Brian Rice

Background: In rural Alaska, where most communities lack road access, air ambulances serve as a critical link between remote communities and hospital services. Objective: To examine how air ambulance utilization decisions are made within the rural Alaskan health organization ecosystem. Methods: We conducted semi-structured interviews with five key stakeholders. Transcripts were analyzed thematically using the Practical, Robust Implementation and Sustainability Model (PRISM). Results: Four themes emerged that collectively define the structure of the air ambulance decision-making system, each reflecting interactions across multiple PRISM domains: limited formal guidance, logistical constraints affecting transport options, variability driven by communication and clinician risk tolerance, and reliance on unwritten rules. Together, these describe decision-making in a limited information state with incomplete feedback, leading to experiential and locally evolved practices. Conclusion: Air ambulance use reflects clinical–logistical constraints rather than critical illness alone. Linking transport decisions to outcomes and developing structured decision support may improve consistency and optimize care delivery.

Purpose: Rural Alaska relies on air ambulances (“medevacs”) to connect remote villages off the road system to higher-level care. Yet medevac decision-making is complicated by limited guidelines, high costs, and delayed transport. We aimed to quantify variability in physician medevac decisions and identify contributing factors. Methods: We conducted a cross-sectional survey of physicians (n=20) within the Alaska Tribal Health System using 20 clinical vignettes representing varying levels of transport complexity. Participants selected transport modality (medevac, commercial, or remain in village) and rated confidence. Interrater agreement was assessed using Gwet’s AC1. A generalized linear mixed model evaluated predictors of medevac use. Findings: Agreement was “almost perfect” for clearly critical cases (“always medevac,” AC1=0.97) but declined substantially for intermediate scenarios (“possible medevac” AC1=0.39; “never medevac” AC1=0.26; “special considerations” AC1=0.12). Confidence strongly correlated with agreement (r=0.83, p<0.001), forming distinct clusters of high-, medium-, and low-certainty decisions. Vignette type was the dominant driver of medevac use (OR 455.5 for “always medevac”), but decisions did not change with experience (OR=0.99, 0.94-1.04). Between-physician variability persisted after adjustment (ICC 3.8%; MOR 1.67), consistent with differences in risk tolerance. Conclusions: Medevac decision-making is highly consistent in certain clinical cases but interrater agreement is limited for intermediate cases. Without a shared gold standard of “correct” medevac decisions, future clinical decision support cannot rely on automated medevac decision making. Efforts to improve care should focus on deconstructing decisions, enhancing outcome feedback, and developing targeted decision support rather than prescriptive decision models.

Comorbidity indices are widely used in clinical research to summarize disease burden. However, traditional indices were developed decades ago in limited populations using fixed weights that do not reflect the diversity of patients in modern healthcare. We present the Personalized Comorbidity Score (PCS), a data-driven framework for context-dependent comorbidity scoring designed to capture patient complexity while remaining accessible for broad research adoption. PCS was developed using Epic Cosmos, a large national EHR network encompassing over 8 million adult inpatient encounters from 2015 to 2020, with comorbidities defined using AHRQ Clinical Classifications Software Refined categories. Models were developed separately across eight age-sex subgroups using LASSO-penalized Cox regression for feature selection and restricted mean survival time for score derivation. PCS is available in two versions: PCS Core, incorporating age, sex, and comorbidities, and PCS Extended, which additionally incorporates socioeconomic and geographic variables. PCS Core and PCS Extended achieved AUROCs of 0.812 and 0.813 for one-year mortality, outperforming traditional indices (AUROC, 0.714-0.730). PCS demonstrated consistently lower subgroup calibration error across demographic and socioeconomic groups without including race or ethnicity as model features. PCS was further evaluated in two complementary external EHR datasets (Stanford Health Care and MIMIC-IV) with distinct patient populations and data structures, where it consistently outperformed traditional indices. Open-source R and Python packages are provided to support broad adoption. PCS provides an updatable framework for comorbidity measurement that is accurate, context-dependent, and designed to evolve alongside clinical practice.

