We conducted a cross-sectional study to investigate racial disparities among patients hospitalized for COVID-19 using electronic health records from a large community-based health care system between March 2020 and February 2021. The primary outcomes were in-hospital mortality and 30-day readmission. Demographics, clinical characteristics, and socioeconomic factors were analyzed using logistic regression, with adjusted odds ratios (AOR) reported. The study identified 90,490 patients, including Hispanic (29%), White (45.5%), Black (17.3%), Asian (3.1%), and Others (4.1%). The mortality model demonstrated a strong performance with an area under curve (AUC) of 0.88 (95%CI=0.877-0.885) and a Brier score of 0.093 (95%CI=0.092-0.095). For the 30-day readmission model, the AUC was 0.726 (95%CI=0.717-0.734) with a Brier score of 0.091 (95%CI=0.088-0.093). Compared with Whites, Hispanics exhibited a higher risk of mortality (AOR=1.23; 95%CI=1.16-1.31) but a lower likelihood of readmission (AOR=0.89; 95%CI=0.83-0.97). These findings highlight the critical need for targeted health management strategies across different racial and ethnic groups.

As data grows exponentially across diverse fields, effectively leveraging big data has become increasingly crucial. In data science and computational genomics, however, minority groups, including African Americans, are significantly underrepresented, coupled with the lack of resources and infrastructure in minority-serving institutions. This paper summarizes the second phase of our funded project that aims to enhance the data science capacity of Meharry Medical College (MMC), a Historically Black College/University (HBCU), by providing training and fostering collaborations between data scientists and researchers in basic science and biomedical fields. Using diverse training approaches and formats, we introduced data science and computational genomics to hundreds of MMC researchers and students in the past 2 years. The training modules designed for dental curriculums introduced artificial intelligence and machine learning to ~250 dental students, 80% of which are African Americans (AA). We have also fostered partnerships between data scientists and other MMC researchers for joint publications and grant applications in various areas that impact the health of AA population. The multiple grants awarded recently to MMC clearly indicate an enhanced data science and genomics capacity of MMC and the impact of our work on the local community.

As data grows exponentially across diverse fields, the ability to effectively leverage big data has become increasingly crucial. In the field of data science, however, minority groups, including African Americans, are significantly underrepresented. With the strategic role of minority-serving institutions to enhance diversity in the data science workforce and apply data science to health disparities, the National Institute for Minority Health Disparities (NIMHD) provided funding in September 2021 to six Research Centers in Minority Institutions (RCMI) to improve their data science capacity and foster collaborations with data scientists. Meharry Medical College (MMC), a historically Black College/University (HBCU), was among the six awardees. This paper summarizes the NIMHD-funded efforts at MMC, which include offering mini-grants to collaborative research groups, surveys to understand the needs of the community to guide project implementation, and data science training to enhance the data analytics skills of the RCMI investigators, staff, medical residents, and graduate students. This study is innovative as it addressed the urgent need to enhance the data science capacity of the RCMI program at MMC, build a diverse data science workforce, and develop collaborations between the RCMI and MMC's newly established School of Applied Computational Science. This paper presents the progress of this NIMHD-funded project, which clearly shows its positive impact on the local community.

This paper describes our efforts to enhance the research capacity of the Research Centers in Minority Institutions (RCMI) program at Meharry Medical College and foster collaborations between RCMI and a newly established School of Applied Computational Sciences (SACS). These efforts include offering subawards to collaborative research groups, financial support to graduate students for their participation in collaborative research, performing need assessment to understand the needs of the community so as to guide our project implementation, and providing data science training to enhance the data analytics skills of the RCMI investigators, staff, medical residents, and graduate students. This paper presents the progress of this project, which clearly indicates its positive impact on the local community.

