Journal of the American Medical Informatics Association最新文献

Objective: The aim of this project was to create time-aware, individual-level risk score models for adverse drug events related to multiple sclerosis disease-modifying therapy and to provide interpretable explanations for model prediction behavior.

Materials and methods: We used temporal sequences of observational medical outcomes partnership common data model (OMOP CDM) concepts derived from an electronic health record as model features. Each concept was assigned an embedding representation that was learned from a graph convolution network trained on a knowledge graph (KG) of OMOP concept relationships. Concept embeddings were fed into long short-term memory networks for 1-year adverse event prediction following drug exposure. Finally, we implemented a novel extension of the local interpretable model agnostic explanation (LIME) method, knowledge graph LIME (KG-LIME) to leverage the KG and explain individual predictions of each model.

Results: For a set of 4859 patients, we found that our model was effective at predicting 32 out of 56 adverse event types (P < .05) when compared to demographics and past diagnosis as variables. We also assessed discrimination in the form of area under the curve (AUC = 0.77 ± 0.15) and area under the precision-recall curve (AUC-PR = 0.31 ± 0.27) and assessed calibration in the form of Brier score (BS = 0.04 ± 0.04). Additionally, KG-LIME generated interpretable literature-validated lists of relevant medical concepts used for prediction.

Discussion and conclusion: Many of our risk models demonstrated high calibration and discrimination for adverse event prediction. Furthermore, our novel KG-LIME method was able to utilize the knowledge graph to highlight concepts that were important to prediction. Future work will be required to further explore the temporal window of adverse event occurrence beyond the generic 1-year window used here, particularly for short-term inpatient adverse events and long-term severe adverse events.

The integration of large language models (LLMs) like ChatGPT into medical education presents potential benefits and challenges. These technologies, aligned with constructivist learning theories, could potentially enhance critical thinking and problem-solving through inquiry-based learning environments. However, the actual impact on educational outcomes and the effectiveness of these tools in fostering learning require further empirical study. This technological shift necessitates a reevaluation of curriculum design and the development of new assessment methodologies to measure its effects accurately. Additionally, the use of LLMs introduces significant ethical concerns, particularly in addressing inherent AI biases to ensure equitable educational access. LLMs may also help reduce global disparities in medical education by providing broader access to contemporary medical knowledge and practices, though their deployment must be managed carefully to truly support the training of competent, ethical medical professionals.

Importance and objective: Identifying sources of sex-based disparities is the first step in improving clinical outcomes for female patients. Using All of Us data, we examined the association of biological sex with cost-related medication adherence (CRMA) issues in patients with cardiovascular comorbidities.

Materials and methods: Retrospective data collection identified the following patients: 18 and older, completing personal medical history surveys, having hypertension (HTN), ischemic heart disease (IHD), or heart failure (HF) with medication use history consistent with these diagnoses. Implementing univariable and adjusted logistic regression, we assessed the influence of biological sex on 7 different patient-reported CRMA outcomes within HTN, IHD, and HF patients.

Results: Our study created cohorts of HTN (n = 3891), IHD (n = 5373), and HF (n = 2151) patients having CRMA outcomes data. Within each cohort, females were significantly more likely to report various cost-related medication issues: being unable to afford medications (HTN hazards ratio [HR]: 1.68, confidence interval [CI]: 1.33-2.13; IHD HR: 2.33, CI: 1.72-3.16; HF HR: 1.82, CI: 1.22-2.71), skipping doses (HTN HR: 1.76, CI: 1.30-2.39; IHD HR: 2.37, CI: 1.69-3.64; HF HR: 3.15, CI: 1.87-5.31), taking less medication (HTN HR: 1.86, CI: 1.37-2.45; IHD HR: 2.22, CI: 1.53-3.22; HF HR: 2.99, CI: 1.78-5.02), delaying filling prescriptions (HTN HR: 1.83, CI: 1.43-2.39; IHD HR: 2.02, CI: 1.48-2.77; HF HR: 2.99, CI: 1.79-5.03), and asking for lower cost medications (HTN HR: 1.41, CI: 1.16-1.72; IHD HR: 1.75, CI: 1.37-2.22; HF HR: 1.61, CI: 1.14-2.27).

Discussion and conclusion: Our results clearly demonstrate CRMA issues disproportionately affect female patients with cardiovascular comorbidities, which may contribute to the larger sex-based disparities in cardiovascular care. These findings call for targeted interventions and strategies to address these disparities and ensure equitable access to cardiovascular medications and care for all patients.

