Clinical depression affects 17.3 million adults in the U.S. However, 37% of these adults receive no treatment, and many symptoms remain unmanaged. buy SGC-CBP30 Mobile health apps may complement in-person treatment and address barriers to treatment, yet their quality has not been systematically appraised. We conducted a systematic review of apps for depression by searching in three major app stores. Apps were selected using specific inclusion and exclusion criteria. The final apps were downloaded and independently evaluated using the Mobile Application Rating Scale (MARS), IMS Institute for Healthcare Informatics functionality score, and six features specific to depression self-management. Mobile health apps for depression self-management exhibit a wide range of quality, but more than half (74%) of the apps had acceptable quality, with 32% having MARS scores ≥ 4.0 out of 5.0. These high scoring apps indicate that mobile apps have the potential to improve patient self-management, treatment engagement, and mental health outcomes.As part of a larger project to co-design and create a mHealth tool to support caregivers of children with cancer, we performed a pilot, qualitative study. For this portion of the project, we engaged with caregivers of children with cancer to co-create and refine a low-fidelity prototype of the Children’s Oncology Planning for Emergencies mHealth tool. Testing was accomplished through recorded semi-structured interviews with each caregiver as they interacted with a low-fidelity wireframe using Adobe Xd. Through the engagement of our key stakeholders, we were able to refine the COPE tool to provide the key elements they desired including pertinent patient medical information, checklist for planning when seeking urgent care, and coordination of care with the medical team and other caregivers.Clinical documentation burden has been broadly acknowledged, yet few interprofessional measures of burden exist. Using interprofessional time-motion study (TMS) data, we evaluated clinical workflows with a focus on electronic health record (EHR) utilization and fragmentation among 47 clinicians 34 advanced practice providers (APPs) and 13 registered nurses (RNs) from an acute care unit (n=15 observations [obs]), intensive care unit (nobs=14), ambulatory clinic (nobs=3), and emergency department (nobs=15). We examined workflow fragmentation, task-switch type, and task involvement. In our study, clinicians on average exhibited 1.4±0.6 switches per minute in their workflow. Eighty-four (19.6%) of the 429 task-switch types presented in the data accounted for 80.1% of all switches. Among those, data viewing- and data entry-related tasks were involved in 48.2% of all switches, indicating documentation burden may play a critical role in workflow disruptions. Therefore, interruption rate evaluated through task switches may serve as a proxy for measuring burden.

Characterize key tasks and information needs for heart failure disease management (HF-DM) in the distinct care setting of skilled nursing facility (SNF) staff in partnership with community-based clinical stakeholders. Develop design recommendations contextualized to the SNF setting for informatics interventions for improved HF-DM in the SNF setting.

Semi-structured interviews with fifteen participants (registered nurses, licensed practical nurses, certified nursing aides and physicians) from 8 Denver-metro SNFs. Data coded using a data-driven, inductive approach.

Key tasks of HF-DM symptom assessment, communicating change in condition, using equipment, documentation of daily weights, and monitoring patients. Themes 1) HF-DM is challenged by a culture of verbal communication; 2) staff face knowledge barriers in HF-DM that are partially attributed to unmet information needs. HF-DM information needs identification of HF patients, HF signs and symptoms, purpose of daily weights, indicators of worsening HF, purpose of sodium restricted diet, and materials to improve patients’ understanding of HF.

HF-DM information needs are not fully supported by current SNF information systems.

HF-DM information needs are not fully supported by current SNF information systems.Answering many of the research questions in the field of cancer informatics requires incorporating and centralizing data that are hosted by different parties. Federated Learning (FL) has emerged as a new approach in which a global model can be generated without disclosing private patient data by keeping them at their original location. Flexible, user-friendly, and robust infrastructures are crucial for bringing FL solutions to the day-to-day work of the cancer epidemiologist. In this paper, we present an open source priVAcy preserviNg federaTed leArninG infrastructurE for Secure Insight eXchange, VANTAGE6. We provide a detailed description of its conceptual design, modular architecture, and components. We also show a few examples where VANTAGE6 has been successfully used in research on observational cancer data. Developing and deploying technology to support federated analyses – such as VANTAGE6 – will pave the way for the adoption and mainstream practice of this new approach for analyzing decentralized data.A bleeding event is a common adverse drug reaction amongst patients on anticoagulation and factors critically into a clinician’s decision to prescribe or continue anticoagulation for atrial fibrillation. However, bleeding events are not uniformly captured in the administrative data of electronic health records (EHR). As manual review is prohibitively expensive, we investigate the effectiveness of various natural language processing (NLP) methods for automatic extraction of bleeding events. Using our expert-annotated 1,079 de-identified EHR notes, we evaluated state-of-the-art NLP models such as biLSTM-CRF with language modeling, and different BERT variants for six entity types. On our dataset, the biLSTM-CRF surpassed other models resulting in a macro F1-score of 0.75 whereas the performance difference is negligible for sentence and document-level predictions with the best macro F1-scores of 0.84 and 0.96, respectively. Our error analyses suggest that the models’ incorrect predictions can be attributed to variability in entity spans, memorization, and missing negation signals.