Objectives. This research describes the prevalence and covariates associated with opioid-induced constipation (OIC) in an observational cohort study utilizing a national veteran cohort and integrated data from the Center for Medicare and Medicaid Services (CMS). Methods. A cohort of 152,904 veterans with encounters between 1 January 2008 and 30 November 2010, an exposure to opioids of 30 days or more, and no exposure in the prior year was developed to establish existing conditions and medications at the start of the opioid exposure and determining outcomes through the end of exposure. OIC was identified through additions/changes in laxative prescriptions, all-cause constipation identification through diagnosis, or constipation related procedures in the presence of opioid exposure. The association of time to constipation with opioid use was analyzed using Cox proportional hazard regression adjusted for patient characteristics, concomitant medications, laboratory tests, and comorbidities. Results. The prevalence of OIC was 12.6%. Twelve positively associated covariates were identified with the largest associations for prior constipation and prevalent laxative (any laxative that continued into the first day of opioid exposure). Among the 17 negatively associated covariates, the largest associations were for erythromycins, androgens/anabolics, and unknown race. Conclusions. There were several novel covariates found that are seen in the all-cause chronic constipation literature but have not been reported for opioid-induced constipation. Some are modifiable covariates, particularly medication coadministration, which may assist clinicians and researchers in risk stratification efforts when initiating opioid medications. The integration of CMS data supports the robustness of the analysis and may be of interest in the elderly population warranting future examination.

BACKGROUND:The development and adoption of health care common data models (CDMs) has addressed some of the logistical challenges of performing research on data generated from disparate health care systems by standardizing data representations and leveraging standardized terminology to express clinical information consistently. However, transforming a data system into a CDM is not a trivial task, and maintaining an operational, enterprise capable CDM that is incrementally updated within a data warehouse is challenging.OBJECTIVES:To develop a quality assurance (QA) process and code base to accompany our incremental transformation of the Department of Veterans Affairs Corporate Data Warehouse health care database into the Observational Medical Outcomes Partnership (OMOP) CDM to prevent incremental load errors.METHODS:We designed and implemented a multistage QA) approach centered on completeness, value conformance, and relational conformance data-quality elements. For each element we describe key incremental load challenges, our extract, transform, and load (ETL) solution of data to overcome those challenges, and potential impacts of incremental load failure.RESULTS:Completeness and value conformance data-quality elements are most affected by incremental changes to the CDW, while updates to source identifiers impact relational conformance. ETL failures surrounding these elements lead to incomplete and inaccurate capture of clinical concepts as well as data fragmentation across patients, providers, and locations.CONCLUSION:Development of robust QA processes supporting accurate transformation of OMOP and other CDMs from source data is still in evolution, and opportunities exist to extend the existing QA framework and tools used for incremental ETL QA processes.

BackgroundThe optimal structure and intensity of interventions to reduce hospital readmission remains uncertain, due in part to lack of head-to-head comparison. To address this gap, we evaluated two forms of an evidence-based, multi-component transitional care intervention.MethodsA quasi-experimental evaluation design compared outcomes of Transition Care Coordinator (TCC) Care to Usual Care, while controlling for sociodemographic characteristics, comorbidities, readmission risk, and administrative factors. The study was conducted between January 1, 2013 and April 30, 2015 as a quality improvement initiative. Eligible adults (N = 7038) hospitalized with pneumonia, congestive heart failure, or chronic obstructive pulmonary disease were identified for program evaluation via an electronic health record algorithm. Nurse TCCs provided either a full intervention (delivered in-hospital and by post-discharge phone call) or a partial intervention (phone call only).ResultsA total of 762 hospitalizations with TCC Care (460 full intervention and 302 partial intervention) and 6276 with Usual Care was examined. In multivariable models, hospitalizations with TCC Care had significantly lower odds of readmission at 30 days (OR = 0.512, 95% CI 0.392 to 0.668) and 90 days (OR = 0.591, 95% CI 0.483 to 0.723). Adjusted costs were significantly lower at 30 days (difference = $3969, 95% CI $5099 to $2691) and 90 days (difference = $5684, 95% CI $7602 to $3627). The effect was similar whether patients received the full or partial intervention.ConclusionAn evidence-based multi-component intervention delivered by nurse TCCs reduced 30- and 90-day readmissions and associated health care costs. Lower intensity interventions delivered by telephone after discharge may have similar effectiveness to in-hospital programs.