Objectives: To understand how frailty and healthcare delays differentially mediate the association between sexual and gender minority older adults (OSGM) status and healthcare utilization.

Materials and methods: Data from the All of Us Research Program participants ≥50 years old were analyzed using marginal structural modelling to assess if frailty or healthcare delays mediated OSGM status and healthcare utilization. OSGM status, healthcare delays, and frailty were assessed using survey data. Electronic health record (EHR) data was used to measure the number of medical visits or mental health (MH) visit days, following 12 months from the calculated All of Us Frailty Index. Analyses adjusted for age, race and ethnicity, income, HIV, marital status ± general MH (only MH analyses).

Results: Compared to non-OSGM, OSGM adults have higher rates of medical visits (adjusted rate ratio [aRR]: 1.14; 95% CI: 1.03, 1.24) and MH visits (aRR: 1.85; 95% CI: 1.07, 2.91). Frailty mediated the association between OSGM status medical visits (Controlled direct effect [Rcde] aRR: 1.03, 95% CI [0.87, 1.22]), but not MH visits (Rcde aRR: 0.37 [95% CI: 0.06, 1.47]). Delays mediated the association between OSGM status and MH visit days (Rcde aRR: 2.27, 95% CI [1.15, 3.76]), but not medical visits (Rcde aRR: 1.06 [95% CI: 0.97, 1.17]).

Discussion: Frailty represents a need for medical care among OSGM adults, highlighting the importance of addressing it to improve health and healthcare utilization disparities. In contrast, healthcare delays are a barrier to MH care, underscoring the necessity of targeted strategies to ensure timely MH care for OSGM adults.

Objective: The All of Us Research Program aims to return value to participants by developing research capacity in communities. We describe a novel set of introductory exercises (Data Sandboxes) and specialized trainings to orient researchers to the Researcher Workbench to foster health equity research.

Materials and methods: We developed a tailored training to familiarize researchers with the All of Us Research Program: (1) orientation, (2) tailored "data treasure hunt" using the Public Data Browser, and (3) overview of the analyses tools and platform.

Results: Participants' pre- and post-knowledge of the contents and structure of the All of Us dataset scores increased significantly after training. These trainings effectively engaged researchers in exploring this rich dataset.

Conclusion: We describe ways of orienting and familiarizing a wide variety of researchers with the All of Us Research Program dataset, sparking their interest, and "jump-starting" their research.

Objective: We aimed to evaluate the feasibility of using ChatGPT as programming support for nursing PhD students conducting analyses using the All of Us Researcher Workbench.

Materials and methods: 9 students in a PhD-level nursing course were prospectively randomized into 2 groups who used ChatGPT for programming support on alternating assignments in the workbench. Students reported completion time, confidence, and qualitative reflections on barriers, resources used, and the learning process.

Results: The median completion time was shorter for novices and certain assignments using ChatGPT. In qualitative reflections, students reported ChatGPT helped generate and troubleshoot code and facilitated learning but was occasionally inaccurate.

Discussion: ChatGPT provided cognitive scaffolding that enabled students to move toward complex programming tasks using the All of Us Researcher Workbench but should be used in combination with other resources.

Conclusion: Our findings support the feasibility of using ChatGPT to help PhD nursing students use the All of Us Researcher Workbench to pursue novel research directions.

Objective: This study leverages the rich diversity of the All of Us Research Program (All of Us)'s dataset to devise a predictive model for cardiovascular disease (CVD) in breast cancer (BC) survivors. Central to this endeavor is the creation of a robust data integration pipeline that synthesizes electronic health records (EHRs), patient surveys, and genomic data, while upholding fairness across demographic variables.

Materials and methods: We have developed a universal data wrangling pipeline to process and merge heterogeneous data sources of the All of Us dataset, address missingness and variance in data, and align disparate data modalities into a coherent framework for analysis. Utilizing a composite feature set including EHR, lifestyle, and social determinants of health (SDoH) data, we then employed Adaptive Lasso and Random Forest regression models to predict 6 CVD outcomes. The models were evaluated using the c-index and time-dependent Area Under the Receiver Operating Characteristic Curve over a 10-year period.

Results: The Adaptive Lasso model showed consistent performance across most CVD outcomes, while the Random Forest model excelled particularly in predicting outcomes like transient ischemic attack when incorporating the full multi-model feature set. Feature importance analysis revealed age and previous coronary events as dominant predictors across CVD outcomes, with SDoH clustering labels highlighting the nuanced impact of social factors.

