Generative large language models are all-purpose text analytics engines: text-to-text learning is all your need.

OBJECTIVE: To solve major clinical natural language processing (NLP) tasks using a unified text-to-text learning architecture based on a generative large language model (LLM) via prompt tuning. METHODS: We formulated 7 key clinical NLP tasks as text-to-text learning and solved them using one unified generative clinical LLM, GatorTronGPT, developed using GPT-3 architecture and trained with up to 20 billion parameters. We adopted soft prompts (ie, trainable vectors) with frozen LLM, where the LLM parameters were not updated (ie, frozen) and only the vectors of soft prompts were updated, known as prompt tuning. We added additional soft prompts as a prefix to the input layer, which were optimized during the prompt tuning. We evaluated the proposed method using 7 clinical NLP tasks and compared them with previous task-specific solutions based on Transformer models. RESULTS AND CONCLUSION: The proposed approach achieved state-of-the-art performance for 5 out of 7 major clinical NLP tasks using one unified generative LLM. Our approach outperformed previous task-specific transformer models by ∼3% for concept extraction and 7% for relation extraction applied to social determinants of health, 3.4% for clinical concept normalization, 3.4%-10% for clinical abbreviation disambiguation, and 5.5%-9% for natural language inference. Our approach also outperformed a previously developed prompt-based machine reading comprehension (MRC) model, GatorTron-MRC, for clinical concept and relation extraction. The proposed approach can deliver the "one model for all" promise from training to deployment using a unified generative LLM.

Identifying social determinants of health from clinical narratives: A study of performance, documentation ratio, and potential bias.

OBJECTIVE: To develop a natural language processing (NLP) package to extract social determinants of health (SDoH) from clinical narratives, examine the bias among race and gender groups, test the generalizability of extracting SDoH for different disease groups, and examine population-level extraction ratio. METHODS: We developed SDoH corpora using clinical notes identified at the University of Florida (UF) Health. We systematically compared 7 transformer-based large language models (LLMs) and developed an open-source package - SODA (i.e., SOcial DeterminAnts) to facilitate SDoH extraction from clinical narratives. We examined the performance and potential bias of SODA for different race and gender groups, tested the generalizability of SODA using two disease domains including cancer and opioid use, and explored strategies for improvement. We applied SODA to extract 19 categories of SDoH from the breast (n = 7,971), lung (n = 11,804), and colorectal cancer (n = 6,240) cohorts to assess patient-level extraction ratio and examine the differences among race and gender groups. RESULTS: We developed an SDoH corpus using 629 clinical notes of cancer patients with annotations of 13,193 SDoH concepts/attributes from 19 categories of SDoH, and another cross-disease validation corpus using 200 notes from opioid use patients with 4,342 SDoH concepts/attributes. We compared 7 transformer models and the GatorTron model achieved the best mean average strict/lenient F1 scores of 0.9122 and 0.9367 for SDoH concept extraction and 0.9584 and 0.9593 for linking attributes to SDoH concepts. There is a small performance gap (∼4%) between Males and Females, but a large performance gap (>16 %) among race groups. The performance dropped when we applied the cancer SDoH model to the opioid cohort; fine-tuning using a smaller opioid SDoH corpus improved the performance. The extraction ratio varied in the three cancer cohorts, in which 10 SDoH could be extracted from over 70 % of cancer patients, but 9 SDoH could be extracted from less than 70 % of cancer patients. Individuals from the White and Black groups have a higher extraction ratio than other minority race groups. CONCLUSIONS: Our SODA package achieved good performance in extracting 19 categories of SDoH from clinical narratives. The SODA package with pre-trained transformer models is available at https://github.com/uf-hobi-informatics-lab/SODA_Docker.

A scoping review of fair machine learning techniques when using real-world data.

