IntraCranial pressure prediction AlgoRithm using machinE learning (I-CARE): Training and Validation Study.

OBJECTIVES: Elevated intracranial pressure (ICP) is a potentially devastating complication of neurologic injury. Developing an ICP prediction algorithm to help the clinician adjust treatments and potentially prevent elevated ICP episodes. DESIGN: Retrospective study. SETTING: Three hundred thirty-five ICUs at 208 hospitals in the United States. SUBJECTS: Adults patients from the electronic ICU (eICU) Collaborative Research Database was used to train an ensemble machine learning model to predict the ICP 30 minutes in the future. Predictive performance was evaluated using a left-out test dataset and externally evaluated on the Medical Information Mart for Intensive Care-III (MIMIC-III) Matched Waveform Database. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Predictors included age, assigned sex, laboratories, medications and infusions, input/output, Glasgow Coma Scale (GCS) components, and time-series vitals (heart rate, ICP, mean arterial pressure, respiratory rate, and temperature). Each patient ICU stay was divided into successive 95-minute timeblocks. For each timeblock, the model was trained on nontime-varying covariates as well as on 12 observations of time-varying covariates at 5-minute intervals and asked to predict the 5-minute median ICP 30 minutes after the last observed ICP value. Data from 931 patients with ICP monitoring in the eICU dataset were extracted (46,207 timeblocks). The root mean squared error was 4.51 mm Hg in the eICU test set and 3.56 mm Hg in the MIMIC-III dataset. The most important variables driving ICP prediction were previous ICP history, patients' temperature, weight, serum creatinine, age, GCS, and hemodynamic parameters. CONCLUSIONS: IntraCranial pressure prediction AlgoRithm using machinE learning, an ensemble machine learning model, trained to predict the ICP of a patient 30 minutes in the future based on baseline characteristics and vitals data from the past hour showed promising predictive performance including in an external validation dataset.

Higher platelet counts and platelet factors are associated with a reduction in Plasmodium falciparum parasite density in young Malian children.

OBJECTIVES: The association between thrombocytopenia and parasite density or disease severity is described in numerous studies. In recent years, several studies described the protective role of platelets in directly killing Plasmodium parasites, mediated by platelet factor 4 (PF4) binding to Duffy antigen. This study aimed to evaluate the protective role of platelets in young children who are Duffy antigen-negative, such as those in sub-Saharan Africa. METHODS: A zero-inflated negative binomial model was used to relate platelet count and parasite density data collected in a longitudinal birth cohort. Platelet factors were measured by enzyme-linked immunosorbent assay in samples collected from malaria-infected children who participated in a cross-sectional study. RESULTS: We described that an increase of 10,000 platelets/µl was associated with a 2.76% reduction in parasite count. Increasing levels of PF4 and CXCL7 levels were also significantly associated with a reduction in parasite count. CONCLUSIONS: Platelets play a protective role in reducing parasite burden in Duffy-negative children, possibly mediated through activation of the innate immune system.

Case study of semaglutide and cardiovascular outcomes: An application of the Causal Roadmap to a hybrid design for augmenting an RCT control arm with real-world data.

Introduction: Increasing interest in real-world evidence has fueled the development of study designs incorporating real-world data (RWD). Using the Causal Roadmap, we specify three designs to evaluate the difference in risk of major adverse cardiovascular events (MACE) with oral semaglutide versus standard-of-care: (1) the actual sequence of non-inferiority and superiority randomized controlled trials (RCTs), (2) a single RCT, and (3) a hybrid randomized-external data study. Methods: The hybrid design considers integration of the PIONEER 6 RCT with RWD controls using the experiment-selector cross-validated targeted maximum likelihood estimator. We evaluate 95% confidence interval coverage, power, and average patient time during which participants would be precluded from receiving a glucagon-like peptide-1 receptor agonist (GLP1-RA) for each design using simulations. Finally, we estimate the effect of oral semaglutide on MACE for the hybrid PIONEER 6-RWD analysis. Results: In simulations, Designs 1 and 2 performed similarly. The tradeoff between decreased coverage and patient time without the possibility of a GLP1-RA for Designs 1 and 3 depended on the simulated bias. In real data analysis using Design 3, external controls were integrated in 84% of cross-validation folds, resulting in an estimated risk difference of -1.53%-points (95% CI -2.75%-points to -0.30%-points). Conclusions: The Causal Roadmap helps investigators to minimize potential bias in studies using RWD and to quantify tradeoffs between study designs. The simulation results help to interpret the level of evidence provided by the real data analysis in support of the superiority of oral semaglutide versus standard-of-care for cardiovascular risk reduction.