Discussion: The development of both Cox-based predictive model and Random Forest Regression model represents the extensive application of the All of Us, in integrating EHR and patient surveys to enhance precision medicine. And the inclusion of SDoH clustering labels revealed the significant impact of sociobehavioral factors on patient outcomes, emphasizing the importance of comprehensive health determinants in predictive models. Despite these advancements, limitations include the exclusion of genetic data, broad categorization of CVD conditions, and the need for fairness analyses to ensure equitable model performance across diverse populations. Future work should refine clinical and social variable measurements, incorporate advanced imputation techniques, and explore additional predictive algorithms to enhance model precision and fairness.

Conclusion: This study demonstrates the liability of the All of Us's diverse dataset in developing a multi-modality predictive model for CVD in BC survivors risk stratification in oncological survivorship. The data integration pipeline and subsequent predictive models establish a methodological foundation for future research into personalized healthcare.

Objectives: To evaluate the NIH All of Us Research Program database as a potential data source for studying allostatic load and stress among adults in the United States (US).

Materials and methods: We evaluated the All of Us database to determine sample size significance for original-10 allostatic load biomarkers, Allostatic Load Index-5 (ALI-5), Allostatic Load Five, and Cohen's Perceived Stress Scale (PSS). We conducted a priori, post hoc, and sensitivity power analyses to determine sample sizes for conducting null hypothesis significance tests.

Results: The maximum number of responses available for each measure is 21 participants for the original-10 allostatic load biomarkers, 150 for the ALI-5, 22 476 for Allostatic Load Five, and n = 90 583 for the PSS.

Discussion: The NIH All of Us Research Program is well-suited for studying allostatic load using the Allostatic Load Five and psychological stress using PSS.

Conclusion: Improving biomarker data collection in All of Us will facilitate more nuanced examinations of allostatic load among US adults.

Objective: Summaries of health research can be a complementary way to return value to participants. We assess how research participants engage with summaries via email communication and how this can be improved.

Materials and methods: We look at correlations between demographic subgroups and engagement in a longitudinal dataset of 305 626 participants (77% are classified as underrepresented in biomedical research) from the All of Us Research Program. We compare this against engagement with other program communications and use impact evaluations (N = 421 510) to measure the effect of tailoring communication by (1) eliciting content preferences, (2) Spanish focused content, (3) informational videos, and (4) article content in the email subject line.

Results: Between March 2020 and October 2021, research summaries reached 67% of enrolled participants, outperforming other program communication (60%) and return of results (31%), which have a high uptake rate but have been extended to a subset of eligible participants. While all demographic subgroups engage with research summaries, participants with higher income, educational attainment, White, and older than 45 years open and click content most often. Surfacing article content in the email subject line and Spanish focused content had negative effects on engagement. Video and social media content and eliciting preferences led to a small directional increase in clicks.

Discussion: Further individualization of tailoring efforts may be needed to drive larger engagement effects (eg, delivering multiple articles in line with stated preferences, expanding preference options). Our findings are likely a conservative representation of engagement effects, given the coarseness of our click rate measure.

Conclusions: Health research summaries show promise as a way to return value to research participants, especially if individual-level results cannot be returned. Personalization of communication requires testing to determine whether efforts are having the expected effect.

Objectives: Despite easy-to-use tools like the Cohort Builder, using All of Us Research Program data for complex research questions requires a relatively high level of technical expertise. We aimed to increase research and training capacity and reduce barriers to entry for the All of Us community through an R package, allofus. In this article, we describe functions that address common challenges we encountered while working with All of Us Research Program data, and we demonstrate this functionality with an example of creating a cohort of All of Us participants by synthesizing electronic health record and survey data with time dependencies.

Target audience: All of Us Research Program data are widely available to health researchers. The allofus R package is aimed at a wide range of researchers who wish to conduct complex analyses using best practices for reproducibility and transparency, and who have a range of experience using R. Because the All of Us data are transformed into the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM), researchers familiar with existing OMOP CDM tools or who wish to conduct network studies in conjunction with other OMOP CDM data will also find value in the package.

Scope: We developed an initial set of functions that solve problems we experienced across survey and electronic health record data in our own research and in mentoring student projects. The package will continue to grow and develop with the All of Us Research Program. The allofus R package can help build community research capacity by increasing access to the All of Us Research Program data, the efficiency of its use, and the rigor and reproducibility of the resulting research.