Objective Cirrhotic patients are at high hospitalisation risk with subsequent high mortality. Current risk prediction models have varied performances with methodological room for improvement. We used current analytical techniques using automatically extractable variables from the electronic health record (EHR) to develop and validate a posthospitalisation mortality risk score for cirrhotic patients and compared performance with the model for end-stage liver disease (MELD), model for end-stage liver disease with sodium (MELD-Na), and the CLIF Consortium Acute Decompensation (CLIF-C AD) models. Design We analysed a retrospective cohort of 73 976 patients comprising 247 650 hospitalisations between 2006 and 2013 at any of 123 Department of Veterans Affairs hospitals. Using 45 predictor variables, we built a time-dependent Cox proportional hazards model with all-cause mortality as the outcome. We compared performance to the three extant models and reported discrimination and calibration using bootstrapping. Furthermore, we analysed differential utility using the net reclassification index (NRI). Results The C-statistic for the final model was 0.863, representing a significant improvement over the MELD, MELD-Na, and the CLIF-C AD, which had C-statistics of 0.655, 0.675, and 0.679, respectively. Multiple risk factors were significant in our model, including variables reflecting disease severity and haemodynamic compromise. The NRI showed a 24% improvement in predicting survival of low-risk patients and a 30% improvement in predicting death of high-risk patients. Conclusion We developed a more accurate mortality risk prediction score using variables automatically extractable from an EHR that may be used to risk stratify patients with cirrhosis for targeted postdischarge management.

Background Early hospital readmission for patients with cirrhosis continues to challenge the healthcare system. Risk stratification may help tailor resources, but existing models were designed using small, single-institution cohorts or had modest performance. Aims We leveraged a large clinical database from the Department of Veterans Affairs (VA) to design a readmission risk model for patients hospitalized with cirrhosis. Additionally, we analyzed potentially modifiable or unexplored readmission risk factors. Methods A national VA retrospective cohort of patients with a history of cirrhosis hospitalized for any reason from January 1, 2006, to November 30, 2013, was developed from 123 centers. Using 174 candidate variables within demographics, laboratory results, vital signs, medications, diagnoses and procedures, and healthcare utilization, we built a 47-variable penalized logistic regression model with the outcome of all-cause 30-day readmission. We excluded patients who left against medical advice, transferred to a non-VA facility, or if the hospital length of stay was greater than 30 days. We evaluated calibration and discrimination across variable volume and compared the performance to recalibrated preexisting risk models for readmission. Results We analyzed 67,749 patients and 179,298 index hospitalizations. The 30-day readmission rate was 23%. Ascites was the most common cirrhosis-related cause of index hospitalization and readmission. The AUC of the model was 0.670 compared to existing models (0.649, 0.566, 0.577). The Brier score of 0.165 showed good calibration. Conclusion Our model achieved better discrimination and calibration compared to existing models, even after local recalibration. Assessment of calibration by variable parsimony revealed performance improvements for increasing variable inclusion well beyond those detectable for discrimination.

Rationale: Intensive care unit (ICU) delirium is highly prevalent and a potentially avoidable hospital complication. The current cost of ICU delirium is unknown. Objectives: To specify the association between the daily occurrence of delirium in the ICU with costs of ICU care accounting for time-varying illness severity and death. Research Design: We performed a prospective cohort study within medical and surgical ICUs in a large academic medical center. Subjects: We analyzed critically ill patients (N=479) with respiratory failure and/or shock. Measures: Covariates included baseline factors (age, insurance, cognitive impairment, comorbidities, Acute Physiology and Chronic Health Evaluation II Score) and time-varying factors (sequential organ failure assessment score, mechanical ventilation, and severe sepsis). The primary analysis used a novel 3-stage regression method: first, estimation of the cumulative cost of delirium over 30 ICU days and then costs separated into those attributable to increased resource utilization among survivors and those that were avoided on the account of delirium’s association with early mortality in the ICU. Results: The patient-level 30-day cumulative cost of ICU delirium attributable to increased resource utilization was $17,838 (95% confidence interval, $11,132–$23,497). A combination of professional, dialysis, and bed costs accounted for the largest percentage of the incremental costs associated with ICU delirium. The 30-day cumulative incremental costs of ICU delirium that were avoided due to delirium-associated early mortality was $4654 (95% confidence interval, $2056–7869). Conclusions: Delirium is associated with substantial costs after accounting for time-varying illness severity and could be 20% higher (∼$22,500) if not for its association with early ICU mortality.