OBJECTIVE: The integration of artificial intelligence (AI) and machine learning (ML) in health care to aid clinical decisions is widespread. However, as AI and ML take important roles in health care, there are concerns about AI and ML associated fairness and bias. That is, an AI tool may have a disparate impact, with its benefits and drawbacks unevenly distributed across societal strata and subpopulations, potentially exacerbating existing health inequities. Thus, the objectives of this scoping review were to summarize existing literature and identify gaps in the topic of tackling algorithmic bias and optimizing fairness in AI/ML models using real-world data (RWD) in health care domains. METHODS: We conducted a thorough review of techniques for assessing and optimizing AI/ML model fairness in health care when using RWD in health care domains. The focus lies on appraising different quantification metrics for accessing fairness, publicly accessible datasets for ML fairness research, and bias mitigation approaches. RESULTS: We identified 11 papers that are focused on optimizing model fairness in health care applications. The current research on mitigating bias issues in RWD is limited, both in terms of disease variety and health care applications, as well as the accessibility of public datasets for ML fairness research. Existing studies often indicate positive outcomes when using pre-processing techniques to address algorithmic bias. There remain unresolved questions within the field that require further research, which includes pinpointing the root causes of bias in ML models, broadening fairness research in AI/ML with the use of RWD and exploring its implications in healthcare settings, and evaluating and addressing bias in multi-modal data. CONCLUSION: This paper provides useful reference material and insights to researchers regarding AI/ML fairness in real-world health care data and reveals the gaps in the field. Fair AI/ML in health care is a burgeoning field that requires a heightened research focus to cover diverse applications and different types of RWD.

Model tuning or prompt Tuning? a study of large language models for clinical concept and relation extraction.

OBJECTIVE: To develop soft prompt-based learning architecture for large language models (LLMs), examine prompt-tuning using frozen/unfrozen LLMs, and assess their abilities in transfer learning and few-shot learning. METHODS: We developed a soft prompt-based learning architecture and compared 4 strategies including (1) fine-tuning without prompts; (2) hard-prompting with unfrozen LLMs; (3) soft-prompting with unfrozen LLMs; and (4) soft-prompting with frozen LLMs. We evaluated GatorTron, a clinical LLM with up to 8.9 billion parameters, and compared GatorTron with 4 existing transformer models for clinical concept and relation extraction on 2 benchmark datasets for adverse drug events and social determinants of health (SDoH). We evaluated the few-shot learning ability and generalizability for cross-institution applications. RESULTS AND CONCLUSION: When LLMs are unfrozen, GatorTron-3.9B with soft prompting achieves the best strict F1-scores of 0.9118 and 0.8604 for concept extraction, outperforming the traditional fine-tuning and hard prompt-based models by 0.6 âˆ¼ 3.1 % and 1.2 âˆ¼ 2.9 %, respectively; GatorTron-345 M with soft prompting achieves the best F1-scores of 0.8332 and 0.7488 for end-to-end relation extraction, outperforming other two models by 0.2 âˆ¼ 2 % and 0.6 âˆ¼ 11.7 %, respectively. When LLMs are frozen, small LLMs have a big gap to be competitive with unfrozen models; scaling LLMs up to billions of parameters makes frozen LLMs competitive with unfrozen models. Soft prompting with a frozen GatorTron-8.9B model achieved the best performance for cross-institution evaluation. We demonstrate that (1) machines can learn soft prompts better than hard prompts composed by human, (2) frozen LLMs have good few-shot learning ability and generalizability for cross-institution applications, (3) frozen LLMs reduce computing cost to 2.5 âˆ¼ 6 % of previous methods using unfrozen LLMs, and (4) frozen LLMs require large models (e.g., over several billions of parameters) for good performance.

Geographic variation and racial disparities in adoption of newer glucose-lowering drugs with cardiovascular benefits among US Medicare beneficiaries with type 2 diabetes.

