Selection and Implementation of Virtual Scribe Solutions to Reduce Documentation Burden: A Mixed Methods Pilot.

Electronic health record (EHR) documentation is a leading reason for clinician burnout. While technology-enabled solutions like virtual and digital scribes aim to improve this, there is limited evidence of their effectiveness and minimal guidance for healthcare systems around solution selection and implementation. A transdisciplinary approach, informed by clinician interviews and other considerations, was used to evaluate and select a virtual scribe solution to pilot in a rapid iterative sprint over 12 weeks. Surveys, interviews, and EHR metadata were analyzed over a staggered 30 day implementation with live and asynchronous virtual scribe solutions. Among 16 pilot clinicians, documentation burden metrics decreased for some but not all. Some clinicians had highly positive comments, and others had concerns regarding scribe training and quality. Our findings demonstrate that virtual scribes may reduce documentation burden for some clinicians and describe a method for a collaborative and iterative technology selection process for digital tools in practice.

A cost-aware framework for the development of AI models for healthcare applications.

Accurate artificial intelligence (AI) for disease diagnosis could lower healthcare workloads. However, when time or financial resources for gathering input data are limited, as in emergency and critical-care medicine, developing accurate AI models, which typically require inputs for many clinical variables, may be impractical. Here we report a model-agnostic cost-aware AI (CoAI) framework for the development of predictive models that optimize the trade-off between prediction performance and feature cost. By using three datasets, each including thousands of patients, we show that relative to clinical risk scores, CoAI substantially reduces the cost and improves the accuracy of predicting acute traumatic coagulopathy in a pre-hospital setting, mortality in intensive-care patients and mortality in outpatient settings. We also show that CoAI outperforms state-of-the-art cost-aware prediction strategies in terms of predictive performance, model cost, training time and robustness to feature-cost perturbations. CoAI uses axiomatic feature-attribution methods for the estimation of feature importance and decouples feature selection from model training, thus allowing for a faster and more flexible adaptation of AI models to new feature costs and prediction budgets.

Development and evaluation of a mobile application for case management of small and sick newborns in Bangladesh.

BACKGROUND: In low-income settings, community health workers (CHWs) are frequently the first point of contact for newborns. Mobile technology may aid health workers in classifying illness and providing referral and management guidance for newborn care. This study evaluates the potential for mobile health technology to improve diagnosis and case management of newborns in Bangladesh. METHODS: A mobile application based on Bangladesh's Comprehensive Newborn Care Package national guidelines (mCNCP) was developed to aid CHWs in identifying and managing small and sick infants. After a 2-day training, CHWs assessed newborns at Sylhet Osmani Medical College Hospital and in the Projahnmo research site (Sylhet, Bangladesh) using either mCNCP or a comparable paper form (pCNCP), similar to standard IMCI-formatted paper forms. CHWs were randomized to conduct a block of ~ 6 newborn assessments starting with either mCNCP or pCNCP, then switched to the alternate method. Physicians using mCNCP served as gold standard assessors. CHW performance with mCNCP and pCNCP were compared using chi-squared tests of independence for equality of proportions, and logistic regressions clustered by CHW. RESULTS: Two hundred seven total CHW assessments were completed on 101 enrolled infants. mCNCP assessments were more often fully completed and completed faster than pCNCP assessments (100% vs 23.8%, p < 0.001; 17.5 vs 23.6 min; p < 0.001). mCNCP facilitated calculations of respiratory rate, temperature, and gestational age. CHWs using mCNCP were more likely to identify small newborns (Odds Ratio (OR): 20.8, Confidence Interval (CI): (7.1, 60.8), p < 0.001), and to correctly classify 7 out of 16 newborn conditions evaluated, including severe weight loss (OR: 13.1, CI: (4.6, 37.5), p < 0.001), poor movement (OR: 6.6, CI: (2.3, 19.3), p = 0.001), hypothermia (OR: 14.9, CI: (2.7, 82.2), p = 0.002), and feeding intolerance (OR: 2.1, CI: (1.3, 3.3), p = 0.003). CHWs with mCNCP were more likely to provide counseling as needed on 4 out of 7 case management recommendations evaluated, including kangaroo mother care. CONCLUSIONS: CHWs in rural Bangladesh with limited experience using tablets successfully used a mobile application for neonatal assessment after a two-day training. mCNCP may aid frontline health workers in Bangladesh to improve completion of neonatal assessment, classification of illnesses, and adherence to neonatal management guidelines.

An integrated approach to surgery and primary care systems strengthening in low- and middle-income countries: building a platform to deliver across the spectrum of disease.

BACKGROUND: Surgical services in low- and middle income countries (LMICs) must be considered within the context of a coordinated strategy for building primary care systems. Weak front-line primary care systems lead to delayed presentation and poor follow-up of patients with surgical illness, increasing the risk of poor outcomes. METHODS: Here we propose a framework to integrating surgery and primary care, organized around basic primary care principles of access, longitudinal care, coordination, integration and equity. RESULTS: Making surgical care accessible will require frontline provider capacity to screen for and recognize common surgical conditions, as well as to deliver certain basic surgical services themselves. Making this care effective will require strengthening the capacity of interdisciplinary teams to provide longitudinal care, involving coordinated networks for referral, communication with and mentorship by more specialized providers, and postoperative follow-up. Innovative approaches to information and communication technology can help to overcome the transportation and infrastructure barriers that jeopardize both access and effectiveness. Explicit integration of surgical and primary care programs at the managerial and administrative levels, as well as at the point-of-care, will also be critical. Taking a pro-equity approach can ensure that populations with the greatest unmet needs are effectively reached. CONCLUSION: Utilizing the pillars of effective primary care as a guiding framework to design, implement, and scale surgical programs in LMICs offers an opportunity for strengthening and enhancing the quality of health systems as a whole.

Factors associated with adherence to antiretroviral therapy among adolescents living with HIV/AIDS in low- and middle-income countries: a systematic review.

Adolescents living in low- and middle-income countries (LMICs) are disproportionately burdened by the global HIV/AIDS pandemic. Maintaining medication adherence is vital to ensuring that adolescents living with HIV/AIDS receive the benefits of antiretroviral therapy (ART), although this group faces unique challenges to adherence. Knowledge of the factors influencing adherence among people during this unique developmental period is needed to develop more targeted and effective adherence-promoting strategies. This systematic review summarizes the literature on quantitative observational studies examining correlates, including risk and resilience-promoting factors, of ART adherence among adolescents living with HIV/AIDS in LMICs. A systematic search of major electronic databases, conference-specific databases, gray literature, and reference lists of relevant reviews and documents was conducted in May 2014. Included studies examined relationships between at least one factor and ART adherence as an outcome and were conducted in primarily an adolescent population (age 10-19) in LMICs. The search identified 7948 unique citations from which 15 studies fit the inclusion criteria. These 15 studies identified 35 factors significantly associated with ART adherence representing a total of 4363 participants across nine different LMICs. Relevant studies revealed few consistent relationships between measured factors and adherence while highlighting potentially important themes for ART adherence including the impact of (1) adolescent factors such as gender and knowledge of serostatus, (2) family structure, (3) the burdensome ART regimens, route of administration, and attitudes about medication, and (4) health care and environmental factors, such as rural versus urban location and missed clinic appointments. Rates of adherence across studies ranged from 16% to 99%. This review identifies unique factors significantly related to ART adherence among adolescents living in LMICs. More research using longitudinal designs and rigorous measures of adherence is required in order to identify the range of factors influencing ART adherence as adolescents living with HIV/AIDS in LMICs grow into adulthood.