More than 50 years after Hart proposed the inverse care law, artificial intelligence (AI) in health care risks repeating the same pattern: Those who could benefit most have the least access. Using publicly available U.S. data, we show that rural areas face greater health burdens but have fewer health care resources and lower capacity to implement AI solutions. These disparities most likely extend beyond the United States. Age-adjusted mortality rates, chronic disease prevalence, and socioeconomic challenges increase with rurality, while the health care workforce and infrastructure decline. This mismatch creates contexts where AI could be most beneficial. However, AI implementation capacity declines from metropolitan to rural areas across key indicators, including interoperability infrastructure, AI adoption, and large language model readiness. In addition, clinical AI systems trained predominantly on urban populations raise concerns about distribution shift and transportability when applied to rural populations. Without deliberate intervention, AI risks amplifying rather than addressing existing disparities. Addressing this misalignment requires coordinated policy, research, and regulatory efforts that explicitly account for geography and equity. Policy should support foundational infrastructure in underserved systems, research should evaluate AI against current care alternatives rather than ideal standards, and regulation should address disparities in access and diffusion to ensure AI benefits reach areas of greatest clinical need.

American Indian and Alaska Native (AI/AN) communities are at a critical juncture in health research, where combining participatory methods with advancements in artificial intelligence and machine learning (AI/ML) can promote equity. Community-based participatory research methods which emerged to help Alaska Native communities navigate the complicated legacy of historical research abuses provide a framework to allow emerging AI/ML technologies to align with their unique world views, community strengths, and healthcare goals. A consortium of researchers (including Alaska Native Tribal Health Consortium, the Center for Alaska Native Health Research at University of Alaska, Fairbanks, Stanford University, Southcentral Foundation, and Maniilaq Association) is using community-engaged AI/ML methods to address air medical ambulance (medevac) utilization in rural communities within the Alaska Tribal Health System (ATHS). This mixed-methods convergent triangulation study uses qualitative and quantitative analyses to develop AI/ML models tailored to community needs, provider concerns, and cultural contexts. Early successes have led to a second funded project to expand community perspectives, pilot models, and address issues of governance and ethics. Using the Ethical, Legal, and Social Implications of Research framework to address implementation of AI/ML in AI/AN communities, this second grant expands community engagement, technical capacity, and creates a body within the ATHS able to provide recommendations about AI/ML security, privacy, governance and policy. These two projects have the potential to provide equitable AI/ML implementation in Alaska Native healthcare and provide a roadmap for researchers and policy makers looking to effect similar change in other AI/AN and marginalized communities.

Emergency care-sensitive conditions (ECSCs) require rapid identification and treatment and are responsible for over half of all deaths worldwide. Prehospital emergency care (PEC) can provide rapid treatment and access to definitive care for many ECSCs and can reduce mortality in several different settings. The objective of this study is to propose a method for using artificial intelligence (AI) and machine learning (ML) to transcribe audio, extract, and classify unstructured emergency call data in the Serviço de Atendimento Móvel de Urgência (SAMU) system in southern Brazil. The study used all “1-9-2” calls received in 2019 by the SAMU Novo Norte Emergency Regulation Center (ERC) call center in Maringá, in the Brazilian state of Paraná. The calls were processed through a pipeline using machine learning algorithms, including Automatic Speech Recognition (ASR) models for transcription of audio calls in Portuguese, and a Natural Language Understanding (NLU) classification model. The pipeline was trained and validated using a dataset of labeled calls, which were manually classified by medical students using LabelStudio. The results showed that the AI model was able to accurately transcribe the audio with a Word Error Rate of 42.12% using Wav2Vec 2.0 for ASR transcription of audio calls in Portuguese. Additionally, the NLU classification model had an accuracy of 73.9% in classifying the calls into different categories in a validation subset. The study found that using AI to categorize emergency calls in low- and middle-income countries is largely unexplored, and the applicability of conventional open-source ML models trained on English language datasets is unclear for non-English speaking countries. The study concludes that AI can be used to transcribe audio and extract and classify unstructured emergency call data in an emergency system in southern Brazil as an initial step towards developing a decision-making support tool.