BACKGROUND: Prior studies have shown disparities in the uptake of cardioprotective newer glucose-lowering drugs (GLDs), including sodium-glucose cotranwsporter-2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1a). This study aimed to characterize geographic variation in the initiation of newer GLDs and the geographic variation in the disparities in initiating these medications. METHODS: Using 2017-2018 claims data from a 15% random nationwide sample of Medicare Part D beneficiaries, we identified individuals diagnosed with type 2 diabetes (T2D), who had ≥1 GLD prescriptions, and did not use SGLT2i or GLP1a in the year prior to the index date,1/1/2018. Patients were followed up for a year. The cohort was spatiotemporally linked to Dartmouth hospital-referral regions (HRRs), with each patient assigned to 1 of 306 HRRs. We performed multivariable Poisson regression to estimate adjusted initiation rates, and multivariable logistic regression to assess racial disparities in each HRR. RESULTS: Among 795,469 individuals with T2D included in the analyses, the mean (SD) age was 73 (10) y, 53.3% were women, 12.2% were non-Hispanic Black, and 7.2% initiated a newer GLD in the follow-up year. In the adjusted model including clinical factors, compared to non-Hispanic White patients, non-Hispanic Black (initiation rate ratio, IRR [95% CI]: 0.66 [0.64-0.68]), American Indian/Alaska Native (0.74 [0.66-0.82]), Hispanic (0.85 [0.82-0.87]), and Asian/Pacific islander (0.94 [0.89-0.98]) patients were less likely to initiate newer GLDs. Significant geographic variation was observed across HRRs, with an initiation rate spanning 2.7%-13.6%. CONCLUSIONS: This study uncovered substantial geographic variation and the racial disparities in initiating newer GLDs.

A Fair Individualized Polysocial Risk Score for Identifying Increased Social Risk in Type 2 Diabetes.

Background: Racial and ethnic minority groups and individuals facing social disadvantages, which often stem from their social determinants of health (SDoH), bear a disproportionate burden of type 2 diabetes (T2D) and its complications. It is crucial to implement effective social risk management strategies at the point of care. Objective: To develop an electronic health records (EHR)-based machine learning (ML) analytical pipeline to address unmet social needs associated with hospitalization risk in patients with T2D. Methods: We identified real-world patients with T2D from the EHR data from University of Florida (UF) Health Integrated Data Repository (IDR), incorporating both contextual SDoH (e.g., neighborhood deprivation) and individual-level SDoH (e.g., housing instability). The 2015-2020 data were used for training and validation and 2021-2022 data for independent testing. We developed a machine learning analytic pipeline, namely individualized polysocial risk score (iPsRS), to identify high social risk associated with hospitalizations in T2D patients, along with explainable AI (XAI) and fairness optimization. Results: The study cohort included 10,192 real-world patients with T2D, with a mean age of 59 years and 58% female. Of the cohort, 50% were non-Hispanic White, 39% were non-Hispanic Black, 6% were Hispanic, and 5% were other races/ethnicities. Our iPsRS, including both contextual and individual-level SDoH as input factors, achieved a C statistic of 0.72 in predicting 1-year hospitalization after fairness optimization across racial and ethnic groups. The iPsRS showed excellent utility for capturing individuals at high hospitalization risk because of SDoH, that is, the actual 1-year hospitalization rate in the top 5% of iPsRS was 28.1%, ~13 times as high as the bottom decile (2.2% for 1-year hospitalization rate). Conclusion: Our ML pipeline iPsRS can fairly and accurately screen for patients who have increased social risk leading to hospitalization in real word patients with T2D.

Enriching Real-world Data with Social Determinants of Health for Health Outcomes and Health Equity: Successes, Challenges, and Opportunities.

OBJECTIVE: To summarize the recent methods and applications that leverage real-world data such as electronic health records (EHRs) with social determinants of health (SDoH) for public and population health and health equity and identify successes, challenges, and possible solutions. METHODS: In this opinion review, grounded on a social-ecological-model-based conceptual framework, we surveyed data sources and recent informatics approaches that enable leveraging SDoH along with real-world data to support public health and clinical health applications including helping design public health intervention, enhancing risk stratification, and enabling the prediction of unmet social needs. RESULTS: Besides summarizing data sources, we identified gaps in capturing SDoH data in existing EHR systems and opportunities to leverage informatics approaches to collect SDoH information either from structured and unstructured EHR data or through linking with public surveys and environmental data. We also surveyed recently developed ontologies for standardizing SDoH information and approaches that incorporate SDoH for disease risk stratification, public health crisis prediction, and development of tailored interventions. CONCLUSIONS: To enable effective public health and clinical applications using real-world data with SDoH, it is necessary to develop both non-technical solutions involving incentives, policies, and training as well as technical solutions such as novel social risk management tools that are integrated into clinical workflow. Ultimately, SDoH-powered social risk management, disease risk prediction, and development of SDoH tailored interventions for disease prevention and management have the potential to improve population health, reduce disparities, and improve health equity.