BACKGROUND In discharge planning, a patient needs assessment helps to identify risk factors that should be addressed to promote a safe and effective transition in care. Low health literacy is associated with worse postdischarge outcomes, but little research has examined its relation to other addressable risk factors. OBJECTIVE To examine the association of health literacy with the number and type of transitional care needs (TCN) among patients being discharged to home. DESIGN, SETTING, PARTICIPANTS A cross‐sectional analysis of patients admitted to an academic medical center. MEASUREMENTS Nurses administered the Brief Health Literacy Screen and documented TCNs along 10 domains: caregiver support, transportation, healthcare utilization, high‐risk medical comorbidities, medication management, medical devices, functional status, mental health comorbidities, communication, and financial resources. RESULTS Among the 384 patients analyzed, 113 (29%) had inadequate health literacy. Patients with inadequate health literacy had needs in more TCN domains (mean = 5.29 vs 4.36; P < 0.001). In unadjusted analysis, patients with inadequate health literacy were significantly more likely to have TCNs in 7 out of the 10 domains. In multivariate analyses, inadequate health literacy remained significantly associated with inadequate caregiver support (odds ratio [OR], 2.61; 95% confidence interval [CI], 1.37‐4.99) and transportation barriers (OR, 1.69; 95% CI, 1.04‐2.76). CONCLUSIONS Among hospitalized patients, inadequate health literacy is prevalent and independently associated with other needs that place patients at a higher risk of adverse outcomes, such as hospital readmission. Screening for inadequate health literacy and associated needs may enable hospitals to address these barriers and improve postdischarge outcomes.

A novel clustering method is proposed for mammographic mass segmentation on extracted regions of interest (ROIs) by using deterministic annealing incorporating circular shape function (DACF). The objective function reported in this study uses both intensity and spatial shape information, and the dominant dissimilarity measure is controlled by two weighting parameters. As a result, pixels having similar intensity information but located in different regions can be differentiated. Experimental results shows that, by using DACF, the mass segmentation results in digitized mammograms are improved with optimal mass boundaries, less number of noisy patches, and computational efficiency. An average probability of segmentation error of 7.18% for well-defined masses (or 8.06% for ill-defined masses) was obtained by using DACF on MiniMIAS database, with 5.86% (or 5.55%) and 6.14% (or 5.27%) improvements as compared to the standard DA and fuzzyc-means methods.

Objective: Support vector machines (SVMs) have drawn considerable attention due to their high generalisation ability and superior classification performance compared to other pattern recognition algorithms. However, the assumption that the learning data is identically generated from unknown probability distributions may limit the application of SVMs for real problems. In this paper, we propose a vicinal support vector classifier (VSVC) which is shown to be able to effectively handle practical applications where the learning data may originate from different probability distributions.Methods: The proposed VSVC method utilises a set of new vicinal kernel functions which are constructed based on supervised clustering in the kernel-induced feature space. Our proposed approach comprises two steps. In the clustering step, a supervised kernel-based deterministic annealing (SKDA) clustering algorithm is employed to partition the training data into different soft vicinal areas of the feature space in order to construct the vicinal kernel functions. In the training step, the SVM technique is used to minimise the vicinal risk function under the constraints of the vicinal areas defined in the SKDA clustering step. Results: Experimental results on both artificial and real medical datasets show our proposed VSVC achieves better classification accuracy and lower computational time compared to a standard SVM. For an artificial dataset constructed from non-separated data, the classification accuracy of VSVC is between 95.5% and 96.25% (using different cluster numbers) which compares favourably to the 94.5% achieved by SVM. The VSVC training time is between 8.75 s and 17.83 s (for 2-8 clusters), considerable less than the 65.0 s required by SVM. On a real mammography dataset, the best classification accuracy of VSVC is 85.7% and thus clearly outperforms a standard SVM which obtains an accuracy of only 82.1%. A similar performance improvement is confirmed on two further real datasets, a breast cancer dataset (74.01% vs. 72.52%) and a heart dataset (84.77% vs. 83.81%), coupled with a reduction in terms of learning time (32.07 s vs. 92.08 s and 25.00 s vs. 53.31 s, respectively). Furthermore, the VSVC results in the number of support vectors being equal to the specified cluster number, and hence in a much sparser solution compared to a standard SVM.Conclusion: Incorporating a supervised clustering algorithm into the SVM technique leads to a sparse but effective solution, while making the proposed VSVC adaptive to different probability distributions of the training data. (C) 2014 Elsevier B.V. All rights reserved.

