 days (range, 1-358 days). Tachyarrhythmia was more common in infants with congenital heart disease (4.0% vs 2.4%; P = .015) and in females (2.7% vs 2.0%; P < .001). The median length of an episode was 7.3 minutes (range, 60 seconds to 5.4 hours) and the probability of an episode lasting longer than 45 minutes was 16.8% (95% CI, 15.4%-18.3%). CONCLUSIONS: We found the cumulative incidence of tachyarrhythmia among infants using direct-to-consumer HR monitors to be higher than previously reported in studies relying on clinical diagnosis. This finding may represent previously undetected subclinical disease in young infants, the significance of which remains uncertain. Clinicians should be prepared to discuss these events with parents.
Assuntos
Triagem e Testes Direto ao Consumidor , Determinação da Frequência Cardíaca/instrumentação , Monitorização Ambulatorial/instrumentação , Taquicardia/diagnóstico , Triagem e Testes Direto ao Consumidor/métodos , Feminino , Determinação da Frequência Cardíaca/métodos , Humanos , Incidência , Lactente , Masculino , Monitorização Ambulatorial/métodos , Estudos Prospectivos , Taquicardia/epidemiologia , Estados Unidos/epidemiologiaRESUMO
The most accurate prognostic approach for follicular lymphoma (FL), progression of disease at 24 months (POD24), requires two years' observation after initiating first-line therapy (L1) to predict outcomes. We applied machine learning to structured electronic health record (EHR) data to predict individual survival at L1 initiation. We grouped 523 observations and 1933 variables from a nationwide cohort of FL patients diagnosed 2006-2014 in the Veterans Health Administration into traditionally used prognostic variables ("curated"), commonly measured labs ("labs"), and International Classification of Diseases diagnostic codes ("ICD") sets. We compared performance of random survival forests (RSF) vs. traditional Cox model using four datasets: curated, curated + labs, curated + ICD, and curated + ICD + labs, also using Cox on curated + POD24. We evaluated variable importance and partial dependence plots with area under the receiver operating characteristic curve (AUC). RSF with curated + labs performed best, with mean AUC 0.73 (95% CI: 0.71-0.75). It approximated, but did not surpass, Cox with POD24 (mean AUC 0.74 [95% CI: 0.71-0.77]). RSF using EHR data achieved better performance than traditional prognostic variables, setting the foundation for the incorporation of our algorithm into the EHR. It also provides for possible future scenarios in which clinicians could be provided an EHR-based tool which approximates the predictive ability of the most accurate known indicator, using information available 24 months earlier.
Assuntos
Linfoma Folicular , Veteranos , Registros Eletrônicos de Saúde , Humanos , Classificação Internacional de Doenças , Linfoma Folicular/diagnóstico , Aprendizado de MáquinaRESUMO
It is widely appreciated that the spectrum of traumatic brain injury (TBI), mild through severe, contains distinct clinical presentations, variably referred to as subtypes, phenotypes, and/or clinical profiles. As part of the Brain Trauma Blueprint TBI State of the Science, we review the current literature on TBI phenotyping with an emphasis on unsupervised methodological approaches, and describe five phenotypes that appear similar across reports. However, we also find the literature contains divergent analysis strategies, inclusion criteria, findings, and use of terms. Further, whereas some studies delineate phenotypes within a specific severity of TBI, others derive phenotypes across the full spectrum of severity. Together, these facts confound direct synthesis of the findings. To overcome this, we introduce PhenoBench, a freely available code repository for the standardization and evaluation of raw phenotyping data. With this review and toolset, we provide a pathway toward robust, data-driven phenotypes that can capture the heterogeneity of TBI, enabling reproducible insights and targeted care.
Assuntos
Lesões Encefálicas Traumáticas , Aprendizado de Máquina , Lesões Encefálicas Traumáticas/classificação , Lesões Encefálicas Traumáticas/diagnóstico , Humanos , Fenótipo , Padrões de ReferênciaRESUMO
Rationale: Computerized severity assessment for community-acquired pneumonia could improve consistency and reduce clinician burden. Objectives: To develop and compare 30-day mortality-prediction models using electronic health record data, including a computerized score with all variables from the original Pneumonia Severity Index (PSI) except confusion and pleural effusion ("ePSI score") versus models with additional variables. Methods: Among adults with community-acquired pneumonia presenting to emergency departments at 117 Veterans Affairs Medical Centers between January 1, 2006, and December 31, 2016, we compared an ePSI score with 10 novel models employing logistic regression, spline, and machine learning methods using PSI variables, age, sex and 26 physiologic variables as well as all 69 PSI variables. Models were trained using encounters before January 1, 2015; tested on encounters during and after January 1, 2015; and compared using the areas under the receiver operating characteristic curve, confidence intervals, and patient event rates at a threshold PSI score of 970. Results: Among 297,498 encounters, 7% resulted in death within 30 days. When compared using the ePSI score (confidence interval [CI] for the area under the receiver operating characteristic curve, 0.77-0.78), performance increased with model complexity (CI for the logistic regression PSI model, 0.79-0.80; CI for the boosted decision-tree algorithm machine learning PSI model using the Extreme Gradient Boosting algorithm [mlPSI] with the 19 original PSI factors, 0.83-0.85) and the number of variables (CI for the logistic regression PSI model using all 69 variables, 0.84-085; CI for the mlPSI with all 69 variables, 0.86-0.87). Models limited to age, sex, and physiologic variables also demonstrated high performance (CI for the mlPSI with age, sex, and 26 physiologic factors, 0.84-0.85). At an ePSI score of 970 and a mortality-risk cutoff of <2.7%, the ePSI score identified 31% of all patients as being at "low risk"; the mlPSI with age, sex, and 26 physiologic factors identified 53% of all patients as being at low risk; and the mlPSI with all 69 variables identified 56% of all patients as being at low risk, with similar rates of mortality, hospitalization, and 7-day secondary hospitalization being determined. Conclusions: Computerized versions of the PSI accurately identified patients with pneumonia who were at low risk of death. More complex models classified more patients as being at low risk of death and as having similar adverse outcomes.