An application of the Causal Roadmap in two safety monitoring case studies: Causal inference and outcome prediction using electronic health record data.

Background: Real-world data, such as administrative claims and electronic health records, are increasingly used for safety monitoring and to help guide regulatory decision-making. In these settings, it is important to document analytic decisions transparently and objectively to assess and ensure that analyses meet their intended goals. Methods: The Causal Roadmap is an established framework that can guide and document analytic decisions through each step of the analytic pipeline, which will help investigators generate high-quality real-world evidence. Results: In this paper, we illustrate the utility of the Causal Roadmap using two case studies previously led by workgroups sponsored by the Sentinel Initiative - a program for actively monitoring the safety of regulated medical products. Each case example focuses on different aspects of the analytic pipeline for drug safety monitoring. The first case study shows how the Causal Roadmap encourages transparency, reproducibility, and objective decision-making for causal analyses. The second case study highlights how this framework can guide analytic decisions beyond inference on causal parameters, improving outcome ascertainment in clinical phenotyping. Conclusion: These examples provide a structured framework for implementing the Causal Roadmap in safety surveillance and guide transparent, reproducible, and objective analysis.

A causal roadmap for generating high-quality real-world evidence.

Increasing emphasis on the use of real-world evidence (RWE) to support clinical policy and regulatory decision-making has led to a proliferation of guidance, advice, and frameworks from regulatory agencies, academia, professional societies, and industry. A broad spectrum of studies use real-world data (RWD) to produce RWE, ranging from randomized trials with outcomes assessed using RWD to fully observational studies. Yet, many proposals for generating RWE lack sufficient detail, and many analyses of RWD suffer from implausible assumptions, other methodological flaws, or inappropriate interpretations. The Causal Roadmap is an explicit, itemized, iterative process that guides investigators to prespecify study design and analysis plans; it addresses a wide range of guidance within a single framework. By supporting the transparent evaluation of causal assumptions and facilitating objective comparisons of design and analysis choices based on prespecified criteria, the Roadmap can help investigators to evaluate the quality of evidence that a given study is likely to produce, specify a study to generate high-quality RWE, and communicate effectively with regulatory agencies and other stakeholders. This paper aims to disseminate and extend the Causal Roadmap framework for use by clinical and translational researchers; three companion papers demonstrate applications of the Causal Roadmap for specific use cases.

Applicability of the Vascular Quality Initiative mortality prediction model for infrainguinal revascularization in a tertiary limb preservation center population.

OBJECTIVE: Patients undergoing revascularization for chronic limb-threatening ischemia (CLTI) are at elevated risk for both mortality and limb loss. To facilitate therapeutic decision-making, a mortality prediction model derived from the Vascular Quality Initiative (VQI) database has stratified patients into low, medium, and high risk, defined by 30-day mortality estimates of ≤3%, 3%-5%, or >5% and 2-year mortality estimates of ≤30%, 30%-50%, or ≥50%, respectively. The purpose of this study was to compare expected mortality risk derived from this model with observed outcomes in a tertiary center. METHODS: Consecutive patients treated at a single center between 2016 and 2019 were analyzed. Baseline demographics, approach, and mortality events were reviewed. Observed mortality was obtained using life-table methods and compared using a log-rank test with the expected mortality risk that was calculated using the VQI model. RESULTS: This study cohort consisted of 195 revascularization procedures in 169 unique patients stratified into 128 (66%) low-, 50 (26%) medium-, and 17 (8%) high-risk cases based on the VQI model. Ninety percent of revascularizations were performed for tissue loss. Compared with the VQI population, comorbidities were prevalent and included unstable angina or myocardial infarction within 6 months (6% vs 2.4% in VQI; P < .001), congestive heart failure (30% vs 23%; P < .001), and dialysis dependence (14% vs 0.9%; P < .001). Patients were also older (31% vs 21% ≥80 years old; P < .001) and more likely to be frail (45% vs 64% independent; P < .001). High-risk patients were more prevalent in the endovascular group (11% of 132 endovascular interventions vs 3% of 63 bypasses; P = .056). Thirty-day observed mortality exceeded expected VQI prediction model mortality in all groups, although was not statistically significant. The VQI model adequately stratified the studied population into risk groups (P < .001). Low-risk patients with CLTI (65% of the overall cohort) experienced 2-year mortality of 18.9%. However, observed mortality rates for medium- and high-risk VQI strata were similar. After a median follow-up of 28 months, medium-risk patients incurred a significantly higher mortality than predicted (53.5% ± 2.1% vs 36.8% ± 1.1%; P = .016). CONCLUSIONS: The VQI mortality prediction model discriminates mortality risk after limb revascularization in CLTI, accurately identifying a majority subgroup of patients who are suitable for either open or endovascular intervention. However, it may underestimate mortality in a tertiary referral population with high comorbidity burden and was not well calibrated for the medium-risk group. It may be more appropriate to dichotomize patients with CLTI who are candidates for limb salvage into an average-risk and high-risk group.