Disparities Contributing to Late-Stage Diagnosis of Lung, Colorectal, Breast, and Cervical Cancers: Rural and Urban Poverty in Florida.

Despite advances in cancer screening, late-stage cancer diagnosis is still a major cause of morbidity and mortality in the United States. In this study, we aim to understand demographic and geographic factors associated with receiving a late-stage diagnosis (LSD) of lung, colorectal, breast, or cervical cancer. (1) Methods: We analyzed data of patients with a cancer diagnosis between 2016 and 2020 from the Florida Cancer Data System (FCDS), a statewide population-based registry. To investigate correlates of LSD, we estimated multi-variable logistic regression models for each cancer while controlling for age, sex, race, insurance, and census tract rurality and poverty. (2) Results: Patients from high-poverty rural areas had higher odds for LSD of lung (OR = 1.23, 95% CI (1.10, 1.37)) and breast cancer (OR = 1.31, 95% CI (1.17,1.47)) than patients from low-poverty urban areas. Patients in high-poverty urban areas saw higher odds of LSD for lung (OR = 1.05 95% CI (1.00, 1.09)), breast (OR = 1.10, 95% CI (1.06, 1.14)), and cervical cancer (OR = 1.19, 95% CI (1.03, 1.37)). (3) Conclusions: Financial barriers contributing to decreased access to care likely drive LSD for cancer in rural and urban communities of Florida.

Unraveling Racial Disparities in Supportive Care Medication Use among End-of-Life Pancreatic Cancer Patients: Focus on Pain Management and Psychiatric Therapies.

BACKGROUND: Supportive care medication use differences may contribute to racial disparities observed in health-related quality of life in patients with pancreatic cancer. METHODS: In this observation study using the Surveillance, Epidemiology, and End Results-Medicare linked database, we sought to examine supportive care medication use disparities in patients with pancreatic cancer from 2005 to 2017 by race and ethnicity. RESULTS: Among 74,309 patients included in the final analysis, racial and ethnic disparities in the use of supportive care medications were identified. After adjustment for confounding factors and compared with non-Hispanic Whites, minorities had significantly less use of opioids [Black: adjusted OR (aOR), 0.84; 95% confidence interval (CI), 0.79-0.88; Asian: aOR, 0.84; 95% CI, 0.79-0.90), and skeletomuscular relaxants (Black: aOR, 0.90; 95% CI, 0.82-0.99; Hispanic: aOR, 0.82; 95% CI, 0.74-0.91; Asian: aOR, 0.59; 95% CI, 0.51-0.68), and increased use of non-opioid analgesics (Hispanic: aOR, 1.16; 95% CI, 1.01-1.14; Asian: aOR, 1.37; 95% CI, 1.26-1.49). Racial and ethnic minorities had less use of antidepressants (Black: aOR, 0.56; 95% CI, 0.53-0.59; Hispanic: aOR, 0.77; 95% CI, 0.73-0.82; Asian: aOR, 0.47; 95% CI, 0.44-0.51), anxiolytics (Black: aOR, 0.78; 95% CI, 0.74-0.82; Hispanic: aOR, 0.66; 95% CI, 0.62-0.71; Asian: aOR, 0.52; 95% CI, 0.48-0.57), and antipsychotics (Hispanic: aOR, 0.90; 95% CI, 0.82-0.99; Asian: aOR, 0.84; 95% CI, 0.74-0.95). CONCLUSIONS: Racial and ethnic disparities in the use of supportive care medications among patients with pancreatic cancer were observed, with the differences unexplained by sociodemographic factors. IMPACT: Future studies should identify strategies to promote equitable use of supportive care medications among racial minorities and explore factors that may influence their use in these populations.

