RESUMEN

Background: Patients newly diagnosed with diabetes mellitus (diabetes), who require insulin must acquire diabetes "survival" skills prior to discharge home. COVID-19 revealed considerable limitations of traditional in-person, time-intensive delivery of diabetes education and survival skills training (diabetes survival skills training). Furthermore, diabetes survival skills training has not been designed to meet the specific learning needs of patients with diabetes and their caregivers, particularly if delivered by telehealth. The objective of the study was to identify and understand the needs of users (patients newly prescribed insulin and their caregivers) to inform the design of a diabetes survival skills training, specifically for telehealth delivery, through the application of user-centered design and adult learning and education principles. Methods: Users included patients newly prescribed insulin, their caregivers, and laypersons without diabetes. In semi-structured interviews, users were asked about experienced or perceived challenges in learning diabetes survival skills. Interviews were audio-recorded and transcribed. Investigators performed iterative rounds of coding of interview transcripts utilizing a constant comparative method to identify themes describing the dominant challenges users experienced. Themes were then mapped to adult learning and education principles to identify novel educational design solutions that can be applied to telehealth-based learning. Results: We interviewed 18 users: patients (N = 6, 33 %), caregivers (N = 4, 22 %), and laypersons (N = 8, 44 %). Users consistently described challenges in understanding diabetes survival skills while hospitalized; in preparing needed supplies to execute diabetes survival skills; and in executing diabetes survival skills at home. The challenges mapped to three educational strategies: (1) spiral learning; (2) repetitive goal directed practice and feedback, which have the potential to translate into design solutions supporting remote/virtual learning; and (3) form fits function organizer, which supports safe organization and use of supplies to execute diabetes survival skills independently. Conclusion: Learning complex tasks, such as diabetes survival skills, requires time, repetition, and continued support. The combination of a user-centered design approach to uncover learning needs as well as identification of relevant adult learning and education principles could inform the design of more user-centered, feasible, effective, and sustainable diabetes survival skills training for telehealth delivery.

RESUMEN

Background and Purpose: Accurate prehospital diagnosis of stroke by emergency medical services (EMS) can increase treatments rates, mitigate disability, and reduce stroke deaths. We aimed to develop a model that utilizes natural language processing of EMS reports and machine learning to improve prehospital stroke identification. Methods: We conducted a retrospective study of patients transported by the Chicago EMS to 17 regional primary and comprehensive stroke centers. Patients who were suspected of stroke by the EMS or had hospital-diagnosed stroke were included in our cohort. Text within EMS reports were converted to unigram features, which were given as input to a support-vector machine classifier that was trained on 70% of the cohort and tested on the remaining 30%. Outcomes included final diagnosis of stroke versus nonstroke, large vessel occlusion, severe stroke (National Institutes of Health Stroke Scale score >5), and comprehensive stroke center-eligible stroke (large vessel occlusion or hemorrhagic stroke). Results: Of 965 patients, 580 (60%) had confirmed acute stroke. In a test set of 289 patients, the text-based model predicted stroke nominally better than models based on the Cincinnati Prehospital Stroke Scale (c-statistic: 0.73 versus 0.67, P=0.165) and was superior to the 3-Item Stroke Scale (c-statistic: 0.73 versus 0.53, P<0.001) scores. Improvements in discrimination were also observed for the other outcomes. Conclusions: We derived a model that utilizes clinical text from paramedic reports to identify stroke. Our results require validation but have the potential of improving prehospital routing protocols.

Asunto(s)

RESUMEN

This data article provides the summary data from tests comparing various Gaussian process software packages. Each spreadsheet represents a single function or type of function using a particular input sample size. In each spreadsheet, a row gives the results for a particular replication using a single package. Within each spreadsheet there are the results from eight Gaussian process model-fitting packages on five replicates of the surface. There is also one spreadsheet comparing the results from two packages performing stochastic kriging. These data enable comparisons between the packages to determine which package will give users the best results.

Asunto(s)