Associations between social determinants of health and interpersonal violence-related injury in Cameroon: a cross-sectional study.

INTRODUCTION: Risk factors for interpersonal violence-related injury (IPVRI) in low-income and middle-income countries (LMICs) remain poorly defined. We describe associations between IPVRI and select social determinants of health (SDH) in Cameroon. METHODS: We conducted a cross-sectional analysis of prospective trauma registry data collected from injured patients >15 years old between October 2017 and January 2020 at four Cameroonian hospitals. Our primary outcome was IPVRI, compared with unintentional injury. Explanatory SDH variables included education level, employment status, household socioeconomic status (SES) and alcohol use. The EconomicClusters model grouped patients into household SES clusters: rural, urban poor, urban middle-class (MC) homeowners, urban MC tenants and urban wealthy. Results were stratified by sex. Categorical variables were compared via Pearson's χ2 statistic. Associations with IPVRI were estimated using adjusted odds ratios (aOR) with 95% confidence intervals (95%CI). RESULTS: Among 7605 patients, 5488 (72.2%) were men. Unemployment was associated with increased odds of IPVRI for men (aOR 2.44 (95% CI 1.95 to 3.06), p<0.001) and women (aOR 2.53 (95% CI 1.35 to 4.72), p=0.004), as was alcohol use (men: aOR 2.33 (95% CI 1.91 to 2.83), p<0.001; women: aOR 3.71 (95% CI 2.41 to 5.72), p<0.001). Male patients from rural (aOR 1.45 (95% CI 1.04 to 2.03), p=0.028) or urban poor (aOR 2.08 (95% CI 1.27 to 3.41), p=0.004) compared with urban wealthy households had increased odds of IPVRI, as did female patients with primary-level/no formal (aOR 1.78 (95% CI 1.10 to 2.87), p=0.019) or secondary-level (aOR 1.54 (95% CI 1.03 to 2.32), p=0.037) compared with tertiary-level education. CONCLUSION: Lower educational attainment, unemployment, lower household SES and alcohol use are risk factors for IPVRI in Cameroon. Future research should explore LMIC-appropriate interventions to address SDH risk factors for IPVRI.

Right population, right resources, right algorithm: Using machine learning efficiently and effectively in surgical systems where data are a limited resource.

There is a growing interest in using machine learning algorithms to support surgical care, diagnostics, and public health surveillance in low- and middle-income countries. From our own experience and the literature, we share several lessons for developing such models in settings where the data necessary for algorithm training and implementation is a limited resource. First, the training cohort should be as similar as possible to the population of interest, and recalibration can be used to improve risk estimates when a model is transported to a new context. Second, algorithms should incorporate existing data sources or data that is easily obtainable by frontline health workers or assistants in order to optimize available resources and facilitate integration into clinical practice. Third, the Super Learner ensemble machine learning algorithm can be used to define the optimal model for a given prediction problem while minimizing bias in the algorithm selection process. By considering the right population, right resources, and right algorithm, researchers can train prediction models that are both context-appropriate and resource-conscious. There remain gaps in data availability, affordable computing capacity, and implementation studies that hinder clinical algorithm development and use in low-resource settings, although these barriers are decreasing over time. We advocate for researchers to create open-source code, apps, and training materials to allow new machine learning models to be adapted to different populations and contexts in order to support surgical providers and health care systems in low- and middle-income countries worldwide.