Individual and social determinants of adherence to sodium-glucose cotransporter 2 inhibitor therapy: A trajectory analysis.

BACKGROUND: Sodium-glucose cotransporter 2 inhibitors (SGLT2is) are known to improve cardiovascular and renal outcomes in patients with type 2 diabetes (T2D). Understanding the longitudinal patterns of adherence and the associated predictors is critical to addressing the suboptimal use of this outcome-improving treatment. OBJECTIVE: To characterize the distinct trajectories of adherence to SGLT2is in patients with T2D and to identify patient characteristics and social determinants of health (SDOHs) associated with SGLT2i adherence. METHODS: In this retrospective cohort study, we identified patients with T2D who initiated and filled at least 1 SGLT2i prescription according to 2012-2016 national Medicare claims data. The monthly proportion of days covered with SGLT2is for each patient was incorporated into group-based trajectory models to identify groups with similar adherence patterns. A multinomial logistic regression model was constructed to examine the association between patient characteristics and group membership. In addition, the association between context-specific SDOHs (eg, neighborhood median income and neighborhood employment rate) and adherence to an SGLT2i regimen was explored in both the overall cohort and the racial and ethnic subgroups. RESULTS: The final sample comprised 6,719 patients with T2D. Four trajectories of SGLT2i adherence were identified: continuously adherent users (49.6%), early discontinuers (27.5%), late discontinuers (14.5%), and intermediately adherent users (8.4%). Patient age, sex, race, diabetes duration, and Medicaid eligibility were significantly associated with trajectory group membership. Areas with a higher unemployment rate, lower income level, lower high school education rate, worse nutrition environment, fewer health care facilities, and greater Area Deprivation Index scores were found to be associated with low adherence to SGLT2is. CONCLUSIONS: Four distinct trajectories of adherence to SGLT2is were identified, with only half of the patients remaining continuously adherent to their treatment regimen during the first year after initiation. Several contextual SDOHs were associated with suboptimal adherence to SGLT2is.

Examining the relationship between metformin dose and cancer survival: A SEER-Medicare analysis.

Cancer is a major health problem in the U.S and type 2 diabetes mellitus (T2DM) is known to increase the risk for the development of many cancers. Metformin, a first-line therapy for treating T2DM, is increasingly being used for its anticancer effects; however, the literature is limited on the effect of metformin dose on overall survival in patients with stage IV cancer. Overall survival was defined as the time interval from the date of diagnosis to the last known follow-up or death from any cause. Subjects who were alive on December 31, 2016 were censored. In this cohort study we examined the relationship between metformin dose and overall survival in persons with both T2DM and stage IV lung, breast, colorectal, prostate, or pancreas cancers. We used a retrospective study design with Cox proportional hazards regression analysis of the 2007-2016 of the Surveillance Epidemiology and End Results-Medicare (SEER) dataset. Of the 7,725 patients, 2,981(38.5%) had been prescribed metformin. Patients who used metformin had significantly better overall survival in both unadjusted (Unadjusted HR, 0.73; 95% CI, 0.69-0.76; p < 0.001) and adjusted models (adjusted HR, 0.77; 95% CI, 0.73-0.81; p < 0.001). The overall survival between patients who took metformin with average daily dose ≥ 1000mg or < 1000mg were not statistically significant (aHR, 1.00; 95% CI, 0.93-1.08; p = 0.90). Metformin use regardless of dose is associated with increased overall survival in older adults with stage IV cancer.

Endocrine adherence in male versus female breast cancer: a seer-medicare review.