Objectives: Pediatric fluid resuscitation in sub-Saharan Africa has traditionally occurred in inpatients. The landmark Fluid Expansion as Supportive Therapy (FEAST) trial showed fluid boluses for febrile children in this inpatient setting increased mortality. As emergency care expands in sub-Saharan Africa, fluid resuscitation increasingly occurs in the emergency unit. The objective of this study was to determine the mortality impact of emergency unit fluid resuscitation on febrile pediatric patients in Uganda. Methods: This retrospective cohort study used data from 2012–2019 from a single emergency unit in rural Western Uganda to compare three-day mortality for febrile patients that did and did not receive fluids in the emergency unit. Propensity score matching was used to create matched cohorts. Crude and multivariable logistic regression analysis (using both complete case analysis and multiple imputation) were performed on matched and unmatched cohorts. Sensitivity analysis was done separately for patients meeting FEAST inclusion and exclusion criteria. Results: The analysis included 3087 febrile patients aged 2 months to 12 years with 1,526 patients receiving fluids and 1,561 not receiving fluids. The matched cohorts each had 1,180 patients. Overall mortality was 4.0%. No significant mortality benefit or harm was shown in the crude unmatched (Odds Ratio [95% Confidence Interval] = 0.88 [0.61–1.26] or crude matched (1.00 [0.66–1.50]) cohorts. Adjusted cohort analysis (including both complete case analysis and multiple imputation) and sensitivity analysis of patients meeting FEAST inclusion and exclusion criteria all also failed to show benefit or harm. Post-hoc power calculations showed the study was powered to detect the absolute harm seen in FEAST but not the relative risk increase. Conclusions: This study’s primary finding is that fluid resuscitation in the emergency unit did not significantly increase or decrease three-day mortality for febrile children in Uganda. Universally aggressive or fluid-sparing emergency unit protocols are unlikely to be best practices, and choices about fluid resuscitation should be individualized.

Objectives Emergency care services are rapidly expanding in Africa; however, development must focus on quality. The African Federation of Emergency Medicine consensus conference (AFEM-CC)-based quality indicators were published in 2018. This study sought to increase knowledge of quality through identifying all publications from Africa containing data relevant to the AFEM-CC process clinical and outcome quality indicators. Design We conducted searches for general quality of emergency care in Africa and for each of 28 AFEM-CC process clinical and five outcome clinical quality indicators individually in the medical and grey literature. Data sources PubMed (1964—2 January 2022), Embase (1947—2 January 2022) and CINAHL (1982—3 January 2022) and various forms of grey literature were queried. Eligibility criteria Studies published in English, addressing the African emergency care population as a whole or large subsegment of this population (eg, trauma, paediatrics), and matching AFEM-CC process quality indicator parameters exactly were included. Studies with similar, but not exact match, data were collected separately as ‘AFEM-CC quality indicators near match’. Data extraction and synthesis Document screening was done in duplicate by two authors, using Covidence, and conflicts were adjudicated by a third. Simple descriptive statistics were calculated. Results One thousand three hundred and fourteen documents were reviewed, 314 in full text. 41 studies met a priori criteria and were included, yielding 59 unique quality indicator data points. Documentation and assessment quality indicators accounted for 64% of data points identified, clinical care for 25% and outcomes for 10%. An additional 53 ‘AFEM-CC quality indicators near match’ publications were identified (38 new publications and 15 previously identified studies that contained additional ‘near match’ data), yielding 87 data points. Conclusions Data relevant to African emergency care facility-based quality indicators are highly limited. Future publications on emergency care in Africa should be aware of, and conform with, AFEM-CC quality indicators to strengthen understanding of quality.

Emergency medical services (EMS) are a critical but often overlooked component of essential public health care delivery in low- and middle-income countries (LMICs). Few countries in Africa have established EMS and there is scant literature to provide guidance for EMS growth. This study aimed to characterize EMS utilization in Harare, Zimbabwe in order to guide system strengthening efforts. We performed a retrospective chart review of patient care reports (PCR) generated by the City of Harare ambulance system for patients transported and/or treated in the prehospital setting over a 14-month period (February 2018 – March 2019). A total of 875 PCRs were reviewed representing approximately 8% of the calls to EMS. The majority of patients were age 15 to 49 (76%) and 61% were female patients. In general, trauma and pregnancy were the most common chief complaints, comprising 56% of all transports. More than half (51%) of transports were for inter-facility transfers (IFTs) and 52% of these IFTs were maternity-related. Transports for trauma were mostly for male patients (63%), and 75% of the trauma patients were age 15–49. EMTs assessed and documented pulse and blood pressure for 72% of patients. In this study, EMS cared primarily for obstetric and trauma emergencies, which mirrors the leading causes of premature death in LMICs. The predominance of requests for maternity-related IFTs emphasizes the role for EMS as an integral player in peripartum maternal health care. Targeted public health efforts and chief complaint-specific training for EMTs in these priority areas could improve quality of care and patient outcomes. Moreover, a focus on strengthening prehospital data collection and research is critical to advancing EMS development in Zimbabwe and the region through quality improvement and epidemiologic surveillance.