Background Restricted interests are a class of repetitive behavior in autism spectrum disorders ( ASD ) whose intensity and narrow focus often contribute to significant interference with daily functioning. While numerous neuroimaging studies have investigated executive circuits as putative neural substrates of repetitive behavior, recent work implicates affective neural circuits in restricted interests. We sought to explore the role of affective neural circuits and determine how restricted interests are distinguished from hobbies or interests in typical development. Methods We compared a group of children with ASD to a typically developing ( TD ) group of children with strong interests or hobbies, employing parent report, an operant behavioral task, and functional imaging with personalized stimuli based on individual interests. Results While performance on the operant task was similar between the two groups, parent report of intensity and interference of interests was significantly higher in the ASD group. Both the ASD and TD groups showed increased BOLD response in widespread affective neural regions to the pictures of their own interest. When viewing pictures of other children's interests, the TD group showed a similar pattern, whereas BOLD response in the ASD group was much more limited. Increased BOLD response in the insula and anterior cingulate cortex distinguished the ASD from the TD group, and parent report of the intensity and interference with daily life of the child's restricted interest predicted insula response. Conclusions While affective neural network response and operant behavior are comparable in typical and restricted interests, the narrowness of focus that clinically distinguishes restricted interests in ASD is reflected in more interference in daily life and aberrantly enhanced insula and anterior cingulate response to individuals’ own interests in the ASD group. These results further support the involvement of affective neural networks in repetitive behaviors in ASD .

Children born with an inhibited temperament are at heightened risk for developing anxiety, depression and substance use. Inhibited temperament is believed to have a biological basis; however, little is known about the structural brain basis of this vulnerability trait. Structural MRI scans were obtained from 84 (44 inhibited, 40 uninhibited) young adults. Given previous findings of amygdala hyperactivity in inhibited individuals, groups were compared on three measures of amygdala structure. To identify novel substrates of inhibited temperament, a whole brain analysis was performed. Functional activation and connectivity were examined across both groups. Inhibited adults had larger amygdala and caudate volume and larger volume predicted greater activation to neutral faces. In addition, larger amygdala volume predicted greater connectivity with subcortical and higher order visual structures. Larger caudate volume predicted greater connectivity with the basal ganglia, and less connectivity with primary visual and auditory cortex. We propose that larger volume in these salience detection regions may result in increased activation and enhanced connectivity in response to social stimuli. Given the strong link between inhibited temperament and risk for psychiatric illness, novel therapeutics that target these brain regions and related neural circuits have the potential to reduce rates of illness in vulnerable individuals.

Although an extensive literature exists on the neurobiological correlates of dyslexia (DYS), to date, no studies have examined the neurobiological profile of those who exhibit poor reading comprehension despite intact word-level abilities (specific reading comprehension deficits [S-RCD]). Here we investigated the word-level abilities of S-RCD as compared to typically developing readers (TD) and those with DYS by examining the blood oxygenation-level dependent response to words varying on frequency. Understanding whether S-RCD process words in the same manner as TD, or show alternate pathways to achieve normal word-reading abilities, may provide insights into the origin of this disorder. Results showed that as compared to TD, DYS showed abnormal covariance during word processing with right-hemisphere homologs of the left-hemisphere reading network in conjunction with left occipitotemporal underactivation. In contrast, S-RCD showed an intact neurobiological response to word stimuli in occipitotemporal regions (associated with fast and efficient word processing); however, inferior frontal gyrus (IFG) abnormalities were observed. Specifically, TD showed a higher-percent signal change within right IFG for low-versus-high frequency words as compared to both S-RCD and DYS. Using psychophysiological interaction analyses, a coupling-by-reading group interaction was found in right IFG for DYS, as indicated by a widespread greater covariance between right IFG and right occipitotemporal cortex/visual word-form areas, as well as bilateral medial frontal gyrus, as compared to TD. For S-RCD, the context-dependent functional interaction anomaly was most prominently seen in left IFG, which covaried to a greater extent with hippocampal, parahippocampal, and prefrontal areas than for TD for low- as compared to high-frequency words. Given the greater lexical access demands of low frequency as compared to high-frequency words, these results may suggest specific weaknesses in accessing lexical-semantic representations during word recognition. These novel findings provide foundational insights into the nature of S-RCD, and set the stage for future investigations of this common, but understudied, reading disorder.