PURPOSE: Breast cancer in men (BC-M) is almost exclusively hormone receptor positive. We conducted a large review of the SEER-Medicare linked database to compare endocrine therapy adherence, discontinuation, and survival outcomes of male versus female patients with breast cancer. METHODS: Study data were obtained through the SEER-Medicare linked database. The study included patients age ≥ 65 years-old diagnosed with breast cancer between 2007 and 2015. The primary endpoints were rates of adherence and discontinuation of endocrine therapy (ET). Adherence was defined as a gap of less than 90 days in-between consecutive Medicare prescriptions. Discontinuation was defined as a gap of greater than 12 months in-between Medicare prescriptions. Secondary endpoint was the association of use of ET with overall survival (OS). RESULTS: Of the 363 male patients on ET, 214 patients (59.0%) were adherent to the therapy, and 149 patients (41.0%) were nonadherent. Of the 20,722 females on ET, 10,752 (51.9%) were adherent to the therapy, and 9970 (48.1%) were nonadherent. 39 male patients (10.7%) discontinued therapy, while 324 (89.3%) did not discontinue therapy. 1849 female patients (8.9%) discontinued therapy, while 18,873 (91.1%) patients did not. Men were significantly more adherent than women (p = 0.008), but there was no significant difference in discontinuation among men and women (p = 0.228). Survival was significantly improved in both men (HR 0.77, 95% CI 0.60-0.99, p = 0.039) and women (HR 0.84, 95% CI 0.81-0.87, p < 0.001) on ET. CONCLUSION: Identification of contributing factors impacting adherence and discontinuation is needed to allow physicians to address barriers to long term use of ET.

Assessment of structured data elements for social risk factors.

OBJECTIVES: Computable social risk factor phenotypes derived from routinely collected structured electronic health record (EHR) or health information exchange (HIE) data may represent a feasible and robust approach to measuring social factors. This study convened an expert panel to identify and assess the quality of individual EHR and HIE structured data elements that could be used as components in future computable social risk factor phenotypes. STUDY DESIGN: Technical expert panel. METHODS: A 2-round Delphi technique included 17 experts with an in-depth knowledge of available EHR and/or HIE data. The first-round identification sessions followed a nominal group approach to generate candidate data elements that may relate to socioeconomics, cultural context, social relationships, and community context. In the second-round survey, panelists rated each data element according to overall data quality and likelihood of systematic differences in quality across populations (ie, bias). RESULTS: Panelists identified a total of 89 structured data elements. About half of the data elements (n = 45) were related to socioeconomic characteristics. The panelists identified a diverse set of data elements. Elements used in reimbursement-related processes were generally rated as higher quality. Panelists noted that several data elements may be subject to implicit bias or reflect biased systems of care, which may limit their utility in measuring social factors. CONCLUSIONS: Routinely collected structured data within EHR and HIE systems may reflect patient social risk factors. Identifying and assessing available data elements serves as a foundational step toward developing future computable social factor phenotypes.

Disparities in Pediatric Patient Portal Activation and Feature Use.

OBJECTIVE: Disparities in adult patient portal adoption are well-documented; however, less is known about disparities in portal adoption in pediatrics. This study examines the prevalence and factors associated with patient portal activation and the use of specific portal features in general pediatrics. MATERIALS AND METHODS: We analyzed electronic health record data from 2012 to 2020 in a large academic medical center that offers both parent and adolescent portals. We summarized portal activation and use of select portal features (messaging, records access and management, appointment management, visit/admissions summaries, and interactive feature use). We used logistic regression to model factors associated with patient portal activation among all patients along with feature use and frequent feature use among ever users (ie, ≥1 portal use). RESULTS: Among 52 713 unique patients, 39% had activated the patient portal, including 36% of patients aged 0-11, 41% of patients aged 12-17, and 62% of patients aged 18-21 years. Among activated accounts, ever use of specific features ranged from 28% for visit/admission summaries to 92% for records access and management. Adjusted analyses showed patients with activated accounts were more likely to be adolescents or young adults, white, female, privately insured, and less socioeconomically vulnerable. Individual feature use among ever users generally followed the same pattern. CONCLUSIONS: Our findings demonstrate that important disparities persist in portal adoption in pediatric populations, highlighting the need for strategies to promote equitable access to patient portals.

Extracting social determinants of health from electronic health records using natural language processing: a systematic review.

OBJECTIVE: Social determinants of health (SDoH) are nonclinical dispositions that impact patient health risks and clinical outcomes. Leveraging SDoH in clinical decision-making can potentially improve diagnosis, treatment planning, and patient outcomes. Despite increased interest in capturing SDoH in electronic health records (EHRs), such information is typically locked in unstructured clinical notes. Natural language processing (NLP) is the key technology to extract SDoH information from clinical text and expand its utility in patient care and research. This article presents a systematic review of the state-of-the-art NLP approaches and tools that focus on identifying and extracting SDoH data from unstructured clinical text in EHRs. MATERIALS AND METHODS: A broad literature search was conducted in February 2021 using 3 scholarly databases (ACL Anthology, PubMed, and Scopus) following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 6402 publications were initially identified, and after applying the study inclusion criteria, 82 publications were selected for the final review. RESULTS: Smoking status (n = 27), substance use (n = 21), homelessness (n = 20), and alcohol use (n = 15) are the most frequently studied SDoH categories. Homelessness (n = 7) and other less-studied SDoH (eg, education, financial problems, social isolation and support, family problems) are mostly identified using rule-based approaches. In contrast, machine learning approaches are popular for identifying smoking status (n = 13), substance use (n = 9), and alcohol use (n = 9). CONCLUSION: NLP offers significant potential to extract SDoH data from narrative clinical notes, which in turn can aid in the development of screening tools, risk prediction models, and clinical decision support systems.

Racial, Ethnic, and Geographic Disparities in Cardiovascular Health Among Women of Childbearing Age in the United States.

Background In the United States, large disparities in cardiovascular health (CVH) exist in the general population, but little is known about the CVH status and its disparities among women of childbearing age (ie, 18-49 years). Methods and Results In this cross-sectional study, we examined racial, ethnic, and geographic disparities in CVH among all women of childbearing age in the United States, using the 2011 to 2019 Behavioral Risk Factor Surveillance System. Life's Simple 7 (ie, blood pressure, glucose, total cholesterol, smoking, body mass index, physical activity, and diet) was used to examine CVH. Women with 7 ideal CVH metrics were determined to have ideal CVH. Among the 269 564 women of childbearing age, 13 800 (4.84%) had ideal CVH. After adjusting for potential confounders, non-Hispanic Black women were less likely to have ideal CVH (odds ratio, 0.54; 95% CI, 0.46-0.63) compared with non-Hispanic White women, and with significantly lower odds of having ideal metrics of blood pressure, blood glucose, body mass index, and physical activity. No significant difference in CVH was found between non-Hispanic White and Hispanic women. Large geographic disparities with temporal variations were observed, with the age- and race-adjusted ideal CVH prevalence ranging from 4.05% in the District of Columbia (2011) to 5.55% in Maine and Montana (2019). States with low ideal CVH prevalence and average CVH score were mostly clustered in the southern United States. Conclusions Large racial, ethnic, and geographic disparities in CVH exist among women of childbearing age. More efforts are warranted to understand and address these disparities.

Comparing the downstream costs and healthcare utilization associated with the use of low-dose computed tomography (LDCT) in lung cancer screening in patients with and without alzheimer's disease and related dementias (ADRD).

OBJECTIVE: This study aims to compare the downstream costs and healthcare utilization associated with using low-dose computed tomography (LDCT) for lung cancer screening in patients with and without Alzheimer's disease and related dementias (ADRD). METHODS: Based on data from IBM MarketScan Commercial Claims Databases (2014-2018), we have identified four study cohorts: ADRD and non-ADRD patients who went through LDCT screening; ADRD and non-ADRD patients without LDCT screening. Annually healthcare utilization and cost were grouped into outpatient, inpatient, and pharmacy. We used difference-in-differences (DID) models to estimate the downstream healthcare utilization and cost associated with LDCT screening in both ADRD and non-ADRD population. We used a difference-in-difference-in-differences (DDD) model to explore whether LDCT screening was associated with higher downstream cost and healthcare utilization in ADRD population than non-ADRD population. RESULT: Compared to individuals without LDCT screening, LDCT screening was associated with increased outpatient visits (2.1, 95% CI 0.7, 3.4) and outpatient cost ($2301.0, 95% CI 296.2, 4305.8) in the ADRD population and increased outpatient visits (0.6, 95% CI 0.1, 1.1) in the non-ADRD population within 1 year after screening. Compared with the non-ADRD population, LDCT screening was found to be associated with an additional 1.5 (95% CI 0.2, 2.8) outpatient visits, 0.7 (95% CI 0.1, 1.3) days of inpatient stays, and $4,960.4 (95% CI 532.7, 9388.0) in overall healthcare costs within 1-year after LDCT in the ADRD population (all p < .5). CONCLUSION: The downstream cost and healthcare utilization associated with LDCT screening were found to be higher in the ADRD population compared to the average population.

A varied approach to left ventricular assist device follow-up improves cost-effectiveness.

BACKGROUND: Left ventricular assist device (LVAD) implantation improves outcomes in advanced heart failure, however, the optimal frequency of outpatient assessments to improve cost-effectiveness and potentially avert readmissions is unclear. METHODS: To test if varying the frequency of follow-up after LVAD implantation reduces readmissions and improves cost-effectiveness, a less intensive follow-up (LIFU) strategy with scheduled visits at 1 month and then every 6 months was compared to an intensive follow-up (IFU) group with scheduled visits at 1, 2, and 4 weeks, and then every 3 months post-implant. We developed a decision-tree model to evaluate the cost-effectiveness of different follow-up schedules at 3, 6, and 12-months. The readmission rates for LIFU and IFU, along with the associated costs, were estimated using data from the IBM MarketScan Commercial Claims Databases (2015-2018). A total of 349 patients were enrolled, with 193 and 156 in the IFU and LIFU groups. RESULTS: Patients with IFU were found to have a lower risk for readmission at 3 months (HR: 0.69, 95% confidence interval (CI): 0.60-0.79), but this difference diminished overtime at 6 months (HR: 0.84, 95% CI: 0.73-0.96) and 12 months (HR: 0.94, 95% CI: 0.83-1.06). The incremental net benefit of IFU, when compared with LIFU, is greatest in the first 3 months and also diminishes over time (3 months: $19616, 6 months $9257, 12 months $717). CONCLUSIONS: An initial IFU strategy, followed by a period of de-escalation at the 6-month post-implant mark in lower-risk patients, may be a more cost-effective strategy to provide follow-up care while not predisposing patients to a higher risk of readmission.

Semantic standards of external exposome data.

An individual's health and conditions are associated with a complex interplay between the individual's genetics and his or her exposures to both internal and external environments. Much attention has been placed on characterizing of the genome in the past; nevertheless, genetics only account for about 10% of an individual's health conditions, while the remaining appears to be determined by environmental factors and gene-environment interactions. To comprehensively understand the causes of diseases and prevent them, environmental exposures, especially the external exposome, need to be systematically explored. However, the heterogeneity of the external exposome data sources (e.g., same exposure variables using different nomenclature in different data sources, or vice versa, two variables have the same or similar name but measure different exposures in reality) increases the difficulty of analyzing and understanding the associations between environmental exposures and health outcomes. To solve the issue, the development of semantic standards using an ontology-driven approach is inevitable because ontologies can (1) provide a unambiguous and consistent understanding of the variables in heterogeneous data sources, and (2) explicitly express and model the context of the variables and relationships between those variables. We conducted a review of existing ontology for the external exposome and found only four relevant ontologies. Further, the four existing ontologies are limited: they (1) often ignored the spatiotemporal characteristics of external exposome data, and (2) were developed in isolation from other conceptual frameworks (e.g., the socioecological model and the social determinants of health). Moving forward, the combination of multi-domain and multi-scale data (i.e., genome, phenome and exposome at different granularity) and different conceptual frameworks is the basis of health outcomes research in the future.